Suppression in Bitterness Intensity of Bitter Basic Drug by Chlorogenic Acid

Abstract

The purpose of the study was to evaluate suppression of the bitterness intensity of bitter basic drugs by chlorogenic acid (CGA) using the artificial taste sensor and human gustatory sensation testing and to investigate the mechanism underlying bitterness suppression using ¹H-NMR. Diphenhydramine hydrocholoride (DPH) was the bitter basic drug used in the study. Quinic acid (QNA) and caffeic acid (CFA) together form CGA. Although all three acids suppressed the bitterness intensity of DPH in a dose-dependent manner as determined by the taste sensor and in gustatory sensation tests, CFA was less effective than either CGA or QNA. Data from ¹H-NMR spectroscopic analysis of mixtures of the three acids with DPH suggest that the carboxyl group, which is present in both QNA and CGA but not CFA, interact with the amine group of DPH. This study showed that the bitterness intensity of DPH was suppressed by QNA and CGA through a direct electrostatic interaction with DPH as confirmed in ¹H-NMR spectroscopic analysis. CGA and QNA may therefore be useful bitterness-masking agents for the basic drug DPH.

Taste plays an important role in determining the acceptability of a pharmaceutical formulation. Many active pharmaceutical ingredients exhibit an unpleasant taste, making taste masking an important step in formulation development. The use of an ‘electronic tongue’ or taste sensor for pharmaceutical purposes is a useful innovation, as it reduces dependence on human gustatory sensation testing. The taste sensor is an analytical sensor array system which is able to detect specific chemical properties (or tastes) of substances using electrochemical techniques and involving different artificial membranes.

Uchida et al. have reported a quantitative analytical method for the evaluation of the bitterness of various basic pharmaceutical products using a taste sensor.^1–10) This taste sensor, an ‘electric tongue’ with global selectivity, was initially developed by Toko et al.¹¹⁾ It comprises several lipid/polymer membranes capable of transforming information about the substances which produce taste into electrical signals.^12,13) The sensor output exhibits different patterns for chemical substances with different taste qualities, such as saltiness, sourness, bitterness and umami, while exhibiting similar patterns for chemical substances with similar tastes.

Diphenhydramine hydrochloride (DPH) is an important antihistamine that reduces the effects of the natural chemical histamine in the human body. It can be used to treat sneezing, watery eyes, itching, hives, runny nose, skin rash and other cold or allergy symptoms. It is known to have a high bitterness intensity.

Chlorogenic acid (CGA) is an ester formed between caffeic acid (CFA) and quinic acid (QNA). CGA and QNA are found in coffee,¹⁴⁾ plums,¹⁵⁾ berries,¹⁶⁾ fruit juice¹⁷⁾ and other foods. Like other phenolic compounds, CGA is generally thought of as a secondary plant product which protects against environmental stress. There is growing scientific evidence to support the health benefits of CGA.¹⁸⁾ In particular, studies indicate that CGA acts as a protective agent, inhibiting or reducing the oxidative stress of cell structures and functions. CGA has also been investigated for its positive effect on blood pressure and glucose regulation. In a previous study,¹⁹⁾ we reported that the addition of a commercial coffee drink to rebamipide solution successfully suppressed the bitterness of rebamipide. This suggested that CGA might be capable of suppressing the bitterness of some bitter drugs.

The aim of this study was to evaluate the ability of CGA to suppress the bitterness intensity of DPH using an artificial taste sensor and human gustatory sensation testing and to use ¹H-NMR to explore the mechanism underlying this bitterness suppression by examining the interaction between CGA and DPH.

Experimental

Materials

DPH was purchased from Wako Pure Chemical Industries, Ltd. (Japan), quinine hydrochloride from Sigma Chemical Co. (U.S.A.), CGA from Cayman Chemical Co. (U.S.A.), and CFA and QNA from Tokyo Chemical Industry Co. (Japan). All other reagents were special reagent grade.

The structures of CGA, QNA, CFA and DPH are shown in Fig. 1.

Fig. 1. Structures of (a) CGA, (b) QNA, (c) CFA and (d) DPH

The Taste Sensor

The taste sensor, SA402B (Intelligent Sensor Technology Inc., Atsugi, Japan) was used to determine the bitterness and sourness intensities of the sample solutions using sensor AN0, developed specifically to detect the bitterness of basic substances. In the first step of the procedure, a reference solution (corresponding to saliva) is measured and the electric potential obtained (mV) is defined as V_r. Then a sample solution is measured and the electric potential obtained defined as V_s. The relative sensor output (R), represented by the difference (V_s−V_r) between the potentials of the sample and the reference solution, corresponds to the ‘taste immediately after putting in the mouth.’ The electrodes are subsequently rinsed with a fresh reference solution for 6 s. When the electrode is dipped into the reference solution again, the new potential of the reference solution is defined as V_r0. The difference (V_r0−V_r) between the potentials of the reference solution before and after sample measurement is the ‘Change in membrane Potential caused by Adsorption’ (CPA) and corresponds to the so-called ‘aftertaste.’ The value obtained when CPA is divided by R is defined as the adsorption ratio.

In this study, the CPA of AN0 (CPA_AN0) was taken as the predicted bitterness intensity of the basic drug tested. In a previous study,²⁰⁾ we showed that the bitterness intensity of DPH could be evaluated using CPA values of the taste sensor BT0 and AN0. Both of sensors show bitter substance dose dependent output and superior in quantifiablity. Sensor BT0 is beneficial for evaluation in a small amount of bitter substance because of its high sensitivity. However it is inferior in durability compared to sensor AN0. Sensor AN0 does not have high sensitivity as the same as sensor BT0, however it is superior in durability compared to sensor BT0. Sensor AN0 was used in this study because bitter substance (DPH) concentration was in measurable range in sensor AN0 and durability was important for evaluation of bitterness suppression in various measurement condition.

Half millimolar DPH solution with/without CGA, QNA or CFA at concentrations of 0.1, 0.5 and 1.0 mM were prepared as the samples.

Gustatory Sensation Tests

Six healthy female subjects, 27±10 years old, participated in the tests in which various tastes and textures were evaluated. No subject reported having a cold or other respiratory tract infection in the week prior to testing. The subjects were asked to refrain from eating, drinking, or chewing gum for at least 1 h prior to testing. All subjects were non-smokers and signed an informed consent before the experiments. The experimental protocol of this study (No. 15–86) was approved in advance, on Feb. 29, 2016, by the ethical committee of Mukogawa Women’s University.

The gustatory sensation test to measure bitterness intensity was performed with six well-trained volunteers according to a modified previously described method.²¹⁾ Quinine hydrochloride solution at concentrations of 0.01, 0.03, 0.1, 0.3 and 1.0 mM were prepared as the standard for bitterness. Bitterness scores of 0, 1, 2, 3 and 4 were allocated to these increasing concentrations of the standard solution. Half millimolar DPH solution with/without CGA, QNA or CFA at concentrations of 0.1, 0.5 and 1.0 mM were prepared as the samples. The sample concentrations were the same as that in evaluation using taste sensor. Before testing, the volunteers were asked to keep 2 mL of standard quinine hydrochloride solution in their mouths for 5 s and were told the concentration and bitterness score of each solution. In the test, the volunteers were asked to keep 2 mL of sample solution in their mouths for 5 s and were evaluated each of the samples bitterness scores. After tasting each sample, subjects gargled well and waited for at least 20 min before tasting the next sample.

¹H-NMR Spectroscopic Analysis

The ¹H-NMR spectra were measured on a JEOL 500 MHz spectrometer using DMSO-d₆ as a solvent and tetramethylsilane (TMS) as an internal standard. ¹H-NMR spectra were acquired at 308 K (35°C) with a 7 s relaxation delay. DPH, CGA, QNA or CFA solutions, DPH with CGA, QNA or CFA solutions were prepared as sample solutions. The mixing ratio of CGA, QNA or CFA to DPH in the sample solution is 0.2, 1 and 2 by molar ratio. The mixing ratio was the same as that of evaluation in the taste sensor and gustatory sensation test.

Experiments

The bitterness intensities of DPH solutions, in the absence and presence of CGA, QNA and CFA, were evaluated in the taste sensor using membrane AN0 and in human gustatory sensation tests. ¹H-NMR spectroscopic analysis was performed to investigate the interaction between DPH and the three acids.

Statistical Analysis

Ekuseru-Toukei 2010 (Social Survey Research Information Co., Ltd., Tokyo, Japan) was used for statistical analysis. The Bonferroni test was used for multiple comparisons. Correlation was examined using Spearman’s correlation test. The 5% level of probability was considered significant.

Results and Discussion

The Influence of CGA on the Bitterness Intensity of DPH Measured by the Taste Sensor and Gustatory Sensation Tests

Figure 2 shows the effect of dose-dependent addition of the three acids (CGA, QNA and CFA) on CPA_AN0 of DPH solution. CPA means change in sensor membrane potential caused by adsorption of bitter substance to sensor membrane. CPA is not influenced by pH and electrical conductivity and consequently has high selectivity for bitter substance. Especially, in the case of using sensor AN0, bitterness taste information converted from CPA outputs.²²⁾ And, we also showed that the bitterness intensity of DPH could be evaluated using CPA values of sensor AN0 using the taste sensor in the previous study.²⁰⁾ Therefore, the CPA of AN0 (CPA_AN0) was taken to predict the bitterness intensity of the DPH in this study.

Fig. 2. The Dose-Dependent Effect of CGA, QNA and CFA on the CPA_AN0 of DPH Solution

n=3, the mean±S.D., *** p<0.001 vs. added acid 0 mM, ^††† p<0.001 vs. added acid 0.1 mM, ^# p<0.05, ^### p<0.001 vs. added acid 0.5 mM (Tukey test).

The CPA_AN0 values of 0.5 mM DPH solution mixed with 0.1, 0.5 and 1 mM of all three acids were decreased in a dose-dependent manner, although to a lesser degree with CFA than with CGA or QNA. In particular, CPA_AN0 values of 0.5 mM DPH solution with 0.5 and 1 mM of the acids were significantly decreased by dose-dependent addition of the acids.

Figure 3 shows the dose-dependent effect of the addition of the three acids (CGA, QNA and CFA) on the bitterness intensity of DPH solution. The bitterness intensities of all concentrations of all three acids measured alone were less than tau 1, which was the bitterness threshold (data not shown). Coffee is a rich source of polyphenol compounds, such as CGA, which exert an antioxidant activity.²³⁾ CGA is reportedly the main source of the bitter taste of coffee.²⁴⁾ However, the bitterness intensity of 1 mM CGA, the maximum concentration used in this study, was still less than tau 1, which is the bitterness threshold. This therefore suggests that the bitterness intensity of CGA itself would have no influence on bitterness intensity of DPH in this study. In fact, the bitterness intensity of a 0.5 mM DPH solution was decreased to a similar extent by the addition of 0.1, 0.5 and 1 mM solutions of all three acids in a dose-dependent manner.

Fig. 3. The Dose-Dependent Effect of CGA, QNA and CFA on the Bitterness Intensity of DPH Solution as Measured in Gustatory Sensation Tests

n=10, the mean±S.D., * p<0.05, ** p<0.01, *** p<0.001 vs. added acid 0 mM, ^† p<0.05 vs. added acid 0.1 mM (Tukey test).

Figure 4 shows the correlation between CPA_AN0 and bitterness intensities of 0.5 mM DPH solution with 0.1, 0.5 and 1 mM concentrations of the three acids. The reduction in the taste sensor outputs of DPH caused by addition of CGA, QNA and CFA correlates well with the human sensory scores. There was a significantly high correlation between CPA_AN0 and bitterness intensities of 0.5 mM DPH solution with 0.1, 0.5, 1 mM solutions of all three acids (r=0.78, Spearman’s correlation test, p<0.05).

Fig. 4. The Correlation between CPA_AN0 and Bitterness Intensities by Gustatory Sensation Test at 0.5 mM of DPH Solution with 0, 0.1, 0.5 and 1 mM Concentrations of CGA, QNA or CFA

r=0.78, p<0.05 (Spearman’s correlation test).

These results show that the suppression of bitterness intensity of the bitter basic drug DPH by CGA, QNA and CFA was accurately predicted by the taste sensor.

¹H-NMR Spectroscopic Analysis of the Interaction between CGA and DPH

¹H-NMR was used to evaluate the interaction between CGA and DPH in order to understand the mechanism underlying bitterness suppression of DPH by CGA. ¹H-NMR data of DPH with/without CGA, QNA or CFA are shown in Table 1. The ¹H-NMR spectrum of DPH is enumerated as follows; 2.763 ppm (6H, s, (CH₃)₂), 3.323 ppm (2H, t, J=5.9 Hz, NCH₂), 3.710 ppm (2H, t, J=5.9 Hz, CH₂O), 5.551 ppm (1H, s, CH), 7.242–7.424 ppm (10H, m, Ar-H). In the ¹H-NMR spectrum of the mixture of DPH and three acids, the signals of dimethyl proton 12, 19 (proton No. shows in Fig. 1(d)) of DPH were shifted slightly downfield in three acids dose-dependent manner. In the ¹H-NMR spectrum of DPH with CGA, dimethyl proton shifted from 2.763 (DPH only) to 2.767 ppm (mixing ratio of CGA to DPH; 0.2) (Δ0.004), to 2.780 ppm (mixing ratio of CGA to DPH; 1) (Δ0.017), to 2.779 ppm (mixing ratio of CGA to DPH; 2) (Δ0.016 ppm). In the ¹H-NMR spectrum of DPH with QNA, dimethyl proton shifted from 2.763 (DPH only) to 2.767 ppm (mixing ratio of QNA to DPH; 0.2) (Δ0.004 ppm), to 2.771 ppm (mixing ratio of QNA to DPH; 1) (Δ0.008 ppm), to 2.789 ppm (mixing ratio of QNA to DPH; 2) (Δ0.026 ppm). In the ¹H-NMR spectrum of DPH with CFA, dimethyl proton shifted from 2.763 (DPH only) to 2.768 ppm (mixing ratio of CFA to DPH; 0.2) (Δ0.005), to 2.773 ppm (mixing ratio of CFA to DPH; 1) (Δ0.010), to 2.776 ppm (mixing ratio of CFA to DPH; 2) (Δ0.013 ppm). It suggests that the electron density was decreased at this location, and that a downfield shift occurred due to a deshielding effect. Ikeda et al. reported that the electron density near to the nitrogen atom was decreased by interaction.²⁵⁾ It is therefore suggested that DPH interacts with CGA, CFA and QNA in the vicinity of the nitrogen atom of DPH. Especially in case of mixing ratio of three acids to DPH was 2, signal shift of DPH with three acids showed high value in order of QNA, CGA and CFA. It was suggested that QNA part in the CGA structure was easier to interact with DPH than CFA.

The spectrum of three acids obtained in the present study agreed with the results of each pure acid reported in previous studies.^26–28) From the spectrum of CGA mixed with DPH, it was suggested that CGA was not hydrolyzed to CFA and QNA. CGA was not decomposed in the condition of this study. In the ¹H-NMR spectrum of three acids with DPH, the signals of vicinity of carboxyl group of three acids shifted slightly upfield in DPH dose-dependent manner. ¹H-NMR data of vicinity of carboxyl group of CGA with/without DPH (proton 7, 11, 12) are shown in Table 2. In the ¹H-NMR spectrum of CGA with DPH, multiplet signal of proton 11, 12 of CGA shifted from 1.776–2.037 ppm (CGA only) to 1.790–2.045 ppm (mixing ratio of DPH to CGA; 0.5) (Δ0.014–Δ0.008 ppm), to 1.793–2.047 ppm (mixing ratio of DPH to CGA; 1) (Δ0.017–Δ0.010 ppm), to 1.798–2.048 ppm (mixing ratio of DPH to CGA; 5) (Δ0.022–Δ0.011 ppm). ¹H-NMR data of vicinity of carboxyl group of QNA with/without DPH (proton 4, 5) are shown in Table 3. In the ¹H-NMR spectrum of QNA with DPH, multiplet signal of proton 11, 12 of QNA shifted from 1.675–1.900 ppm (QNA only) to 1.681–1.906 ppm (mixing ratio of DPH to QNA; 0.5) (Δ0.006–Δ0.006 ppm), to 1.680–1.909 ppm (mixing ratio of DPH to QNA; 1) (Δ0.005–Δ0.009 ppm), to 1.691–1.927 ppm (mixing ratio of DPH to QNA; 5) (Δ0.016–Δ0.027 ppm). ¹H-NMR data of vicinity of carboxyl group of CFA with/without DPH (proton 7) are shown in Table 4. In the ¹H-NMR spectrum of CFA with DPH, proton 7 of CFA shifted from 6.160 (CFA only) to 6.172 ppm (mixing ratio of DPH to CFA; 0.5) (Δ0.012), to 6.174 ppm (mixing ratio of DPH to CFA; 1) (Δ0.014), to 6.200 ppm (mixing ratio of DPH to CFA; 5) (Δ0.040 ppm). It suggests that the electron density was increased at this location, and that an upfield shift occurred due to a shielding effect. It is therefore suggested that three acids interact with DPH in the vicinity of the carboxyl group.

Table 1. ¹H-NMR Data of DPH with/without CGA, CFA and QNA

	Mixing ratio to DPH	Proton
		12, 19	10	9	7	2–6, 14–18
DPH	—	2.763	3.323	3.710	5.551	7.242–7.424
DPH+CGA	0.2	2.767	3.331	3.717	5.554	7.243–7.428
	1	2.780	3.329	3.693	5.550	7.246–7.422
	2	2.779	3.332	3.693	5.548	7.248–7.421
DPH+QNA	0.2	2.767	3.335	3.729	5.561	7.240–7.454
	1	2.771	3.332	3.708	5.551	7.243–7.444
	2	2.789	3.341	3.690	5.551	7.256–7.419
DPH+CFA	0.2	2.768	3.331	3.729	5.561	7.240–7.439
	1	2.773	3.331	3.705	5.550	7.243–7.427
	2	2.776	3.324	3.685	5.547	7.251–7.427

(ppm).

Table 2. ¹H-NMR Data of Vicinity of Carboxyl Group of CGA with/without DPH

	Mixing ratio to CGA	Proton
		7	11, 12
CGA	—	6.150	1.776–2.037
DPH+CGA	0.5	6.156	1.790–2.045
	1	6.159	1.793–2.047
	5	6.157	1.798–2.048

(ppm).

Table 3. ¹H-NMR Data of Vicinity of Carboxyl Group of QNA with/without DPH

	Mixing ratio to QNA	Proton
		4, 5
QNA	—	1.675–1.900
DPH+QNA	0.5	1.681–1.906
	1	1.680–1.909
	5	1.691–1.927

(ppm).

Table 4. ¹H-NMR Data of Vicinity of Carboxyl Group of CFA with/without DPH

	Mixing ratio to CFA	Proton
		7
CFA	—	6.160
DPH+CFA	0.5	6.172
	1	6.174
	5	6.200

(ppm).

To discuss the influence of interaction in DPH and acids for bitterness suppression, ionization of each substance was predicted using Marvin sketch software. All of amine group of DPH in 0.5 mM DPH solution with three acids at 0.1, 0.5 and 1 mM were predicted to be ionized. Many carboxyl group of CGA, CFA and QNA were predicted to be ionized in 0.5 mM DPH solution with three acids at 0.1, 0.5 and 1 mM. The percentage of ionized carboxyl group were predicted enumerated follows, CGA; 53.8–88.0% (pH 3.39–4.10), QNA 46.5–89.7% (pH 3.47–4.24) and CFA; 58.9–90.0% (pH 3.89–4.56). From data of ¹H-NMR and prediction in ionization of each substance using Marvin sketch software, electrostatic interactions between the positive charge on the amine group of DPH and the negative charges on the carboxyl groups of CGA, QNA and CFA are indicated. In particular, it is suggested that the structure of QNA, which is also part of CGA, is important in the interaction with DPH.

Bitter basic drugs are adsorbed on the positively charged and hydrophobic part of the taste sensor membrane and probably cause a change in membrane potential by changing the charge density of the taste sensor output.²²⁾ It is suggested that the positively charged amine group of DPH is adsorbed on the AN0 taste sensor membrane. If CGA is added to the DPH solution, the adsorption of DPH to the taste sensor membrane decreases due to electrostatic interactions between the positive charge on the amine group of DPH and the negative charge on the carboxyl group of CGA. The size of the decrease of sensor output values varies according to the acid used. The reason for this variation is unclear. It is thought that the relative strengths of the electrostatic interaction between DPH and the different acids vary with the acidity of the acid. Thus, the degrees of sensor output attenuation in 0.5 mM DPH solution with CGA and QNA were greater than with CFA because of the more potent acidities of CGA (pK_a 3.33) and QNA (pK_a 3.46) compared with CFA (pK_a 3.65). Another reason for the variation according to the acid used may be the number of hydrophilic groups contained in each structure. Three acids have hydroxyl group in addition to carboxyl group as hydrophilic groups. Both of aromatic and non-aromatic hydroxyl group cause OH–π interaction with aromatic rings by hydrogen bond.^29,30) The numbers of hydroxyl group contained in CGA, QNA and CFA are 5, 4 and 2, respectively. Many hydroxyl groups may be easy to cause OH–π interaction. The number of OH–π interactions by hydrogen bond of CGA, QNA or CFA with aromatic rings of DPH may also be related to the variation of bitterness suppression according to the acid. Ogata et al. reported that the taste-masking effect of a combination of basic propiverine and acidic compounds is caused by an interaction between the nitrogen of propiverine and the acids.³¹⁾ In our study, DPH with CGA/QNA is also a combination of a basic and an acidic compound. This electrostatic interaction was not involved in the almost case of bitterness suppression between basic drug and bitterness-masking agents. However, CGA and QNA were suggested to suppress the bitterness of bitter substance which has negative ion-charged group as bitterness-masking agent by this electrostatic interaction just only mixing basic drug and bitterness-masking agent.

Conclusion

The three acids CGA, QNA and CFA all suppressed the bitterness intensity of DPH in a dose-dependent manner, although the effect was greater with CGA and QNA than with CFA. ¹H-NMR analysis suggested that an electrostatic interaction between the amine group of DPH and the carboxyl group of CGA/QNA reduced the membrane adsorption of DPH in the taste sensor and binding with the bitterness receptor in gustatory sensation tests. CGA and QNA may therefore be useful bitterness-masking agents for the basic drug DPH, with an electrostatic interaction as the underlying mechanism.

Conflict of Interest

The authors declare no conflict of interest.

References

• 1) Uchida T., Miyanaga H., Tanaka H., Wada K., Kurosaki S., Ohki T., Yoshida M., Matsuyama K., Chem. Pharm. Bull., 48, 1845–1848 (2000).
• 2) Uchida T., Tanigake A., Miyanaga Y., Matsuyama K., Kunitomo K., Kobayashi K., Ikezaki H., Taniguchi A., J. Pharm. Pharmacol., 55, 1479–1485 (2003).
• 3) Miyanaga Y., Mukai J., Mukai T., Odomi M., Uchida T., Chem. Pharm. Bull., 52, 490–493 (2004).
• 4) Kataoka M., Miyanaga Y., Tsuji E., Uchida T., Int. J. Pharm., 279, 107–114 (2004).
• 5) Mukai J., Miyanaga Y., Ishizaka T., Asaka K., Nakai Y., Tsuji E., Uchida T., Chem. Pharm. Bull., 52, 1416–1421 (2004).
• 6) Harada T., Narazaki R., Nagira S., Ohwaki T., Aoki S., Iwamoto K., Chem. Pharm. Bull., 54, 1072–1075 (2006).
• 7) Tokuyama E., Matsunaga C., Yoshida K., Mifsud J. C., Irie T., Yoshida M., Uchida T., Chem. Pharm. Bull., 57, 382–387 (2009).
• 8) Harada T., Uchida T., Yoshida M., Kobayashi Y., Narazaki R., Ohwaki T., Chem. Pharm. Bull., 58, 1009–1014 (2010).
• 9) Ito M., Yoshida M., Kobayashi Y., Hiraoka M., Ikezaki H., Uchida T., Sens. Mater., 23, 483–492 (2011).
• 10) Uchida T., Yoshida M., Hazekawa M., Haraguchi T., Furuno H., Teraoka M., Ikezaki H., J. Pharm. Pharmacol., 65, 1312–1320 (2013).
• 11) Toko K., Matsuno T., Yamafuji K., Hayashi K., Ikezaki H., Sato K., Toukubo R., Kawarai S., Biosens. Bioelectron., 9, 359–364 (1994).
• 12) Kashima A., Kobayashi T., Kobayashi M., Tanaka E., Fukumoto K., Ueno K., Jpn. J. Health Care Sci., 32, 511–516 (2006).
• 13) López-Froilán R., Ramírez-Moreno E., Podio N. S., Pérez-Rodríguez M. L., Cámara M., Baroni M. V., Wunderlin D. A., Sánchez-Mata M. C., Food Funct., 7, 2706–2711 (2016).
• 14) Hashimoto Y., Matsunaga C., Tokuyama E., Tsuji E., Uchida T., Okada H., Chem. Pharm. Bull., 55, 739–746 (2007).
• 15) Stacewicz-Sapuntzakis M., Crit. Rev. Food Sci. Nutr., 53, 1277–1302 (2013).
• 16) Khanal R., Howard L. R., Prior R. L., J. Agric. Food Chem., 62, 3987–3996 (2014).
• 17) Hao Y., Gao R., Liu D., He G., Tang Y., Guo Z., Food Chem., 200, 215–222 (2016).
• 18) Lewandowska H., Kalinowska M., Lewandowski W., Stepkowski T. M., Brzoska K., J. Nutr. Biochem., 32, 1–19 (2016).
• 19) Haraguchi T., Miyazaki A., Yoshida M., Uchida T., Atarashii Ganka (Journal of the Eye), 30, 1619–1622 (2013).
• 20) Ito M., Yoshida M., Kobayashi Y., Hiraoka M., Ikezaki H., Uchida T., Sens. Mater., 23, 483–492 (2011).
• 21) Katsuragi Y., Mitsui Y., Umeda T., Otsuji K., Yamasawa S., Kurihara K., Pharm. Res., 14, 720–724 (1997).
• 22) Kobayashi Y., Habara M., Ikezaki H., Chen R., Naito Y., Toko K., Sensers, 10, 3411–3443 (2010).
• 23) Clifford M. N., Planta Med., 70, 1103–1114 (2004).
• 24) Bicchi C. P., Binello A. E., Pellegrino G. M., Vanni A. C., J. Agric. Food Chem., 43, 1549–1555 (1995).
• 25) Ikeda H., Moriwaki H., Yukawa M., Iwase Y., Aki H., Yakugaku Zasshi, 130, 1589–1595 (2010).
• 26) Forino M., Tenore G. C., Tartaglione L., Carmela D., Novellino E., Ciminiello P., Food Chem., 178, 306–310 (2015).
• 27) López-Martínez L. M., Santacruz-Ortega H., Navarro R. E., Sotelo-Mundo R. R., González-Aguilar G. A., PLoS ONE, 10, PMID: 0140242 (2015).
• 28) Iwai K., Kishimoto N., Kakino Y., Mochida K., Fujita T., J. Agric. Food Chem., 52, 4893–4898 (2004).
• 29) Arunan E., Desiraju G. R., Klein R. A., Sadlej J., Scheiner S., Alkorta I., Clary D. C., Crabtree R. H., Dannenberg J. J., Hobza P., Kjaergaard H. G., Legon A. C., Mennucci B., Nesbitt D. J., Pure Appl. Chem., 83, 1619–1636 (2011).
• 30) Babu M. M., Nucleic Acids Res., 31, 3345–3348 (2003).
• 31) Ogata T., Tanaka D., Ozeki T., Drug Dev. Ind. Pharm., 40, 1084–1091 (2014).