Total Synthesis of Resiniferatoxin Enabled by Radical-Mediated Three

Oct 17, 2017 - Graduate School of Pharmaceutical Sciences, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan ...... However, their use ...
0 downloads 6 Views 3MB Size
Article Cite This: J. Am. Chem. Soc. 2017, 139, 16420-16429

pubs.acs.org/JACS

Total Synthesis of Resiniferatoxin Enabled by Radical-Mediated Three-Component Coupling and 7-endo Cyclization Satoshi Hashimoto, Shun-ichiro Katoh, Takehiro Kato, Daisuke Urabe,† and Masayuki Inoue* Graduate School of Pharmaceutical Sciences, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan S Supporting Information *

ABSTRACT: Resiniferatoxin (1) belongs to a daphnane diterpenoid family and has strong agonistic effects on TRPV1, a transducer of noxious temperature and chemical stimuli. The densely oxygenated trans-fused 5/7/6-tricarbocycle (ABC-ring) of 1 presents a daunting challenge for chemical synthesis. Here we report the development of a novel radical-based strategy for assembling 1 from three components: A-ring 9, allyl stannane 18b, and C-ring 17b. The 6-membered 17b, prepared from D-ribose derivative 19, was designed to possess the caged orthoester structure with α-alkoxy selenide as a radical precursor. Upon treatment of 17b with 18b, 9, and V-40, the potently reactive α-alkoxy bridgehead radical was generated from 17b and then sequentially coupled with 9 and 18b to yield 16b. This first radical reaction formed the hindered C9,10-linkage between the A and C-rings and extended the C4-chain on the A-ring in a stereoselective fashion. After derivatization of 16b into 15, the remaining 7-membered B-ring was cyclized in the presence of n-Bu3SnH and V-40 by utilizing the xanthate on the C-ring as the radical precursor and the allylic dithiocarbonate as the terminator. The second radical reaction thus enabled not only the 7endo cyclization but also construction of the C8-stereocenter and the C6-exo olefin. Tricycle 14 was elaborated into the targeted 1 by a series of highly optimized chemoselective reactions. The present total synthesis of 1 demonstrates the advantages of radical reactions for linking hindered bonds within carbocycles without damaging preexisting functionalities, thereby offering a new strategic design for multistep target-oriented synthesis.



INTRODUCTION Euphorbiaceae is a large family of higher plants distributed in tropical and temperate regions. These plants contain numerous bioactive natural products used traditionally as medicines for the treatment of edema, gonorrhea, migraine, and warts.1 To date, over 650 diterpenoids isolated from the plant family have been shown to possess various medically important properties (e.g., antitumor, cytotoxic, antiviral, and anti-inflammatory activities). These diterpenoids are classified into more than 20 subcategories according to their skeletal types. In 1975, Hecker et al. isolated resiniferatoxin (1, Figure 1) as an irritant component from the latex of Euphorbia resinifera.2 The structure elucidation in 1982 revealed that 1 belongs to a daphnane diterpenoid family.3 Compound 1 was shown to exhibit its biological activity by potently activating transient receptor potential vanilloid 1 (TRPV1).4 TRPV1 is an ion channel protein in the plasma membrane of sensory neurons that transduces multiple painful stimuli, including noxious heat, extracellular pH, and chemicals such as capsaicin from hot chili peppers.5,6 TRPV1 recently attracted a great deal of attention as a potential therapeutic target for treating chronic inflammatory, neuropathic, and cancer pains. Since 1 has exceedingly strong analgesic properties through the desensitization of nociceptive neurons, it holds great promise for alleviating multiple modalities of pain.7,8 © 2017 American Chemical Society

Figure 1. Structures of resiniferatoxin (1) and related diterpenoids.

As exemplified by 1, genkwanine D,9 and daphnetoxin,10 the daphnane diterpenoids are characterized by a trans-fused 5/7/6-tricarbocycle (ABC-ring), a C9,13,14-orthoester motif, Received: September 28, 2017 Published: October 17, 2017 16420

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society

coupling strategies motivated us to integrate powerful radical reactions for the synthesis of 1.23 We envisaged that achieving such a radical-based assembly of this extremely complex diterpenoid would lead to a true methodological innovation in the field of synthetic organic chemistry. Herein, we detail the development of a novel route to resiniferatoxin (1). The complex 5/7/6-tricyclic structure of 1 was efficiently built by inter- and intramolecular radical reactions: a three-component coupling reaction between A-, C-rings, and a side-chain, and following 7-endo B-ring cyclization.24 The new strategy and tactics developed here will have further applications to total syntheses of numerous daphnane and tigliane diterpenoids with identical ring substructures.

and a C13-isopropenyl group on the C-ring. While these three natural products differ in their oxidation and unsaturation levels at the AB-ring, they share the C-ring moiety except for the attached orthoesters. On the other hand, phorbol11 and crotophorbolone12 are a tigliane diterpenoid and its derivative, respectively, and possess the same AB-ring substructure as 1 but differently substituted C-rings. The distinct functional group patterns on the common tricyclic framework impart diverse biological functions to these diterpenoids. For example, genkwanine D, daphnetoxin, and the acylated phorbol display cytotoxic, anti-HIV, and tumor-promoting activities, respectively. Daphnanes, tiglianes and their derivatives together have attracted a great deal of attention from the synthetic community due to their significant biological activities and architecturally complex structures.13,14 The most sought-after target among them to date is phorbol. In 1989, the Wender group accomplished a racemic total synthesis of phorbol,15 followed by the asymmetric version in 1997.16 Cha’s group disclosed their own solution to this problem in 2001,17 and Baran’s group described the elegant biosynthesis-inspired total synthesis in 2016.18 In 2015, our group achieved a chemical synthesis of crotophorbolone, a derivative of phorbol, by a radical cyclization.19 Despite many attempts over 40 years from the isolation, the sole successful synthesis of daphnanes is the de novo construction of 1 by Wender and co-workers in 1997.20 Wender ingeniously synthesized 1 via intramolecular 1,3dipolar cycloaddition and zirconium-mediated cyclization to assemble the tricyclic ring system (Scheme 1). After



RESULTS AND DISCUSSION Synthetic Plan for Resiniferatoxin. The intricate tricyclic structure of 1 possesses three double bonds (C1,6,15) and seven contiguous stereogenic centers, including three tetrasubstituted carbons (C4,9,13) (Scheme 2B). The 5membered A- and 6-membered C-rings are linked directly at C9 and C10 and indirectly at C4 and C8 through the fourcarbon spacer (C5−7,20), and these two-rings and spacer together constitute the seven-membered B-ring. We planned to employ radical-based methodologies to form all three ringconnecting bonds (C4−5, C7−8, and C9−10) highlighted in color in Scheme 2. To simplify the overall synthetic scheme, we were particularly interested in utilizing the threecomponent radical coupling reaction to realize single-step formation of the C4−5 and C9−10 bonds. Such radical multicomponent couplings, however, have been underexplored in comparison to their ionic counterparts.25 The feasibility of the idea was validated by our recent model study (Scheme 2A).26 Specifically, 2,4,10-trioxadamantane orthoester 8, cyclopentenone 9, and allyl stannane 1027 underwent coupling by the intermediacy of 11 and 12, producing 13 with the new C4−5 and C9−10 linkages. αAlkoxy bridgehead radical 11, generated from α-alkoxy selenide 8, was tied back by the cage structure. 28 Consequently, the radical center was more exposed and thus more reactive, and the O-substituted stereocenter was fixed. These features permitted the reaction to proceed with stereospecific bond formation at the sterically hindered C9postion of 13.29 We decided to implement this powerful methodology in our synthetic scheme for resiniferatoxin (1) (Scheme 2B). Accordingly, 1 was retrosynthetically disassembled into αalkoxy selenide (C-ring) 17b, cyclopentenone (A-ring) 9, and branched allyl stannane 18b. The three-component reaction of these fragments would be significantly more challenging than the model reaction because the C9-radical-generating position is shielded by the proximal C8- and C11-substituents and the cyclohexane ring is locked into the twist-boat conformation by the orthoester at the C9,13,14-OHs. Nevertheless, the singlestep reaction would rapidly build up structurally complex intermediate 16b, which would be easily derivatized into 15, the substrate for the second radical reaction. Radical C7−8 bond formation from 15 was expected to cyclize the B-ring with the C6-exo olefin. The functional groups would then be adjusted and attached to transform 14 into the target 1. The pivotal intermediate 17b of this radical-based strategy was planned to be synthesized from D-ribose derivative 19. Stereoselective Synthesis of the C-Ring. We prepared the three radical donors 38, 17a, and 17b from starting

Scheme 1. Wender’s Total Synthesis of Resiniferatoxin (1)

preparation of 3 from penta-1,4-dien-3-ol (2), oxidopyrylium ion 4, generated by treating 3 with DBU, reacted with the terminal olefin to simultaneously form the B- and C-rings of 5. Elongation of the carbon chains at the C4,10-positions from 5 gave rise to 6, which was treated with Cp2ZrBu2 to produce the crucial tricycle 7. Then, subsequent functional group transformations from 7 resulted in the total synthesis of 1 (44 total steps from 2). Radical-based carbon−carbon (C−C) bond formation is in principle an ideal reaction for the total synthesis of densely oxygenated natural products because it exhibits high efficiency under neutral conditions without affecting polar functional groups.21,22 Our continued interest in developing radical 16421

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society

ketone of 24 via protonation of the C9-enol intermediate from the convex face.33 The reaction thus set the correct C11configuration and differentiated the C9- and C13-oxidation levels. The NOESY correlations of H8−H11 and H11−H13 corroborated the C8,11,13-equatorially oriented chair conformation of 24. Then, the lithiated ethyl vinyl ether equatorially attacked the β-face of C9-ketone 24, stereoselectively furnishing the carbon-extended 25. After oxidation of the secondary C13-hydroxy group of 25 to the C13-ketone of 26 by TPAP catalysis,34 isopropenyl lithium was added from the β-face to produce 27 with the requisite C13stereocenter. Hence, construction of the four stereogenic centers corresponding to the C-ring of 1 was completed at this stage. The unusual orthoester-attached α-alkoxy selenide structure of 36 was prepared from 27. The acidic hydrolysis of the acetonide and the enol ether transformed diol 27 to tetraol 28 with the methyl ketone at the C9-position. Benzoylation of the secondary C8-hydroxy group of 28 occurred siteselectively over the C14-hydroxy group by reflecting its subtle steric differences, leading to monobenzoylated 29. The remaining three free hydroxy groups of 29 were simultaneously capped using trimethyl orthoacetate in acidic media to generate the caged ortho acetate 30. The C9-methyl ketone of 30 was in turn converted to the phenyl selenide of 36 with the following four steps. Treatment of 30 with TMSOTf and Et3N afforded TMS-enol ether 31. The electron-rich C9′-olefin of 31 was oxidized with m-CPBA chemoselectively over the C15-olefin to provide α-hydroxy ketone 32, which was oxidatively cleaved with Pb(OAc)4 in one pot to furnish carboxylic acid 33. Mesyl ester 34 was formed from 33 with MsCl and Et3N35 and then transformed into Barton ester 35 using 2-mercaptopyridine N-oxide sodium salt (B) and DMAP.36 Upon photoirradiation of unstable 35 without isolation, the C9-bridgehead radical was generated, and in situ-captured by (SePh)2, resulting in the construction of α-alkoxy selenide 36.37 The thus-obtained 36 was utilized as the intermediate for synthesis of the three radical donors 38, 17a, and 17b, which differ only at the C8-positions (α-OAc for 38, β-OAc for 17a, and β-OTMS for 17b). Basic methanolysis of 36 afforded αalcohol 37, which was acetylated to obtain 38. On the other hand, the C8-stereocenter of 37 was inverted through oxidation with DMSO and Ac2O38 and reduction of 39 with NaBH4 and CeCl3·7H2O,39 leading to β-alcohol 40. Secondary alcohol 40 was protected as the acetate and TMS ether under standard conditions to provide 17a and 17b, respectively. Overall, 38 and 17a/b were synthesized from 19, in 17 and 19 steps, respectively. Optimization of Three-Component Radical Coupling Reactions. To attain an efficient three-component radical coupling reaction, we separately screened the components (Scheme 4) and thus employed the three C-rings (38, 17a, and 17b), the one A-ring (9),40 and the two branched allyl stannane (18a and 18b).41 As shown in Scheme 4, the combination of 17b, 9, and 18b delivered the coupling adduct in the highest yield. When 38 was heated to 130 °C with 9 (5 equiv) and 18a (5 equiv) in the presence of V-40 (0.5 equiv) in chlorobenzene, a sequential radical reaction underwent to give adduct 43. The obtained inseparable mixture of 43 and 9 was treated with K2CO3 in MeOH to produce pure 44, albeit in only 9% yield from 38. Although this basic treatment

Scheme 2. (A) Model Study of the Three-Component Radical Coupling Reaction. (B) Synthetic Plan for Resiniferatoxin (1)

material 19 to illuminate the structural factors for an efficient three-component coupling reaction (Scheme 3). First, 6membered ring 23 was constructed from the commercially available D-ribose derivative 19. Compound 19 was treated with n-BuLi to deprotonate the acidic protons and then with vinylmagnesium bromide to provide triol 20 as a single C13stereoisomer. Oxidative cleavage of the resultant 1,2-diol of 20 generated hemiacetal 21, which was sequentially subjected to n-BuLi and isopropenylmagnesium bromide to afford 22 in a C9-stereoselective manner. Diene 22 was in turn treated with the Hoveyda-Grubbs second-generation catalyst30,31 (A, 2 mol %) at 80 °C in (CH2Cl)2, leading to 6-membered ring 23. In this ring-closing metathesis reaction, 1,4-benzoquinone effectively impeded the olefin isomerization of 22.32 The stereostructure of product 23 was confirmed by X-ray crystallographic analysis (see CIF file for 23). Next, the C9,11,13-stereocenters were installed by taking advantage of the convex/concave bias forged by the acetonide-protected cis-fused ring systems. Interestingly, Pd/ C and H2 isomerized the allylic C9-alcohol of 23 to the C916422

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society Scheme 3. Stereoselective Synthesis of the C-Ringa

Reagents and conditions: (a) n-BuLi, vinylmagnesium bromide, THF, 80%; (b) NaIO4, THF, H2O, 0 °C; (c) n-BuLi, isopropenylmagnesium bromide, THF, 77% (2 steps); (d) A (2 mol %), 1,4-benzoquinone, (CH2Cl)2, 80 °C, 81%; (e) H2, Pd/C, EtOAc, hexane, 0 °C, 71%; (f) t-BuLi, ethyl vinyl ether, THF, 0 °C, 91%; (g) TPAP, 4-methylmorpholine N-oxide, CH2Cl2, MS4A; (h) t-BuLi, 2-bromopropene, TMEDA, THF, −45 °C, 54% (2 steps) (recovered 26: 24%); (i) Dowex 50W, THF, H2O, 90 °C; (j) BzCl, pyridine, CH2Cl2, 0 °C, 83% (2 steps); (k) MeC(OMe)3, (+)-CSA, benzene, 85%; (l) TMSOTf, Et3N, CH2Cl2, 87%; (m) m-CPBA, CH2Cl2, 0 °C; TBAF, 0 °C; Pb(OAc)4, K2CO3, toluene; (n) MsCl, Et3N, CH2Cl2, 0 °C; (o) B, DMAP, toluene; hν, (SePh)2, 31% (3 steps); (p) K2CO3, MeOH, 0 °C, 87%; (q) Ac2O, DMAP, pyridine, CH2Cl2, 87%; (r) DMSO, Ac2O, 35 °C; (s) NaBH4, CeCl3·7H2O, MeOH, −78 °C; (t) Ac2O, DMAP, pyridine, CH2Cl2, 84% (3 steps); (u) TMSOTf, 2,6-lutidine, CH2Cl2, 0 °C, 88% (3 steps). a

Scheme 4. Optimization of Three-Component Radical Coupling Reactions

comprised the multiple reactions (elimination of the C1-siloxy group, detachment of the C8O-acetyl group, and oxy-Michael addition of the C8-OH to the enone), the low isolated yield of 44 clearly reflected the low efficiency of the coupling of 38. We postulated that the steric shielding effect of the equatorially disposed C8-acetoxy group lowered the reactivity of C9-bridgehead radical 41. In fact, the C8-epimeric 17a with

the axial C8-acetoxy group was determined to be a better radical precursor. After submitting 17a, 9, and 18a to the radical and basic conditions, enone 47 was obtained in 27% yield. No concomitant oxy-Michael addition from 47 was likely to originate from the different spatial relationship between C1 and C8-OH. To increase the trapping rate of the radical, tri-n-butyl allyl stannane 18a was switched to triphenyl 16423

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society

would lead to undesired pathways, thus decreasing the overall efficiency.42 The structure of 16b was determined by detailed NMR experiments (Figure 2A). The coupling constants of H1−H10 (J = 2.9 Hz) and H4−H10 (J = 3.7 Hz) indicated pseudo axial orientations of the C1,4,10-substituents, and the NOE correlations of H1−H11, H4−H8, and H5−H10 prove the trans-relationships of these three groups. The DFToptimized structure of 16b at the M06−2X/6-31g(d) level of theory (298 K, 1 atm)43 further supports the unique threedimensional (3D) shape of 16b (Figure 2B). It was remarkable that the three-component reaction of 16b realized sequential attachment of the A-ring and the branched allyl chain and introduction of the C4,10-stereocenters with stereochemical retention of the bridgehead tetrasubstituted C9-carbon (Scheme 4). The chemo- and stereoselective two C−C bond formations would be controlled by the intrinsic characters of the radical species and the 3D-structures of the reactants. The nucleophilic α-alkoxy bridgehead radical 45b generated via homolytic cleavage of the C−Se bond of 17b first reacts with the electron-deficient olefin of 9 from the opposite face of the bulky TBSO-group to set the C10stereocenter of 46b. Then, electrophilic radical 46b adds to the electron-rich olefin of 18b in a trans-manner to the large C-ring to introduce the C4-stereocenter of 16b. Construction of the Tricyclic Framework by Radical 7-endo Cyclization. Therefore, a radical three-component coupling reaction of 9, 17b, and 18b intermolecularly linked the two hindered ring-connecting bonds of 16b (C9−10 and C4−5) and markedly increased the complexity of the structure in a single step (Scheme 5). We envisaged that

Figure 2. (A) The NMR data of 16b. (B) The DFT-optimized structure of 16b (the TBDPS group was replaced with the Me group. M06−2X/6-31g(d), 298 K, and 1 atm).

counterpart 18b. The modification indeed resulted in the higher yielding generation of 47 (33%) after the two sequential reactions (17a → 16a → 47). Most importantly, changing the protective group of the C8OH from Ac (17a) to TMS (17b) doubled the yield of the coupling adduct. Namely, treatment of 17b with 9 (5 equiv), 18b (5 equiv), and V-40 (0.4 equiv) at 130 °C delivered 16b in 52% yield as a single isomer after purification. This dramatic improvement was likely due to the potential disadvantage of the acetyl group of 45a compared with the silyl group of 45b. Neighboring group participation of the acetyl CO double bond of 45a to the C9-radical center

Scheme 5. Synthesis of the Tricyclic Framework by Employing the Two Radical Reactionsa

Reagents and conditions: (a) 9 (5 equiv), 18b (5 equiv), V-40 (0.4 equiv), chlorobenzene, 130 °C, 52%; (b) NaN(TMS)2, THF, 0 °C, 77%; (c) TBSOTf, Et3N, CH2Cl2, 0 °C; (d) m-CPBA, NaHCO3, hexane, CH2Cl2, 0 °C, 59% (2 steps); (e) DIBAL-H, CH2Cl2, −93 °C; (f) 2,2dimethoxytetrahydrofuran, (+)-CSA, benzene, 50 °C, 60% (dr = 5:3, 2 steps); (g) TBAF, THF; (h) NaH, CS2, MeI, THF, 90% (2 steps); (i) xylene, 110 °C; n-Bu3SnH, V-40, 180 °C (microwave), 71%. a

16424

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society

supporting the observed C4-stereoselectivity of 49. The higher energy of TS2 in comparison to TS1 would be attributable to its higher tortional strain: the dihedral angle of C5−C4− C10−C9 (38.9°) of TS2 is smaller than that of TS1 (67.4°).44 Since the steric interaction between the C5- and C9-positions becomes larger in TS2, epoxidation occurs via TS1 to produce 58a selectively over 58b, and 58a is hydrolyzed to β-alcohol 59a. The tricyclic framework 14 was constructed from 50 by the subsequent 5 steps (Scheme 5). DIBAL-H-reduction stereoselectively converted α-hydroxy ketone 50 into syn-1,2-diol 51, which was protected as the oxacyclopentylidene orthoester using 2,2-dimethoxy tetrahydrofuran and (+)-CSA to produce 52.45 After the TBDPS and TMS groups of 52 were removed with TBAF, the two hydroxy groups of 53 were simultaneously derivatized into the corresponding xanthates of 54 by treatment with NaH, CS2, and MeI.46 Before forming the 7-membered ring, we differentiated the radicalgenerating propensity of the C7- and C8-positions of bisxanthate 54. This was realized simply by heating 54 to 110 °C in xylene, which promoted the [3,3]-sigmatropic rearrangement of the allylic xanthate of 54 into the less reactive dithiocarbonate of 15.47 Then, 15 was in situ-heated to a higher temperature (180 °C) under microwave irradiation in the presence of n-Bu3SnH and V-40 to deliver tricycle 14 as the only isolable product in 71% yield. Under these conditions, the secondary C8-radical 55 was chemoselectively formed from 15 via C8−O homolysis and added to the less hindered C7-position rather than the C6-position. The subsequently generated tertiary C6-radical 56 in turn ejected the dithiocarbonate functionality via the C20−S homolysis, thus terminating the radical reaction.48 Consequently, this SH2′ reaction realized 7-endo cyclization of the B-ring,49 attachment of the C6-exo olefin, and introduction of the correct C8-stereocenter. The lack of participation of the five oxygen-based functionalities (C3,4,9,13,14) and the two olefins (C1,15) demonstrated the excellent chemoselectivity of the radical reaction. The perfect C8-stereoselectivity of the cyclization indicated that the C7-olefin of 55 approached the C-ring only from the α-face. To identify structural elements responsible for the phenomenon, the DFT calculation (UM06-2X/6-31g(d), 298 K, 1 atm) was performed using 60 (Scheme 7), in which the C13- and C20-substituents of 55 were omitted, and the oxacyclopentylidene orthoester was replaced with acetonide. As a result, TS3 (α-face approach), which leads to 61a, was calculated to be energetically favored by 2.69 kcal/mol (ΔG) over TS4 (β-face approach), which leads to 61b. The higher energy of TS4 is likely to come from the close contacts of the two atoms within the sum of the van der Waals radii [H−H (2.20 Å) and H−O (2.52 Å)]. Whereas both TS3 and TS4 have the three disfavored interactions indicated in the blue dotted lines, the atom distances of TS4 [H−H (2.06 Å), H− O (2.11, 2.27 Å)] are shorter than those of TS3 [H−H (2.11 Å), H−O (2.43, 2.49 Å)]. Thus, the calculation results clarified that the formation of 14 was preferred due to the less steric interactions of the corresponding transition state. Scheme 8 illustrates the two selected negative data to show the difficulties associated with B-ring construction. Substrate 62, which only differs from 54 in its C3,4-oxidation states and capped protective group, similarly underwent thermal rearrangement to afford dithiocarbonate 63. Submission of 63 to radical conditions (n-Bu3SnH, V-40, xylene, 180 °C,

the remaining ring-connecting C7−8 bond would be constructed by an intramolecular radical reaction. Prior to this second radical reaction, the A-ring was functionalized. NaN(TMS)2-induced elimination of the C1-siloxy group transformed 16b to enone 48, which was treated with TBSOTf and Et3N to afford TBS-enol ether 49. m-CPBAoxidation of 49 then provided tertiary C4-alcohol 50 as the sole isomer, the C4-configuration of which was unambiguously assigned by X-ray crystallographic analysis of its derivative (see 67 of Scheme 9). Accordingly, the oxidation of 49 proceeded from the same face with the sterically cumbersome C-ring structure upon establishing the desired C4-stereogenic center. To rationalize this counterintuitive stereochemical outcome, we performed DFT calculations of the transition states (M062X/6-31g(d), 298 K, 1 atm, Scheme 6). In facilitating the calculation, the C3O-TBS, C7O-TBDPS and C8O-TMS, and C13-C(CH2)CH3 of 49 were simplified into the C3O-TMS, C7O-Me, C8O-Me, and C13-H of 57, respectively, while mCPBA was replaced with peroxybenzoic acid. The calculated activation energy of TS1 (β-face approach) is smaller than that of TS2 (α-face approach) by 2.61 kcal/mol (ΔG), Scheme 6. Rationale of the C4-Stereoselectivitya

a

Values in parentheses are relative free energies: ΔG, 298 K, 1 atm. 16425

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society Scheme 7. Rationale of the C8-Stereoselectivitya

a

functional group transformations needed to be properly orchestrated and judiciously optimized to achieve the total synthesis (Scheme 9). Before constructing the C3-ketone and the ortho phenylacetate, pentaol 67 was prepared from 14. Full deprotection turned out to be problematic due to the presence of the two acid-sensitive allylic and tertiary hydroxy groups in 67. Importantly, the oxacyclopentylidene orthoester of 14 allowed us to use relatively mild conditions for this particular reaction. Specifically, treatment of 14 with 1.5 M HCl in aqueous MeOH simultaneously detached the two orthoesters to furnish pentaol 67 after the addition of LiOH to the mixture to remove the partially remaining acetyl group. While orthoacetate 68 was recovered as the minor compound, this was resubjected to the same conditions to obtain additional 67. The intricate 5/7/6-tricyclic structure of 67 with 8 contiguous stereocenters was clearly established by X-ray crystallographic analysis (see CIF file for 67). Next, the allylic C3-hydroxy group of 67 was chemoselectively oxidized over the secondary C14-hydroxy group utilizing catalytic AZADO, CuCl, 2,2′-bipyridyl, and DMAP under air to generate the C3-ketone of 69.50 The ortho phenylacetate was introduced to the C9,13,14-triol system of 69 to yield 70 by C14-esterification with phenyl acetic acid by the action of activated ester C and DMAP at 0 °C,51 and subsequent treatment with 2,4,6-trichlorobenzoic acid at 50 °C. The last remaining free C4-alcohol of 70 was protected as the TMS ether with TMSOTf and 2,6-lutidine, leading to 71. Then, C2-functionalization and subsequent C7-hydroxylation transformed 71 into 77. After 1,4-reduction of C3-enone 71 to C3-ketone 73, the C2-position was methylated using LiN(TMS) 2 and MeI to provide 74. N-Methyl-N(trimethylsilyl)trifluoroacetamide (D) 52 was applied to promote TMS-enol ether formation from 74, and the resulting 75 was treated with NBS to afford C2-brominated 76. Although 76 has the three potential reactive sites (C5,7,16), allylic hydroxylation of 76 with SeO2 in t-BuOH at 80 °C gave rise to β-oriented C7-alcohol 77 as the major product (77: C5-alcohol 78 = 5:1), while leaving the most kinetically protected C16-position intact.53 Interestingly, the C5/7-site-selectivity of the allylic oxidation was significantly influenced by the A-ring structure. SeO2-induced allylic oxidation of 71 exclusively produced undesired C5-hydroxylated compound 72. The unsaturated A-ring of 71 was assumed to affect the conformation and steric environment of the 7-membered B-ring, thereby changing the favorable pathway.54 The last four steps completed the total synthesis of 1. The reagent combination of Li2CO3 and LiBr transformed αbromo ketone 77 to α,β-unsaturated ketone 79 through the elimination of HBr. Allylic C7-alcohol 79 was in turn treated with SOCl2 to induce the SN2′ substitution, producing allylic C20-chloride 80 with the requisite C7-double bond. The introduced chloride of 80 underwent SN2 displacement upon treatment with cesium carboxylate E.55 Finally, TBAFpromoted removal of the TMS group on the tertiary C4OH and TBS group on the phenolic OH, furnishing the targeted resiniferatoxin (1). All of the analytical data including 1 H, 13C NMR, IR, and [α]D value of the synthetic 1 were identical to those of the natural counterpart.

Values in parentheses are relative free energies: ΔG, 298 K, 1 atm.

Scheme 8. Attempted B-Ring Cyclizations

microwave), however, led to only decomposition without providing 64, presumably because the electron-rich C8-radical intermolecularly added to the electron-deficient carbonylconjugated C1-olefin in the course of the reaction. On the other hand, ruthenium catalyst A promoted no ring-closing metathesis between the two sterically hindered olefins of 65. These experiments accentuated the importance of the substrate design and reaction mode for achieving successful cyclization of the complex intermediate. Total Synthesis of Resiniferatoxin. The remaining key issues from 14 to 1 were modification of the carbon framework at C2,3,7,20 and decoration of the surrounding motifs (the ortho phenylacetate and homovanillate). These 16426

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society Scheme 9. Total Synthesis of Resiniferatoxin (1)a

Reagents and conditions: (a) 1.5 M HCl in aqueous MeOH, 30 °C, LiOH, 51% for 67 and 36% for 68 from 14; 57% for 67 from 68 (recovered 68: 28%); (b) AZADO, CuCl, 2,2′-bipyridyl, DMAP, CH3CN, air, 0 °C, 90%; (c) C, DMAP, toluene, THF, 0 °C; 2,4,6-trichlorobenzoic acid, 50 °C, 53% (recovered 69: 15%); (d) TMSOTf, 2,6-lutidine, CH2Cl2, 74%; (e) SeO2, t-BuOH, 80 °C, 43% from 71; (f) LiBH(s-Bu)3, THF, −78 °C, 83%; (g) LiN(TMS)2, THF, 0 °C; MeI, −20 °C, 94%; (h) D, DMAP, DABCO, CH3CN, 110 °C; (i) NBS, THF, 0 °C, 88% (2 steps); (j) SeO2, t-BuOH, 80 °C, (77: 78 = 5:1); (k) Li2CO3, LiBr, DMF, 150 °C; (l) SOCl2, pyridine, Et2O, 25% (3 steps); (m) E, DMF; (n) TBAF, THF, 0 °C, 92% (2 steps). a



SUMMARY In summary, we developed a novel radical-based synthetic strategy for the chemical construction of resiniferatoxin (41 total steps from 19 to 1). The synthetic route is highlighted by two powerful radical reactions. Design of the caged C-ring structure 17b with the α-alkoxy selenide enabled us to realize the three-component coupling reaction with A-ring 9 and 18b by the intermediacy of bridgehead radical 45b to produce 16b. After derivatization of 16b into bisxanthate 54, the selective [3,3]-sigmatropic rearrangement of the allylic xanthate to the allylic dithiocarbonate, and the in situ radical reaction of 15 attained 7-endo cyclization to deliver 14 via the secondary radical formation and the dithiocarbonate expulsion. These inter- and intramolecular radical transformations thus annulated the 7-membered B-ring by stereoselectively linking the three hindered bonds (C4−5, C7−8, and C9−10) that connect the A- and C-rings without touching many potentially reactive functionalities. Other notable features of the successful synthesis include: (1) stereoselective preparation of 17b with five consecutive stereocenters by taking advantage of the 3D-shape of the intermediates, (2) stereoselective introduction of the tertiary C4-hydroxy group of 50 by exploiting the torsional straincontrolled reaction, (3) C7-site-selective allylic oxidation of 76 with the properly functionalized A-ring, and (4) construction of the allylic C20-homovanillate by an SN2′-reaction of 79 with SOCl2, followed by an SN2 reaction with cesium carboxylate E. Because of their flexibility and robustness, the strategy and tactics developed here would be applicable to the synthesis of other biologically active daphnanes and tiglianes.

Collectively, we hope that the present nonconventional route to 1 will enrich the science of radical chemistry and offer new perspectives in the strategic design of multistep targetoriented synthesis beyond this work.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b10177. Experimental procedures, characterization data, and NMR spectra of all newly synthesized compounds (PDF) Crystallographic structure for 23 (CIF) Crystallographic structure for 67 (CIF)



AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Masayuki Inoue: 0000-0003-3274-551X Present Address †

Faculty of Engineering, Toyama Prefectural University, 5180 Kurokawa, Imizu-shi, Toyama 939−0398, Japan. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was financially supported by Grant-in-Aids for Scientific Research (S) (JSPS, 17H06110) to M.I. and for 16427

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society

(q) Hassan, A. H. E.; Lee, J. K.; Pae, A. N.; Min, S.-J.; Cho, Y. S. Org. Lett. 2015, 17, 2672. (r) Li, Y.; Wei, M.; Dai, M. Tetrahedron 2017, 73, 4172. (15) (a) Wender, P. A.; Lee, H. Y.; Wilhelm, R. S.; Williams, P. D. J. Am. Chem. Soc. 1989, 111, 8954. (b) Wender, P. A.; Kogen, H.; Lee, H. Y.; Munger, J. D., Jr.; Wilhelm, R. S.; Williams, P. D. J. Am. Chem. Soc. 1989, 111, 8957. (c) Wender, P. A.; McDonald, F. E. J. Am. Chem. Soc. 1990, 112, 4956. (16) Wender, P. A.; Rice, K. D.; Schnute, M. E. J. Am. Chem. Soc. 1997, 119, 7897. (17) Lee, K.; Cha, J. K. J. Am. Chem. Soc. 2001, 123, 5590. (18) Kawamura, S.; Chu, H.; Felding, J.; Baran, P. S. Nature 2016, 532, 90. (19) (a) Asaba, T.; Katoh, Y.; Urabe, D.; Inoue, M. Angew. Chem., Int. Ed. 2015, 54, 14457. (b) Urabe, D.; Asaba, T.; Inoue, M. Bull. Chem. Soc. Jpn. 2016, 89, 1137. For a chemical conversion of phorbol to crotophorbolone, see: (c) Wender, P. A.; Kee, J.-M.; Warrington, J. M. Science 2008, 320, 649. (20) (a) Wender, P. A.; Jesudason, C. D.; Nakahira, H.; Tamura, N.; Tebbe, A. L.; Ueno, Y. J. Am. Chem. Soc. 1997, 119, 12976. For a synthesis of daphnane congeners, see: (b) Wender, P. A.; Buschmann, N.; Cardin, N. B.; Jones, L. R.; Kan, C.; Kee, J.-M.; Kowalski, J. A.; Longcore, K. E. Nat. Chem. 2011, 3, 615. (21) For recent reviews on radical reactions, see: (a) Rowlands, G. J. Tetrahedron 2009, 65, 8603. (b) Rowlands, G. J. Tetrahedron 2010, 66, 1593. (c) Yan, M.; Lo, J. C.; Edwards, J. T.; Baran, P. S. J. Am. Chem. Soc. 2016, 138, 12692. (22) For selected recent examples of intermolecular radical addition reactions in the total synthesis of terpenoids, see: (a) Schnermann, M. J.; Overman, L. E. Angew. Chem., Int. Ed. 2012, 51, 9576. (b) Sun, Y.; Li, R.; Zhang, W.; Li, A. Angew. Chem., Int. Ed. 2013, 52, 9201. (c) Wang, L.; Wang, H.; Li, Y.; Tang, P. Angew. Chem., Int. Ed. 2015, 54, 5732. (d) Tao, D. J.; Slutskyy, Y.; Overman, L. E. J. Am. Chem. Soc. 2016, 138, 2186. (e) Slutskyy, Y.; Jamison, C. R.; Lackner, G. L.; Müller, D. S.; Dieskau, A. P.; Untiedt, N. L.; Overman, L. E. J. Org. Chem. 2016, 81, 7029. (f) Slutskyy, Y.; Jamison, C. R.; Zhao, P.; Lee, J.; Rhee, Y. H.; Overman, L. E. J. Am. Chem. Soc. 2017, 139, 7192. (23) Inoue, M. Acc. Chem. Res. 2017, 50, 460. (24) Murai, K.; Katoh, S.; Urabe, D.; Inoue, M. Chem. Sci. 2013, 4, 2364. This preliminary study describes the synthesis of a tricyclic carbon framework of 1 with the incorrect C10-stereochemistry. In the present study, we significantly shortened the synthetic route to the C-ring (−7 steps), realized the construction of the tricycle with the correct C10-stereochemistry, and accomplished the total synthesis of 1. (25) For recent reviews on multicomponent reactions, see: (a) Multicomponent Reactions; Zhu, J., Bienaymé, H., Eds.; Wiley-VCH: Weinheim, 2005. (b) Nicolaou, K. C.; Edmonds, D. J.; Bulger, P. G. Angew. Chem., Int. Ed. 2006, 45, 7134. (c) Godineau, E.; Landais, Y. Chem. - Eur. J. 2009, 15, 3044. (d) Touré, B. B.; Hall, D. G. Chem. Rev. 2009, 109, 4439. (e) Pellissier, H. Chem. Rev. 2013, 113, 442. (26) Urabe, D.; Yamaguchi, H.; Inoue, M. Org. Lett. 2011, 13, 4778. (27) Keck, G. E.; Yates, J. B. J. Am. Chem. Soc. 1982, 104, 5829. (28) Walton, J. C. Chem. Soc. Rev. 1992, 21, 105. (29) α-Alkoxy bridgehead radical reactions were utilized for the two- and three-component coupling reactions in our laboratory. (a) Urabe, D.; Nagatomo, M.; Hagiwara, K.; Masuda, K.; Inoue, M. Chem. Sci. 2013, 4, 1615. (b) Kamimura, D.; Urabe, D.; Nagatomo, M.; Inoue, M. Org. Lett. 2013, 15, 5122. (c) Nagatomo, M.; Koshimizu, M.; Masuda, K.; Tabuchi, T.; Urabe, D.; Inoue, M. J. Am. Chem. Soc. 2014, 136, 5916. (d) Nagatomo, M.; Hagiwara, K.; Masuda, K.; Koshimizu, M.; Kawamata, T.; Matsui, Y.; Urabe, D.; Inoue, M. Chem. - Eur. J. 2016, 22, 222. (e) Kamimura, D.; Nagatomo, M.; Urabe, D.; Inoue, M. Tetrahedron 2016, 72, 7839. (30) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2000, 122, 8168. (31) Metathesis in Natural Product Synthesis: Strategies, Substrates and Catalysts; Cossy, J., Arseniyadis, S., Meyer, C., Eds.; Wiley-VCH: Weinheim, 2010.

Young Scientists (A) (JSPS, 16H06213) to D.U. Determination of X-ray crystallographic structures was financially supported by Nanotechnology Platform (MEXT, 12024046). Fellowships to S.H. and S.K. from JSPS are gratefully acknowledged.



REFERENCES

(1) For recent reviews on daphnane diterpenoids, see: (a) Shi, Q.W.; Su, X.-H.; Kiyota, H. Chem. Rev. 2008, 108, 4295. (b) Liao, S.G.; Chen, H.-D.; Yue, J.-M. Chem. Rev. 2009, 109, 1092. (c) Vasas, A.; Hohmann, J. Chem. Rev. 2014, 114, 8579. (2) Hergenhahn, M.; Adolf, W.; Hecker, E. Tetrahedron Lett. 1975, 16, 1595. (3) Adolf, W.; Sorg, B.; Hergenhahn, M.; Hecker, E. J. Nat. Prod. 1982, 45, 347. (4) Szallasi, A.; Blumberg, P. M. Neuroscience 1989, 30, 515. (5) For a review on TRPV1, see: Voets, T.; Talavera, K.; Owsianik, G.; Nilius, B. Nat. Chem. Biol. 2005, 1, 85. (6) (a) Liao, M.; Cao, E.; Julius, D.; Cheng, Y. Nature 2013, 504, 107. (b) Elokely, K.; Velisetty, P.; Delemotte, L.; Palovcak, E.; Klein, M. L.; Rohacs, T.; Carnevale, V. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, E137. (7) (a) Brederson, J.-D.; Kym, P. R.; Szallasi, A. Eur. J. Pharmacol. 2013, 716, 61. (b) Brown, D. C. Pharmaceuticals 2016, 9, 47. (8) For recent reviews on the roles of natural product structures in drug discovery, see: (a) Harvey, A. L.; Edrada-Ebel, R.; Quinn, R. J. Nat. Rev. Drug Discovery 2015, 14, 111. (b) Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2016, 79, 629. (c) Allred, T. K.; Manoni, F.; Harran, P. G. Chem. Rev. 2017, 117, 11994. (9) Zhan, Z.-J.; Fan, C.-Q.; Ding, J.; Yue, J.-M. Bioorg. Med. Chem. 2005, 13, 645. (10) (a) Stout, G. H.; Balkenhol, W. J.; Poling, M.; Hickernell, G. L. J. Am. Chem. Soc. 1970, 92, 1070. (b) Vidal, V.; Potterat, O.; Louvel, S.; Hamy, F.; Mojarrab, M.; Sanglier, J.-J.; Klimkait, T.; Hamburger, M. J. Nat. Prod. 2012, 75, 414. (11) Hoppe, W.; Brandl, F.; Strell, I.; Röhrl, M.; Gassmann, I.; Hecker, E.; Bartsch, H.; Kreibich, G.; Szczepanski, Ch. v. Angew. Chem., Int. Ed. Engl. 1967, 6, 809. (12) Thielmann, H. W.; Hecker, E. Liebigs Ann. Chem. 1969, 728, 158. (13) For recent reviews on the total synthesis of terpenoids, see: (a) Maimone, T. J.; Baran, P. S. Nat. Chem. Biol. 2007, 3, 396. (b) Urabe, D.; Asaba, T.; Inoue, M. Chem. Rev. 2015, 115, 9207. (c) Brill, Z. G.; Condakes, M. L.; Ting, C. P.; Maimone, T. J. Chem. Rev. 2017, 117, 11753. (14) For synthetic studies of 1 and related daphnane and tigliane diterpenoids, see: (a) Rigby, J. H.; Kierkus, P. C.; Head, D. Tetrahedron Lett. 1989, 30, 5073. (b) Harwood, L. M.; Ishikawa, T.; Phillips, H.; Watkin, D. J. Chem. Soc., Chem. Commun. 1991, 527. (c) Shigeno, K.; Sasai, H.; Shibasaki, M. Tetrahedron Lett. 1992, 33, 4937. (d) Page, P. C. B.; Jennens, D. C. J. Chem. Soc., Perkin Trans. 1 1992, 2587. (e) Dauben, W. G.; Dinges, J.; Smith, T. C. J. Org. Chem. 1993, 58, 7635. (f) Paquette, L. A.; Sauer, D. R.; Edmondson, S. D.; Friedrich, D. Tetrahedron 1994, 50, 4071. (g) Sugita, K.; Neville, C. F.; Sodeoka, M.; Sasai, H.; Shibasaki, M. Tetrahedron Lett. 1995, 36, 1067. (h) Lee, K.; Cha, J. K. Org. Lett. 1999, 1, 523. (i) Ovaska, T. V.; Reisman, S. E.; Flynn, M. A. Org. Lett. 2001, 3, 115. (j) Jackson, S. R.; Johnson, M. G.; Mikami, M.; Shiokawa, S.; Carreira, E. M. Angew. Chem., Int. Ed. 2001, 40, 2694. (k) Page, P. C. B.; Hayman, C. M.; McFarland, H. L.; Willock, D. J.; Galea, N. M. Synlett 2002, 2002, 583. (l) Wender, P. A.; Bi, F. C.; Buschmann, N.; Gosselin, F.; Kan, C.; Kee, J.-M.; Ohmura, H. Org. Lett. 2006, 8, 5373. (m) Wender, P. A.; D’Angelo, N.; Elitzin, V. I.; Ernst, M.; Jackson-Ugueto, E. E.; Kowalski, J. A.; McKendry, S.; Rehfeuter, M.; Sun, R.; Voigtlaender, D. Org. Lett. 2007, 9, 1829. (n) Stewart, C.; McDonald, R.; West, F. G. Org. Lett. 2011, 13, 720. (o) Catino, A. J.; Sherlock, A.; Shieh, P.; Wzorek, J. S.; Evans, D. A. Org. Lett. 2013, 15, 3330. (p) Tong, G.; Liu, Z.; Li, P. Org. Lett. 2014, 16, 2288. 16428

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429

Article

Journal of the American Chemical Society (32) Hong, S. H.; Sanders, D. P.; Lee, C. W.; Grubbs, R. H. J. Am. Chem. Soc. 2005, 127, 17160. (33) For recent examples of related reactions, see: (a) Constantino, M. G.; Beatriz, A.; da Silva, G. V. J.; Zukerman-Schpector, J. Synth. Commun. 2001, 31, 3329. (b) Wang, Q.; Chen, C. Org. Lett. 2008, 10, 1223. (34) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis 1994, 1994, 639. (35) Nicolaou, K. C.; Baran, P. S.; Zhong, Y.-L.; Choi, H.-S.; Fong, K. C.; He, Y.; Yoon, W. H. Org. Lett. 1999, 1, 883. (36) Barton, D. H. R.; Bridon, D.; Zard, S. Z. Tetrahedron Lett. 1984, 25, 5777. (37) (a) Barton, D. H. R.; Crich, D.; Motherwell, W. B. J. Chem. Soc., Chem. Commun. 1983, 939. For reviews on Barton esters, see: (b) Crich, D.; Quintero, L. Chem. Rev. 1989, 89, 1413. (c) Saraiva, M. F.; Couri, M. R. C.; Le Hyaric, M.; de Almeida, M. V. Tetrahedron 2009, 65, 3563. (38) Albright, J. D.; Goldman, L. J. Am. Chem. Soc. 1967, 89, 2416. (39) Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226. (40) (a) Curran, T. T.; Hay, D. A.; Koegel, C. P.; Evans, J. C. Tetrahedron 1997, 53, 1983. (b) Watson, T. J. N.; Curran, T. T.; Hay, D. A.; Shah, R. S.; Wenstrup, D. L.; Webster, M. E. Org. Process Res. Dev. 1998, 2, 357. (41) Weigand, S.; Brückner, R. Synthesis 1996, 1996, 475. (42) (a) Giese, B.; Gröninger, K. S.; Witzel, T.; Korth, H.-G.; Sustmann, R. Angew. Chem., Int. Ed. Engl. 1987, 26, 233. (b) Korth, H.-G.; Sustmann, R.; Gröninger, K. S.; Leisung, M.; Giese, B. J. Org. Chem. 1988, 53, 4364. (43) (a) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215. (b) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2008, 112, 1095. (44) (a) Cheong, P. H.-Y.; Yun, H.; Danishefsky, S. J.; Houk, K. N. Org. Lett. 2006, 8, 1513. (b) Wang, H.; Houk, K. N. Chem. Sci. 2014, 5, 462. (45) Kennedy, R. M.; Abiko, A.; Takemasa, T.; Okumoto, H.; Masamune, S. Tetrahedron Lett. 1988, 29, 451. (46) Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans. 1 1975, 1574. (47) (a) Harano, K.; Taguchi, T. Chem. Pharm. Bull. 1972, 20, 2348. (b) Nakai, T.; Ari-Izumi, A. Tetrahedron Lett. 1976, 17, 2335. (48) Ziegler, F. E.; Zheng, Z.-L. Tetrahedron Lett. 1987, 28, 5973. (49) (a) Lee, E.; Lim, J. W.; Yoon, C. H.; Sung, Y.; Kim, Y. K.; Yun, M.; Kim, S. J. Am. Chem. Soc. 1997, 119, 8391. (b) Justicia, J.; Oller-López, J. L.; Campaña, A. G.; Oltra, J. E.; Cuerva, J. M.; Buñuel, E.; Cárdenas, D. J. J. Am. Chem. Soc. 2005, 127, 14911. (c) Yokoe, H.; Mitsuhashi, C.; Matsuoka, Y.; Yoshimura, T.; Yoshida, M.; Shishido, K. J. Am. Chem. Soc. 2011, 133, 8854. (50) Sasano, Y.; Nagasawa, S.; Yamazaki, M.; Shibuya, M.; Park, J.; Iwabuchi, Y. Angew. Chem., Int. Ed. 2014, 53, 3236. (51) Kawanami, Y.; Dainobu, Y.; Inanaga, J.; Katsuki, T.; Yamaguchi, M. Bull. Chem. Soc. Jpn. 1981, 54, 943. (52) Donike, M. J. Chromatogr. 1969, 42, 103. (53) (a) Sharpless, K. B.; Lauer, R. F. J. Am. Chem. Soc. 1972, 94, 7154. (b) Arigoni, D.; Vasella, A.; Sharpless, K. B.; Jensen, H. P. J. Am. Chem. Soc. 1973, 95, 7917. (54) Rationale of the C5/7-site-selectivity was described in the Supporting Information. (55) (a) Kruizinga, W. H.; Strijtveen, B.; Kellogg, R. M. J. Org. Chem. 1981, 46, 4321. (b) Sanna, V.; Pintus, G.; Roggio, A. M.; Punzoni, S.; Posadino, A. M.; Arca, A.; Marceddu, S.; Bandiera, P.; Uzzau, S.; Sechi, M. J. Med. Chem. 2011, 54, 1321.

16429

DOI: 10.1021/jacs.7b10177 J. Am. Chem. Soc. 2017, 139, 16420−16429