Zinc–gold cooperative catalysis for the direct alkynylation of benzofurans

The direct alkynylation of benzofurans was achieved for the first time using the hypervalent iodine reagent 1-[(triisopropylsilyl)ethynyl]-1,2-benziodoxol-3(1H)-one (TIPS-EBX) based on the cooperative effect between a gold catalyst and a zinc Lewis acid. High selectivity was observed for C2-alkynylation of benzofurans substituted with alkyl, aryl, halogen and ether groups. The reaction was also successful in the case of the more complex drug 8-methoxypsoralen (8-MOP).

Keywords: alkynylation; benzofurans; cooperative catalysis; direct functionalization; gold catalysis; hypervalent iodine

Benzofurans are important heterocycles frequently encountered in both bioactive compounds and organic materials (Figure 1). For example, members of the furocoumarin class of natural products including psoralen (1), 8-methoxypsoralen (2) and angelicin (3) can cross-link with DNA upon light irradiation. They have consequently been used for the treatment of skin diseases such as cancer or psoriasis [1-4]. The natural product coumestrol (4) is found especially in soy beans and has estrogenic activity [5]. Synthetic bioactive compounds containing benzofurans are also important, as exemplified by amiodarone (5), as antiarrythmic drug [6,7]. Finally, benzofurans have also emerged recently as important structural elements for organic materials, such as the organic transistor 6 [8].

Due to the importance of benzofurans, the discovery of new efficient methods for their synthesis and functionalization is an intensive field of research [9-11]. Especially interesting would be methods allowing the direct and regioselective C–H functionalization of benzofurans [12]. In this context, the introduction of an alkyne would be particularly useful, as acetylenes are important building blocks in synthetic chemistry, chemical biology and materials science [13]. Nevertheless, to the best of our knowledge, the direct alkynylation of benzofurans is still an unknown process.

Since 2009, our group has developed a mild gold-catalyzed [14-17] method for the alkynylation of electron-rich aryls such as indoles and pyrroles [18], thiophenes [19], anilines [20] and furans [21]. Key for success was the use of ethynylbenziodoxolones, which are cyclic hypervalent iodine reagents [22,23]. Nevertheless, the conditions we have used for other heterocycles gave only very low yields in the case of benzofurans. Herein, we would like to report the first catalytic direct C2-alkynylation of benzofurans 7 based on a cooperative effect between a gold catalyst and a zinc Lewis acid using 1-[(triisopropylsilyl)ethynyl]-1,2-benziodoxol-3(1H)-one (TIPS-EBX, 8) as reagent (Scheme 1). The reaction proceeded under mild conditions (60 °C under air) and could also be used to alkynylate the more complex polycyclic natural product 8-methoxypsoralen (2).

Benzofuran (7a) is less reactive then furans and indeed no product was observed under the conditions optimized for the latter [21] at room temperature or at 60 °C using the commercially available electrophilic alkynylation reagent TIPS-EBX (8) (Table 1, entries 1 and 2) [24-27]. Fortunately, benzofuran (7a) was also more stable in the presence of acidic additives, and co-activation became possible, whereas Zn(OTf)₂ was superior to trifluoroacetic acid (TFA) at 60 °C (Table 1, entries 3 and 4) [19,28]. No product was observed in the absence of AuCl, demonstrating the cooperative effect of the two metals (Table 1, entry 5). Lower or higher temperatures did not increase the yield (Table 1, entries 6 and 7). Finally, using a larger excess of TIPS-EBX (8) and Zn(OTf)₂ gave 75% yield of alkynylation product 9a (Table 1, entry 8). The use of Zn(OTf)₂ in catalytic amount led to a lower yield (Table 1, entry 9), and a larger excess resulted in decomposition of the starting material only (Table 1, entry 10). Although other Lewis acids could also be used (Table 1, entries 11 and 12) [29], no better results than with Zn(OTf)₂ were obtained (Table 1, entry 7). Importantly, in contrast to our previous work with benzothiophenes [19], high selectivity for C2 alkynylation was observed.

Entrya	Equiv 8	Additiveb	T [°C]	Yield
1	1.2	–	23	<5%
2	1.2	–	60	<5%
3	1.2	TFA	60	42%
4	1.2	Zn(OTf)2	60	56%
5	1.2	Zn(OTf)2c	60	<5%
6	1.2	Zn(OTf)2	40	48%
7	1.2	Zn(OTf)2	82	36%
8	2	Zn(OTf)2	60	75%
9	2	Zn(OTf)2d	60	37%
10	2	Zn(OTf)2e	60	0%
11	2	Zn(NTf)2	60	57%
12	2	Yb(OTf)3	60	62%

^aReaction conditions: 7a (0.20 mmol) and AuCl (0.01 mmol) in acetonitrile (0.8 mL) under air for 26 h, isolated yield; ^bsame amount as 8; ^cwithout gold catalyst; ^d0.2 equiv; ^e4.0 equiv.

The scope of the reaction was then investigated (Scheme 2). Substitution by diverse functional groups was first examined on the C5 position. An electron-rich methoxy group was well tolerated, giving the desired alkynylation product 9b in 73% yield. The reaction was also successful with a bromide substituent (product 9c), making the method orthogonal to classical cross-coupling chemistry [30]. In presence of an aryl or an alkyl substituent, alkynylation was also obtained in 72% and 50% respectively (products 9d and 9e). Benzofurans substituted at the C7 position could also be used, as demonstrated by the efficient formation of alkynes 9f and 9g. In contrast, when 7-methoxybenzofuran (7j) was used, no C2 alkynylation product could be isolated. Instead, a mixture of C4 and C6 alkynylated benzofurans 9j and 9j’ was obtained (Scheme 3) [31]. Substitution on the furan ring was also possible at the C3 position (product 9h), but the use of 2-methylbenzofuran (7i) led to very a low yield in the alkynylation reaction.

Finally, we wondered if the alkynylation method could also be successful in the case of more complex benzofuran-containing natural products and drugs. We were pleased to see that the alkynylation of 8-methoxypsoralen (2) was indeed possible. The major product 10 bearing the acetylene group at the C5’ position was obtained in 37% yield (Scheme 4) [32]. Although the yield was still moderate, this was one of the first examples of direct alkynylation of a marketed drug. It also gave access in a single step to an interesting furocoumarin derivative with an extended chromophore, which could be important for phototherapy.

Mechanistically, the reaction could proceed either via π-activation of the triple bond by the gold catalyst followed by conjugate addition of the benzofuran, α-elimination and 1,2-shift, or oxidative addition of TIPS-EBX (8) onto the gold catalyst (either at the Au(I) or Au(0) oxidation level) followed by electrophilic auration and reductive elimination [33]. The role of the zinc Lewis acid is not completely clear at this stage, but it may act by complexing the carboxylate group of the hypervalent iodine reagent, enhancing its electrophilic reactivity [19,34]. In fact, a complete shift of the ¹H NMR signals of TIPS-EBX (8) was observed when Zn(OTf)₂ was added, whereas no signal shift was observed when mixing the Lewis acid and benzofuran (7a) [35].

In conclusion, the first direct alkynylation method of benzofurans has been developed. Key for success was a cooperative effect between a gold catalyst and a zinc Lewis acid, together with the use of the hypervalent iodine reagent TIPS-EBX (8). Preliminary results obtained with 8-methoxypsoralen (2) demonstrated that the reaction could also be applied to more complex furocoumarin natural products.

General procedure for the alkynylation of benzofurans: TIPS-EBX (8, 342 mg, 0.800 mmol, 2.0 equiv), AuCl (4.6 mg, 0.020 mmol, 0.050 equiv), Zn(OTf)₂ (289 mg, 0.800 mmol, 2.0 equiv) and benzofuran 7 (0.40 mmol, 1.0 equiv) were added into CH₃CN (2.0 mL) under air. The mixture was stirred for 26 hours at 60 °C. Then the mixture was concentrated in presence of silica gel and purified directly by column chromatography.

Supporting Information File 1: Experimental part.
Format: PDF	Size: 1.1 MB	Download

1. Song, P.-S.; Tapley, K. J., Jr. Photochem. Photobiol. 1979, 29, 1177. doi:10.1111/j.1751-1097.1979.tb07838.x
   Return to citation in text: [1]
2. Gasparro, F. P.; Dallamico, R.; Goldminz, D.; Simmons, E.; Weingold, D. Yale J. Biol. Med. 1989, 62, 579.
   Return to citation in text: [1]
3. Caffieri, S. Photochem. Photobiol. Sci. 2002, 1, 149. doi:10.1039/b107329j
   Return to citation in text: [1]
4. Santana, L.; Uriarte, E.; Roleira, F.; Milhazes, N.; Borges, F. Curr. Med. Chem. 2004, 11, 3239. doi:10.2174/0929867043363721
   Return to citation in text: [1]
5. Bickoff, E. M.; Booth, A. N.; Lyman, R. L.; Livingston, A. L.; Thompson, C. R.; Deeds, F. Science 1957, 126, 969. doi:10.1126/science.126.3280.969-a
   Return to citation in text: [1]
6. Mason, J. W. N. Engl. J. Med. 1987, 316, 455. doi:10.1056/NEJM198702193160807
   Return to citation in text: [1]
7. Bogazzi, F.; Tomisti, L.; Bartalena, L.; Aghini-Lombardi, F.; Martino, E. J. Endocrinol. Invest. 2012, 35, 340. doi:10.3275/8298
   Return to citation in text: [1]
8. Mitsui, C.; Soeda, J.; Miwa, K.; Tsuji, H.; Takeya, J.; Nakamura, E. J. Am. Chem. Soc. 2012, 134, 5448. doi:10.1021/ja2120635
   Return to citation in text: [1]
9. Cacchi, S.; Fabrizi, G.; Goggiomani, A. Heterocycles 2002, 56, 613. doi:10.3987/REV-01-SR(K)1
   Return to citation in text: [1]
10. Hou, X.-L.; Yang, Z.; Yeung, K.-S.; Wong, H. N. C. Five-membered ring systems: furans and benzofurans; Progress in Heterocyclic Chemistry, Vol. 18; Elsevier, 2007; pp 187–217. doi:10.1016/S0959-6380(07)80011-0
    Return to citation in text: [1]
11. De Luca, L.; Nieddu, G.; Porcheddu, A.; Giacomelli, G. Curr. Med. Chem. 2009, 16, 1–20. doi:10.2174/092986709787002826
    Return to citation in text: [1]
12. Yu, J.-Q.; Shi, Z. C–H Activation; Topics in Current Chemistry, Vol. 292; Springer: Heidelberg, 2010. doi:10.1007/978-3-642-12356-6
    Return to citation in text: [1]
13. Diederich, F.; Stang, P. J.; Tykwinski, R. R. Acetylene chemistry: chemistry, biology, and material science; Wiley-VCH: Weinheim; Germany, 2005.
    Return to citation in text: [1]
14. Fürstner, A.; Davies, P. W. Angew. Chem., Int. Ed. 2007, 46, 3410. doi:10.1002/anie.200604335
    Return to citation in text: [1]
15. Hashmi, A. S. K. Chem. Rev. 2007, 107, 3180. doi:10.1021/cr000436x
    Return to citation in text: [1]
16. Gorin, D. J.; Sherry, B. D.; Toste, F. D. Chem. Rev. 2008, 108, 3351. doi:10.1021/cr068430g
    Return to citation in text: [1]
17. Arcadi, A. Chem. Rev. 2008, 108, 3266. doi:10.1021/cr068435d
    Return to citation in text: [1]
18. Brand, J. P.; Charpentier, J.; Waser, J. Angew. Chem., Int. Ed. 2009, 48, 9346. doi:10.1002/anie.200905419
    Return to citation in text: [1]
19. Brand, J. P.; Waser, J. Angew. Chem., Int. Ed. 2010, 49, 7304. doi:10.1002/anie.201003179
    Return to citation in text: [1] [2] [3] [4]
20. Brand, J. P.; Waser, J. Org. Lett. 2012, 14, 744. doi:10.1021/ol203289v
    Return to citation in text: [1]
21. Li, Y.; Brand, J. P.; Waser, J. Angew. Chem., Int. Ed. 2013, 52, 6743. doi:10.1002/anie.201302210
    Return to citation in text: [1] [2]
22. Zhdankin, V. V. Curr. Org. Synth. 2005, 2, 121. doi:10.2174/1570179052996982
    Return to citation in text: [1]
23. Brand, J. P.; Fernández González, D.; Nicolai, S.; Waser, J. Chem. Commun. 2011, 47, 102. doi:10.1039/c0cc02265a
    Return to citation in text: [1]
24. Zhdankin, V. V.; Kuehl, C. J.; Krasutsky, A. P.; Bolz, J. T.; Simonsen, A. J. J. Org. Chem. 1996, 61, 6547. doi:10.1021/jo960927a
    Return to citation in text: [1]
25. Brand, J. P.; Waser, J. Synthesis 2012, 44, 1155. doi:10.1055/s-0031-1290589
    Return to citation in text: [1]
26. Bouma, M. J.; Olofsson, B. Chem.–Eur. J. 2012, 18, 14242. doi:10.1002/chem.201202977
    Return to citation in text: [1]
27. Brand, J. P.; Waser, J. Chem. Soc. Rev. 2012, 41, 4165. doi:10.1039/c2cs35034c
    See for a review on electrophilic alkynylation.
    Return to citation in text: [1]
28. Wang, Y.; Liu, L.; Zhang, L. Chem. Sci. 2013, 4, 739. doi:10.1039/c2sc21333h
    See for a recent example of cooperative gold-zinc catalysis.
    Return to citation in text: [1]
29. See Supporting Information for a full list of tested Lewis acids.
    Return to citation in text: [1]
30. No product was obtained with strongly electron-withdrawing substituents, such as cyanide.
    Return to citation in text: [1]
31. A 5:1 mixture of non-separable products was obtained, which prevented complete assignment of the structure of the major regioisomer.
    Return to citation in text: [1] [2]
32. The regiochemistry of the alkynylation was determined by 2D NMR experiments after removal of the silyl protecting group on alkyne 10. One other non-identified isomer was observed in the crude mixture by 1H NMR (yield < 5%).
    Return to citation in text: [1]
33. Brand, J. P.; Chevalley, C.; Scopelliti, R.; Waser, J. Chem.–Eur. J. 2012, 18, 5655. doi:10.1002/chem.201200200
    Return to citation in text: [1]
34. Koller, R.; Stanek, K.; Stolz, D.; Aardoom, R.; Niedermann, K.; Togni, A. Angew. Chem., Int. Ed. 2009, 48, 4332. doi:10.1002/anie.200900974
    Return to citation in text: [1]
35. See Figure S1 in the Supporting Information File 1.
    Return to citation in text: [1]