The 20 commercially available EOs were donated by Nagaoka Perfumery Co., Ltd. (Osaka, Japan). The 17 types of hydrodistilled EOs used in the study were from herbal plants, including basil (Ocimum basilicum L., produced in the Union of the Comoros), caraway (Carum carvi L., Europe), carrot seed (Daucus carota L., France), celery seed (Apium graveolens L., India), chamomile (Matricaria chamomilla L., Egypt), citronella (Cymbopogon winterianus Jowitt, Indonesia), clary sage (Salvia sclarea L., USA), clove (Syzygium aromaticum L., Madagascar), cumin (Cuminum cyminum L., India), eucalyptus (Eucalyptus globulus L., China), lemongrass [Cymbopogon citratus (DC.) Stapf, India], marjoram (Majorana hortensis Moench., France), nutmeg (Myristica fragrans Houtt., Indonesia), sage (Salvia officinalis L., Albania), sandalwood (Santalum album L., India), spearmint (Mentha spicata L., USA), and thyme (Thymus vulgaris L., India). The cold-pressed EOs were derived from 3 species of citrus fruits: lemon (Citrus limon L., USA), lime (C. aurantifolia Swingle, Mexico) and orange (C. sinensis L., Portugal). Five terpenoids, citral (3,7-dimethyl-2,6-octadien-1-al), geraniol, geranyl acetate, linalool and camphene were purchased from Sigma-Aldrich, Inc. (St. Louis, MO, USA). Tranilast and glycyrrhetinic acid, anti-allergic and anti-inflammatory controls, respectively, were also obtained from Sigma-Aldrich, Inc. Geranial was prepared by the oxidation of geraniol with manganese dioxide, as previously described (12). For western blot analyses, anti-nuclear factor-κB (NF-κB) p65 and anti-inhibitor of NF-κB (IκB)-α antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Horseradish peroxidase-conjugated anti-rabbit immunoglobulin G (IgG) antibody was obtained from Thermo Scientific (Kanagawa, Japan). All other reagents were of analytical grade and were obtained from Nacalai Tesque, Inc. (Kyoto, Japan).

RBL-2H3 rat basophilic leukemia and RAW264.7 murine macrophage cell lines were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). The RBL-2H3 cells were cultured in Eagle’s minimum essential medium supplemented with 4.5 g glucose/l plus 10% fetal bovine serum (FBS), 5 mM L-glutamine, 50 U/ml penicillin and 50 U/ml streptomycin. The RAW264.7 cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 4.5 g glucose/l plus 10% FBS, 5 mM L-glutamine, 50 U/ml penicillin and 50 U/ml streptomycin. All cells were cultured at 37°C in their respective media in a humidified atmosphere of 5% CO₂/95% air.

Mast cell degranulation-mediated histamine release is accompanied by the release of β-hexosaminidase (13). Therefore, we performed a β-hexosaminidase release assay, as previously described (14) with slight modifications to determine the degranulation of mast cells as a measure of the anti-allergenic activity of the examined compounds. Briefly, the RBL-2H3 cells at 8×10⁴ cells/well in 24-well plates were washed with Tyrode’s buffer (137 mM NaCl, 5.6 mM glucose, 11.9 mM NaHCO₃, 2.7 mM KCl and 0.32 mM NaH₂PO₄) containing 1 mM CaCl₂ and 0.5 mM MgCl₂, and the EOs (in 0.5% DMSO) were then added to the culture wells at a final concentration of 100 μg/ml. The cells were then stimulated with 5 μM A23187 and incubated for 30 min. The cell supernatant and total cell lysate dissolved in 2% Triton X-100 were collected and mixed with substrate solution (2 mM p-nitrophenyl-N-acetyl-β-D-glucosaminide in 0.1 M sodium citrate buffer, pH 4.5). The mixture was incubated for 90 min at 37°C and the reaction was then terminated by the addition of a stopping buffer containing a 0.2 M glycine buffer at pH 11.0. Absorbance at 405 nm was measured using a microplate reader (Vmax-K; Molecular Devices, LLC, Sunnyvale, CA, USA). The EO-mediated inhibition of β-hexosaminidase release and activity was expressed as the percentage inhibition and calculated using the following formula: % activity = [(β-hexosaminidase release in the absence of EO − β-hexosaminidase release in the presence of EO)/β-hexosaminidase release in the absence of EO] ×100.

The RAW264.7 cells were placed in a 12-well plate at 5×10⁴ cells/well and incubated for 24 h. The cells were then pre-treated with each EO at a final concentration of 100 μg/ml in 0.5% DMSO for 30 min prior to the addition of LPS (100 ng/ml). Following LPS stimulation for 24 h, the cell culture medium was collected to quantify the secreted TNF-α using a commercially available enzyme-linked immunosorbent assay (ELISA) development system (Bay Bioscience Co., Ltd., Kobe, Japan), in accordance with the manufacturer’s instructions. The EO-mediated inhibition of TNF-α production was expressed as the percentage inhibition, calculated using the following formula: % activity = [(TNF-α production in the absence of EO − TNF-α production in the presence of EO)/TNF-α production in the absence of EO] ×100.

GC-MS was performed using an Agilent 6890 gas chromatograph coupled with an Agilent 5972A MSD mass spectrometer (Agilent Technologies Japan, Ltd., Tokyo, Japan). The chromatograph was fitted with an InertCap WAX (polyethylene glycol)-fused silica capillary column (60 m × 0.25 mm i.d.; film thickness, 0.25 μm; GL Sciences, Inc., Tokyo, Japan), with purified helium (≥99.99995%) carrier gas at 149 kPa inlet pressure, in split/splitless mode and at 1.2 ml/min. The oven temperature was programmed as follows: 5 min at 70°C and then 3°C increments/min to 240°C, with both the injector and transfer line held at 240°C. The mass spectrometer operated at a 70 eV electron impact, with an ion source temperature of 150°C, and the data collected in full scan mode over a mass scan range revealed an m/z of 27–400. Each 0.6 μl sample was injected in a split ratio of 1:60. A qualitative analysis was performed by similarity searches of private and commercial mass spectral databases (Wiley 275 and the National Institute of Standards and Technology 02).

The RAW264.7 cells were pre-treated with 10 μg/ml of a test compound for 30 min followed by treatment with 100 ng/ml LPS. After 30 min, the nuclear proteins and whole cell lysates were collected, using the methods described in the study by Nishiumi et al (15). The protein concentrations of the nuclear proteins and whole cell lysates were measured using the BCA™ Protein Assay kit (Pierce, Rockford, IL, USA), according to the manufacturers instructions. Briefly, 25 μl of each sample or 2 mg/ml BSA solution (as a standard) were added to a 96-well microplate well, and then 200 μl of BCA Working Reagent were added to each well. The microplate was incubated at 37°C for 30 min, and the protein concentration was determined by measuring the absorbance at 575 nm. The nuclear proteins and whole cell lysates were subjected to western blot analysis to evaluate the nuclear translocation of NF-κB and the protein expression of IκB-α.

The nuclear proteins (30–50 μg of protein/sample) and whole cell lysates (50 μg of protein/sample) were boiled in a quarter volume of sample buffer [1 M Tris-HCl, pH 7.5, 640 mM 2-mercaptoethanol, 0.2% bromphenol blue, 4% sodium dodecyl sulfate (SDS) and 20% glycerol] and then separated on 10% SDS-polyacrylamide gels. The proteins in the gels were transferred onto a PVDF membrane. The membrane was blocked with 1% skimmed milk (for NF-κB) or 5% bovine serum albumin in TBS-T (10 mM Tris-HCl, 100 mM NaCl and 0.5% Tween-20) (for IκB-α) and probed with an anti-NF-κB p65 (1:1,000) or anti-IκB-α (1:1,000) antibody and subsequently with the horseradish peroxidase-conjugated secondary antibody. The protein-antibody complex was detected using ChemiLumi-ONE (Nacalai Tesque, Inc.) and an Image Reader (ImageQuant Imager 350/350Lumi; GE Healthcare Life Sciences, Uppsala, Sweden). The intensity of each band was analyzed using ImageJ software, which was developed at the National Institutes of Health. The results were expressed as a percentage of the mean protein expression ratio of the test compound-treated samples compared with that of the control group (i.e., the cells incubated in the absence of the test compound).

Female 6-week-old imprinting control region (ICR) mice (25–27 g body weight) were obtained from Japan SLC, Inc. (Hamamatsu, Japan), maintained on a standard moderate fat (MF) diet (Oriental Yeast Co., Ltd., Osaka, Japan), and provided with water ad libitum. All animal experiments were approved by the Kobe Gakuin University Animal Committee (Kobe, Japan) according to the university guidelines for the care and use of laboratory animals.

The passive cutaneous anaphylaxis (PCA) reaction was measured according to a previous study (14). Mice were sensitized by an intradermal injection of 0.1 μg of anti-dinitrophenyl (DNP)-IgE in the ear and intravenously challenged 4 h later with 0.2 ml (1 mg/ml) of a DNP-labeled human serum albumin solution containing 2% Evans blue dye. In a series of experiments, a test compound (100 mg/kg) was administered orally 2 h prior to antigen challenge. The animals in the control group received saline. The mice were subsequently sacrificed and the ears removed and weighed 30 min after the challenge. After dissolution of the ears in 200 μl of 1 N potassium hydroxide (KOH), they were incubated overnight at 37°C for the measurement of the amount of Evans blue dye present in the exudates. For this purpose, the dissolved tissue solution was added to 400 μl of a mixture of acetone and 0.6 N phosphoric acid (5/13, v/v) and the optical density at 620 nm was measured. The amount of dye in the exudates was calculated from an Evans blue standard curve and the results expressed as a percentage of the mean exudate dye amount in the ear samples of the treated test mice compared with that of the controls.

The mouse inflammatory test was performed according to the method described in the study by Gschwendt et al (16). Briefly, an acetone solution containing a test compound at 500 μg/20 μl or a vehicle control of 20 μl of acetone was applied to the inner section of a mouse ear and, 30 min later, an acetone solution of 12-O-tetradecanoylphorbol-13-acetate (TPA, a chemical edema inducer) at 0.5 μg/20 μl was applied to the same part of the ear; acetone, followed by a TPA application, served as a control. After 7 h, a 6-mm diameter disk was obtained from the ear and weighed. The anti-inflammatory activity was determined by comparing the ear disk weight of the test group and the control group animals. The results were expressed as a percentage and were derived using the following formula: percentage activity = [(TPA only) − (tested compound plus TPA)]/[(TPA only) − (vehicle)] ×100.

All the results are expressed as the means ± standard deviation (SD) of at least 3 independent determinations for each experiment. Statistical significance between each experimental group was analyzed using the Student’s t-test, and a probability value of 0.01 and 0.05 was used as the criterion of significance.

First, the inhibitory effects of 20 species of EOs on mast cell degranulation were investigated. β-hexosaminidase release is widely used as an indicator for evaluating the extent of degranulation by mast cells and the corresponding release of histamine and other chemical mediators, which is an important process in initiating the immediate type of hypersensitive reaction. The effects of the EOs (at 100 μg/ml) on the release of a chemical mediator (β-hexosaminidase) was investigated in the RBL-2H3 cells treated with the calcium ionophore, A23187. The EOs did not influence the growth of RBL-2H3 cells and did not inhibit β-hexosaminidase enzyme activity (data not shown). The degree of degranulation was calculated by assessing the β-hexosaminidase activity in the supernatant and cell lysate. Among the 20 EOs examined, treatment with EOs from chamomile, lemongrass and sandalwood produced >40% inhibition of mast cell degranulation, as assessed by β-hexosaminidase release. Moreover, of the EOs examined, lemongrass EO showed the strongest inhibitory activity (Fig. 1A). On the other hand, EOs from eucalyptus and lime did not influence β-hexosaminidase release. A23187 is known to induce a calcium flux, which may further induce protein kinase C (PKC) activation, a process essential for mast cell degranulation (8). It is hence plausible that the β-hexosaminidase inhibition induced by treatment with EOs may affect PKC activity. The mechanisms through which EOs inhibit mast cell degranulation require further investigation.

We then investigated whether plant EOs, at 100 μg/ml, inhibit the production of TNF-α induced by LPS stimulation in cultured cells. The inflammatory cytokine, TNF-α, activates the NF-κB signaling pathway by binding to the TNF-α receptor, thereby initiating an inflammatory response implicated in various inflammatory diseases (17). In cultured RAW264.7 murine macrophage cells, none of the compounds examined showed cytotoxicity at 100 μg/ml (data not shown); the EOs had no effect on the proliferation of peritoneal macrophages. The RAW264.7 cells produced 693 pg/ml of TNF-α following stimulation with LPS. The rate of TNF-α production in the controls was set at 0%. Among the EOs examined, those derived from lemongrass showed the strongest effect on TNF-α production (>50% inhibition) and those from chamomile, lemon and sandalwood displayed the second strongest effects (30–50% inhibition) (Fig. 1B). EOs derived from eucalyptus, lime and nutmeg failed to affect TNF-α production (0% inhibition).

The possible association between the observed inhibition of β-hexosaminidase in mast cells and TNF-α inhibition in macrophages was confirmed by comparing the effects of the 20 EOs on these biological activities. The inhibition of β-hexosaminidase release showed a high correlation with TNF-α inhibition, with a significant correlation coefficient (R²=0.783). These results led us to speculate that the anti-allergic activity related to the inhibition of β-hexosaminidase release and the anti-inflammatory effects induced by the suppression of TNF-α may be EO-related events within a shared biological pathway.

Among the 20 EOs examined, lemongrass EO produced the strongest inhibitory activities in the cultured cells (Fig. 1); thereafter we focused on lemongrass EO.

The association between the biological activity of lemongrass EO and EO composition was clarified by an analysis of the primary EO components by GC-MS. In total, 138 compounds were detected in the EO, of which 40 components were identified, and the 15 major compounds are listed in Table I. The 2 main compounds representing >30% of the total content were geranial and neral (40.16 and 34.24%, respectively). A mixture of these stereoisomic monoterpene aldehydes is referred to as citral. Of the identified components, 5 compounds, citral [1], geraniol [3], geranyl acetate [4], linalool [5] and camphene [6], were commercially available. Geranial [2] was chemically synthesized. Preparations of compounds 1–6 contain terpenoids of low molecular weight, and their corresponding chemical structures are shown in Fig. 2.

Based on the above results, the inhibitory effects of 100 μg/ml of lemongrass EO and its major components, compounds 1–6, on β-hexosaminidase release by mast cells and TNF-α production by macrophages were investigated. In the RBL-2H3 and RAW264.7 cells, these compounds did not show cytotoxicity at 100 μg/ml (data not shown).

Citral [1] (a mixture of neral and geranial) and geranial [2], which is a trans-isomer of citral, markedly inhibited the release of β-hexosaminidase induced by the calcium ionophore, A23187, and this inhibitory effect was similar to that elicited by lemongrass EO (Fig. 3A). On the other hand, 4 other components, geraniol [3], geranyl acetate [4], linalool [5] and camphene [6], exerted weak or moderate effects, and the inhibitory effect on the release of β-hexosaminidase elicited by compounds 3–6 was 2-fold weaker than that elicited by compounds 1–2 or lemongrass EO.

The effects of compounds 1–6 on TNF-α production were similar to those on β-hexosaminidase release, and compounds 1–2 and lemongrass EO exerted stronger inhibitory effects than compounds 3–6 (Fig. 3B). These results suggest that citral [1] and/or geranial [2] are the bioactive components conferring the biological activities to lemongrass EO. Therefore, we sought to explore compounds 1–2 further.

NF-κB activation is known to be the rate-controlling factor during an inflammatory response. Therefore, the inhibitory effects of lemongrass EO and its major components, citral [1] and geranial [2], on the LPS-induced nuclear translocation of NF-κB were examined in RAW264.7 cells (Fig. 4A). Western blot analysis revealed that 10 μg/ml of citral [1] and geranial [2] inhibited the LPS-induced NF-κB nuclear translocation, with 40 and 52% inhibition, respectively. Lemongrass EO, at 10 μg/ml, also inhibited the nuclear translocation of NF-κB, and this inhibitory activity was similar to that of compounds 1–2.

Stimulation with LPS results in the activation of Toll-like receptor 4 and that of downstream IκB kinases (IKKs), which in turn phosphorylate IκB, resulting in IκB degradation and the translocation of NF-κB into the nucleus (18). IκB-α inhibits NF-κB by masking the nuclear localization signals (NLS) of NF-κB proteins and maintaining them in a sequestered and inactive state in the cytoplasm (19). In addition, IκB-α blocks the ability of the NF-κB transcription factor to bind to DNA, which is required for the proper functioning of NF-κB (20). Therefore, the effects of lemongrass EO and compounds 1–2 on the LPS-induced decrease in the protein expression of IκB-α were examined in the RAW264.7 cells (Fig. 4B). IκB-α protein expression was reduced by 59% following LPS stimulation (Fig. 4B). The amount of IκB-α protein in the cells treated with compounds 1–2 (10 μg/ml each) was almost the same as that in the LPS-treated cells (the IκB-α protein level was approximately 35% compared to that in the control untreated cells), suggesting that compounds 1–2 did not influence the protein expression of IκB-α. These results demonstrate that lemongrass EO and its major components suppress the NF-κB nuclear translocation through an IκB-α-independent mechanism. The effects of these compounds at other points in the NF-κB signaling pathway required further investigation.

The IgE-mediated PCA reaction in vivo is a method used for studying the mechanisms of the immediate hypersensitivity reaction. As shown in Fig. 5A, 100 mg/kg of tranilast, a commonly used anti-allergic drug that targets mast cell degranulation and inhibits the PCA reaction, was used as a positive control. Lemongrass EO and its major components, citral [1] and geranial [2], at 100 mg/kg also inhibited the PCA reaction in the mice by 60.7, 57.7 and 77.4%, respectively. The inhibitory effects of lemongrass EO and compounds 1–2 were >2-fold stronger than those of tranilast, suggesting a great potential for use as anti-allergic compounds.

We examined the anti-inflammatory activity of lemongrass EO, citral [1] and geranial [2] by examining their effects on TPA-induced inflammatory edema of the mouse ear in vivo. The application of TPA (0.5 μg) to the mouse ears induced edema, resulting in a 241% increase in the weight of the ear disk 7 h after application. Pre-treatment with 500 μg/ear of these compounds significantly suppressed inflammation, and the strength of the inhibitory effect was ranked as follows: geranial [2] > lemongrass EO > citral [1] (Fig. 5B). Therefore, the in vivo anti-inflammatory effects of these compounds displayed the same order as their in vivo anti-allergic effects (Fig. 5A). These results suggest that the anti-allergic effects of the components of lemongrass EO positively correlate with the anti-inflammatory activity observed. Glycyrrhetinic acid (500 μg/ear), a known anti-inflammatory agent, had a 40.5% anti-inflammatory effect in this assay system, and these compounds were found to be approximately 1.5-fold more effective than this agent.

The biological effects elicited by compounds 1–2 were similar to those produced by almost the same concentrations of lemongrass EO (Figs. 3–5), suggesting that the anti-allergy and anti-inflammatory effects of lemongrass EO are largely due to its 2 major components, citral [1] and/or geranial [2].