C. cochinchinensis samples were collected from the mountain areas from Taiwan region, air dried, and then kept in a cool place for further use. Liquiritin was purchased from ChromaDex (Santa Ana, CA, USA). Methanol, acetonitrile, ethyl acetate (EtOAc), and n-hexane were purchased from Merck (Darmstadt, Germany). DPPH (1, 1-diphenyl-2-picrylhydrazyl), potassium ferricyanide (III) [K₃Fe(CN)₆], potassium dihydrogenphosphate, dipotassium hydrogenphosphate, phosphoric acid, trichloroacetic acid, iron (III) chloride, butylated hydroxyanisole (BHA), and ascorbic acid were purchased from Sigma (St. Louis, MO, USA). L-3,4-dihydroxyphenylalanine (L-DOPA), kojic acid, and dimethyl sulfoxide (DMSO) were purchased from Acros Organics (Fair Lawn, NJ, USA). Sodium phosphate dibasic anhydrous was purchased from J. T. Baker (Petaling Jaya, Selangor, Malaysia).

A sample of pulverized C. cochinchinensis (1.0 g) was sonicated in an ultrasonic bath (Chrom Tech, Taipei, Taiwan region) for 20 min with 7 mL of 70% methanol (methanol/H₂O = 7/3, v/v). The suspension was centrifuged at 6000 rpm (HERMLE Z206A, Germany) for 10 min. The supernatant was collected and run through a 0.45 μm filter. The residual solids were extracted with fresh 70% methanol. After the C. cochinchinensis sample was extracted three times, all the collected supernatants were mixed together. Subsequently, 70% methanol was used to make up the total volume to 20 mL. Moreover, 1.0 mg liquiritin was dissolved in 10 mL of 70% methanol and used as the internal standard (IS) solution. Before performing the HPLC analysis, 100 μL of the extracted solution was mixed with 100 μL of the IS solution.

Tetrahydroxyflavanonol (THF) was isolated from C. cochinchinensis stem by Chen et al. [18] , based on the modified method described by Kobayashi et al. [19] . Dried C. cochinchinensis samples were extracted four times by methanol under reflux. The extract was partitioned using a mixture of EtOAc and water (1:1, v/v). The EtOAc extract was run through a silica gel column, and then eluted with a mixture of n-hexane and EtOAc/MeOH. Eight fractions were collected during elution. Fraction number 3 was further separated by high performance liquid chromatography (HPLC) using a Cosmosil 5C18-AR column (Nacalai Tesque, Tokyo, Japan) to obtain THF.

HPLC analysis was performed on an Agilent 1200 system with a reverse phase column (Cosmosil 5C18-AR II, 5 μm, 25 cm × 4.6 mm I.D.; Nacalai Tesque, Kyoto, Japan). The detection wavelength was set at 254 nm. The flow rate was 0.8 mL/min with a linear solvent gradient of A-B (A = 10 mM KH₂PO₄, pH 4.6; B = CH₃CN/CH₃OH/H₂O, 1.5/2.5/1, v/v/v). as follows: 0 min, 40% B; 10 min, 40% B; 20 min, 60% B; 30 min, 70% B; and 50 min, 100% B.

The total phenolic content was measured following the method described by Singleton et al. [20] . An amount of 200 μL of different concentrations of samples was mixed with 200 μL of 0.5 N Folin-Ciocalteu reagent, to which 200 μL of 10% (w/v) Na₂CO₃ and 40 μL of distilled water were added. The mixture was incubated at room temperature for 1 h in the dark. After incubation, the mixture was centrifuged at 5000 rpm for 10 min. An amount of 200 μL of the supernatant was transferred to a 96-well plate and the absorbance of each well was measured using an ELISA reader at a wavelength of 700 nm. Gallic acid was used as a positive control. Each measurement was performed at least in duplicate.

The flavonoid content was measured according to the method described by Chandra et al. [21] . Different concentrations of samples (50 μL) were mixed with 100 μL of 10% (w/v) AlCl₃. The mixture was incubated at room temperature for 10 min in the dark. The absorbance of the mixture at 430 nm wavelength was measured using an ELISA reader. Quercetin was used as a positive control. Each measurement was performed at least in duplicate.

Radical scavenging activities of THF marker standards and C. cochinchinensis extracts were measured respectively using the methods of Singh and Rajini and Chan et al. and Azman et al. [22] [23] [24] . The radical scavenging activity of ascorbic acid, used as a positive control, was also measured. The sample (50 μL) was mixed with 50 μL of freshly prepared 160 μM DPPH in ethanol. The mixture was kept in the dark for 30 min. The absorbance of the mixture at 517 nm wavelength was measured using an ELISA reader (TECAN^R, Austria). Each measurement was performed at least in duplicate. The radical scavenging activity was calculated as follows:

$\text{DPPH radical scavenging activity}\left(\%\right)=\left(1-\frac{A_{\text{Sample}}}{A_{\text{Blank}}}\right)\times 100\%$ (2)

where A_Sample and A_Blank represent the absorbance of sample and blank solution, respectively.

The reducing ability of the samples was measured following the method described by Canabady-Rochelle et al. [25] . Samples of different concentrations (100 μL each) were individually mixed with 100 μL of 1% (w/v) K₃Fe(CN)₆ and 100 μL of 2 mM phosphate buffer (pH 6.6). The mixture was incubated at 50˚C for 20 min. After incubation, 100 μL of 10% (w/v) trichloroacetic acid was added to it, and the mixture was centrifuged at 3000 rpm for 2 min. An amount of 100 μL of the supernatant was transferred to a 96-well plate. Each well contained 100 μL of distilled water and 20 μL of 0.1% (w/v) FeCl₃ solution. BHA was used as a positive control. The absorbance of each well was measured using an ELISA reader at 700 nm wavelength. Each measurement was performed at least in duplicate.

An amount of 20 μL of extracted C. cochinchinensis pith solution (500 μg/mL, in 3.3% of DMSO) was placed in a 96-well plate. Then, 40 μL of tyrosinase solutions of various concentrations (0.277, 0.554, 1.662, 3.324, and 6.648 μg/mL) and 0.1 mM of L-DOPA solution (dissolved in a sodium phosphate buffer at pH 6.8) were added to it.

Another 20 μL of extracted C. cochinchinensis pith solution (31.25, 62.5, 125, 250, and 500 μg/mL, in 3.3% of DMSO) was placed in a 96-well plate, to which 40 μL of tyrosinase solution (6.648 μg/mL) and 0.1 mM of L-DOPA solution (dissolved in a sodium phosphate buffer at pH 6.8) were added. These mixed solutions were kept at room temperature (25˚C) for 25 min. The absorbance was measured at 475 nm [12] [26] using the Microplate-Reader (Sunrise Basic, Grödig, Austria). Kojic acid was used as a positive control. The tyrosinase inhibition rate (%) was calculated from the following equation:

$\text{The inhibition rate}\left(\%\right)=\left(1-\frac{{\text{OD}}_{\text{sample}}}{{\text{OD}}_{\text{control}}}\right)\times 100\%$ (3)

The absorbance of sample (OD_sample) and control (OD_control) was measured at 475 nm. The IC₅₀ value was determined by regression of a constructing dose- response curve at which 50% target activity was lost.

In a 96-well plate, 20 μL of extracted C. cochinchinensis pith solution (7.8125, 15.625, 31.25, and 62.5 μg/mL, in 3.3% of DMSO) was placed, to which 40 μL of tyrosinase solutions (6.648 μg/mL) were added. The substrate was L-DOPA solution, which was prepared by dissolving L-DOPA (0.1, 0.3, 0.5, 0.7, and 1.0 mM) in sodium phosphate buffer at pH 6.8. The Line weaver-Burk plot was obtained by plotting the inverse values of reaction rate (V) and concentration of L-DOPA (Equation (1)).

Statistical evaluation was performed by running one-way analysis of variance (ANOVA) with SAS^R software (version 6.08, SAS Institute Inc., Cary, NC, USA). All data were presented as mean ± standard deviation (SD). Differences were considered to be statistically significant when the p-value was less than 0.05.

The total phenolic content of C. cochinchinensis extracts is shown in Figure 1(a) . The pith extract contained more phenolic components than the xylem and bark extracts. No significant difference was observed in the total phenolic content between the bark and xylem extracts. Flavonoid contents of C. cochinchinensis extracts are shown in Figure 1(b) , which were in the order pith > xylem > bark. Thus, the pith portion of C. cochinchinensis stem was expected to have better antioxidant activity.

The DPPH radical scavenging activities of the extracts were measured as the decrease of absorbance at a wavelength of 517 nm, and the results are shown in Figure 1(c) . The DPPH radical scavenging activity of the THF standard is shown in Figure 1(d) . The IC₅₀ value of the THF standard was 0.122 mg/mL and those of the pith, xylem, and bark extracts were 0.769, 2.809, and 3.34 mg/mL, respectively. The pith portion of C. cochinchinensis stem showed better DPPH radical scavenging activity. The reducing ability was measured as the change in

absorbance at a wavelength of 700 nm. Figure 1(e) and Figure 1(f) show the reducing ability of C. cochinchinensis extracts and the THF standard, respectively. Generally, higher flavonoid contents result in better reducing ability. On the other hand, better reducing ability represents stronger antioxidant activity.

Figures 2(a)-(c) show the representative HPLC chromatograms of C. cochinchinensis stem extracts from the bark, xylem, and pith portions respectively. THF was well separated by HPLC with a retention time of 15 min. Based

on the chromatographic analysis results, THF was found only in xylem and pith extracts of C. cochinchinensis (691.5 ± 3.2 and 1,489.7 ± 5.5 μg/g, respectively). The THF content of bark extract was not detectable. The pith portion of C. cochinchinensis stem contained more THF than the xylem portion. Therefore, the pith extracts showed better radical scavenging activity and reducing ability. Although xylem and bark extracts had similar antioxidant results, only xylem extracts were found to contain THF. Bark extracts may contain some other ingredients which may also provide antioxidant activity. More studies are required to explain why bark extracts had similar antioxidant activity without containing THF.

Derivatives of both flavonol and flavanone have been approved as good antioxidants [27] . Flavonol derivatives such as THF and quercetin have similar chemical structures ( Figure 3 ). Multiple hydroxyl groups, especially on the B-ring, improve the antioxidant activity of flavonoids [28] . Likewise, flavanone derivatives such as tetrahydroxyflavanone and luteolin also have similar chemical structures ( Figure 3 ). The difference between THF and tetrahydroxyflavanone is that THF has one extra hydroxyl group on the B-ring. Consequently, THF has better antioxidant activity [28] . Because the pith portion of C. cochinchinensis extracts contained more THF, pith extracts had a better reducing ability than other extracts. Likewise, the bark portion contained the least THF and had the worst reducing ability.

Pith extracts of C. cochinchinensis showed the ability to inhibit the formation of DOPA chrome, which can be detected with a spectrophotometer at a wavelength of 475 nm. When 0.1 mM of L-DOPA was used as the substrate, tyrosinase activity was increased with the addition of more tyrosinase. A linear relationship of the first-order was observed between the tyrosinase activity and tyrosine concentration ( Figure 4 ). However, tyrosinase activity reduced with the addition of

more C. cochinchinensis pith extracts. This confirmed that the extracted C. cochinchinensis pith solution could inhibit tyrosinase activity. In addition, the slope of the linear lines decreased with the addition of more C. cochinchinensis pith extracts ( Figure 4 ).

The ability of C. cochinchinensis extracts to inhibit tyrosinase activity could be attributed to the presence of phenolics in the extracts. The IC₅₀ value (36.3 μg/mL) of ethanol extracted C. cochinchinensis stem solution was reported by Zheng et al. [8] . In this study, the pith portion of C. cochinchinensis stem was further extracted with methanol. Figure 5(a) shows the inhibition rate of tyrosinase activity using C. cochinchinensis pith extracts, which reduced the tyrosinase activity in a dose-dependent manner. The slope and intercept of the linear regression line were 0.5015 and 41.914, respectively. The IC₅₀ value of methanol extracted C. cochinchinensis pith solution was calculated to be 16.1 μg/mL. Comparing this result with the results reported by Zheng et al., C. cochinchinensis pith extracts were found to exhibit a better inhibitory ability. Figure 5(b) shows a comparison of tyrosinase inhibitory rate between C. cochinchinensis pith extracts and kojic acid, which was used as a positive control. Based on Figure 5(b) , the inhibition rate of tyrosinase activity was 70.4% when 250 μg/mL of C. cochinchinensis pith extracts were added. This inhibition rate was close to that of kojic acid (70.5%), at a concentration of 62.5 μg/mL. Although the inhibitory ability of C. cochinchinensis pith extracts was approximately 25% of that of kojic acid, C. cochinchinensis extracts are natural ingredients and may possibly be used in cosmetic products.

In this study, it was required to determine whether the inhibitory activity of C. cochinchinensis pith extracts was competitive or noncompetitive. Figure 6

shows the Line weaver-Burk double reciprocal plot of C. cochinchinensis pith solutions. The substrate was L-DOPA. Based on Figure 6 , the x-intercept (−1/K_m) remained the same but the y-intercept (1/V_max) increased with increasing concentrations of C. cochinchinensis pith extracts. As a result, K_m remained unchanged (0.23 mM), but V_max decreased by the introduction of an inhibitor. The binding of C. cochinchinensis pith extracts to tyrosinase had no effect on

the binding of L-DOPA to tyrosinase. The binding sites of L-DOPA and C. cochinchinensis pith extracts to tyrosinase were different. Based on the Lineweaver-Burk double reciprocal plot shown in Figure 6 , the inhibitory activity was determined to be noncompetitive.