Twenty-four different sesame accessions (belonging to Turkey and 11 other countries) were used as plant material in this study (Table 1). The sesame accessions were procured from the Aegean Agricultural Research Institute (Kahramanmaras, Balikesir-Kepsut, Mugla-Koycegiz, Kilis, Bursa-Orhangazi, Malatya-Ispendere, Orhangazi-99), United States Department of Agriculture (India, Afghanistan, Libya, Greece1, Greece 2, Zaire, USA, Egypt1, Egypt 2, Ethiopia) and the other seven were personally collected by the author (Iran1, Iran 2, Iran 3, Iran 4, Iran 5 (Tabriz), Iraq (Karkuk), and Pakistan (Multan)). The field experiment was conducted at the Experimental Farm of Cukurova University (41°04´N, 36°71´E, and 36 m), in Adana (Turkey) for two consecutive years in 2015 and 2016. In the Adana province, a typical Mediterranean climate prevails and the winters are warm and rainy whereas the summers are hot and dry. During the study period, monthly average day temperatures ranged from 25.0 ^oC to 30.0 ^oC in 2015 and 27.1 ^oC to 29.9 ^oC in 2016. The total rainfall was 146.1 mm and 93.8 mm during the growing periods in 2015 and 2016, respectively. The average relative humidity ranged from 63.4 to 69.8% in 2015 and 66.1 to 69.0% in 2016 (Table 2).

The experiments were conducted in a Randomized Complete Block design with three replications. Before planting, 200 kg·ha^-1 of DAP (36 kg·ha^-1 N, 92 kg·ha^-1 P) fertilizers were applied. Ammonium nitrate (33%N) at the rate of 200 kg·ha^-1 was applied once before the first irrigation. Twenty-four different sesame populations (belong to Turkey and different countries) were used as plant material in this study (Table 1). Sesame seeds were sown in the second week of June, 2015 and 2016. The accessions were grown in four row plots of 5 m row length with a row spacing of 70 cm and intra-row spacing of 15 cm. Thinning was carried out after 25 days of sowing to secure one plant per 15 cm. Sprinkler irrigation was established immediately after sowing and thereafter used when necessary based on soil and plant conditions. Weeding was carried out by hand and no herbicides were applied during the growing seasons. All the plants were harvested in the last week of September, in 2015 and 2016.

The samples of 24 genotypes grown in 2015 and 2016 were subjected to oil extraction using a Soxhlet apparatus and a gravimetric method. The experiments were performed in a Randomized Complete Block design with three replicates and seeds were bulked (20 g from each replication) and 5 g clean and mature seed samples were taken for oil content and fatty acid analysis. The oil content was determined by comparing the weights of 5 g seed samples before and after extraction using a Soxhlet apparatus (FOSS) with petroleum ether for 3 h.

An oil sample of 500 mg was dissolved in 2 ml isooctane followed by 1.5 ml of 0.5 M methanolic NaOH. The tube was then vortexed and held in boiling water for 7 min and allowed to cool to room temperature. Two ml of BF3 (Boron trifluoride) were added, vortexed, and held in boiling water for 5 min and allowed to come down to room temperature. The tube was vortexed after adding 5 ml NaCl, centrifuge at 4,000 rpm for 10 min. The supernatant was used for GC analyses (AOAC 1984).

The fatty acid (FA) composition was analyzed using a GC Clarus 500 with auto sampler (Perkin Elmer, USA) equipped with a flame ionization detector and a fused silica capillary SGE column (30 m • 0.32 mm, ID • 0.25 lm, BP20 0.25 UM, USA). The oven temperature was brought to 140 °C for 5 min, then raised to 200 °C at a rate of 4 °C/min and to 220 °C at a rate of 1°C/min, while the injector and the detector temperatures were set at 220 °C and 280 °C, respectively. All analyses were performed in triplicate, and their means were reported. The data were statistically analyzed using a JUMP 8.1.0 statistical software package. The Least Significant Differences (LSD) test was used to compare the treatments at the 0.05 level.

Within the desaturation pathway, the oleic desaturation ratio (ODR) and the linoleic desaturation ratio (LDR) (Pleines and Friedt, 1988) were used to determine the efficiency of the desaturation from OA to LA (ODR) and from LA to ALA (LDR). They were calculated as follows:

The data were statistically analyzed by One-Way ANOVA using the SPSS software.

The oil content and fatty acid composition in oilseed crops is affected by genotype, location, temperature, moisture content, growing conditions, planting date, fertilization, and the interaction these factors (Ul-Hassan and Ahmed, 2012).

A significant variation in oil content was found among the 24 sesame accessions as shown in Table 3. The oil content of the 24 sesame accessions varied between 42.8-54.37% over a two-year period, the average being 47.7%. The highest oil content was obtained from the Mugla-Koycegiz (T) Libya accession, whereas the lowest value was obtained from the Iran 1 accession. The range of oil content found was narrower than that in the Turkish (Uzun et al., 2008), worldwide (Yermanos et al., 1972), and Mediterranean core collection (Yol et al., 2015). These differences may have originated due to different genetic backgrounds and growing conditions.

The fatty acid compositions of the 24 sesame accessions are shown in Tables 3-5. In sesame oil, oleic acid and linoleic acid are the major fatty acids which form more than 80% of the total oil content. These were found in large amounts in the oil of all sesame accessions. The data belonging to the oil content and fatty acid composition for 2015, 2016 and in a two-year average are presented in Tables 3–5. Statistically significant differences were observed among the accessions for oil content and fatty acid composition in 2015, 2016 and in a two-year average. The oil contents varied between 42.9-53.5% in 2015 and between 42.8-54.37% in 2016 (Tables 3-4). According to a two-year average, the highest oil content was recorded for the Greece 2 (53.12%) accession, while the lowest oil content was recorded for Iraq (44.63%).

The highest oleic (43.09%) and the lowest linoleic acid (40.39%) percentages were obtained from the Iraq accession for 2015 and the highest linoleic acid (46.38%) and the lowest oleic acid (36.14%) percentages were obtained from the Iran 3 accession in 2015 (Table 3). The Kilis (T) accession had the highest oleic (43.63%) and the lowest linoleic (39.13%) acid contents; while the Libya accession had the higest linoleic (46.13%) and the lowest oleic (36.65%) acid contents in 2016 (Table 4). The oleic acid contents varied between 36.14 and 43.63% with a mean value of 39.02% (Table 5). Linoleic acid in sesame seed oil ranged from 36.13 to 46.38% with an average value of 43.64%. The linoleic acid content was thus found to be higher than the oleic acid content in sesame seed oil. Oleic acid content percentages in earlier reports by Uzun et al., (2008) and Were et al., (2006) were lower than those in the present study, although they reported higher linoleic acid contents than that found in the present study.

Bhunia et al., (2015) reported that the percentage of oleic acid varied between 38 and 50%, and that linoleic acid ranged from 18 to 43%. Although linolenic acid is one of the most important unsaturated fatty acids, the linolenic acid content was very low in all sesame accessions varying from 0.28 to 0.40% (the highest in Iran1 and Iran 2),with an average of 0.32%. Over a two-year period, the highest average oleic acid percentage was obtained from Kilis (T) (43.63%) and the lowest from the Iran 3 (36.14%) sesame accession. Over a two-year period, the highest average linoleic acid content was recorded in Egypt 1 (45.74%), and the lowest was in the Kilis (T) (39.83%) sesame accession (Table 5).Sesame has more unsaturated fatty acids than many other vegetable oils, and its higher proportion of unsaturated to saturated fatty acids makes it a potentially important dietary source of essential fatty acids (Bedigian, 2010). Linoleic acid is required for cell membranes, for transportation of cholesterol in the bloodstream, and for blood clotting (Salunkhe et al., 1991).

In the present study, the total content of these unsaturated fatty acids was 83.9%. This value is similar to those reported by Uzun et al., (2008) (83.1%) and Kurt et al., (2016) (83%), higher than those reported by Bhunia et al., (2015) (75%) and Yol et al., (2016) (80.3%), and lower than those reported by Were et al., (2006) (85.9%) and the World collection (Yermanos, 1972) (85.6%). A high level of unsaturated fatty acids increases oil quality for human consumption.

Palmitic and stearic acids were the predominant saturated fatty acids in sesame oil, and were in the range of 8.19 to 10.26% and 4.63 to 6.35%, respectively. Palmitic acid is used in industrial products such as soaps, esters, and plasticizers (Ndangui et al., 2010).

Germ plasm, year and germ plasm x year effects were significant for all the fatty acids and oil contents. Most fatty acids did not differ significantly in their variation for the germ plasm x year interaction, while all the tested parameters showed significant variation in germ plasm (Table 6).

Correlations between different traits are generally due to the presence of linked genes and the epistatic effect of different genes (Baloch et al., 2014). The results of the correlation analyses of oil content and fatty acids in the sesame accessions examined are shown in Table 7. Oil content was negatively and significantly correlated in oleic and stearic acids, but not in palmitic and linoleic acids. A similar relationship between oil content and oleic acid has been reported in sesame seeds (Yol et al., 2015; Were et al., 2006).

Stearic acid was significantly positively correlated with oleic and arachidic acid, while it had an inverse association with both linoleic and linolenic acids. Such an association was also reported by Yol et al., 2015. Linoleic acid was significantly negatively correlated with arachidic acid. Were et al., (2006) also reported a similar relationship between oleic and arachidic acids. Oleic acid had a strong negative correlation with linoleic acid (r = -0.919) (Table 7). This was expected because linoleic acid is synthesized from oleic acid (Yol et al., 2015). A similar finding has been reported by other researchers for other oilseed crops including peanut (Mora-Escobedo et al., 2015) and soybean (Farno, 1996). The results indicated that different enzymatic processes of fatty acid synthesis are the principal determinant of the fatty acid composition in sesame seeds (Were et al., 2006). This inverse relationship between oleic and linoleic acids may be helpful in evaluating varieties that are rich in either oleic acid or linoleic acid (Pleines and Friedt, 1988; Baydar et al., 1999).

The nutritional quality of edible seed oils is determined ideally by the presence of a lower level of saturated fatty acids (SFA) [e.g. palmitic acid (C16:0), stearic acid (C18:0)], and a high level of unsaturated fatty acids [e.g. oleic acid (OA), linoleic acid (LA) and α-linolenic acid (ALA)]. Variations in the UFA/SFA ratio in different accessions are shown in Table 8. Taking into account all fatty acids, the average content of unsaturated fatty acids was 83.93%, the highest UFA value being obtained from the Iran 2 (84.82%) accession. A high level of unsaturated fatty acids (UFA) enhances the quality of oil for human consumption. In particular, a high level of linoleic acid in the oil decreases blood cholesterol levels and plays an important role in preventing atherosclerosis (Ghafoorunissa, 1994). Among the accessions, the highest O:L ratio was recorded in Kilis (T) (1.07); while the lowest was in Ethiopia (0.82).

It is difficult to evaluate suitable cultivars with beneficial health attributes solely on the basis of their phenotypic characteristics. The fatty acid biosynthetic pathway is highly inter-connected and any breeding modification will affect the entire system (Pleines and Friedt, 1988). Therefore, the two ratios were additionally developed and used to overcome this problem, the oleic de-saturation ratio (ODR) and the linoleic de-saturation ratio (LDR) (Bhunia et al., 2015; Mondal et al., 2015).

The variations in the ODR and LDR ratios are shown in Table 8. Among all the accessions, LDR values were found to be very low because of the low level of linolenic acid in sesame oil. The average ODR value was 0.53, while the average LDR value was 0.007. While the average ODR value found was higher than that reported by Bhunia et al., (2015) (0.434) because of lower oleic acid content, the average LDR value was lower than that reported by Bhunia et al., (2015) (0.013) because of higher linoleic acid and lower linolenic acid contents. Mondal et al., (2010) reported that the mean value of ODR was 0.5, while the mean value of LDR was 0.01.

The oleic acid and linoleic acid ratio varied between 0.82 and 1.07. Among the accessions, the highest O/L ratio was 1.07 in Kilis (T), and the lowest was 0.82 in Ethiopia because of higher linoleic acid contents. As can be seen in Table 8, the Turkish sesame accessions O/L ratio was higher than other countries.