Plant Foods for Human Nutrition

, Volume 67, Issue 4, pp 384–392

Chemical Composition and Anti-inflammatory Properties of the Unsaponifiable Fraction from Awara (Astrocaryum vulgare M.) Pulp Oil in Activated J774 Macrophages and in a Mice Model of Endotoxic Shock


  • Emilie Bony
    • Laboratoire de Pharmacologie et Physiopathologie Expérimentales, UMR Qualisud, Faculté de PharmacieUniversité Montpellier I
    • Centre de Coopération Internationale en Recherche Agronomique pour le Développement (CIRAD), Département PERSYSTUMR Qualisud
  • Frédéric Boudard
    • Laboratoire d’Immunologie, Faculté de PharmacieUniversité Montpellier I
  • Emilie Dussossoy
    • Laboratoire de Pharmacologie et Physiopathologie Expérimentales, UMR Qualisud, Faculté de PharmacieUniversité Montpellier I
  • Karine Portet
    • Laboratoire de Pharmacologie et Physiopathologie Expérimentales, UMR Qualisud, Faculté de PharmacieUniversité Montpellier I
  • Pierre Brat
    • Centre de Coopération Internationale en Recherche Agronomique pour le Développement (CIRAD), Département PERSYSTUMR Qualisud
  • Jean Giaimis
    • Laboratoire d’Immunologie, Faculté de PharmacieUniversité Montpellier I
    • Laboratoire de Pharmacologie et Physiopathologie Expérimentales, UMR Qualisud, Faculté de PharmacieUniversité Montpellier I
Original Paper

DOI: 10.1007/s11130-012-0323-z

Cite this article as:
Bony, E., Boudard, F., Dussossoy, E. et al. Plant Foods Hum Nutr (2012) 67: 384. doi:10.1007/s11130-012-0323-z


Awara (Astrocaryum vulgare M.) pulp oil has been shown to possess anti-inflammatory properties in vivo, and contains an unsaponifiable matter rich in bioactive compounds. This study focused on the ethanolic unsaponifiable fraction (EUF) of awara pulp oil. Its chemical composition has been characterized: carotenoid, phytosterol, and tocopherol contents represent 125.7, 152.6, and 6.8 μg/mg of EUF, respectively. We further evaluated this fraction for anti-inflammatory properties in J774 macrophages activated by lipopolysaccharide (LPS) plus interferon (IFN) γ to understand the biological effects of awara pulp oil. EUF strongly decreased nitric oxide (NO), prostaglandin E2, tumour necrosis factor (TNF) α, and interleukin (IL) -6 and -10 production in activated J774 cells. Moreover, it inhibited expression of inducible NO synthase and cyclooxygenases-2 in vitro. The anti-inflammatory properties of EUF were also confirmed in vivo by modulation of TNFα, IL-6 and IL-10 serum concentration in an endotoxic shock model. Pre-treatment with awara oil fraction offers promise as a protective means to lower the production of excessive amounts of pro-inflammatory molecules.


Awara (Astrocaryum vulgare M.)Unsaponifiable fractionCarotenoidsPhytosterolsTocopherolsAnti-inflammatoryJ774 macrophagesEndotoxic shock





Ethanolic unsaponifiable fraction








Nitric oxide synthase


Oxygen radical absorbance capacity




Reactive oxygen species


Sodium nitroprusside


Tumour necrosis factor


Awara (Astrocaryum vulgare M.) belongs to the Arecaceae family and is principally used in nutrition because of the high carotenoid content of its pulp oil. Moreover, therapeutic uses are given in the traditional pharmacopoeia of French Guiana for skin and eye diseases [1]. In a previous study, we have shown that awara pulp oil exhibits anti-inflammatory properties in vivo, both in acute and chronic inflammation models [2]. We hypothesized that these anti-inflammatory properties are mainly or partly due to the unsaponifiable matter that represents almost 1 % of the total lipid fraction. The unsaponifiable fraction is a potential source of bioactive molecules such as phytosterols, carotenoids, or tocopherols and tocotrienols [3]; compounds that are partly responsible for biological properties of vegetable oils [4]. Indeed, the anti-inflammatory properties of olive oil have been attributed in part to minor compounds such as tyrosol, squalene or β-sitosterol [5, 6]. Palm oil also possesses beneficial effects due to its high carotenoid and tocotrienol contents [7]. We therefore decided to specifically study the unsaponifiable fraction of awara pulp oil in order to explain its anti-inflammatory properties.

Many human diseases such as septic shock, rheumatoid arthritis, asthma or cancer, are associated with acute or chronic inflammation [8, 9], and with overproduction of reactive oxygen species (ROS) leading to imbalance between oxidative stress and antioxidant defences [10]. Inflammation is a complex biological response which is regulated by a large number of inflammatory mediators released from cells such as macrophages, lymphocytes, leukocytes, and mast cells. They included arachidonic acid metabolites, reactive oxygen species (ROS), nitric oxide (NO), and cytokines such as interleukin (IL-6 or -10) and tumour necrosis factor (TNF) α. NO is a reactive radical molecule playing an important role in physiological conditions; however, excessive production of NO can result in inflammatory reactions. Indeed, three isoforms of nitric oxide synthases (NOS) have been found, and it is the inducible NOS (iNOS) that is implicated in inflammatory response [11]. Cyclooxygenases (COX) are enzymes responsible for production of important biological mediators called prostanoids (including prostaglandins (PG), prostacyclin, and thromboxane). Two isoforms of COX have been found: COX-1 which is constitutive and COX-2 which is induced during inflammation and responsible for the production of large amounts of pro-inflammatory prostaglandins, such as PGE2 regulating vascular permeability, platelet aggregation, and thrombus formation at the inflammation site [12]. TNFα plays a critical role in the inflammation process and is produced in early stages by activated macrophages. IL-6 is known to be a multifunctional cytokine that regulates immune responses, haematopoiesis, acute phase response, and inflammation. IL-10 is an immunosuppressive and anti-inflammatory cytokine that modulates the function of several adaptive immune-related cells [13]. Therefore, the discovery of plant extracts that can down-regulate the expression or activity of iNOS and COX-2 and also cytokine production, may lead to the identification of potent anti-inflammatory compounds.

In the present study, we first investigated the ethanolic unsaponifiable fraction (EUF) of awara pulp oil for carotenoid, tocopherol, tocotrienol, and phytosterol content. Then, we evaluated the in vitro inhibitory effects of EUF on NO, PGE2, IL-6, TNFα, and IL-10 production induced by LPS/IFNγ in J774 macrophages. We also determined the effects of the EUF on iNOS and COX-2 expression. To confirm potential anti-inflammatory effects of EUF in vitro, we therefore evaluated whether EUF also decreased IL-6, TNFα, and IL-10 serum concentration in a mice endotoxic shock model.

Materials and Methods


All solvents were HPLC or analytical grade, purchased from Carlo Erba. Methyl tert-butyl ether (MTBE), H2PO4, KOH, BSA, LPS (E. coli, 055: B5), Tris HCl, NaCl, Triton X100, DTT, EDTA, EGTA, PMSF, indomethacin, fluorescein, 6-hydroxy-2, 5, 7, 8-tetramethyl-2-carboxilic acid (Trolox), trypan blue, β-mercaptoethanol, bromophenol blue, sulfanilamide, naphthyl-ethylenediamide, stigmasterol, campesterol, β-sitosterol, fucosterol, and 5α-cholestane-3β-ol were from Sigma-Aldrich. 2-2’-azobis (2-amidinopropane) dihydrochloride (AAPH) was purchased from Wako Chemicals. RPMI 1640 with Glutamax®, fetal bovine serum, streptomycin, and penicillin were purchased from Gibco, Invitrogen. Recombinant murine IFNγ was purchased from Genzyme (Tebu-Bio). Na2SO4 anhydrous and KOH were purchased from Carlo Erba; tocopherol standards from VWR International SAS; cycloartenol and β-carotene from Extrasynthese; Tween 20, glycerol and SDS from Euromedex.

Plant Material and Sample Preparation

Awara fruits (Astrocaryum vulgare) were obtained from CIRAD French Guiana (Combi station, Sinnamary, March 2008) and were stored at −20 °C until analysis to avoid enzymatic degradation. Oil extraction and saponification were performed as previously described [2]. The unsaponifiable extract is water insoluble, and for cell culture studies, the samples tested must be soluble in water. This is why we prepared an ethanolic unsaponifiable fraction (EUF) by dissolving 20 mg of the unsaponifiable extract in 500 μl ethanol and filtered on a 0.45 μm filter. Thereafter chemical and biological analyses were performed on the EUF.

Chemical Analysis

Identification and quantification of tocopherols, tocotrienols, phytosterols and carotenoids in EUF were performed as previously described [2] by HPLC-fluorometric detection, GC-MS, and HPLC-DAD-MS, respectively. Analyses were performed in triplicate, and independent measurements and results were expressed as μg tocopherols per mg of EUF±sem, μg 5α-cholestane-3β-ol equivalent per mg of EUF±sem, and as μg β-carotene equivalent per mg of EUF±sem.

COX Inhibitory Assay Kit

The ability of EUF and indomethacin to inhibit in vitro ovine COX-1 and COX-2 was determined using an enzyme immuno assay kit: COX (ovine) Inhibitor Screening Assay® (Cayman Chemical), which measures prostaglandins F2α by SnCl2 reduction of COX-derived prostaglandins FH2 produced in the COX reaction. Briefly, tested compounds were incubated with COX-1 or COX-2 enzyme in buffer in the presence of heme at 37 °C. The substrate of COXs: arachidonic acid was then added and the level of prostaglandins F2α was dosed. In the presence of COX inhibitors, the level of prostaglandins F2α decreased.

NO Scavenging Activity

NO scavenging activity was measured by evaluation of nitrite production after photochemical degradation of sodium nitroprusside (SNP). Briefly, 50 μl of SNP (2.5 μM) were left to incubate with 50 μl of EUF (5–40 μg/ml) for 1 h under a daylight lamp to allow photodegradation of SNP. Nitrite production was then measured by Griess reaction [15].

Macrophage Culture

The J774 macrophage cell line was obtained from the American Type Culture Collection (ATCC, TIB67; Rockville, MD). Cell viability was assessed by MTS/PMS assay as previously described [14].

Cell Treatments

The cells (5 × 105 cells/well) were seeded onto a 24-well culture plate for 1 h in complete RPMI 1640 medium, and treated with various concentrations of EUF (5–40 μg/ml). After 4 h incubation at 37 °C, the cells were stimulated with LPS/IFNγ (1 ng/ml + 10 UI/ml). The supernatants were collected at 6, 24 and 48 h for mediator measurements (stored at −20 °C until use). For Western-blotting, cells were seeded onto a 6-well culture plate at 15 × 105 cells/well in complete RPMI 1640 medium.

Quantification of Nitrite Production

Nitrite concentration in the cultured medium was determined via the Griess reaction as described by Boudard et al. [15]. EUF did not induce nitrite production in unstimulated macrophages (data not shown).

PGE2 and Cytokine Measurements

PGE2 concentrations in the 24 h supernatants were measured using an EIA kit (Cayman, Spi Bio). Cytokine levels (TNFα, IL-6, and IL-10) were measured in supernatants of cell culture, tissues homogenates and in serum using an ELISA kit (eBioscience). Detection limits were 4, 4, and 2 pg/ml for TNFα, IL-6, and IL-10, respectively. EUF did not induce PGE2, TNFα, IL-6, and IL-10 production in unstimulated macrophages (data not shown).

Western-Blot Analysis

Cells were lysed in cold buffer (phosphate-buffered saline 1 % Triton X100 and inhibitor protease cocktail, Roche) for 30 min, scrapped and centrifuged at 13,000 g for 2 min. The supernatants were boiled for 10 min in mixed buffer (125 mM Tris, 10 % glycerol, 2 % SDS, 5 % β-mercaptoethanol, 0.5 % blue bromophenol) and stored at −20 °C until analysis. Proteins were separated on an 8 % SDS-polyacrylamide gel electrophoresis and transferred to polyvinyl difluoride membranes (PVDF, Immobilion-P, Millipore). The membranes were then blocked with 5 % fat-free dry milk in TBST-T pH 7.5 for 30 min, and then incubated for 1 h at room temperature with rabbit polyclonal anti-iNOS antibody (M-19) or anti-COX-2 antibody (M-19) diluted to 1:1,000, and with monoclonal anti-β-actin (C4) diluted to 1:5,000 (Santa Cruz Biotechnology). The membranes were then incubated with peroxidase-conjugated anti-rabbit IgG (1/10,000, Jackson ImmunoResearch), and bands were detected by enhanced chemiluminescence detection (ECL, Amersham). Signal intensities were evaluated by densitometric analysis (ImageJ software) and normalized with β-actine. EUF did not induce COX-2 and iNOS expression in unstimulated macrophages (data not shown).

Endotoxic Shock Model

Animals were obtained from Janvier® (France) and were kept at an ambient temperature of 22 ± 2 °C under a 12 h normal phase light-dark cycle. Food and water were provided ad libitum, and mice were acclimated for one week prior to the start of the experiment. Six-week-old male Balb/c mice were randomly assigned to five groups of eight mice. The control group was IP injected with saline. The LPS 1.5 h and LPS 6 h groups were IP injected with LPS at 20 mg/kg. The EUF 1.5 h and EUF 6 h groups were pre-treated by IP with EUF (25 mg/kg) 2 h before LPS injection. Rats were anaesthetized and blood was collected by cardiac puncture 1.5 h or 6 h after LPS injection, and were kept for 30 min at room temperature for clotting, and centrifuged at 3,000 g for 20 min at 4 °C. Serum was stored at −20 °C until cytokine analysis. ORAC (oxygen radical absorbance capacity) assays were performed as previously described by Dussossoy et al. [14]. The ORAC values were expressed as μM Trolox equivalents per ml of serum.

Statistical Analysis

The data are mean±sem from n independent experiments performed in triplicate or quintuplicate, as mentioned in the captions for each figure, and are normalized to 100 with the positive control when appropriate. The data were statistically analysed by one-way analysis of variance followed by Bonferroni multiple comparison test, or by unpaired Student’s t-test, as appropriate, using the software GraphPad Prism 5. When homogeneity of variance was not respected (Bartlett’s test), a logarithmic transformation to the data or a non-parametric statistic was applied, in this case, the Kruskal-Wallis test. A p-value of less than 0.05 was considered statistically significant.

Animal Care and Use

These experiments were carried out in accordance with the Declaration of Helsinki and with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the US National Institutes of Health. Our laboratory practice was approved by the “Comité d’Ethique pour l’Expérimentation Animale Languedoc-Roussillon”


Phytochemical Analysis

We first measured the content of tocopherols, tocotrienols, phytosterols, and carotenoids in EUF. The HPLC-fluorometric detection analysis only revealed the presence of α-tocopherol in EUF with a content of 6.75 μg ± 0.08 per mg of EUF (mean±sem of three independent determinations).

EUF was injected in HPLC-DAD-MS for tentative identification and quantification of carotenoids (see Electronic Supplementary Material 1). The different isoforms of phytoene, phytofluene, α, β, δ, γ, and ζ-carotene detected in EUF have previously been identified in awara pulp oil [2]. The phytoene, phytofluene and ζ-carotene isomers were quantified at their maximum absorption wavelength (290, 350, and 400 nm, respectively), and expressed as β-carotene equivalent at 450 nm. The β-carotene isomers and 13-cis-β-carotene are the main carotenoids identified, representing almost 20 % of the total carotenoid content, followed by all-trans-β-carotene (6 %) and 15-cis-β-carotene (4 %). Phytoene and phytofluene isomers represent almost 19 and 10 % of the total carotenoid content, respectively. The total amount of carotenoids was 125.7 ± 2.7 μg of β-carotene equivalent/mg of EUF.

GC-MS analyses allowed identification and quantification of phytosterols in EUF (see Electronic Supplementary Material 2), which have previously been detected in awara pulp oil [2]. The total content of phytosterols was 152.6 ± 3.2 μg of 5α-cholestane-3β-ol equivalent/mg of EUF, and β-sitosterol represents almost 37 % of the phytosterols.

Effect of EUF on COX-1 and COX-2 Activities

We evaluated the capacity of EUF to inhibit COX-1 and COX-2 activities by an in vitro test, and compared it to indomethacin (reference NSAIDs). As shown in Fig. 1, EUF induced a strong and dose-dependent inhibition. At 250 μg/ml, EUF decreased COX-1 and COX-2 activities by 41 and 71 %, respectively. In comparison, indomethacin decreased COX-1 (10 μM) and COX-2 (0.5 μM) activities by 37 and 47 %, respectively.
Fig. 1

In vitro effect of EUF and indomethacin on COX-1 and COX-2 activities. Data (n = 6) are mean±sem from two independent experiments performed in triplicate. * p < 0.05 vs. positive control

NO Scavenging Activity

We evaluated the NO scavenging activity of EUF on photodegradation of SNP (see Electronic Supplementary Material 3, n = 6). EUF showed a significant scavenging activity on NO production, with 77 and 83 % nitrite production for 40 and 20 μg/ml respectively, compared to 100 % nitrite production for SNP.

Cell Viability

The toxic effect of EUF on the cells was examined by MTS/PMS assay (see Electronic Supplementary Material 4, n = 10). The results demonstrated that EUF was not cytotoxic for all tested concentrations. This concentration of 40 μg/ml corresponds to a 0.2 % ethanol vehicle and was also tested on MTS assay, and did not show any toxic effect. The observed effects were not due to cell death by EUF or by the vehicle.

Effect of EUF on Nitrite Production and iNOS Expression by LPS/IFNγ-Activated J774 Macrophages

NO production was assessed by measuring the accumulation of nitrites, a stable metabolite of NO, by colorimetric assay based on the Griess reaction, in the 48 h culture supernatant. Compared to unstimulated macrophages, activation by LPS/IFNγ dramatically increased nitrite production (Fig. 2a). Treatment with EUF induced significant inhibition of nitrite production, and this effect was dose-dependent with a maximal inhibition at 40 μg/ml. At this dose the NO scavenging activity of EUF induced a 23 % reduction of nitrite. To further elucidate the effects of EUF, we examined by Western-blot analysis the protein expression of iNOS (Fig. 2b) in LPS/IFNγ-activated J774 macrophages. EUF inhibited iNOS expression in a dose-dependent manner, with 79 % inhibition at 40 μg/ml. The nitrite reduction of EUF can be attributed to both NO scavenging and iNOS inhibition properties.
Fig. 2

Effects of EUF on nitrite production at 48 h (a), iNOS expression at 24 h (b), PGE2 production at 24 h (c), COX-2 expression at 24 h (d), TNFα production at 6 h (e), IL-6 production at 24 h (f), and IL-10 production at 24 h (g), by LPS/IFNγ-activated J774 macrophages. Cells were pre-treated with EUF (40, 20, 10, 5 μg/ml) for 4 h and then activated with LPS (1 ng/ml) + IFNγ (10 UI/ml). For a, c, e, f, and g, data (n = 9) are mean±sem from three independent experiments performed in triplicate and are normalized to 100 with positive control (LPS/IFNγ); * p < 0.05 vs. LPS/IFNγ activation. Mean absolute values for LPS/IFNγ were 37 μM, 4023, 1612, 90301, and 22801 pg/ml for nitrites, PGE2, TNFα, IL-6, and IL-10, respectively. For b and d, the data are a representative profile of three independent experiments

Effect of EUF on PGE2 Production and COX-2 Expression by LPS/IFNγ-Activated J774 Macrophages

PGE2 was quantified in 24 h culture supernatant using EIA kit. Compared to unstimulated macrophages, activation by LPS/IFNγ dramatically increased PGE2 production. EUF induced a dose-dependent inhibition of the PGE2 production, with significant inhibition (58 %) at 40 μg/ml (Fig. 2c). Western-blot analysis of COX-2 (Fig. 2d) in LPS/IFNγ-activated J774 macrophages showed that EUF inhibited COX-2 expression in a dose-dependent manner with a 55 % inhibition at 40 μg/ml.

Effect of EUF on TNFα, IL-6, and IL-10 Production by LPS/IFNγ-Activated J774 Macrophages

Compared to unstimulated macrophages, activation by LPS/IFNγ drastically increased TNFα production at 6 h, and production of IL-6 and IL-10 at 24 h (Fig. 2e, f, g). Treatment with EUF induced a dose-dependent inhibition of TNFα, IL-6 and IL-10 production, with respectively 42, 62 and 90 % significant inhibition at 40 μg/ml.

Effect of EUF on Serum Cytokine Concentration and Antioxidant Capacity in LPS-Induced Endotoxic Shock in Mice

In our experimental model, LPS induced a strong increase in TNFα, IL-6 and IL-10 serum concentration. TNFα concentration (Fig. 3a) significantly increased at 1.5 h (2,785 ± 327 pg/ml) as compared to the control (11 ± 7 pg/ml), and subsequently decreased to reach the control values at 6 h (27 ± 22 pg/ml). IL-6 concentration (Fig. 3b) significantly and gradually increased from 1.5 h (23,911 ± 2,182 pg/ml) to 6 h (44,861 ± 1009 pg/ml) as compared to control (6 ± 3 pg/ml). IL-10 concentration (Fig. 3c) significantly increased at 1.5 h (6,178 ± 517 pg/ml) and decreased at 6 h (2,521 ± 163 pg/ml) (control value: 222 ± 142 pg/ml). Pre-treatment by EUF (25 mg/kg, IP) 2 h before LPS injection strongly and significantly decreased TNFα serum concentration, with 46 % inhibition at 1.5 h (1,516 ± 257 pg/ml). At 6 h after LPS injection, IL-6 serum concentration decreased with 66 % inhibition (15,214 ± 1,016 pg/ml), and IL-10 with 45 % inhibition (1,386 ± 169 pg/ml).
Fig. 3

Effects of EUF on serum cytokine concentration and antioxidant capacity by LPS-induced endotoxic shock in mice. Mice were pre-treated with EUF (25 mg/kg IP) 2 h before LPS injection (20 mg/kg IP). TNFα (a), IL-6 (b), and IL-10 (c) were measured using ELISA in serum at 1.5 h and 6 h after LPS injection. d Antioxidant capacities of serum were measured using the ORAC method at 1.5 h and 6 h after LPS injection. Values are mean±sem (n = 8); * and #p < 0.05 vs. LPS 1.5 h and LPS 6 h, respectively

Compared to the control (9,222 ± 456 μM Trolox eq.), LPS-treated mice also exhibited an increase in antioxidant capacity at 1.5 h (10,975 ± 213 μM Trolox eq.) and a subsequent decrease at 6 h (9,976 ± 287 μM Trolox eq.) (Fig. 3d). EUF significantly inhibited the antioxidant capacity induced by LPS, decreasing the value to the control level at 1.5 h (9,171 ± 358 μM Trolox eq.), and reducing it below the control level at 6 h (8,435 ± 367 μM Trolox eq.).


A diet rich in fruits and vegetables providing phytonutrients such as carotenoids, phytosterols, tocopherols, tocotrienols, and polyphenols, has been associated with the prevention of several diseases [10]. Indeed, carotenoids are natural pigments which possess antioxidant properties and prevent inflammatory diseases such as cardiovascular disease and rheumatoid arthritis [16]. Phytosterols have shown anti-inflammatory properties in vivo by decreasing edema induced by carrageenan or 12-0-Tetradecanoylphorbol acetate [17, 18], and also by decreasing pro-inflammatory cytokines production (TNFα, IL-6) and increasing anti-inflammatory (IL-10) production in Apo E-KO mice [19]. Tocopherols which are considered as the most effective lipid soluble antioxidants also possess anti-inflammatory properties in vivo and in vitro [20].

In order to explain the anti-inflammatory properties of awara pulp oil previously described [2], we hypothesized that these effects are mainly or partly due to the unsaponifiable matter. Thus, we first decided to characterize EUF for carotenoid, phytosterol, and tocopherol contents. Carotenoids and phytosterols identified in EUF were similar to awara pulp oil composition, representing 125.7 and 152.6 μg/mg of EUF, respectively; α-tocopherol represented only 6.75 μg/mg of EUF. The carotenoids identified were cyclic and acyclic carotenes, with β-carotene isomers as major compounds, and β-sitosterol as the major phytosterol. The main differences with awara pulp oil were observed in β-carotene and arundoin content due to their low ethanolic solubility.

To assess the in vitro anti-inflammatory effect of EUF, we first investigated the ability of EUF to inhibit COXs activities in vitro. EUF exhibited a non-specific inhibition of COXs activity, which is a common mechanism of non-steroidal anti-inflammatory drugs (NSAIDs), and exhibited a more potent inhibition for COX-2.

Then, we used murine J774 macrophage cell line activated by LPS/IFNγ association. In this model, EUF strongly inhibited nitrite production, which is due to both inhibition of iNOS expression and direct NO scavenging activity of EUF. Likewise, it strongly inhibited PGE2 production in activated macrophages by inhibition of both activity and expression of COX-2 enzyme. These effects may be explained by anti-inflammatory properties of compounds identified in EUF. Indeed, β-sitosterol in RAW 264.7 macrophages stimulated by PMA [5] and β-carotene in LPS-stimulated macrophages [21] possess anti-inflammatory properties by decreasing PGE2 and NO production via inhibition of iNOS and COX-2 expression. Phytoene and phytofluene, carotenoid precursors, are known for their anti-inflammatory properties in human dermal fibroblasts treated with UV radiation [22]. Tocopherols inhibit PGE2 production and COX-2 activity, but did not affect COX-2 protein expression in LPS-stimulated RAW 264.7 macrophages [23]. Then, we showed that EUF inhibited pro-inflammatory (IL-6 and TNFα) cytokine production in LPS/IFNγ-activated J774 macrophages. Experimental studies have also shown that β-carotene [21] and β-tocopherol [24] possess anti-inflammatory properties in vitro by inhibiting cytokine production (TNFα, IL-1β) or mRNA expression (IL-1β and IL-6). Unfortunately, EUF also decreased the anti-inflammatory IL-10 production in LPS/IFNγ-activated J774 cells. The cytokines network is so complicated that it has been difficult to fully understand until now. TNFα plays a critical role in the inflammation process and is secreted in the first stage, IL-6 in the second stage, and IL-10 in the last stage. Taken together, EUF inhibited a wide range of inflammatory mediators and cytokines, which suggests a general inhibitory effect on macrophage activation. NF-κB, a dimeric transcription factor, is a key molecule in the inflammatory pathway. It is a critical factor for expression of various inflammatory cytokines (TNFα, IL-6, -10) and expression of iNOS and COX-2 [25]. NF-κB activation has been already inhibited in vitro by β-sitosterol [5], cycloartenol [26], tocopherol [27] and β-carotene [21]. Thus, the general inhibitory effect of EUF on macrophage activation may be explained by the inhibition of NF-κB activation. EUF contains lower concentrations of micronutrients compared to the active dose given in the literature of compounds tested alone; this can be explained by a synergic action of identified molecules. Indeed, the highest concentration (40 μg/ml) tested in our model can be expressed as 9.4 μM of total carotenoids (β-carotene equivalent), 14.7 μM of total phytosterols (β-sitosterol equivalent) and 2.5 nM total tocopherols (α-tocopherol equivalent), while concentrations tested with a similar activity in the literature are around 50 μM for β-carotene [21], α-tocopherol [24] and β-sitosterol [5].

To confirm the in vitro anti-inflammatory properties of EUF, we thereafter used an in vivo model of endotoxic shock. Septic shock is a serious clinical problem with high mortality caused in most cases by Gram-negative bacterial endotoxins such as LPS, which is a potent activator of the immune system. Septic shock has been described to result from a multitude of systemic and cellular processes and the overproduction of inflammatory mediators such as TNFα, IL-6 and IL-10 leading to multiple organ failure [28]. TNFα and IL-6 are implicated in the early phase of endotoxic shock. TNFα is one of the most important pro-inflammatory cytokines involved in endotoxic shock, and IL-6 has both pro- and anti-inflammatory activities in acute inflammation. Previous studies have already demonstrated that β-carotene treatment [21] or a diet enriched with α-tocopherol and β-carotene [29] decrease production of inflammatory mediators, such as TNFα and IL-6 in endotoxic shock. Phytosterols also decreased pro-inflammatory cytokines production (TNFα, IL-6) and increased anti-inflammatory (IL-10) production in Apo E-KO Mice [19].

In our experimental model, LPS induced a strong and significant increase in TNFα, IL-6 and IL-10 serum cytokine concentration with different kinetics, according to the cytokine. EUF pre-treatment prevents this increase induced by LPS in serum, which confirms the results obtained in vitro. However, compared to results obtained with awara pulp oil [2], we observed that EUF reduces serum TNFα concentration at 1.5 h less efficiently than serum IL-6 concentration at 6 h. EUF did not induce an increase in IL-10 serum concentration at 1.5 h but decreased it at 6 h. Differences in unsaponifiable content of EUF and oil lead to a difficult comparison. Indeed, the relative percentage of arundoin in oil is 3.6 times higher than in EUF. Arundoin is a triterpene methyl ether less soluble in ethanol than other phytosterols and, as far as we know, its biological properties have not yet been characterized. In the same way, all-trans-β-carotene is less soluble in ethanol than other carotenoids, thus, the relative percentage in oil is eight times higher than in EUF. Likewise, the fatty acid content of oil could also explain the observed differences in serum cytokine modulation.

Septic shock is also associated with increased oxidative stress and decreased antioxidant defences. Thus, antioxidant supplementation is potentially useful in the management of septic shock [30]. Therefore, we also measured the antioxidant capacity of serum after LPS-induced endotoxic shock. We showed a significant increase in antioxidant capacity at 1.5 h after LPS injection, whereas EUF pre-treatment prevents this increase. Similar results were obtained with awara pulp oil both in endotoxic shock and pulmonary inflammation model [2]. EUF pre-treatment decreased the LPS-induced inflammation; this may be due to inhibition of cell activation and thus to decreased ROS production, which secondarily induces a lower response from antioxidant mechanisms.

To conclude, the endotoxic shock model confirmed the anti-inflammatory effects of EUF observed in vitro on activated macrophages. These inhibitory effects on inflammatory mediator production may be due to inhibition of macrophage activation by identified compounds, i.e., carotenoids, phytosterols and tocopherols which might act in a synergistic way. These results confirmed that the anti-inflammatory properties of awara pulp oil [2] are in part due to the unsaponifiable fraction, and that consumption of awara fruit may have potential preventive effects on several inflammatory diseases. Nevertheless, further studies are needed to better understand these effects and in particular the involvement of fatty acids and arundoin. Indeed, the role of vegetable oils, like olive and red palm oils, in the improvement of several diseases may be due to both fatty acid composition and antioxidant and/or anti-inflammatory micronutrients [4, 7].

Supplementary material

11130_2012_323_MOESM1_ESM.doc (160 kb)
ESM 1(DOC 159 kb)

Copyright information

© Springer Science+Business Media New York 2012