Antioxidant and anti-inflammatory activities of Lonicera japonica Thunb. var. sempervillosa Hayata flower bud extracts prepared by water, ethanol and supercritical fluid extraction techniques.

Lonicera japonica Thunberg (LJ) has long been used as an antipyretic, anti-inflammatory and anti-infectious agent in East Asia. The subspecies L. japonica Thunb. var. sempervillosa Hayata (LJv) is a variant that mainly grows in Taiwan. This study examined the antioxidant and anti-inflammatory activities of the extracts from the flower buds of these two species. The extracts were obtained by three extraction methods: water extraction, ethanol extraction, and supercritical-CO2 fluid extraction (SFE). The antioxidant activities of dry LJ (dLJ) extracts were superior to those of LJv extracts. Water extracts possessed higher activities than that prepared by ethanol or SFE. The total polyphenols content, total flavonoids content, and the amount of chlorogenic acid and luteolin-7-O-glucoside were all higher in the water extracts compared to the other two. The SFE extracts of these two species all exhibited excellent anti-inflammatory activities. Although the water and ethanol extracts of dLJ extracts had higher anti-inflammatory activity than that of LJv extracts, the SFE extracts prepared from fresh LJv flower buds (fLJv) exhibited the highest activity among all extracts. The SFE effectively isolates the bioactive components of L. japonica and can obtain the L. japonica extracts with high anti-inflammatory activity.

Introduction

Lonicera japonica Thunb. (LJ), belonging to the Caprifoliaceae family, is known as Japanese honeysuckle, Jin Yin Hua and Ren Dong. It is a traditional Chinese medicine endowing with antidote, diuretic, tonic, antipyretic, anti-inflammatory and anti-infectious activities. It has been widely used in treating exopathogenic wind-heat, furuncles, epidemic febrile diseases, carbuncles, sores and some infectious diseases. It is also applied to treat chronic enteritis, pneumonia, acute tonsillitis, nephritis, acute mastitis, leptospirosis in some folk prescriptions. Recent uses include the prevention and treatment of human and animal viruses, such as SARS coronavirus and swine H1N1 flu virus (Shang et al., 2011). Several pharmacological studies have shown that LJ and its active ingredients possess wide bioactivities, such as antioxidant, anti-inflammatory, antibacterial, antiviral, blood fat reducing, antipyretic and antiendotoxin (Shang et al., 2011). Therefore, in addition to its folk medicinal uses, LJ is also used in health food and cosmetics.

More than 150 chemical compounds have been isolated from LJ. The major compositions are essential oils, flavones, saponins, iridoids, and organic acids. Chlorogenic acid, luteolin, luteolin-7-O-glucoside and essential oils have good pharmacological effects and are believed to be the active ingredients of LJ (Ikeda et al., 1994, Wang, 2010). Different plant parts, including flowers, flower buds, leaves and whole plant, contain different ingredients and have different activities. Among them, flowers and flower buds of LJ are the most frequently used plant parts in Taiwan. However, differences in habitat, harvest time, extraction methods, and flower preparation (fresh versus dry) also cause variations in chemical compositions and the quality of LJ extract (Shang et al., 2011).

L. japonica Thunb. var. sempervillosa Hayata (LJv, Mao Rong Tong) is a variant subspecies of LJ that mainly grows in Pingtung County and Taitung County, Taiwan (Hayata, 1919). The plant LJv resembles LJ except for its leaves, which are thinly hairy, densely villose beneath, and its flowers spread a special fresh fragrance. To the best of our knowledge, this study is the first to examine the bioactivities of LJv.

The L. japonica is usually extracted with water or organic solvents, such as methanol or ethanol. One alternative to traditional extraction by organic solvents is supercritical fluid extraction (SFE). The most commonly used supercritical fluid is CO2 because of its favorable operating temperature and pressure for extracting thermolabile ingredients. Supercritical CO2 is also non-toxic, non-flammable, widely available, chemically inert, and has low viscosity, low surface tension, high diffusivity and favorable density (Wang, 2011, Quitain et al., 2013). Compared to conventional organic solvent extraction methods, the advantages of SFE are its non-toxicity and high separation selectivity. Especially, the CO2 gas used in SFE operation can be easily recycled to avoid the concern of greenhouse effects. Therefore, the use of SFE for extracting natural products is rapidly increasing (King, 2014, Uddin et al., 2015, Yen et al., 2015).

This work compared extracts from dry and fresh flower buds of LJv and the dry flower buds of LJ. The fresh flower buds of LJ were not included because the fresh samples produced at the same area and the same season as the dry flower buds of LJ were difficult to purchase throughout the experiment duration. These extracts were prepared with water, ethanol, and supercritical-CO2 fluid. Their antioxidant, anti-inflammatory activities and their chemical compositions were then analyzed.

Materials and methods

Chemicals and plant materials

(+)-Catechin, chlorogenic acid, dimethyl sulfoxide (DMSO), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), 1,1-diphenyl-2-picrylhydrazyl (DPPH), Escherichia coli lipopolysaccharide (LPS), Folin-Ciocalteu reagent, gallic acid, hypoxanthine (HPX), luteolin-7-O-glucoside, N G-monomethyl-l-arginine acetate (L-NMMA), nitroblue tetrazolium (NBT), penicillin, streptomycin, xanthine oxidase (XOD) were purchased from Sigma Chemicals Co. (St Louis, MO, USA). Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum were from Gibco (Grand Island, New York, USA). All chemicals used in this study were of reagent or higher grade.

The reference dry flower buds of LJ (dLJ) were purchased from Henan, China. Flower buds of LJv (fLJv) freshly harvested from Jhutian Township, Pingtung (Taiwan) were obtained from a local farmer, Mr. Hong Ying. Some of these fresh flower buds were air-dried at 40 °C for 24 h, and then stored at −20 °C (designated as “dLJv”). All samples were harvested in April. The morphological characteristics of LJv were compared with those of a sample authenticated by Prof. Den-En Shieh (Department of Food Science and Technology, Tajen University of Technology, Pingtung, Taiwan). A DNA analysis of the plant materials was performed as described in our previous study (Chiou et al., 2007). The similarity of ITS2 sequences between LJv and LJ exceeded 99%, indicating that these two materials are indistinguishable at the species level.

Preparation of LJv and LJ extracts

The ethanol and water extracts of dLJv or dLJ were prepared by crushing or drenching 300 g of flower buds in 1800 ml ethanol (50, 75 or 95%) or water for 1 day. After repeating the extraction procedure three times, medicinal gauze was used to filter out insoluble debris. The filtrates were collected and concentrated with a vacuum evaporator. The extract samples were then dried in a freeze-dryer.

The ethanol and water extracts of fLJv were prepared by crushing 800 g of fresh flower buds and drenching them in 4800 ml ethanol (50, 75 or 95%) or water for one day, and repeated this operation for three times. The other operations were the same as above.

The supercritical-CO2 fluid extracts were prepared by processing the crushed flower buds (1 kg each) in the supercritical fluid extractor (5 l/1000 bar R&D unit, Natex, Ternitz, Austria). The dynamical extractions were performed at 150, 250 and 350 bar at 45 °C for 2 h. The extract samples were dried in a freeze-dryer.

Determination of antioxidant activity

Determination of scavenging activity on DPPH radicals

The scavenging activity against DPPH free radical was measured using the method of Hsu et al. (2005). In brief, 0.25 ml of 0.5 M DPPH ethanolic solution was mixed with 1.0 ml extract solution in an eppendorf tube. The absorbance at 517 nm was measured after keeping the solution in the dark for 30 min. The control was the measurement using ethanol to replace the extract sample in the reaction solution. The blank was measured by using ethanol to replace DPPH in the solution. After conducting the scavenging activity measurements under different concentrations of samples, the IC50 value, i.e. the concentration of sample that causes 50% inhibition, was estimated from the plot of scavenging activity against the sample concentration.

Determination of scavenging activity on superoxide anions

Superoxide anion radicals were generated in a HPX-XOD system by HPX oxidation and assayed by NBT reduction (Hsu et al., 2005). The solutions of NBT (300 μM), HPX (1.1 mM) and XOD (1.67 IU/ml) were prepared separately in a 0.1 M sodium phosphate buffer (pH 7.4). The HPX (760 μl) was mixed with 100 μl NBT and 100 μl extract solution in a 1-ml cuvette. Next, 40 μl XOD was added. The decrease in absorbance at 560 nm was measured every 15 s for 6 min. The rate of decreasing, designated as R1, was estimated by enzyme kinetic function of the spectrophotometer (Ultrospec 2100 pro, GE Healthcare, Amersham Place, UK). The control was the measurement using phosphate buffer to replace the extract sample in the reaction solution, and the decreasing rate of the absorbance is designated as R0. The O2·− scavening activity was determined by the following equation:O

Determination of total polyphenols content

The total amount of phenolic compounds of each extract was determined according to the method of Zielinski and Kozlowska (2000). A 0.15 ml aliquot of the diluted extract solution was mixed with 0.75 ml of 0.2 N Folin-Ciocalteu reagent and 0.6 ml of 7.5% sodium carbonate solution. The mixture was allowed to stand at room temperature for 30 min and then measured at 765 nm with a spectrophotometer. A calibration curve was constructed using gallic acid as a standard. The total polyphenols content is expressed as percentage of gallic acid equivalents of dry extract.

Determination of total flavonoids content

The total flavonoids content was determined according to Ismail et al. (2010) with some modifications. An aliquot of 0.3 ml sample, 1.2 ml of deionized water, and 0.075 ml of 15% Na2CO3 were added to a test tube. Five minutes later, 0.15 ml of 10% AlCl3 was added. After 6 min at room temperature, 0.5 ml of 1 M NaOH and 0.275 ml deionized water were added. The absorbance of the solution was measured against a blank at 510 nm. A calibration curve was constructed using catechin as a standard. The total flavonoids content is expressed as percentage of catechin equivalents of dry extract.

Determination of anti-inflammatory activity

The anti-inflammatory activity was determined by the suppressing effect on nitric oxide (NO) production in LPS-stimulated RAW264.7 macrophage cells (Bioresource Collection and Research Center; Hsinchu, Taiwan). The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10%(v/v) fetal bovine serum, 1% penicillin/streptomycin and 1.5 g/l sodium bicarbonate. The cells were cultivated in a humidified incubator at 37 °C with 5% CO2 and 95% air. The inhibitory effect of extract samples on NO production was measured using the method of Hong et al. (2014). RAW264.7 cells (4 × 105) were seeded into 100 μl DMEM medium in 96-well culture plates and incubated at 37 °C with 5% CO2 for 24 h. The cells were then incubated with 1 μl LPS (1 μg/ml) and 0.5 μl sample solution, which was the freeze-dried extract dissolved in DMSO. Following incubation for 24 h, 80 μl of the supernatant were collected for nitrite assay. The residual medium was removed, and the cell viability was evaluated using the MTT method (Hong et al., 2014). The nitrite concentration in the medium was measured as an index of NO production by using the Griess reaction. Griess reagent was freshly prepared from reagents A (1% sulfanilamide in 2.5% phosphoric acid) and B (0.1% N-1-naphthylethylenediamide dihydrochloride in water) at a ratio of 1:1. Equal volume of Griess reagent was added to the supernatants from the cells treated with the extract samples. Absorbance was measured on an ELISA reader (Model 550, Bio-Rad Laboratories, Hercules, CA, USA) at 540 nm. The NO-suppressing effect was expressed as the IC50 which denotes the concentration of extract sample causing 50% inhibition of NO production by LPS-activated RAW 264.7 cells. The L-NMMA, a clinical use iNOS inhibitor, was used as the positive control.

Analysis of chemical compositions

The sample analysis was conducted by an Acquity UPLC system coupled to a TQD mass spectrometer equipped with a Z-Spray ESI source (Waters, Manchester, UK). The UPLC system was equipped with an Acquity UPLC BEH C18 column (2.1 × 100 mm, 1.7 μm) and maintained at 50 °C with a column oven. The mobile phase consisted of 0.1% acetic acid in water in channel A and 100% acetonitrile in channel B. An isocratic elution was used with a flow rate of 0.4 ml/min at a ratio of 95% A and 5% B for a total of 18 min and the injection volume was 5 μl. The mass spectrometer was operated in electrospray positive mode with the capillary voltage set at 3.2 kV. The TQD mass spectrometer was operated with the following optimized source-dependent parameters (ESI source): capillary voltage 3.2 kV, cone voltage 40 V, desolvation temperature 300 °C, desolvation gas flow 500 l/h, cone gas flow 50 l/h, source temperature 120 °C and scanning from 100 to 800  m/z. The MassLynx 4.1 with QuanLynx software (Waters) was used for data processing.

Gas chromatography (GC) analysis was performed using Thermo Trace GC Ultra system (Thermo Fisher Scientific, Waltham, MA, USA). The GC column was DB-WAX capillary column (30 m × 0.32 mm, film thickness 0.25 μm, Agilent Technologies, Santa Clara, CA, USA). Injector and detector temperatures were set at 250 °C and 280 °C, respectively. Oven temperature was kept at 90 °C for 1 min, raised to 180 °C by a rate of 15 °C/min, kept at 180 °C for 3 min, and then raised to 240 °C by a rate of 5 °C/min, kept at 240 °C for 18 min. The carrier gas was helium at a flow rate of 1.8 ml/min. Diluted samples of 2.0 μl was injected manually and in the splitless mode.

Determination of the amounts of chlorogenic acid and luteolin-7-O-glucoside

The amount of chlorogenic acid and luteolin-7-O-glucoside were determined by HPLC (L-7100, Hitachi, Tokyo, Japan). The extracts were dissolved in 50% methanol and filtered with a 0.22 μm filter. The diluted samples were analyzed by a Cosmosil 5C18-MS-II column (250 mm × 4.6 mm, 5 μm; Nacalai Tesque, Kyoto, Japan) at column oven of 35 °C. The sample injection size was 20 μl. The flow rate was 1.0 ml/min, and the detection was carried out at 330 nm. For chlorogenic acid analysis, the solution of 0.4% phosphoric acid/acetonitrile (90:10, v/v) was used as the mobile phase. For luteolin-7-O-glucoside analysis, the mobile phase consisted of 0.4% phosphoric acid (channel A) and acetonitrile (channel B) using the gradient program as follows: 0–25 min, 90%–85% A; 25–40 min, 85%–80% A; 40–45 min, 80%–70% A; 45–50 min, 70%–60% A.

Statistical analysis

All experiments were conducted for three to five independent replicates. The data are expressed in terms of mean and standard deviation. The experimental data were analyzed using Microsoft Excel software (Microsoft Software Inc., USA).

Results and discussion

Extraction yield

In this study, 300 g of dry flower buds or 800 g of fresh flower buds (dry weight = 135.2 g) were extracted by ethanol and water, respectively. Additionally, 1 kg each of the dry or fresh flower buds were extracted by supercritical-CO2 fluid. Fig. 1 shows the extraction yields from different materials and different extraction methods. Because the fresh flower buds contained 83.1% water, for comparison under the same standard, their yields were calculated on the basis of dry material.

Fig. 1

Extraction yield of different extracts. An amount of 300 g each of the crushed dry flower buds or 800 g crushed fresh flower buds was extracted by ethanol and water, respectively. An amount of 1 kg each of the crushed dry or fresh flower buds was extracted by supercritical-CO2 fluid at different pressures. The extracted samples were dried in a freeze-dryer and weighed, and the extraction yield was calculated.

The yields of dLJv, fLJv and dLJ extracted by water were all around 38%. In ethanol extraction of dLJv, the yield decreased as the ethanol concentration increased. For fLJv and dLJ, the extraction with 75% ethanol obtained the highest yield. In SFE extraction, increasing the extraction pressure increased the yield of dLJv extracts but decreased the yield of dLJ extracts. The yields of the fLJv extracts from different extraction pressures were all similar (around 10%).

Antioxidant activity

As reported in the literature, the ethyl acetate extract and the methanol extract of LJ flower buds had good ROS scavenging activities (Cai et al., 2004, Choi et al., 2007). Myung et al. (2004) demonstrated that the LJ extracts prepared by 70% methanol or 70% acetone extraction had good tyrosinase inhibition, xanthine oxidase inhibition, and nitrite scavenging activities. Choi et al. (2007) reported that the ingredients of ethyl acetate extract, such as luteolin, caffeic acid, protocatechuic acid, isorhamnetin-3-O-d-glucopyranoside, quercetin 3-O-d-glucopyranoside, and luteolin 7-O-d-glucopyranoside, had high antioxidant activity. Most of these bioactive components are flavonoids.

This study investigated the antioxidant activities of different extracts by analyzing their DPPH and SOD scavenging activities. The DPPH radicals scavenging activity of antioxidants are attributed to their hydrogen donating abilities. While the SOD anion scavenging activity denotes the ability to remove free radicals, such as peroxyl, alkoxyl, hydroxyl, and nitric oxide, which formed from superoxide anions through the Fenton reaction, lipid oxidation or nitric oxidation (Ambrosio and Flaherty, 1992, Kostyuk and Potapovich, 1998). For all extracts examined in this study, radical scavenging activities were dose-dependent within the concentration range tested. The IC50 values of DPPH and SOD scavenging activities of different L. japonica extracts are compared in Fig. 2 A and B. In these analyses, the highest concentration tested was 1000 μg/ml. Scavenging activities were high for both DPPH and SOD in all water extracts. The dLJ extracts showed much higher antioxidant activities compared to the LJv extracts. The dLJ extract prepared by 75% ethanol had the lowest IC50 values of DPPH (56.8 ± 0.5 μg/ml) and SOD (134.1 ± 5.7 μg/ml) scavenging activities. For comparison, the IC50 values of DPPH and SOD scavenging activities of catechin, which was used as the positive control, were 5.6 ± 1.2 μg/ml and 22.7 ± 0.7 μg/ml, respectively. All SFE extracts had low antioxidant activities.

Fig. 2

Antioxidant activities and bioactive component contents of different extracts. (A) DPPH radical scavenging activity (expressed by IC50 values); (B) superoxide anion scavenging activity (expressed by IC50 values); (C) total polyphenols content; (D) total flavonoids content.

Phenolic compounds and flavonoids are considered powerful antioxidants because they donate hydrogen or electrons to form stable radical intermediates (Wang et al., 2006, Vijayalaxmi et al., 2015). The antioxidant activities of many fruits, vegetables and edible plant extracts have a strong correlation with their total polyphenols content (TPC) and total flavonoids content (TFC) (Katsube et al., 2004, Kiselova et al., 2006, Klimczak et al., 2007, Jayaprakasha et al., 2008, Tsai et al., 2014). Fig. 2C and D show the TPC and TFC values of these L. japonica extracts. High TPC and TFC were found for dLJ water and ethanol extracts, but TPC and TFC were all low for all LJv extracts and the SFE extracts.

The extracts of dLJv, fLJv and dLJ prepared by 75% ethanol were further analyzed by UPLC–MS/MS (Fig. 3 A–C). Eight components were identified as chlorogenic acid (Compound 1, RT = 2.7 min), loganin (Compound 2, RT = 3.4 min), secologanic acid (Compound 3, RT = 3.8 min), sweroside (Compound 4, RT = 4.9 min), secoxyloganin (Compound 5, RT = 6.0 min), epi-vogeloside (Compound 6, RT = 6.5 min), luteolin-7-O-glucoside (Compound 7, RT = 7.5 min), 3,5-di-O-caffeoylquinic acid (Compound 8, RT = 8.2 min). The dLJ contained higher amounts of chlorogenic acid, secologanic acid and epi-vogeloside than those of the other two. Chlorogenic acid and luteolin-7-O-glucoside are the bioactive components of LJ and are known to have potent antioxidant and anti-inflammatory activities (Choi et al., 2007, Huo et al., 2003, Guo et al., 2014). Further analyses for determining their concentrations were performed by HPLC. The chlorogenic acid concentration of dLJv, fLJv and dLJ ethanol (75%) extracts were 0.89%, 0.21% and 1.75%, respectively. The luteolin-7-O-glucoside concentration of dLJv, fLJv and dLJ ethanol (75%) extracts were 0.017%, 0.009% and 0.035%, respectively. The dLJ extracts had the highest chlorogenic acid content and the highest luteolin-7-O-glucoside content, which partly explains why the dLJ extracts had the best antioxidant activity.

Fig. 3

UPLC–MS/MS spectrometry profile of L. japonica extracts. (A) dLJv 75% ethanol extract; (B) fLJv 75% ethanol extract; (C) dLJ 75% ethanol extract.

Anti-inflammatory activity

Many studies have shown that different extracts of LJ can inhibit various inflammatory reactions, and suppress inflammatory cytokines expression. In a mouse model of proteinase activated receptor 2 (PAR2)-mediated edema in the paw, the water extract of LJ significantly inhibited the change in paw thickness, vascular permeability, PAR2 agonists-induced myeloperoxidase (MPO) activity, and TNF-α expression in paw tissue (Tae et al., 2003). Aqueous extract of LJ also attenuated trypsin-induced mast HMC-1 cell activation through inhibition of ERK phosphorylation and the inhibition of trypsin activity (Kang et al., 2004). Water extract of LJ inhibited both COX-1 and COX-2 activity, the expression of IL-1β-induced COX-2 protein and mRNA in A549 cells (Xu et al., 2007). All of these reports suggest that LJ is a good anti-inflammatory agent for treating inflammatory disorders.

This study assayed the anti-inflammatory properties of L. japonica extracts by suppressing the effects of NO production in LPS-activated RAW264.7 cells. Fig. 4 A shows the IC50 values for the NO-suppressing activities of these extracts. The relative effectiveness of these extracts for suppressing NO formation was SFE extracts > ethanol extracts > water extracts; and dLJ > dLJv > fLJv of ethanol extracts and water extracts. However, for the SFE extracts, the fLJv exhibited the highest activity among the three materials. The IC50 values of the fLJv extracts prepared at 150, 250 and 350 bar were 69.3 ± 6.0, 71.0 ± 6.8, 64.3 ± 7.5 μg/ml, respectively. The IC50 values of the dLJ extracts prepared at 150, 250 and 350 bar were 184.7 ± 13.1, 177.6 ± 8.9, 218.9 ± 18.3 μg/ml, respectively. In contrast, the IC50 values of the positive control L-NMMA was 22.4 ± 2.0 μg/ml. Moreover, these extracts did not significantly retard cell growth in a 24-h treatment under a concentration below their respective NO-suppressing IC50 value. Fig. 4B shows that the fLJv SFE extract prepared at 350 bar revealed a dose-dependent inhibitory effect on NO formation of RAW264.7 cells, and this extract had no cytotoxic effects on RAW264.7 cells.

Fig. 4

Anti-inflammatory activity assay using LPS-induced RAW264.7 cells. (A) The NO-suppressing activity (expressed by IC50 values) of different extracts. (B) Concentration effects of fLJv SFE extract prepared at 350 bar on NO formation of LPS-induced RAW264.7 cells and on the viability of RAW264.7 cells. PC: Positive control L-NMMA; V: cells were stimulated with LPS but without sample treatment.

The chemical compositions of these extracts were compared by gas chromatography. The analytical chromatograms show that the ethanol extracts of dLJ, dLJv and fLJv (Fig. 5 A–C) had comparable chemical composition patterns, but their compositions differed from those of the SFE extracts (Fig. 5D–F). Additionally, the dLJ SFE extract (Fig. 5D) and dLJv SFE extract (Fig. 5E) had comparable compositions, but the fLJv SFE extract (Fig. 5F) had more ingredients than the other two extracts, especially the ingredients at the retention time between 7 and 13 min.

Fig. 5

Comparison of chemical composition of different extracts by gas chromatography. (A) dLJ 75% ethanol extract; (B) dLJv 75% ethanol extract; (C) fLJv 75% ethanol extract; (D) dLJ SFE extract prepared at 350 bar; (E) dLJv SFE extract prepared at 350 bar; (F) fLJv SFE extract prepared at 350 bar.

Most of the fresh L. japonica samples had a high water content. Thus, a drying step is usually required before extraction. However, the drying process may result in the loss of some bioactive components. Ji et al. (1990) reported that the content of bioactive ingredient linalool in volatile oil from fresh LJ flowers was more than 14%, and other oil compositions were low-boiling point unsaturated terpenes; while the content of hexadecanoic acid was more than 26%, and linalool was less than 0.4% in volatile oil from dry LJ flowers. Fragrant composites are lost by heating and lighting in the drying process. Moreover, different extraction methods would result in different contents and compositions of LJ essential oils (Du et al., 2009).

An important feature of supercritical CO2 extraction is its low polarity. Therefore, it can only extract the non-polar components (Quitain et al., 2013). The properties of CO2 in supercritical state resemble those of the solvent ethyl acetate. Thus, more polar impurities are eliminated by SFE than by conventional aqueous and ethanol extractions. This feature leads to the concentration of non-polar bioactive components in the SFE extracts and results in high anti-inflammatory activity. Furthermore, CO2 extraction prevents degradation of extracts because it is performed in a non-oxidizing atmosphere (Jaime et al., 2007).

Conclusions

This study examined and compared the antioxidant and anti-inflammatory activities of LJv and LJ extracts and compared the activities of extracts prepared by different extraction methods. To our knowledge, this study is the first to compare the bioactivities of these two species. The antioxidant activities of dLJ extracts were superior to those of LJv extracts. In terms of anti-inflammatory activity, SFE was a good separation tool that provided excellent activities of L. japonica extracts. Since SFE concentrates non-polar bioactive components of LJ and LJv in the extraction step, it can obtain the extracts with high anti-inflammatory activity. Additionally, SFE is non-toxic, non-flammable, and easy recycle use of CO2. Therefore, using SFE to recover valuable components from L. japonica is effective and clean, and is a good alternative extraction method.

Although the water and ethanol extracts of dLJ extracts had higher anti-inflammatory activity compared to LJv extracts, the fLJv SFE extracts had superior activity to dLJ extracts. Further studies are needed to compare the active ingredients of fLJv, dLJv and dLJ SFE extracts; and to investigate the other bioactivities of LJv and LJ extracts, such as anti-microbial, antiviral and antipyretic activities.