A New Catalytic Cu(II)/Sparteine Oxidant System for ,-Phenolic Couplings of Styrenyl Phenols: Synthesis of Carpanone and Unnatural Analogs

A new catalytic Cu(II)/sparteine system has been developed to promote , -phenolic couplings of styrenyl phenols en route to carpanone and related unnatural congeners in yields exceeding 85%. Benzoxanthenones are a class of lignan natural products, exemplified by a highly oxygenated tetracyclic ring system with four to five contiguous stereocenters, isolated as single diastereomers. Notable members include (Figure 1) carpanone (1), polemannone (2), sauchinone (3), and the unnatural CLL-19 (4). Benzoxanthenones have garnered a great deal of attention since the classic biomimetic synthesis of carpananone (1) by Chapman in 1971 (Scheme 1), which utilized PdCl2 and NaOAc to promote the , -phenolic coupling and subsequent endoselective, inverse-electron demand Diels-Alder reaction. Chapman’s approach afforded carpanone (1) in 46% yield as a single diastereomer, which was confirmed by single X-ray crystallography. After this intial report, several laboratories disclosed additional oxidative systems, both stoichiometric and catalytic, to produce carpanone including metal(II) salen/O2 (metal)Co, Mn, Fe), O2 (hν, rose bengal), AIBN, dibenzoyl peroxide, and AgO in yields ranging from 14-94%. In 2001, Ley reported on the total (1) Brophy, G.; Mohandas, J.; Slaytor, M.; Sternhell, S.; Watson, T.; Wilson, L. Tetrahderon Lett. 1969, 10, 5159–5162. (2) Jaupovic, J.; Eid, F. Phytochemistry 1987, 26, 2427–2429. (3) Sung, S.; Kim, Y. C. J. Nat. Prod. 2000, 63, 1019–1021. (4) Goess, B. C.; Hannoush, R. N.; Chan, L. K.; Kirchhausen, T.; Shair, M. D. J. Am. Chem. Soc. 2006, 128, 5391–5403. (5) Chapman, O. L.; Engel, M. R.; Springer, J. P.; Clardy, J. C. J. Am. Chem. Soc. 1971, 93, 6696–6698. Figure 1. Structures of benzoxanthone natural and unnatural products. ORGANIC LETTERS 2008 Vol. 10, No. 18 4097-4100 10.1021/ol801643t CCC: $40.75  2008 American Chemical Society Published on Web 08/27/2008 synthesis of carpanone employing only solid-supported reagents and scavengers. Around the same time, Lindsley and Shair described a hetero, -phenolic coupling reaction, facilitated by IPh(OAc)2, to deliver heterotetracyclic analogs of carpanone; however, this oxidant system was unable to produce carpanone itself but was able to produce less electron-rich homodimers. We were interested in alternative oxidant systems to promote the , -phenolic coupling reaction, and one that might afford enantioselectivity. Upon perusal of the literature, we were attracted to the work of Hovorka, in which a CuCl2/tert-butyl amine system (4.0 equiv CuCl2, 16.0 equiv tert-butyl amine, 1.0 equiv of each naphthol) was able to promote highly selective oxidative cross-couplings of substituted 2-naphthols, 5 and 6, to afford unsymmetrical 1,1′binaphthols 7 in >90% yields (Scheme 2). Subsequently, catalytic, enantioselective variations were developed that afforded unsymmetrical 1,1′-binaphthols with excellent enantioselection (27-99% ee) employing (-)-sparteine in place of tert-butylamine. However, these conditions had never been applied to , -phenolic coupings of styrenyl phenols. In order to extend this system to , -phenolic couplings and the synthesis of carpanone and related analogs, we first had to prepare the requisite styrenyl phenols. Starting from commercially available 5-methoxysalicylaldehyde 8, an E-selective Wittig reaction afforded styrenyl phenols 9 and 10 in >85% yield and 11 in 30% yield (Scheme 3). The key styrenyl phenol 13 to access carpanone was prepared according to literature precendent from sesamol 12 in three steps. Our studies began by exposing 10 to 4.0 equiv of CuCl2 and 16.0 equiv of tert-butylamine in nondegassed MeOH exposed to air at room temperature for different reaction times (Scheme 4). When the reaction was quenched with saturated NH4Cl after 45 min, the desired homocoupled product 14 was isolated in 80% yield as a single diastereomer, and the relative stereochemistry was confirmed by NOE measurements. When reactions were quenched after 8 h, two products were isolated in ∼1:1 ratio: the desired 14 along with a product 15 consistent with the conjugate addition of MeOH to 14, which afforded a single diastereomer containing six contiguous stereocenters. If the reaction was allowed to proceed in excess of 16 h, the conjugate (6) Matsumoto, M.; Kuroda, K. Tetrahedron Lett. 1981, 22, 4437–4440. (7) Iyer, M. R.; Trivedi, G. K. Bull. Chem. Soc. Jpn. 1992, 65, 1662– 1664. (8) Baxendale, I. R.; Lee, A. I.; Ley, S. V. Synlett 2001, 9, 1482–1484. (9) Lindsley, C. W.; Chan, L. K.; Goess, B. C.; Joseph, R.; Shair, M. D. J. Am. Chem. Soc. 2000, 122, 422–423. (10) Hovorka, M.; Gunterova, J.; Zavada, J. Tetrahedron Lett. 1990, 31, 413–416. (11) Nakajima, M.; Miyoshi, I.; Kanayama, K.; Hashimoto, S-I.; Noji, M.; Koga, K. J. Org. Chem. 1999, 64, 2264–2271. (12) Li, X.; Yang, J.; Kozloski, M. C. Org. Lett. 2001, 3, 1137–1140. (13) Suzuki, Y.; Takahashi, H. Chem. Pharm. Bull. 1983, 31, 1751– 1753. (14) See Supporting Information for full experimental details. Scheme 1. Biomimetic Synthesis of Carpanone Scheme 2. CuCl2/tert-Butylamine Oxidative Coupling of 2-Naphthols To Yield Unsymmetrical 1,1′-Binaphthols Scheme 3. Synthesis of Styrenyl Phenols 9-11 and 13 Scheme 4. CuCl2/tert-Butylamine Oxidative , -Phenolic Coupling To Afford 14 and 15 4098 Org. Lett., Vol. 10, No. 18, 2008 addition product 15 formed exclusively with isolated yields of 89% as a single diastereomer due to selective addition to the convex face of the rigid tetracyclic scaffold. Again, NOE measurements established the relative stereochemistry for 15. We were surprised by the complex molecular architecture of 15 that could arise in a single pot from a starting material devoid of any chiral centers by a , -phenolic coupling, inverse-electron demand DA, and subsequent conjugate addition reaction cascade. Our attention now turned to optimization of these two reactions and evaluation of chiral amine ligands to provide enantioselectivity in the , -phenolic coupling. Utilizing 10, we next surveyed a variety of chiral amine ligands 16-19, both monodentate and bidentate, under a variety of temperatures (-20 °C to rt), concentrations, solvent systems, and copper source with both stoichiometric and catalytic manifolds in order to determine if alternative amine/ copper complexes would promote the , -phenolic coupling reaction and engender a degree of enantioselectivity in the product 14 (Figure 2). As shown in Table 1, we first examined conversion to 14 employing various Cu(I) and Cu(II) salts with the four chiral amines 16-19 at -10 °C in nondegassed MeOH. At this temperature, the standard conditions with tert-butylamine (entry 9) suffered a dimunition in yield, whereas the bidentate ligand 16 afforded excellent conversion to 14 and an 80% isolated yield (entry 1). Catalytic quantities of amine ligand also afforded good conversion to 14 with excess copper. We next examined the conversion of 10 to 14 when utilizing only 10 mol % copper source with 10 mol % of 16-19 in MeOH at -20 °C for 24 h (Table 2). In addition to CuCl2, CuCl and Cu(OTf)2 afforded good results, whereas CuI and CuBr2 faired less well, delivering the Michael adduct 15 as a major side product. Our original conditions of CuCl2/tert-butylamine failed entirely under these low temperature, catalytic conditions. Finally, we evaluated the effect of solvent on the conversion of 10 to 14 (Table 3). For this study, we maintained 10 mol % copper, 10 mol % (-)-sparteine at -20 °C for 24 h and examined CH2Cl2, CH3CN, and MeOH. Clearly, MeOH is the optimal solvent to faciliate the , -phenolic coupling reaction. After an exhaustive survey, only poor enantioselectivity (<5% ee) was observed by analytical chiral LC; however, we noted that bidentate (-)-sparteine was superior Figure 2. Chiral amine ligands surveyed to promote the , phenolic coupling reaction and potentially provide % ee. Table 1. Metal-Catalyzed , -Phenol Homocoupling