Phytochemical Composition, Antioxidant Activity, and Neuroprotective Effect of Terminalia chebula Retzius Extracts

The objectives of this study were to determine phytochemical compositions, chemiluminescence antioxidant activities, and neuroprotective effects on PC12 cells for water, methanol, and 95% ethanol extracts of the air-dried fruit of Terminalia chebula Retzius. The water extract afforded the greatest yield, and total phenolic and tannin content. The methanol extract yielded the greatest total triterpenoid content. Based on four chemiluminescence antioxidant assays, the three extracts showed various degrees of antioxidant activity. The methanol extract showed good antioxidant activity based on the horseradish peroxidase-luminol-hydrogen peroxide (H2O2) assay. The water extract appeared to have good antioxidant activities in cupric sulfate-Phen-Vc-H2O2 and luminol-H2O2 assays. Pyrogallol-luminol assay showed the 95% ethanol extract to have good antioxidant activity. The methanol and water extracts presented neuroprotective activities on H2O2-induced PC12 cell death at 0.5–5.0 μg/mL. Further investigations are necessary to verify these activities in vivo.


Introduction
The auto-oxidation of lipids generates reactive oxygen species (ROS) such as superoxide anion radicals ( • O 2 − ), hydroxyl radicals ( • OH), and hydrogen peroxide (H 2 O 2 ) [1]. Excessive production of ROS is implicated in ageing as well as in many diseases, including atherosclerosis, cancer, and inflammatory disease [2]. Antioxidants are very important for human health, and thus antioxidant supplementation is recommended to provide cellular protection from the deleterious effects of excessive ROS concentrations [3]. In this study, we applied chemiluminescence techniques to assay three Terminalia chebula Retzius extracts for antioxidant activity. Chemiluminescence antioxidant assay is a simple, direct, and effective method well suited to free radicals and antioxidant study [4].
T. chebula is a traditional medicine belonging to the genus Terminalia, family Combretaceae, and is extensively cultivated in Taiwan. The dried ripe fruit of T. chebula is an important Indian herb used extensively in the indigenous system of medicine (ayurvedic) for its homeostatic, antitussive, laxative, diuretic, and cardiotonic activities [5].

Methanol, Water
, and 95% Ethanol Extraction. T. chebula was pulverized into fine powder using a stainless steel blender (Waring Commercial, Torrington, Conn, U.S.A.). Two-gram aliquots of the dried powder were each extracted three times with methanol (20 mL), deionized water (20 mL), and 95% ethanol (20 mL). The mixtures were agitated in an ultrasonic cleaner (model DC200H, Chemist Scientific Corporation, Taipei, Taiwan) for 15 min at room temperature then filtered. The methanol, deionized water, and 95% ethanol filtrates were individually pooled and each solvent removed at 40 • C, under reduced pressure by rotary evaporator (Rotavapor R210, Buchi, Postfach, Flawil, Switzerland). Finally, each extract was dried overnight in a freeze dryer (model FD3-12P-80 • C, Kingmech Corporation, Taipei, Taiwan) before calculating the yield of each extract. All of the dried extracts were brown solids and were stored at −20 • C prior to phytochemical composition analyses and bioassays.

Total Phenolic Content.
The total phenolic contents of the three extracts were determined using the Folin-Ciocalteu assay [16]. Briefly, 0.6 mg samples of each of the three extracts were dissolved into methanol (1 mL), deionized water (1 mL), and 95% ethanol (1 mL), respectively, and then 11.4 μL aliquots of each of these solutions were mixed with Na 2 CO 3 (2%, 227.32 μL). The mixtures were stood at room temperature for 2 min before the addition of Folin-Ciocalteau reagent (50%, 11.4 μL) to each sample mixture. After incubation at room temperature for 30 min, the absorbances of the reaction mixtures were measured at 750 nm using a tunable microplate reader (VersaMax, Molecular Devices Corporation, Sunnyvale, Calif, U.S.A.). Gallic acid (0.2-1.0 mg/mL in methanol) was used as a standard, and the total phenolic contents of three extracts were expressed in milligram gallic acid equivalents (mg gallic acid/g extract).

Total Triterpenoid Content.
After optimizing all experimental parameters, total triterpenoid content was determined by colorimetry using the following procedure [17]. Briefly, 10 mg of each of the three extracts was individually dissolved in 1 mL of methanol, deionized water, and 95% ethanol. Then, 100 μL of each of these solutions was mixed with vanillin/glacial acetic acid (150 μL, 5% w/v) and perchloric acid solution (500 μL). The sample solutions were heated for 45 min at 60 • C and then cooled in an icewater bath to the ambient temperature. After the addition of glacial acetic acid (2.25 mL), each sample solution's absorbance was measured at 548 nm, using a UV-visiblelight recording spectrophotometer (UV-160 A; Shimadzu Corporation, Kyoto, Japan). Ursolic acid (0.025-0.5 mg/mL in methanol) was used as a standard. Results were expressed as milligram ursolic acid equivalents (mg ursolic acid/g extract).

Total Tannin Content.
Analysis of total tannin was based on a titrimetric method [18]. Zinc ion reacts with tannin compounds in alkali solution, to form complexes. Residual zinc ion is then titrated with EDTA, and zinc complexed tannin is determined from EDTA consumption and total zinc content. The methanol, water, and 95% ethanol extracts (1 mg each) were each placed into glass vials and dissolved with 1 mL of deionized water. The vials were warmed in a water bath for 5 min at 35 ± 2 • C. ZnAc (1 M, 0.4 mL) and NH 3 (0.28 mL) were mixed together, and the warmed 1-mL extract solutions were added. The solutions were replaced in the water bath for 30 min at 35 ± 2 • C. Deionized water (8.92 mL) was added to make the final volume up to 10.6 mL. After careful filtering, sample solutions were obtained. The solutions (0.8 mL) were further diluted with 5.2 mL of deionized water, and 0.5 mL of NH 3 -NH 4 Cl buffer (pH 10) was added. Finally, the mixture was titrated with 0.05 M EDTA. The blank was detected without addition of the extract. The total tannin content (%/mg extract) of each extract was calculated as follows: . V blank and V extract represent the EDTA titration volumes (mL) recorded for the respective blank and extract solutions. W extract represents the weight (mg) of each extract [19].   15% v/v) was added to yield a final volume of 237.5 μL. The background was detected without the addition of H 2 O 2 solution. Chemiluminescence was measured for 10 min at 25 • C with a microplate luminometer. The scavenging activity ratio (%) of each sample was calculated using the formula described for the pyrogallol-luminol system [21]. 2.10. Protective Effect. PC12 cells were plated on poly-L-lysine hydrobromide-coated 100 mm cell culture dishes and grown in DMEM, supplemented with 10% HS, 1% FBS, a mixture of 1% penicillin/streptomycin, and 1% Lglutamine at 37 • C in a 95% humidified air-5% CO 2 chamber. Cells were subcultured for up to ten passages. Cellular viability was determined using the trypan blue exclusion test. Only cell preparations with 95% or greater viability were used. PC12 cells were seeded in poly-L-lysine hydrobromidecoated 24-well cell culture plates (1.25 × 10 5 cells/well) with complete DMEM for 24 h. H 2 O 2 was added to induce PC12 cell death. To study the protective effect of test samples on the PC12 cells, we renewed the media and preincubated the cells for 12 h, either with or without the presence of 500 μL test samples, to obtain sample concentrations of 0, 0.5, 2.5, and 5.0 μg/mL. Thereafter, the media was replaced, and 500 μL H 2 O 2 was added to a concentration of 40 μM H 2 O 2 , and the mixture was incubated for an additional 12 h. The control was incubated without the addition of either sample or H 2 O 2 solution.
All extracts were dissolved in DMSO. The concentration of DMSO in the final culture medium was 0.5%, which had no observable effect on cell viability as determined by MTT reduction assay, following Choi et al.'s method [22] with slight modifications. Upon completion of incubation, MTT solution (50 μL, 1.0 mg/mL) was added to the culture medium, and the cells were incubated for 2 h at 37 • C. The medium was then removed, and 300 μL of DMSO was added to the well to dissolve the formazan, derived from live cell mitochondrial cleavage of the tetrazolium ring. The formazan solutions were incubated for 30 min at 25 • C, and formazan solutions (250 μL) were placed in 96-well plates. The amount of MTT formazan product was determined by measuring optical density (OD) with a tunable microplate reader at a test wavelength of 570 nm and a reference wavelength of 655 nm. MTT reduction was calculated as OD 570 nm − OD 655 nm . Cell viability (%) was determined as (OD sample /OD control ) × 100%. Where OD sample represents OD sample (570 nm) − OD sample (655 nm), and OD control represents the mean value of OD control (570 nm) − OD control (655 nm) . The control cells exhibited 100% cell viability. Gallic acid and AC-DEVD-CHO were the positive controls. Gallic acid exists in T. chebula, and AC-DEVD-CHO is a caspase 3 inhibitor.
2.11. Statistical Analysis. The IC 50 value (the concentration of a sample that is required for 50% inhibition in vitro) was determined using linear regression. Each phytochemical characteristic and chemiluminescence antioxidant activity was determined three times, using the same extract in order to determine reproducibility and to provide a mean ± standard deviation (SD) using Microsoft Excel 2003. Cell viability was measured by MTT reduction assay. Cell viability (%) represents three replicates per treatment. For each sample's "protective effect" against H 2 O 2 effects on PC12 cells, only data concerning PC12 cell group exposure to H 2 O 2 was considered. All data were represented as means ± SD based on triplicate determinations. Data were analyzed for statistical significance using one-way ANOVA, followed by Tukey's test as a post-hoc test with SPSS software (SPSS for Windows, Version 10).

Extraction Yield, and Total Phenolic, Triterpenoid, and Tannin Content of Methanol, Water, and 95% Ethanol
Extracts. Table 1 presents the yield, and total phenolic, triterpenoid, and tannin content of the three extracts. The yield of the three extracts varied from 21.7% to 39.4%. The polarity (δP) of methanol, water, and ethanol lays in the following descending order: water (16.0) > methanol (12.3) > ethanol (8.8). The water extract had the greatest yield of the three extracts. The lowest extraction yield of the 95% ethanol extract may be due to the high δP of the major components of T. chebula.
Total phenolic content in the three extracts was determined from a linear gallic acid standard curve. The total phenolic content of the three extracts varied from 867.2 to 1041.8 mg gallic acid/g extract. This result suggests that the water extract provided the greatest concentration of phenolic compounds of the three extracts. The total triterpenoid content of the three extracts was evaluated by colorimetry, using ursolic acid as the standard. The total triterpenoid content of the three extracts varied widely from 0.8 to 4.2 mg ursolic acid/g extract. The lowest total triterpenoid content occurred for the water extract, whereas the methanol extract provided the greatest triterpenoid content of the group. The triterpenoid is a low polarity compound, and nine oleananetype triterpenoids were isolated from the methanol extract of T. chebula [23]. The total tannin content of the three extracts varied from 33.9 to 40.3%/mg extract. The greatest total tannin content was detected in the water extract. Finally, these results show that the water extract produced a higher total phenolic and tannin content than the methanol and 95% ethanol extracts did. This may be the cause of the best extraction yield seen for the T. chebula water extract.

Antioxidant Activity of the HRP-Luminol-H
The antioxidant activity of three extracts was evaluated using a chemiluminescence assay method ( Table 2). The scavenging effect of luminol radicals was observed upon addition of the three extracts. The antioxidant IC 50 values for the three extracts ranged from 6.9 to 42.4 μg/mL. The antioxidant abilities of the three extracts compared with vitamin C and trolox were in the order trolox > methanol extract > vitamin C > 95% ethanol extract > water extract. However, other workers using a different chemiluminescence method reported the scavenging activity of luminol radicals by the T. chebula water extract to be strong [24]. This finding shows that the high antioxidant activity observed for the methanol extract is a consequence of its total triterpenoid content.

Antioxidant Activity of the Pyrogallol-Luminol System.
Pyrogallol was autoxidized under the alkaline condition to generate the • O 2 − radical ion, and the scavenging effect of this radical was observed on addition of each of the three extracts. The IC 50 values for the three extracts' antioxidant activity ranged from 146.0 to 235.5 μg/mL ( Table 2). The antioxidant properties of the three extracts compared with vitamin C and trolox are in the descending order of 95% ethanol extract > methanol extract > trolox > water extract > vitamin C. Another research group, using a different method [6,25], reported the IC 50 value of the • O 2 − radical ion's scavenging activity for 70% methanol, methanol, and water extracts of T. chebula as 13.4, 480.0, and 730.0 μg/mL, respectively. These findings show that each of the three extracts exhibit different antioxidant abilities. The greater scavenging effect of the • O 2 − radical ion by the 95% ethanol extract relates to the extract's total triterpenoid content. The good • O 2 − scavenging effect of the methanol extract results from the total triterpenoid and tannin content.

Antioxidant Activity of the CuSO 4 -Phen-Vc-H
The • OH scavenging effect was observed by adding each of the three extracts. The IC 50 values for the three extract's antioxidant activity ranged from 4.5 to 6.7 μg/mL ( Table 2). The antioxidant properties of the three extracts compared with vitamin C and trolox are in the decreasing order of water extract > trolox > methanol extract > 95% ethanol extract > vitamin C. Compared to another published method, the IC 50 value of the • OH scavenging activity for 70% methanol extract of T. chebula was reported to be 72.0 μg/mL [25]. This finding shows that the three extracts had different • OH scavenging activities. The higher • OH scavenging activity of the water extract is related to the amount of total phenolic and tannin content.  The data are presented as mean ± SD for three replicates. a % w/w. b mg gallic acid/g extract. c mg ursolic acid/g extract. d %/mg extract.  extracts, against both vitamin C and trolox, were in the order of trolox > water extract > methanol extract > 95% ethanol extract > vitamin C. Compared to another method, the IC 50 value of the H 2 O 2 scavenging activity for 70% methanol extract of T. chebula was found to be much higher than can be represented [25]. Thus, all three extracts were able to scavenge H 2 O 2 . The higher antioxidant activity of the water extract arises from the amount of total phenolic and tannin content.

Antioxidant Activity of the
3.6. Protective Effect. In order to ascertain the neuroprotective effects of the T. chebula extracts, we applied an in vitro "Inhibition of H 2 O 2 -induced PC12 cell death" model to estimate the neuroprotective effect. Application of H 2 O 2 induced a dose-dependent loss in viability of PC12 cells. The cell viability decreased to 12.2% when exposed to 80 μM H 2 O 2 for 12 h, as shown in Figure 1. In order to obtain a detectable effect, cells were treated with 40 μM H 2 O 2 for 12 h, causing cell viability to decrease to 47.3%. Table 3 shows the results of this protective study; we found that cell viability decreased to 59.0 ± 5.1% when exposed to H 2 O 2 (40 μM) for 12 h. The methanol and water extracts inhibited H 2 O 2 -induced cytotoxicity at 0.5-5.0 μg/mL. The water extract showed the greatest neuroprotective activity among the three extracts. The data of water extract is two times bigger than that of control because it alone did not show any cytotoxicity at 0.1-20.0 μg/mL [15]. Meanwhile, the total time of this protective experiment was 24 h, and the water extract could induce cell proliferation at 0.5-5.0 μg/mL for up to 72 h [15]. The result of the protective study did not show any cytotoxicity and neuroprotection when PC12 cells were pretreated with the water extract at 0.  [26][27][28]. Based on the results of this study and findings of our previous report [29], we propose that the effective neuroprotective activity of the water extract is a consequence 6 Evidence-Based Complementary and Alternative Medicine

Conclusions
We demonstrated and compared for the first time the phytochemical compositions, chemiluminescence antioxidant activities, and neuroprotective effects of water, methanol, and 95% ethanol extracts of the air-dried fruit of T. chebula Retzius. The three extracts of T. chebula were the richest of phenolic compounds compared with total triterpenoid and tannin content as presented in Table 1. The methanol extract had the greatest total triterpenoid content and exhibited good antioxidant activity in the HRP-luminol-H 2 O 2 assay. The water extract appeared to have the greatest total phenolic and tannin content and showed good antioxidant activities in both CuSO 4 -Phen-Vc-H 2 O 2 and luminol-H 2 O 2 assays. The 95% ethanol extract exhibited good antioxidant activity in the pyrogallol-luminol assay. Thus, the three extracts present various levels of ROS scavenging efficiency due to differences between the mechanisms of the four ROS chemiluminescence systems. The three extracts are new potential sources of natural antioxidants for food and nutraceutical products. The methanol and water extracts exhibit neuroprotective activities against H 2 O 2 -induced toxicity toward PC12 cells and are potential candidates for the treatment of H 2 O 2induced neurodegenerative disease. Further investigations are required to elucidate the exact mechanisms in vivo that these candidates operate by and to isolate and purify the major compound of each of the extracts.