Fatty acids, 4-desmethylsterols, and triterpene alcohols from Tunisian lentisc (Pistacia lentiscus) fruits
Abstract
A comparative study was performed to determine the fatty acid, 4-desmethylsterol and triterpenic alcohols compositions of three different Tunisian populations of Pistacia lentiscus fruit Rimel (RM), Korbous (KO), and Tebaba (TB). Fruits are rich in lipids, which varied from 39.37% (KO) to 42.48% (TB) on a dry weight basis. Qualitatively, fatty acid, sterol, mono- and dimethylsterol composition is identical for all populations. Oleic acid was the major fatty acid for all samples, accounting from 40.49% in TB population to 50.72% in RM population followed by the palmitic and linoleic acids. Other fatty acids are present at lower levels. Total sterol amount varied from 109.72 mg/100 g of oil (KO) to 434.26 mg/100 g of oil (RM) with an average of 248.74 mg/100 g of oil. The major 4-desmethylsterol component in all studied Tunisian populations of P. lentiscus oil was ß-sitosterol followed by campesterol in TB and KO, and by stigmasterol in RM. The amount of total triterpenic alcohols varied from 42.39 mg/100 g of oil in RM population to 70.41 mg/100 g oil in TB population. The quantitative difference in the fatty acids and 4-desmethylsterols of the different populations studied could be due to the effect of geographic region and soil type.
Introduction
Pistacia lentiscus L. (Anacardiaceae) is a common evergreen dioecious shrub occurring in Mediterranean ecosystems in a wide variety of habitats. It is a member of a heterogeneous family of 11 species. In Tunisia, it has a large geographical and bioclimatic distribution, extending from the humid to the arid areas 1, 2. Fruits of P. lentiscus give an edible oil which is rich in unsaturated fatty acids such as oleic and linoleic acid 3. Oils with a high proportion of oleic acid are more stable than others and contribute to reduction in cardiovascular diseases in humans 4. On the other hand, linoleic acid is an essential fatty acid; the high content of linoleic acid makes the oil very important to the industries. The linoleic acid can be used in protective coatings, plastics, urethane derivatives, surfactant, dispersants, cosmetics, lubricants, and varieties of synthetic intermediate, stabilizers in plastic formulations and in the preparations of other long chain compounds. The high content of linoleic acid in seed oil is very important to the production of oleo-chemicals 5. Therefore, the fatty acids profile is a main determinant of the oil quality. In Tunisia, the oil of lentisc is used by the population in traditional medicine in many ways, as an anti-diarrheal, anti-inflammatory, and in asthma treatment. In addition to fatty acids, vegetable oils contain phytosterols, which are the major components of the unsaponifiable matter. In vegetable matrices, these lipid compounds are present in three main classes names 4-desmethylsterols (sterols), 4-monomethylsterols, and 4,4′-dimethylsterols (triterpene alcohols) 6 and attract the interest of food chemists because they are of great importance for food labeling and nutritional purposes. They are also characteristic of the genuineness of vegetable oils 7. Phytosterols play major roles in several areas namely in pharmaceuticals (production of therapeutic steroids), nutrition and cosmetics (creams, lipstick). Additionally, it has been recognized as an important component of healthy diets and diets designed to reduce hypercholesterolemia and have anti-inflammatory and anti-carcinogenic effects 8-10. Their primary mechanism of lowering blood cholesterol levels is the competitive replacement of cholesterol in bile salt micelles, resulting in reduced absorption of unesterified cholesterol from the small intestine 11. The most abundant plant sterols are β-sitosterol, campesterol, and stigmasterol and the daily dietary intake of plant sterols ranges between 167 and 437 mg 12.
Owing to phytosterols' important health effects, nowadays the objective of food industry is to isolate plant matrices rich in these healthy compounds.
There is a small number of reports available in the literature studying the fatty acid composition of P. lentiscus fruit 13 but there is no study done on its 4-desmethyl-, 4,4′-dimethyl- and 4-monomethylsterols compositions. Hence, the aim of the present study is to identify and compare differences among three Tunisian P. lentiscus populations Korbous (KO), Tebaba (TB), and Rimel (RM) by determining their sterol and fatty acid profiles.
Materials and methods
Samples
P. lentiscus L. (Lentisc) fruits were collected from plants growing wild in the three Tunisian regions of Korbous (KO), Tebaba (TB), and Rimel (RM) in January 2009. From each region, fruits were collected from 10 trees, mixed and placed in an oven at 60°C and were weighed each day until the difference between successive weight is less than or equal to 5 mg. Table 1 shows the different geographic and climatic data and the soil type of the three sampling stations.
| Sampling location | Code | Region | Latitude | Longitude | Altitude over the sea surface | Climate | Soil type | Other climatic data recorded during 2008 | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Maximum temperature | Average annual rainfall | Humidity | |||||||||
| Summer | Winter | ||||||||||
| Korbous | KO | Nabeul (Northeast) | 36° 49'N | 10° 34′ E | 419 m | Subhumide(warm winter variant) | Sandy clay | 26–33 | 15–22 | 211 mm | 70–83 |
| Tebaba | TB | Beja (Northwest) | 37° 00'N | 9° 06′ E | 250 m | Humide(mild and temperate winter variant) | Sandy clay | 26–37 | 15–20 | 490 mm | 50–80 |
| Rimel | RM | Bizerte (North) | 37° 14′N | 9° 54′ E | 6 m | Subhumide(mild winter variant) | Send dune limestone | 26–33 | 16–21 | 423 mm | 75–90 |
Reagent and standards
Petroleum ether was acquired from Fisher Scientific SA (Loughborough, Spain). Methanol and n-hexane 95%, solvents of HPLC grade, were purchased from Panreac Quimica SA (Barcelona, Spain). Sodium methylate and Acetyl chloride from Alfa Aesar GmbH & Co (Germany). Ethanol was purchased from Scientific Limited (Northampton, UK). TLC silica plates (silica gel 60G F254, 20/20 cm, 0.25 mm thickness) and Potassium hydroxide pellets were obtained from Merck (Darmstadt, Germany) and the Fatty acid and sterol standards were acquired from Sigma–Aldrich (Paris, France).
Lipid extraction
The dried samples were reduced in powder. Oil was extracted by a soxhlet extractor for 6 h using Petroleum ether as solvent. The solvent was evaporated under reduced pressure, using a rotary evaporator at 50°C. The lipid content was determined according to the AOCS 14 method. The oil was dried by using a stream of nitrogen and stored at −20°C.
GC analysis of fatty acids
After their methylation according to the method reported by Piombo et al. 15, FAMEs were analyzed by GC using a Agilent 6890 chromatograph (Bios Analytique, France) series with a Innowax capillary column (SGE, Courtaboeuf, France) having the following characteristics: length, 30 m; internal diameter, 0.32 mm; film thickness 0.25 µm. The carrier gas was helium, at a flow through the column of 2 mL/min; split of ratio was 1:80; the injector temperature was maintained at 250°C and the flame-ionization detector (FID) at 270°C. The oven temperature was heated from 185 to 225°C at 4°C/min and held at 225°C for 20 min. The fatty acids were identified by comparison of their retention times with those of standards.
Analysis of triterpene alcohols
Saponification
Unsaponifiable lipids were determined by saponifying 5 g of lipid extracts with 50 mL ethanolic KOH 12% (w/v) mixed with both 200 µL of 5-α-cholestanol 0.2% (w/v) solution and 100 µL of 1-eicosanol 0.5% (w/v) (internal standards for 4-desmethylsterols and triterpene alcohols) and heating at 60°C for 1.30 h. After cooling, 50 mL of H2O was added and the unsaponifiable matter was extracted four times with 50 mL petroleum ether. The combined ether extract was washed with 50 mL of EtOH–H2O (1:1). The ether extracted was dried over anhydrous Na2SO4 and evaporated. The dry residues were dissolved in chloroform for TLC analysis.
Thin-layer chromatography
The unsaponifiable matter was separated into subfractions on preparative silica gel thin-layer plates (silica gel 60 G F254), using 1-dimensional TLC with hexane-Et2O (65:35 by volume) as the developing solvent. The unsaponifiable (4 mg in l00 µL CHCl3) was applied on the silica gel plates in 3 cm bands. To correctly identify the sterols, and triterpene alcohols bands, reference samples of purified 5-α-cholestanol and 1-eicosanol were applied on the left side of the TLC plates. After development the plate was sprayed with 2′,7′-dichlorofluorescein and viewed under UV light. On the basis of the reference spot, the sterols and alcohols bands were identified. The bands corresponding to 4-desmethylsterols and triterpene alcohols were scraped off separately and was extracted three times with CHCl3–Et2O (1:1), filtered to remove the residual silica, dried in a rotary evaporator and stored at 10°C for further analysis.
GC–MS analysis
Each phytosterol fraction was silylated with 100 µL of Bis (trimethylsilyl) Trifluoro-acetamide +1% Trimethylchlorosilane agent at 60°C for 30 min. The derivatized sterols and alcohols fractions were immediately injected separately into a GC (Hewlett Packard 6890) coupled to a HP 5973 mass selective detector (Agilent technologies), set to scan from 20 to 550 m/z. The system was fitted with a capillary HP-5 column (5% phenyl methyl siloxane, 30 m × 0.25 mm, 0.25 µm film thickness) and Helium was used as the carrier gas at 1.5 mL/min. GC–MS operating temperatures were as follows: injector 250°C, detector 310°C and oven temperature was programmed from 120 to 320°C at 10°C/min and held at 320°C for 10 min. The ionization energy was 70 eV. Manual injection of 1 µL of the solution of sterols or triterpene alcohols was performed in the split mode at a 1:50 split ratio. The sterols were identified by comparing the relative retention times to a standards and mass spectra with those previously published 16. The peaks were also confirmed with Wiley 275.L Mass Spectral Library. Compounds were quantified by directly comparing their total ion chromatogram peak areas with that of an internal standard (5-α-cholestanol or eicosanol).
Results and discussion
Lentisc oil content
Oil content of fully ripened P. lentiscus fruit from different Tunisian populations varied from 39.37 ± 0.16% (KO) to 42.48 ± 0.01% (TB) on a dry weight basis (Table 2). Our result was similar to that found by Charef et al. 13, about 38.8% in Algerian P. lentiscus fruit. The percentage of oil of two related species of the Anacardiaceae family Pistacia atlantica and Pistacia terebinthus varies respectively between 45% 17 and 41, 2% 18. As a result of the high oil content, fruit of lentisc seem to be an interesting source for the production of vegetable oil.
| Origins | Oil content (% d/w) | Total unsaponifiable. (%) |
|---|---|---|
| Korbous | 39.37 ± 0.16 | 0.68 ± 0.04 |
| Tebaba | 42.48 ± 0.01 | 0.43 ± 0.14 |
| Rimel | 42.54 ± 1.74 | 0.61 ± 0.09 |
| Mean ± SD | 41.46 ± 1.81 | 0.57 ± 0.13 |
- Each value is mean ± SD of a triplicate analysis performed on different samples % d/w; percentage of dried weight.
Total unsaponifiable content
The unsaponifiable matter of the three Tunisian samples studied ranged from 0.43 ± 0.14% (TB) to 0.61 ± 0.09% (RM), and 0.68 ± 0.04% (KO) of total lipid with an overall mean of 0.57 ± 0.13% (Table 1). Compared to other oil crops, these values are low. Indeed, the rate of unsaponifiable varies with species and provenance. In Mediterranean, Pinus pinea the unsaponifiable value is between 1.32 and 2.09% 19 while it can reach more than 10% in certain plants 20. However, these values are a source of information on the characterization and authentication of the lentisc fruit. In fact, the unsaponifiable fraction contains a variety of bioactive substances, which include sterols, hydrocarbons, tocopherols, terpenes, phenols, and others. These minor compounds are more useful in authenticity tests 21, 22.
The addition of unsaponifiable matter isolated from wheat germ, corn or olive oil was found to retard oxidation in vegetable oils and model lipids subjected to heating 23.
Fatty acid composition
Only nine different fatty acids were identified (Table 3). Qualitatively, fatty acid composition is identical for all populations. Oleic acid was the major fatty acid for all samples, accounting from 40.49% in TB population to 50.72% in RM population followed by the palmitic and linoleic acids with an overall mean respectively of 24.39 and 24.11%. Stearic acid was detected in lower amount in the fruit oil of P. lentiscus only 1.13% in RM population to 1.43% in TB population. However, there were negligible levels of fatty acids detected. For example, linolenic (0.74%), gadoleic (0.11%), and arachidic acid which is detected as trace. Also, our results agree well with the data recorded by Charef at al. 13.
| Fatty acid | Population | Mean ± SD (n = 3) | ||
|---|---|---|---|---|
| KO | TB | RM | ||
| C16:0 | 23.23 | 26.78 | 23.16 | 24.39 ± 2.07 |
| C16:1 | 1.03 | 2.69 | 1.28 | 1.66 ± 0.90 |
| C18:0 | 1.30 | 1.43 | 1.13 | 1.28 ± 0.15 |
| C18:1n9 | 48.75 | 40.49 | 50.72 | 46.65 ± 5,43 |
| C18:1n7 | 0.62 | 1.25 | 0.99 | 0.95 ± 0,32 |
| C18:2n6 | 24.07 | 26.52 | 21.75 | 24.11 ± 2.39 |
| C18:3n3 | 0.65 | 0.79 | 0.78 | 0.74 ± 0.08 |
| C20:0 | 0.13 | tr | 0.06 | 0.09 ± 0.05 |
| C20:1 | 0.13 | tr | 0.09 | 0.11 ± 0.03 |
| Sum of SFA | 24.66 | 28.21 | 24.35 | 25.74 ± 2.14 |
| Sum of MUFA | 50.53 | 44.43 | 53.08 | 49.34 ± 4.44 |
| Sum of PUFA | 24.72 | 27.31 | 22.53 | 24.85 ± 2.39 |
| Sum of UFA | 75.25 | 71.74 | 75.61 | 74.2 ± 2.14 |
- KO, Korbous; TB, Tebaba; RM, Rimel; tr, trace.
- Values are mean of two repetitions.
It appears that lentisc oil is characterized by the predominance of MUFA 44.43% for the TB population and more than 50% for the KO and RM populations (Table 3). The high MUFA diet reduced total- and LDL-cholesterol concentrations and total cholesterol/HDL-cholesterol ratio 24. If we compare the results for fatty acid composition of P. lentiscus from different Tunisian populations with those for fatty acid composition of P. atlantica 25, we see that the FA composition is very near to that of P. atlantica oil from Iran and Algeria and P. terebinthus oil from Turkey. Our results show that the oil has a higher content of unsaturated FA (oleic and linoleic 70.76%) and that the oil of P. lentiscus fruit can be classified as an oleo-linoleic vegetable oil.
Sterol composition
Recently, the cosmetic, drug and alimentary/nutraceutical industries have focused attention on low-cost renewable resources, rich in lipid-related compounds such as phytosterols which are an important part of the unsaponifiable matter of vegetable oils. The analysis of the sterols provides rich information about the quality and the identity of the oil investigated, and for the detection of oil and mixtures not recognized by their fatty acids profile 26.
Only four sterols were identified and quantified. The amount of total sterols varied from 109.72 mg/100 g of oil in Korbous population to 434.26 mg/100 g of oil in Rimel population with an average of 248.74 mg/100 g of oil (Table 4). A significant difference was recorded by comparing the sterol content of the lentisc fruit from the three studied provenances. Indeed, Rimel population contains the highest amount of total phytosterol. This difference may be due essentially to geographical conditions, climate and soil. Sterol contents in RM population were higher than that in KO (109.72 mg/100 g), in TB (202.26 mg/100 g) population, and in other plants such as olive oil (150 mg/100 g) and lower than those in crude corn oil (850 mg/100 mg) and rapeseed oil (820 mg/100 mg) 27, 28. The major sterol components in all studied Tunisian populations of P. lentiscus oil was ß-sitosterol. That amount changed depending on the geographical origin, from (99.61 mg/100 g of oil in KO) to (389.50 mg/100 g of oil in RM). Likewise, the effect of provenance is always highlighted for other molecular species of P. lentiscus sterols. Indeed, for KO and TB populations and in descending order, campesterol comes after ß-sitosterol followed by cholesterol in KO and Stigmasterol in TB. The RM population shows a sterol profile quantitatively different from the other studied populations. Cholesterol reaches a value >4% in KO. Many sterol benefits for human health have been identified. ß-sitosterol supplementation inhibits MCF-7 and MDA-MB-231 cell growth and activates Fas signaling in human breast cancer cells 29. This gives the P. lentiscus oil significant nutritional and therapeutic value.
| KO | TB | RM | Mean ± SD | ||||
|---|---|---|---|---|---|---|---|
| Content | % | Content | % | Content | % | ||
| ß-Sitosterol | 99.61 | 90.78 | 186.84 | 92.37 | 389.50 | 89.69 | 225.31 ± 148.73 |
| Campesterol | 5.15 | 4.69 | 9.30 | 4.59 | 7.68 | 1.76 | 7.37 ± 2.09 |
| Cholesterol | 4.96 | 4.52 | tr | tr | 3.97 | 0.91 | 4.46 ± 0.70 |
| Stigmasterol | tr | tr | 6.12 | 3.02 | 33.11 | 7.62 | 19.61 ± 19.08 |
| Total sterols | 109.72 | 100 | 202.26 | 100 | 434.26 | 100 | 248.74 ± 167.19 |
- Values are mean of two repetitions.
- KO, Korbous; TB, Tebaba; RM, Rimel; tr, trace.
Triterpenic alcohols composition
Table 5 shows the 4-monomethylsterols and 4-4 dimethylsterols compounds isolated from three different populations of P. lentiscus fruit oil unsaponifiable matter, identified by GC–MS and quantified by using 1-eicosanol as internal standard. Only six molecular species were investigated in lentisc fruits for the first time: obtusifoliol, butyrospermol, ß-amyrine, cycloartenol, 24-methylene cycloartenol, and citrostadienol. Significant differences were observed in the composition of the studied provenances of lentisc fruit. Tebaba oil was the richest one in terms of all of the 4-mono- and 4,4′-dimethylsterols under study. Indeed, the amount of total triterpenic alcohols varied from 42.39 mg/100 g of oil (180.32 mg/kg of dry weight) in RM population to 70.41 mg/100 g oil (299.1 mg/kg d.w) in TB population. Cycloartenol was the predominant compound, with an average of 37.19 ± 15.24 mg/100 g oil, followed by 24-methylene cycloartenol with an average of 10.71 ± 3.03 mg/100 g of oil and by ß-amyrine. The 4-monomethylsterols as Obtusifoliol, Butyrospermol and Citrostadienol were also detected with respectively an average of 1.12; 1.31 ± 0.22 and 1.88 ± 0.23 mg/100 g of oil. These results are much lower than those found by Herchi 30 and Tlili et al. 31. The remarkable change between the populations studied in the quantitative composition in some triterpene compounds may be due to geographical and climatic effects that differ from one station to another 32. The triterpenic compounds have an important role in the prevention of many diseases. Cycloartenol and 24-methylene cycloartenol, the major 4,4′-dimethylstrol in P. lentiscus fruit oil have an anti-inflammatory effect against many factors inducing inflammation in mice 33.
| Component | KO | TB | RM | Mean ± SD | |||
|---|---|---|---|---|---|---|---|
| Content | % | Content | % | Content | % | ||
| Obtusifoliol | 1.12 | 2.13 | tr | tr | tr | tr | 1.12 |
| Butyrospermol | 1.16 | 2.21 | 1.47 | 2.08 | tr | tr | 1.31 ± 0.22 |
| ß-amyrine | 3.7 | 7.06 | 5.42 | 7.69 | 3.56 | 8.39 | 4.22 ± 1.04 |
| Cycloartenol | 31.19 | 59.52 | 54.53 | 77.44 | 25.87 | 61.02 | 37.19 ± 15.24 |
| 24-Methylene cycloartenol | 13.27 | 25.32 | 7.36 | 10.45 | 11.5 | 27.12 | 10.71 ± 3.03 |
| Citrostadienol | 1.96 | 3.74 | 1.63 | 2.31 | 2.06 | 4.85 | 1.88 ± 0.23 |
| Total of triterpenic alcohols | 52.4 | 100 | 70.41 | 100 | 42.39 | 100 | 55,06 ± 14.2 |
- Values are mean of two repetitions.
- KO, Korbous; TB, Tebaba; RM, Rimel; tr, trace.
Nine fatty acids, four sterols, and six triterpenic alcohols were determined in summer in P. lentiscus fruit oil from different geographical region. Results show that qualitatively, the fatty acid, sterol and alcohol composition is the same for the three studied populations. Contrariwise, their quantities vary according to the provenance. This is may be due to soil and climate factors that differ between the regions. Due to the richness of P. lentiscus oil in bioactive molecules such as phytosterols, omega-6 and MUFA this study consolidates the possibility of incorporating P. lentiscus oils into food, cosmetic and pharmaceutical products.
Acknowledgements
We gratefully thank the members of the Mass spectrometry laboratory (University of Ottawa, Ottawa, Canada) and the members of CIRAD (Montpellier, France).
The authors have declared no conflict of interest.
