Fresh whole gingerbread plum kernels (Neocarya macrophylla) were obtained from two different locations (one from Birni N’Gaouré, southern region of Republic of Niger and the other from Gaoul in the Boke region, Republic of Guinea). The kernels were kept dried in desiccators at room temperature. To extract the oils, the kernels were milled using a laboratory scale hammer miller. The resulting paste was dispersed in n-hexane at a paste to n-hexane ratio of 1:5 (w/v) and stirred for 15 min at room temperature and let stand for 2 h. The supernatant, which is the mixture of n-hexane and oil, was decanted and the cake was re-extracted with n-hexane in the same conditions. This process was repeated until no trace of oil was found in the solvent. Then, the different supernatants were combined and the n-hexane was evaporated using a rotary evaporator at 40 °C. The extracted oils were transferred into glass vials, flushed with nitrogen and maintained at −18 °C until use.

IV, SV, FFA, AV, PV, color and pH were the indices measured for the extracted oils. FFA, IV and SV were determined according to the methods Ca5a-40, Cd 125 and Cd 3-25, respectively, from the AOCS Official Methods of Analysis (AOCS, 1993a; AOCS, 1993b; AOCS, 1993c). The PV was determined by method Cd 8-53 of the AOCS Official Methods of Analysis (AOCS, 1998) and AV was determined based on the AOAC Official Methods of analysis (AOAC, 1998). To determine the color parameters, a Hunter Lab digital colorimeter was used to measure gingerbread plum kernel oil’s color and color scales L*a*b* values were recorded.

The fatty acid composition of gingerbread plum kernel oil from Niger (GPKON) and gingerbread plum kernel oil from Guinea (GPKOG) was determined as reported by Amza et al. (2010). Fat was extracted with methyl ether which was prepared directly with the treatment of the fat with sodium methoxide. A gas chromatography/mass spectra (GC/MS) system was used to identify and quantify the fatty acids of the product developed on a FINNIGAN TRACE MS gas chromatograph/mass spectra equipped with a 30 m×0.25 mm Ov-1701 column. The column flow rate w as 0.8 mL·min⁻¹ with helium as the carrier gas, the split was 64 mL·min⁻¹ and the source temperature was 270 °C. The fatty acid methyl esters were identified by comparison with the retention times of NU CHECK Inc. standards (Elysian, 1L) and quantified by internal normalization.

A standard solution was prepared using a gallic acid solution at a concentration of 0.1–1.0 mg·mL⁻¹. The reaction mixture is composed of 50 µL of standard or sample solution, 200 µL of freshly prepared Folin–Ciocalteau reagent, and 3 mL of distilled water. The mixture is left at room temperature for 10 min and 500 µL of 20% sodium carbonate was added. The solution was mixed and incubated in water bath at 40 °C for 20 min and the reaction was stopped in an ice bath. The absorbance was measured at 765 nm and distilled water was used as a blank (Zhou and Yu, 2004). TPC contents were quantified and expressed as Gallic Acid Equivalent (GAE) from a calibration curve; y=0.0095x±0.046 (R²=0.9883).

Rutin is used as standard for the estimation of total flavonoids. 10 mg of rutin were dissolved in 10 mL of methanol and a calibration curve was made from aliquots of 0.1, 0.2, 0.4, 0.6, 0.8 and 1.0 mL from the above stock solution. To each concentration, 1.5 mL of methanol, 0.1 mL of 10% aluminum chloride, 0.1 mL of 1M potassium acetate and 2.8 mL of distilled water were added. The reaction mixture was kept at room temperature for 30 min and the volume was made up with water. The absorbance of the resulting solutions was measured at 415 nm against a blank. The calibration curve was prepared by plotting absorbance Vs concentration and it was found to be linear over this concentration range of 10–100 µg·mL⁻¹. The samples (1 m·mL⁻¹) were prepared in the same conditions as described above. The concentration of total flavonoids in the test samples was determined by extrapolation from the calibration curve. The total flavonoid content was expressed as µg·mL⁻¹.

β-carotene was isolated and quantified as described by Gimeno et al. (2000). 400 mg of sample were inserted into tubes with a screw-on top and 0.2 g of ascorbic acid, 15 mL of absolute ethanol and 4 mL of potassium hydroxide (76% m/V) were added for the saponification process, which was performed under nitrogen atmosphere. The tubes were continuously shaken for 30 min at 70 °C and the samples were then cooled in an ice bath and 5 mL of NaCl (2.5% m/V) were added to provoke phase separation. A mixture of n-hexane/ethyl acetate (85:15 v/v) was also added in order to extract the non-saponifiable fraction. The organic phases obtained were evaporated to dryness at 40 °C using a rotor vapor. The residue was then re-dissolved in methanol. In this process, the chromatographic injection must be performed as early as possible to avoid the oxidation and decomposition of the β-carotene. Alternatively, the sample can be stored at −20 °C for a maximum of one week.

For the preparation of unsaponifiable matter, mainly sterols; each oil sample (400 mg) was mixed with KOH (10%) in methanol (50 mL) using sonication and horizontally shaken at 30 °C for 24 h. After incubation, distilled water (10 mL), ethanol (1 mL) and hexane (20 mL) were added. The samples were vigorously shaken in separatory funnels and after partition, the hexane layer was collected. The remaining aqueous/alcohol phase was re-extracted twice with hexane (20 mL). The hexane extracts were pooled and dried over anhydrous MgSO₄ and the solvent removed in vacuo. The dry extracts were suspended in dichloromethane (10 mL). Prior to GC-MS, 50 mL (dried under nitrogen) of these suspensions were derivatized with bis (trimethylsilyl) trifluroacetamide (BSTFA) (100 µL) at 37 °C for 15 min.

Normal phase HPLC with fluorescence detection (excitation 292 nm, emission 325 nm) was used to analyze tocopherols (Kamal-Eldin et al., 2000). The HPLC system consisted of a pump (Waters 510; Waters Corp., Milford, MA), an auto-sampler with a cooling module (Waters 712), a scanning fluorescence detector (Waters 474), and an Inertsil silica column (5 mm, 250 mm* 4.6 mm; Varian Chromapack, Middelburg, Netherlands) with a silica guard column (Guard-Pak Silica, Waters). The temperature of the column oven was 30 °C. Separation of the vitamers was based on isocratic elution. The mobile phase contained 3% 1, 4-dioxane and 97% n-hexane. The flow rate of the mobile phase was 2 mL·min⁻¹. Tocopherols were quantified with an external standard method in which quantification was based on peak areas. The method was validated by determining the following parameters: detection and determination limits, range of linearity and repeatability. Detection limits were defined as a signal three times the height of the noise. Determination limits were defined as three times the detection limit.

All experiments were conducted in triplicate with SPSS Inc. software (version 13.0). One-way analysis of variance (ANOVA) was used to determine significant differences among means, with the significance level taken at a=0.05. Tukey’s HSD test was used to perform multiple comparisons among means.

The FA compositions of GPKON and GPKOG are presented in Table 2. The major mono-unsaturated fatty acid (MUFA) in both samples was oleic acid; 42.46% (GPKON) and 41.43% (GPKOG). Linoleic acid and arachidonic acid were the polyunsaturated fatty acids (PUFA) with arachidonic acid present in higher levels (17.67% and 21.72% in GPKON and GPKOG, respectively) than those in common vegetable oils. The main SFA were palmitic acid and stearic acid and the remaining fatty acids occurred in small amounts (<1%). The percentages of SFA, MUFA, and PUFA and the ratio of MUFA/PUFA for both samples were also investigated. Total SFA values were 15.90 and 17.18 for GPKON and GPKOG, respectively. These results are in the range of those (14.87 and 20.88%) reported by Noorali et al. (2014) for several olive cultivars grown in different areas, but lower than the values for coconut (86.5%) and plam oils (47.8%) (Foster et al., 2009). The MUFA contents in GPKON and GPKOG; (48.31 and 42.51% respectively) (Table 2) were much higher than those reported for safflower (12.0%), sunflower (20.5%), soybean (21.3%), corn (29.9%) and palm (37.1%) oils (Zhang et al., 2009). While PUFA values were found to be 35.08% (GPKON) and 40.40% (GPKOG) due to the high levels of linoleic and arachidonic acids found in this study. The MUFA/PUFA ratio is an important parameter for oil stability in highly unsaturated oils. Thus, regarding fatty acid composition, GPKON can be considered as a little bit more stable due to its relatively higher MUFA/PUFA ratio. However, the MUFA/PUFA values found in this study (1.38 for GPKON and 1.05 for GPKOG) are much lower than those reported by Moayedi et al. (2011) for several wild almond specie oils and are categorized as AZ (3.86), AH (3.24) and AJ (3.34).

Preliminary antioxidant tests revealed DPPH radical scavenging activity and reducing power in a dose–dependent manner (Fig. 2 and Fig. 3). The IC₅₀ values of DPPH radical scavenging activity and AC_0.5 values for the reducing power of GPKON and GPKOG are shown in Table 3. Lower IC₅₀ or AC_0.5 values indicate stronger antioxidant activities. The results of DPPH radical-scavenging activities and reducing power showed significant differences (p<0.05). GPKOG exhibited two times greater DPPH radical scavenging activity, (IC₅₀=0.24 mg·mL⁻¹) than GPKON (IC₅₀=0.56 mg·mL⁻¹) while GPKON had better reducing power with an AC_0.5 value of 1.76 mg·mL⁻¹ (Table 3). On the other hand, GPKON and GPKOG showed higher DPPH radical scavenging activity when compared with different palm oils categorized as crude palm olein (IC₅₀=19.2 mg·mL⁻¹), crude palm oil (IC₅₀=20.9 mg·mL⁻¹) and crude palm stearin (IC₅₀=30.7 mg·mL⁻¹) (Kumar and Krishna, 2014). As for β-carotene and flavonoid contents, the values were found to be 1.53; 2.06 mg·mL⁻¹ and 86.25; 71.11 µg·mL⁻¹ for GPKON and GPKOG, respectively (Table 3). It is well-documented that β–carotene and flavonoids are natural antioxidants. Flavonoids can directly react with superoxide anions and lipid peroxyl radicals and consequently inhibit or break the chain of lipid peroxidation. In the case of β-carotene, some authors reported values of 6.43 mg·kg⁻¹ and 5 mg·kg⁻¹ for extra virgin olive oil (Anniva and Tsimidou, 2009).

The total phenol contents of GPKON and GPKOG are expressed as their equivalent gallic acid in mg per kg oil. The values are 184.22 mg·kg⁻¹ (GPKON) and 250.77 mg·kg⁻¹ (GPKOG) (Table 3). The value for regular almond was 37.7 mg·kg⁻¹ (Moayedi et al., 2011), which is much more inferior to the values found in this study. Wijeratne et al. (2006) reported a total phenolic content of 8 mg (quercetin equiv·g⁻¹ of ethanolic extract) in whole almond seed while Laincer et al. (2014) found a level of 218.66 mg·kg⁻¹ total phenolic contents present in olive oils from Algeria. Our results are in the range of these findings. A direct relationship has been found between the contents of total phenolics and the antioxidant capacity of plants. Thus, GPKOG, presenting the higher level of total phenolic content, appeared also to be more active as an antioxidant source.

The tocols (tocopherols and tocotrionols) are the most important class of antioxidants naturally present in vegetables oils and fats. These natural fat-soluble antioxidants exist in at least seven forms with α, β, γ, and δ predominating in most vegetable oils. Data about the qualitative composition of tocopherols are summarized in Table 2 and Fig. 4. Three tocopherols were found, wherein α-tocopherol constituted 97.88 and 62.96% of the total in GPKON and GPKOG, respectively, followed by β-tocopherol and δ-tocopherols in both samples (Table 2). ϒ-tocopherol and α-tocopherol proved to be the main tocopherols in vegetable oils and fats (Schwartz et al., 2008) even though γ-tocopherol was not detected in the present study. ϒ-tocopherol is the major tocopherol present in soybean oil with the sigma, alpha and beta compounds present in decreasing quantities; 62.8, 26.7, 9.3 and 1.2%, respectively (Gunstone et al., 2002). Values of 42.5% β-tocopherol, 41.3% γ-tocopherol, 15.9% α-tocopherol and 0.3% δ-tocopherol were found in canterbury bell (Campanula medium) seed oil (Hassaniena et al., 2014). The antioxidant properties of tocopherols may be involved in combating atherosclerosis by preventing the oxidation of low-density lipoproteins. The levels of tocopherols detected in gingerbread plum kernel may contribute to the stability of GPKON and GPKOG toward oxidation.

The sterol composition of GPKON and GPKOG is shown in Table 4. Eleven sterol compounds were detected in both samples, wherein 24-hydroxy-24-methyl cholesterol, sitosterol and clerosterol accounted for 68.5% and 66.33% in GPKON and GPKOG respectively. The sterol marker was 24-hydroxy-24-methyl cholesterol (39.34% in GPKON and 41.55% in GPKOG) followed by sitosterol and clerosterol. Other components such as 7-α-hydroxycholesterol, Δ⁵-avenasterol, 24-dihydrolanosterol, sitostanol, β-amyrin, 24-hydroxycholesterol, campesterol and phytol were found in low levels (<7%), (Table 4). Among the different phytosterols, β-sitosterol has been most intensively investigated with respect to its beneficial and physiological effects on health. β-sitosterol lowers cholesterol levels, enhances immunity, and has anti-inflammatory, antipyretic and anti-carcinogenic effects (prostate essentially) (Pegel, 1997; Kritchevsky and Shirley, 2005). However, compared with other vegetable oils, GPKON and GPKOG contain less sitosterol; 16.18% and 13.17%, respectively. Reported values of β-sitosterol were 88% and 91% in wild pistachio and wild almond seed oils respectively (Givianrad et al., 2013), 54.1% in soybean, 60.3% in corn, and 46.1% in coconut oils (Gunstone et al., 2002) while, Δ⁵-avenasterol was found to be 2.5%, 10.5% and 27.4% in soybean, corn and coconut oils respectively and campesterol accounted for 18.1%, 17.2% and 8.8% in soybean, corn and coconut oils (Gunstone et al., 2002). The values for Δ⁵-avenasterol and campesterol in GPKON and GPKOG were 4.47%, 6.73% and 4.82, 2.36%, respectively (Table 4). Since the structure of phytosterols resembles that of cholesterol, these compounds may be involved in similar oxidative reactions. Przybylski and Eskin (1988) found some oxidation products formed from plant sterols during the storage of fried food products. However, as sterols are affected by processing; about 40% of these components can be removed from the oil during deodorization.