A novel C–C radical–radical coupling reaction promoted by visible light: facile synthesis of 6-substituted N-methyl 5,6-dihydrobenzophenanthridine alkaloids

Abstract

A novel photoredox-mediated direct intermolecular C–H functionalization of N-methyl 5,6-dihydrobenzophenanthridine is developed utilizing the visible light-induced reductive quenching pathway of photocatalyst Ir(ppy)₃. In the proposed coupling mechanism, an α-amino C-radical is generated at the 6-position of N-methyl 5,6-dihydrobenzophenanthridine which is capable of coupling with α-EWG (electron withdrawing group) substituted C-radicals. The utility of this methodology has been demonstrated via rapid access to the analogue of natural 6-substituted N-methyl 5,6-dihydrobenzophenanthridine alkaloids.

---

Naturally occurring quaternary benzophenanthridine alkaloids (QBAs) such as sanguinarine and chelerythrine (Fig. 1, compounds 1) are of great interest for their biological activities, such as anti-cancer,¹ anti-bacterial,² anti-fungal,³ anti-inflammatory⁴ and enzyme inhibitory potencies.⁵ According to previous literature, natural QBAs are rare because of the stable fused aromatic benzene rings and are mostly 2,3,8,9-oxygenated or 2,3,7,8-oxygenated due to the unique biosynthetic pathway of QBAs.⁶ However, 6-substituted N-methyl 5,6-dihydrophenanthridines have been isolated from medicinal plants.^3a,7 The electrophilic C-6 atom of QBAs mainly accounts for the formation of this type of 6-substituted benzophenanthridine. Nucleophilic addition of the C=N double bond leads to the formation of 6-hydroxyl, 6-ethoxyl and acetonyl 5,6-dihydro dihydrobenzophenanthridines (Fig. 1 compounds 2) during the isolation process.^7a,8 However, classical nucleophilic addition reactions cannot explain the formation of 6-hydroxylethyl dihydrosanguinarine (compound 3, Fig. 1).^8,9 We thought that the synthesis of compound 3 would be convenient if methyl 2′-(5,6-dihydrosanguinarine-6-yl)acetate (compound 4, Fig. 1) was available. Interestingly, compounds 3 and 4 are naturally occurring^8,9 and their synthetic method was not reported until now.

Fig. 1 Structures of sanguinarine and its typical derivatives.

Radical-based transformations are fundamental synthetic tools which show good functional group compatibility.¹⁰ Recently, great efforts have been made in photoredox catalysis and provided a series of model radical reactions.¹¹ Tertiary amine represents a model reaction substrate in photoredox reaction. As shown in Fig. 2, tertiary amine 6 can be oxidized by visible light excited photocatalysts to give aminium radical cation 7 through a single electron transfer (SET) process. If a good hydrogen atom acceptor is present in the reaction, it may abstract a hydrogen atom from intermediate 7, converting the aminium radical cation 7 to the iminium ion 8 as an electrophilic reaction intermediate. The aminium radical cation 7 may also be deprotonated by base at the α-position to yield α-amino C-radical 9 which can undergo radical addition or coupling reactions. The applications of iminium ion 8 as a kind of electrophilic intermediate¹² and α-amino C-radical 9 as radical addition reaction intermediate¹³ have received most of the research interests in photoredox catalyzed transformations of tertiary amines according to previously literature. However, the applications of α-amino C-radical 9 in radical–radical coupling reaction were reported in rare cases.¹⁴

Fig. 2 Visible light promoted transformation of tertiary amine.

In this study, we sought to develop a method to synthesize the analogue of compound 4 based on a visible light promoted C–C radical–radical coupling reaction with dihydrosanguinarine 5 (Fig. 3), a tertiary amine, as starting material. Dihydrosanguinarine 5 is known as the biosynthetic precursor of sanguinarine, and can be naturally oxidized to sanguinarine in solvents by oxygen or catalyzed by DBOX, a biosynthetic enzyme in medicinal plants.⁶ From this standpoint, we postulated that dihydrosanguinarine 5 can be oxidized by Ir^IV(ppy)₃⁺, an Ir species generated from the photocatalytic cycle (Fig. 3). In this photocatalytic cycle, visible light excited *Ir^III(ppy)₃ (E^IV/*III_1/2 = −1.73 V vs. SCE)^11a is capable of reducing ethyl bromoacetate 10 to give corresponding α-carbonyl C-radicals 11 and Ir^IV(ppy)₃⁺. Subsequently reductive quenching of the Ir^IV(ppy)₃⁺ (E^IV/III_1/2 = +0.77 V vs. SCE)^11a with dihydrosanguinarine 5 results in neutral Ir^III(ppy)₃ complex and aminium radical cation 12, which is further deprotonated by base to α-amino C-radical 13. C-radical 13 can be stabilized by the π-orbital of benzene and p-orbital of nitrogen atom.^11a Finally, the subsequent radical–radical coupling between radicals 11 and 13 is capable of giving target compound 14a.

Fig. 3 Proposed photoredox C–C radical–radical coupling mechanism.

In the initial experiments to examine this possibility, Ir(ppy)₃ was firstly employed as photocatalyst and Na₂HPO₄ was used as base. A solution of dihydrosanguinarine, ethyl bromoacetate, photocatalyst and Na₂HPO₄ was firstly bubbled with nitrogen for 10 minutes and then irradiated with 25 W household compact fluorescent lamp under nitrogen atmosphere. After 24 h reaction, target compound 14a was obtained in 63% isolated yields (Table 1, entry 1). Replacement of photocatalyst Ir(ppy)₃ with Ru(bpy)₃Cl₂ (E^III/*II_1/2 = −0.81 V vs. SCE)^11a was found not suitable for the radical coupling reaction (Table 1, entry 2) and sanguinarie was detected as major product. When inorganic base Na₂HPO₄ was changed to Na₂CO₃, a significant decrease of target compound was observed (Table 1, entry 3). It was interesting that a coupling product of dihydrosanguinarine with solvent DMF was detected as major byproduct if Na₂CO₃ was used (see ESI†). Moreover, replacement of solvent DMF to MeCN also led to decreased yields of compound 14a (Table 1, entry 4). In this case, recovery of dihydrosanguinarine 5 was observed. To identify the role of visible light, inorganic base Na₂HPO₄ and photocatalysts in the coupling reaction respectively, several control experiments were conducted as shown in Table 1. When the reaction was carried out without the irradiation of visible light, coupling reaction was totally shutdown which led to the recovery of starting material 5 (Table 1, entry 5). The absence of Na₂HPO₄ in the reaction led to significant decreased yield of 14a (Table 1, entry 6). In this case, sanguinarine was detected as the major byproduct (Table 1, entry 6). Moreover, it should be noted that replacement of Na₂HPO₄ with organic base i-Pr₂NEt only resulted in trace of target compounds with the starting material 5 recovered (Table 1, entry 7). According to our hypothesis in Fig. 3, it was the reducible organic base i-Pr₂NEt but not dihydrosanguinarine that would be oxidized by Ir species Ir^IV(ppy)₃⁺ generated from *Ir^III(ppy)₃ firstly.¹⁵ Thus, the formation of aminium radical cation 12 could be blocked leading to obviously decreased yield of 14a. Finally, only trace of sanguinarine could be detected with the starting material 5 recovered without the addition of photocatalyst Ir(ppy)₃ (Table 1, entry 8).

Table 1 Visible light promoted C–C radical–radical coupling to compound 14^a

Entry	Photocatalysts	Bases	Solvents	Yields^b
a Unless otherwise indicated, the reactions were performed with 0.2 mmol of 5 and 0.6 mmol of ethyl bromoacetate, 2 mol% of photocatalysts and 0.6 mmol of base in 2 mL of solvent at ambient temperature for 24 under nitrogen atmosphere with visible light irradiation (25 W household compact fluorescent lamp).b Isolated yields.c Under darkness.
1	Ir(ppy)3	Na2HPO4	DMF	63%
2	Ru(bpy)3Cl2·6H2O	Na2HPO4	DMF	0%
3	Ir(ppy)3	Na2CO3	DMF	27%
4	Ir(ppy)3	Na2HPO4	MeCN	13%
5^c	Ir(ppy)3	Na2HPO4	DMF	0%
6	Ir(ppy)3	None	DMF	8%
7	Ir(ppy)3	i-Pr2NEt	DMF	Trace
8	None	Na2HPO4	DMF	0%

---

Having identified optimal conditions for this photoredox catalyzed C–C radical coupling reaction, we next investigated the scope of bromide coupling partners. As summarized in Fig. 4, ethyl 2-bromopropionate was firstly used instead of ethyl bromoacetate, and the desired compound 14b was obtained in 44% isolated yield with d.r. value of 1 : 1. However, when ethyl bromoacetate was replaced by diethyl bromomalonate, target compound 14c was only isolated in 8% yield because compound 14c was unstable during the silica gel column chromatography purification process (see ESI†). When bromoacetonitrile was used as radical precursor, coupling compound 14d was also obtained in 57% yield. Furthermore, 2-bromoacetophenone, 2-bromo-4′-methoxyacetophenone and 2-bromo-4′-nitroacetophenone were found suitable for the radical–radical coupling reaction, and compounds 14e–14g were obtained in moderate yields (47–61%). Dihydrochelerythrine was also investigated as coupling partner, and corresponding 6-substituted dihydrochelerythrines were obtained in 34–56% yields (14h–14j). Yields of these dihydrochelerythrine derivatives (14h–14j) were slightly lower than corresponding sanguinarine derivatives, which could be tentatively attributed to the higher steric shielding of the methoxy group in dihydrochelerythrine. Finally, it should be noted that sanguinarine or chelerythrine were detected as the major byproducts in the coupling reactions due to the hydrogen atom abstraction side reaction as shown in Fig. 2.

Fig. 4 Scope of bromoalkanes with electron withdrawing group (EWG) at α-position in the radical–radical coupling reaction^a. ^aUnless otherwise indicated, the reactions were performed with 0.2 mmol of 5 and 0.6 mmol of bromides, 2 mol% of Ir(ppy)₃ and 0.6 mmol of Na₂HPO₄ in 2 mL of DMF at ambient temperature for 24 under nitrogen atmosphere with irradiation of 25 W household compact fluorescent lamp. ^bStarting material 5 was used at 0.6 mmol scales.

With the scope of bromoalkanes established in Fig. 4, we next turned to the synthesis compound 4 (Fig. 1). As shown in Fig. 5, compound 14a can be easily reduced to 6-hydroxylethyl derivative 4 in 92% isolated yield by stirring a mixture of compound 14a, 10 equiv. of LiAlH₄ in THF at 0 °C for 6 h.

Fig. 5 Further derivatization of coupling compound 14a.

To further understand the coupling mechanism, a radical trapping experiment was carried out. Under standard reaction conditions, 3.0 equiv. of 2,2,6,6-tetramethylpiperidinooxy free radical (TEMPO) was added to the reaction system (Fig. 6A). After 24 h reaction, HPLC-HRMS analysis of the crude product showed coupling reaction was shutdown. However, the trapping product 15 (Fig. 6A) was not detected. We supposed that compound 15 was able to hydrolyze to give sanguinarine under the reaction conditions. On the other hand, TEMPO could act as a hydrogen atom acceptor of aminium radical cation 12 in Fig. 3 and facilitate the generation of sanguinarine through a hydrogen-atom transfer (HAT) process (see ESI†) as shown in Fig. 2. However, the coupling product 17 (Fig. 6) of TEMPO with α-carbonyl C-radical 11 was detected by GC-MS analysis (see ESI†). Additionally, we found that the use of sanguinarine as coupling partner resulted only in the recovery of starting materials (Fig. 6B). It should be noted that the present coupling reaction showed specificity to natural dihydrobenzophinanthridine alkaloid as coupling partner. As shown in scheme (C) in Fig. 6, when dihydrophenanthridines 18 were used as coupling partner, only lactams 19a and 19b were isolated in 51% and 55% yields under the optimized reaction conditions. Thus it was suggested that larger conjugate system in the dihydrobenzophinanthridine substrates 5 probably facilitated the coupling reaction.

Fig. 6 Further reaction mechanism research.

In summary, we have successfully developed a new method for synthesis of 6-substituted N-methyl 5,6-dihydrobenzophenanthridine alkaloids based on photoredox catalysis with natural 5,6-dihydrobenzophenanthridine alkaloids and commercial available α-bromo carbonyl compounds as starting material. Based on control experiments and previous literature this reaction is supposed to proceed via radical pathway in which visible light trigged a novel C–C radical–radical coupling reaction at room temperature. This synthetic method is simple and mild which is expected to be applied in synthesis of other functionalized alkaloids in the future.

Acknowledgements

This research was financially supported by National Natural Science Foundation of China (No. 31402109).

Notes and references

1. (a) H. Chen, H. Long, X. Cui, J. Zhou, M. Xu and G. Yuan, J. Am. Chem. Soc., 2014, 136, 2583; (b) T. Nakanishi, M. Suzuki, A. Saimoto and T. Kabasawa, J. Nat. Prod., 1999, 62, 864; (c) Y.-H. Zhang, A. Bhunia, K. F. Wan, M. C. Lee, S.-L. Chan, V. C.-K. Yu and Y.-K. Mok, J. Mol. Biol., 2006, 364, 536; (d) F. R. Stermitz, K. A. Larson and D. K. Kim, J. Med. Chem., 1973, 8, 939.
2. (a) T. K. Beuria, M. K. Santra and D. Panda, Biochemistry, 2005, 44, 16584; (b) R. Jaiswal, T. K. Beuria, R. Mohan, S. K. Mahajan and D. Panda, Biochemistry, 2007, 46, 4211; (c) P. Kosina, J. Gregorova, J. Gruz, J. Vacek, M. Kolar, M. Vogel, W. Roos, K. Naumann, V. Simanek and J. Ulrichova, Fitoterapia, 2010, 81, 1006.
3. (a) W. G. Ma, Y. Fukushi, S. Tahara and T. Osawa, Fitoterapia, 2000, 71, 527; (b) S. E. Newman and M. J. Roll, HortScience, 1999, 34, 686.
4. (a) G. Mazzanti, A. D. Sotto, A. Franchitto, G. L. Mammola, P. Mariani, S. Mastrangelo, F. Menniti-Ippolito and A. Vitalone, J. Ethnopharmacol., 2009, 126, 518; (b) A. Zdarilova, E. Vrublova, J. Vostalova, B. Klejdus, D. Stejskal, J. Proskova, P. Kosina, A. Svobodova, R. Vecera, J. Hrbac, D. Cernochova, J. Vicar, J. Ulrichova and V. Simanek, Food Chem. Toxicol., 2008, 46, 3721; (c) Z. Qiang, W. Jiang, Z. Zhang, M. Qian and B. Du, J. Ethnopharmacol., 2012, 144, 145; (d) K. Pencikova, P. Kollar, V. M. Zavalova, E. Taborska, J. Urbanova and J. Hosek, Phytomedicine, 2012, 19, 890.
5. (a) P. Cheng, J. Zhou, Z. Qing, W. Song, S. Liu, W. Liu, H. Xie and J. Zeng, Bioorg. Med. Chem. Lett., 2014, 24, 2712; (b) M. Janovska, M. Kubala, V. Simanek and J. Ulrichova, Toxicol. Lett., 2010, 196, 56; (c) S. S. Lee, M. Kai and M. K. Lee, Phytother. Res., 2001, 15, 167; (d) A. Sedo, K. Vlasicova, P. Bartak, R. Vespalec, J. Vicar, V. Simanek and J. Ulrichova, Phytother. Res., 2002, 16, 84.
6. (a) P. Cheng and J. Zeng, Chin. J. Org. Chem., 2012, 32, 1605; (b) M. H. Zenk, Pure Appl. Chem., 1994, 66, 2023.
7. (a) Z. Dvorak, V. Kuban, B. Klejdus, J. Hlavac, J. Vicar, J. Ulrichova and V. Simanek, Heterocycles, 2006, 68, 2403; (b) J. Gregorova, J. Babica, R. Marek, H. Paulova, E. Taborska and J. Dostal, Fitoterapia, 2010, 81, 565.
8. J. E. Park, T. D. Cuong, T. M. Hung, I. Lee, M. Na, J. C. Kim, S. Ryoo, J. H. Lee, J. S. Choi, M. H. Woo and B. S. Min, Bioorg. Med. Chem. Lett., 2011, 21, 6960.
9. S. Koul, T. K. Razdan, C. S. Andotra, A. K. Kalla, S. Koul and S. C. Taneja, Planta Med., 2002, 68, 262.
10. N. Zhang, S. R. Samanta, B. M. Rosen and V. Percec, Chem. Rev., 2014, 114, 5848.
11. For representative review on photoredox catalysis, see: (a) C. K. Prier, D. A. Rankic and D. W. C. MacMillan, Chem. Rev., 2013, 113, 5322; (b) R. A. Angnes, Z. Li, C. R. D. Correia and G. B. Hammond, Org. Biomol. Chem., 2015, 35, 9152; (c) J. W. Beatty and C. R. J. Stephenson, Acc. Chem. Res., 2015, 48, 1474.
12. For representative research on iminium ion as electrophilic intermediate, see: (a) M. Rueping, C. Vila and T. Bootwicha, ACS Catal., 2013, 3, 1676; (b) G. Bergonzini, C. S. Schindler, C.-J. Wallentin, E. N. Jacobsen and C. R. J. Stephenson, Chem. Sci., 2014, 5, 112; (c) L. R. Orgren, E. E. Maverick and C. C. Marvin, J. Org. Chem., 2015, 80, 12653; (d) X. Li, X. Gu, Y. Li and P. Li, ACS Catal., 2014, 4, 1897; (e) S. Cai, X. Zhao, X. Wang, Q. Liu, Z. Li and D. Z. Wang, Angew. Chem., Int. Ed., 2012, 51, 8050; (f) A. G. Condie, J. C. Gonzalez-Gomez and C. R. J. Stephenson, J. Am. Chem. Soc., 2010, 132, 1464; (g) D. B. Freeman, L. Furst, A. G. Condie and C. R. J. Stephenson, Org. Lett., 2012, 14, 94.
13. For representative research on α-amino C-radical as radical addition intermediate, see: (a) X. Dai, D. Cheng, B. Guan, W. Mao, X. Xu and X. Li, J. Org. Chem., 2014, 79, 7212; (b) X. Liu, X. Ye, F. Bures, H. Liu and Z. Jiang, Angew. Chem., Int. Ed., 2015, 54, 11443; (c) H. Zhou, P. Lu, X. Gu and P. Li, Org. Lett., 2013, 15, 5646; (d) P. Kohls, D. Jadhav, G. Pandey and O. Reiser, Org. Lett., 2012, 14, 672; (e) P. Zhang, T. Xiao, S. Xiong, X. Dong and L. Zhou, Org. Lett., 2014, 16, 3264; (f) A. Noble and D. W. C. MacMillan, J. Am. Chem. Soc., 2014, 136, 11602; (g) Y. Miyake, K. Nakajima and Y. Nishibayashi, J. Am. Chem. Soc., 2012, 134, 3338; (h) X. Dai, R. Mao, B. Guan, X. Xu and X. Li, RSC Adv., 2015, 5, 55290.
14. D. Uraguchi, N. Kinoshita, T. Kizu and T. Ooi, J. Am. Chem. Soc., 2015, 137, 13768.
15. (a) Q. Tang, X. Liu, S. Liu, H. Xie, W. Liu, J. Zeng and P. Cheng, RSC Adv., 2015, 5, 89009; (b) M. J. Schnermann and L. E. Overman, Angew. Chem., Int. Ed., 2012, 51, 9576.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra05927a

---