An Improved Method of Theabrownins Extraction and Detection in Six Major Types of Tea (Camellia sinensis)

Department of Tea Science, Anxi College of Tea Sciences, Fujian Agriculture and Forestry University, Fuzhou 350002, China Department of Foods and Pharmaceutical Engineering, Wuzhou University, Wuzhou 543002, Guangxi, China Faculty of Applied Sciences, Ton Duc ang University, Ho Chi Minh 700000, Vietnam Department of Horticulture, Guangxi University, Nanning 530004, Guangxi, China Fujian Provincial University Key Laboratory of Green Energy and Environment Catalysts, College of Chemistry and Materials, Ningde Normal University, Ningde, Fujian 352100, China


Introduction
Tea is either prepared by infusion or decoction of Camellia sinensis dried leaves, which are classi ed into six major types (green, oolong (or cyan), black, white, yellow, and dark) based on the manufacturing process. Both black and oolong teas are two kinds of aerated tea leaves. After plucking, these fresh leaves are processed by withering, rolling, aeration, and then inactivation by drying (Figure 1(a)). During the aeration process, polyphenols are oxidized by endogenous enzymes such as polyphenol oxidase and peroxidase, thus producing many oxidative compounds. Polyphenol oxidation results in the pigment's dynamic transition into tea pigments that include TFs, TRs, and TBs [1,2]. e relationship among TFs, TRs, and TBs is shown in Supplementary Figure 1. e transition induces color changes in tea leaves ranging from green to red brown [3]. All the pigments a ect the color and avor of liquid tea [3][4][5][6][7]. TBs give tea leaves a brown color and can be formed by TFs or TRs oxidation or polyphenol aggregation with other sugars and acidic compounds [8,9]. e leaves of oolong tea are green with red edges (Figure 1(b)III) due to the formation and accumulation of oxidized polyphenols from leaf margins during the aeration process. As the aeration period proceeds, the surfaces of all tea leaves become reddish brown (Figure 1(b)I) and black tea leaves are produced. erefore, oolong tea is "semioxidized" tea, and black tea is "fully oxidized" tea [10]. White tea is obtained by only withering and drying, as more prolonged withering leads to a severe water deficiency that induces membrane disintegration and polyphenol oxidization by endogenous enzymes. In addition, white tea can be stored for a long time, during which chemical reactions such as catechin and amino acid oxidation can occur that induce better flavor and taste (more sweetness and smoothness) and more health benefits [11][12][13]. e objective of our study was to "improve" ( Figure 2) the abovementioned "traditional" method by adding an ethyl acetate extraction step between liquid tea production and n-butanol extraction. Moreover, our method's new equations for the TB content (%) of the six major types of tea extraction had different absorbance values at 380 nm (A 380 ) and slope compared to black tea analyses by the traditional method.

Samples.
e six major types of tea samples obtained are listed in Table 1. All were purchased from local tea markets located in Anxi, Fujian Province, Dali, Yunnan Province, and Chongqing City, China. Each tea sample was ground into tea powder using a grinder (Joyoung Co., Shandong, China) and then stored in sealed cans until analysis.

Preparation of Liquid Tea Samples.
Boiling distilled and deionized water (d.d. H 2 O; 125 mL) was added to 3.0 grams of tea powder in a 250 mL conical flask and shaken in a water bath at 90°C for 10 min. e liquid tea was then centrifuged at 1,800 × g for 10 min at 4°C. e supernatant was diluted to 125 mL with d.d. H 2 O and then divided into two aliquots, one for TB extraction by the traditional method and the other by our improved method ( Figure 2).

TB Extraction.
e traditional method was based on Yao et al. [33] and Roberts and Smith [35,36], with a few modifications. Briefly, 25 mL of tea solution was extracted with the same volume of n-butanol (Xilong Scientific Co., Guangdong, China) and shaken for 3 min. e lower layer of the solution was then centrifuged at 1,800 × g for 10 min at 4°C. Two mL of the supernatant was then mixed with 2 mL of saturated (10.2%) oxalic acid (SINOPHARM Co., Shanghai, where "w" is the weight (gram) of the tea sample, "M" is the moisture content (%) of the tea sample, and 21.18 is the inverse slope of A 380. e improved method with TB concentrations from TB extracted samples was modified from the abovementioned traditional method. Mainly, 35 mL of ethyl acetate (SINO-PHARM Co., Shanghai, China) was added to samples of the six tea types (35 mL) in a 125 mL separating funnel and shaken for 3 min. After discarding the upper layer, the lower layer (25 mL) was then extracted with 25 mL of n-butanol (Xilong Scientific Co., Guangdong, China) and shaken for 3 min; the lower layer solution was used for centrifugation at 1,800 × g for 10 min at 4°C. e A 380 detections of the solutions were done the same way as in the abovementioned traditional method, and TB contents (%) were determined by the six equations for the improved method. TB in the six tea types was extracted by the improved method and then freeze-dried. Freeze-dried samples were dissolved in d.d. H 2 O, various concentrations of TB solution (2 mL) were mixed with 2 mL of saturated oxalic acid, 6 mL of d.d. H 2 O was added, and the samples were then diluted to 25 mL with 95% (v/v) ethanol. Samples were then measured at A 380 .

Moisture.
Moisture in the tea powder was measured using a vacuum oven according to an international standard method (ISO1573 (BS6049-2), 1980).

Data Analysis. Paired data with
A 380 values of both traditional and improved methods were subjected to paired t-tests using Microsoft Excel 2019. TB contents (%) are presented as mean values ± standard deviations (SD) of twelve independent sets of experiments with similar results. Paired t-tests were calculated with high significance at p ≤ 0.01using SPSS version 23.0 (SPSS, Chicago, USA). Linear equations were established by regression analysis between A 380 measurements and TB concentrations of the six tea types using SigmaPlot ver. 12.5 (SYSTAT Software, San Jose, CA). Figure 3 illustrates A 380 values from TB solutions extracted by traditional and improved methods. Readings from the improved method were significantly lower (around 80-90%) compared to the traditional method in all six tea types, indicating that ethyl acetate extraction removes TFs and portions of TRs from liquid tea [33] and decreases TB A 380 values. TB compositions were different between these extraction methods, and revised parameters for the improved method should therefore be established.

Comparisons of TB Extractions between Traditional and Improved Methods.
After plucking, yellow, green, and dark tea leaves are first fixed by steaming or pan-frying (Figure 1(a)) to inactivate all endogenous enzymes. Green tea is produced after rolling and drying. Yellow tea is obtained when tea leaves are kept wet and under high temperatures for 6 to 12 h (Figure 1(a)) between fixing and drying, during which chlorophylls are degraded and polyphenols are auto-oxidized. Dark tea is harvested after leaves have been kept wet and microorganisms grow on their surfaces for many days between fixing and drying (Figure 1(a)). During this piling process, microbes (mainly Aspergillus fumigatus, Aspergillus Niger, and Saccharomyces cerevisiae) secrete many enzymes to induce polyphenol oxidation, cell wall degradation, and fermentation [14][15][16][17]. TB content increases during dark tea processing [18], and the leaves become dark brown (Figure 1(b)).
Similar plant secondary metabolites can reduce the risk of age-related chronic diseases and promote health benefits [19,20]. TBs have physiological functions such as reducing blood lipid and blood sugar levels [21][22][23]; controlling of diabetes mellitus [24]; attenuation of hypercholesterolemia [25,26]; reducing serum levels of total cholesterol, lowdensity cholesterol, and triglycerides [27]; and osteoclastogenesis suppression and prevention of bone loss [28], together with inhibition of cell cycling and tumor cell growth [29]. In addition, TB content is a positive parameter in evaluating fragrance and flavor in dark tea and white tea [9,18,30,31]. Zhu et al. [32] and Cheng et al. [18] reported that stringent taste levels were decreased and stale and fungal aromas increased with TB content. Furthermore, dark tea and white tea leaves can be stored for a long time, and longer storage times result in higher TB content [14]. In contrast, during storage, an increased TB content worsens tea quality in both black tea and oolong tea due to the loss of aroma and sweetness molecules [5,7,33]. erefore, TB content may be an objective measure of tea quality [34]. Yao et al. [33] combined three methods [35][36][37] to analyze TF, TR, and TB contents and constructed a "traditional" method of TB extraction ( Figure 2). However, TBs are defined as a variety of water-soluble compounds but are not dissolved in n-butanol and ethyl acetate [22,27]. In the traditional method, without ethyl acetate extraction, TFs or TRs would be retained in the TB layer solution and thus affect the accuracy of TB content analysis, because A 380 is detected from TF, TR, and TB solutions. In addition, the TB equation was based on black tea samples extracted from the traditional method [33,35,36]. Whether the slop parameter of the black tea equation is the same as other types of tea remains unknown.

Equations Established for the Improved Method.
e traditional method's empirical equation for determining TB content (%) is [21.18 × 2 × A 380 × 100%]/w (1-M), where "w" stands for the weight (in grams) of tea sample powder and 21.18 is the inverse slope of A 380 based on black tea [33,35,36]. In our study, six major types of tea were extracted by the improved method, their solutions were freeze-dried, and TB powder was collected. After TB powders were dissolved in d.d. H 2 O, different A 380 values were observed from the samples. Figure 4 shows that A 380 values were significantly (p < 0.0001) and positively correlated with the TB concentrations of green tea (r = 0.9963, R 2 = 0.9926), yellow tea (r = 0.9961, R 2 = 0.9922), oolong tea (r = 0.9977, R 2 = 0.9954), black tea (r = 0.9987, R 2 = 0.9975), white tea (r = 0.9973, R 2 = 0.9947), and dark tea (r = 0.9975, R 2 = 0.9950).
Regression equation slopes for dark tea, white tea, black tea, oolong tea, yellow tea, and green tea were 0.0704, 0.0475, 0.0381, 0.0350, 0.0166, and 0.0160, respectively. Furthermore, the inverse slopes of dark tea, white tea, black tea, oolong tea, yellow tea, and green tea were 14.205, 21.053, 26.247, 28.571, 60.241, and 62.5, respectively. e equations for the six tea types using the improved method were as follows: dark tea:    In those empirical equations, "M" means the tea sample's moisture content (%) and "w" stands for the weight (in grams) of tea sample powder. Figure 4 shows that two types of aerated tea (black tea and oolong tea) had similar variations, but green tea and yellow tea almost completely overlapped, indicating that the TB compositions were similar in these two types of tea. Wang et al. [38] demonstrated that green tea and yellow tea had similar chemical components according to metabolome analysis. After hightemperature treatments in fixation and drying processes during storage, isomer flavanols and chlorophyll metabolites (e.g., pheophytins) were produced in the autooxidation process, and the color of tea leaves turned olive brown [39][40][41][42][43]. e slope of white tea was higher than that of black tea and lower than that of dark tea (Figure 4), implying that the TB composition in white tea is notably different from that in dark tea and black tea. Although liquid chromatography-mass spectrometry analysis was used to determine the chemical composition in white tea after longterm storage [11,12,44], the TB composition still remained unknown. In addition, the TB of dark tea contains some fungi-specific metabolites detected by gas chromatographymass spectrometry or liquid chromatography-mass spectrometry-based metabolomics [18,23,32,[45][46][47][48][49].

Comparison of TB Content between Traditional and
Improved Methods in Tea Extractions. TBs were extracted by both traditional and improved methods, and calculations were made with appropriate equations at A 380 . Figure 5 shows that the TB contents of these two methods were significantly different (p < 0.01) in all types of liquid teas.
We repeated many studies by using the traditional extraction method and found that dark tea had the highest TB content (10%∼14%) of the six tea types [14,15,50,51], followed by black tea (7%∼9%) [14,50,52], and ranges of 2%∼3.5% in green tea and yellow tea [50,52]. e improved method showed that black tea had the highest TB content (7.97%∼11.19%, average 9.75%), and oolong tea ranged  2.65%∼6.23% with an average of 5.25%. Black tea and oolong tea were oxidized by endogenous enzymes in a longer aeration process and had higher TB contents. e TB contents of white tea ranged from 5.35% to 8.02%, averaging 7.03%, suggesting that four years of storage was sufficient to transfer high levels of TB content in the leaves. e TB contents of dark tea ranged from 5.66 to 7.66%, averaging 6.96%, while green tea and yellow tea ranged from 2.63 to 4.46% with an average of 4.03% and 3.87 to 5.2% with an average of 4.61%, respectively, indicating that their values differed significantly from levels derived from the traditional extraction method. Dark tea had the highest A 380 of the six tea types ( Figure 3); however, high slop in TB concentrations ( Figure 4) resulted in TB content not being as high as with the traditional method ( Figure 5). TB composition is formed during tea processing, and the aeration procedure is one of the most important steps in black tea and oolong tea processing, while it does not occur in other teas. erefore, different equations are proposed to calculate TB content in different types of tea, especially dark tea, yellow tea, and green tea.

Conclusion
is study provided an improved method for the analysis of TB content (%) from six major types of tea. is method decreases the A 380 values of TBs with differing TB absorbance capabilities in six major types of tea. Six equations have been developed to analyze TB content (%) in those six tea types extracted by an improved method, and the ranges of TB content (%) showed significant differences compared to the traditional method.

SD: Standard deviation
Data Availability e data used to support this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this article.

Authors' Contributions
T-C L. designed the experiments; T-C L. and Q-N Z. collected the sample, conducted experiments, and analyzed the data. T-C L. and K-H L. wrote the article; T-C L., Q-N Z., H-L H., P-Y L., J-Y Z., C-K., and Y-J L. collected the sample and investigated the study; T-H Ko projectedministration and reviewed and edited the manuscript.  Figure 5: Paired comparisons of TB contents (%) of six major types of tea extracts between the traditional method (a) and improved method (b). Each bar represents the mean ± SD of twelve independent measurements. Asterisk ( * * ) indicates high significance atp ≤ 0.01.
Journal of Chemistry 7