Tocopherols, natural antioxidant molecules that scavenge free radicals in biological systems, are abundantly present in plant seed oil. They exist in four isomers, i.e. α, β, γ and δ tocopherol, depending upon the number and position of methyl substituents on the chromanol ring, with α tocopherol having the maximum of three, β and γ tocopherol having two, and δ tocopherol having only one substituent. An increased tendency to release hydrogen atoms occurs with increasing numbers of methyl substituents on the chromanol ring. Thus, the antioxidative activities of tocopherols in biological systems against lipid peroxidation are α > β > γ > δ (100, 50, 10 and 3% relative activity, respectively) [1]. Moreover, α tocopherol is preferentially retained and distributed in the body [2]. Medical evidence has indicated that an intake of 400 IU of tocopherols results in a decreased risk for atherosclerosis, cancer, degenerative diseases like Alzhemeir and Parkinson’s disease and an improved immune system [3, 4]. With regard to the contribution of tocopherols in the protection of oxidation-sensitive fatty acids from the free radicals (under in vitro conditions), the relative anti-oxidant activities of tocopherols are α < β < γ < δ which is the reverse of the in vivo activities.

Soybean oil constitutes about 30% of worldwide consumption of vegetable oils; and unlike other major vegetable oils available in the market, it contains all four isomers of tocopherols with the γ isomer possessing the maximum (60%) proportion of total tocopherols content [5]. All four naturally occurring tocopherol isomers in crude oil decrease in the refining process to varying degree [6, 7] mainly in the deodorization step, where γ tocopherol suffers the greatest loss [6]. This necessitates the need for a breeding program with the objective of enhancing the content of tocopherols in soybean seeds. As genotypic variability is the basis of any breeding program pertaining to the development of genotypes with an increased level of a particular quality trait, it becomes imperative to study the genetic variability for tocopherols in soybean genotypes to identify genotypes with high levels of tocopherols. However, very limited information concerning genotypic variation for tocopherols isomers in soybean seed oil is available in the literature [8, 9]. McCord et al. [8] investigated three isomers of tocopherol (α, γ and δ tocopherol) in three populations derived from three single crosses between normal linolenate and reduced linolenate lines. Panizzi and Erhan [9] investigated seeds of 161 Brazilian soybean cultivars for all four isomers of tocopherols. The tocopherol contents of Indian soybean genotypes have not yet been investigated.

Soybean seed oil contains three major unsaturated fatty acids viz. oleic acid (23%), linoleic acid (53%) and linolenic acid (8%). The rate of oxidation of linolenic acid, linoleic acid and oleic acid is in the ratio of 21.6:10.3:1 [10], respectively indicating that these fatty acids play an important role in the oxidation of soybean oil. Since the oxidative stability of vegetable oils is a function of tocopherols as well as the fatty acid composition, the relationship between tocopherol isomers and different unsaturated fatty acids needs to be investigated. In view of the limited reports focusing on the association of different tocopherols with fatty acids in soybean seed oil available in the literature [8, 11, 12], the present investigation was undertaken to survey the genotypic variability for tocopherols among soybean genotypes recommended for use in India and to study the relationship of tocopherols isomers with each other as well as with fatty acids and oil content.

Materials and Methods

Materials

Sixty-six cultivars of soybean were raised following standard agronomic practices in single row plots of 5 m long with 0.45 m spacing between rows in the experimental field of the National Research Centre for Soybean, Indore, (22°N), situated in the Malwa region of Madhya Pradesh, India, in the 2005 cropping season. Madhya Pradesh is the hub of soybean cultivation in India; and during the cropping season 2005, at Indore, the average daily mean temperature was 25.9 °C and the rainfall received was 620.4 mm. Of the selected genotypes, some of the genotypes have one parent from maturity group V–VII. Seeds of all the genotypes harvested at their respective maturity were evaluated for oil content, fatty acids and four tocopherols (i.e. α, β, γ, and δ tocopherol).

Methods

(1) Extraction and estimation of oil content. Oil from ground seeds (1 g) was extracted with 180 ml hexane in an automated Soxhlet unit (Pelican Equipments, Chennai, India) for 3 h. Percent oil content was determined by weight differences.

(2) Determination of fatty acid composition. Ground seeds (200 mg) were soaked in 2.5 ml petroleum ether (boiling point 40–60 °C) overnight at room temperature. The petroleum ether–oil mixture was decanted and kept in a water bath at 75 °C till the petroleum ether was completely removed. The oil extracted was trans-esterified using 1 N sodium methoxide in anhydrous methanol [13]. Fatty acid methyl esters (FAMEs) were prepared, separated and analyzed using Shimadzu GC 17A, using capillary column (SGE BPX70), with a length and diameter of 30 m and 0.32 mm, respectively. The oven temperature of GC was programmed at 140 °C for 3.6 min, then increased to 170 °C at the rate of 13.5°C/min and maintained for 3.8 min and finally increased to 182 °C at 5 °C/min. The flame ionization detector (FID) and injector were maintained at 240 °C. Nitrogen, the carrier gas used, was maintained at a flow rate of 15 ml/min with a column pressure of 90 kPa. Hydrogen and air were maintained at 50 kPa for proper detector operation. The peaks obtained for fatty acid methyl esters were identified by comparing the retention times with those of standard fatty acid methyl esters (Sigma-Aldrich, India).

(3) Extraction and determination of tocopherols using HPLC. Twenty randomly taken seeds of each soybean genotype, in triplicate, were ground using a metallic pestle and mortar into a fine flour and sieved (500 μm size). Oil from the fine soy flour was extracted by soaking the flour in HPLC-grade hexane for 8 h. The hexane–oil mixture was transferred into vials and the solvent was evaporated using a freeze drier. The weight of the oil was determined gravimetrically in each vial and the samples were re-dissolved in HPLC-grade n-hexane to a fixed volume (1.0 ml). Tocopherol composition was determined using a Shimadzu HPLC system equipped with a UV detector and a silica-NH2column (5 μm; Phenomenex; Spinco Biotech, India) with dimensions of 250 × 4.6 mm. Twenty microlitres of a syringe-filtered sample were injected into the column and eluted isocratically with HPLC-grade n-hexane and ethyl acetate (70:30 v/v) at a flow rate of 1.0 ml/min. The tocopherols were detected with a UV detector (SPD 10 AT vp) at a wavelength of 295 nm. The relative amounts of tocopherols were calculated by comparing their peak areas with a standard curve generated using different amounts of external standards of α, β, γ and δ tocopherol (Sigma-Aldrich, India). Total tocopherols content was computed by summing up the values of all the four isomers. The tocopherols data were expressed as μg/g oil basis. The contents of α, β, γ and δ tocopherol were also determined as a percentage of total tocopherol.

(4) Computation of Vitamin E activity. The vitamin E activity of soybean oil from different cultivars was taken as the sum of multiplication of α, β, γ and δ tocopherol contents by 1.0, 0.5, 0.1, and .03, respectively as previously reported [14].

Statistical Analyses

Genotypes were arranged in randomized block design and data were analyzed using the statistical programme MSTAT-C version 2.1 (Russell D. Freed, Michigan State University). Mean values were used for correlation studies.

Results and Discussion

Values for all the four isomers of tocopherols of the Indian genotypes investigated are presented in Table 1. Genotypic variation was observed for the four isomers of tocopherol, total tocopherol content and tocopherols’ composition. The predominant tocopherol was γ tocopherol followed by δ tocopherol, α tocopherol and β tocopherol. α Tocopherol ranged from 58 μg/g for ‘MACS13’ to 794 μg/g for ‘Co Soya2’, with an overall mean value of 269 μg/g of oil. β Tocopherol ranged from 12 for ‘Pusa16’ to 121 for ‘Co Soya2’ with an overall mean value of 40 μg/g oil. γ Tocopherol ranged from 240 μg/g for ‘SL96’ to 1,794 μg/g for ‘Co Soya2’, with an overall mean value of 855 μg/g of oil while δ tocopherol ranged from 66 μg/g for ‘KB79’ to 602 μg/g for ‘Co Soya2’, with an overall mean value of 241 μg/g of oil. Total tocopherol content ranged from 422 μg/g for ‘SL96’ to 3,311 μg/g for ‘Co Soya2’, with an overall mean value of 1,405 μg/g of oil. α/γ Tocopherol ranged from 0.17 for ‘MACS13’ to 0.52 for ‘JS2’ with a overall mean value 0.308. With regard to percent tocopherol composition, α-tocopherol ranged from 11 for ‘MACS13’ to 26 for ‘MACS124’ with an overall mean value of 18%. β Tocopherol ranged from 2% for ‘VLS1’ to 6 for ‘NRC2’ with an overall mean value of 3%. γ Tocopherol ranged from 51 for ‘JS2’ to 69; ‘VLS1’ and ‘PK327’ with an overall mean value of 60% whereas δ tocopherol ranged from 15 for ‘JS90-41’ to 26 for ‘NRC2’ with an overall mean value of 18%. Total vitamin E activity (activity as α tocopherol equivalents) ranged from 102 for genotype ‘MACS13’ to 1,052 μg/g of oil for genotype ‘Co Soya2’.

Table 1 Tocopherols content (mean and standard deviation, n = 3) of some Indian soybean genotypes
Full size table

Genotypic variability was also observed for oil content and all the five major fatty acids viz. palmitic acid (C16:0), stearic acid (C18:0), oleic acid (C18:1), linoleic acid (C18:2), linolenic acid (C18:3) content (Table 2). The oil content ranged from 13 (genotypes ‘JS79-81’, ‘NRC12’, ‘PK471’) to 22 (variety ‘VLS1’) with an overall mean value of 17%. Of all the genotypes studied, three genotypes viz. ‘‘VLS1’, ‘JS71-05’, and ‘Pusa24’ exhibited an oil content of more than 20%. Among saturated fatty acids, palmitic acid (C16:0) ranged from 9% for genotype ‘VLS1’ to 15% for genotype ‘KHSb2’ while stearic acid ranged from 2% for genotype ‘Ankur’, ‘Samrat’, ‘Shilajeet’, and ‘Shivalik’ to 8% for genotype ‘VLS1’. Overall average values for palmitic and stearic acid among all the genotypes studied were 11 and 4%, respectively. Monounsaturated fatty acid, i.e. oleic acid (C18:1) ranged from 19% for genotype ‘PK327’ to 36% for genotype ‘Bhatt yellow’, with an overall mean value of 25%. Linoleic acid (C18:2) and linolenic acid (C18:3) are two major polyunsaturated fatty acids found in soybean. Linoleic acid ranged from 42% for genotype ‘Bhatt Black’ to 59% for ‘Pusa22’ while linolenic acid ranged from 5% for ‘NRC48’ and ‘Pusa24’ to 11% for genotype ‘SL96’. The overall mean linoleic and linolenic acid values, of all the soybean genotypes studied were 53 and 7% of the total fatty acids, respectively. Averaged overall the genotypes studied, mean value for M (monounsaturated fatty acid):P (polyunsaturated fatty acids) ratio, an indicator of oxidative stability of oil as linoleic acid and linolenic acid oxidize 10 and 21 times faster than oleic acid [10], was 0.44. In general, soybean oil has an M:P ratio of 0.5 which is much lower than the value of 1.95, 1.65 and 4.0 for canola, peanut and olive oil, respectively, indicating that soybean oil is less oxidatively stable than canola, peanut and olive oils.

Table 2 Range and overall mean values for oil content and fatty acid composition of some of Indian soybean genotypes
Full size table

The ratios between the lowest and the highest values for α, β, γ, δ and total tocopherol in the present study were 1:13.6, 1:10.4, 1:7.5, 1:9.1, 1:7.9 as compared to 1:17.4, 1:10.7, 1:4.4, 1: 3.3, 1:3.5 reported in Brazilian cultivars, respectively [9]. Indian genotypes exhibited more variability for γ- and δ- and total tocopherol while lower variability for α tocopherol than observed in Brazilian genotypes. McCord et al. [8] investigated 20 normal and 20 reduced linolenate soybean lines for tocopherols and reported the ranges of 133–182, 713–740, 395–440 and 1,268–1,348 μg/g of oil for α, γ, δ and total tocopherol content, respectively with the ranges of 10.5–13.5 (% α), 53.85–67.2 (% γ) and 31.15–32.4 (% δ) tocopherol. In this study [8], β tocopherol was integrated into the γ tocopherol due to insufficient separation of β and γ tocopherol. Higher average percent α tocopherol content and lower average percent γ and δ tocopherol content in our study were observed compared to the values for the corresponding isomers in the results of McCord et al. [8]. These differences may be due to genotypic differences or may be attributed to the growing temperature as suggested by Britz and Kremer [15], who reported an increase in free α tocopherol content in seeds of soybean under warmer growing conditions during seed maturation. More specifically, they found an increase in ratio of α tocopherol/total tocopherol from 0.130 at 23 °C to 0.372 at 28 °C in the ‘Essex’ variety while this ratio increased from 0.128 to 0.280 in the ‘Forrest’ variety. Interestingly, in the current study, the average value for α-/total tocopherol was 0.192 and the prevailing average daily temperature during seed development was 26.27 °C. The value obtained for α tocopherol/total tocopherol in our investigation lies in between the values obtained for the ratio at 23 and 28 °C in the earlier study [15]. Genotypic variability for tocopherol isomers has been reported in other oilseed crops as well [11, 16, 17]. Dolde et al. [11] reported ranges of 504 to 687 and 534 to 1,858 μg/g of oil in canola and sunflower, respectively. In sunflower, Velasco et al. [16] also reported genotypic variation for tocopherol content ranging from 562 to 1,872 μg/g of oil and the α tocopherol, being the predominant fraction of total tocopherol content in sunflower seeds, ranged from 88.4 to 96.3% of the total tocopherol. In corn, Rocheford et al. [17] reported ranges of 11.8–66 and 43–229 μg/g of oil for α and γ tocopherol, respectively in 45 hybrids of corn. In this study, α and γ tocopherol accounted approximately 20 and 80% of the total tocopherol content, respectively.

Table 3 shows the correlation of different tocopherols with oil content and fatty acids. Our results indicated a highly significant (p < 0.001) positive correlation of all the four tocopherols isomers and total tocopherol content with each other. Similar results were also shown in an earlier investigation [8], but with lower level of significance (p < 0.01). Furthermore, in the present study, soybean genotypes did not exhibit any significant association of α, β, γ, δ tocopherol and total tocopherol with linolenic acid, which is in contrast to the results of Dolde et al. [11]. These authors [11] found a positive association of linolenic acid with γ tocopherol but a negative association with α tocopherol. However, as the investigation was conducted only in four genotypes, the correlations obtained in their study may be misleading. A couple of reports demonstrating positive association between linolenic acid and tocopherols isomers are also available in the literature [8, 12]. Almonor et al. [12], in an investigation of 14 soybean genotypes, reported a significant positive association between tocopherols and linolenic acid [12]. McCord et al. [8] also observed positive association of α, δ tocopherol and total tocopherol with linolenic acid in an analysis of tocopherols in 20 normal and 20 low linolenic acid soybean lines. The authors suggested that it was possible to breed low linolenate soybean lines with reasonable amount of total tocopherols as some lines with low linolenic acid showed appreciable amount of tocopherols. Significant positive correlations (p < 0.05) of total tocopherols and γ tocopherol content with linoleic acid as well as PUFA content observed in the present study indicated that the alleles that govern the expression of ω-6 desaturase gene, responsible for conversion of oleic acid (C18:1) to linoleic acid (C18:2), influence the tocopherols levels as well. A similar observation was reported in cv ‘N85-2176’ carrying homozygous recessive allele for ω-6 desaturase when compared with cv ‘Dare’ carrying homozygous dominant allele for ω-6 desaturase [12]. Significant positive correlation between α tocopherol content and PUFA content has been reported in corn hybrids also [18]. In the current study, no correlation between tocopherols fractions and oil content was observed. Reports focusing on the study of the relationship between tocopherols and oil content in soybean seeds are unavailable, however, significant association of tocopherols with oil content has been reported in other oilseed crops [19, 20]. Negative association of α and γ tocopherol with oil content in canola seeds [19] whereas a positive association of β, and γ tocopherol with oil content has been reported in Cannabis sativa seeds [20]. In the present study, oil content exhibited a significant (p < 0.01) negative correlation with oleic acid and M:P.

Table 3 Correlation studied between four tocopherol isomers, total tocopherols, total vitamin E activity, unsaturated fatty acids and monounsaturated (M):polyunsaturated fatty acids (P)
Full size table

α Tocopherol is biosynthesized from γ tocopherol by γ tocopherol methyltransferase (γ TMT). Shintani and DellaPenna [21] achieved 95% of the total tocopherol pool as α-tocopherol by overexpression of γ tocopherol methyltransferase (γ TMT) gene in Arabidopsis seeds. Apart from this transgenic approach, enhanced levels of a specific tocopherol can be achieved through traditional or molecular breeding. A highly significant correlation (p < 0.001) of tocopherol isomers with each other and total tocopherol contents in our study suggests that any breeding program focusing on enhancement of one type of tocopherols would result in enhancement of other three tocopherols. Therefore, enhancement of total tocopherols content remains the only way to enhance specific type of tocopherol isomer. Furthermore, molecular markers for high levels of tocopherols need to be identified in soybean, as accomplished in corn [18], to facilitate the selection of genotypes with high levels of tocopherols in large populations in breeding programs. Genotypes, ‘CoSoya2’ and ‘Ankur’, exhibited comparatively higher values for total tocopherols content, seem to be good donors for breeding programs focusing on the enhancement of total tocopherol content. Notwithstanding, the significant (p < 0.05) positive relationship observed for γ-tocopherol and total tocopherol content with linoleic acid in the present study, there are some cultivars, which showed a comparatively high total tocopherol with low linoleic acid suggesting that it may be possible to improve the oxidative stability of soybean oil by decreasing the linoleic acid content without compromising the tocopherol content.