Research Article - (2017) Volume 5, Issue 1
Flavan-3-ol derivatives are common plant-derived bioactive compounds with strong various biological activities. In particular, (–)-epigallocatechin-3-O-gallate exhibits a variety of moderate biological activities without severe toxicity. Its health-promoting effects have been widely studied because it is a main ingredient in green tea and is commercially available at low cost. In general, the galloyl moieties of flavan-3-ol derivatives are thought to be essential to their strong biological activities. However, it is not clear which position is most effective for modification with the galloyl moiety to strengthen the biological activities because various galloylated analogs are difficult to obtain in pure form in a quantity large enough for assays. Therefore, we synthesized various galloylated flavan-3-ol derivatives in a stereoselective and regioselective manner. We reported in a previous paper that 3,5-digalloyl-(–)-epicatechin displayed much stronger inhibitory activity than 3,5-digalloyl-(+)-catechin did against HeLa S3 cell proliferation. In this study, we describe synthetic studies of 7-galloyl- and 3,7-digalloyl-modified (–)-epicatechin and (+)-catechin and compared their biological activities, HeLa S3 cell proliferation inhibitory activity and DPPH radical scavenging activity with those of 3,5- digalloyl- (–)-epicatechin. The results indicated that 3,7-digalloyl derivative did not inhibit HeLa S3 cell proliferation. Furthermore, 3,5-digalloyl- (–)-epigallocatechin was synthesized to evaluate the importance of the number of phenolic hydroxyl groups on the B-ring. Contrary to our expectations, 3,5-digalloyl-(–)-epigallocatechin exhibited a weaker inhibition of HeLa S3 cell proliferation than 3,5-digalloyl- (–)-epicatechin did. The DPPH radical scavenging activity of the synthesized compounds suggested that the galloyl-modified position and the number of phenolic hydroxyl groups on B-ring affected to the radical scavenging ability. In conclusion, we found that 3,5-digalloyl-(–)-epicatechin exhibited a superlative HeLa S3 cell proliferation inhibitory effect among the galloylated flavan-3-ol derivatives we synthesized and DPPH radical scavenging activity was affected by the galloyl-moiety-introduced position and the number of phenolic hydroxyl groups on the B-ring.
Keywords: Digalloyl-flavan-3-ols; DPPH radical scavenging activity; Stereoselective and regioselective synthesis; Inhibitory activity; HeLa S3 cells proliferation
EGCG: Epigallocatechin-3-O-gallate; DPPH: 2,2-diphenyl-1-picrylhydrazyl; BSA: Bovine serum albumin; TBS: tertbutyldimethylsilyl; DCC: Dicyclohexylcarbodiimide; TBAF: Tetra-nbutylammonium fluoride; EDC: 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride; THF: Tetrahydrofuran; DMF: N,Ndimethylformamide; TFA: Trifluoroacetic acid; DMAP: N,Ndimethyl- 4-aminopyridine; MTT: 3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyltetrazolium bromide; SAR: Structure activity relationship; HMQC: Heteronuclear multiple quantum correlation; HMBC: heteronuclear multiple bond coherence.
There is currently great interest in the investigation of compounds with strong biological activities that are derived from food sources because these compounds are generally considered safe owing to their consumption as part of the general daily diet. Polyphenols are thought to have a variety of health benefits and are found in many health foods, vegetables, and fruits [1,2]. However, the structure-activity relationships (SARs) of polyphenols are not well understood because polyphenols are typically obtained as a mixture of various analogs, which makes purification difficult. The flavan-3-ol series is one of the most well-known groups of biologically active polyphenol compounds and includes (–)-epigallocatechin-3-O-gallate (EGCG) (1) (Figure 1). (–)-EGCG (1) exhibited higher anti-proliferative activity against the cervical epithelioid carcinoma cell line, HeLa S3, then (–)-epicatechin (2) and (+)-catechin (3) did (Figure 1) owing to the greater number of phenolic hydroxyl groups on the B-ring and the presence of galloyl moieties. Although other flavan-3-ols are found naturally as minor components in plants, it has thus far been difficult to conduct detailed investigations of the wide variety of these compounds with standard biological assays because they are only available in limited quantities.
In our previous paper, we reported that multiple gallo-moieties in one molecule of flavan-3-ol affect inhibitory activity against HeLa S3 cell proliferation [3]. Interestingly, the effect of gallo-moieties is potentially limited depending on the stereochemistry. We synthesized the flavan-3-ols 4 and 5, which were modified with two galloyl moieties and derived from (–)-epicatechin and (+)-catechin, respectively (Figure 2). Among them, only compound 4 showed strong inhibitory activity against the proliferation of HeLa S3 cells.
In this study, we described the synthesis of 7-galloyl-(–)-epicatechin and (+)-catechin series 6−9 to evaluate the role of the position of the galloyl moiety. In addition, we synthesized 3,5-di-galloyl-(–)- epigallocatechin (10) (5-galloyl derivative of EGCG (1)) to assess the importance of the number of phenolic hydroxyl groups on the B-ring (Figure 3). These compounds could be synthesized because of the discovery of the deprotection method of tert-butyldimethylsilyl (TBS) ether in the 7-position on the A-ring in basic conditions. In a previous paper, we reported the selective deprotection method of TBS ether in the 5-position on the A-ring of TBS-protected (–)-epicatechin (2) and (+)-catechin (3) in acidic conditions for the synthesis of 5-modified flavan-3-ols 4 and 5 regioselectively [3]. This acidic deprotection method is generally applicable to other TBS-protected flavan-3-ols to enable the regioselective synthesis of 10. Our previous [3] and current study using the synthesized galloyl-modified flavan-3-ol series, which are usually difficult to obtain from plants, indicated that the structure of 3,5-digalloyl and two hydroxyl groups on the B-ring were important for inhibitory activity against HeLa S3 cell proliferation.
General
All commercially available chemicals for chemical synthesis were used without further purification. All reactions were performed under an argon atmosphere and monitored using thin-layer chromatography TLC) with 0.25 mm pre-coated silica gel plates 60F254 (Art 5715, Merck KGaA, Darmstadt, Germany). An ATAGO (Minato, Japan) AP-300 spectrometer was used to measure optical rotation. 1H- and 13C-NMR spectra (400/100 MHz) were recorded on a DD2 NMR Spectrometer (Agilent, Santa Clara, CA, USA). A microTOFfocus mass spectrometer (Bruker Daltonics, Billerica, MA, USA) was used to acquire electrospray ionization (ESI) mass spectra. Synthesized compounds were dissolved in dimethyl sulfoxide (DMSO) and stored at –20°C. HPLC purification was carried out on an Ascentis® column (SUPELCO® analytical, Sigma Aldrich Co., St. Louis, MO, USA; 25 cm × 21.5 mm, 5 μm) using the solvents (A) 0.05% HCOOH in CH3CN and (B) 0.05% HCOOH and 10% CH3CN in H2O. Elution was done with a linear gradient 20%–100% B in 25 min (flow rate, 4.0 mL/min). The human cervical adenocarcinoma cell line, HeLa-S3, was provided by the RIKEN BRC through the National Bio-Resource Project of the MEXT (Tsukuba, Japan). The inhibitory activity of HeLa S3 cells proliferation and the DPPH scavenging activity were estimated with a microplate reader (Filter Max F5 multi-mode microplate reader; Molecular Devices, Downingtown, PA, USA).
Synthesis
Deprotection of 7- or 5-O-TBS group of compound 11 (13 and 15): To a solution of 5,7,3’,4’-tetra-O-TBS-(−)-epicatechin (11) (0.55 g, 0.74 mmol) in MeOH (40 mL), CH2Cl2 (10 mL), K2CO2 (0.41 g; 3.00 mmol) was added slowly at 0°C to RT. After stirring for 2 h, the pale yellow reaction mixture was quenched with sat. NH4Cl aq. The aqueous solution was extracted with CHCl3 and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica-gel column purification (n-hexane/EtOAc, 20:1 to 5:1) afforded 0.12 g of 13 (0.19 mmol, 26%) and 0.10 g of 15 (0.16 mmol, 22%) as white powder, respectively. And 0.046 g (0.061 mmol, 8.2%) of 11 was recovered. Data for 13: [α]25 D –5.9 (c 0.34, CHCl3); 1H NMR (400 MHz, CDCl3) 6.96 (1H, d, J = 1.8 Hz), 6.92 (1H, dd, J = 1.8, 8.3 Hz), 6.86 (1H, d, J = 8.3 Hz), 6.11 (1H, d, J = 2.4 Hz), 6.00 (1H, d, J = 2.4 Hz), 4.90 (1H, br s), 4.56 (1H, s), 4.24-4.22 (1H, m), 2.86-2.88 (2H, m), 1.68 (1H, d, J = 5.9 Hz), 1.01 (9H, s), 1.00 (9H, s), 0.99 (9H, s), 0.25 (6H, s), 0.21 (3H, s), 0.21 (3H, s), 0.20 (6H, s); 13C-NMR (100 MHz, CDCl3) 155.6, 155.5, 154.9, 147.0, 146.7, 131.0, 121.0, 119.3, 119.2, 103.3, 99.4, 96.7, 78.1, 66.5, 28.5, 25.93, 25.91, 25.7, 18.44, 18.42, 18.2, −4.10, −4.12, −4.21, −4.24; ESILRMS (m/z) 1292 (13), 1291 (30), 1290 (61), 1289 (85), 1288 (85), 657 (25), 656 (58), 655 ([M+Na]+, 100), 635 (11), 634 (26), 633 ([M+H]+, 51); ESIHRMS calcd. for C33H57O6Si3,633.3463; found 633.3458.
Deprotection of 7- or 5-O-TBS group of compound 12 (14 and 16): To a solution of 5,7,3’,4’-tetra-O-TBS-(+)-catechin (12) (1.00 g, 1.34 mmol) in MeOH (40 mL), CH2Cl2 (10 mL), K2CO2 (0.80 g, 5.78 mmol) was added slowly at 0°C to RT. After stirring for 2 h, the pale-yellow reaction mixture was quenched with sat. NH4Cl aq. The aqueous solution was extracted with CHCl3 and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica-gel column purification (n-hexane/EtOAc, 20:1 to 5:1) afforded 0.15 g of 14 (0.23 mmol, 17%) and 0.16 g of 16 (0.25 mmol, 19%) as white powder, respectively. And 0.042 g (0.056 mmol, 4.2%) of 12 was recovered. Data for 14: [α]25 D +13.6 (c 0.22, CHCl3); 1H-NMR (400 MHz, CDCl3) 6.87 (1H, br s), 6.85 (2H, br s), 6.07 (1H, d, J = 2.4 Hz), 5.99 (1H, d, J = 2.4 Hz), 4.66 (1H, d, J = 7.6 Hz), 3.99 (1H, ddd, J = 5.6, 7.6, 8.4 Hz), 2.94 (1H, dd, J = 5.6, 16.4 Hz), 2.58 (1H, dd, J = 8.4, 16.4 Hz), 1.69 (1H, d, J = 4.0 Hz), 0.99 (9H, s), 0.99 (9H, s), 0.96 (9H, s), 0.24 (3H, s), 0.23 (3H, s), 0.20 (3H, s), 0.20 (3H, s), 0.16 (3H, s), 0.16 (3H, s); 13C-NMR (100 MHz, CDCl3) 155.6, 155.2, 155.0, 147.2, 147.1, 130.8, 121.2, 119.9, 119.6, 104.4, 99.2, 96.4, 81.1, 68.4, 27.6, 25.9 (2), 25.7, 18.42, 18.40, 18.2, –4.09, –4.10, –4.14, –4.15, –4.21, –4.23; ESILRMS (m/z) 1291 (12), 1290 (22), 1289 (31), 1288 (30), 658 (7.3), 657 (24), 656 (56), 655 ([M+Na]+, 100), 635 (11), 634 (26), 633 ([M+H]+, 53); ESIHRMS calcd. for C33H57O6Si3, 633.3463; found 633.3458.
(2R,3S)-3’,4’-5-Tri-O-TBS-(+)-catechin-7-O-(tri-O-benzyl) gallate (18): To a solution of 14 (80.4 mg, 0.13 mmol) and 3,4,5-tribenzyloxybenzoic acid (99.1 mg, 0.22 mmol) was added EDC (31.8 mg, 0.17 mmol) and DMAP (1.0 mg, 8.19 μmol) in CH2Cl2 (2 mL) at 0°C. After stirring for 24 h, the reaction mixture was quenched with water. The aqueous solution was extracted with CHCl3 and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica gel column purification (n-hexane/ EtOAc; 10:1 to 2:1) afforded 76.0 mg of 18 (0.075 mmol, 54%) as an amorphous solid. [α]26 D–4.5 (c 0.22, CHCl3); 1H-NMR (400 MHz, CDCl3) 7.52-7.28 (15H, m), 7.51 (2H, s), 6.90 (1H, br s), 6.87 (2H, br s), 6.42 (1H, d, J = 2.2 Hz), 6.28 (1H, d, J = 2.2 Hz), 5.17 (4H, s), 5.15 (2H, s), 4.66 (1H, d, J = 7.7 Hz), 4.01 (1H, ddd, J = 2.1, 7.7, 8.6 Hz), 3.07 (1H, dd, J = 2.1, 16.4 Hz), 2.66 (1H, dd, J = 8.6, 16.4 Hz), 1.71 (1H, d, J = 3.8 Hz), 1.01 (9H, s), 1.00 (9H, s), 0.98 (9H, s), 0.26 (3H, s), 0.26 (3H, s), 0.21 (3H, s) 0.21 (3H, s), 0.19 (3H, s), 0.18 (3H, s); 13C-NMR (100 MHz, CDCl3) 165.5, 155.5, 154.7, 152.6, 150.0, 147.4, 147.2, 142.9, 137.4, 136.6, 130.5, 128.6, 128.5, 128.2, 128.1, 128.0, 127.6, 124.5, 121.2, 120.1, 119.7, 109.9, 109.7, 104.9, 103.1, 81.3, 75.2, 71.3, 68.2, 28.1, 25.9, 25.9, 25.7, 18.4, 18.4, 18.2, –4.05, –4.07, –4.09, –4.12, –4.18, –4.21; ESILRMS (m/z) 1081 (19), 1080 (46), 1079 (81), 1078 ([M+Na]+, 100), 1058 (11), 1057 (20), 1056 ([M+H]+, 25), 656 (10), 655 (18); ESIHRMS calcd. for C61H79O10Si3, 1055.4981; found 1055.4976.
(2R,3R)-(–)-epicatechin-7-O-(tri-O-benzyl)gallate (19): To a solution of 13 (41.9 mg, 66.2 μmol) and 3,4,5-tribenzyloxybenzoic acid (47.6 mg, 0.10 mmol), EDC (24.8 mg, 0.13 mmol) and DMAP (1.00 mg, 8.19 μmol) were added in CH2Cl2 (2 mL) at 0°C to RT. After stirring for 24 h, the reaction mixture was quenched with water. The aqueous solution was extracted with CHCl3 and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica gel column purification (n-hexane/EtOAc; 10:1 to 2:1) afforded crude 17 as an amorphous solid. A solution of crude in THF (3 mL) was added dropwise to triethylamine trihydrofluoride (TEA, 3HF) (50 μL, 0.30 mmol) at 0°C to RT. Concentration and silica-gel column (CHCl3/MeOH; 10:1 to 1:2) afforded 26.7 mg of 19 (37.5 μmol, 57%) as an amorphous solid. [α]24 D−36.3 (c 1.1, MeOH); 1H-NMR (400 MHz, CD3OD) 7.52 (2H, s), 7.49-7.22 (15H, m), 6.99 (1H, d, J = 1.9 Hz), 6.81 (1H, dd, J = 1.9, 8.2 Hz), 6.76 (1H, d, J = 8.2 Hz), 6.27 (1H, d, J = 2.2 Hz), 6.24 (1H, d, J = 2.2 Hz), 5.18 (4H, s), 5.10 (2H, s), 4.90 (1H, br s), 4.23 (1H, dd, J = 2.4, 4.8 Hz), 2.96 (1H, dd, J = 4.8, 17.6 Hz), 2.85 (1H, dd, J = 2.4, 17.6 Hz); 13C-NMR (100 MHz, CD3OD) 166.2, 158.1, 157.3, 154.0, 151.5, 146.0, 145.9, 143.8, 138.7, 138.1, 132.0, 129.7, 129.6, 129.2, 129.1, 128.9, 126.0, 119.4, 115.9, 115.3, 110.3, 106.6, 102.4, 101.9, 80.1, 76.2, 72.2, 67.1, 29.6; ESILRMS (m/z) 1450 (22), 1449 (45), 1148 (47), 737 (12), 736 (46), 735 ([M+Na]+, 100), 580 (15), 579 (45), 414 (20), 413 (75); ESIHRMS calcd. for C43H36O10Na, 735.2206; found 735.2250.
(2R,3S)-(+)-catechin-7-O-(tri-O-benzyl)gallate (20): A solution of 18 (61.0 mg, 57.8 μmol) in THF (2 mL) was added dropwise to triethylamine trihydrofluoride (60 μL, 0.36 mmol) at 0°C to RT. Concentration and silica-gel column (CHCl3/MeOH; 10:1 to 1:2) afforded 36.4 mg of 20 (51.1 μmol, 88%) as an amorphous solid. [α]24 D –11 (c 0.35, MeOH); 1H-NMR (400 MHz, CD3OD3) 7.51 (2H, s), 7.48-7.22 (15H, m), 6.84 (1H, d, J = 1.9 Hz), 6.74 (1H, d, J = 8.3 Hz), 6.72 (1H, dd, J = 1.9, 8.3 Hz), 6.23 (1H, d, J = 2.2 Hz), 6.22 (1H, d, J = 2.2 Hz), 5.17 (4H, s), 5.10 (2H, s), 4.65 (1H, d, J = 7.3 Hz), 4.05 (1H, ddd, J = 5.4, 7.3, 8.0 Hz), 2.92 (1H, dd, J = 5.4, 16.6 Hz), 2.61 (1H, dd, J = 8.0, 16.6 Hz); 13C-NMR (100 MHz, CD3OD) 166.2, 157.7, 156.9, 154.0, 151.6, 146.4, 146.3, 138.7, 138.2, 131.9, 129.8, 129.6, 129.2, 129.14, 129.13, 128.9, 126.0, 120.0, 116.1, 115.2, 110.4, 107.3, 102.0, 101.7, 83.0, 76.2, 72.3, 68.3, 28.6; ESILRMS (m/z) 1450 (28), 1449 (60), 1448 (64), 737 (11), 736 (45), 735 ([M+Na]+, 100), 714 (8), 713 ([M+H]+, 16), 580 (20), 579 (61), 451 (8), 450 (31); ESIHRMS calcd. for C43H37O10, 713.2387; found 713.2406.
(2R,3R)-(–)-epicatechin-7-O-gallate (6): A solution of 19 (31.4 mg, 44.1 μmol) in THF/MeOH/H2O (20:1:1, 8.8 mL) was hydrogenated over 20% Pd(OH)2/C (1 mg) for 12 h at RT. Filtration and concentration afforded a pale brown solid, which was purified using HPLC purification to give 11.5 mg of pure 6 (26.0 μmol, 59%) as a pale brown powder. [α]24 D –63.9 (c 0.31, MeOH); 1H-NMR (400 MHz, CD3OD) 7.15 (2H, s), 6.99 (1H, d, J = 1.6 Hz), 6.80 (1H, dd, J = 1.6, 8.3 Hz), 6.75 (1H, d, J = 8.3 Hz), 6.23 (1H, d, J = 2.0 Hz), 6.21 (1H, d, J = 2.0 Hz), 5.00-4.80 (1H, m), 4.23 (1H, br s), 2.95 (1H, dd, J = 4.4, 17.4 Hz), 2.84 (1H, dd, J = 2.4, 17.4 Hz); 13C-NMR (100 MHz, CD3OD) 167.1, 158.0, 157.3, 151.7, 146.7, 146.0, 145.9, 140.5, 132.0, 120.8, 119.4, 115.9, 115.3, 110.5, 106.3, 102.4, 101.9, 80.1, 67.2, 29.6; ESILRMS (m/z) 466 (19), 465 ([M+Na]+, 76), 451 (26), 450 (10), 426 (11), 425 (42); ESIHRMS calcd. for C22H18O10Na, 465.0798; found 465.0792.
(2R,3S)-(+)-catechin-7-O-gallate (8): A solution of 20 (36.6 mg, 51.4 μmol) in THF/MeOH/H2O (20:1:1, 8.8 mL) was hydrogenated over 20% Pd(OH)2/C (1 mg) for 12 h at RT. Filtration and concentration afforded a pale brown solid, which was purified using HPLC purification to give 18.9 mg of pure 8 (42.7 μmol, 83%) as a pale brown powder. [α]19 D +27.9 (c 0.36, Me2CO) {lit.[4] [α]27 D +38.9 (c 0.81; Me2CO)}; 1H-NMR (400 MHz, CD3OD 3) 7.26 (2H, s), 6.95 (1H, d, J = 1.9 Hz), 6.87 (1H, d, J = 8.3 Hz), 6.82 (1H, dd, J = 1.9, 8.3 Hz), 6.31(1H, d, J = 2.2 Hz), 6.29 (1H, d, J = 2.2 Hz), 476 (1H, d, J = 7.3 Hz), 4.15 (1H, ddd, J = 5.4, 7.3, 7.9 Hz), 3.02 (1H, dd, J = 5.4, 17.0 Hz), 2.71 (1H, dd, J = 7.9, 17.0 Hz); 13C-NMR (100 MHz, CD3OD) 167.0, 157.6, 156.8, 151.8, 146.7, 146.34, 146.29, 140.5, 131.9, 120.6, 119.9, 116.1, 115.1, 110.5, 107.0, 102.0, 101.8, 83.0, 68.4, 28.5; ESILRMS (m/z) 466 (24), 465 ([M+Na]+, 100), 451 (25), 450 (95), 426 (10), 425 (40); ESIHRMS calcd. for C22H18O10Na, 465.0798; found 465.0836.
(2R,3R)-(–)-epicatechin-3,7-di-O-(tri-O-benzyl)gallate (23): To a solution of 13 (80.5 mg, 0.13 mmol) and 3,4,5-tribenzyloxybenzoic acid (0.16 g, 0.34 mmol), DCC (70.6 g, 0.34 mmol) and DMAP (1.00 mg, 8.19 μmol) were added in CH2Cl2 (5 mL) at 0°C to RT. After stirring for 3 days, the reaction mixture was quenched with water. The aqueous solution was extracted with CHCl3 and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica gel column purification (n-hexane/EtOAc; 15:1 to 2:1) afforded crude 21 as an amorphous solid. A solution of crude in THF (2 mL) was added dropwise to triethylamine trihydrofluoride (76 μL, 0.46 mmol) at 0°C. Concentration and silica-gel column (n-hexane/EtOAc; 5:1 to 1:5) afforded 20.8 mg of 23 (18.3 μmol, 14%) as an amorphous solid. [α]26 D –74.4 (c 1.7, CHCl3); 1H-NMR (400 MHz, CDCl3) 7.50 (2H, s), 7.44-7.27 (30H, m), 7.21 (2H, s), 6.77-6.75 (3H, m), 6.55 (1H, d, J = 2.0 Hz), 6.31 (1H, d, J = 2.0 Hz), 5.56 (1H, br s), 5.15-5.05 (14H, m), 3.10-3.04 (2H, m); 13C-NMR (100 MHz, CDCl3) 165.4, 165.2, 155.7, 155.1, 152.7, 152.3, 150.2, 143.7, 143.2, 143.1, 142.3, 137.3, 137.2, 136.8, 136.4, 130.1, 128.8, 128.56, 128.55, 128.51, 128.21, 128.20, 128.1, 128.03, 127.97, 127.94, 127.58, 127.5, 124.6, 124.1, 119.1, 115.1, 113.6, 109.7, 109.0, 104.5, 102.5, 101.88, 101.87, 75.2, 75.1, 71.3, 70.9, 68.52, 68.48, 25.6; ESILRMS (m/z) 1159 (20), 1158 (46), 1157 ([M+Na]+, 60), 805 (29), 804 (55), 652 (18), 651 (79), 608 (20), 607 (100); ESIHRMS calcd. for C71H58O14Na, 1157.3724; found 1157.3757.
(2R,3S)-3’,4’-5-Tri-O-TBS-(+)-catechin-3,7-di-O-(tri-Obenzyl) gallate (22): To a solution of 14 (52.1 mg, 82.3 μmol) and 3,4,5-tribenzyloxybenzoic acid (0.10 g, 0.22 mmol), DCC (56.9 mg, 0.28 mmol) and DMAP (1.0 mg, 8.19 μmol) were added in CH2Cl2 (2 mL) at 0°C. After stirring for 24 h, the reaction mixture was quenched with water. The aqueous solution was extracted with CHCl3 and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica gel column purification (n-hexane/ EtOAc; 10:1 to 2:1) afforded 107.9 mg of 22 (73.0 μmol, 89%) as an amorphous solid. [α]25 D +20.8 (c 8.6, CHCl3); 1H-NMR (400 MHz, CDCl3) 7.50 (2H, s), 7.45-7.27 (30H, m), 7.24 (2H s), 6.86 (1H, br s), 6.77 (2H, br s), 6.52 (1H, d, J = 2.2 Hz), 6.33 (1H, d, J = 2.2 Hz), 5.60- 5.40 (1H, m) 5.20 (1H, d J = 5.5 Hz), 5.15 (4H, s), 5.14 (2H, s), 5.08 (2H, s), 5.07 (4H, s), 2.90-2.88 (2H, m), 0.96 (9H, s), 0.94 (9H, s), 0.91 (9H, s), 0.22 (3H, s), 0.21 (3H, s), 0.15 (3H, s), 0.14 (3H, s), 0.10 (3H, s), 0.08 (3H, s); 13C-NMR (100 MHz, CDCl3) 165.2, 164.3, 155.1, 154.6, 152.6, 152.4, 150.3, 147.2, 146.8, 142.9, 142.5, 137.43, 137.37, 136.7, 136.6, 131.0, 128.57, 128.56, 128.54, 128.52, 128.23, 128.20, 128.07, 128.01, 127.99, 127.95, 127.61, 127.59, 125.0, 124.9, 121.1, 119.4, 118.7, 109.7, 109.1, 108.8, 105.1, 103.1, 77.9, 75.2, 75.1, 71.3, 71.1, 25.91, 25.88, 25.7, 18.45, 18.42, 18.23, –4.09, –4.16, –4.20, –4.22, –4.25; ESILRMS (m/z) 1503 (22), 1502 (44), 1501 (66), 1500 ([M+Na]+, 57), 437 (77), 413 (100); ESIHRMS calcd. for C89H100NaO14Si3 1499.6319; found 1499.6313.
(2R,3S)-(+)-catechin-3,7-di-O-(tri-O-benzyl)gallate (24): A solution of 22 (77.5 mg, 52.4 μmol) in THF (2 mL) was added dropwise to triethylamine trihydrofluoride (50 mL, 0.30 mmol) at 0°C. Concentration and a short silica-gel column (n-hexane/EtOAc; 5:1 to 1:5) afforded 31.5 mg of 24 (27.7 μmol, 53%) as an amorphous solid. [α]27 D +8.7 (c 6.8, CHCl3); 1H-NMR (400 MHz, CDCl3) 7.50 (2H, s), 7.44-7.19 (32H, m), 6.87 (1H, d, J = 1.8 Hz), 0.86 (1H, d, J = 8.3 Hz), 6.79 (1H, dd, J= 1.8, 8.3 Hz), 6.46 (1H, d, J = 2.0 Hz), 6.25 (1H, d, J = 2.4 Hz), 5.70 (1H, br s), 5.63 (1H, br s), 5.54 (1H, br s), 5.44 (1H, ddd, J = 2.4, 5.6, 5.6 Hz), 5.20 (1H, d, J = 5.6 Hz), 5.15-5.06 (12H, m), 2.76 (1H, dd, J = 5.6, 17.2 Hz), 2.59 (1H, dd, J = 2.4, 17.2 Hz); 13C-NMR (100 MHz, CDCl3) 166.4, 166.1, 155.4, 154.6, 152.7, 152.4, 149.7, 144.2, 143.6, 143.3, 142.7, 137.2, 137.1, 136.5, 136.3, 130.4, 128.54, 128.52, 128.50, 128.20, 128.18, 128.08, 128.04, 128.00, 127.97, 127.6, 127.53, 127.52, 124.6, 123.7, 118.7, 116.3, 112.6, 109.7, 109.2, 105.1, 102.0, 101.6, 77.2, 75.2, 75.1, 71.2, 71.1, 69.7, 29.7; ESILRMS (m/z) 1159 (33), 1158 (78), 1157 ([M+Na]+, 100), 805 (7), 804 (14), 580 (22), 579 (62); ESIHRMS calcd. for C71H58O14Na, 1157.3724; found 1157.3763.
(2R,3R)-(–)-epicatechin-3,7-di-O-gallate (7): A solution of 23 (17.5 mg, 15.4 μmol) in THF/MeOH/H2O (20:1:1, 8.8 mL) was hydrogenated over 20% Pd(OH)2/C (1.0 mg) for 12 h at RT. Filtration and concentration afforded a pale brown solid, which was purified using HPLC purification to give 4.9 mg of pure 7 (8.24 μmol, 54%) as a pale brown powder. [α]21 D–180 (c 0.50, MeOH); 1H-NMR (400 MHz, CD3OD) 7.17 (2H, s), 6.95 (2H, s), 6.94 (1H, d, J = 2.0 Hz), 6.83 (1H, dd, J = 2.0, 8.4 Hz), 6.69 (1H, d, J = 8.4 Hz), 6.30 (1H, d, J = 2.0 Hz), 6.24 (1H, d, J = 2.0 Hz), 5.57 (1H, br s), 5.11 (1H, br s), 3.09 (1H, dd, J = 3.4, 16.4 Hz), 2.95 (1H, d, J = 16.4 Hz); 13C-NMR (100 MHz, CD3OD) 167.5, 167.1, 157.9, 157.2, 151.8, 146.7, 146.3, 146.1, 146.0, 140.5, 139.8, 131.1, 121.4, 120.7, 119.4, 116.0, 115.1, 110.6, 110.2, 105.6, 102.3, 102.1, 78.9, 69.6, 27.2; ESILRMS (m/z) 619 (8), 618 (31), 617 ([M+Na]+, 100), 580 (17), 579 (49), 451 (9), 450 (34); ESIHRMS calcd. for C29H22O14Na, 617.0907; found 617.0905.
(2R,3S)-(+)-catechin-3,7-di-O-gallate (9): A solution of 24 (29.0 mg, 25.5 μmol) in THF/MeOH/H2O (20:1:1, 8.8 mL) was hydrogenated over 20% Pd(OH)2/C (1.0 mg) for 12 h at RT. Filtration and concentration afforded a pale brown solid, which was purified using HPLC purification to give 13.9 mg of pure 9 (23.4 μmol, 92%) as a pale brown powder. [α]21 D +23 (c 0.43, MeOH); 1H-NMR (400 MHz, CD3OD) 7.17 (2H, s), 6.97 (2H, s), 6.84 (1H, s), 6.728 (1H, s), 6.726 (1H, s), 6.29 (1H, d, J = 2.2 Hz), 6.25 (1H, d, J = 2.2 Hz), 5.44 (1H, ddd, J = 5.3, 5.4, 5.5 Hz), 5.17 (1H, d, J = 5.5 Hz), 2.85 (1H, dd, J= 5.3, 17.2 Hz), 2.81 (1H, dd, J = 5.4, 17.2 Hz); 13C-NMR (100 MHz, CD3OD) 167.5, 167.1, 157.7, 156.4, 152.0, 146.7, 146.4 (2), 146.3, 140.6, 139.9, 131.2, 121.3, 120.6, 119.0, 116.3, 114.2, 110.5, 110.1, 105.8, 102.1, 102.0, 79.4, 70.6, 24.2; ESILRMS (m/z) 617 (19), 617 ([M+Na]+, 62), 414 (26), 413 (100); ESIHRMS calcd. for C29H22O14Na, 617.0907; found 617.0914.
(2R,3R)-3’,4’,5,5’,7-penta-O-TBDMS-(−)-epigallocatechin-3- O-(tri-O-TBDMS)gallate (25): To a solution of EGCG (1) (5.07 g, 1.11 mmol) in THF-DMF (25 ml, 4:1), TBDMSCl (16.7 g, 0.11 mol) was added at 0°C. After stirring for 2 days, the reaction mixture was quenched with water. The aqueous solution was extracted with EtOAc and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica gel column purification (n-hexane/EtOAc; 150:1 to 110:1) afforded 7.73 g of 25 (0.56 mmol, 51%) as a white powder. [α]21 D −57.5 (c 1.6, CHCl3); 1H-NMR (400 MHz, CDCl3) 7.05 (2H, s), 6.58 (2H, s), 6.16 (1H, d, J = 2.2 Hz), 5.93 (1H, d, J = 2.2 Hz), 5.58 (1H, br s), 5.01 (1H, br s), 2.93 (2H, br s), 0.98 (9H, s), 0.97 (9H, s), 0.95 (9H, s), 0.95 (9H, s), 0.87 (18H, s), 0.86 (18H, s), 0.202 (3H, s), 0.199 (3H, s), 0.18 (3H, s), 0.13 (6H, s), 0.12 (6H, s), 0.11 (6H, s), 0.10 (3H, s), 0.10 (6H, s), 0.08 (6H, s), 0.06 (3H, s); 13C-NMR (100 MHz, CDCl3) 165.0, 155.6, 154.9, 154.7, 148.4, 148.2, 142.9, 137.8, 129.9, 121.8, 115.4, 112.4, 103.8, 103.7, 101.6, 76.9, 67.9, 26.8, 26.2, 26.14 (2), 26.07, 25.7, 25.6, 18.74, 18.67, 17.41, 17.39, 18.2, 18.1, −3.68, −3.69, −3.71, −3.72, −3.74, −3.83, −3.92, −3.96, −4.02, −4.17, −4.35, −4.40, −4.41; ESILRMS (m/z) 1397 (46), 1396 (80), 1395 ([M+Na]+, 100), 1394 (85), 1374 (54), 1373 (67), 1372 ([M+H]+, 57); ESIHRMS calcd. for C70H131O11Si8, 1371.78400; found. 1371.7889.
(2R,3R)-3’,4’,5’,7-tetra-O-TBDMS-(−)-epigallocatechin-3-O- (tri-O-TBDMS)gallate (26): A solution of 25 (2.19 g, 1.6 mmol) in CH2Cl2 (20 mL) was added dropwise to TFA (0.45 mL, 5.88 mmol) at 0°C to RT. After stirring for 2 hours, the pale-yellow reaction mixture was quenched with sat. NaHCO3. The aq. solution was extracted with CHCl3 and the organic phase was washed with water and brine and dried (MgSO4). Filtration, concentration and silica gel column purification (n-hexane/EtOAc, 30:1 to 5:1) afforded 1.10 g of 26 (0.87 mmol, 54%) as an amorphous solid. [α]24 D −73.2 (c 1.2, CHCl3); 1H-NMR (400 MHz, CDCl3) 7.04 (2H, s), 6.60 (2H, s), 6.13 (1H, d, J = 2.4 Hz), 5.93 (1H, d, J = 2.4 Hz), 5.64 (1H, d, J = 2.8, 4.0 Hz), 5.01 (1H, br s), 4.67 (1H, s), 3.00 (1H, dd, J = 4.0, 16.1 Hz), 2.95 (1H, dd, J = 2.8, 16.1 Hz), 0.98 (9H, s), 0.97 (9H, s), 0.95 (9H, s), 0.92 (18H, s), 0.86 (18H, s), 0.21 (3H, s), 0.21 (3H, s), 0.14 (6H, s), 0.12 (6H, s), 0.11 (6H, s), 0.10 (9H, s), 0.09 (3H, s), 0.09 (3H, s), 0.07 (3H, s); 13C-NMR (100 MHz, CDCl3) 165.1, 155.8, 155.3, 154.6, 148.4, 148.2, 142.9, 137.9, 129.7, 121.7, 115.3, 112.4, 101.0, 100.2, 99.5, 77.2, 67.4, 26.21, 26.16, 26.15, 26.12, 26.09, 25.6, 18.74 (x2), 18.68, 18.4, 18.1, −3.67, −3.70, −3.73, −3.76, −3.79, −3.90, −3.91, −4.0, −4.3, −4.5; ESILRMS (m/z) 1281 (1), 1280 ([M+Na]+, 1), 806 (14), 805 (54), 804 (100), 414 (22), 413 (81); ESIHRMS calcd. for C64H116O11Si7Na, 1279.6800; found 1279.6795.
(2R,3R)-3’,4’,5’,7-tetra-O-TBDMS-5-O-(tri-O-benzyl)galloyl- (−)-epigallocatechin-3-O- (tri-O-TBDMS)gallate (27): To a solution of 26 (0.10 g, 79.5 μmol) and 3,4,5-tribenzyloxybenzoic acid (48.8 mg, 0.11 mmol), DCC (21.4 mg, 0.10 mmol) and DMAP (1.0 mg, 8.19 μmol) were added in CH2Cl2 (10 ml) at 0°C. After stirring for 2 days, the reaction mixture was quenched with water. The aqueous solution was extracted with CHCl3 and the organic phase was washed with water and brine and then dried (MgSO4). Filtration, concentration and silica gel column purification (n-hexane/EtOAc; 30:1 to 5:1) afforded 0.11 g of 27 (65.4 μmol, 82%) as an amorphous solid. [α]23 D 0 (c 1.6, CHCl3); 1H-NMR (400 MHz, CDCl3) 7.52-7.23 (17H, m), 7.06 (2H, s), 6.60 (2H, s), 6.45 (d, J = 2.4 Hz), 6.25 (1H, d, J = 2.4 Hz), 5.62 (1H, br s), 5.15 (6H, s), 5.01 (1H, br s), 2.95 (1H, dd, J = 4.0, 16.8Hz), 2.80 (1H, d, J = 16.8 Hz), 0.99 (9H, s), 0.98 (9H, s), 0.96 (9H, s), 0.89 (18H, s), 0.88 (18H, s), 0.24 (3H, s), 0.23 (3H, s), 0.14 (6H, s), 0.13 (6H, s), 0.12 (6H, s), 0.11 (9H, s), 0.10 (3H, s), 0.09 (3H, s), 0.07 (3H, s); 13C-NMR (100 MHz, CDCl3) 164.8, 163.8, 155.7, 155.1, 152.6, 150.3, 148.5, 148.2, 143.0, 142.9, 138.0, 137.3, 136.5, 129.4, 128.6, 128.5, 128.2, 128.03, 128.00, 127.5, 124.0, 121.6, 115.2, 112.3, 109.7, 107.1, 106.0, 105.5, 77.2, 75.2, 71.3, 66.9, 27.0, 26.20, 26.16, 26.14, 26.06, 25.6, 18.73 (x2), 18.68, 18.4, 18.1, −3.70, −3.72, −3.73, −3.74, −3.77, −3.87, −3.93, −4.0, −4.4, −4.5; ESILRMS (m/z) 1703 (5), 1702([M+Na]+, 6), 804 (39), 803 (73), 413 (100); ESIHRMS calcd. for C92H140O16Si7, 1696.8526; found 1696.8521.
(2R,3R)-5-O-galloyl-(−)-epigallocatechin-3-O-gallate (10): A solution of 27 (47.1 mg, 28.0 μmol) in THF (5 mL) was added dropwise to triethylamine trihydrofluoride (93 μL, 0.56 mmol) at 0°C. Concentration and a short silica-gel column (CHCl3/MeOH; 6:1 to 3:1) afforded clude product. And then a solution of clude product in THF/ MeOH/H2O (20:1:1, 4.4 mL) was hydrogenated over 20% Pd(OH)2/C (1.0 mg) for 12 h at RT. Filtration and concentration afforded a pale brown solid, which was purified using HPLC purification to give 4.1 mg of pure 10 (6.7 μmol, 24%) as a pale brown powder. 1H-NMR (400 MHz, CD3OD): 7.16 (2H, s), 6.92 (2H, s), 6.50 (2H, s), 6.36 (1H, d, J = 2.4 Hz), 6.25 (1H, d, J = 2.4 Hz), 5.49 (1H, br s), 5.04 (1H, br s), 2.99 (1H, dd, J = 4.0, 17.6 Hz), 2.76 (1H, dd, J = 2.4, 17.6 Hz); 13C-NMR (100 MHz, CD3OD) 167.6, 166.5, 158.2, 157.3, 152.0, 146.8 (x2), 146.3, 140.7, 140.0, 133.9, 130.3, 121.3, 120.2, 110.6, 110.3, 106.9, 105.0, 103.8, 102.0, 78.9, 69.3, 27.1; ESILRMS (m/z) 1245 (15), 1244 (36), 1243 (58), 634 (32), 633 (100). [α]27 D −7.4 (c 0.27, MeOH) (lit.[5] [α] D −13); 1H-NMR (400 MHz, CD3OD): 7.16 (2H, s), 6.92 (2H, s), 6.50 (2H, s), 6.36 (1H, d, J = 2.4 Hz), 6.25 (1H, d, J = 2.4 Hz), 5.49 (1H, br s), 5.04 (1H, br s), 2.99 (1H, dd, J = 4.0, 17.6 Hz), 2.76 (1H, dd, J = 2.4, 17.6 Hz); 13C-NMR (100 MHz, CD3OD) 167.6, 166.5, 158.2, 157.3, 152.0, 146.8 (x2), 146.3, 140.7, 140.0, 133.9, 130.3, 121.3, 120.2, 110.6, 110.3, 106.9, 105.0, 103.8, 102.0, 78.9, 69.3, 27.1; ESILRMS (m/z) 1245 (15), 1244 (36), 1243 (58), 634 (32), 633 (100); ESIHRMS calcd. for C29H22O15Na, 633.0856; found 633.0828.
Inhibitory Activity of HeLa S3 Cell Proliferation: 104 cells per well with 100 μL of medium in a 37°C incubator equilibrated with a 5% CO2: 95% humidified air atmosphere. D-MEM (Dulbecco’s Modified Eagle’s Medium; Gibco® (Life Technologies, Grand Island, NY, USA) supplemented with 5% fetal calf serum and 1% Pen-Strep; Invitrogen™ (Life Technologies). After 24 h of incubation, 1 μL of synthesized six compounds in 10% DMSO were added (final 100, 50 μM) and incubated for 48 h (The final concentration of DMSO in each well is 0.1% v/v). As negative controls, medium alone well and 10% DMSO added to the well were prepared at the same time. After the medium was removed and the cell was washed with PBS, 90 μL of new medium and 10 μL of the MTT solution (3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyltetrazolium bromide, 5 mg/mL) was added to each well and incubated at 37°C for 2.5 h. After incubation, the reaction medium was removed and 100 μL of DMSO was added to each well and mix thoroughly with the pipette. Following which, viable cells were assessed using a microplate reader (Filter Max F5 multi-mode microplate reader; Molecular Devices) to measure the OD at 570 nm. And the absorbance values converted into the percentage cell proliferation inhibitory activity as follows: [(absorbance of the control − absorbance of the sample)/absorbance of the control] × 100.
DPPH Radical Scavenging Activity: DPPH radical scavenging activity was measured with general procedure [6]. A solution of DPPH radical in EtOH (30 μM, 1.0 mL) was added to 1 μL of the synthesized each compound in DMSO, and incubated at 30°C for 30 min (n = 6). The scavenging activity was estimated with a microplate reader (Filter Max F5 multi-mode microplate reader; Molecular Devices, Downingtown, PA, USA) to measure the OD at 515 nm. Negative controls, the samples that 1 μL of DMSO added to the 1.0 mL of EtOH were prepared at the same time. And the absorbance values converted into the percentage radical scavenging activity as follows: [(absorbance of the control − absorbance of the sample)/absorbance of the control] × 100.
Synthesis
First, 7-galloyl-(–)-epicatechin (6) and 7-galloyl-(+)-catechin (8) were synthesized based on our strategy, deprotection of one TBS moiety, as shown in Scheme 1. Since its first isolation from Sanguisorba officinals [4], there have been several reports regarding the isolation and biological activities of compound 8 as a minor constituent of Paeoniae obovate [7]; a moderate antiproliferation constituent against B16F10 [8], HeLa [8], and MK-1 [8]; a cytotoxic constituent against HepG2 [9] isolated from the seeds of Rhynchosia volubilis; and an antioxidant isolated from Acacia nilotica [10]. In addition, synthesis of 8 from a catechol-protected-(+)-catechin derivative, moderate antiplasmodial activity against FcB1, and cytotoxicity against MRC-5 cells were reported [11].
Deprotection of TBS ether of (–)-epicatechin derivative 11 in basic conditions afforded a mixture of 7-hydroxyl and 5-hydroxyl, compounds 13 and 15, which were separable by silica gel column chromatography with 17% and 19% yields, respectively. (+)-Catechin derivative 12 was similarly converted to 14 and 16 with 26% and 22% yields, respectively. Determination of the deprotected 7-position was confirmed using HMQC and HMBC methodology and compared with 5-deprotected compounds reported previously [3]. TBS-protected- 7-hydroxyl-(–)-epicatechin derivative 13 was condensed using 3,4,5-tribenzyloxybenzoic acid by 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC), then deprotection of all TBS groups by treatment with triethylamine trihydrofluoride afforded 19 with a 57% yield (2 steps). A galloyl moiety was selectively introduced at the 7-position when EDC was used as condensing agent. Hydrogenolysis of the benzyl ether of 19 in the presence of Pd(OH)2/C, a catalyst, gave 7-galloyl-(–)-epicatechin (6) with a 59% yield. TBS-protected-7-hydroxyl-(+)-catechin derivative 14 was also condensed using 3,4,5-tribenzyloxybenzoic acid to give compound 18 with a 54% yield. Deprotection of the TBS ether of 18 using triethylamine trihydrofluoride gave compound 20 with an 88% yield, then hydrogenolysis of benzyl ether gave 7-galloyl-(+)-catechin (8) with an 83% yield (Scheme 1).
Thereafter, the 3,7-digalloyl derivatives, 3,7-digalloyl-(–)- epicatechin (7) and 3,7-digalloyl-(+)-catechin (9) were synthesized as shown in Scheme 2. Compound 9 was isolated from Acacia gerrafdii and the data of methylated 9 were reported [12]. The TBSprotected- 7-hydroxyl series, 13 and 14, were condensed using 3,4,5-tribenzyloxybenzoic acids with dicyclohexylcarbodiimide (DCC) as a condensing agent. Because compound 21, derived from 13, could not be separated from DCC residue, purification was performed after deprotection of the TBS moiety with triethylamine trihydrofluoride to give 23 with a 14% yield (2 steps). Compound 22, derived from 14, was purified to afford pure product with an 89% yield. TBS-deprotection of 22 with triethylamine trihydrofluoride gave 24 with a 53% yield. 23 and 24 were hydrocracked of benzyl ether to give the 3,7-digalloyl derivatives, 3,7-digalloyl-(–)-epicatechin (7) and 3,7-digalloyl-(+)- catechin (9), with 54% and 92% yields, respectively (Scheme 2).
3,5-Digalloylated (–)-epigallocatechin (10), the compound used to evaluate the enhancing effect of the three hydroxyl groups on the B-ring for inhibitory activity against HeLa S3 cell proliferation, was synthesized as shown in Scheme 3. Compound 10, one of the commercially available EGCG derivatives, was reported to be a minor constituent of green tea leaf [5] and Oolong tea [13] with a variety of biological activities.
The eight phenolic hydroxyl groups of (–)-EGCG (1) were all protected by TBS groups to give 25 with a 51% yield. Thereafter, regioselective deprotection of the 5-TBS ether on the A-ring of 25 proceeded smoothly to give the 5-hydoroxyl compound 26 with a 54% yield. 26 was condensed with 3,4,5-tribenzyloxybenzoic acid by DCC to give digalloyl derivative 27. Because 26 was inseparable from the DCC residue, purification was performed after deprotection of the TBS ethers. TBS-deprotection of 27 with triethylamine trihydrofluoride gave 3,5-digalloylated (–)-epigallocatechin (10) with a 24% yield (2 Steps) (Scheme 3).
Inhibitory activity against cervical epithelioid carcinoma cell line, HeLa S3, proliferation
The inhibitory activities of the synthesized flavan-3-ol derivatives against HeLa S3 cell proliferation are shown in Figure 4. As noted previously, compound 8 was reported to moderately inhibit the proliferation of B16F10 [8], HeLa [8], and MK-1 [8] cells and to be a moderately cytotoxic compound to HepG2 [9] and MRC-5 cells [11]. Compounds 6−10 exhibited significantly lower activity compared to 3,5-digalloyl-(–)-epicatechin (4). The IC50 values of 4 and 9 were 26.0 μM and 85.2 μM, respectively. Surprisingly, 3,5-digalloylated (–)-epigallocatechin (10), in which the number of hydroxyl groups on the B-ring was greater than that in 4, did not efficiently inhibit cell proliferation. Compound 10 was identified as a strong inhibitor of lymphoid tyrosine phosphatase [14], but HeLa S3 cell proliferation was not suppressed by 10 in our assay system.
These results demonstrated that the galloyl moiety at the 5-position and the two hydroxyl groups on the B-ring of compound 4 were significant for the inhibition of HeLa S3 cell proliferation. Our recent report described the synthesis and SAR studies of dimeric flavan- 3-ols, which consisted of (–)-epicatechin (2) and (+)-catechin (3) components, and the fine structure of flavan-3-ols contributed to inhibitory activity against HeLa S3 cell proliferation [15]. In the SAR study using monomeric flavan-3-ol, slight structural differences greatly influenced the cell proliferation inhibitory properties.
2,2-Diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity
Polyphenol compounds are strong antioxidants and radical scavengers. In previous studies, we reported the DPPH radical scavenging activity of synthesized flavan-3-ols, oligomeric flavan-3-ols, and their derivatives to investigate the relationship between radical scavenging activity and structure [2,15,16]. We reported previously that the correlation between the number of phenolic hydroxyl groups with DPPH radical scavenging ability of 5-galloyl series of (–)-epicatechin and (+)-catechin and dimeric flavan-3-ol series was low [3,15]. In Figure 5, the DPPH radical scavenging activities of compounds 1, 4, and 6–10 at final concentrations of 1, 1.5, 2, 2.5, 5, and 10 μM are shown. The SC50 values (the concentration of 50% scavenging activity) of these compounds were 2.2, 2.1, 2.8, 1.4, 2.3, 1.3, and 1.4 μM, respectively. The SC50 values of these compounds indicated that the number of phenolic hydroxyl groups was relatively correlated with DPPH radical scavenging activity and suggested that 3,7-digalloyl compounds 7 and 9 were good scavengers of the DPPH radical compared to compounds 4. Furthermore, the three phenolic hydroxyl groups on the B-ring were important for radical scavenging activity based on a comparison between compound 4 and 10. These results suggested that DPPH radical scavenging activity was affected by the galloyl-moiety-introduced position. In addition, we confirmed that the DPPH radical scavenging activity of synthesized compounds poorly correlated with inhibitory activity against HeLa S3 cell proliferation.
We synthesized 7-galloyl- and 3,7-digalloyl-(–)-epicatechin and (+)-catechin series 6−9 using 7-TBS deprotection in basic conditions to evaluate which position was most effective for modification with the galloyl moiety to strength the inhibition of HeLa S3 cell proliferation. The 3,7-digalloyl derivatives inhibited HeLa S3 proliferation to a lesser extent than 3,5-digalloyl-(–)-epicatechin (4). In addition, we synthesized 3,5-digalloyl- (–)-epigallocatechin (10) using 5-TBS deprotection in acidic conditions to evaluate the enhancing effect of the three-hydroxyl groups on the B-ring. Contrary to our expectations, 3,5-digalloyl-(–)-epigallocatechin inhibited HeLa S3 cell proliferation to a lesser extent than 3,5-digalloyl-(–)-epicatechin. The DPPH radical scavenging activity of the synthesized compounds indicated that the galloyl-modified position was affected by the galloyl-moiety-introduced position. In this study, we found that 3,5-digalloyl-(–)-epicatechin had the superlative inhibitory effect among the galloylated flavan-3- ol derivatives we synthesized. These experiments using synthesized galloyl-modified flavan-3-ols, which are usually difficult to obtain from plants at the same time, indicated that the structure of 3,5-digalloyl and the two hydroxyl groups on the B-ring were important for inhibitory activity against HeLa S3 cell proliferation.
This work was supported by BRAIN: Program for Promotion of Basic and Applied Research for Innovations in Bio-oriented Industry (A. Saito), Technology Research Promotion Program for Agriculture, Forestry, Fisheries, and Food Industry (A. Saito), the Tojuro Iijima Foundation for Food and Technology (A. Saito), the Kurata Memorial Hitachi Science and Technology Foundation (A. Saito) and the Science Research Promotion Fund 2016 from the Japan Private School Promotion Foundation (A. Saito).
T.H. and Y.H. synthesized all compounds for biological assays; T.H. and S.K. measured the inhibitory activity against HeLa S3 cell proliferation and the DPPH scavenging activity; M.H., N.N., T.K. and A.S. supervised the study; and T.H. and A.S. wrote the manuscript.
The authors declare no conflict of interest.