Effect of tea infusion-derived pectin on the astringent taste of catechins

Abstract

We examined the effect of pectin refined from sen-cha tea infusions on the astringency of catechins using a taste sensor system. Pectin was fractionated from tea infusions, and the pectin of a comparable size to the pectin that was observed to increase in amount in a steaming-time-dependent manner in tea infusions (first decoction) was obtained. The tea infusion-derived pectin significantly suppressed the strong initial astringency imparted by gallate-type catechins, EGCg and ECg, and the weak astringency imparted by non-gallate-type catechins, EGC and EC. On the other hand, the tea infusion-derived pectin did not suppress the residual catechin-derived astringency. When the tea infusion-derived pectin was added to the tea infusion, a weak, but concentration-dependent inhibitory effect on the initial astringency was observed.

Introduction

Of the two types of sen-cha, fukamushi-cha is a deep-steamed tea that is steamed more than two to three times longer than normal-steamed tea; the production method was established in the 1960s to alleviate the astringent taste of tea. The content of catechins, which are responsible for the astringency, in tea leaves does not change significantly with steaming time, whereas the content of water-soluble pectin (WSP) does. According to the results of a recent study (Hirono et al., 2020), the concentration of pectin in an infusion of fukamushi-cha was about two and three times higher than that of normal-steamed tea in the first and second decoction, respectively, which were compared under general drinking conditions using a teapot. Pectin, which is increased by deep steaming, is a kind of polysaccharide that has no taste of its own. However, following the report on the masking effect of pectin on persimmon astringency (Taira et al., 1997), it has been reported that catechin astringency is reduced by the complexation of gallate-type catechins with pectin from citrus (Hayashi et al., 2005).

The content of WSP in tea leaves is increased not only by steaming, but also by roasting (Hirono and Mizukami, 2018). This is thought to be due to the β-elimination reaction that occurs during heating, changing water-insoluble pectin into WSP. However, since the WSP itself decomposes at the same time, the molecular weight distribution of pectin in the tea infusion changes significantly to a lower molecular weight distribution after heating above 140 °C, such as in hoji-cha.

In this study, we first showed that in the case of deep steaming, the molecular weight distribution of pectin in the tea infusion did not change significantly in the first decoction compared to normal steaming, unlike that observed for roasting. Then, as the main focus, pectin was isolated from the infusion of normal-steamed tea, and its effect on the astringent taste of catechins was directly verified using a taste sensor.

A taste sensor is a device that mimics the taste recognition mechanism of an organism. In living organisms, when a taste substance is adsorbed to the membrane surface of taste cells on the tongue, a potential change occurs in the cell membrane. The information of this potential change is transmitted to the cerebrum, which recognizes it as taste. In the taste recognition system developed by Intelligent Sensor Technology Co., Ltd., taste substances are adsorbed onto artificial lipid membranes in place of taste cells, and a sensor that responds specifically to each basic taste detects the membrane potential change and sends a signal to a computer. We can then analyze the intensity of the taste from the sensor output value (Ikezaki, 2011). In this paper, we used a sensor specific to the astringent taste to verify the effect of tea infusion-derived pectin on the astringent taste of catechins.

Materials and Methods

Preparation of tea infusions    Tea leaves were cultivated at the National Institute of Fruit Tree and Tea Science plantation in Kanaya-Shishidoi, Shizuoka, Japan. Tea leaves ('Yabukita’ cultivar) were harvested during the first crop season in April 2018. Harvested tea leaves were immediately steamed for about 50 s, then processed into sen-cha (unrefined tea). Six grams of unrefined tea leaves (not powdered) were infused with 600 mL of hot water, immediately after boiling, using a GV-3 teapot (Celec Co., Ltd., Tajimi, Japan) and left at room temperature for 5 min. The infusion without tea leaves was freeze-dried, and the lyophilized concentrate was then dissolved in distilled water. A precipitate was formed by the addition of ethanol to a final concentration of 70% ethanol, and was then suspended in 10 mM MES (pH 5.6).

Anion-exchange chromatography    The sample in 10 mM MES (pH 5.6) was centrifuged, then the supernatant was filtrated using a Minisart Plus Syringe Filter (Sartorius Stedim Biotech, Goettingen, Germany). The filtrate was applied to a 20-mL HiPrep 16/10 DEAE FF column (GE Healthcare, Little Chalfont, UK) equilibrated with 10 mM MES (pH 5.6). After washing with 200 mL of 10 mM MES (pH 5.6), the column was developed with a 0-1 M NaCl gradient (200 mL). The eluted fractions (10 mL each) were collected using a CHF122SC fraction collector (Advantec Toyo Kaisha, Ltd., Tokyo, Japan). Determination of the pectin concentration in the eluted fractions was conducted using the 3,5-dimethylphenol method (Scott, 1979), a colorimetric test with D-(+)-galacturonic acid as a standard. The content of total neutral sugars (D-glucose equivalent) in the same fractions was determined by the phenol-sulfuric acid method (Dubois et al., 1956), a colorimetric test with D-(+)-glucose as a standard, after correction for interference from galacturonic acid.

Size exclusion chromatography    Fractions containing pectin, in which the polysaccharides composed of neutral sugars were eliminated by anion-exchange chromatography, were collected, and then desalted and condensed using a Vivaflow 200 Laboratory Cross Flow Cassette (Sartorius Stedim Biotech). The desalted solution was freeze-dried and the lyophilized concentrate was dissolved in a mixture of 0.1 M sodium acetate and 0.15 M sodium chloride aqueous solution (pH 5.6), then applied to a HiPrep 26/60 Sephacryl S-500 HR column (GE Healthcare). Size exclusion chromatography was performed as described previously (Hirono and Mizukami, 2018). Fractions (3.0 mL) were collected and analyzed for galacturonic acid content by the 3,5-dimethylphenol method, while measuring ultraviolet absorbance at 260 nm and 280 nm for the measurement of concentrations of nucleic acids and proteins. Fractions containing pectin, in which nucleic acids or proteins had been eliminated by size exclusion chromatography, were collected, and then desalted and condensed using a Vivaflow 200 Laboratory Cross Flow Cassette. The desalted solution was further desalted and condensed using a Vivaspin 20 (Sartorius Stedim Biotech), and used as the tea infusion-derived pectin in this study.

Monosaccharide analysis    The tea infusion-derived pectin was hydrolyzed in 8 M trifluoroacetic acid at 100 °C for 3 h. Then, the acid hydrolysate was labeled with aminobenzoic acid ethyl ester and analyzed by high-performance liquid chromatography (HPLC). A PN-PAK C18 column (3.0 × 75 mm; Protenova Co., Ltd., Higashikagawa, Japan) was used with 0.2 M potassium borate buffer (pH 8.9)/acetonitrile solvent at a flow rate of 0.5 mL/min at 26 °C, and detection was performed using a fluorescence monitor (excitation at 305 nm, emission at 360 nm).

Astringency measurement using the taste sensor system    Evaluation of the astringency of catechin solutions or tea infusions was conducted using the SA402B taste sensor system (Intelligent Sensor Technology Co., Ltd., Atsugi, Japan) fitted with a sensor probe for astringency (SB2AE1) and a reference probe as described by Hayashi et al. (2005). In this study, two types of astringency, i.e., the initial astringency and the residual astringency, were evaluated. The initial astringency is the taste we sense the moment we take a drink. On the other hand, the residual astringency is the taste we experience after swallowing. First, the electrical potential in the standard solution (30 mM KCl, 0.3 mM tartaric acid, pH 3.5) corresponding to human saliva (Vr) was measured. Then, the electrical potential in the sample solution (Vs) was measured, and the electrical potential difference (Vs - Vr) was defined as the initial astringency. After washing the sensor with standard solution, the electrical potential in the standard solution (Vr′) was once again measured, and the electrical potential difference (Vr′ - Vr) was defined as the residual astringency. The average of three measurements, excluding the initial measurement cycle, was adopted as the electrical potential for each sample solution. The catechin solutions were prepared as follows. Catechins (EGCg, ECg, EGC, and EC) at a concentration of 0.65 mM were respectively dissolved in a 4.7 mM KCl aqueous solution, then divided into two groups: one containing the tea infusion-derived pectin-additive (60 mg/L) and the other containing a distilled water-additive in place of the pectin. For EGCg and ECg, the tea infusion-derived pectin was added at various concentrations (0, 20, 40, and 60 mg/L). The tea infusion for the taste sensor was prepared as follows. The same tea leaves that were used to isolate the pectin were used. Tea leaves (5.0 g) were infused with 150 mL of water at 80 °C for 60 s (Hirono et al., 2020). The concentrations of pectin and catechins in the tea infusions were determined by the 3,5-dimethylphenol method and HPLC method (Hirono et al., 2020), respectively. To this infusion, the tea infusion-derived pectin was individually added at 0, 20, 40, and 60 mg/L, and the samples were applied to the sensor.

Results and Discussion

Changes in the molecular weight distribution of WSP in tea leaves and pectin in tea infusions according to steaming time    We previously reported the changes in the WSP content of unrefined tea leaves and tea infusions according to steaming time (Hirono et al., 2020). In the present study, we report the changes in the molecular weight distribution of WSP according to steaming time (Fig. 1). The preparation method of tea leaves and tea infusions was as reported by Hirono et al. (2020) and the molecular weight determination method was as reported by Hirono and Mizukami (2018). These data showed that higher molecular weight pectin from whole WSP in tea leaves was present in the infusions. Furthermore, the first decoction contained roughly equivalently sized pectin with peaks at elution volumes of around 150–160 mL that increased with increasing steaming time.

Fig. 1.

Changes in molecular weight distribution of water-soluble pectin in tea leaves (a), and pectin in the first (b) and second (c) decoctions of unrefined teas according to steaming time. Steaming times are indicated in the figures. Arrows in the figures show molecular weight markers (dextrans).

Anion-exchange chromatography    We attempted to obtain refined pectin from tea infusions, in which the size of the pectin was the same as that in the actual drinking form of tea, based on the above results. The solution of high molecular weight substances, precipitated by adding 70% ethanol to the tea infusions, was applied to an anion-exchange resin. The chromatographic elution profile is shown in Fig. 2. Total neutral polysaccharides that did not adsorb to the resin were eluted at an elution volume of 20–40 mL. Pectin, acidic polysaccharides, and neutral sugars as a side chain of pectin that adsorbed to the resin were eluted at an elution volume of 250–330 mL. These pectin fractions, free of neutral polysaccharides, were collected.

Fig. 2.

Anion-exchange chromatography elution profile of the tea infusion.

Size exclusion chromatography    To further eliminate other high molecular weight substances from pectin, size exclusion chromatography was performed using a HiPrep 26/60 Sephacryl S-500 HR column. The chromatographic elution profile is shown in Fig. 3. Measurement of concentrations of nucleic acids and proteins was conducted by measuring ultraviolet absorbance at 260 nm and 280 nm, respectively. Fractions eluted at an elution volume of 120-210 mL, including a peak around 160 mL, which were free of nucleic acids and proteins and showed a similar size to the pectin that increased in amount with increasing steaming time (shown in Fig. 1), were collected.

Fig. 3.

Size exclusion chromatography elution profile of the tea infusion.

Arrows in the figure show molecular weight markers (dextrans).

Monosaccharide analysis    The tea infusion-derived pectin was hydrolyzed by acid treatment and the monosaccharaide composition was determined. HPLC analysis of the aminobenzoic acid ethyl ester-labeled sample revealed the presence (mol%) of galacturonic acid (77.55), arabinose (8.11), galactose (6.32), rhamnose (2.56) and glucose (2.37). Although the compositional ratio was different, the major sugar composition was similar to the monosaccharide composition of pectin fractions obtained from green tea leaves under acidic extraction (Ele-Ekouna et al., 2011).

Astringency measurement using the taste sensor system    Fig. 4(a) shows the potential difference (Vs - Vr), defined as the initial astringency, for the four kinds of catechins (as the major catechins) in the tea infusions. Fig. 5(a) shows the electrical potential difference (Vr’ - Vr), defined as the residual astringency, for the four kinds of catechins. As the degree of astringency increased, the potential difference decreased (the negative value increased). First, among the four catechins, EGCg and ECg, which are gallate-type catechins (around −100 to −90 mV in the initial astringency and around −45 mV in the residual astringency), were shown to be stronger in both the initial and residual astringency than EGC and EC, which are non-gallate-type catechins (around −20 mV in the initial astringency and less than −5 mV in the residual astringency). This was consistent with a previous report in which gallate-type catechins exhibited strong astringency among catechins (Nakagawa, 1972). The tea infusion-derived pectin suppressed the initial astringency of all catechins [Fig. 4(a)]. In particular, it significantly (more than 10 mV) suppressed the strong initial astringency imparted by the gallate-type catechins, EGCg and ECg. For the initial astringency detected for EGCg and ECg, the tea infusion-derived pectin exhibited a concentration-dependent suppressive effect [Fig. 4(b)]. On the other hand, the tea infusion-derived pectin did not suppress the residual astringency of any of the catechins [Fig. 5(a,b)]. Hayashi et al. (2005) reported that citrus-derived pectin suppressed the initial astringency of gallate-type catechins, but not that of non-gallate-type catechins. These results are in agreement with our results, in which the tea infusion-derived pectin exhibited a stronger suppressive effect on the initial astringency of gallate-type catechins than on non-gallate-type ones. The astringent taste of catechins is thought to be recognized through the hydrophobic association with phospholipids or proteins on the cell membrane of tongue cells (Kajiya et al., 2001) or salivary proteins (Murray et al., 1994). The artificial lipid-based membrane on the surface of the astringency sensor electrode used in this study consisted of positively charged lipids, tetradodecylammonium bromide, dioctyl phenylphosphonate as a plasticizer, and polyvinyl chloride, a polymeric material (Kobayashi et al., 2010). Although the membrane did not contain proteins that make up cell membranes or saliva, it was designed to have a charge density and high hydrophobicity such that it exhibits a highly pronounced selective responsiveness to negatively charged and highly hydrophobic astringents, such as catechins (Nakayama, 2013). The effect of catechins on the astringency sensor electrode was thought to be related to the hydrophobicity of the catechins (Kajiya et al., 2001), e.g., the higher hydrophobicity of gallate-type catechins resulted in a stronger astringent response than that of non-gallate-type catechins. Hayashi et al. (2005) demonstrated that the closer complexation between gallate-type catechins and pectin is a factor in the astringency reduction of catechins. Therefore, the complexation of gallate-type catechins with the tea infusion-derived pectin may have partially inhibited the hydrophobic association between these catechins and the sensor membrane, thereby reducing the degree of the membrane potential change and resulting in the inhibitory effect of the pectin on the initial astringency of the catechins. However, the binding affinity of the pectin to gallate-type catechins was not so strong and the pectin was washed out; thus, the amount of catechins that stayed bound to the sensor membrane remained constant regardless of the addition of the pectin, and the pectin did not show a suppressive effect on the residual astringency measured after the sensor was washed. As for the tea infusion, which had a pectin concentration of 19 mg/L, a weak, but concentration-dependent inhibitory effect on the initial astringency was observed from the tea infusion-derived pectin [Fig. 4(c)]. The addition of 60 mg/L of pectin solution suppressed the astringency by approximately 5 mV. The concentrations of catechins in this tea infusion were: EGCg, 0.54 mM; ECg, 0.12 mM; EGC, 1.1 mM; and EC, 0.37 mM. The concentration of EGCg in the tea infusion was not obviously lower than that of EGCg alone (0.65 mM); however, the degree of suppression of the initial astringency of the tea infusion by the tea infusion-derived pectin was lower than that of EGCg alone. This might be due to the interaction of EGCg with other components in the tea infusion, which weakened the interaction between pectin and EGCg. Regarding the residual astringency, no inhibitory effect from the tea infusion-derived pectin was observed [Fig. 5(c)], as in the case of catechins alone.

Fig. 4.

Initial astringency of (a) the four catechins, (b) EGCg and ECg, and (c) the tea infusion.

(a), (b) The electrical potential difference of 4.7 mM KCl solution, the solvent of catechins, is shown as 0 mV.

The electrical potential difference of 60 mg/L pectin was 0.81mV.

Fig. 5.

Residual astringency of (a) the four catechins, (b) EGCg and ECg, and (c) the tea infusion.

The electrical potential difference of 60 mg/L pectin was −0.07mV.

In this study, we examined for the first time the effect of pectin refined from sen-cha tea infusions on the astringency of catechins using a taste sensor. In past studies, the examination of the astringent taste of teas using a taste sensor has largely been reported only for the residual astringency (Hayashi et al., 2006; Matsuo et al., 2012; Kubo et al., 2014); in this study, we also examined the initial astringency. For reference, the average value of the electrical potential difference of the residual astringency, which corresponds to one unit in the evaluation scale for estimating the astringency (when a person could recognize a difference in taste intensity), has been reported to be 4.18 mV (Ujihara et al., 2017). The results showed that the tea infusion-derived pectin significantly (more than 4.18 mV) suppressed the initial astringency of the gallate-type catechins and tea infusions. The concentration of pectin (60 mg/L) used in this study was thought to reflect the actual concentration of pectin in fukamushi-cha (Hirono et al., 2020). These results suggested that the pectin in sen-cha tea infusions might have an inhibitory effect on the astringent taste that is perceived when sen-cha is taken into the mouth, and might also contribute to the reduction of the astringency in fukamushi-cha. In this study, relatively high molecular weight pectin, the amount of which is increased with increasing steaming time, was used to verify the effect of pectin on catechin astringency. In the case of low molecular weight pectin, the amount of which is increased by roasting, such as in hoji-cha, the effect of pectin on catechin astringency may be different, i.e., it may be much weaker.

Acknowledgements    This research was partly supported by a Research Grant from the National Agriculture and Food Research Organization Gender Equality Program. The author would like to thank Masako Ishikawa for the research assistance.

Conflict of interest    The author declares that there are no conflicts of interest associated with this manuscript.

Abbreviations

EGCg

(−)-epigallocatechin-3-gallate

ECg

(−)-epicatechin-3-gallate

EGC

(−)-epigallocatechin

EC

(−)-epicatechin

References

•  Dubois,  M.,  Gilles,  K. A.,  Hamilton,  J. K.,  Rebers,  P.A., and  Smith,  F. (1956). Colorimetric method for determination of sugars and related substances. Anal. Chem., 28, 350-356.
•  Ele-Ekouna,  J.-P.,  Pau-Roblot,  C.,  Courtois,  B., and  Courtois,  J. (2011). Chemical characterization of pectin from green tea (Camellia sinensis). Carbohydr. Polym., 83, 1232-1239.
•  Hayashi,  N.,  Ujihara,  T., and  Kohata,  K. (2005). Reduction of catechin astringency by the complexation of gallate-type catechins with pectin. Biosci. Biotechnol. Biochem., 69, 1306-1310.
•  Hayashi,  N.,  Chen,  R.,  Ikezaki,  H.,  Yamaguchi,  S.,  Maruyama,  D.,  Yamaguchi,  Y.,  Ujihara,  T., and  Kohata,  K. (2006). Techniques for universal evaluation of astringency of green tea infusion by the use of a teaste sensor system. Biosci. Biotechnol. Biochem., 70, 626-631.
•  Hirono,  H. and  Mizukami,  Y. (2018). Dynamics of water-soluble pectin in the roasting process during green tea manufacturing. Food Sci. Technol. Res., 24, 177-181.
•  Hirono,  H.,  Goto,  S., and  Aiba,  S. (2020). A Study on the content of water soluble pectin and catechins in sen-cha (unrefined tea) of various degree of steaming. J. Jpn. Soc. Food Sci. Technol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 67, 264-270. (in Japanese)
•  Ikezaki,  H. (2011). Intelligent Sensor Technology Inc. aims to establish a global standard for the measurement of taste. J. Inst. Image Inform. TV. Engnr., 65, 1575-1579.
•  Kajiya,  K.,  Kumazawa,  S., and  Nakayama,  T. (2001). Steric effects on interaction of tea catechins with lipid bilayers. Biosci. Biotechnol. Biochem., 65, 2638-2643.
•  Kobayashi,  Y.,  Habara,  M.,  Ikezazki,  H.,  Chen,  R.,  Naito,  Y., and  Toko,  K. (2010). Advanced taste sensors based on artificial lipids with global selectivity to basic taste qualities and high correlation to sensory scores. Sensors., 10, 3411-3443.
•  Kubo,  T.,  Fujiwara,  T., and  Tomizawa,  Y. (2014). Evaluation of astringency and umami of green tea infusions with different elution conditions using a taste sensor system. J. Jpn. Soc. Food Sci. Technol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 61, 192-198. (in Japanese)
•  Matsuo,  H.,  Hayashi,  N.,  Ujihara,  T.,  Fujita,  S.,  Tastuno,  T.,  Mitarai,  M.,  Gejima,  Y.,  Toyomitsu,  Y.,  Kinoshita,  O., and  Taniguchi,  T. (2012). Astringency of Kamairi-cha and Sen-cha. J. Jpn. Soc. Food Sci. Technol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 59, 6-16. (in Japanese)
•  Murray,  N.J.,  Williamson,  M.P.,  Lilley,  T.H., and  Haslam,  E. (1994). Study of the interaction between salivary proline-rich proteins and a polyphenol by ¹H-NMR spectroscopy. Eur. J. Biochem., 219, 923-935.
•  Nakagawa,  M. (1972). The relationship between astringency and the reaction aspects of astringents for protein. J. Jpn. Soc. Food Sci. Technol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 19, 531-537. (in Japanese)
•  Nakayama,  T. (2013). Molecular interaction of tea catechins with biological substances. Bull. Nippon Vet. Life Sci. Univ. (in Japanese), 62, 1-7
•  Scott,  R.W. (1979). Colorimetric determination of hexuronic acids in plant materials. Anal. Chem., 51, 936-941.
•  Taira,  S.,  Ono,  M., and  Matsumoto,  N. (1997). Reduction of persimmon astringency by complex formation between pectin and tannins. Postharvest Biol. Technol., 12, 265-271.
•  Ujihara,  T.,  Hayashi,  N.,  Chen,  R.,  Habara,  M., and  Ikezaki,  H. (2017). Deviation in astringent and umami taste intensity values of leaf-type green tea using taste sensor system in interlaboratory trial. J. Jpn. Soc. Food Sci. Technol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 64, 74-80. (in Japanese)