HETEROCYCLES

Review | Regular issue | Vol. 87, No. 11, 2013, pp. 2225-2248
Received, 11th September, 2013, Accepted, 4th October, 2013, Published online, 10th October, 2013.
DOI: 10.3987/REV-13-782

■ Recent Syntheses of Proanthocyanidins

Hidefumi Makabe*

Sciences of Functional Foods, Graduate School of Agriculture, Shinshu University, 8304 Minamiminowa-mura, Kamiina-gun, Nagano 399-4598, Japan

Abstract

Recently proanthocyanidins have been paid much attention due to their significant biological activities and health benefitical effects. The author reviewed recent progress of the syntheses of proanthocyanidins including our work within this decade.

INTRODUCTION
Proanthocyanidins are known as condensed or non-hydrolysable tannins whose structures are basically consisted of flavan-3-ols.1 These compounds are widely distributed in the vegetables, fruits and plant-originated beverage such as tea and wines.2 Proanthocyanidins have been reported to exhibit strong free-radical scavenging and antioxidative activities.3 Many significant biological activities such as antitumor,4,5 antiviral,6 anti-inflammatory,7 and the inhibition of DNA polymerase were reported.8 Thus proanthocyanidins are increasingly recognized as possessing health beneficial effects for humans. There are various type of proanthocyaidins in the nature. The large structural diversity of these compounds exhibits various types of the configurational differences such as C-3- and C-4 stereocenter and the regioisomers and stereochemistries of the inter-flavan bonds. Because their identification as well as purification is extremely difficult even using modern methods of isolation technique such as HPLC and UPLC, further investigation of the biological activities, i. e. mechanism of action of bioactivities remains unknown. In these days, in order to obtain pure proanthocyanidins, synthetic studies have been devoted.9-11 As a result, some of total syntheses of procyanidins have been accomplished and their biological activities and their structure-activity relationship studies have been begun to be reported. However, the syntheses of high degree of oligomerised proanthocyanidins are still difficult although the degree of polimerization was reported to be enhanced biological activities.12,13 Here the author wish to introduce the recent synthetic approaches toward procyanidins and prodelphinidins (Figure 1, 2).

SYNTHESIS OF DIMERIC PROCYANIDINS
Synthesis of procyanidin B1 (1)-B4 (4)
Early synthetic studies of procyanidin dimer can be dated back to 1965 reported by Creasy and Swain.14 They condensed 4-hydroxylated catechin with catechin using aqueous 0.4 M HCl to give corresponding dimer in good yield. However, the selectivity of regio- and stereochemistry of newly formed interflavan bond was not determined. They showed that 4-hydroxylated flavan was activated by acid and protonation at the C-4 hydroxy group generates cationic spieces. The synthesis of catechin dimers using non-protected catechins was ideal, however, the difficulty to purify the products because of the sensitivity by air and acids made synthesis problematic. Thus protected catechin derivatives were used to prepare dimeric products using Lewis acids. In 1991, Kawamoto and co-workers reported the condensation of per-O-benzylated catechin derivative 17 as a nucleophile using five times excess amount with electrophile 18 using TiCl4 as a Lewis acid to give the corresponding dimer 19 in high yield with good selectivity (α:β = 5:1).15 It is notable that attack at the C-8 position was occurred exclusively (Scheme 1).

Since this report had been appeared, further synthetic studies have been reported. Ubukata and Nakajima reported excellent regio- and stereoselective condensation using 4-ethoxyethoxy group as an electrophile and TMSOTf as a Lewis acid and they synthesized procyanidin B3 (3) (Scheme 2).16

The neighboring group participation of C-3 acetoxy group seemed to be effective to form 3,4-anti selective interflavan bond (Figure 3).16,17

They also accomplished stereoselective condensation of catechin and/or epicatechin nucleophile and electrophile and total synthesis of procyanidins B1 (1), B2 (2), and B4 (4), respectively (Scheme 3).18

The same group also reported the synthesis of procyanidin B2 and B3 gallates using same strategy (Scheme 4).8,19

Using protected catechin delivatives blocks provided promising dimeric products with excellent regio- and stereoselectivity. However, this reaction requires a large excess amount of the nucleophile (4-5 equiv.) in order to suppress forming the higher oligomers. In view of purification after condensation, the remained nucleophile needs to be removed. To overcome this problem, Makabe and co-workers investigated Lewis acid mediated equimolar condensation.20,21 They chose tetrabenzylated catechin 17, as a nucleophilic unit, prepared by the Kawamoto’s procedure 15 and electrophile unit 43 prepared by the Saito’s method.16 Equimolar condensation of 17 with 43 at room temperature was examined using various Lewis acids including rare earth metal at room temperature (Table 1).

Using TiCl4 and BF3•Et2O gave sluggish results. The next attempt was examined using late transition metals as Lewis acids.　Among them, AgBF4 gave a good selectivity with moderate chemical yield. They further paid attention to rare metal Lewis acids such as Sc and La. While Sc gave poor stereoselectivity, La afforded high selectivity although the chemical yield was only 34%. Finally, replacing La to Yb furnished condensed product in 64% yield with excellent selectivity. Interestingly, Gd(OTf)3 and Lu(OTf)3 did not give any condensed product. Next, they expanded the condensation of the combination of catechin nucleophile 17 and epicatechin nucleophile 26 with catechin electrophile 43 and/or epicatechin electrophile 44 using Yb(OTf)3 as a Lewis acid. In each case, the reaction worked well. In view of the stereoselectivty, however, the epicatechin electrophile 44 gave a little bit poor results compared to catechin electrophile 43. Condensed compounds 45, 46, 22, and 47 were subjected to the hydrolysis of the acetate with K2CO3 in MeOH followed by debenzylidation by Pd(OH)2 to give procyanidin B1 (1)-B4 (4), respectively (Scheme 5).

The author’s group also reported procyanidin B2 and B3 gallates and their anticancer activity using equimolar condensation (Scheme 6).22

Synthesis of procyanidin B6 (5)
Quite recently, Ohmori and Suzuki et al. accomplished first total synthesis of procyanidin B6 (5) which has a rare 4,6-inter-flavan bond. They succeeded regioselective linkage by the halogen-capping strategy followed by removal of the benzyl groups and halogen-caps by one-pot hydrogenolysis (Scheme 7).23

SYNTHESIS OF TRIMERIC PROCYANIDINS
Synthesis of procyanidin C1 (6) and C2 (7)
Saito and Nakajima et al. reported TMSOTf mediated condensation of monomeric catechin electrophile 21 with nucleophilic dimer 19 for synthesis of procyanidin C2 (7). However, this reaction needed a large excess amount of nucleophilic dimer 19 (4.0-4.5 equiv.) to avoid further oligomer formation (Scheme 8).24,25

To overcome this problem, Ohmori and Suzuki et al. developed the orthogonal approach for synthesis of catechin trimer.26 The acetoxy group served as an orthogonal leaving group that was activated hard Lewis acid (BF3·OEt2). Another important chemical feature of this reaction is the Br-capping of the C-8 position to suppress the self-condensation. Thus the amount of nucleophile could be reduced to 1.2 equivalent (Scheme 9).

The same group also synthesized epiafzelechin-epigallogatechin-catechin (EZ-EG-CA) (61) using similar strategy (Scheme 10).27

Makabe and co-workers also reported the synthesis of catechin and epicatechin trimer (procyanidin C1 (6) and C2 (7)) using equimolar condensation.28,29 Equimolar condensation of 19 with 43 was examined using transition metal Lewis acids and Yb(OTf)3 in CH2Cl2. They paid attention to silver Lewis acid because condensation could be performed under neutral conditions. According to Ferreira and co-workers’ report, using AgBF4 as the thiophilic Lewis acid offered advantages to control the level of oligomeration in the procyanidin B1-B4 and C2 synthesis.30 As shown in Table 1, AgBF4 and AgOTf gave 53 in excellent yield, respectively. However, Yb(OTf)3 afforded low yield. Due to the steric hindrance of Yb(OTf)3, it seemed to be difficult for dimeric nucleophile 19 to attack C-4 position of electrophile 43 (Table 2).

Condensed product 53 was successfully converted into procyanidin C2 (7) and its peracetate (63). The 1H NMR spectral data of peracetate 63 was in good agreement with that of the reported value.31 (Scheme 11).

As to the synthesis of procyanidin C1 (6), equimolar condensation of 27 with 44 was examined using transition metal Lewis acids and Yb(OTf)3 in CH2Cl2. Condensation using AgBF4 and AgOTf resulted condensed product 66 in low yield. Next we examined equimolar condensation of 27 with various 4-alkoxy-epicatechin derivatives 44, 64, 65, and 24 using Yb(OTf)3. As shown in Table 3, 4-(2”-ethoxyethoxy) derivative 24 afforded condensed product 66 in 57% yield. Other 4-alkoxy derivatives gave 66 in low yield and both of the nucleophile and electrophile remained. We found that 4-(2”-ethoxyethoxy) group was suitable for Yb(OTf)3 mediated activation at C-4 position of electrophile.

The condensed product 66 was transformed into triol 67 using n-Bu4NOH. Finally deprotection of the benzyl group of 67 subsequent lyophilization afforded procyanidin C1 (6) in good yield.29 The 1H NMR spectral datum of peracetate 68 was in good agreement with that of the reported value (Scheme 12).31

SYNTHESIS OF HIGHER OLIGOMERS
First challenge of the synthesis procyanidin oligomer was reported by Kozikowski and co-workers.32 Thio ether 69 and epicatechin dimer derivative 70 were treated with excess amount of AgBF4 gave multiple oligomerized products. Each compound was separated by HPLC and isolated products were evaluated antitumor activity (Scheme 13).

Saito and Nakajima et al. reported TMSOTf mediated condensation of monomeric epicatechin electrophile 24 with nucleophilic trimer 67 to afford 71. This reaction also needed a large excess amount of nucleophilic trimer 67 (4.0 equiv.) to avoid further oligomer formation. The condensed product 71 was successfully converted to cinnamtannin A2 (8) (Scheme 14).33

Several studies demonstrated non-oligomeric selective self-polymerization. Kondo and co-workers performed non-selective oligomerization of 3,4-diacetoxy-flavan 72 mediated by B(C6F5)3. They obtained the mixture of higher oligomers up to pentadecamer. The stereochemistry of newly formed interflavan bonds was not determined (Scheme 15).34

Recently, an elegant selective synthesis of higher oligomers up to the tetracosamer was accomplished using orthogonal strategy by Ohmori and Suzuki et al.35 The important matter for the orthogonal activation was the choice of leaving groups at C-4 position. They selected 4-ethoxyethoxy group for hard activation and the 2,6-xylylthio (SXy) group for soft activation. The bromo-capped electrophiles could be subjected to equimolar condensation with large nucleophiles (up to dodecamer). First, they prepared catechin hexamers 75 and 77 in a stereoselective manner. To confirm a stereochemistry of newly formed inter-flavan bond, they synthesized hexamer 75 independently by {2 + 4} coupling using structurally defined dimer 78 and tetramer 79 (Scheme 16).

Next, they prepared two dodecamers 82 and 84 using {6 + 6} coupling. The stereochemistry of newly formed interflavan bonds was not determined (Scheme 17).

Finally, the condensation between dodecamers 82 and 84 using I2 and Ag2O gave desired tetracosamer 85 cleanly in 80% yield (Scheme 18).

SYNTHESIS OF PRODELPHINIDINS
As describes above, many examples of the syntheses of procyanidins were reported in this decade including our syntheses, however, synthetic studies on prodelphinidins are quite limited due to the difficulty in obtaining (−)-gallocatechin or (+)-epigallocatechin as synthetic starting materials.36 Until now only an example of total synthesis of prodelphinidin B3 (14) and C2 (16) has been reported by Makabe and co-workers using equimolar coupling between nucleophilic and electrophilic partners in the presence of Lewis acid.5 The gallocatechin-derived building block 88 was constructed as Chan and co-workers reported with slight modification.37 DDQ oxidation in the presence of methanol or ethoxyethanol gave electrophiles 89 and 90, respectively (Scheme 19).

They examined the condition of equimolar condensation of catechin nucleophile 17 with gallocatechin electrophile 89 or 90. As shown in Table 4, 4-(2”-ethoxyethoxy) derivative 90 afforded condensed product in good yield when Yb(OTf)3 was used as Lewis acid. On the other hand, using methoxy derivative 89 gave 91 in very low yield. The choice of leaving the group at the C-4 position was important in this condensation (Table 4).

The condensed product 91 was transformed into diol 92 using n-Bu4NOH, subsequent deprotection of the benzyl ethers of 92 afforded prodelphinidin B3 (14) in good yield. The 1H NMR spectral data of peracetate 93 was in good agreement with that of the reported value (Scheme 20).31

Diol 92 was used as a nucleophile for the synthesis of prodelphinidin C2 (16). They examined equimolar condensation of 92 with electrophile 89 or 90. The authors have found that silver Lewis acids were effective for the construction of catechin trimer derivatives.28,29 Thus they used AgBF4 and AgOTf as Lewis acid, respectively. As shown in Table 5, the 4-methoxy derivative 89 afforded condensed product 94 in good yield when AgOTf was used. The reaction using AgBF4 as Lewis acid and 4-methoxy derivative 89 as an electrophile afforded 94 in poor yield. The combination of the C-4 leaving group and silver Lewis acid was very important (Table 5).

The condensed product 94 was transformed into triol 95 using n-Bu4NOH.32 Finally deprotection of the benzyl ethers of 95 and subsequent lyophilization afforded prodelphinidin C2 (16) in good yield. The 1H NMR spectral data of peracetate 96 was in good agreement with that of the reported value (Scheme 21).31

The authors examined the antitumor activities of synthesized prodelphinidins B3 (14) and C2 (16) against PC-3 prostate cancer cell lines together with procyanidin B3 (3), C1 (6) and C2 (7) which were prepared by same group (Figure 4).

Results were obtained by two independent methods: cell count measurement and MTT assay. Epigallocatechin gallate (EGCG), which is well known as an antitumor agent, was used as a positive control. As shown in Figure 4, EGCG, prodelphinidin B3 (14) and C2 (16) exhibited significant cytotoxic activity with IC50 values below 50 μM. Making a comparison of the potencies of 14 with procyanidin B3 (3), suggested that the cytotoxic effects were clearly associated with the presence of the pyrogallol moiety. The PDB3 (14) and PCB3 (3) have the same carbon skeleton. The only difference is that PDB3 (14) has an additional hydroxy group at the B ring. The authors showed that this hydroxy group greatly affected the cytotoxic effect. As for 16 and procyanidin C1 (6) or C2 (7), the data showed that the pyrogallol moiety was essential for their activity. This tendency was also observed in the MTT assay. These findings might be useful in searching for antitumor compounds among the proanthocyanidins (Figure 5).

CONCLUSION
Proanthocyanidins have been paid attention to the synthetic and biological researchers due to their unique structures and significant biological activities. Synthetic efforts toward proanthocyanins have performed mainly using condensation of C-4 to C-8 flavan bonds. Flavan-3-ols as nucleophiles and flavan-3,4-diols and its derivatives were subjected for the condensation in the presence of a Lewis acid. Recent progress of this reaction made a regio- and stereoselective synthesis of proanthocyanidin oligomers in good yield. However, there is still much room to develop synthetic methodologies especially for the highly polymerized proanthocyanidins. When synthetic methods of this complex molecule will be fully developed, proanthocyanidins will be candidate drugs for treating various diseases.

References

1. D. Ferreira and X.-C. Li, Nat. Prod. Rep., 2000, 17, 193. CrossRef
2. D. Ferreira and X.-C. Li, Nat. Prod. Rep., 2002, 19, 517. CrossRef
3. N. Seeram, M. Aviram, Y. Zhang, S. M. Hennings, L. Feng, M. Dreher, and D. Heber, J. Agric. Food Chem., 2008, 56, 1415. CrossRef
4. S. Mitsuhashi, A. Saito, N. Nakajima, H. Shima, and M. Ubukata, Molecules, 2008, 13, 2998. CrossRef
5. W. Fujii, K. Toda, K. Kawaguchi, S.-i. Kawahara, M. Katoh, Y. Hattori, H. Fujii, and H. Makabe, Tetrahedron, 2013, 69, 3543. CrossRef
6. H. Y. Cheng, T. C. Lin, C. M. Yang, D. E. Shieh, and C. C. Lin, J. Sci. Food Agric., 2005, 85, 10. CrossRef
7. X. Terra, J. Valls, X. Vitrac, J. M. Merrillon, L. Arola, A. Ardevol, C. Blade, J. Fernandez-Larrea, G. Pujadas, J. Salvado, and M. Blay, J. Agric. Food Chem., 2007, 55, 4357. CrossRef
8. A. Saito, M. Emoto, A. Tanaka, Y. Doi, K. Shoji, Y. Mizushina, H. Ikawa, H. Yoshida, N. Matsuura, and N. Nakajima, Tetrahedron, 2004, 60, 12043. CrossRef
9. D. Ferreira and C. M. Coleman, Planta Med., 2011, 77, 1071. CrossRef
10. K-i. Oyama, K. Yoshida, and T. Kondo, Curr. Org. Chem., 2011, 15, 2567. CrossRef
11. K. Ohmori and K. Suzuki, Curr. Org. Chem., 2012, 16, 566. CrossRef
12. A. P. Kozikowski, W. Tückmantel, G. Böttcher, and L. J. Romanczyk, Jr., J. Org. Chem., 2003, 68, 1641. CrossRef
13. P. W. Caton, M. R. Pothecary, D. M. Lees, N. Q. Khan, E. G. Wood, T. Shoji, T. Kanda, G. Rull, and R. Corder, J. Agric. Food Chem., 2010, 58, 4008. CrossRef
14. L. L. Creasy and T. Swain, Nature, 1965, 208, 151. CrossRef
15. H. Kawamoto, F. Nakatubo, and K. Murakami, Mokuzai Gakkaishi, 1991, 37, 488.
16. A. Saito, N. Nakajima, A. Tanaka, and M. Ubukata, Tetrahedron, 2002, 58, 7829. CrossRef
17. A. Saito, N. Nakajima, A. Tanaka, and M. Ubukata, Biosci. Biotechnol. Biochem., 2002, 66, 1764. CrossRef
18. A. Saito, N. Nakajima, N. Matsuura, A. Tanaka, and M. Ubukata, Heterocycles, 2004, 62, 479. CrossRef
19. A. Saito, Y. Mizushina, H. Ikawa, H. Yoshida, Y. Doi, A. Tanaka, and N. Nakajima, Bioorg. Med. Chem., 2005, 13, 2759. CrossRef
20. Y. Mohri, M. Sagehashi, T. Yamada, Y. Hattori, K. Morimura, T. Kamo, M. Hirota, and H. Makabe, Tetrahedron Lett., 2007, 48, 5891. CrossRef
21. Y. Mohri, M. Sagehashi, T. Yamada, Y. Hattori, K. Morimura, Y. Hamauzu, T. Kamo, M. Hirota, and H. Makabe, Heterocycles, 2009, 79, 549. CrossRef
22. M. Suda, M. Katoh, K. Toda, K. Matsumoto, K. Kawaguchi, S.-i. Kawahara, Y. Hattori, H. Fujii, and H. Makabe, Bioorg. Med. Chem. Lett., 2013, 23, 4935. CrossRef
23. G. Watanabe, K. Ohmori, and K. Suzuki, Chem. Commun., 2013, 49, 5210. CrossRef
24. A. Saito, A. Tanaka, M. Ubukata, and N. Nakajima, Synlett, 2004, 1069. CrossRef
25. A. Saito, Y. Doi, A. Tanaka, N. Matsuura, M. Ubukata, and N. Nakajima, Bioorg. Med. Chem., 2004, 12, 4783. CrossRef
26. K. Ohmori, N. Ushimaru, and K. Suzuki, Proc. Natl. Acad. Sci. USA, 2004, 101, 12002. CrossRef
27. T. Yano, K. Ohmori, H. Takahashi, T. Kusumi, and K. Suzuki, Org. Biol. Chem., 2012, 10, 7685. CrossRef
28. Y. Oizumi, Y. Mohri, Y. Hattori, and H. Makabe, Heterocycles, 2011, 83, 739. CrossRef
29. Y. Oizumi, M. Katoh, Y. Hattori, K. Toda, K. Kawaguchi, H. Fujii, and H. Makabe, Heterocycles, 2012, 85, 2241. CrossRef
30. P. J. Steynberg, R. J. J. Nel, H. van Rensberg, B. C. B. Bezuidenhoudt, and D. Ferreira, Tetrahedron, 1998, 54, 8153. CrossRef
31. J. A. Delcour and S. A. R. Vercruysse, J. Inst. Brew., 1986, 92, 244. CrossRef
32. A. P. Kozikowski, W. Tückmantel, G. Böttcher, and L. J. Romanczyk, Jr., J. Org. Chem., 2003, 68, 1641. CrossRef
33. A. Saito, Y. Mizushina, A. Tanaka, and N. Nakajima, Tetrahedron, 2009, 65, 7422. CrossRef
34. K. I. Oyama, M. Kuwano, M. Ito, K. Yoshida, and T. Kondo, Tetrahedron Lett., 2008, 49, 3176. CrossRef
35. K. Ohmori, T. Shono, Y. Hatakoshi, T. Yano, and K. Suzuki, Angew. Chem. Int. Ed., 2011, 50, 4862. CrossRef
36. K. Krohn, I. Ahmed, M. John, M. C. Letzel, and D. Kuck, Eur. J. Org. Chem., 2010, 2544. CrossRef
37. S. B. Wan, Q. P. Dou, and T. H. Chan, Tetrahedron, 2006, 62, 5897. CrossRef