Gallic acid (GA), (−)-epicatechin gallate (ECG), (−)-epicatechin (EC), (−)-gallocatechin gallate (GCG), (−)-epigallocatechin (EGC), and (−)-epigallocatechin gallate (EGCG) were purchased from Wako Pure Chemical Ind. (Osaka, Japan). Myricetin, quercetin, kaempferol, croton oil, carrageenin, Evans blue, Cyclosporin A (Cs A), dexamethasone acetate (DEX), and LPS from Escherichia coli 055:B5 were purchased from Sigma Chemical Co. (St. Louis, Mo, USA). P. acnes (ATCC6919) used in this study had previously been prepared by our laboratory.

Tea flowers were procured from Guangdong Tea Experiment Substation, Yingde, Guangdong, China. Dried tea flowers were treated with 20 parts of boiling water for 5 min. After filtration and evaporation of water, the residue was powdered under frozen decompression. The recovery rate was 19.5%. The extract residue was immediately dissolved in water before the experiment.

The concentrations of tea catechins in tea flower buds was determined by HPLC under the following conditions: Cosmosil 5PE-MS column (Nakalai Tesuque, Kyoto, Japan, 4.6 mm × 150 mm, 5 μm); mobile phase, eluent A 0.05% trifluoroacetic acid (TFA) in water; eluent B 0.05% TFA in acetonitrile; flow rate, 2 mL/min; column temperature, 40°C; detection wavelength of 280 nm. The analysis was performed with a gradient program of 10% eluent B for 5 min, 21% for 8 min, and 90% for 6 min [¹⁰]. The quantification of individual catechins was determined by standard calibration curves calculated from those of authentic reference standards. The total flavonoid content was determined using a colorimetric method as described by Wu et al. [¹¹]. As shown in Table 1, EGCG, ECG, were identified as major polyphenol constituents of tea flower, while GCG and GA were present in smaller quantities. The total catechins and flavonoid contents in tea flowers were 37.46 and 11.61 mg/g, respectively. All the analyses have been performed in triplicate and the results are reported on wet basis.

Eight-week-old male KM mice were purchased from the Guangdong Medical Laboratory Animal Center (Guangzhou, China). All mice were kept in a pathogen-free animal room under controlled conditions at 23 ± 1°C. A 12 h light-dark cycle was maintained, with lights on from the time of 06:00 to 18:00. The mice were provided with standard laboratory diet and water. The animals were allowed to acclimatize to the environment for 1 week before the experiment. Procedures for animal experiments were conducted in accordance with the Guiding Principles for the Care and Use of Laboratory Animals as adopted and promulgated by the United States National Institutes of Health.

The croton oil ear test was performed according to [¹², ¹³]. A total of 25 μL of 0.3% croton oil (dissolved in acetone) was topically applied to the inner surface of the right ear of each mouse. DEX (1 mg/kg, served as positive control) and TFE (50 mg/kg, 100 mg/kg, and 200 mg/kg) were orally administered to mice 30 min prior to the application of croton oil. The left ear remained untreated as negative control. Three hours after croton oil being applied, 1% Evans blue (dissolved in saline) was injected via the tail vein at a dose of 10 mL/kg body weight. The animals were sacrificed 1 h later and both the treated and untreated ears were collected for further determination. The ears were soaked in 1 mL of 1 M KOH at 37°C for 24 h and 9 mL 0.6 M H₃PO₄-acetone (5 : 13) was added afterwards. The mixture was then centrifuged at 3000 rpm for 15 min. The absorbance of supernatants was detected at 620 nm [¹²]. The quantification of Evans blue in mice ear, as a major indicator of capillary permeability was determined by standard calibration curves.

The carrageenan-induced paw edema model in mice was used to study the effect of TFE on acute inflammation [¹⁴]. Mice were treated with DEX (1 mg/kg, served as positive control), TFE (50 mg/kg, 100 mg/kg, and 200 mg/kg) or the same volume of saline by oral administration. After 30 min, edema was induced by injecting 10 μL of carrageenan (2.0%, w/v) into subplantar area of the right hind paw. Measurement of paw size was carried out with caliper rule (Digimatic, Mitutoyo 500, Japan) 4 h following carrageenan injection. The extent of the edema was measured by the difference in size between the two hind paws.

TFE (50 mg/kg, 100 mg/kg, and 200 mg/kg) and Cs A (25 mg/kg, served as positive control) were orally administrated to mice for 7 days, while the normal control group and model group received saline only. Liver inflammation was induced by injecting heat-killed P. acnes in saline (0.2 mg/10 g body weight) via the tail vein 2 d after TFE, Cs A and saline administration. On the seventh day, LPS (5 μg/10 g weight) was injected to all mice intravenously except normal control group. The animals were sacrificed under anesthesia with diethyl ether for liver inflammation analysis 5 h after LPS injection [¹⁵]. Blood samples were taken from the hearts of the mice and transferred into a tube containing 2% sodium heparin. The tubes were centrifuged at 5000 rpm for 5 min and supernatants were used for following experiment. Plasma alanine aminotransferase (ALT) activity, which is a marker of hepatocyte injury, was determined by using commercial kits and absorbance was detected with MK3 microplate reader (Labsystems Co., Finland) at 492 nm.

For histologic analysis, small sections of fresh liver tissues were immediately fixed in 10% buffered formalin and embedded in paraffin wax. Tissue sections (4 μm) were stained with haematoxylin and eosin (H&E) and were examined for liver damage under light microscopy.

NO concentration in liver tissue was determined by the Griess reaction [¹⁶]. Briefly, liver tissues were homogenated in saline, a 40 μL 20% liver homogenate sample was transferred into 96-well microplates, and then 160 μL Griess reagent (1% sulfanilamide in 5% H₃PO₄ and 0.1% N-(1-naphthyl) ethylenediamine dihydrochloride) was added. The absorbance of the chromophore was read at 540 nm. A standard curve was constructed using known concentrations of sodium nitrite [¹⁷].

Total RNA was extracted from the liver samples by using Trizol Reagent according to the protocol of the manufacturer (Invitrogen, Carlsbad, CA, USA). The concentration of the extracted RNA was calculated by measuring the optical density at 260 nm. Aliquots of 2 μg RNA were transformed to first strand complementary DNA (cDNA) at 42°C for 1 h in a 20 μL reaction mixture containing mouse Moloney leukemia virus reverse transcriptase (Invitrogen, CA, USA) with oligo (dT)₁₅ primers (Tiangen, Beijing, China). PCR amplification was carried out with 1 μL of the resulting cDNA, 12.5 μL of 2× Taq reaction buffer (Tiangen, Beijing, China), 2 μL of dNTP mixture, 1 μM forward primer, 1 μM reverse primer, and 1 μL of Taq polymerase (Tiangen, Beijing) in a total volume of 25 μL. The PCR primers for mouse TNF-α (forward, 5′-GGCGGTGCCTATGTCTC-3′; reverse, 5′-GCAGCCTTGTCCCTTGA-3′), IL-1β (forward, 5′-GCTGGAGAGTGTGGAT-3′; reverse, 5′-CTTGTGAGGTGCTGATG-3′), β-actin (forward, 5′-ACCCAGATCATGTTTGAGACC-3′; reverse, 5′-CATACCCAAGAAGGAAGGCT-3′) were used. The thermocycler conditions were 94°C for 30 s, with an annealing temperature of 60°C for 40 s and an elongation temperature of 72°C for 50 s for the first 30 cycles, followed by an elongation temperature of 72°C for 7 min. After the reaction was completed, the amplified product was removed from the tubes and run on 2% agarose gel and visualized by ethidium bromide staining. The band intensity of ethidium bromide fluorescence was measured using an image analysis system (Bio-Rad, Hercules, CA, USA), then quantified with Quantity One analysis software (Bio-Rad, Hercules, CA, USA), and expressed as the ratio to β-actin.

The data were presented as mean ± S.D. and analyzed statistically by one-way analysis of variance (ANOVA) followed by Dunnett's significant post-hoc test for pair-wise multiple comparisons. Differences in data of biochemical parameters among the different groups were considered to be statistically significant at P < 0.05.

Topical application of croton oil-induced edema at the ears of mice and significant increase in the weight of treated right ear when compared with the untreated left ear. As shown in Table 2, the oral administration of TFE produced a dose-dependent suppression of ear edema in mice. The edema inhibitory rates of TFE were 5.65%, 13.04%, and 25.40% at the dose of 50, 100, and 200 mg/kg, respectively. Whereas DEX (1 mg/kg) gave rise to a significant inhibition of 90.79% in ear weight compared to control. Croton oil-induced capillary permeability leaded to the increase of Evans blue leakage in the mice ear. TFE at doses of 100 and 200 mg/kg showed significant inhibitory effects on croton oil-induced capillary permeability (Table 2), with 21.76% and 25.13% inhibition of dye leakage at doses of 100 mg/kg and 200 mg/kg, respectively. DEX (1 mg/kg) gave a significant inhibition of dye leakage at 90.16%.

In the carrageenan-induced edema test, the paw sizes and percentages of inhibition by TFE and DEX are shown in Table 3. A maximum increase of paw thickness, 1.22 ± 0.15 mm, was obtained 4 h following carrageenan injection in control mice. TFE significantly decreased the carrageenan-induced edema at doses of 200 mg/kg and achieved its maximal inhibitory effects of 19.67%. At a dose of 1 mg/kg, DEX produced a greater inhibition of edema by 69.67%.

The effect of TFE on P. acnes plus LPS-induced liver inflammation was investigated by the histologic analysis and plasma ALT level. As shown in Figure 1, there was no obvious pathological abnormality in normal control group, liver parenchyma was in good morphology and hepatocytes were arranged around the central vein, no congestion and inflammation were observed in the sinusoids (Figure 1(a)). Compared with normal mice, liver tissue sections from P. acnes plus LPS challenged mice showed severe pathological alterations, such as confusion of hepatic lobule structure with inflammatory cell infiltration, serious vacuolation of hepatic cells, necrosis and erythrocyte influx (Figure 1(b)). However, in TFE (50, 100, and 200 mg/kg) treated groups and Cs A group, the area and extent of necrosis were attenuated and the immigration of inflammatory cells was reduced (Figures 1(c)–1(f)). As shown in Figure 2, plasma ALT activity in P. acnes plus LPS-treated mice was significantly higher than the normal group (P < 0.01). TFE significantly reversed the ALT activity by 16.49%, 46.05%, and 49.42% at doses of 50 mg/kg, 100 mg/kg, and 200 mg/kg, respectively. These data indicate that TFE attenuates liver inflammation induced by P. acnes plus LPS injection.

NO is a common inflammatory mediator involved in the pathogenesis of inflammation. In this study, the effect of TFE on P. acnes plus LPS induced liver inflammation was also investigated by determining the NO production in liver tissues. As shown in Figure 3, compared with the normal control mice, the concentration of NO was markedly increased in model group (P < 0.01). However, significant concentration-dependent inhibition of NO production was detected when mice were treated with TFE (100, 200 mg/kg), with inhibitory of 28.6% and 34.9%, respectively.

To further explore the mechanisms of the protective effects of TFE on immunological liver inflammation induced by P. acnes and LPS, mRNA expression of TNF-α and IL-1β mRNA in liver tissue were determined. Compared with the normal control group, treatment with P. acnes and LPS led to significant increases in TNF-α and IL-1β mRNA levels (Figure 4). However, this effect was reversed in mice treated with TFE (50, 100, and 200 mg/kg), with similar result from mice treated with Cs A.