Antioxidant and Cytoprotective Effects of Tibetan Tea and Its Phenolic Components

Tibetan tea (Kangzhuan) is an essential beverage of the Tibetan people. In this study, a lyophilized aqueous extract of Tibetan tea (LATT) was prepared and analyzed by HPLC. The results suggested that there were at least five phenolic components, including gallic acid, and four catechins (i.e., (+)-catechin, (−)-catechin gallate (CG), (−)-epicatechin gallate (ECG), and (−)-epigallocatechin gallate). Gallic acid, the four catechins, and LATT were then comparatively investigated by four antioxidant assays: ferric reducing antioxidant power, 2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl 3-oxide radical (PTIO•) scavenging, 1,1-diphenyl-2-picryl-hydrazl radical scavenging, and 2,2′-azino-bis(3-ethylbenzo-thiazoline-6-sulfonic acid) radical scavenging assays. In these assays, LATT, along with the five phenolic components, increased their antioxidant effects in a concentration-dependent manner; however, the half maximal scavenging concentrations of ECG were always lower than those of CG. Gallic acid and the four catechins were also suggested to chelate Fe2+ based on UV-visible spectral analysis. Ultra-performance liquid chromatography coupled with electrospray ionization quadrupole time-of-flight tandem mass spectrometry (UPLC−ESI−Q−TOF−MS/MS) analysis suggested that, when mixed with PTIO•, the five phenolic components could yield two types of radical adduct formation (RAF) products (i.e., tea phenolic dimers and tea phenolic-PTIO• adducts). In a flow cytometry assay, (+)-catechin and LATT was observed to have a cytoprotective effect towards oxidative-stressed bone marrow-derived mesenchymal stem cells. Based on this evidence, we concluded that LATT possesses antioxidative or cytoprotective properties. These effects may mainly be attributed to the presence of phenolic components, including gallic acid and the four catechins. These phenolic components may undergo electron transfer, H+-transfer, and Fe2+-chelating pathways to exhibit antioxidative or cytoprotective effects. In these effects, two diastereoisomeric CG and ECG showed differences to which a steric effect from the 2-carbon may contribute. Phenolic component decay may cause RAF in the antioxidant process.


Introduction
Tibetan tea ("Zangcha" in Chinese, Figure S1), a member of the fully fermented tea (dark tea) family, has a history of approximately 1000 years in China.It has become an essential beverage for millions of Tibetan people.The Tibetan Plateau varies in altitude between 3000 and 5000 meters and is not well-suited for cultivating vegetables, fruit, or trees.Thus, Tibetan people eat various meat products, such as beef, milk, and butter tea.These high-protein, high-lipid foods can effectively help them to resist the severe cold of high altitude.However, this diet may increase the risk of cardiovascular and indigestion diseases.
It has been reported that Tibetan tea can effectively lower blood pressure, remove blood lipids, and reduce the generation of atherosclerosis [1,2].Intake of Tibetan tea is also hypothesized to blunt the effects of high-altitude hypoxia [3].This tea is indispensable for Tibetan life and is thus called "Tibetan tea" by the Chinese, even though the tea is actually manufactured in Sichuan (a Han nationality Province in China).The tea is processed through five main working procedures with 30 processing technologies; these technologies were listed as Chinese Intangible Cultural Heritage in 2008 [4].The whole process takes approximately 6 months, during which some chemical components have possibly degraded and some fungal populations have also been altered [5].All of these give Tibetan tea a characteristic taste and potential beneficial effects.
Nevertheless, as a traditional tea with ethnological characteristics, Tibetan tea has received very little interest from scientists.In the PubMed database, only three papers about Tibetan tea can be retrieved from 1991 to 2016 [3,6,7].
According to their working procedures and materials, Tibetan tea products are classified into three main types: Kangzhuan, Jinjian, and Kangjian.The most popular type is Kangzhuan, i.e., brick tea.The present study attempted to use Kangzhuan as a representative to estimate the antioxidant or cytoprotective effects of Tibetan tea and examined the possible mechanisms.

Results and Discussion
In this study, the prepared LATT was analyzed by the HPLC method, and at least five phenolic components were found, including gallic acid (GA) and four catechins [(+)-catechin and its galloylated derivatives CG, ECG, and EGCG, Figure 1].The contents of GA, (+)-catechin, EGCG, ECG, and CG were calculated as 0.44 ± 0.02 g/100 g, 0.11 ± 0.01 g/100 g, 0.24 ± 0.01 g/100 g, 0.13 ± 0.01 g/100 g, and 0.04 ± 0.01 g/100 g, respectively.Each of their contents in LATT was lower than in green tea [8].The difference was reported to be due to the comprehensive working protocols (especially fermentation [9]) and production seasons [10].In the present study, their lower levels may also be related to the sample preparation.Our preliminary experiment using distilled water to dissolve LATT found the huge peaks of caffeine and gallic acid to severely interfere with the determination of other peaks.Thus, we had to use methanol to prepare LATT solution for determination.Methanol however cannot fully extract the phenolic components from a lyophilized aqueous extract.This also led to a lower level of caffeine (0.72 ± 0.01 g/100 g).
Our previous study demonstrated that the antioxidant capacity of plants can be mainly attributed to the existence of total phenolics [11].In the present experiments, five phenolic components and LATT were comparatively evaluated using various antioxidant assays, including the ferric reducing antioxidant power (FRAP), PTIO•-scavenging, ABTS + •-scavenging, and DPPH•-scavenging assays.The half maximal scavenging concentration (SC 50 ) values in these assays revealed that the five phenolic components always exhibited stronger antioxidant abilities than LATT itself (Table 1).On the basis of this evidence and previous reports [12,13], it can be assumed that these phenolic components may be responsible for the antioxidant effects of LATT.This is consistent with data from the flow cytometry assay, in which (+)-catechin (68.2%) gave a higher uninjured cell population than LATT (56.9%) (Figure 2).The study only measured (+)-catechin for cytoprotection because (+)-catechin possesses the basic, most typical skeleton (Figure 1), and other galloylated catechins (especially EGCG) have been widely studied for their cytoprotection previously [14][15][16][17].
As mentioned in the previous literature, the FRAP assay at pH 3.6 is essentially an electron-transfer (ET) process [18].The data (Figure S2 and Table 1) showed that the five phenolic components in LATT could effectively increase the FRAP percentages, suggesting that these phenolic components may undergo the ET pathway to scavenge free radicals.
The ET pathway, however, is usually accompanied by an H + -transfer pathway [18].The PTIO•-scavenging assay, an H + -transfer-involved reaction [19,20], was hence performed in the study.The five phenolic components and LATT dose-dependently increased their PTIO•-scavenging efficacies (Figure S3), indicating that H + -transfer may be involved in the antioxidant process. in LATT could effectively increase the FRAP percentages, suggesting that these phenolic components may undergo the ET pathway to scavenge free radicals.
The ET pathway, however, is usually accompanied by an H + -transfer pathway [18].The PTIO•scavenging assay, an H + -transfer-involved reaction [19,20], was hence performed in the study.The five phenolic components and LATT dose-dependently increased their PTIO•-scavenging efficacies (Figure S3), indicating that H + -transfer may be involved in the antioxidant process.   in LATT could effectively increase the FRAP percentages, suggesting that these phenolic components may undergo the ET pathway to scavenge free radicals.
The ET pathway, however, is usually accompanied by an H + -transfer pathway [18].The PTIO•scavenging assay, an H + -transfer-involved reaction [19,20], was hence performed in the study.The five phenolic components and LATT dose-dependently increased their PTIO•-scavenging efficacies (Figure S3), indicating that H + -transfer may be involved in the antioxidant process.
SC 50 values are expressed as the mean ± SD (n = 3).Mean values with different letters (a, b, c, d, e, or f) in same column are significantly different (p < 0.05).All dose-response curves are detailed in Figures 2-5.
It is worth mentioning that, after the ET and H + -transfer reaction, phenolic antioxidants may be transferred into free radicals.The phenolic antioxidant radical may form a covalent adduct with another free radical in the decay process.This is called a radical adduct formation (RAF) reaction [21].As shown in Figure 3A,B, the reaction product of PTIO• with (+)-catechin gave a peak with m/z 579.2 at 1.03 min in the ultra-performance liquid chromatography coupled with electrospray ionization quadrupole time-of-flight tandem mass spectrometry (UPLC−ESI−Q−TOF−MS/MS) analysis.The m/z value (580) of the product was exactly double the molecular weight of (+)-catechin, and its secondary MS resembled (+)-catechin itself (Figure 3C), suggesting a dimerization reaction in the RAF pathway.In addition, (+)-catechin can react with PTIO• to give (+)-catechin-PTIO (m/z 520.9, Figure 3D-F).
As a galloylated derivative of (+)-catechin, EGCG was also found to yield a relevant EGCG-PTIO product (m/z 689.2, Figure 3G-I) and dimeric EGCG-EGCG (m/z 915.2, Figure 3J-K).In addition to EGCG, other galloylated derivatives (CG and ECG) similarly gave RAF products (Figures S6 and S7).GA also gave a dimer of GA-GA.The dimer was identified by a peak with m/z 339.0, which is exactly double the molecular weight of GA in primary MS spectra, which yielded secondary MS spectra similar to GA itself, i.e., m/z 125, 170 (Figure 3L-N and Figure S8).Taken together, GA and the four catechins could generate RAF products when mixing with PTIO• radicals.It can be deduced that GA and the four catechins undergo the RAF pathway to exert their antioxidant actions.
Reactive oxygen species generation is usually linked to transition metals (especially Fe 2+ and Cu + ).For instance, Fe 2+ can catalyze H 2 O 2 to form •OH radicals via the Fenton reaction (Fe 2+ + H 2 O 2 → Fe 3+ + •OH + OH − ) [26].Thus, chelation of Fe 2+ in solution can greatly reduce the catalytic potential to inhibit •OH generation [23,30].In the present study, the five phenolic components and LATT exhibited Fe 2+ -chelating abilities.As shown in Figure 4A,B, GA bound Fe 2+ to generate a blue solution with λ max 583 nm absorption.The EGCG-Fe 2+ complex produced a dark-blue solution with λ max 594 nm.Each of the CG-Fe 2+ and ECG-Fe 2+ complexes gave a navy solution with λ max 585 nm.The (+)-catechin-Fe 2+ complex yielded a green solution with λ max = 462 nm.These data imply that all five phenolic compounds and LATT exhibit Fe 2+ -chelating abilities and that Fe 2+ -chelation may play a role in the antioxidative process of LATT [31].
Molecules 2018, 23, x 5 of 12 phenolic compounds and LATT exhibit Fe 2+ -chelating abilities and that Fe 2+ -chelation may play a role in the antioxidative process of LATT [31].It must be emphasized that CG and ECG are two diastereoisomers: CG and ECG take (2S, 3R) and (2R, 3R) absolute configurations, respectively.Thus, the sole difference between CG and ECG is the absolute steric configuration at the 2-carbon (Figures 1 and 6).This difference gives them distinctive molecular shapes.In the CG molecule, the B ring and galloyl moiety are arranged on two different sides, while they appear on the same side in ECG (Figure 5).Thus, an unbalanced array of chemical moieties enhances the polarity of ECG, and ECG (16.5 min) is eluted ahead of CG (17.0 min) in reversed-phase column HPLC (Figure 1).More importantly, such different steric configurations It must be emphasized that CG and ECG are two diastereoisomers: CG and ECG take (2S, 3R) and (2R, 3R) absolute configurations, respectively.Thus, the sole difference between CG and ECG is the absolute steric configuration at the 2-carbon (Figures 1 and 6).This difference gives them distinctive molecular shapes.In the CG molecule, the B ring and galloyl moiety are arranged on two different sides, while they appear on the same side in ECG (Figure 5).Thus, an unbalanced array of chemical moieties enhances the polarity of ECG, and ECG (16.5 min) is eluted ahead of CG (17.0 min) in reversed-phase column HPLC (Figure 1).More importantly, such different steric configurations can also affect antioxidant abilities, including radical-scavenging and Fe 2+ -chelation.As shown in Table 1, ECG always exhibited higher antioxidant levels than CG in DPPH•-scavenging, ABTS + •-scavenging, PTIO•-scavenging, and FRAP assays.This may be because two very large ipsilateral moieties (B ring and galloyl moiety) may lead to repulsion to increase molecular energy and ET potential.This presumption is consistent with previous findings that cis-catechins exhibit a higher antioxidant ability than their epimers [32,33].
Molecules 2018, 23, x 6 of 12 can also affect antioxidant abilities, including radical-scavenging and Fe 2+ -chelation.As shown in Table 1, ECG always exhibited higher antioxidant levels than CG in DPPH•-scavenging, ABTS + •scavenging, PTIO•-scavenging, and FRAP assays.This may be because two very large ipsilateral moieties (B ring and galloyl moiety) may lead to repulsion to increase molecular energy and ET potential.This presumption is consistent with previous findings that cis-catechins exhibit a higher antioxidant ability than their epimers [32,33].In Fe 2+ -chelation, there is also a difference between CG and ECG.As illustrated in Figure 4B, ECG-Fe 2+ gave stronger UV absorbance than ECG itself, while CG-Fe 2+ exhibited weaker UV absorbance than CG itself.This difference can be explained by the absolute configurational difference at the 2-carbon.As two moieties with metal chelation potential [34,35], the B ring and the galloyl residue, are on the same side in ECG, this configuration can promote their synergistic binding to Fe 2+ .This synergistic effect, however, cannot exist in CG, because they are separated distantly (Figure 5).
In a word, there are substantial differences between CG and ECG in terms of molecular polarity, radical scavenging, and Fe 2+ -chelation.This stereochemistry may help us to further understand differences between catechins and corresponding "epi"-catechins [36,37].In Fe 2+ -chelation, there is also a difference between CG and ECG.As illustrated in Figure 4B, ECG-Fe 2+ gave stronger UV absorbance than ECG itself, while CG-Fe 2+ exhibited weaker UV absorbance than CG itself.This difference can be explained by the absolute configurational difference at the 2-carbon.As two moieties with metal chelation potential [34,35], the B ring and the galloyl residue, are on the same side in ECG, this configuration can promote their synergistic binding to Fe 2+ .This synergistic effect, however, cannot exist in CG, because they are separated distantly (Figure 5).
In a word, there are substantial differences between CG and ECG in terms of molecular polarity, radical scavenging, and Fe 2+ -chelation.This stereochemistry may help us to further understand differences between catechins and corresponding "epi"-catechins [36,37].can also affect antioxidant abilities, including radical-scavenging and Fe 2+ -chelation.As shown in Table 1, ECG always exhibited higher antioxidant levels than CG in DPPH•-scavenging, ABTS + •scavenging, PTIO•-scavenging, and FRAP assays.This may be because two very large ipsilateral moieties (B ring and galloyl moiety) may lead to repulsion to increase molecular energy and ET potential.This presumption is consistent with previous findings that cis-catechins exhibit a higher antioxidant ability than their epimers [32,33].In Fe 2+ -chelation, there is also a difference between CG and ECG.As illustrated in Figure 4B, ECG-Fe 2+ gave stronger UV absorbance than ECG itself, while CG-Fe 2+ exhibited weaker UV absorbance than CG itself.This difference can be explained by the absolute configurational difference at the 2-carbon.As two moieties with metal chelation potential [34,35], the B ring and the galloyl residue, are on the same side in ECG, this configuration can promote their synergistic binding to Fe 2+ .This synergistic effect, however, cannot exist in CG, because they are separated distantly (Figure 5).
In a word, there are substantial differences between CG and ECG in terms of molecular polarity, radical scavenging, and Fe 2+ -chelation.This stereochemistry may help us to further understand differences between catechins and corresponding "epi"-catechins [36,37].).An annexin V/propidium iodide (PI) assay kit was purchased from Abcam (Cambridge, UK).Methanol and water were of HPLC grade.All other reagents used in this study were purchased as analytical grade from the Guangzhou Chemical Reagent Factory (Guangzhou, China).

Preparation of Lyophilized Aqueous Extract of Tibetan Tea (LATT)
Tibetan tea was extracted based on the guidance of the manufacturer.In brief, the newlypurchased Tibetan tea (Kangzhuan) was decocted twice using 50-fold distilled water at 100 • C for 5 min.The extract was concentrated and lyophilized to prepare a lyophilized aqueous extract of Tibetan tea (LATT, Figure S10 and Figure 6).LATT was stored at 4 • C for further exploration.

Preparation of Lyophilized Aqueous Extract of Tibetan Tea (LATT)
Tibetan tea was extracted based on the guidance of the manufacturer.In brief, the newlypurchased Tibetan tea (Kangzhuan) was decocted twice using 50-fold distilled water at 100 °C for 5 minutes.The extract was concentrated and lyophilized to prepare a lyophilized aqueous extract of Tibetan tea (LATT, Figures S10 and 6).LATT was stored at 4 °C for further exploration.

HPLC Analysis for Phenolic Components in LATT
The LATT methanolic solution (48.6 mg/mL) was analyzed using an HPLC method.The HPLC analysis was performed on a Shimadzu LC-20A (Shimadzu Co., Kyoto, Japan) equipped with an Agilent

Ferric Reducing Antioxidant Power (FRAP) Assay
The FRAP assay was carried out based on Benzie and Strain's method [36].Briefly, the FRAP reagent was prepared freshly by mixing 10 mM TPTZ, 20 mM FeCl 3 , and 0.25 M acetate buffer (pH 3.6) at 1:1:10.The test sample (x = 2-10 µL) was added to (20 − x) µL of 95% ethanol followed by 80 µL of FRAP reagent.After incubation for 30 min, the mixture was measured for the A 593nm value at ambient temperature using distilled water as the blank.The relative reducing antioxidant power of the sample compared to the maximum absorbance was calculated by the following formula: Relative reducing e f f ect where A min is the lowest A 593nm value in the experiment, A is the A 593nm value of the reaction mixture with sample, and A max is the greatest A 593nm value in the experiment.

Free Radical-Scavenging Assays In Vitro
The free radical-scavenging assays included the PTIO•-scavenging assay, DPPH•-scavenging assay, and ABTS + •-scavenging assay.The PTIO•-scavenging assay was based on our method [22].In brief, 80 µL of aqueous PTIO• solution (0.1 mM) was mixed with 20 µL of an aqueous or alcoholic solution of sample at various concentrations.The mixture was incubated at 50 • C for 30 min, and the absorbance at 560 nm was measured using a microplate reader (Multiskan FC, Thermo Scientific, Shanghai, China).The PTIO•-scavenging capacity % was calculated as: where A 0 is the absorbance of the control without sample and A is the absorbance of the reaction mixture with sample.The ABTS + •-scavenging assay was based on a previous report [37].ABTS + • was produced by mixing 200 µL of (NH 4 ) 2 ABTS (7.4 mM) with 200 µL of K 2 S 2 O 8 (2.6 mM).After incubation in the dark for 12 h, the mixture was diluted with methanol (approximately 1:20) so that the absorbance at 734 nm was 0.30 ± 0.01.Then, the diluted ABTS + • solution (80 µL) was added to 20 µL of an ethanolic (or aqueous) solution of sample at various concentrations and mixed thoroughly.After the reaction mixture stood for 3 min, the absorbance at 734 nm was measured using a microplate reader.The ABTS + •-scavenging capacity % was calculated using the formula described in the PTIO•-scavenging assay.
The DPPH•-scavenging assay was based on previous reports [38].In the assay, 80 µL of an ethanolic solution of DPPH• (0.1 mM) was mixed with 20 µL of an ethanolic (or aqueous) solution of sample at various concentrations.After incubation for 30 min, the mixture was measured for absorbance at 519 nm.The DPPH•-scavenging capacity % was calculated using the formula described in the PTIO•-scavenging assay.SC 50 was defined as the antioxidant concentration to give a scavenging capacity % half the value of that in the absence of antioxidant.

UPLC−ESI−Q−TOF−MS/MS Analysis of Reaction Products of PTIO• with Phenolic Components
UPLC−ESI−Q−TOF−MS/MS spectra of reaction products of PTIO• with phenolic components were determined according to the method described by our report [39].The methanolic solutions of phenolic components were mixed with a solution of PTIO• radicals in methanol at a molar ratio of 1:2, and the resulting mixtures were incubated for 24 h at room temperature.The product mixtures were filtered through a 0.22-µm filter and measured using a UPLC-ESI-Q-TOF-MS/MS system equipped with a C 18 column (2.0 mm i.d.× 100 mm, 2.2 µm, Shimadzu Co., Kyoto, Japan).The mobile phase was used for elution and consisted of a mixture of methanol (phase A) and water (phase B).The column was eluted at a flow rate of 0.3 mL/min with the following gradient elution program: 0-10 min, 60-100% A; 10-15 min, 100% A. The sample injection volume was set at 1 µL to separate the components, column temperature was 40 • C; Q-TOF-MS/MS analysis was conducted with a Triple TOF 5600 plus mass spectrometer (AB SCIEX, Framingham, MA, USA) equipped with an ESI source, which was run in the negative ionization mode.The scan range was set at 100-2000 Da.The system was run with the following parameters: ion spray voltage, −4500 V; ion source heater, 550 • C; curtain gas (CUR, N 2 ), 30 psi; nebulizing gas (GS1, air), 50 psi; Tis gas (GS2, air), 50 psi.The declustering potential (DP) was set at −100 V, whereas the collision energy (CE) was set at −40 V with a collision energy spread (CES) of 20 V. The RAF products were quantified by extracting the corresponding formula (e.g., [C 28 H 30 N 2 O 8 -H] − for (+)-catechin-PTIO• and [C 30 H 28 O 12 -H] − for (+)-catechin dimer) from the total ion chromatogram and integrating the corresponding peak.

Cytoprotective Effect Assay
The MSC culture was performed in terms of our previous reports with slight modifications [14,40].Briefly, bone marrow was obtained from the femur and tibia of rats.The marrow samples were diluted with L-DMEM containing 10% FBS.MSCs were prepared by gradient centrifugation at 900× g for 30 min on 1.073 g/mL Percoll.The prepared cells were transferred into culture flasks at a density of 1 × 10 4 /cm 2 .The MSCs at passage 3 were estimated for cultured cell homogeneity using detection of CD44 by flow cytometry and used for the investigation.Injured MSCs were prepared by treatment with H 2 O 2 (200 µM).After incubation for 3 h, MSCs in the control group and H 2 O 2 group were incubated for 24 h in DMEM, while MSCs in the sample groups were incubated for 24 h in DMEM with (+)-catechin and LATT.
After culturing, cells were assayed using flow cytometry [41].Briefly, the cultured cells were harvested with trypsin (0.05%) digestion in phosphate-buffered saline (PBS).The cells (5 × 10 5 ) were collected by centrifugation and then resuspended in 500 µL of binding buffer.At the same time, 5 µL of annexin V-FITC and 5 µL of propidium iodide were added.After incubation at room temperature for 5 min in the dark, fluorescence was measured by flow cytometry (Accuri C6, Franklin Lakes, BD, USA) with standard software.

UV Spectral Determination of Fe 2+ -Chelation
UV spectra for the Fe 2+ -chelation of LATT and its phenolic components were determined according to the method described by our report [42][43][44].Briefly, 75 µL of a methanolic solution of sample (1 mg/mL) and 25 µL of an aqueous solution of FeCl 2 •4H 2 O (100 mg/mL) were added to 400 µL of an aqueous mixture of distilled water and methanol (1:1).The solution was mixed vigorously.The resulting mixture was incubated at room temperature for 1 h.Subsequently, the product mixture was collected and a spectrum was obtained using a UV-visible spectrophotometer (Unico UV 2600A, Unico Co., Shanghai, China) from 200 to 900 nm.Next, 200 µL of the supernatant was transferred to a 96-well plate and photographed using a smartphone.

Statistical Analysis
The SC 50 values were obtained by linear regression analysis.All linear regression analysis was conducted with Origin 6.0 professional software.Measurement of significant differences between the mean SC 50 values of the sample and positive controls was performed using one-way ANOVA or the T-test.The analysis was performed using SPSS software 13.0 (SPSS Inc., Chicago, IL, USA) for Windows.p < 0.05 was considered to be significant.

Figure 5 .
Figure 5. Stick-models of preferential conformations of CG and ECG (Red is for O atom; White is H and gray is C. Double bonds are not shown.).

Figure 5 .
Figure 5. Stick-models of preferential conformations of CG and ECG (Red is for O atom; White is H and gray is C. Double bonds are not shown.).

Figure 5 .
Figure 5. Stick-models of preferential conformations of CG and ECG (Red is for O atom; White is H and gray is C. Double bonds are not shown.).

Table 1 .
The FRAP and SC 50 values of