Phenolic compounds and the biological effects of Pu-erh teas with long-term storage

ABSTRACT

Pu-erh tea has gained more and more popularity and attracted much attention for its various biological effects. The objective of this study was to determine the active phenolic compounds and the biological effects of 15 differently aged Pu-erh teas. The results showed that 43 active phenolics, containing 7 flavan-3-ols, 11 organic acids, and esters, 3 proanthocyanidin dimers, 2 benzotropolones, and 20 flavonoid glycosides were identified based on high-performance liquid chromatography coupled to electrospray ionisation and quadrupole time-of-flight tandem mass spectrometry (HPLC-ESI-QTOF-MS/MS). In particular, 2’-O-galloylhyperin and quercetin-3-O-p-coumaroyl-rhamnosyl-arabinosyl-hexoside were identified from constituents of tea for the first time. The 15 Pu-erh teas exhibited strong DPPH and ABTS scavenging activity in a concentration-dependent manner with IC₅₀ 0.99–2.12 and 0.97–2.67 mmol Trolox equivalent antioxidant capacity/g, respectively. All of the Pu-erh teas tested significantly inhibited the proliferation of SMMC-7721 cells, in a dosage-dependent manner. Of the 15 Pu-erh teas examined, the cake tea P6 had the strongest effect on SMMC-7721 cells with IC₅₀ = 65.88 ± 3.53 µg/mL. The others Pu-erh teas had an IC₅₀ = 96 – 509 µg/mL. Our results also indicated that not all the older Pu-erh teas display stronger anti-activities and anti-cancer than the younger ones and the youngest Pu-erh tea did not possess the highest level of active components. This study provides useful information for consumers to deeply understand the chemical constituents and bioactivity of Pu-erh tea preserved for a long time.

KEYWORDS:

• Pu-erh tea
• Phenolic compounds
• HPLC-ESI-QTOF-MS/MS
• Radical scavenging activity
• Human hepatocellular carcinoma

Introduction

Pu-erh tea is a unique fermented tea, which is originally produced in Yunnan Province, southwest of China using the leaves of the cultivated variety large-leaf tea of Camellia sinensis. It has been confirmed that Pu-erh tea exhibits multiple beneficial health effects, such as antioxidant,^[1,2] anticancer,^[3] antimutagenic,^[4] antimicrobial,^[4] anti-atherogenic, antiobesity,^[5,6] as well as hepatoprotective^[7] activities. The health benefits attributed to Pu-erh tea has increased its popularity.

In the production of Pu-erh tea, the sun-dried green tea leaves are autoclaved and dried to make Pu-erh raw tea and stored in natural conditions to make aged Pu-erh tea, or the leaves are artificially post-fermented with microorganisms for several months to 1 year to produce ripened Pu-erh tea. For fully fermented black tea, the oxidation of the tea polyphenols is performed mainly by polyphenol oxidases from the tea leaves. However, the polyphenol oxidases are inactivated by heat during the steam-compression process. Thus, the tea polyphenols in Pu-erh tea are oxidized mainly by the action of microorganisms.^[8] The microorganisms oxidize tea polyphenols more completely than polyphenol oxidase in black tea, leading to lower concentrations of tea polyphenols and tea catechins in Pu-erh tea.

It is generally believed that the longer the aged Pu-erh tea is kept, the better it will be in both quality and taste. Thus, the commercial value of older Pu-erh tea is much higher than that of the younger ones. However, there is still an argument whether aged Pu-erh teas with longer storage times have stronger bioactivities. Some researchers showed that the chemical constituents and beneficial health effects were significantly different in the Pu-erh teas with different storage years.^[9] Others have reported that younger Pu-erh teas had stronger free radical scavenging activities than older ones of the same type.^[10] Furthermore, another study demonstrated that the quality of tea infusion could reach levels as high as the long-term storage Pu-erh teas after the fresh tea leaves were artificially inoculated with the appropriate microorganisms for a few weeks.^[11] However, little research has been carried out on the bioactivities and biochemical components of the aged Pu-erh teas with 25 or much longer storage years due to their rarity and precious.

Modern pharmacological research suggests that the health benefits of tea consumption are mainly due to the tea polyphenols. Many types of polyphenols have been identified in tea, including catechins, flavonols, flavones, gallic acid (GA), quinic esters of gallic, caffeic and coumaric acid, theaflavins, thearubigins, and proanthocyanidins.^[12,13] Identification and determination of the previously mentioned phenolic components has been performed using hyphenated analytical techniques, such as liquid chromatography-mass spectrometry (LC-MS).

In the present study, 15 aged Pu-erh teas with storage periods from 10 to 105 years were used to evaluate their phenolic constituents and bioactivities. The phenolic compounds of these aged Pu-erh teas were identified by the high-performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry (HPLC-QTOF-MS) technique. Their antioxidant activities were determined by both diphenylpicryl-hydrazyl (DPPH) and 2,2′-azinobis(3-ethylbenzthiazoline-sulfonic acid) diammonium salt (ABTS) assays. And their effects on the proliferation of human hepatocellular carcinoma were investigated using the 3-(4,5-dimethylthiazolyl)-2,5-diphenyl-tetrazolium bromide (MTT) method.

Materials and methods

Tea samples

Fifteen Pu-erh tea samples with storage periods from 10 to 105 years were obtained from Kemingyuan Tea Shop, Xiamen, Fujian, China, as shown in Table S1. Cake tea P7 (ripened pu-erh tea) is manufactured according to artificially post-fermented process in Yunnan province, and then stored at room temperature; others are raw Pu-erh teas, which are stored at natural conditions for a long time (25–105 years) to ferment, that so-called aged Pu-erh teas. More detailed information on these Pu-erh tea samples is summarized in the supporting information (Fig. S1). The tea samples were all kept in desiccators at room temperature. All of the experiments were carried out in 2015.

Chemicals

The standard samples of GA, caffeine, theobromine, theophylline, (+)-catechin (C), (–)-epicatechin (EC), (–)-gallocatechin (GC), (–)-epigallocatechin (EGC), (-)-epicatechin gallate (ECG), (–)-epigallocatechin gallate (EGCG), and (–)-gallocatechin gallate (GCG) were purchased from Sigma (St. Louis, MO). DPPH was obtained from Aldrich (Milwaukee, WI). Acetonitrile, methanol, and formic acid were high-performance liquid chromatography (HPLC)-grade and obtained from Merck (Darmstadt, Germany). Deionized water was prepared using a Millipore Milli Q-Plus system (Millipore, Billerica, MA). Fetal bovine serum (FBS) was obtained from Hyclone. RPMI-1640 medium, penicillin, streptomycin and 0.25% trypsin were purchased from Gibco. Dimethylsulfoxide (DMSO) was obtained from Amresco and MTT was obtained from Sigma.

Preparation of tea infusion samples

Two grams of each tea sample and 50 mL of boiling water were added into a conical flask and heated in a water bath (95°C) for 15 min. The tea infusions were filtrated through 0.22 μm membrane filter after cooling down. Finally, the resulting tea infusions were lyophilized.

HPLC-QTOF-MS analyses

HPLC-QTOF-MS analysis was performed using a Agilent 1260 Series Rapid Resolution LC system equipped with a binary pump with degasser, high-performance autosampler and thermostated column compartment (Agilent Corporation, Milford, MA), coupled with a Agilent 6520 Series quadrupole time-of-flight mass spectrometer. Chromatographic separations were performed on a 2.1 mm × 150 mm 1.8 μm ZORBAX SB-C18 (Agilent, USA) chromatography column. The column was maintained at 35°C and the injection volume was 10 µL. The mobile phase consisted of solvent A (0.5% formic acid water solution) and solvent B (0.5% formic acid acetonitrile solution) at a flow rate of 0.30 mL/min: 0–8 min, 5–5% B; 20–38 min, 5–95% B; 38–45 min, 95–95% B; 47–55 min, 5–5% B. The ESI source was operated in negative and positive mode. Nitrogen (N₂) was used as sheath, auxiliary, and collision gas. The drying temperature and drying gas flow were 350°C and 5 L/min, respectively. The nebulizer pressure, collision energy, fragmentor, and skimmer were 40 psi, 35, 175, and 65 V, respectively. The mass spectra were obtained at a mass-to-charge ratio (m/z) scan range from 100 to 3000. Auto MS/MS parameters were used for the analysis in a full-scan mode.

Determination of aged Pu-erh tea antioxidant activity

DPPH radical-scavenging activity of the tea infusions was determined by the method described by Gaulejac et al.^[14] with some modifications. Briefly, 0.1 mL of lyophilized tea powder water solution of various concentrations was added to 3 mL of the DPPH methanol solution (6 × 10⁻⁵ mol/L). The samples were incubated at room temperature for 30 min. Subsequently, the absorbance of each sample was measured at 517 nm using a Shimadzu ultraviolet (UV)-1602PC spectrophotometer (Kyoto, Japan). The radical-scavenging ability was calculated by the formula: scavenging ability (%) = 100 × [(A₀ – A_i /A₀)]. Where A_i is absorbance of the tea sample; A₀ is absorbance of the methanol blank control.

The ABTS ⁺ radical scavenging activity was measured using the method described by Antolovich et al.^[15] with minor changes. First, the ABTS radical cation (ABTS·⁺) work solution was prepared by mixing ABTS·⁺ stock solution (7 mmol/L in water) with 2.5 mM potassium persulfate for 12 h at 4°C in the dark and then the work solution was diluted with ethanol to an absorbance of approximately 1.0 at 734 nm. Second, 0.1 mL of various concentrations of lyophilized tea powder water solution was mixed with 3.9 mL ABTS·⁺ work solution, and then the samples were incubated at room temperature for 6 min. Finally, the absorbance of each sample was measured at 734 nm using the Shimadzu UV-1602PC spectrophotometer (Kyoto, Japan) with ethanol as a blank control.

All of the experiments were repeated three times. The IC₅₀ values (50% inhibition of the DPPH and ABTS radical formation, mg/mL) of the tea infusions and Trolox were calculated, respectively. The antioxidant activity of the tea sample was converted to the Trolox equivalent antioxidant capacity (TEAC) and calculated as follows: TEAC (mmol Trolox/g) = IC₅₀ (Trolox) /IC₅₀ (Pu-erh tea sample).

Determination of anticancer activity in the aged Pu-erh teas

The anticancer activity of the aged Pu-erh teas was determined using the MTT assay described by Van de Loosdrecht et al.^[16] Human hepatocarcinoma cancer cell line SMMC-7721 was provided by Xiamen Overseas Chinese Subtropical Plant Introduction Garden (Fujian, China). The SMMC-7721 cells were cultured at 37°C with 5% CO₂ in a RPMI-1640 medium (pH = 7.2–7.4) supplemented with 10% FBS, 100 U/mL penicillin and 100 U/mL streptomycin. Exponentially growing cells were harvested and prepared into cell suspensions. Subsequently, 200 μL cell suspensions were inoculated into each well of the 96-well plates and the cells were incubated at 37°C for 24 h to achieve cell attachment. After that, the SMMC-7721 cells were treated with 180 μL/well tea infusion samples with different concentration for 24 h. Next, 20 μL MTT solution (5 mg/mL) was added to each well and the plates were incubated at 37°C for another 4 h. After the supernatants were removed, 150 μL DMSO was added to each well to dissolve the formazan crystals. Finally, the absorbance of each sample was read at 490 nm using a microplate reader. The cell survival percentage was calculated using the formula: Survival (%) = OD_{tea exposure} /OD_control × 100. The IC₅₀ values (50% inhibition of cell growth) were calculated. All of the experiments were repeated five times.

Total content determination of the phenolics, flavonoids, catechins, GA, and purine alkaloids

Total polyphenol content of the Pu-erh tea samples was determined by the Folin-Ciocalteu method described by Singleton et al. with minor modification.^[17] First, 1 mL of the tea infusion samples were mixed with 5 mL of 10% (v/v) Folin-Ciocalteu’s phenol reagent and 4 mL of 7.5% (w/v) Na₂CO₃ solution. Second, the mixtures were incubated for 1 h at room temperature and then their absorbance at 765 nm was measured. The total polyphenol content of each tea infusion sample was calculated according to a standard curve of GA solution and converted into milligrams of GA equivalents per gram of the Pu-erh tea of sample (mg/g). Each experiment was conducted in triplicate.

The total flavonoid content was analyzed using the Dowd method with minor modification.^[18] Briefly, 2 mL tea infusion sample, 8 mL of 1.5% (w/v) AlCl₃, 4 mL HAc-NaAc buffer (pH = 5.5), and 11 mL of 50% (v/v) ethanol were added to 25-mL volumetric flask, respectively. After incubation for 30 min, the absorbance was measured at 415 nm using a UV-vis spectrophotometer. Rutin was used as a standard solution. Total flavonoid content was expressed as milligrams of rutin equivalents per gram of the Pu-erh teas. All experiments were carried out in triplicate.

The content of free amino acid (glutamate as a standard solution) and total soluble saccharides (glucose as a standard solution) were determined according to the methods described by Liang et al.^[19] Total contents of GA, purine alkaloids and catechins were separately measured on the Agilent series 1100 (Agilent Technologies) liquid chromatograph. A Welch Ultimate XB-C18 column (250 mm × 4.6 mm i.d., 5 μm) with a precolumn was used in this study. The separation was achieved by a gradient elution of 0.1% formic acid solution (solvent A) and acetonitrile (solvent B) at a flow rate of 1 mL/min: 0–20 min, 5–17% B; 20–40 min, 17–17% B. The equilibration time was 20 min and the injection volume was 20 μL. The detection wavelength was set at 278 nm. The column was re-equilibrated with the initial conditions for 15 min before the next run.

Statistical analysis

Correlations (r) of data were evaluated through Pearson’s correlation analysis using SPSS 19.0 software, and a p-value of less than 0.05 was considered to be statistically significant.

Results and discussion

Polyphenolic identification by HPLC-QTOF-MS/MS analysis

Some studies demonstrated that the potential health benefits of the Pu-erh teas could be mainly attributed to their polyphenolic compounds.^[20,21] Accordingly, the polyphenolic components in the 15 aged Pu-erh teas were identified by the HPLC-QTOF-MS/MS technique in this study. Combined with the relevant literature, the components of the tea polyphenols were putatively annotated by comparison of their accurate mass and MS/MS fragmentation patterns with those of the authentic standards.^[22–26] Chromatogram profiles of the aged Pu-erh teas recorded as negative modes are shown in Fig. S2. The retention times, MS and MS² spectral data and the identification results of 43 compounds in the aged Pu-erh teas are listed in Table 1.

Table 1. Retention times, MS spectral characteristics, and identify of phenolic compound present in 15 Pu-erh teas.

Peak	Retention time (min)	Identified compounds	[M–H]^– (m/z)	MS^Citation2 (m/z)
1	1.184	Ethyl 3,4-dihydroxybenzoate	181.068	135.9104
2	1.299	Quinic acid	191.0559	127.0389
3	1.734	Purpurogallin	221.0256	113.0205
4	2.757	Unknown	151.0263	108.0209
5	4.053	Gallic acid	169.0147	125.0239
6	5.505	3-O-galloylquinic acid	343.0676	191.0557
7	9.511	GC	305.0668	167.0336, 165.0161
8	16.242	(Epi)theaflagalli	399.1055	275.1064, 110.0260
9	16.893	Unknown	194.0417	108.0464
10	17.505	Unknown	465.1066	331.0780, 245.0806
11	17.546	3-p-coumaroylquinic acid	337.0908	119.0500, 163.0387, 191.026
12	17.861	Caffeic acid	179.0580	135.0308, 164.0328, 107.0125, 122.0358
13	17.894	4-p-coumaroylquinic acid	337.0919	163.0035, 125.0259
14	17.973	C	289.0731	123.0450
15	18.155	EGC	305.0669	219.0645, 167.0348, 165.0212
16	18.142	(Epi)catechin(epi)catechin	577.1380	289.0709, 125.0231, 407.0747
17	18.400	(Epi)catechin(epi)catechin	577.1337	407.0780, 289.0716, 125.0247
18	18.7	Unknown	209.0636	137.0239
19	19.042	Unknown	545.1238	125.0365
20	19.313	Dihydromyricetin	319.0416	203.0337, 175.0404,137.0263, 109.0304
21	19.417	(Epi)catechin(epi)catechin	577.1279	289.0658, 407.0719, 125.0210
22	19.505	Unknown	465.103	331.0819, 205.0502, 123.0444, 179.0344
23	19.612	Unknown	361.0561	165.0189, 137.0289, 123.0463, 109.0307
24	19.975	3-caffeoylquinic acid	353.0872	191.0553, 179.0367, 173.0452, 135.0449
25	20.573	EG C G	457.0769	305.0606, 169.0136
26	20.963	Unknown C-glycosylated flavones	595.1636	475.1165, 383.0935, 355.0826
27	20.837	EC	289.0714	137.0241, 203.0678, 123.0446, 125.0238
28	21.764	5-p-coumaroylquinic acid	337.0926	191.0556, 173.0447, 163.0381
29	22.4	6,8-C-diglucosylapigenin	593.1544	473.1105, 413.0821, 383.0786, 353.0680
30	22.749	ECG	441.0857	289.0694, 169.0129
31	23.051	CG	441.0820	289.0701, 169.0141
32	23.575	6-C-arabinosyl-8-C-glucosyl apigenin	563.1396	473.1096, 443.0984, 413.0889, 383.0756, 353.0644
33	23.833	8-C-arabinosyl-6-C-glucosyl apigenin	563.1426
34	24.522	Myricetin 3-O-galactoside	479.0816	316.0220, 317.0247
35	24.756	Unknown	400.142	205.0505, 203.0691, 161.0599, 123.0452, 137.0218
36	24.917	Vitexin	431.0989	311.0553, 283.0599
37	24.736	2’-O-galloylhyperin	615.0925	463.0774, 301.0267
38	25.241	Unknown	400.1398	205.0496, 161.0599, 123.0492
39	25.315	Isovitexin	431.0996	311.0510, 283.0592
40	25.715	Apigenin 6,8-C-depentoside	533.1318	473.1049, 443.0953, 413.0793, 383.0763, 353.0643
41	25.961	Quercetin-3-O-dirhamnosylglucoside	755.1969	609.1176, 301.0321
42	26.147	Rutin	609.1448	301.0333
43	26.374	Ellagic acid	300.9992
44	26.831	Kaempferol-3-O-rhamnosyl-galactoside	593.1435	285.0354
45	27.137	Kaempferol-rhamnose-hexose-rhamnose conjugate	739.2004	593.1507, 285.0440
46	27.190	Myricetin	317.0309	151.0025, 137.0235, 109.0296
47	27.281	Kaempferol-3-O-rhamnosyl-glucoside	593.1440	285.0342
48	27.71	Unknown	401.0908	341.0692, 233.0438, 191.0344, 163.0375, 151.0396, 123.0452
49	29.379	Quercetin	301.0349	151.0033, 121.0297, 107.0141
50	29.902	Quercetin-3-O-p-coumaroyl-rhamnosyl-arabinosyl-hexoside	887.2237	741.1873,595.1124, 301.0348
51	30.332	Quercetin-3-O-p-coumaroyl-dirhamnosylhexoside	901.2346	755.1764, 609.1458, 301.0336
52	31.172	Kaempferol	285.0388
53	36.377	Trigalloylglucose	635.0416	112.9858

Flavan-3-ols

Peaks 7, 14, 15, 25, 27, 30, and 31 were identified as GC, C, EGC, EGCG, EC, ECG, and CG by comparing the retention time, [M–H]^– ions and MS² fragment ions with those of authentic standard compounds and literature data. The parent ion [M–H]^– at m/z 289.0714, generated MS² fragments at m/z 203.0678 and 137.0241, corresponding to the loss of one C₄H₆O₂ and C₈H₈O₃ moieties, respectively. The parent ion [M–H]^– at m/z 305.0669, generated MS² fragments at m/z 219.0645, 167.0348, and 165.0212, corresponding to the loss of one C₄H₆O₂, C₇H₆O₃, and C₇H₈O₃ moieties, respectively. The parent ion [M–H]^– at m/z 441.0857 produced MS² fragments 289.0694 and 169.0129 corresponding to (epi) catechin and GA moieties, respectively. The parent ion [M–H]^– at m/z 457.0769 produced MS² fragments 305.0606 and 169.0136 corresponding to (epi) gallocatechin and GA moieties, respectively.

Proanthocyanidin dimers

Peak 16, 17, and 21 have the same [M–H]^– at m/z 577.1337 and MS² ion at m/z 407.0780 and 289.0716 corresponding to the cleavage of GA (170 Da) and epi(catechin); therefore, these compounds were tentatively identified as (epi)catechin(epi)catechin. These proanthocyanidin dimers were formed through quinone-methide (QM), retro-Diels-Alder (RDA), and heterocyclic ring fission (HRF) pathways.^[10]

Organic acids and their esters

Peak 1 was identified as ethyl 3,4-dihydroxybenzoate because it had [M–H]^– at m/z 181.068 and MS² fragments at m/z 135.9104 corresponding to the loss of a OC₂H₅ moiety. Peak 2 was identified as quinic acid because it has [M–H]^– at m/z 191.0559 and MS² fragments at m/z 127.0389. Peak 5 was identified as GA, which had a [M–H]^– at m/z 169.0147 and MS² ions at m/z 125.0239 with the loss of one CO₂ molecule. Peak 6 was tentatively identified as 3-O-galloylquinic acid, which produced a [M–H]^– at m/z 343.0676 and prominent MS² ions at m/z 191.0557 (quinic acid residue). Peak 11, 13, and 28 have the same [M–H]^– at m/z 337.0919 and fragmented ions at m/z 163.0035 corresponding to the deprotonated ion of cinnamoyl; therefore, peaks 11, 13 and 28 were 3-p-coumaroylquinic acid, 4-p-coumaroylquinic acid, and 5-p-coumaroylquinic acid, respectively, according to the elute order described by Clifford et al.^[27] Peak 12 was determined to be caffeic acid, which had a [M–H]^– at m/z 179.0580. Peak 24 was determined to be 3-caffeoylquinic acid, which had a [M–H]^– at m/z 353.0872 and MS² fragment at m/z 191.0553 corresponding to the loss of a caffeoyl unit. Peak 43 was identified as ellagic acid, it had a [M–H]^– at m/z 300.9992 that is almost identical to peak 49 (quercetin) with [M–H]^– at m/z 301.0349. Peak 53 was tentatively identified as trigalloylglucose because it had a [M–H]^– at m/z 635.0416 and MS² fragments at m/z 112.9858.

Flavonols

Peak 49 had a [M–H]^– ion at m/z 301.0349 and MS² ion at m/z 151.0033, 121.0297, 107.0141 was tentatively identified as quercetin. Peak 52 with [M–H]^– ion at m/z 285.0388 was tentatively identified as kaempferol. Peak 46 with [M–H]^– ion at m/z 317.0309 and MS² ion at m/z 151.0025, 137.0235, and 109.0296 was tentatively identified as myricetin. Peak 20 with a [M–H]^– ion at m/z 319.0416, and MS² ion at m/z 203.0337, 175.0404, 137.0263, 109.0304, has a molecular weight 2 amu higher than compound myricetin was tentatively identified as dihydromyricetin.

O-Glycosylated flavonols

A total of six myricetin, quercetin, or kaempferol glycosides conjugated to a range of sugars, including glucose, galactose, and rhamnose, as mono-, di-, and trisaccharides were identified, which were universally present in tea. Peak 34 had a [M–H]^– at m/z 479.0816, which produced MS² ions at m/z 316.0220 and 317.0247 (myricetin) with the loss of m/z 162 corresponding to the cleavage of hexose sugar; thus, peak 34 was determined as myricetin 3-O-galactoside.

Peak 41 had a [M–H]^– at m/z 755.1969 and MS² fragments at m/z 609.1176 and 301.0321 corresponding to the loss of 146 Da and 454 Da fragments. The loss of 146 Da was in keeping with the cleavage of a rhamnosyl unit, 301.0321 corresponding to quercetin. Generally, the more sugar moieties attached to the flavonoid aglycone, the earlier the retention time in reversed phase HPLC. Therefore, peak 41 was identified as quercetin-3-O-dirhamnosylglucoside.

Peak 42 was identified as rutin with [M–H]^– at m/z 609.1448 and MS² fragment at m/z 301.0333 by losing a sugar molecule of glucose and rhamnose from aglycone quercetin. Peak 44 and 47 have the same [M–H]^– at m/z 593.1435 and MS² fragment at m/z 285.0354 by losing a sugar molecule of hexose-rhamnose from aglycone kaempferol. Galactosides elute earlier than the corresponding glucosides; therefore, peak 44 and 47 were identified as kaempferol-3-O-rhamnosyl-galactoside and kaempferol-3-O-rhamnosyl-glucoside, respectively. Peak 45 was tentatively identified as kaempferol-rhamnose-hexose-rhamnose with a [M–H]^– at m/z 739.2004 and MS2 fragment at m/z 593.1507 corresponding to the loss of a 146 Da fragment (equates with the cleavage of a rhamnosyl unit) and 285.0440 (kaempferol).

C-Glycosylated flavones

Eight C-glycosylated apigenins were identified in 15 aged Pu-erh teas. Peak 29 had a [M–H]^– ion at m/z 593.1544 and MS² ions at m/z 473.1105 (M–H–Hexosyl)^–, 383.0786, 353.0680 (M–H–dihexosyl)^–, showing the existence of two glycosyls; thus, peak 29 was tentatively identified as 6,8-C-diglucosylapigenin. Peak 26 had a [M–H]^– ion at m/z 595.1636 and predominant MS² ions at m/z 475.1165, 385.0935, and 355.0826 (M–H–dihexosyl)^–, which has a molecular weight 2 amu higher than compound 29; therefore, it was identified as an unknown C-glycosylated flavones.

Peaks 32 and 33 had the same [M–H]^– at m/z 563.1396 and MS² ions at m/z 473.1096 (M–H–pentosyl)^–, 443.0984 (M–H–hexosyl)^–, 383.0756 and 353.0644, which were assigned as 6-C-arabinosyl-8-C-glucosyl, apigenin, and 6-C-glucosyl-8-C-arabinosyl apigenin, respectively based on the different ratio of ions m/z 443:473. Peak 40 was tentatively identified as apigenin 6,8-C-depentoside because it had a [M–H]^– ion at m/z 533.1318 and MS² fragments at 473.1049 (due to loss of 60 amu from one pentosyl), 443.0953, 413.0793 (due to loss of two pentosyls of 60 amu), 383.0763, and 353.0643.

Peak 36 and 39 have the same [M–H]^– ion at m/z 431.0989 and main MS² fragment ions at m/z 311.0553 (M–H–hexosyl)^–, 283.0599, which were assigned as vitexin and isovitexin. Peak 37 had a [M–H]^– ion at m/z 615.0925 and main MS² fragment ions 463.0774 and 301.0267; thus, peak 37 was tentatively identified as 2’-O-galloylhyperin, which was not previously reported in tea.

Acylated glycosylated flavonols

Two acylated flavonol glycosides contained one p-coumaroyl group were identified in the tested Pu-erh teas. Peak 50 had a [M–H]^– at m/z 887.2237 and MS² fragment at m/z 741.1873 corresponding to the loss of a 146 Da. This 146 amu loss was in keeping with the cleavage of a p-coumaroyl group. The ion 741.1873 produced 595.1124 and 301.0348 ions corresponding to the loss of a 146 Da (a rhamnose moiety) and 294 Da (a sugar molecule of arabinosyl and hexosyl) fragments. The fragment ion at m/z 301 corresponded to the aglycone quercetin. Therefore, peak 50 was tentatively identified as quercetin-3-O-p-coumaroyl-rhamnosyl-arabinosyl-hexoside, which is a new compound identified in teas.

Peak 51 was speculated as quercetin-3-O-p-coumaroyl-dirhamnosylhexoside with [M–H]^– at m/z 901.2346 and MS² fragment at m/z 755.1764 corresponding to the loss of a 146 Da fragment that was in keeping with the cleavage of a p-coumaroyl group. The fragment ion at m/z 755.1764 was similar to the deprotonated ion of compound xx, which produced fragment ions at m/z 609.1458 and 301.0336.

Substance of benzotropolones

Peak 3 had a [M–H]^– ion at m/z 221.0256 and a predominant MS² ion fragment at m/z 113.0205, corresponding to the loss of a pyrogallol unit; thus, peak 3 was tentatively identified as purpurogallin. Purpurogallin is well known to play a key role in affecting the availability of micronutrients in the soil, the bioavailability of metals in the body and the oxidation of theaflavins and their physiological activity.^[28] Peak 8 had a [M–H]^– ion at m/z 399.1055 and main MS² ion fragments at m/z 275.1064 and 110.0260, corresponds with (C₁₄H₁₂O₆–H)^– ion and C₆H₆O₂; therefore, peak 8 was tentatively identified as (epi) theaflagallin, which was formed from the oxidation of (epi) gallocatechin and pyrogallol.

In total, 43 phenolic compounds including 7 catechins, 3 proanthocyanidin dimers, 11 organic acids, and their esters, 4 flavonols, 6 O-glycosylated flavonols, 8 C-glycosylated flavones, 2 acylated glycosylated flavonols, and 2 benzotropolones were identified in 15 Pu-erh teas. Many of the previously stated phenolic compounds have been reported to exhibit a broad range of biological activities, especially the catechins, which are strong antioxidants in vitro and in vivo. For example, Qian et al.,^[10] reported that 10 major free radical scavengers were found in Pu-erh tea, and nine of them were identified as GA, GC, EGC, C, EC, EGCG, ECG, rutin, and quercetin-3-glucoside by HPLC-DAD-MS coupled online with ABTS-based analysis.

Proanthocyanidin dimers are known to be good free radical scavengers because it contains many more hydroxyl groups than a monomer. Because hydroxyl groups in dimers are conjugated with a double bond, dimers may act as a hydrogen donor, thereby stabilizing the electron distribution in the molecule and enhancing antioxidative properties.^[29] Flavonoids in tea exist in aglycone-free and glycoside forms. The major flavonoids in tea are the conjugates of quercetin and kaempferol with conjugating moiety varying from mono- to tetra-glycosides.^[30–33] GA, quinic esters of gallic, coumaric, and caffeic acids are the common phenolic acids in teas.^[32] Flavonoids and phenolic acids have been demonstrated to possess many biological activities, such as antioxidant capacity, antiproliferative, apoptotic activity, antiviral, and antimicrobial.^[34–37]

DPPH and ABTS radical scavenging activity

Free radicals can damage diverse cellular macromolecules, including proteins, carbohydrates, lipids, and nucleic acids due to their high reactivity, further leading to many diseases (such as diabetes mellitus, cancer, atherosclerosis, neurodegenerative diseases, and arthritis).^[38] Previous studies have reported that the Pu-erh tea extracts can effectively eliminate the free radicals, preventing living organisms from oxidative damage.^[2,39] Therefore, the radical-scavenging activities of 15 aged Pu-erh teas with the storage periods from 10 to 105 years were separately evaluated by the DPPH and ABTS assays in present study.

The IC₅₀ values of DPPH and ABTS assay are shown in Table 2. The results showed that the brick tea B4 (75-year storage) exhibited simultaneously the lowest DPPH and ABTS radical-scavenging activities. In the DPPH assay, the bowl tea T1 (75-year storage) was observed to possess the highest DPPH radical-scavenging activity, followed by the bowl tea T2 (71-year storage). In the ABTS assay, the cake tea P3 (75-year storage) appeared to have the highest ABTS radical-scavenging activity. Furthermore, the DPPH and ABTS radical-scavenging activities of the 15 aged Pu-erh teas occurred in a concentration-dependent manner (Fig. S3). The DPPH and ABTS radical-scavenging rates of the 14 aged pu-erh tea extracts with concentrations from 0.6 to 2.5 mg/mL could all reach at least 70% except the extracts of the brick tea B4.

Table 2. IC₅₀ values of DPPH and ABTS assay of 15 Pu-erh teas.

Pu-erh tea type	Number	IC50 mmol TEAC/g
		DPPH	ABTS
Bowl tea	T1	2.12 ± 0.03	1.43 ± 0.08
	T2	2.08 ± 0.01	1.74 ± 0.06
	T3	1.83 ± 0.03	2.10 ± 0.10
Cake tea	P1	1.13 ± 0.01	2.07 ± 0.15
	P2	1.28 ± 0.02	1.97 ± 0.06
	P3	1.62 ± 0.01	2.67 ± 0.02
	P4	1.17 ± 0.01	1.40 ± 0.06
	P5	1.18 ± 0.01	1.36 ± 0.03
	P6	1.48 ± 0.00	1.90 ± 0.10
	P7	1.10 ± 0.01	1.27 ± 0.06
Brick tea	B1	1.79 ± 0.01	1.36 ± 0.13
	B2	1.95 ± 0.02	2.13 ± 0.04
	B3	1.44 ± 0.04	2.23 ± 0.12
	B4	0.99 ± 0.04	0.97 ± 0.02
	B5	1.74 ± 0.03	2.23 ± 0.14

Our results showed that the radical-scavenging activities of different aged Pu-erh teas varied, and not all the older Pu-erh teas exhibit stronger biological ability than the younger ones. This result is not in accordance with an earlier study performed by Qian et al., in which the ABTS radical-scavenging activities of the aged Pu-erh teas decreased with the increase of the Pu-erh tea storage years.^[10] Phenolic compounds have been reported to be excellent free radical scavengers due to their strong hydrogen-donating ability. The inconsistency between these results might be attributed to the difference in phenolic constituents of the aged Pu-erh tea samples used by these two independent studies.

Anticancer activity of aged Pu-erh teas

MTT is one of the most common in vitro assays to estimate the cytotoxic and/or cytostatic, or apoptosis-inducing effects of anticancer agents prior to preclinical and clinical testing. In the present study, effects of the aged Pu-erh tea samples on the proliferation of human hepatoma SMMC-7721 cells were examined in vitro using MTT assay with the standard substances of EGCG as a positive control. As shown in Fig. 1, growth and proliferation of SMMC-7721 cells were inhibited in a dose-dependent manner by the treatments of both EGCG and the Pu-erh tea samples for 24 h. The IC₅₀ values of 15 aged Pu-erh teas and EGCG were summarized in Table 3. The results showed that the cake tea P6 (25-year storage) exhibited the strongest inhibitory activity on the proliferation of the SMMC-7721 cells with an IC₅₀ value of 65.88 ± 3.53 µg/mL. Simultaneously, the cell viability of the SMMC-772 cells was less than 10% when the cells were treated with 120 µg/mL tea extracts of the cake tea P6 for 24 h. The IC₅₀ value of the bowl tea T3 (27-year storage) was 96.94 ± 2.15 µg/mL, while those of the other 13 Pu-erh teas were from 154 to 509 µg/mL. Although the anticancer activities of the 15 aged Pu-erh tea samples were all weaker than that of the positive control, pure EGCG, of which the IC₅₀ value was 24.24 ± 0.92 µg/mL, the Pu-erh teas used in this study all exhibited potent anticancer activities, revealing their beneficial health effects.

Table 3. IC₅₀ values of MTT assay of 15 Pu-erh teas.

Pu-erh tea type	Number	IC50 μg/mL
Bowl tea	T1	508.88 ± 11.21
	T2	467.1 ± 7.56
	T3	96.94 ± 2.15
Cake tea	P1	384.62 ± 12.30
	P2	289.02 ± 7.06
	P3	160.12 ± 9.1
	P4	404.67 ± 19.41
	P5	191.09 ± 15.89
	P6	65.88 ± 3.53
	P7	154.39 ± 14.67
Brick tea	B1	477.87 ± 4.20
	B2	208.93 ± 7.34
	B3	250.15 ± 16.51
	B4	414.89 ± 19.36
	B5	163.18 ± 27.88

Figure 1. Effects of 15 Pu-erh tea extracts at various concentrations on relative viability of SMMC-7721 cells 24 h after treatment (n = 5).

Comparison of the contents of major active compounds of various aged Pu-erh teas

Previous reports showed that levels of active compounds in Pu-erh tea are affected by raw materials with varied cultivation environments, different manufactures, storage conditions, and so on, leading to the differences of partly beneficial health effects. The contents of total polyphenols, total flavonoids, total catechins, free amino acid, total soluble saccharides, purine alkaloids, and GA in the 15 aged Pu-erh teas with long-tern storage were determined (Table 4).

Table 4. The contents of the total polyphenols, total flavonoids, total catechins, free amino acid, total soluble saccharides, purine alkaloids, and gallic acid.

Pu-erh tea	Number			mg/g DW
		Total polyphenols	Total flavonoids	Free amino acid	Total solublesaccharides	Gallic acid	Totalcatechins	Purinealkaloidal
Bowl tea	T1	47.02 ± 0.30^c	14.04 ± 0.61^ab	3.86 ± 0.25^bc	26.46 ± 1.86^ab	0.17 ± 0.01^ij	1.94	20.59
	T2	38.82 ± 0.44^ef	10.66 ± 0.32^g	3.88 ± 0.03^bc	21.76 ± 1.15^bc	0.60 ± 0.01^hi	1.43	18.69
	T3	58.92 ± 0.71^b	11.79 ± 1.40^cd	4.15 ± 0.14^a	22.47 ± 0.54^bc	10.12 ± 0.3^a	0.9	26.32
Cake tea	P1	29.66 ± 1.70^h	12.16 ± 0.67^cd	3.79 ± 0.04^cd	14.72 ± 3.12^d	1.41 ± 0.35^g	0.29	23.8
	P2	41.50 ± 0.49^de	12.29 ± 1.33^cd	3.67 ± 0.07^cd	27.64 ± 0.66^a	2.60 ± 0.34^e	1.15	26.55
	P3	63.91 ± 3.22^a	12.72 ± 1.08^bc	3.69 ± 0.01^cd	29.04 ± 4.12^a	5.81 ± 0.82^b	0.96	27.18
	P4	32.33 ± 1.73^h	14.26 ± 1.77^a	3.51 ± 0.19^ef	14.30 ± 1.32^d	0.28 ± 0.05^ij	0.89	28.29
	P5	25.76 ± 1.81^i	10.90 ± 0.37^fg	3.42 ± 0.07^g	6.20 ± 1.38^e	0.14 ± 0.02^j	0.43	26.45
	P6	37.56 ± 1.72^fg	12.86 ± 0.98^bc	3.87 ± 0.02^bc	22.26 ± 4.75^bc	1.51 ± 0.30^g	1.02	26.23
	P7	23.96 ± 1.09^i	11.25 ± 1.06^ef	3.58 ± 0.02^de	13.25 ± 1.31^d	0.91 ± 0.07^h	0.63	30.29
Brick tea	B1	25.21 ± 0.40^i	11.45 ± 0.41^de	3.90 ± 0.14^bc	13.83 ± 1.50^d	0.54 ± 0.08^hi	0.70	19.3
	B2	42.99 ± 1.81^d	13.02 ± 0.41^ab	4.04 ± 0.21^ab	21.42 ± 0.12^b	3.09 ± 0.72^d	0.73	21.90
	B3	35.34 ± 3.30^g	12.94 ± 0.311^ab	3.61 ± 0.15^de	28.75 ± 0.42^a	4.01 ± 0.19^c	0.98	23.46
	B4	17.34 ± 1.88^j	8.99 ± 1.41^h	3.45 ± 0.02^fg	6.16 ± 0.96^e	0.13 ± 0.01^j	0.27	22.94
	B5	64.33 ± 3.98^a	13.90 ± 0.34^ab	3.74 ± 0.02^cd	17.46 ± 1.23^cd	2.06 ± 0.26^f	0.72	23.7

The difference letter in the same row indicated that the difference among the tea samples is significantly through Duncan test (p < 0.05).

The brick tea B5 (45-year storage) was found to have the highest total polyphenol content (64.33 ± 3.98 mg/g), followed by the cake tea P3 (75-year storage, 63.91 ± 3.22 mg/g), compared with the total content of polyphenol of 17.34–58.92 mg/g in other 13 tea tested samples. The contents of total catechin in the 15 aged Pu-erh teas were 0.27–1.94 mg/g, which are significantly lower than that of 38.69 ± 16.51 mg/g in aged pu-erh teas (6-years storage) reported by Zhang et al.^[40] The highest total catechins content (1.94 mg/g) was found in the bow l tea T1, whereas the lowest level (0.27 mg/g) was observed in brick tea B4 with the similar storage times (75-years storage). Low content of total catechins in present study is observed because tea catechins could be oxidized into orthoquinones, which could be condensed to further form theaflavins, thearubigins or others compounds during their long-term preservation process.^[40]

The bowl tea T3 (27-year storage) had the highest GA content with 10.12 mg/g, which was 78 times higher than that of brick tea B4 (75-year storage, 0.13 mg/g). The GA contents in other 13 Pu-erh teas were 0.14–5.81 mg/g. All the GA levels of 15 aged Pu-erh teas are lower than the one (18.27–68.25 mg/g) of 6 aged Pu-erh teas reported by Qian et al ^[10]. The total flavonoid, free amino acid, purine alkaloid and soluble saccharide content in the 15 Pu-erh tea samples were 8.99–14.26, 3.42–4.15, 18.69–30.29, and 6.16–29.04 mg/g, respectively.

15 aged Pu-erh teas were divided in to four storage stages (10–30, 40–60, 70–90, >100 years), and the average contents of important chemical components were summarized in Table 5. The average contents of total polyphenols, total flavonoids, GA, and total catechins in aged Pu-erh tea at different storage stages showed no significant difference (p > 0.05). The results indicated that the youngest Pu-erh tea did not possess the highest level of active components.

Table 5. Average contents of the main active compounds of Pu-erh tea at four different storage stages.

Storage time	Number of samples	mg/g DW
		total polyphenols	total flavonoids	gallic acid	total catechins
Short turn (10–30 year)	P5, P6, P7, T3	36.55 ± 16.08^a	11.70 ± 0.85^a	3.17 ± 4.66^a	0.75 ± 0.27^a
Middle turn (40–60 year)	B5, P4	48.33 ± 22.63^a	14.08 ± 0.25^a	1.17 ± 1.26^a	0.81 ± 0.12^a
Longer turn (70–90 year)	B3, B4, T1, T2, P1, P2, P3	39.08 ± 14.50^a	11.97 ± 1.66^a	2.10 ± 2.16^a	1.00 ± 0.60^a
Long turn (>100 year)	B1, B2	34.10 ± 12.57^a	12.24 ± 1.11^a	1.82 ± 1.80^a	0.72 ± 0.02^a

The letter “a” in the same row indicated that there is no significant difference among the tea samples (p > 0.05).

The coefficient of correlation (r) between the IC₅₀ values of radical-scavenging and anticancer activities and the contents of the active components in the aged Pu-erh teas were analyzed (Table 6). The results indicated that the total polyphenol and catechin content were significantly correlated with the DPPH radical scavenging activities of all of the samples (p < 0.05). While the total polyphenol and GA content were significantly correlated with the ABTS radical scavenging activities of all of the samples (p < 0.01), only the GA content was correlated with the anti-cancer abilities across all samples (p < 0.05).

Table 6. Correlation coefficients between antioxidant and anticancer activities and the main active compound contents.

	Correlation coefficientof IC50 (DPPH)	Correlation coefficientof IC50 (ABTS)	Correlation coefficientof IC50 (MTT)
Total polyphenols	0.585*	0.770**	–0.399
Total flavonoids	0.321	0.467	–0.124
Gallic acid	0.257	0.666**	–0.561*
Total catechins	0.676**	0.136	0.251

Significant correlation by Student t-test.

*p < 0.05; **p < 0.01.

Conclusion

Generally, the older the Pu-erh tea, the higher commercial value it possesses. Our results indicated that the anti-oxidant and anti-cancer activities of the 15 aged Pu-erh teas with different storage years displayed minor differences. Not all the older Pu-erh teas exhibit stronger biological ability than the younger ones and the oldest Pu-erh tea did not have the lowest content of active components. The value of Pu-erh tea is affect by the taste, odor, beneficial component amounts, and others. Thus, to assess the value of Pu-erh tea preserved at different times based on our research results is not enough. Forty-three active components were identified in the 15 Pu-erh tea samples by the HPLC-ESI-QTOF-MS/MS method, including 7 flavan-3-ols, 11 organic acids and esters, 3 proanthocyanidin dimers, 2 benzotropolones, and 20 flavonoid glycosides. Importantly, our results demonstrated that: (1) both total polyphenol and catechin content were significantly correlated with the DPPH radical scavenging activities; (2) the total polyphenol and GA content were significantly correlated with the ABTS radical scavenging activities; and (3) only GA content was correlated with anti-cancer abilities. Anyway, this study could provide useful information of chemical constituents and bioactivity of aged Pu-erh teas with different storage years for consumers to understand the value of aged Pu-erh teas deeply.

Acknowledgments

The authors would like to thank Ji-Duan Qiu at Xiamen Kemingyuan Tea Shop for supplying the Pu-erh tea samples.

Funding

The authors are grateful for financial support from the Special Fund of the Fujian Key Science and Technology Special Projects—Key Agricultural Science and Technology Special Project (No. 2015NZ0003) and the Fujian Special Fund for Scientific Research Institutes in the Public Interest (No. 2016R1017-3).

Additional information

Funding

The authors are grateful for financial support from the Special Fund of the Fujian Key Science and Technology Special Projects—Key Agricultural Science and Technology Special Project (No. 2015NZ0003) and the Fujian Special Fund for Scientific Research Institutes in the Public Interest (No. 2016R1017-3).