Acetonitrile for HPLC analyses was purchased from Merck Chemical Co. (NJ, USA). 2, 2–Azobis–3–ethylbenzthiazoline–6–sulphonic acid (ABTS) was purchased from Bio Basic Inc. (Markham, Ontario, Canada). Folin–Ciocalteu phenol reagent, α, α–diphenyl–β–picrylhydrazyl (DPPH), α–, γ– and δ–Tocopherol standards were purchased from Sigma Chemical Co., (St. Louis, MO, USA). Boron trifluoride–methanol, hexane, potassium hydroxide and other chemicals were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All reagents were of analytical grade and the Methanol for HPLC was of chromatographic grade.

The seeds of Sacha Inchi were harvested from Xishuangbanna at optimum maturity, dried in the shade, and stored in 10 kg bags in a dark room at room temperature for later pressing. Oil samples were obtained by cold–pressing raw de-hulled seeds, using a screw press (KOMET, DD85G, German) in a warm press procedure at 50 °C. The extracted oil was decanted into dark bottles and stored at –20 °C until analysis.

The physicochemical characteristics of the oil samples were tested using the official American Oil Chemicals’ Society methodology including acid value (Ca 5a–40); peroxide value (Cd 8–53); saponification value (Cd 3–25); iodine value (Cd 1–25); unsaponifiable matter (Ca 6a–40); moisture and volatile compound contents (Ca 2c–25); insoluble impurities (Ca 3a–46); refractive index (Cc 7–25); specific extinction (Ch 5–91) (AOCS, 2009). The oil characterization analyses were performed in duplicate.

The fatty acid composition of the oil was analyzed by gas chromatography with a flame ionization detector (GC–FID) following Fanali et al. (2011). The fatty acids were converted into their methyl esters (FAMEs) by adding 2 mL of sodium methylate in methanol (2.5 mol·L^-1) to 100 mg of oil and incubating at 100 °C for 15 min. After the mixture was cooled, 1 mL of boron trifluoride–methanol (20% BF₃) reagent was added, and the solution was heated at 100 °C for 15 min. The solution was then cooled and mixed with 1 mL of n–hexane and 1 mL of water prior to being centrifuged for 2 min (1300 rpm). The upper layer containing FAMEs was transferred to a 2 mL vial and stored at 4 °C until GC–FID analysis (Christie, 2003).

The analysis was carried out on an Agilent–6820 gas chromatograph with a manual injector, and FID detector. FAMEs separation was performed on Agilent J&W GC columns (DBFFAP; 30 m long × 0.25 mm internal diameter, 0.25 µm film thickness) and gas (He) was used as a carrier at a constant flow of 1.2 mL·min^-1. Oven temperatures were set at 180 °C and 220 °C for start and end points, respectively, and increased at a rate of 10 °C·min^-1 and held for 5 min. The injector and detector temperatures were set at 260 °C. The sample size was 1 mL with a split ratio of 30:1. For each sample triplicates were prepared and injected. Data analysis was carried out using Agilent Cerity software.

The content of tocopherols in Sacha Inchi oil was analyzed using a High Performance Liquid Chromatography/ultraviolet–visible detector (HPLC/UV–VIS) according to a slightly modified method described by Gimeno et al. (2000). The sample was prepared by dissolving 1 g of oil in 1 mL of n–hexane. The dissolved oil samples were run on an Agilent 1100 chromatographic system with a quaternary pump (G1311A), an autosampler (G1313A), a degasser (G1322A), and a dual absorbance detector (G1315B). The chromatographic data were acquired and processed using the software package Agilent. The analytical column was Agilent Zorbax SB C–18 (4.6×250 mm and particle size of 5 µm) from Agela Technologies.

All of the analyses were performed at ambient temperature. The mobile phase was a mixture of acetonitrile–deionized water (98:2, vol/vol), and the absorption wavelength in the UV detector was 292 nm. Samples with 3 µL volume and a flow ratio of 1 mL·min^-1 were used for analysis. The amount of each compound, namely α-tocopherol, γ-tocopherol, and δ-tocopherol, was calculated from the results obtained for each standard.

The analysis of total phenolic compounds was performed using the Folin–Ciocalteu spectrophotometric method described by Hrncirik and Fritsche (2004). Briefly, a sample of 2.5 g oil was first dissolved in 5 mL hexane then extracted with methanol/water (80:20, v/v) for 2 min under nitrogen atmosphere with an electronic blender. Phase separation was achieved by centrifugation for 20 min at 4000 rpm. The extraction procedure was repeated twice, the two methanolic fractions were pooled and stored at –20 °C until analysis. A 2 mL aliquot of the methanolic phase was diluted with water to a total volume of 5 mL followed by an addition of 0.5 mL Folin–Ciocalteu reagent. After 3 min, one mL of sodium carbonate solution (35%, v/v) was added to the reaction mixture, mixed, and diluted with water to a final volume of 10 mL. The absorbance of the solution was measured at 725 nm. The calibration curve was constructed using standard solutions of gallic acid in the range of 50 to 500 µg·mL^-1. The results were expressed as gallic acid equivalent (GAE) in mg.100g^-1 of oil (mg GAE·100 g^-1 oil).

Given the complex properties of the oil, there is no single method that fully reflects the antioxidant profile of a sample, making the data compiled to date difficult to compare and interpret. In our study, radical scavenging activity (RSA) of the oil samples was assayed using both DPPH and ABTS methods. In the DPPH method, both oil samples and methanolic extracts were assayed; for the ABST method, the methanolic extracts were assayed.

A methanolic extract was extracted from 2 g of each of the Sacha Inchi oil samples, mixed with 1 mL of n–hexane, and the mixture vigorously stirred until dissolution. Two ml of a methanol/water (80:20 v/v) mixture was added to each diluted sample, followed by centrifugation (5 min 5000 rpm). The extraction was repeated twice and the methanolic extracts were combined.

The storage stability of Sacha Inchi oil was determined by acid value (AV) and peroxide value (PV) analyses during long time storage. Researchers have shown that light and temperature are important factors contributing to lipid oxidation (Fereidoon, 2005; Stark et al., 2000). The total amount of Sacha Inchi oil was stored for 12 months in colorless and green bottles with polyethylene sealing lids. Three of each type of bottle were then stored at three different temperature: 4 °C, 20 °C and 30 °C. Analyses were conducted at 2–month intervals for AV and PV.

All measurements were repeated at least three times. The values of the different parameters were expressed as mean±standard deviation (SD). The data were analyzed using SPSS software (SPSS. 16.0, USA) and SigmaPlot software (SigmaPlot. 10.0, USA) for Windows.

The Physicochemical Characteristics of Sacha Inchi oil from Xishuangbanna and Peru are shown in Table 1. Acid value (1.79±0.004 mg KOH·g^-1 oil) and peroxide value (1.846±0.067meq O₂·kg^-1 oil) are within the limits (4.0 mg KOH·g^-1 oil and 15 meq O₂·kg^-1 oil, respectively) established by The Codex Alimentarius Commission for cold–pressed and non–refined oil (FAO, 1999). The saponification value and iodine value from Xishuangbanna are slightly lower than the values from Peru. Moisture and volatile compound contents, unsaponification matter, insoluble impurities, refractive index, with specific extinction of Sacha Inchi oil were first characterized.

The fatty acid profile of Sacha Inchi oil was analyzed using GC–FID. Peak identification was obtained by conjunction with pure standard FAMEs. The chromatography results showed that the major fatty acids identified in Sacha Inchi oil are palmitic acid (4.40±0.34%), stearic acid (3.21±0.06%), oleic acid (9.29±0.32%), linoleic acid (39.57±0.20%), and linolenic acid (43.52±0.92%). Similar to the reported fatty acid data on Sacha Inchi oil from Peru, Sacha Inchi oil from Xishuangbanna is also dominated by linolenic and linoleic acids. The remarkably high content of linolenic acid (43.52%) makes the oil especially important as an alternative plant–based ω–3 source in the diet, especially in China. According to a survey on dietary intake across China, people in the Northwest region receive sufficient ω–3 fatty acids from perilla seed oil, but in other regions the ω–3/ω–6 fatty acid is lower than the World Health Organization (WHO) recommended ratio of 1:4. Therefore, it is important for people to increase ω–3 fatty acid intake for optimum health (Zhang, 2004). The fatty acid compositions of Sacha Inchi oil from Xishuangbana and Peru are similar (Table 2), with a much higher content of linolenic acid than other commonly consumed vegetable oils such as soybean and peanut oil (Fanali et al., 2011; Dyer et al., 2008; Ozcan., 2010). Therefore, Sacha Inchi oil has great potential to serve as an alternative linolenic acid source in the diet either as an edible oil or through nutraceutical supplements.

Tocopherols are an important minor component in oil that may be anti-oxidative and slow down the process of oxidation (Kamal–Eldin et al., 1996). Oils with higher levels of tocopherols are expected to exhibit greater oxidative stability. Total tocopherols in Sacha Inchi oil from Xishuangbanna was 161.87±0.85 mg·100g^-1 (Table 2), a little lower than in Sacha Inchi oil from Peru, but higher than in most other edible oils (Fanali, et al., 2011; Gliszczyńska-Świglo and Sikorska, 2004). More importantly, most of the tocopherol content detected in Sacha Inchi oil is γ–tocopherol whereas α–tocopherol is predominant in olive oil and sunflower oil. Tocopherols are the most efficient antioxidant of a lipidic nature in the human diet and it is important in the biosynthesis of related proteins (Procida et al., 2013). Research has also shown that α–tocopherol is more efficient than γ–tocopherol in inhibiting the formation of both hydroxyl– and ketodiene–derivatives at 10 ppm, but this order of efficiency was reversed at 100 ppm and higher concentrations (Makinen et al., 2000).

The total phenolic compound content of Sacha Inchi oil from Xishuangbanna was determined using the Folin–Ciocalteu technique. The concentration of phenols was 6.58±0.27 mg·100 g^-1 of oil expressed as GAE, similar to the reported value of Sacha Inchi oil from Peru (6.2 mg·100 g^-1) (Fanali et al., 2011). According to a report by Valavanidis et al. (2004), among edible oils, extra virgin olive oil has the highest total phenolic content with a range of 17–21 mg GAE·100 g^-1. Soybean oil has mid-range levels of total phenolics at 6–8 mg 100·g^-1. Sunflower oil has even lower levels with a range of 3–4 mg·100 g^-1, and corn oil samples contained less than 1 mg·kg^-1. Sacha Inchi oil from Xishuangbanna appears to have similar levels of total phenolics as soybean oil. Polyphenols are good for the oil's oxidative stability of polyunsaturated fatty acids imparting the characteristic flavor to the oil. Moreover, many polyphenols have beneficial effects on some common diseases including hypertension and atherosclerosis and play roles in the prevention of certain cancers and modification of immune responses (Luaces et al., 2007).

The radical scavenging activity (RSA) of the analyzed methanol extract of Sacha Inchi oil samples was 12.82±0.60 and 8.25±0.50 mM gallic acid·kg^-1 Sacha Inchi oil for DPPH and ABTS methods, respectively. For total Sacha Inchi oil, the total antioxidant activity (TAC) was 133.42±3.63 mM gallic acid·kg^-1 Sacha Inchi oil measured by the DPPH method. Results from the DPPH assay in total oil were much higher than those in the hydrophilic fraction indicating that the main component responsible for antioxidant activity is within the hypdophobic fraction. The high content of tocopherols could play a major antioxidant role in Sacha Inchi oil, or in synergy with phenolics (Espin et al., 2000). The exact roles of tocopherols and phenolics in the antioxidative activity of the oil are worth further study. In a similar study of the total antioxidant capacity of Greek olive oil, Minioti et al. (2010) showed that antioxidant capacities among 50 samples in the hydrophilic fraction were between 1.29–9.95 mM (DPPH) and 5.42–22.5 mM (ABTS) gallic acid/kg oil. While in total oil, the antioxidant activity ranged from 77–177 mM·kg^-1 as assessed by the DPPH method. The mean antioxidant capacity values for DPPH and ABTS in the hydrophilic fraction are 5.4±2.2 and 13.1±4.2 mM gallic acid·kg^-1 oil while values in DPPH in total olive oil is 116.4 Mm gallic acid·kg^-1 oil. The RSA of Sacha Inchi oil appears to be higher than in most Greek virgin olive oils. For Sacha Inchi oil, the total phenolics together with tocopherols not only act as antioxidants but also bring more health benefits beyond essential fatty acids. The health benefits associated with this high radical scavenging activity are worth further investigation.

The storage stability of Sacha Inchi oil was evaluated by acid value (AV) and peroxide value (PV) analyses during 12 months of storage. The formation of free fatty acids is an important measure of food rancidity, which leads to a number of effects such as hydrolyzed and oxidation rancidity. Acid value is a principal indicator of lipid rancidity, the value of the oil samples ranged from 1.790±0.004 to 2.704±0.110 mg KOH·g^-1 oil during the entire storage period (Figure 1). In addition, acid value increased faster at a higher temperature, but no significant difference in AV was observed between colorless and green bottles. This result shows that temperature is a major parameter leading to the rancidity of Sacha Inchi oil.

Peroxide value (PV) is the most common measurement of lipid oxidation. As a measurement of hydroperoxide formation, it has been recognized as a useful index for early stages of oxidation. Usually, in a typical test, PV reaches a maximum value during the process of oxidation followed by a decrease. In Sacha Inchi oil, PV increased steadily in the first 10 months and then plateaued (Figure 2). Compared with oil samples stored in colorless bottles, the samples stored in green bottles showed higher oxidative stability, and the oil samples stored at 4 °C showed the highest oxidative stability. Notably, PV reached a value of 15 meq O₂·kg^-1 oil at about 9 months (Figure 1A). It is generally accepted that PV of oil is not more than 15 meq O₂·kg^-1 oil for safe consumption (FAO, 1999). The PV analyses suggest that Sacha Inchi oil from Xishuangbanna has a stable shelf life of about 9 months in green bottles at 4 °C. The oil remaining stable may be attributed to high levels of tocopherols and phenolics as natural antioxidants. This is the first time the oxidative stability of Sacha Inchi oil has been evaluated and the results may give rise to further research and additional applications of the oil.