Technique for assessing the astringency of persimmon fruit by measuring the liposome aggregation

Abstract

The astringency of persimmon fruits is a significant factor for consumers and the nutritional industry. To date, astringents, such as polyphenols, specifically persimmon condensed tannins, have been assessed using polyphenol quantification assays, such as the Folin–Ciocalteu method, based on their reducing power. However, these methods are influenced by the presence of other reducing substances. In this study, we developed a cost-effective liposome turbidity analysis using a portable visible spectrophotometer based on the interaction between liposomes and astringents. Authentic astringents, such as catechins and theaflavin-3-O-gallate, were analyzed, and their half-maximal effective concentrations (EC₅₀) were calculated. These results indicated that the affinity to the membrane was similar to that of astringency, as determined by sensory analysis. Additionally, the EC₅₀ values of partially purified tannins from non-astringent and astringent persimmons were calculated. In conclusion, we determined the application methods to assess astringent persimmon fruits with and without the removal of astringency.

Introduction

Persimmons (Diospyros kaki L.) are predominantly cultivated in Asian countries, such as China, Japan, and Korea. Each region has numerous local cultivars broadly classified into two types: non-astringent and astringent (Sato and Yamada, 2016). Non-astringent immature fruits exhibit slight astringency, which reduced with maturity. In contrast, astringent fruits exhibit a high astringency during maturation. Because astringency is undesirable to consumers, farmers remove it by incubating harvested fruits with CO₂ or ethanol before shipping. In addition to their agricultural significance, astringents, such as persimmon tannins, are gaining recognition in the nutrition industry because of their diverse bioactivities, such as antihyperlipidemic, anti-inflammatory effects (Zou et al., 2014, Zou et al., 2012), and antibacterial properties (Tomiyama et al., 2016). Therefore, a convenient technique for assessing the astringency of persimmons is required.

Astringency is attributed to polyphenols, such as catechins, anthocyanidins, and proanthocyanidins. Specifically, persimmon condensed tannins are highly polymerized (mean degree of polymerization = 26) and galloylated (72 %) proanthocyanidins consisting of (−)-epicatechin (EC), (−)-epigallocatechin (EGC), (−)-epicatechin gallate (ECg), and (−)-epigallocatechin gallate (EGCg) (Zhu et al., 2019). As polyphenols are reducing agents, persimmon astringency is assessed using polyphenol quantification assays, such as the Folin-Ciocalteu (Folin and Ciocalteu, 1927), Folin-Denis (Swain and Hillis, 1959), and tannin printing methods (Matsuo and Ito, 1977), based on the color change of metals in the reaction solution. However, these methods are significantly influenced by other reducing agents, particularly in assessing persimmons. Niikawa et al. detected 27–132 mg/100 g fresh weight (FW) of vitamin C, a well-known interfering substance in these methods, in 31 diverse typical Japanese persimmon cultivars (Niikawa et al., 2011). We focused on the interactions between polyphenols and biological membranes to enhance the assessment of astringency in persimmons. Although the mechanisms of astringency perception in the oral cavity are unknown, astringency is believed to result from the interaction of astringents with oral compounds, such as tissues, cell membrane proteins, epithelial cells, and mechano- and chemoreceptors (Pires et al., 2020, Reis et al., 2020). To date, the interaction between polyphenols and biological membranes has been demonstrated using various methods, such as catechin incorporation into liposomes (Hashimoto et al., 1999), calcein leakage assays (Kajiya et al., 2004), and isothermal titration calorimetry (ITC) (Šturm et al., 2022, Sun et al., 2009, Virtanen et al., 2022, Zhu et al., 2019). Scanning electron microscopy (SEM) (Zhu et al., 2019), transmission electron microscopy (Šturm et al., 2022), atomic force microscopy (AFM) (Evans et al., 2009), and X-ray diffraction (Sun et al., 2009) have demonstrated changes in membrane morphology. However, these methods require expensive equipment and professional expertise. The addition of catechins or theaflavins, which are astringent compounds found in black tea, into the liposome solution caused liposomal aggregation, leading to cloudiness in the solution (Liu and Tzen, 2022, Nakayama et al., 2023, Narai-Kanayama et al., 2018), which is visible to the naked eye. Moreover, detailed analyses can be performed using an inexpensive portable visible spectrophotometer (Nakayama et al., 2023). In this study, we developed a cost-effective liposome turbidity assay to assess the astringency of persimmon fruit.

Materials and Methods

Materials EC, EGC, ECg, and theaflavin-3-O-gallate (TF2A) were graciously supplied by Mitsui Norin (Shizuoka, Japan). EGCg was purchased from Merck Millipore (Burlington, MA, USA). Dipalmitoylphosphatidylcholine (DPPC) and dimyristoylphosphatidylcholine (DMPC) were purchased from NOF CORPORATION (Tokyo, Japan). All other reagents were of analytical grade, unless specified otherwise. The structures of authentic astringents and phosphatidylcholines (PCs) were shown in Fig. 1.

Fig. 1

Chemical structures of theaflavin, catechins, and phosphatidylcholines.

(A) Catechins: (−)-epicatechin (EC), R₁ = H, R₂ = H; (−)-epigallocatechin (EGC), R₁ = H, R₂ = OH; (−)-epicatechin gallate (ECg), R₁ = galloyl, R₂ = H; (−)-epigallocatechin gallate (EGCg), R₁ = galloyl, R₂ = OH. (B) Theaflavin-3-O-gallate (TF2A). (C) Dipalmitoylphosphatidylcholine (DPPC). (D) Dimyristoylphosphatidylcholine (DMPC).

Preparation of persimmon fruit extracts Mature astringent persimmon fruits (Diospyros kaki cv. Tonewase) were harvested in early October at the Nara Prefecture Agricultural Research and Development Center (Japan). To remove astringency, whole persimmon fruits were stored at 25 °C for 16 h with or without CO₂, and then left in air at 23 °C for 7 days. Subsequently, the fruits were peeled, cut, and stored at −80 °C until use. The stored fruit (1 g) was crushed with 2 mL ethanol using a bead shocker (TAITEC CORPORATION, Saitama, Japan). Following centrifugation at 15 000 × g for 10 min at 4 °C, the supernatant was collected as the persimmon fruit extract. For liposome turbidity analysis, the extracts were diluted with four parts of water.

Preparation of persimmon tannin Mature, non-astringent (Diospyros kaki cv. Gosho), and astringent persimmon fruits (Diospyros kaki cv. Horenbo) were harvested in late November at the Nara Prefecture Agricultural Research and Development Center (Japan). Persimmon tannins were prepared as described previously (Hamasaki, 2010). Briefly, whole persimmon fruits were stored with ethanol for 5 days to insolubilize the soluble tannin. Subsequently, the fruits were peeled, homogenized, and centrifuged at 1 630 × g for 10 min at 4 °C. The fraction containing tannins was collected and lyophilized. The dried powder (100 mg) was resuspended in 10 mL distilled water and autoclaved for 15 min at 121 °C. After cooling, the filtrate was lyophilized and the resulting dried tannin powder was stored.

Preparation of liposome solution The liposome solution was prepared as previously described (Iijima et al., 2022, Nakayama et al., 2023, Narai-Kanayama et al., 2018). In a round-bottom flask, 30 mg DPPC or DMPC was dissolved in chloroform (1.5 mL) and dried using a vacuum pump. The resulting thin film formed on the inner wall was suspended in 10 mL of 15 mM sodium phosphate buffer with 132 mM sodium chloride (pH 6.5), and then transferred to a 15 mL conical tube. The vesicles were prepared using an ultrasonic disruptor (UD-211, Tomy Seiko, Tokyo, Japan) for 10 min at a 50 % duty cycle and 80 % output level. Following centrifugation at 15 000 × g for 15 min at room temperature, the supernatant was collected as the liposome solution and its concentration was determined using a phospholipid test kit (FUJIFILM Wako Pure Chemical Corporation, Tokyo, Japan). Subsequently, a 2.5 mM liposome solution was prepared with the aforementioned sodium phosphate buffer and stored at room temperature.

The particle size analysis The particle size distribution of liposomes was analyzed using a Zetasizer Nano ZS (Malvern Panalytical, Malvern, UK) as previously described (Iijima et al., 2022). The liposome solution (1 µL) was diluted in 150 µL of milliQ and analyzed at 25 °C using a cuvette (ZEN0040, Malvern Panalytical).

Liposome turbidity analysis The sample solution (40 µL) was added to 160 µL of 2.5 mM liposome solution in 0.2 mL PCR tube (FUKAE-KASEI CO., LTD, Kobe, Japan). After vortexing and a 45-second incubation, the optical density (OD₆₁₅) was measured using a portable photo absorbance meter (PiCOEXPLORER, Yamato Scientific Co., Ltd., Tokyo, Japan) in turbidity mode at 575–660 nm with a peak at 615 nm. The background signal was measured using water as a mock control.

Folin-Ciocalteu assay Total phenolic content of the sample solution was determined using the Folin-Ciocalteu assay (Folin and Ciocalteu, 1927) as previously described (Kumazawa et al., 2002). Briefly, the sample solution (100 µL) was mixed with 500 µL of 10 % Folin-Ciocalteu reagent (FUJIFILM Wako Pure Chemical Corporation) and 400 µL of 7.5 % Na₂CO₃. After vortexing and a 1-hour incubation at 25 °C, the absorbance was measured at 765 nm using spectrophotometer (U-2900, Hitachi High-Tech Science Corporation, Tokyo, Japan)

Statistical analysis and curve fitting Statistical analysis and curve fitting were performed with a five-parameter logistic equation using SigmaPlot 15 software (Systat Software Inc., San Jose, CA, USA).

Results and Discussion

Stability of liposome turbidity analysis PC is a significant component in biological membranes. Egg PC and DPPC were used for the liposome turbidity analysis (Nakayama et al., 2023, Narai-Kanayama et al., 2018). To ensure a reliable and sensitive assesing system, saturated PCs, such as DPPC and DMPC, were used as liposomal materials because unsaturated PCs are readily oxidized during prolonged storage. To assess the potential aggregation of DPPC (DPPC-LP) and DMPC (DMPC-LP) liposomes during prolonged storage, aliquots were incubated at 27 °C, assuming room temperature in the dark. The particle size distribution exhibited a single peak for DPPC-LP (78.8 nm) and DMPC-LP (122 nm), respectively (Fig. 2A). The size distributions were consistently observed in independent experiments under the same conditions. DPPC-LP maintained its peak stability for 28 days, whereas DMPC-LP exhibited multiple peaks after 14 days. In turbidity analysis, the 2.5 mM liposome solution (final concentration: 2 mM) was mixed with 2 mM EGCg (final concentration: 400 µM), and turbidity was measured using a portable photo absorbance meter (PiCOEXPLORER). The OD₆₁₅ values of DPPC-LP (0.659 ± 0.010) and DMPC-LP (0.053 ± 0.002) remained constant during storage (Fig. 2B). The large difference in their sensitivities might be due to the membrane properties (Clerc and Thompson, 1995). These results indicate that DPPC-LP is more stable and sensitive than that of DMPC-LP for liposome turbidity analysis.

Fig. 2

Stability of liposome solutions stored at 27 °C.

(A) Changes in the particle size and distribution. (B) Changes in DPPC-LP (closed circles) and DMPC-LP (closed squares) aggregation by EGCg (400 µM). The optical density (OD) is represented as difference between the sample value and that of mock control. Each value represents the mean ± SD (n = 5). Significant differences are analyzed using Tukey–Kramer method (p < 0.05). However, no significant difference is observed during storage for DPPC-LP and DMPC-LP.

Evaluating the polyphenol-dependent aggregation of liposome In a previous study, authentic catechins, theaflavins, and black tea infusions caused the aggregation of DPPC-LP through hydrophobic interactions, and the intensity was assessed by the OD₆₁₅ value at given concentrations (Iijima et al., 2022, Nakayama et al., 2023). Moreover, we realized that the value of the upper asymptote of the dose-response curves was unstable during independent experiments using individually prepared liposome solutions, and was not correlated with astringency. These observations suggest that the OD₆₁₅ values can be used in simplified judgment, but lack universality as an index for astringency. To assess astringency by quantifying the interaction, the half-maximal effective concentration (EC₅₀) was calculated by analyzing the precise dose-response curves of authentic catechins and TF2A. Here, TF2A was chosen to represent theaflavin because it shows the strongest interaction with PC (Nakayama et al., 2023, Narai-Kanayama et al., 2018). Initially, a 2.5 mM liposome solution (final concentration: 2 mM) was combined with 0–5 mM EGCg (final concentration: 0–1000 µM). Turbidity exhibited a sigmoidal transition in the range of 0–600 µM EGCg, and a linear transition in the range of 600–1 000 µM EGCg (Fig. 3A). In accordance with this result, the homogenous aggregation of liposome-like cloudiness was densely distributed in the range of 0–600 µM EGCg, while visible small particles appeared in the range of 600–1000 µM EGCg (Fig. 3B). This phenomenon suggests a transition in the liposome structure due to the interaction with EGCg, which is negatively charged at neutral pH. When charged membrane vesicles, such as liposomes, aggregate, they can transition from initially separated vesicles or hemifusion vesicles to fully fused vesicles (Chernomordik and Kozlov, 2008, Haque et al., 2001, Shi et al., 2022, Siegel, 1986). Shi et al. (2022) observed an alteration in the response of total lipid mixing concerning the ratios of positively charged polymer to potentially negatively charged lipid vesicles. These reports imply that the hemifusion liposome could progress to significant full fusion vesicles at 600 µM EGCg. Thus, to analyze the interaction of astringents with liposomes, dose-response curves were generated from distinct sigmoidal change regions (Fig. 3C and D) and the EC₅₀ was calculated from five independent experiments (Table 1). Among the authentic astringents, TF2A had the lowest EC₅₀, whereas gallate-type catechins (EGCg and ECg) exhibited significantly lower EC₅₀ values than non-gallate-type catechins (EGC and EC). In conclusion, the astringency order was TF2A > EGCg = ECg > EGC = EC, partially aligned with previous studies. The astringency of theaflavins was analyzed through sensory analysis of black tea brews (Ding et al., 1992) and water-containing standard compounds (Scharbert et al., 2004). Ding et al. (1992) suggested that the astringency of black tea correlated with catechins, not theaflavins. However, Scharbert et al. (2004) suggested that the human astringency threshold for authentic theaflavins is approximately one-tenth that of authentic catechins. Moreover, microscopic observation of liposomes indicated that theaflavins induced significant morphological changes compared with catechins (Phan et al., 2014). Previous studies have indicated that theaflavins exhibit stronger interactions with liposomes with greater astringency than catechins when the same amount is considered. For catechins, sensory analysis using water containing standard compounds demonstrated the human astringency thresholds for EC (930 µM), EGC (520 µM), ECg (260 µM), and EGCg (190 µM), respectively (Scharbert et al., 2004). The membrane affinity order for these catechins was ECg > EGCg >> EC > EGC (Kajiya et al., 2001, Kajiya et al., 2002), which differed slightly from the sensory analysis. Although liposome turbidity analysis was based on the interaction between polyphenols and biological membranes, our data aligned with the astringency sensory order (Fig. 3C and D, Table 1) if significant differences were not considered. However, the tendency of the galloyl motif to be significant for astringency and membrane interactions has been observed in both experimental and simulation data (Sirk et al., 2009, Sirk et al., 2011). In summary, we encountered difficulties in distinguishing between EGCg and ECg, and between EGC and EC. However, the EC₅₀ derived from liposome turbidity analysis could serve as a universal index for assessing the astringency of authentic materials.

Fig. 3

Polyphenol-dependent aggregation of liposome.

(A) DPPC-LP with various concentrations of EGCg were incubated at 27 °C for 45 sec. The optical density is represented as difference between sample value and a mock control. Each value represents mean ± SD (n = 10). The line represents a five-parameter logistic sigmoidal curve. (B) Image represents DPPC-LP treated with various EGCg concentrations. (C) DPPC-LP with various concentrations of TF2A (black), EGCg (blue), and ECg (red) were incubated at 27 °C for 45 sec. The optical density is represented as difference between sample value and a mock control. Each value corresponded to mean ± SD (n = 5 experiments). Each line represents a five-parameter logistic sigmoidal curve. (D) DPPC-LP with various concentrations of EGC (orange) and EC (green) were incubated at 27 °C for 45 sec. The optical density is represented as difference between sample value and a mock control. Each value corresponded to mean ± SD (n = 5 experiments). Each line represents a five-parameter logistic sigmoidal curve.

Table 1. The half-maximal effective concentration (EC₅₀) of theaflavin and catechins.

TF2A [µM]	EGCg [µM]	ECg [µM]	EGC [µM]	EC [µM]
a97.8 ± 5.1	b327.3 ± 25.9	b345.7 ± 9.7	c5089.6 ± 262.1	c5178.1 ± 230.3

Each value corresponded to mean ± SD (n = 5 experiments)

Clusters with identical letter codes did not exhibit significant differences according to Tukey’s test (p < 0.05).

Assessing the effects of sugar and ascorbic acid in liposome turbidity analysis Persimmon contains significant amounts of sugars (Hirai and Yamazaki, 1984) and ascorbic acid (Niikawa et al., 2011). Since the presence of certain reducing agents, such as ascorbic acid, in persimmon fruits affects the detection value of the Folin-Ciocalteu method, and sugars can affect the environment of lipid membrane surfaces (Crowe et al., 1984), the effects of these contaminants on liposome turbidity were compared using the Folin-Ciocalteu method. In Folin-Ciocalteu method and liposome turbidity analysis, for the sugar aspect, we utilized solutions containing glucose (22.0 mg/mL), fructose (15.7 mg/mL), sucrose (17.9 mg/mL), or mixed sugar (22.0 mg/mL glucose, 15.7 mg/mL fructose, and 17.9 mg/mL sucrose), respectively, assuming a sugar composition of major non-astringent persimmon fruits (Diospyros kaki cv. Fuyu) (Hirai and Yamazaki, 1984). The results showed no variation in detection values between samples and the mock control, indicating that sugar does not influence the detection value in these analyses (Fig. 4A). In Folin-Ciocalteu assay and liposome turbidity analysis, for ascorbic acid aspect, we utilized solutions, which mimic the fruit juice of mature astringent persimmon (0.167 mg/mL), mature non-astringent persimmon (0.333 mg/mL), immature astringent persimmon (0.667 mg/mL), or immature nonastringent persimmon (1.000 mg/mL), respectively. In the Folin-Ciocalteu assay, the 0.167 mg/mL solution exhibited a significant difference from the mock control, with 1.000 mg/mL solution exhibiting a high value equivalent to the positive control of 2 mM EGCg (final concentration: 400 µM) (Fig. 4B, left). In contrast, no significant difference was observed between the mock control and any of the ascorbic acid solutions (Fig. 4B, right). Therefore, liposome turbidity analysis proved effective in minimizing the background interference of contaminants and was almost negligible for the assessment of persimmon astringency.

Fig. 4

Comparison of Folin-Ciocalteu method with liposome turbidity analysis.

(A) The effects of sugar (Glc: 22.0 mg/mL glucose; Fru: 15.7 mg/mL fructose; Suc: 17.9 mg/mL sucrose; Mix: 22.0 mg/mL glucose, 15.7 mg/mL fructose, and 17.9 mg/mL sucrose) on Folin-Ciocalteu method (left panel) and liposome turbidity analysis (right panel). A 400 µM of EGCg was used as a positive control. Each value corresponded to mean ± SD (n = 5). Significant differences were analyzed by Dunnett’s test (*, p < 0.05 vs. mock control) (B) The effects of ascorbic acid on Folin-Ciocalteu method (left panel) and liposome turbidity analysis (right panel). A 400 µM of EGCg was used as a positive control. Each value corresponded to mean ± SD (n = 5). Significant differences were analyzed by Dunnett’s test (*,p < 0.05 vs. mock control).

Assessing tannin and extraction from persimmon fruits through liposome turbidity analysis Persimmons contain various astringents such as tannins (Kometani and Takemori, 2016, Li et al., 2010, Matsuo and Ito, 1978). To assess whether EC₅₀ values of astringent mixture were similar to that of authentic materials, we used tannins obtained from astringent persimmon “Horenbo” and non-astringent persimmon “Gosho” varieties for liposome turbidity analysis. The dried tannin powder was resuspended in water and combined with liposome solution. No visible particles were observed in the tannin concentration range of 0–0.20 mg/mL. In contrast to authentic materials, the tannin solution exhibited a brown coloration, prompting us to investigate whether the colorimetric value corresponding to the color should be subtracted from the raw OD₆₁₅. The tannin solution was diluted with water to match the assay solution concentration, and its OD₆₁₅ was measured in turbidity mode. Consequently, the EC₅₀ values of Horenbo (0.023 mg/mL) and Gosho (0.066 mg/mL) were determined from their respective sigmoidal curves, considering their color (Fig. 5A, Table 2). However, neglecting the color had no impact on their EC₅₀ values because of its low OD₆₁₅ value (approximately 0.10 at 0.20 mg/mL tannin) associated with color (Table 2). These data indicate that our liposome turbidity analysis was significant in assessing astringent mixtures, with color having a negligible effect for simple analysis. Considering the variation in polyphenol composition of tannins among cultivars (Takemori et al., 2022, Yonemori et al., 1983) and the differing astringency of catechins (Scharbert et al., 2004), it is advisable to assess the astringency of mixed astringencies based on membrane affinity strength rather than total polyphenol content. The liposome turbidity analysis in this study could be considered a valuable assessment technique.

Fig. 5

Tannin or persimmon fruit-dependent aggregation of liposome.

(A) DPPC-LP with various concentrations of tannin (Horenbo: closed circle and Gosyo: open circle) were incubated at 27 °C for 45 sec. The optical density is represented as difference between sample value, and mock control and background of color. Each value corresponds to mean ± SD (n = 5). Lines (Horenbo: solid line and Gosyo: dotted line) represent a five-parameter logistic sigmoidal curve. The data represent one of five individual experiments. (B) DPPC-LP with various concentrations of distinct supernatants from “Tonewase” without removing astringency was incubated at 27 °C for 45 sec. The optical density is represented as difference between sample value, and mock control and background of color. Each value corresponds to mean ± SD (n = 5). A line represents a five-parameter logistic sigmoidal curve. The data represent one of ten individual materials. (C) DPPC-LP with various concentrations of distinct supernatants from “Tonewase” with removing astringency was incubated at 27 °C for 45 sec. The optical density is represented as difference between sample value, and that of mock and background of color. Each value corresponds to mean ± SD (n = 5). A line represents a five-parameter logistic sigmoidal curve. The data represent one of ten individual materials.

Table 2. The half-maximal effective concentration (EC₅₀) of persimmon tannin.

Tannin from “Horenbo” [mg/mL]		Tannin from “Gosho” [mg/mL]
Absorbance of pigment		Absorbance of pigment
Neglect	Consider	Neglect	Consider
0.022 ± 0.002	0.023 ± 0.002	0.066 ± 0.007	0.066 ± 0.007

Each value corresponded to mean ± SD (n = 5 experiments)

We conducted a study on application methods for assessing persimmon fruit using liposome turbidity analysis.

We used the “Tonewase” astringent persimmon with or without removing astringency. In the sample preparation, we utilized water to prepare the extract because high organic solvent concentrations disrupt liposomes. However, the high sugar content in persimmons hinders the distinct separation of supernatants and residual pulp through centrifugation at 15 000 × g for 10 min at 4 °C. Subsequently, extraction with ethanol or methanol contributed to obtaining distinct supernatants through centrifugation at a final concentration of 20 %, which is acceptable. Ethanol was used in this study because of its toxicity to the general public including non-scientists working in agriculture and food industry. Using a portable photo absorbance meter, we investigated whether it was necessary to subtract the color-related values from raw OD₆₁₅. Distinct supernatants were diluted with water, and the OD₆₁₅ was measured. The EC₅₀ value of “Tone wase” with or without removing astringency was finally determined as 2.012 g-FW /mL or 0.021 g-FW/mL from sigmoidal curves considering the color (Figure 5B and C, Table 3). In this case, the effect of color was negligible (Table 3). In pragmatic straightforward qualitative analysis for distinguishing removing astringency, a sigmoidal curve is potentially redundant because the OD₆₁₅ of each 0.02 g-FW/mL solution was distinctly different (Fig. 5B and C). Thus, our liposome turbidity analysis proved valuable for assessing persimmon fruit astringency. However, it remains necessary to determine whether the sensitivity of this method is adequate for distinguishing subtle differences that are comparable to the human astringency threshold. The potential application of this method to various persimmons and other fruit cultivars is currently under investigation.

Table 3. The half-maximal effective concentration (EC₅₀) of persimmon fruits.

Without removing astringency [g-fresh weight/mL]		With removing astringency [g-fresh weight/mL]
Absorbance of pigment		Absorbance of pigment
Neglect	Consider	Neglect	Consider
0.021 ± 0.010	0.021 ± 0.010	2.091 ± 0.306	2.012 ± 0.244

Each value corresponded to mean ± SD (n = 10 materials)

Acknowledgements Funding: This study was supported by the Tojuro Iijima Foundation for Food Science and Technology (Tokyo, Japan). We appreciate providing authentic catechins and theaflavins to Mitsui Norin (Shizuoka, Japan).

Conflict of interest There are no conflicts of interest to declare.

References

•  Chernomordik,  L.V. and  Kozlov,  M.M. (2008). Mechanics of membrane fusion. Nat. Struct. Mol. Biol., 15, 675–683.
•  Clerc,  S.G. and  Thompson,  T.E. (1995). Permeability of dimyristoyl phosphatidylcholine/dipalmitoyl phosphatidylcholine bilayer membranes with coexisting gel and liquid-crystalline phases. Biophys J, 68, 2333–2341.
•  Crowe,  J.H.,  Crowe,  L.M., and  Chapman,  D. (1984). Preservation of membranes in anhydrobiotic organisms: the role of trehalose. Science, 223, 701–703.
•  Ding,  Z.,  Kuhr,  S., and  Engelhardt,  U.H. (1992). Influence of catechins and theaflavins on the astringent taste of black tea brews. Z. Lebensm. Unters. Forsch., 195, 108–111.
•  Evans,  P.J.,  Bird,  M.R.,  Rogers,  D., and  Wright,  C.J. (2009). Measurement of polyphenol-membrane interaction forces during the ultrafiltration of black tea liquor. Colloids Surf. A Physicochem. Eng. Asp., 335, 148–153.
•  Folin,  O. and  Ciocalteu,  V. (1927). On tyrosine and tryptophane determinations in proteins. J. Biol. Chem., 73, 627–650.
•  Hamasaki,  S. (2010). Quick preparation of kaki-tannin from de-astringent persimmon fruit treated by ethanol. Hort. Res. (Japan) (Engeigaku kenkyu), 9, 367–372. (in Japanese)
•  Haque,  M.E.,  McIntosh,  T.J., and  Lentz,  B.R. (2001). Influence of lipid composition on physical properties and peg-mediated fusion of curved and uncurved model membrane vesicles: “nature’s own” fusogenic lipid bilayer. Biochemistry, 40, 4340–4348.
•  Hashimoto,  T.,  Kumazawa,  S.,  Nanjo,  F.,  Hara,  Y., and  Nakayama,  T. (1999). Interaction of tea catechins with lipid bilayers investigated with liposome systems. Biosci. Biotechnol. Biochem., 63, 2252–2255.
•  Hirai,  S. and  Yamazaki,  K. (1984). Studies on sugar components of sweet and astringent persimmon by gas chromatography. Nippon Shokuhin Kagaku Kogaku Kaishi, 31, 24–30. (in Japanese)
•  Iijima,  M.,  Kawaguchi,  A.,  Ogura,  Y.,  Yoshimoto,  R.,  Kaneda,  M.,  Kera,  K.,  Kuroda,  S., and  Nakayama,  T. (2022). Nano-visualization of the in vitro antiviral activity of black tea based on production area using a liposome-based virus membrane model. Biosci. Biotechnol. Biochem., 86, 1658–1669.
•  Kajiya,  K.,  Hojo,  H.,  Suzuki,  M.,  Nanjo,  F.,  Kumazawa,  S., and  Nakayama,  T. (2004). Relationship between antibacterial activity of (+)-catechin derivatives and their interaction with a model membrane. J. Agric. Food Chem., 52, 1514–1519.
•  Kajiya,  K.,  Kumazawa,  S., and  Nakayama,  T. (2001). Steric effects on interaction of tea catechins with lipid bilayers. Biosci. Biotechnol. Biochem., 65, 2638–2643.
•  Kajiya,  K.,  Kumazawa,  S., and  Nakayama,  T. (2002). Effects of external factors on the interaction of tea catechins with lipid bilayers. Biosci. Biotechnol. Biochem., 66, 2330–2335.
•  Kometani,  T. and  Takemori,  K. (2016). Polyphenols from persimmon fruits as a functional foods material. Nippon Shokuhin Kagaku Kogaku Kaishi, 63, 331–337. (in Japanese)
•  Kumazawa,  S.,  Taniguchi,  M.,  Suzuki,  Y.,  Shimura,  M.,  Kwon,  M.S., and  Nakayama,  T. (2002). Antioxidant activity of polyphenols in carob pods. J. Agric. Food Chem., 50, 373–377.
•  Li,  C.,  Leverence,  R.,  Trombley,  J.D.,  Xu,  S.,  Yang,  J.,  Tian,  Y.,  Reed,  J.D., and  Hagerman,  A.E. (2010). High molecular weight persimmon (Diospyros kaki L.) proanthocyanidin: a highly galloylated, A-linked tannin with an unusual flavonol terminal unit, myricetin. J. Agric. Food Chem., 58, 9033–9042.
•  Liu,  C.T. and  Tzen,  J.T.C. (2022). Exploring the relative astringency of tea catechins and distinct astringent sensation of catechins and flavonol glycosides via an in vitro assay composed of artificial oil bodies. Molecules, 27, 5679–5691.
•  Matsuo,  T. and  Ito,  S. (1977). On mechanisms of removing astringency in persimmon fruits by carbon dioxide treatment I. Some properties of the two processes in the deastringency. Plant Cell Physiol., 18, 17–25.
•  Matsuo,  T. and  Ito,  S. (1978). The chemical structure of kakitannin from immature fruit of the persimmon (Diospyros kaki L.). Agric. Biol. Chem., 42, 1637–1643.
•  Nakayama,  T.,  Huriya,  H.,  Kamo,  A.,  Kosako,  K.,  Kozima,  T.,  Sugawara,  M.,  Suzuki,  T.,  Nishizima,  M.,  Yamashita,  A.,  Yano,  M.,  Makino,  S.,  Kera,  K., and  Iijima,  M. (2023). Molecular interaction of black tea components with phospholipid vesicles. Food Sci. Technol. Res., 29, 163–169.
•  Narai-Kanayama,  A.,  Saruwatari,  K.,  Mori,  N., and  Nakayama,  T. (2018). Theaflavin-3-gallate specifically interacts with phosphatidylcholine, forming a precipitate resistant against the detergent action of bile salt. Biosci. Biotechnol. Biochem., 82, 466–475.
•  Niikawa,  T.,  Suzuki,  T.,  Ozeki,  T.,  Miyake,  N., and  Kurata,  T. (2011). Varietal difference in fruit vitamin C content and increase of fruit vitamin C content by water impermeable mulching treatment of Japanese persimmon trees. Hort. Res. (Japan) (Engeigaku kenkyu), 10, 225–231. (in Japanese)
•  Phan,  H.T.,  Yoda,  T.,  Chahal,  B.,  Morita,  M.,  Takagi,  M., and  Vestergaard,  M.C. (2014). Structure-dependent interactions of polyphenols with a biomimetic membrane system. Biochim. Biophys. Acta, 1838, 2670–2677.
•  Pires,  M.A.,  Pastrana,  L.M.,  Fucinos,  P.,  Abreu,  C.S., and  Oliveira,  S.M. (2020). Sensorial perception of astringency: oral mechanisms and current analysis methods. Foods, 9, 1124–1146.
•  Reis,  A.,  Soares,  S.,  Sousa,  C.F.,  Dias,  R.,  Gameiro,  P.,  Soares,  S., and  de Freitas,  V. (2020). Interaction of polyphenols with model membranes: Putative implications to mouthfeel perception. Biochim. Biophys. Acta Biomembr., 1862, 183133–183142.
•  Sato,  A. and  Yamada,  M. (2016). Persimmon breeding in Japan for pollination-constant non-astringent (PCNA) type with marker-assisted selection. Breeding Science, 66, 60–68
•  Scharbert,  S.,  Holzmann,  N. and  Hofmann,  T. (2004). Identification of the astringent taste compounds in black tea infusions by combining instrumental analysis and human bioresponse. J. Agric. Food Chem., 52, 3498–3508.
•  Shi,  S.,  Markl,  A.M.,  Lu,  Z.,  Liu,  R., and  Hoernke,  M. (2022). Interplay of fusion, leakage, and electrostatic lipid clustering: membrane perturbations by a hydrophobic antimicrobial polycation. Langmuir, 38, 2379–2391.
•  Siegel,  D.P. (1986). Inverted micellar intermediates and the transitions between lamellar, cubic, and inverted hexagonal lipid phases. II. Implications for membrane-membrane interactions and membrane fusion. Biophys. J., 49, 1171–1183.
•  Sirk,  T.W.,  Brown,  E.F.,  Friedman,  M., and  Sum,  A.K. (2009). Molecular binding of catechins to biomembranes: relationship to biological activity. J. Agric. Food Chem., 57, 6720–6728.
•  Sirk,  T.W.,  Friedman,  M., and  Brown,  E.F. (2011). Molecular binding of black tea theaflavins to biological membranes: relationship to bioactivities. J. Agric. Food Chem., 59, 3780–3787.
•  Šturm,  L.,  Prislan,  I.,  González-Ortega,  R.,  Mrak,  P.,  Snoj,  T.,  Anderluh,  G., and  Poklar Ulrih,  N. (2022). Interactions of (-)-epigallocatechin-3-gallate with model lipid membranes. Biochim. Biophys. Acta Biomembr., 1864, 183999–184011.
•  Sun,  Y.,  Hung,  W.C.,  Chen,  F.Y.,  Lee,  C.C., and  Huang,  H.W. (2009). Interaction of tea catechin (-)-epigallocatechin gallate with lipid bilayers. Biophys. J., 96, 1026–1035.
•  Swain,  T. and  Hillis,  W.E. (1959). The phenolic constituents of Prunus domestica. I.—The quantitative analysis of phenolic constituents. J. Sci. Fd Agric., 10, 63–68.
•  Takemori,  K.,  Okano,  M.,  Shimeno,  A., and  Kometani,  T. (2022). Polyphenol composition analysis in 30 cultivars of persimmon fruits: comparison of polyphenols between cultivars in Japan, China, and Korea. Food Sci. Technol. Res., 28, 403–413.
•  Tomiyama,  K.,  Mukai,  Y.,  Saito,  M.,  Watanabe,  K.,  Kumada,  H.,  Nihei,  T.,  Hamada,  N., and  Teranaka,  T. (2016). Antibacterial action of a condensed tannin extracted from astringent persimmon as a component of food addictive pancil PS-M on oral polymicrobial biofilms. Biomed. Res. Int., 2016, 5730748–5730754.
•  Virtanen,  V.,  Green,  R.J., and  Karonen,  M. (2022). Interactions between hydrolysable tannins and lipid vesicles from Escherichia coli with isothermal titration calorimetry. Molecules, 27, 3204–3221.
•  Yonemori,  K.,  Matsushima,  J., and  Sugiura,  A. (1983). Differences in tannins of non-astringent and astringent type fruits of Japanese persimmon (Diospyros kaki Thunb.). J. Japan. Soc. Hort. Sci., 52, 135–144. (in Japanese)
•  Zhu,  W.,  Wang,  R.F.,  Khalifa,  I., and  Li,  C.M. (2019). Understanding toward the biophysical interaction of polymeric proanthocyanidins (persimmon condensed tannins) with biomembranes: relevance for biological effects. J. Agric. Food Chem., 67, 11044–11052.
•  Zou,  B.,  Ge,  Z.Z.,  Zhang,  Y.,  Du,  J.,  Xu,  Z., and  Li,  C.M. (2014). Persimmon tannin accounts for hypolipidemic effects of persimmon through activating of AMPK and suppressing NF-kappaB activation and inflammatory responses in high-fat diet rats. Food Funct., 5, 1536–1546.
•  Zou,  B.,  Li,  C.-m.,  Chen,  J.-y.,  Dong,  X.-q.,  Zhang,  Y., and  Du,  J. (2012). High molecular weight persimmon tannin is a potent hypolipidemic in high-cholesterol diet fed rats. Food Res. Int., 48, 970–977.