Chiral Recognition of Diketopiperazine Containing Proline Residues by (−)-Epigallocatechin-3-O-gallate in Water

Abstract

The stoichiometry and precipitate yield of a complex of (−)-epigallocatechin-3-O-gallate (EGCg) and cyclo(Pro-Xxx) (Xxx = phenylalanine (Phe), tyrosine (Tyr)) were evaluated using integrated values of their proton signals by quantitative ¹H-NMR (q NMR). It was determined to be a 1 : 1 complex of EGCg and cyclo(Pro-Xxx). The change in the chemical shift value of proton signals of cyclo(Pro-Xxx) in ¹H-NMR spectra by adding standard amounts of EGCg was investigated. Differences in chemical shift values of H_8α, H_7αβ, H_8β, H₁₀, H₉, and H₃ proton signals between cyclo(L-Pro-L-Phe) and cyclo(D-Pro-D-Phe), and those of H_8α, H_7αβ, H_8β, H₁₀, H₉, H₃, and H₁₃ proton signals between cyclo(L-Pro-L-Tyr) and cyclo(D-Pro-D-Tyr) were observed as a significant difference at 54 mmol/L of EGCg. It was found that their chirality was clearly recognized by EGCg. The significant difference in the change of the chemical shift value of H_8α proton signals between cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) was the largest, and the difference was considered to have resulted from the difference in the ratio of extended conformer in equilibrium between folded and extended conformers. Such a significant difference in change values between cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx) was not observed due to a rigid intramolecular CH–π interaction. EGCg did not clearly recognize the chirality of cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx).

Introduction

When a hot tea beverage cools, turbid and brown-white particles occur and then precipitate. This phenomenon is called the “creaming-down reaction.” It is well-known that tea catechins such as 2,3-cis gallated catechins (−)-epigallocatechin-3-O-gallate (EGCg), (−)-epicatechin-3-O-gallate (ECg), and caffeine are abundantly contained in the precipitate of the creaming-down reaction.¹⁾ Therefore, we attempted crystallization of the precipitate made from an aqueous solution of EGCg and caffeine, and ECg and caffeine. The crystals obtained were determined to be a 2 : 2 complex of EGCg and caffeine^2,3) (Fig. 1a) and a 2 : 4 complex of ECg and caffeine⁴⁾ by X-ray crystallographic analysis.

Fig. 1. Crystal Structure of 2 : 2 Complex of (−)-Epigallocatechin-3-O-gallate (EGCg) and Caffeine (a), and 2 : 2 Complex of EGCg and Cyclo(Pro-Gly) (b)

In the 2 : 2 complex of EGCg and caffeine, caffeine moieties were positioned in the space surrounding the top and bottom walls of B′ rings and left and right walls of A and B rings of EGCg moieties. Since the space formed with three aromatic A, B, and B′ rings was highly hydrophobic, caffeine molecules were captured into the space in water. It was considered that the solubility of the complex in water rapidly decreased, in comparison with that of EGCg and caffeine alone, and these complexes precipitated from the aqueous solutions. Therefore, complexes of precipitates formed from an aqueous solution of EGCg and a variety of heterocyclic compounds were prepared, and the correlation between the chemical structures of the heterocyclic compounds and molecular capture ability was studied.⁵⁾ Furthermore, since the C ring of EGCg has C₂ and C₃ of two chiral carbon atoms, the hydrophobic space was also a chiral space. It was assumed that the special space formed by EGCg could recognize the chirality of compounds captured.

Many pharmaceuticals are used currently in racemic form, although it is desirable to use a single enantiomer with greater potency, from the viewpoints of adverse effects and pharmacokinetics. EGCg and its derivatives, which easily form a complex with a variety of heterocyclic compounds in water and then precipitate from the solution in water, may be potentially new optical resolving agents for pharmaceuticals and natural products.

Diketopiperazine cyclo(Pro-Gly) was selected as a chiral compound of a substrate of EGCg. Cyclo(Pro-Gly) consist of 5- and 6-membered rings similar to caffeine, and their molecular sizes are approximately the same as caffeine. Cyclo(L-Pro-Gly) and cyclo(D-Pro-Gly) formed 2 : 2 EGCg complexes in the same manner as the 2 : 2 complex of EGCg and caffeine^6,7) (Fig. 1b). However, the difference in ¹H-NMR spectra of the 2 : 2 EGCg complex of cyclo(L-Pro-Gly) and that of cyclo(D-Pro-Gly) was only the difference in chemical shift values of proton signals for methylene protons H_7α, H_7β, and H_8α of the proline (Pro) residue. The differences were difficult to observe as chiral recognition due to small differences in their chemical shifts and overlapping of signals. The diketopiperazines cyclo(Pro-Xxx) (Xxx = phenylalanine (Phe), tyrosine (Tyr)), which have an aromatic ring as a side chain, adopted a folded conformation due to intramolecular CH–π interaction,^8,9) and are bulky molecules in water (Fig. 2). In Fig. 2, the notation of α and β for the methylene hydrogen atom of cyclo(Pro-Xxx) (Xxx = Phe, Tyr) means that H_α is a hydrogen atom positioned the same side as the benzene ring, and H_β is a hydrogen atom positioned the different side as that. Thus, we investigated chiral recognition of cyclo(Pro-Xxx) by EGCg.

Fig. 2. Cyclo(Pro-Xxx) (Xxx = Phe, Tyr)

Experimental

Materials

EGCg was purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.), tert-Butoxycarbonyl (Boc)-L-Pro-OH was from Peptide Institute, Ltd., Osaka, Japan, and Boc-D-Pro-OH, H-L-Xxx-OBzl·p-tosylate, and H-D-Xxx-OBzl·p-tosylate (Xxx = Phe, Tyr) were from Kokusan Kagaku Co., Ltd. (Tokyo, Japan).

The detailed method for synthesizing diketopiperazine cyclo (Pro-Xxx) was described in our previous report.¹⁰⁾

The stereochemical structure of cyclo(Pro-Xxx) was constructed using ChemBio3D Ultra 14.0 (PerkinElmer, Inc., MA, U.S.A.), and then optimized by Molecular Dynamics.

NMR Experiments

¹H-NMR spectra were recorded at 30 °C on JEOL JNM-ECZ400R (Tokyo, Japan) operating at 400 MHz using a 5 mm ϕ sample tube. In general, ¹H-NMR experiments were performed with: spectral width, 7.494 KHz; scan times, 8 times; pulse interval, 5 s. Deuterated water D₂O (99.9 atom % D; FUJIFILM Wako Pure Chemical Corp., Osaka, Japan) and deuterated dimethylsulfoxide (DMSO)-d₆ (Eurosotop) were used as measurement solvents. Chemical shift values are expressed in ppm downfield using deuterated sodium 2, 2-dimethyl-2-silapentane-5-sulfonate (DSS-d₆, Wako Pure Chemical Corp.) as an internal reference substance.

Preparation of Precipitate of Complex of EGCg and Cyclo(Pro-Xxx)

Diketopiperazine cyclo(Pro-Xxx) (1.09 × 10⁻² mmol) and EGCg (4.996 mg, 1.09 × 10⁻² mmol) were dissolved in distilled water (80 and 30 µL, respectively). An aqueous solution of cyclo(Pro-Xxx) was poured into an aqueous solution of EGCg, and the mixture was heated at 90 °C to dissolve completely, and then left at room temperature for 1 h followed by 4 °C for 1 d to obtain a supernatant liquid and sticky precipitate. After removing the supernatant liquid, the sticky precipitate was evaporated under reduced pressure and heated at 40 °C for 1 d to create a solid.

Quantitative (q) NMR Experiments

Quantitative ¹H-NMR (q NMR) was performed with the following optimized parameters: probe temperature, 30 °C; spinning, off; scan times, 8 times; pulse interval, 64 s.

The above solid made from an aqueous solution of cyclo(Pro-Xxx) and EGCg was dissolved in DMSO-d₆ (600 µL) containing DSS-d₆ (5.00 mmol), 540 µL of the solution was put in an NMR sample tube, and q NMR was performed. This measurement was performed three times, and the average value of three integrated values of corresponding signals obtained by q NMR measurement was calculated.

Results and Discussion

Stoichiometry of Complex of EGCg and Cyclo(Pro-Xxx) (Xxx = Phe, Tyr)

A mixture of an equimolecular amount of EGCg and cyclo(Pro-Xxx) (Xxx = Phe, Tyr) in aqueous solution afforded precipitates, which were a complex of EGCg and cyclo(Pro-Xxx). The stoichiometry and precipitate yield of the complexes were estimated by q NMR using integrated values of proton signals of H_2,”6” (2H), H_2,’6′ (2H), H₈ (1H), H₆ (1H) of EGCg, H₃ (1H), H₉ (1H), H_8α (1H), H_7α (1H), and H_8β (1H) of cyclo(Pro-Phe), and H₃ (1H), H₉ (1H), H₁₂ (2H), H₁₃ (2H), H_7α (1H), H_8α (1H), and H_8β (1H) of cyclo(Pro-Tyr). A trimethyl group of DSS-d₆ [singlet (s) of trimethyl group (CH₃)₃-] was used not only as an internal reference substance (0 ppm), but also as an internal standard substance (9H). It was determined to be a 1 : 1 complex of EGCg and cyclo(Pro-Xxx), as shown in Table 1.¹¹⁾

Table 1. Stoichiometry and Precipitate Yield of EGCg Complex of Cyclo(L-Pro-L-Phe) (a), Cyclo(D-Pro-D-Phe) (b), Cyclo(L-Pro-D-Phe) (c), and Cyclo(D-Pro-L-Phe) (d)

(a)
Compound Proton	EGCg				Cyclo(L-Pro-L-Phe)
	H2″6″(2H)	H2´6´(2H)	H8(1H)	H6(1H)	H3(1H)	H9(1H)	H8α(1H)
Chemical shift (ppm)	6.842	6.432	5.856	5.960	4.380	4.103	1.443
Integrated value	5.667	5.373	2.677	2.700	3.000	3.010	2.857
Concentration (mmol/L)	14.167	13.433	13.383	13.500	15.000	15.050	14.283
Yield (%)	70.183	66.550	66.303	66.881	74.312	74.560	70.761
Ratioa)	1.054	1.000	0.996	1.003	1.120	1.123	1.066
(b)
Compound Proton	EGCg				Cyclo(D-Pro-D-Phe)
	H2″6″(2H)	H2´6´(2H)	H8(1H)	H6(1H)	H3(1H)	H9(1H)	H8α(1H)
Chemical shift (ppm)	6.842	6.431	5.858	5.959	4.380	4.102	1.442
Integrated value	5.003	4.767	2.427	2.417	2.747	2.727	2.583
Concentration (mmol/L)	12.508	11.917	12.133	12.083	13.733	13.633	12.917
Yield (%)	61.968	59.037	60.110	59.862	68.037	67.541	63.991
Ratioa)	1.050	1.000	1.018	1.014	1.154	1.146	1.077
(c)
Compound Proton	EGCg				Cyclo(L-Pro-D-Phe)
	H2″6″(2H)	H2´6´(2H)	H8(1H)	H6(1H)	H3(1H)	H8β(1H)	H7a(1H)
Chemical shift (ppm)	6.842	6.431	5.857	5.958	4.020	1.978	1.822
Integrated value	5.503	5.280	2.717	2.693	3.237	3.037	2.830
Concentration (mmol/L)	13.758	13.200	13.583	13.467	16.183	15.183	14.150
Yield (%)	68.161	65.394	67.294	66.716	80.174	75.220	70.101
Ratioa)	1.042	1.000	1.029	1.020	1.226	1.150	1.072
(d)
Compound Proton	EGCg				Cyclo(D-Pro-L-Phe)
	H2″6″(2H)	H2´6´(2H)	H8(1H)	H6(1H)	H3(1H)	H8β(1H)	H7α(1H)
Chemical shift (ppm)	6.841	6.432	5.858	5.956	4.022	1.976	1.819
Integrated value	4.333	4.167	2.073	2.033	2.803	2.183	2.270
Concentration (mmol/L)	10.833	10.417	10.367	10.167	14.017	10.917	11.350
Yield (%)	53.670	51.606	51.358	50.367	69.440	54.083	56.229
Ratioa)	1.040	1.000	0.995	0.976	1.345	1.049	1.090

a) Ratio is when the integrated value of the H_2´,6´ proton signal of EGCg is 1.000.

Chiral Recognition of Cyclo(L-Pro-L-Xxx) and Cyclo(D-Pro-D-Xxx) (Xxx = Phe, Tyr) by EGCg

Diketopiperazine cyclo(Pro-Xxx) (Xxx = Phe, Tyr) has two chiral carbons and four stereochemical isomers: cyclo(L-Pro-L-Xxx) LL form, cyclo(D-Pro-D-Xxx) DD form, cyclo(L-Pro-D-Xxx) LD form, and cyclo(D-Pro-L-Xxx) DL form (Fig. 3). The LL and DD forms, and LD and DL forms are enantiomers of each other (Fig. 3), and ¹H-NMR spectra of cyclo(Pro-Phe) are shown in Fig. 4.¹²⁾

Fig. 3. Four Stereochemical Isomers of Cyclo(Pro-Phe)

Fig. 4. ¹H-NMR Spectra of Cyclo(L-Pro-L-Phe), Cyclo(D-Pro-D-Phe) (a), and Cyclo(L-Pro-D-Phe), Cyclo(D-Pro-L-Phe) (b) in D₂O

Concentration of cyclo(Pro-Phe) is 10 mmol/L.

Cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) formed an intramolecular CH–π interaction between H_8α and their benzene rings, and cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx) formed an intramolecular CH–π interaction between H₉ and their benzene rings^8,9) (Figs. 5a, c). In a crystal state, cyclo(L-Pro-L-Phe) (Fig. 5b) and cyclo(L-Pro-L-Tyr) adopted extended and folded conformations, respectively.⁸⁾ In water, cyclo(Pro-Xxx) adopted a folded conformations, and in organic solvent such as methanol, acetone, and dimethyl sulfoxide, cyclo(Pro-Xxx) was in equilibrium between folded and extended conformers (Fig. 5d).

Fig. 5. Stereochemical Structure of Cyclo(L-Pro-L-Xxx) (Xxx = Phe, Tyr) in Water (a), Cyclo(L-Pro-L-Xxx) in Crystal State (b), and Cyclo(D-Pro-L-Xxx) in Water (c), and Equilibrium of Cyclo(L-Pro-L-Xxx) (Xxx = Phe, Tyr) between Folded and Extended Conformations (d)

Change in chemical shift value of proton signals of cyclo(L-Pro-L-Phe) and cyclo(D-Pro-D-Phe) in ¹H-NMR spectra by adding standard amounts of EGCg is shown in Fig. 6.¹³⁾ When the concentrations of EGCg were 10 and 54 mmol/L, changes in the chemical shift values for proton signals of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx), and differences of the change values between cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) are shown in Table 2.

Fig. 6. Change in Chemical Shift Value of Proton Signals of Cyclo(L-Pro-L-Phe) (a) and Cyclo(D-Pro-D-Phe) (b) by Adding EGCg

Measuring solvent is D₂O, and concentration of cyclo(L-Pro-L-Phe) and cyclo(D-Pro-D-Phe) is 10 mmol/L and EGCg is 0–54 mmol/L.

Table 2. Change Value of Chemical Shift of Proton Signals of Cyclo(L-Pro-L-Xxx) (Xxx = Phe, Tyr) and Cyclo(D-Pro-D-Xxx) at 10 and 54 mmol/L of Concentration of EGCg, and Difference in Change Value between Cyclo(L-Pro-L-Xxx) and Cyclo(D-Pro-D-Xxx)

(a)
Diketopiperazine	EGCg	H8α	H7αβ	H8β	H10	H10	H6	H6	H9	H3	Ar H2	Ar H1
Cyclo(L-Pro-L-Phe)	10 mmol/L	0.010	−0.016	0.002	0.000	−0.023	−0.009	−0.006	−0.044	−0.035	−0.013	−0.006
Cyclo(D-Pro-D-Phe)	10 mmol/L	0.019	−0.011	0.008	0.004	−0.030	−0.010	−0.003	−0.034	−0.036	−0.012	−0.004
	Difference	0.009	0.005	0.006	0.004	0.007	0.001	0.003	0.010	0.001	0.001	0.002
Cyclo(L-Pro-L-Phe)	54 mmol/L	0.042	−0.045	0.008	0.011	−0.066	−0.021	−0.011	−0.138	−0.103	−0.037	−0.013
Cyclo(D-Pro-D-Phe)	54 mmol/L	0.092	−0.020	0.033	0.019	−0.100	−0.020	0.002	−0.108	−0.124	−0.042	−0.011
	Difference	0.050	0.025	0.025	0.008	0.034	0.001	0.013	0.030	0.021	0.015	0.002
(b)
Diketopiperazine	EGCg	H8α	H7αβ	H8β	H10	H10	H6	H6	H9	H3	H13	H12
Cyclo(L-Pro-L-Tyr)	10 mmol/L	0.031	−0.008	0.007	−0.002	−0.025	−0.007	−0.003	−0.046	−0.038	0.008	−0.010
Cyclo(D-Pro-D-Tyr)	10 mmol/L	0.044	0.007	0.012	0.005	−0.042	−0.011	−0.004	−0.040	−0.047	0.003	−0.014
	Difference	0.007	0.015	0.005	0.007	0.017	0.004	0.001	0.006	0.009	0.005	0.004
Cyclo(L-Pro-L-Tyr)	54 mmol/L	0.120	−0.005	0.035	0.041	−0.076	−0.010	0.003	−0.143	−0.112	0.038	−0.027
Cyclo(D-Pro-D-Tyr)	54 mmol/L	0.184	0.027	0.061	0.042	−0.125	−0.008	0.006	−0.108	−0.157	0.010	−0.048
	Difference	0.064	0.032	0.026	0.001	0.049	0.002	0.003	0.035	0.045	0.028	0.019

1) Unit of values is ppm. 2) Measurement of chemical shift value of ptoron signal for H_7β and H_8α was impossible due to a overlap with the other proton signal.

When a significant difference at 54 mmol/L of EGCg was estimated as a difference larger than 0.020 ppm judging from the standard deviation of the chemical shift values, differences in the chemical shift values of H_8α, H_7αβ, H_8β, H₁₀, H₉, and H₃ proton signals between cyclo(L-Pro-L-Phe) and cyclo(D-Pro-D-Phe), and those of H_8α, H_7αβ, H_8β, H₁₀, H₉, H₃, and H₁₃ proton signals between cyclo(L-Pro-L-Tyr) and cyclo(D-Pro-D-Tyr) were observed as a significant difference (Table 2). Their chirality was clearly recognized by EGCg.

Regarding the significant differences, the largest differences in change values between cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) (Xxx = Phe, Tyr) were 0.050 and 0.064 ppm of the H_8α proton signal, respectively. The difference in the change of chemical shift values of the H_8α proton signal between cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) by adding EGCg is shown in Fig. 7.

Fig. 7. Change in Chemical Shift Value of H_8α Proton Signals of Cyclo(L-Pro-L-Phe) and Cyclo(D-Pro-D-Phe) (a), Cyclo(L-Pro-L-Tyr) and Cyclo(D-Pro-D-Tyr) (b) by Adding EGCg

Measuring solvent is D₂O, and concentration of cyclo(L-Pro-L-Xxx) (Xxx = Phe, Tyr) and cyclo(D-Pro-D- Xxx) is 10 mmol/L and EGCg is 0–54 mmol/L.

The H_8α proton signals of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx), which were observed at 0.764 and 0.721 ppm, respectively, in a high field due to an intramolecular CH–π interaction between H_8α and their benzene rings, shifted downfield by the addition of EGCg. It was considered that the ratio of extended conformer of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) increased by adding EGCg. Chemical shift values of proton signals for H_8α of cyclo(L-Pro-L-Xxx) (Xxx = Phe, Tyr) on adopting a folded conformation were 0.764 and 0.721 ppm, respectively. When cyclo(L-Pro-L-Xxx) adopted an extended conformation in water, chemical shift values of the proton signals for H_8α were expected to be 1.637 and 1.644 ppm, respectively. The values were derived from the chemical shift value of the H_8β proton signal of cyclo(D-Pro-L-Xxx) without formation of an intramolecular CH–π interaction between H_8β and its benzene ring (Fig. 5). Thus, the ratio of extended conformer of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) was estimated by Eqs. 1 and 2, and the change in the ratio of extended conformer of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) to the concentration of EGCg is shown in Fig. 8. The significant difference in the change of the chemical shift value of H_8α proton signals between cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) was thought to have resulted from the difference in the ratio of extended conformer in equilibrium between folded and extended conformers. Since the ratio of extended conformer of cyclo(Pro-Xxx) increased by adding EGCg in water, it was considered that upon the 1 : 1 complex formation of EGCg and cyclo(Pro-Xxx), cyclo(Pro-Xxx) changed from folded to extended conformers.

  

(1)

Eq. 1. Ratio of Extended Conformer of Cyclo(L-Pro-L-Phe) and Cyclo(D-Pro-D-Phe)

  

(2)

Eq. 2. Ratio of Extended Conformer of Cyclo(L-Pro-L-Tyr) and Cyclo(D-Pro-D-Tyr)

Fig. 8. Change in Ratio of Extended Conformer of Cyclo(L-Pro-L-Phe) and Cyclo(D-Pro-D-Xxx) (a), and Cyclo(L-Pro-L-Phe) and Cyclo(D-Pro-D-Xxx) (b)

Measuring solvent is D₂O, and concentration of cyclo(L-Pro-L-Xxx) (Xxx = Phe, Tyr) and cyclo(D-Pro-D- Xxx) is 10 mmol/L and EGCg is 0–54 mmol/L.

Change of Cyclo(L-Pro-D-Xxx) and Cyclo(D-Pro-L-Xxx) (Phe. Tyr) by Addition of EGCg

Changes in chemical shifts of proton signals of cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx) (Phe. Tyr) in ¹H-NMR spectra by adding standard amounts of EGCg are shown in Fig. 9.¹⁴⁾

Fig. 9. Change in Chemical Shift Value of Proton Signals of Cyclo(L-Pro-D-Phe) (a) and Cyclo(D-Pro-L-Phe) (b) by Adding EGCg

Measuring solvent is D₂O, and concentration of cyclo(L-Pro-L-Phe) and cyclo(D-Pro-D-Phe) is 10 mmol/L and EGCg is 0–54 mmol/L.

The changes of chemical shifts were small, in comparison with those of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) (Fig. 6). It was because the intramolecular CH–π interaction of cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx) between H₉ and their benzene rings was more rigid than that of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) between H_8α and their benzene rings,^8,9) and a change from folded to extended conformation was hard to occur. No significant difference in change values between cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx) at 54 mmol/L of EGCg was observed (Table 3). Thus, EGCg did not clearly recognize their chirality.

Table 3. Change Value of Chemical Shift of Proton Signal of Cyclo(L-Pro-D-Xxx) (Xxx = Phe, Tyr) and Cyclo(D-Pro-L-Xxx), and Difference in Change Value between Cyclo(L-Pro-D-Xxx) and Cyclo(D-Pro-L-Xxx)

(a)
Diketopiperazine	EGCg	H7β	H8β	H9	H10	H6	H6	H3	Ar H2	Ar H1
Cyclo(L-Pro-D-Phe)	10 mmol/L	−0.009	0.009	0.006	−0.008	0.000	−0.008	−0.004	−0.005	0.002
Cyclo(D-Pro-L-Phe)	10 mmol/L	−0.007	0.011	0.005	−0.007	0.002	−0.005	−0.002	−0.008	0.000
	Difference	0.002	0.002	0.001	0.001	0.002	0.003	0.002	0.003	0.002
Cyclo(L-Pro-D-Phe)	54 mmol/L	−0.030	0.037	0.032	−0.024	0.006	−0.028	−0.002	−0.020	−0.001
Cyclo(D-Pro-L-Phe)	54 mmol/L	−0.029	0.038	0.024	−0.020	0.002	−0.012	0.000	−0.026	−0.003
	Difference	0.001	0.001	0.008	0.004	0.004	0.016	0.002	0.006	0.002
(b)
Diketopiperazine	EGCg	H7β	H8β	H9	H10	H10	H6	H6	H3	H13	H12
Cyclo(L-Pro-D-Tyr)	10 mmol/L	−0.019	0.007	0.016	−0.008	−0.015	−0.007	−0.017	−0.012	0.010	−0.009
Cyclo(D-Pro-L-Tyr)	10 mmol/L	−0.002	0.000	0.007	−0.015	−0.011	−0.008	−0.015	−0.016	0.005	−0.014
	Difference	0.007	0.007	0.009	0.007	0.004	0.001	0.002	0.004	0.005	0.005
Cyclo(L-Pro-D-Tyr)	54 mmol/L	−0.054	0.038	0.069	−0.009	−0.033	0.001	−0.047	−0.016	0.046	−0.017
Cyclo(D-Pro-L-Tyr)	54 mmol/L	−0.049	0.026	0.052	−0.036	−0.027	0.003	−0.027	−0.022	0.040	−0.014
	Difference	0.005	0.012	0.017	0.027	0.006	0.002	0.020	0.006	0.006	0.003

1) Unit of values is ppm. 2) Measurement of chemical shift value of ptoron signal for H_7β and H_8α was impossible due to a overlap with the other proton signal.

Conclusion

Chiral recognition of diketopiperazine cyclo(Pro-Xxx) (Phe, Tyr) by EGCg was investigated. A significant difference in the change of the chemical shift value of some proton signals between cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) by adding EGCg was observed. The chirality of cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) was clearly recognized by EGCg. A significant difference in the change of the chemical shift value of H_8α was considered to have resulted from the difference between cyclo(L-Pro-L-Xxx) and cyclo(D-Pro-D-Xxx) in the ratio of extended conformer in equilibrium between folded and extended conformers.

Such a significant difference in change values between cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx) was not observed due to the rigid intramolecular CH–π interaction. EGCg did not clearly recognize the chirality of cyclo(L-Pro-D-Xxx) and cyclo(D-Pro-L-Xxx).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References and Notes

• 1) Maruyama N., Suzuki Y., Sakata K., Yagi A., Ina K., Proc. International Symposium Tea Science, 145–149 (1991).
• 2) Tsutsumi H., Sato T., Ishizu T., Chem. Lett., 41, 1669–1671 (2012).
• 3) Ishizu T., Tsutsumi H., Kinoshita Y., Mukaida H., Sato T., Kajitani S., Chem. Pharm. Bull., 62, 552–558 (2014).
• 4) Ishizu T., Sato T., Tsutsumi H., Yamamoto H., Chem. Lett., 39, 607–609 (2010).
• 5) Tsutsumi H., Sato A., Fujino S., Fujioka Y., Ishizu T., Chem. Pharm. Bull., 67, 501–504 (2019).
• 6) Ishizu T., Tsutsumi H., Yokoyama A., Tetrahedron Lett., 56, 1111–1114 (2015).
• 7) Ishizu T., Tsutsumi H., Yokoyama E., Tanabe H., Yokoyama A., Chem. Pharm. Bull., 64, 142–149 (2016).
• 8) Ishizu T., Tsutsumi H., Yokoyama E., Kawmoto H., Yokota R., Chem. Pharm. Bull., 65, 598–602 (2017).
• 9) Ishizu T., Sato P., Tsuyama S., Nagao R., Fujiki K., Yamaji A., Analytical Science Advances, 3, 38–46 (2022).
• 10) Ishizu T., Tokunaga M., Fukuda M., Matsumoto M., Goromaru T., Takemoto S., Chem. Pharm. Bull., 69, 585–589 (2021).
• 11) Stoichiometry and precipitate yield of EGCg complexes of cyclo(L-Pro-L-Tyr), cyclo(D-Pro-D-Tyr), cyclo(L-Pro-D-Tyr), and cyclo(D-Pro-L-Tyr) is shown in Table 4 of Supplementary Materials.
• 12) ¹H-NMR spectra of cyclo(Pro-Tyr) are shown in Fig. 10 of Supplementary Materials.
• 13) Change in chemical shift value of proton signals of cyclo(L-Pro-L-Tyr) and cyclo(D-Pro-D-Tyr) in ¹H-NMR spectra by adding EGCg is shown in Fig. 11 of Supplementary Materials.
• 14) Change in chemical shift value of proton signals of cyclo(L-Pro-D-Tyr) and cyclo(D-Pro-L-Tyr) in ¹H-NMR spectra by adding EGCg is shown in Fig. 12 of Supplementary Materials.