Argan fruit was collected in Tiout (Taroudant County) in August, 2014 and prepared following the methodology used by the local women’s cooperative. The fruit was air-dried for 3 weeks then mechanically de-hulled (SMIR Technotour, Agadir, Morocco). Manually opened argan nuts contained the kernels that were ground using an endless press (IBG Monforts Oekotec GmbH, Mönchengladbach, Germany). An aliquot of collected argan oil was immediately analyzed. Remaining oil was stored at 60 °C and analyses were repeated after every week (acidity, peroxide value, and p-anisidine index) or after 3, 6, 10, and 12 weeks of storage (tocopherol, sterol, and fatty acid contents).

Cactus fruit was picked in Sidi Ifni in July, 2014. The fruit was manually peeled and cactus seeds were collected, air-dried and finally ground using the same type of endless press as that used for argan oil.

For the study, oil samples were stored in a Memmert UF110 plus oven (Memmert GmbH, Schwabach, Germany) equipped with a Kimo KTT310-RF thermostat at a constant temperature of 60 ± 1 °C.

An aliquot of each oil was analyzed immediately after oil extraction. The remaining oil was stored at 60 °C and an amount necessary for the analysis was subtracted every three weeks. Analyses were performed over a 12 week-period.

The chemical and physical parameters (acidity, peroxide index, p-anisidine value, and fatty acid content) were analyzed, in triplicate, following the analytical methods described in Regulations EC 2568/91 (Commission Regulation, 1991).

Fatty acid composition was determined as their corresponding methyl esters by gas chromatography on a CPWax 52CB column (30 m x 0.25 mm i.d., 0.25 µm film thickness) using He (flow rate 1 mL/mn) as carrier gas. Oven, injector, and detector temperatures were set at 170, 200, and 230 °C; respectively. Injected quantity was 1 µL for each analysis (Gharby et al., 2011).

Sterol composition was determined after trimethylsilylation of the crude sterol fraction using a Varian 3800 instrument equipped with a VF-1 ms column (30 m x 0.25 mm i.d., 0.25 µm film thickness) and using Helium (flow rate 1.6 mL/min) as carrier gas. Column temperature was isothermal at 270 °C, the injector and detector temperature was 300 °C. Injected quantity was 1μL for each analysis (Gharby et al., 2011).

On the basis of the AOCS Official method Ce 8-89 (Gharby et al., 2011), tocopherol content was determined by HPLC using Shimadzu instruments equipped with a C18-Varian column (25 cm x 4 mm). Detection was performed using a fluorescence detector (excitation wavelength 290 nm, detection wavelength 330 nm). The eluent used was a 99:1 isooctane/ isopropanol (V/V) mixture, at a flow rate of 1.2 mL/min.

Values reported are the mean values ± SE of 3 replicates. The significance level was set at P = 0.05. Separation of mean values was performed by Tukey’s test at the 0.05 significance level.

For both oils, initial fatty acid content was found in the range of published values for cactus seed oil from Morocco (Zine et al., 2013; Taoufik et al., 2015), or of geographically close origin (Monia et al., 2005). Accelerated aging did not induce any significant modification in the fatty acid content or distribution and, even after 12 weeks at 60 °C, the linoleic acid content in cactus seed oil was found to be similar to that of the freshly prepared oil (Table 1).

Sterols were the second class of compounds investigated. Cactus seed and argan oils are both rich in phytosterols but they only share Δ⁷-avenasterol as common sterols. In addition to Δ⁷-avenasterol, the sterols in cactus seed oil are campesterol, stigmasterol, β-sitosterol, Δ⁵-avenasterol, and Δ⁷-stigmasterol (El Mannoubi et al., 2009). In addition, argan oil sterols are schottenol, spinasterol, and stigmasta-8,22-diene (Charrouf and Guillaume, 1999). As observed for fatty acids, no significant variations in sterol content was observed for both oils during 12 weeks of accelerated aging (data not shown). In cactus seed oil, the content of β-sitosterol, as the main sterol, remained constant at over 78% and that of campesterol at around 10%.

Tocopherols, whose anti-oxidant properties are presented as important for the oil’s cosmetic properties, were also investigated (Guillaume and Charrouf, 2011). Whereas argan oil contains α-, β-, γ-, and δ-tocopherols (β- being minoritary) (Charrouf and Guillaume, 1999), cactus seed oil does not contain β-tocopherol (El Mannoubi et al., 2009). Therefore, only the changes in those three tocopherols that the two oils have in common were investigated over 12 weeks of accelerated aging conditions.

Tocopherol distribution in freshly prepared cactus seed oil is different from that of argan oil. In argan oil, α- and δ-tocopherol are present in an amount which is 10 times greater than in cactus seed oil (Table 2). The γ-tocopherol content is similar in both oils (Table 2). The concentration in α-tocopherol remained stable in cactus seed oil for 6 weeks, after which it began to significantly decrease, suggesting the occurrence of the oxidative processes. In argan oil, variations in α-tocopherol were not significant over the 12 weeks. The content in γ-tocopherol in cactus seed oil turned out to be more stable over the study period and only began to significantly decrease after the ninth week. At week-12, the γ-tocopherol content in cactus seed oil that had constantly been similar or slightly higher than that of argan oil, became inferior to that of argan oil, likely indicating the intense destruction of γ-tocopherol by oxidizing species. Under the same aging conditions, the γ-tocopherol content in argan oil decreased sooner since a significant variation in γ-tocopherol content was observable after week 3. However, the loss in γ-tocopherol remained moderate throughout the whole study. Initially low δ-tocopherol content remained stable in cactus seed oil over the twelve weeks of study; whereas it had started to significantly decrease in argan oil after week-6. The total amount of tocopherol in both oils became significantly different after 9 weeks of storage.

These differences in tocopherol content variations between the two oils clearly point out that anti-oxidative processes occurring during aging in cactus seed and argan oils are either different or that responses to oxidative processes are different. The differences may reflect the formation of different oxidizing species resulting mainly from oleic (argan oil) or linoleic (cactus oil) acids. They may also reflect a different oxidative behavior of synergestic associations of the fatty acids with other anti-oxidant molecules, possibly the sterols and/or phospholipids (Gharby et al., 2012b; Zaanoun et al., 2014). Despite the observed different reactivity of cactus seed and argan oil tocopherols in accelerated oxidative conditions, it is noteworthy that the initial cactus oil total tocopherol content was 81% of that of argan oil. After 12 weeks of accelerated aging, and different oxidative processes, the total tocopherol content in cactus seed oil was still 81% of that of argan oil, but on a lower total level.

Once a variation in the content of the anti-oxidant tocopherols was identified, it was decided to further evaluate the shelf-life of argan and cactus seed oils from Morocco, by examining the variations in two other key oxidation markers likely to be modified during prolonged storage at 60 °C. Hence, the peroxide and p-anisidine values were determined. The variations in oil acidity, a parameter important for cosmetics, were also examined.

The composition of cactus seed oil depends on its geographical origin (Ciriminna et al., 2017), as does its acidity for which large variations have also been reported. Acidity values as low as 0.56% for oleic acid (Zine et al., 2013) and as high as 5.08% for oleic acid (De Wit et al., 2016) have been reported for cactus oils of various geographical origins. We found an acidity of 1.15% for oleic acid in our cactus seed oil sample (Figure 1). Such acidity, which is twice that found in a previous study on a different batch of Moroccan cactus seed oil (Zine et al., 2013), clearly indicates that the geographical origin is neither the only factor influencing cactus seed oil acidity nor the most influential factor. The maturity of the fruit and/or the seed water-content, two conditions that inevitably superimpose, are two parameters that also likely dramatically influence the acidity of cactus oil. If this later parameter has already been suggested (De Wit et al., 2016) and could be evaluated by a moisture measurement, the huge number of seeds in a cactus fruit and their different degrees of maturity, unfortunately makes a global evaluation of cactus seeds difficult. The presence of enzymes or of unidentified residues formed after oil extraction has also been suggested to explain the high acidity of cactus seed oil (De Wit et al., 2016).

After storage at 60 °C, the calculated acidity of the cactus oil sample increased almost linearly until week 9 to reach 2.87% for oleic acid (Figure 1). The slope of the straight line, which describes hydrolysis, followed a kinetic second order. Such a value corresponds to a hydrolysis rate estimated to be 2x10^-3 mmol of triglyceride/week. The initial acidity value in argan oil was 0.3% for oleic acid, which is very low, far lower than cactus oil. The trend of hydrolysis followed a linear process, such as the cactus oil, during storage at 60 °C for the 12 weeks of our study but its slope was twice as low as that of cactus seed oil (Figure 1). Accordingly, the hydrolysis of argan oil during the storage period occurred at a rate estimated to be 0.75x10^-3 mmol of triglyceride/week, a rate almost thrice lower than that of cactus oil. Hence, cactus seed oil appears to be much more sensitive to hydrolysis than argan oil.

Accelerated aging is a good measure to evaluate lipid peroxidation (Stewart and Bewley, 1980). Therefore, we evaluated the oxidation degree of cactus seed oil by examining its peroxide value (Figure 2). Cactus seed oil had an initial peroxide value of 4.59 meq O₂/kg. If much lower peroxide values have been previously reported (Matthäus and Özcan, 2011; Özcan and Al Juhaimi, 2011) for cactus seed oil, this value is similar to that already reported for Moroccan cactus oil (Zine et al., 2013) and much lower than that determined for South African cactus oil for which a peroxide value as high as 33.6 meq O₂/kg has been observed (De Wit et al., 2016). The first weeks of storage of cactus seed oil at 60 °C could be identified as the oxidative propagation phase and the cactus oil peroxide value for cactus seed oil reached a maximum value of 9.43 meq O₂/kg at week 3. After that, some peroxides began to be broken into secondary oxidation products as attested by the decrease for 2 weeks (from week 3 to week 5) in the peroxide value. After 6 weeks of storage, the kinetics of the peroxidation became faster than that of the secondary oxidation formation and large amounts of peroxides were again detected. After 12 weeks, the peroxide value reached a maximum value of 39.41 meq O₂/kg. At that moment, the γ-tocopherol content underwent a large decrease, likely attesting to its destruction by the massive formation of peroxides. Interestingly, we observed that the α-tocopherol content in cactus oil decreased after 6 weeks. This also suggests an active and early involvement of α-tocopherol in the prevention of peroxide formation. Tocopherol intervention could be sequential, α-tocopherol would participate in cactus seed oil preservation in a first step and γ-tocopherol in a second phase.

In argan oil, the propagation phase lasted 6 weeks (twice as long as that of cactus seed oil) and additional peroxides began to significantly appear after 10 weeks. After 12 weeks, the peroxide value of argan oil reached 33.6 meq O₂/kg, a value almost 15% lower than that of cactus seed oil. Consequently, peroxide formation is much faster in cactus oil than in argan oil.

To get a better picture of the secondary oxidation product formation, the p-anisidine value for our oils as a function of the time of storage were determined. For cactus oil, a strong increase in the p-anisidine value was observed after 2 weeks confirming the decrease in peroxide value observed after 3 weeks. The p-anisidine value continued to increase regularly, attesting to the constant formation of secondary oxidation products during storage at 60 °C. In argan oil, secondary oxidation products were formed after 7 weeks of storage, again confirming the plateau observed in peroxide formation after 8 weeks.