Hallasan wormwood (A. japonica var. Hallaisanensis) was collected more than individual 10 plants in the Jeju Island area of South Korea from May to June. The plant was identified and authenticated by Dr. Hyoun-Chol Kim. After air-drying the aerial parts (leaves and stems) and roots of the plant material at room temperature, the dried material was powdered (< 0.5 ㎜).

Folin Ciocalteu’s phenol reagent was purchased from Hayashi Pure Chemical Inc., Ltd. (Osaka, Japan). Flavonoid standard compound was obtained from ChemFaces Biochemical Co., Ltd. (Wuhan, China). 6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (trolox), phenolic acid standard compound were purchased from Sigma Aldrich Co. (St. Louis, MO, USA). Aluminium nitrate, sodium carbonate and ferric chloride were purchased from Wako Pure Chemical Inc., Ltd. (Kyoto, Japan). Methanol and the other solvents used in the extraction were purchased from Dae-Jung Chemicals and Metals Co., Ltd. (Siheung, Korea). The other chemicals used in experiment were of ACS grade, obtained from Sigma Aldrich Co., Ltd. (St. Louis, MO, USA).

Hallasan wormwood extracts of aerial parts and roots were obtained by extraction with a 99.5% methanol three times each 3 hours at solid-liquid ratio of 20 ㎖/g (ratio of the solvent volume per gram of raw material). After filtering (5A filter paper, Advantech Co., Ltd., Tokyo, Japan), Hallasan wormwood extracts (aerial parts and roots) were concentrated by evaporation to remove methanol at 45℃ under reduced pressure using a rotary evaporator (Hei-VAP Precision, Heidolph, Schwabach, Germany). Methanol was removed from a portion of the Hallasan wormwood extract and it was sequentially fractionated by solvent-solvent fractionation to yield n-hexane (Hex), ethyl acetate (EtOAc), water-saturated butanol (BuOH) and residual water (Water) fractions respectively.

Total phenolic content was measured using the Folin-Ciocalteu method (Li et al., 2008). After diluting the extract with 700 ㎕ of distilled water, 100 ㎕ of 50% (v/v) Folin-Ciocalteu reagent was mixed and treated for 2 hours. After that, 100 ㎕ of 20% sodium carbonate was added to develop the color, and then the absorbance at 750 ㎚ was measured using an i-Mark microplate reader (168-1135, Bio-Rad Inc., Hercules, CA, USA). Gallic acid was used as the standard compound for expressing the total phenolic content (㎎·GAE/g).

Total flavonoid content measured by reaction with aluminum nitrate. One hundred ㎕ of extract and 300 ㎕ of ethanol were mixed, and 20 ㎕ of 10% aluminum nitrate and 20 ㎕ of 1 M potassium acetate were added. The next distilled water was added with 560 ㎕ and reacted for 1 hour. The reaction solution was measured for absorbance at 415 ㎚ with a microplate reader. Quercetin was used as the standard compound for expressing the total flavonoid content (㎎·QE/g).

DPPH radical scavenging activity was measured by applying the method of Blois (1958). 40 ㎕ of extract and 160 ㎕ of 0.15 mM DPPH solution reacted in a dark room. After 30 minutes, the absorbance at 490 ㎚ was measured and the DPPH radical scavenging activity was calculated.

The concentration IC₅₀ (inhibitory concentration, ㎍/㎖) required to eliminate the 50% DPPH radical was calculated using the inhibition rate measured for each concentration.

TEAC measurements were performed by modifying the method of Gliszczyńska-Świgło (2006). In the ABTS solution, 2,2-azino-bis-(3-ethylbenzo-thiazoline-6-sulphonic acid) and 2.45 mM potassium persulfate were mixed and reacted for 12 hours to obtain an absorbance value of 0.70 ± 0.02 at 734 ㎚. After, 950 ㎕ of ABTS solution and 50 ㎕ of extract were mixed, reacted for 5 minutes, and the absorbance at 734 ㎚ was measured. Trolox was used as the standard compound for indicating TEAC (mM·TE/g).

The FRAP measurements were performed by modifying the method of Gliszczyńska-Świgło (2006). After adding 150 ㎕ of FRAP working solution (1 : 1 : 10, 10 mM 2,4,6-tripyridyl-s-triazine, 20 mM FeCl₃, 0.3 M sodium acetate buffer) and 5 ㎕ of extract, the reaction was carried out at 37℃, for 30 minutes. The absorbance at 595 ㎚ was measured using a microplate reader, and the FRAP (mM·FE/g) of the extract was calculated as the absorbance value of FeSO₄ used as a standard substance.

ORAC was performed by modifying the method of Zulueta et al. (2009). Two hundred ㎕ of the extract diluted with 78 nM fluorescein solution was treated at 37℃, for 10 minutes. Next, 50 ㎕ of 221 mM 2,2'-azobis (2-aminopropane) dihydrochloride was added, and fluorescence was measured at 1-minute intervals for 1 hour under the conditions of excitation 485 ㎚ and emission 535 ㎚. At this time, the value of the decreasing fluorescence was calculated by the area under the curve, and the ORAC value was calculated using the decreasing value of the trolox-specific fluorescence.

The phenolic compound analysis of the extracts and fractions was divided into phenolic acid and flavonoid, and the retention time and photodiode array (PDA) were analyzed under the same conditions by HPLC under the conditions shown in next using the respective standard compounds.

Phenolic acid chromatographic separation was performed on a Shimadzu 20A system (Shimadzu Co., Kyoto, Japan) equipped with a UV-DAD detector, using a YMC-Triart C18 column (250 ㎜ × 4.6 ㎜, 5 ㎛ hybrid silica based ODS, YMC Co., Seongnam, Korea). The mobile phase consisted of 0.1% trifluoroacetic acid containing water (A) and methanol (B). A gradient program was used as follows: 10% B in the first 5 min, 20 - 25% B during 10 - 25 min, 25 - 40% B during 30 - 45 min, 60 - 100% B during 50 - 60 min, then B held at 10% for 10 min. The flow rate was set at 1 ㎖/min and the injection volume was 10 ㎕. Column temperature was controlled at 40℃. Spectral data from 190 to 360 ㎚ were recorded and the phenolic acid chromatograms (6.25 - 200.00 ㎍/㎖) were monitored at 245 ㎚.

Flavonoid was performed on a Shimadzu 20A system using a YMC-Triart C18 column. The mobile phase consisted of 0.1% formic acid containing water (A) and acetonitrile (B). A gradient program was used as follows: 10 - 20% B in the first 5 min, B held at 20% for 15 min, 25 - 30% B during 30 - 45 min, 30 - 50% B during 55 - 60 min, 50 - 60% B during 60 - 70 min, 60 - 80% B during 70 - 75 min, B held at 80% for 5 min, then B held at 10% for 5 min. The flow rate was set at 1 ㎖/min and the injection volume was 10 ㎕. Column temperature was controlled at 40℃. Spectral data from 190 to 360 ㎚ were recorded and the flavonoid chromatograms (3.125 - 100.000 ㎍/㎖) were monitored at 280 ㎚.

All data were analyzed in triplicate and the data were expressed as mean ± standard deviation. One-way Analysis of Variance (ANOVA) was used and Duncan`s Multiple Range Test (DMRT) was used to confirm the significance of the differences between groups (*p < 0.05). All statistical analysis were performed on SPSS (Ver. 20.0, SPSS Inc., Chicago, IL, USA)

The total phenolic content of the aerial parts extract was 74.94 ± 1.58 ㎎·GAE/g. Hex (11.67 ± 0.44 ㎎·GAE/g) and water fraction (24.28 ± 1.19 ㎎·GAE/g) showed lower content than the extract in the total phenolic content of fraction, but EtOAc (303.92 ± 19.04 ㎎·GAE/g) and the BuOH fraction (157.40 ± 12.55 ㎎·GAE/g) showed more than twice as high total phenolic content as the extract (Table 1).

The total phenolic content of the roots extract was 53.04 ± 2.50 ㎎·GAE/g, which was lower than that of the aerial parts extract. The total phenolic content of the roots fractions tended to be similar to that of the aerial parts fractions. Hex (7.20 ± 1.31 ㎎·GAE/g) and water fractions (29.45 ± 1.08 ㎎·GAE/g) were lower than the roots extract. EtOAc (420.00 ± 11.99 ㎎·GAE/g) and BuOH fractions (69.33 ± 9.12 ㎎·GAE/g) showed a higher phenolic content than the roots extract (Table 1).

In a study by Ahmed et al. (2014), the hexane fraction reported that the removal of chlorophyll and non-polar components affected improving the extraction efficiency of phenolic components. In this experiment, the total phenolic content from the Hex fraction was low, and the high phenol content was confirmed in the EtOAc and BuOH fractions. It is probable that the Hex fraction obtained a non-phenolic substance and the phenolic component was fractionated with EtOAc and BuOH.

Flavonoids are components belonging to polyphenols, which are plant pigments with 15 carbon skeletons and C3-C6-C3 as the basic structure, and were found to be contained in almost all parts such as leaves, stems, and roots of plants (Hertog et al., 1992). Flavonoids are substances known to have direct protective effects on cells as well as antioxidant, antiinflammatory, tumor, antiviral and antibacterial properties due to their biological properties (Cazarolli et al., 2008). The total flavonoid content of the aerial parts extract and the roots extract was 31.47 ± 0.65 ㎎·QE/g and 52.09 ± 1.53 ㎎·QE/g, respectively (Table 1).

The total phenolic content of the aerial parts extract (74.94 ± 1.58 ㎎·GAE/g) was higher than that of the roots extract (53.04 ± 2.50 ㎎·GAE/g), but the total flavonoid content was found to be higher in the roots extract (52.09 ± 1.53 ㎎·QE/g). Kim et al. (2012) reported that no high correlation was found between the total flavonoid content with the total phenolic content of various plant extracts. However, the total flavonoid content of each fraction showed a tendency similar to the total phenolic content. This is considered as the result of the flavonoids belonging to polyphenols being displayed as being fractionated with EtOAc and BuOH solvent in the process of fractionation.

As a result, the roots EtOAc fraction showed the highest total phenolic content and total flavonoid content, and is thought to have a large amount of phenolic compounds.

DPPH radical scavenging activity, TEAC, FRAP and ORAC analysis were performed to test the antioxidant activity of the Hallasan wormwood extracts and fractions.

The DPPH radical scavenging activity IC₅₀ of the aerial parts extract was 469.78 ± 3.67 ㎍/㎖, and the EtOAc fraction was 121.69 ± 0.37 ㎍/㎖, which was the highest activity (Table 2). The IC₅₀ of the roots extract was 108.66 ± 1.44 ㎍/㎖, which was higher than that of the aerial parts EtOAc fraction, and the roots EtOAc fraction was 7.58 ± 1.14 ㎍/㎖, which was equivalent to that of BHA (8.09 ± 0.17 ㎍/㎖).

As a result of TEAC, the aerial parts extract (369.75 ± 10.34 mM·TE/g) and the root extract (381.67 ± 13.26 mM·TE/g) showed similar activities (Table 2). The TEAC values of the aerial parts EtOAc fraction and roots EtOAc fraction were 1167.38 ± 40.95 and 1283.42 ± 54.49 mM·TE/g, which were the highest among the fractions.

The FRAP method is an experiment to measure the ability to reduce ferric iron (Fe³⁺) to ferrous iron (Fe²⁺), and the reducing power can be compared and analyzed. The ability to donate electrons to reactive oxygen species and free radicals by various mechanisms of antioxidant activity is called the reducing power, and it is known that this is closely related to the antioxidant activity (Siddhuraju et al., 2002).

The FRAP of the aerial parts extract was 1131.29 ± 13.56 mM·FE/g. The highest aerial part fractions were the BuOH fraction (1592.68 ± 25.81 mM·FE/g). The roots extract showed a FRAP value of 473.98 ± 5.22 mM·FE/g, which was lower than that of the aerial parts extract. However, the roots EtOAc fraction was 5046.91 ± 88.39 mM·FE/g, the highest of all samples (Table 3).

The ORAC values are aerial part EtOAc fraction (25.69 ± 1.18 mM·TE/g), BuOH fraction (22.88 ± 0.55 mM·TE/g), extract (20.04 ± 0.54 mM·TE/g), Hex fraction (11.89 ± 0.30 mM·TE/g), and water fraction (5.02 ± 1.46 mM·TE/g) had the highest content in that order (Table 3).

The roots extract showed 12.82 ± 0.48 mM·TE/g, which was lower than that of the aerial parts extract. In addition, the EtOAc fraction (28.57 ± 1.01 mM·TE/g) except for the Hex (15.67 ± 0.29 mM·TE/g), BuOH (21.13 ± 1.18 mM·TE/g), and water (4.37 ± 0.10 mM·TE/g) fractions showed higher activity than the roots extract and fractions (Table 3).

von Gadow et al. (1997) reported that caffeic acid and rutin had higher DPPH radical scavenging activity than BHA. From the results of this study, it was impossible to confirm a direct correlation between the antioxidant activity of the extract and fractions with these contents, but it is speculated that these phenolic compound components indirectly affected the antioxidant activity. In particular, total flavonoid content and antioxidant activity tended to be similar.

Antioxidant activity tends to be similar to the total phenolic content than individual phenolic compound content. Thus, total phenolic content and total flavonoid content were associated with the antioxidant activity of extracts and fractions.

In conclusion, the roots EtOAc fraction had the best antioxidant activity.

Table 4 shows the results of phenolic compound using HPLC for the Hallasan wormwood extract and fractions.

Nine phenolic acid and 1 flavonoid were detected aerial parts from Hallasan wormwood. The main phenolic acids were chlorogenic acid, caffeic acid, and rutin were the main components of flavonoids.

The content of chlorogenic acid, caffeic acid, rutin, and taxifolin in the EtOAc fractions is higher than that of other fractions, and the components belonging to p-coumaric acid, ferulic acid, sinapinic acid and benzoic acid groups are also included.

Five phenolic acid and 1 flavonoid were detected roots from Hallasan wormwood. The main phenolic acids were chlorogenic acid and caffeic acid.

The EtOAc fractional extract with the highest phenolic compound content contains p-coumaric acid and ferrulic acid. However, it was not detected in other root fraction extracts. The components with the highest content in the analyzed phenolic compounds were chlorogenic acid and caffeic acid. Caffeic acid had the high hydrogen donating capacity toward the DPPH radical (von Gadow et al., 1997).

Chlorogenic acid and caffeic acid are components that are related to each other. Caffeic acid has a stronger antioxidant activity than that of chlorogenic acid and chlorogenic acid is hydrolyzed into caffeic acid in the intestine (Sato et al., 2011). Therefore, ingestion of Hallasan wormwood fractional extract with high chlorogenic acid content predicts its potential health benefits.

There was a difference in the phenolic compound content analyzed in the aerial parts and roots of Hallasan wormwood. Jang and Kang (2019) analyzed the phenolic acid and flavonoid composition and content of A. princeps var orientalis extract. As for the phenolic acid, chlorogenic acid was detected in the leaves and stems, and p-coumaric acid in the roots was contained. And the flavonoid found taxifolin on the leaves and stems and quercetin on the roots. Chlorogenic acid, p-coumaric acid, and rutin detected in the present study results were confirmed in Hallasan wormwood. However, quercetin was not detected.

This is probably because Hallasan wormwood is taxonomically another variety. Ryu et al. (2011) reported the contents of 12 phenolic acids (gallic acid, protocatechuic acid, gentisic acid, p-hydroxybenzoic acid, vanillic acid, caffeic acid, chlorogenic acid, salicylic acid, p-coumaric acid, sinapinic acid, ferulic acid, and t-cinnamic acid) and 3 flavonoids (rutin, quercetin and, kaempferol) contained in A. annua leaf and stem extracts. It had a difference from the phenolic compound of A. annua extract.

Seong et al. (2015) collected Perilla frutescens from China, Japan and South Korea and analyzed and compared phenolic compounds. As a result, it was found that the phenol content and composition ratio differed depending on the species collected by Perilla frutescens. This suggests that various Artemisia spp. phenolic compound studies are needed.

The phenolic compounds content analyzed by HPLC and antioxidant activity in this study did not appear to have a significant effect. However, the extract having a high total flavonoid content with the total phenolic content had a high antioxidant activity. Therefore, there will be main phenolic compounds that affect unanalyzed antioxidants, except for the detected phenolic acids and flavonoids.

From the results presented in this study, the EtOAc fractions of Hallasan wormwood root have good antioxidant activities. Total phenolic and total flavonoid contents are the predominant constituents that correlate with antioxidant activity. This study highlights the antioxidant activities of the phenolic constituents of Hallasan wormwood.

Use of Hallasan wormwood extract and fractions should take into account the toxicity. And not all phenolic compounds reported to date have been analyzed in this study. Therefore, for these plants to be useful as therapeutic agents, and toxicity studies of the extract and fractions are necessary. Also, additional investigation of unknown substances that have not been analyzed is required.