Scandium Triflate Catalyzed Nazarov Cyclization of Arylvinyl Epoxides Derived from Alkoxides and Chloro(aryl)carbenes: A Facile Access to Resveratrol-Derived Natural Products

Abstract

The reaction of arylvinyl alkoxides with chloro(aryl)carbenes provided the corresponding arylvinyl epoxides that underwent Nazarov cyclization in a catalytic amount of scandium triflate, providing easy access to several highly substituted indenes, including some resveratrol-derived natural products.

Key words

Nazarov cyclization - arylvinyl alcohol - arylvinyl epoxides - indenes - resveratrol-derived natural products - scandium triflate - chloro(aryl)carbenes

Resveratrol and its oligomers are a highly diverse and privileged class of natural products and are found to exhibit a wide range of biological activities[1] such as antioxidants,[2] anticancer,[3] antidiabetic,[4] cardioprotective,[5] and anti-aging properties.[6] Some Japanese and Chinese folk medicines, which are in a high concentration of resveratrol related compounds, are used to treat ailments related to the liver, skin, heart, and lipid metabolism.[7] The resveratrol monomer is oligomerized to form dimers, trimers, tetramers, and higher-order oligomers, generally up to 8 resveratrol units and these polyphenolic metabolites mainly act as biological defense agents similar to many other secondary metabolites in plants.[8] Pharmacological significance and dimeric skeletons to architecturally complex oligomers from simple resveratrol have been intriguing to the chemical community, thriving to isolate more than 300 resveratrol oligomers having dihydrobenzofuran and indane moieties and bicyclic [3.2.1] and [3.3.0] ring systems.[9] Out of all structural patterns, an indane skeleton containing resveratrol-based natural products (I–XIV, Figure [1]), originated from dimers or substituted dimers of resveratrol, are given particular interest from a synthetic standpoint. Understanding biosynthetic pathways and developing general synthetic strategies that overcome these natural products’ scarcity by isolation were initial aims to address. Synthetic attempts based on the proposed biosynthetic pathways were not satisfactory, often giving unselective and natural and unnatural compounds.[10] However, some synthetic strategies addressed these problems, and some managed to get a common intermediate, thereby accessing a reasonable number of natural products of this class.[11]

Given our ongoing interest in novel electrocyclization precursors,[12] we have reported converting arylvinyl ketones into arylvinyl oxiranes and utilizing them to synthesize various highly substituted indenes.[12e] Further, the synthesis of 3, which was obtained from arylvinyl ketone 1 in 4 steps (Scheme [1]A), served as an advanced intermediate for the total synthesis of resveratrol-based natural products, (±)-isoampelopsin D (I), and (±)-the proposed structure of caraphenol B (V).[12e] Nevertheless, in the present study, we have envisioned that compound 3 could be obtained from 1 via arylvinyl oxirane 2, if successful affording 3 only in two steps, as illustrated in Scheme [1]B. Herein we report a novel method to access indene derivatives, including resveratrol-based natural products, in a highly simplified manner.

Before we embark on precursors that yield resveratrol-derived natural products, we have prepared a model substrate 1a in two steps to evaluate the anticipated outcome. First, reacting 3,5-dimethoxybezaldehyde (7) and (3-methylbut-2-en-2-yl)magnesium bromide followed by oxidation of the resulted 8a afforded 1a (Scheme [2]). Compound 1a was also achieved from compound 9 via compound 10 following our previous report.[12b]

At the outset, treating compound 1a with benzal bromide in a trap solvent system (THF/diethyl ether/pentane, 4:1:1)[13] at –120 °C resulted in 2a with a lower yield (Table [1], entry 1). Whereas temperatures at –110 °C and –78 °C provided little improved yields of 2a (entries 2 and 3). THF or diethyl ether in place of trap solvent diminished the product formation (entries 4 and 5). Conversely, the reaction in CH₂Cl₂ at –78 °C provided 2a in moderate yield (entry 6). In order to further improve the yield in CH₂Cl₂, we have checked with raising the temperature (at 0 °C or 25 °C) and changing the base from n-BuLi to s-BuLi and t-BuLi, but observed only unsatisfactory yields (entries 7–10).

Figure 1 Resveratrol based natural products I–XIV

We found an interesting report in the literature at this stage, converting allylic or benzylic alcohols into an oxirane using dichlorocarbene and chlorophenylcarbene.[14] Inspired by this work, we have envisioned that under similar reaction conditions, 3a could be obtained directly from 8a. If successful, this would further simplify the synthesis of advanced intermediate 3a or its analogues, allowing several indane moiety resveratrol-based natural products. Gratifyingly, the subjection of 8a to benzal chloride (PhCHCl₂)[15] in the presence of KH and KOt-Bu in THF at 0 °C provided 2a in 29% yield (Table [1], entry 11). Two equivalents of benzal chloride under the same conditions doubled the 2a formation (entry 12) and obtained 2a in a similar yield (entry 13), even doubling both benzal chloride and bases (KH and KOt-Bu). Pleasingly, the optimum yield (90%) of 2a was obtained with 2.5 equivalents of benzal chloride, KH, and KOt-Bu (entry 14).[16] However, additional excess equivalents of benzal chloride and bases suppressed 2a yield (62%, entry 15), likely the decomposition of formed product 2a due to excess base.

Scheme 1 (A) Our previous synthetic plan; (B) present plan

Scheme 2 Synthesis of 8a and 1a and converting them into 3a via 2a

With optimized reaction conditions for 2a in hand, next, we turned our attention to the conversion of 2a into 3a, shown in Table [1] (entries 16–23). First, with Brønsted acid TFA, 2a produced 3a in a 27% yield (Table [1], entry 16). The reaction of 2a with Lewis acids, TiCl₄, and BF₃·OEt₂ provided 3a in 50% and 20% yields, respectively (entry 17 and 18). Treatment of 2a with other Lewis acids, Cu(OTf)₂ or AlCl₃, also provided similar yields (entries 19 and 20). However, the subjection of 2a to Sc(OTf)₃ dramatically improved yield, affording 3a in 96% (entry 21).[16] Increasing Sc(OTf)₃ quantity slightly lowered the yields (entries 22 and 23).

Table 1 Optimization of Reaction Conditions for 2a and 3a

Entry	Substrates	Reagent (equiv)	Solvent	Temp (°C)	Time (min)	Yield of 2a and 3a (%)
1	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	tsa	–120	5	2a (10)
2	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	tsa	–110	5	2a (20)
3	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	tsa	–78	5	2a (20)
4	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	THF	–78	60	2a (0)
5	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	Et2O	–78	60	2a (0)
6	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	CH2Cl2	–78	5	2a (40)
7	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	CH2Cl2	0	5	2a (20)
8	1a and PhCHBr2 (1.2)	n-BuLi (1.2)	CH2Cl2	25	5	2a (15)
9	1a and PhCHBr2 (1.2)	s-BuLi (1.2)	CH2Cl2	–78	5	2a (15)
10	1a and PhCHBr2 (1.2)	t-BuLi (1.2)	CH2Cl2	–78	5	2a (10)
11	8a and PhCHCl2 (1.0)	KH/KOt-Bu (1:1)	THF	0	10	2a (29)b
12	8a and PhCHCl2 (2.0)	KH/KOt-Bu (1:1)	THF	0	10	2a (58)b
13	8a and PhCHCl2 (2.0)	KH/KOt-Bu (2:2)	THF	0	10	2a (59)b
14	8a and PhCHCl2(2.5)	KH/KOt-Bu (2.5:2.5)	THF	0	10	2a (90)
15	8a and PhCHCl2(3.0)	KH/KOt-Bu (3.0:3.0)	THF	0	10	2a (62)
16	2a	TFA (0.1)	CH2Cl2	0	10	3a (27)b
17	2a	TiCl4 (0.1)	CH2Cl2	0	10	3a (50)
18	2a	BF3·OEt2 (0.1)	CH2Cl2	0	10	3a (20)
19	2a	Cu(OTf)2 (0.1)	CH2Cl2	0	50	3a (45)b
20	2a	AlCl3 (0.1)	CH2Cl2	0	10	3a (32)b
21	2a	Sc(OTf)3 (0.1)	CH2Cl2	0	45	3a (96)
22	2a	Sc(OTf)3 (0.3)	CH2Cl2	0	45	3a (92)
23	2a	Sc(OTf)3 (0.5)	CH2Cl2	0	45	3a (90)

^a Compounds 1a, 8a, and 2a were used in 1 equiv each in appropriate solvents: trap solvent (ts) THF/diethyl ether/pentane (4:1:1), CH₂Cl₂, and THF in 0.3 M concentration.

^b Based on recovery of the starting material.

Scheme 3 Substrate scope (synthesis of 3b–s). Reagents and conditions: RPhCHCl₂, KH, KOt-Bu, THF, 0 °C; b) Sc(OTf)₃, CH₂Cl₂, 0 °C; c) based on recovery of the starting materials (brsm).

Having had optimum reaction conditions for both key reactions in hand, we have surveyed the substrate scope with various benzal chlorides and different substitutions on arylvinyl alcohols (R¹ and R²), as depicted in Scheme [3]. First, reacting 3,5-dimethoxybenzal chloride with 8a under standard reaction conditions furnished the corresponding product 3b in a 90% yield. Benzal chloride with electron-donating groups/atoms such as –OMe, -SMe, and -F at para position lowered the yields, obtaining 3c–e in 74–76% yield based on the recovery of the starting material (brsm). This implies that chlorophenyl carbenes having electron-donating groups at the para position are less effective than the simple phenyl in this reaction. Furthermore, under standard reaction conditions, electron-withdrawing groups such as –NO₂, –CN on benzal chloride failed to provide the corresponding product 3, it appears that electron-withdrawing groups are detrimental to this reaction. Nonetheless, various alkyl substitutions such as i-Pr, n-Bu, i-Bu, and t-Bu at the para position delivered 3f–i in excellent yields (91–97%). A phenyl substitution containing biphenyl derivative is also equally competitive in providing 3j in 90% yield. Then, the substrate scope with substitutions on the R² position of arylvinyl alcohols 8 was verified. Either alkyl substitutions (Me, Et, and i-Pr) or aryl (Ph) substitution delivered the corresponding indene derivatives 3k–n in good yields. Additionally, substrates without any substitution at R² position were also competitive to deliver expected indene derivatives, affording 3o and 3p in comparable yields. Moreover, tricyclic frameworks 3q and 3r, which are more complex indene derivatives in this series and structurally similar to taiwaniaquinoids,[17] were also achieved from this method. Interestingly, tetracyclic framework compound 3s was also accessed from the corresponding arylvinyl alcohol in good yields (Scheme [3]).

Then, applying this method to synthesize resveratrol-based natural products is planned (Scheme [4]). Accordingly, the advanced intermediates (3t and 3u) were envisioned with the expectation of achieving this class of natural products from a divergent synthetic approach. First, 2t was prepared in 73% yield from reacting 8t under optimized reaction conditions, and the subsequent indene-formation step under optimized conditions resulted in low yields of 3t. Changing the Lewis acid from Sc(OTf)₃ to TiCl₄ surmounted this problem, furnishing 3t in 89% yield. Spectral data of 3t has well resembled with data reported previously,[12e] which is becoming a formal synthesis for (±)-isoampelopsin D (I), and the proposed structure of (±)-caraphenol B (V)/epi-caraphenol B.[11d]

Afterwards, similarly, 3u was prepared in 69% yield from 8u. This advanced intermediate 3u was subjected to DMP to give the corresponding enone 11 in 87% yield. Subsequently, hydrogenation on compound 11 resulted in permethylated epi-caraphenol C 12 in 91% yield. The data of 12 is in good agreement with the data reported in the literature,[11d] which turns into a formal synthesis of the proposed structure of (±)-caraphenol C (VII)/epi-caraphenol C.

In conclusion, we have developed a novel method to highly substituted indenes from arylvinyl alcohols via arylvinyl oxiranes. The method’s scope was demonstrated by preparing several substrates including accessing some advanced intermediates of resveratrol-derived natural products in a highly efficient manner. Further application of this method in a variety of other natural products is in progress.

Scheme 4 Formal synthesis of epi-caraphenol B, isoampelopsine D, and epi-caraphenol C. Reagents and conditions: a) p-MeOPhCHCl₂, KH, KOt-Bu, THF, 0 °C, 73–79%; b) TiCl₄, CH₂Cl₂, –78 °C, 89%; c) DMP, CH₂Cl₂, 0 °C to rt, 2 h, 87%; d) H₂/Pd-C, MeOH, EtOAc, Et₃N, 1 h, 91%.

• References and Notes

• 1 Keylor MH, Matsuura BS, Stephenson R. J. Chem. Rev. 2015; 115: 8976
• 2 He S, Yan X. Curr. Med. Chem. 2013; 20: 1005
• 3a Jang M, Cai L, Udeani GO, Slowing KV, Thomas CF, Beecher CW. W, Fong HH. S, Farnsworth NR, Kinghorn AD, Mehta RG, Moon RC, Pezzuto JM. Science 1997; 275: 218
• 3b Xiao Q, Zhu W, Feng W, Lee SS, Leung AW, Shen J, Gao L, Xu C. Front. Pharmacol. 2019; 9: 1534
• 4a Banks AS, Kon N, Knight C, Matsumoto M, Gutiérrez-Juárez R, Rossetti L, Gu W, Accili D. Cell Metab. 2008; 8: 333
• 4b Milne JC, Lambert PD, Schenk S, Carney DP, Smith JJ, Gagne DJ, Jin L, Boss O, Perni RB, Vu CB, Bemis JE, Xie R, Disch JS, Ng PY, Nunes JJ, Lynch AV, Yang H, Galonek H, Israelian K, Choy W, Iffland A, Lavu S, Medvedik O, Sinclair DA, Olefsky JM, Jirousek MR, Elliott PJ, Westphal CH. Nature 2007; 450: 712
• 5 Frankel EN, Waterhouse AL, Kinsella JE. Lancet 1993; 341: 1103
• 6 Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra A, Prabhu VV, Allard JS, Lopez-Lluch G, Lewis K, Pistell PJ, Poosala S, Becker KG, Boss O, Gwinn D, Wang M, Ramaswamy S, Fishbein KW, Spencer RG, Lakatta EG, Le Couteur D, Shaw RJ, Navas P, Puigserver P, Ingram DK, de Cabo R, Sinclair DA. Nature 2006; 444: 337
• 7 Nonomura S, Kanagawa H, Makimoto A. Yakugaku Zasshi 1963; 83: 988
• 8a Langcake P, Pryce RJ. Physiol. Plant Pathol. 1976; 9: 77
• 8b Langcake P, McCarthy WV. Vitis 1979; 18: 244
• 9 Xiao K, Zhang H.-J, Xuan L.-J, Zhang J, Xu Y.-M, Bai D.-L. In Studies in Natural Products Chemistry, Bioactive Natural Products, Part N, Vol. 34 . Atta-ur-Rahman, Eds.: Elsevier; New York: 2008: 453-646
• 10a Aguirre JM, Alesso EN, Iglesias GY. M. J. Chem. Soc., Perkin Trans. 1 1999; 1353
• 10b Li X.-C, Ferreira D. Tetrahedron 2003; 59: 1501
• 10c Takaya Y, Terashima K, Ito J, He Y.-H, Takeoka M, Yamaguchi N, Niwa M. Tetrahedron 2005; 61: 10285
• 10d Langcake P, Pryce RJ. J. Chem. Soc., Chem. Commun. 1977; 208
• 10e Sako M, Hosokawa H, Ito T, Iinuma M. J. Org. Chem. 2004; 69: 2598
• 10f Wenling L, Li H, Li Y, Hou Z. Angew. Chem. Int. Ed. 2006; 45: 7609
• 11a Klotter F, Studer A. Angew. Chem. Int. Ed. 2014; 53: 2473
• 11b Li W, Yang S, Lv T, Yang Y. Org. Biomol. Chem. 2014; 12: 2273
• 11c Zhong C, Zhu J, Chang J, Sun X. Tetrahedron Lett. 2011; 52: 2815
• 11d Snyder SA, Brill ZG. Org. Lett. 2011; 13: 5524
• 11e Keffrey JL, Sarpong R. Org. Lett. 2009; 11: 5450
• 11f Snyder SA, Breazzano SP, Ross AG, Lin Y, Zografos AL. J. Am. Chem. Soc. 2009; 131: 1753
• 11g Jeffrey JL, Sarpong R. Tetrahedron Lett. 2009; 50: 1969
• 11h Snyder SA, Zografos L, Lin Y. Angew. Chem. Int. Ed. 2007; 46: 8186
• 12a Mahesh SK, Nanubolu JB, Sudhakar G. J. Org. Chem. 2019; 84: 7815
• 12b Sudhakar G, Mahesh SK, Vemulapalli SP. B, Nanubolu JB. Org. Lett. 2017; 19: 4500
• 12c Sudhakar G, Raghavaiah J, Mahesh G, Singarapu KK. Org. Biomol. Chem. 2016; 14: 2866
• 12d Sudhakar G, Reddy KJ, Nanubolu JB. Org. Biomol. Chem. 2015; 13: 8875
• 12e Sudhakar G, Satish K. Chem. Eur. J. 2015; 21: 6475
• 13a Zhao Y, Kang J, Park C.-M, Bagdon PE, Peng B, Xian M. Org. Lett. 2014; 16: 4536
• 13b Hoffmann RW, Brumm K, Bewersdorf M, Mikolaiski W, Kusche A. Chem. Ber. 1992; 125: 2741
• 14 Harada T, Akiba E, Oku A. J. Am. Chem. Soc. 1983; 105: 2771
• 15a Chen J, Lin J.-H, Xiao J.-C. Org. Lett. 2018; 20: 3061
• 15b Kaballa GW, Wu Z. Tetrahedron Lett. 2000; 41: 579
• 15c Han GY, Han PF, Perkins J, McBay HC. J. Org. Chem. 1981; 46: 4695
• 16 Typical Procedure for the Synthesis of 2 To a mixed suspension of potassium hydride (2.5 equiv) and potassium tert-butoxide (2.5 equiv) in THF was added a THF (2.5 mL) solution of 8 (1.0 equiv), and the mixture was stirred at 0 °C under an argon atmosphere for 5 min after that was added a THF (2.5 mL) solution of benzal chloride (2.5 equiv). After completing the starting material, the reaction was quenched with aq NH₄Cl and extracted with EtOAc. The organic layer was washed with aq NaCl, dried over Na₂SO₄, and concentrated under reduced pressure. The crude product was purified by using basic Al₂O₃ column chromatography to isolate 2. Characterization Data of 2a 123.5 mg, 90% yield, dr 1:1 based on ¹H NMR spectroscopy. ¹H NMR (500 MHz, CDCl₃): δ = 7.25–7.14 (m, 5 H), 6.41 (d, J = 2.3 Hz, 2 H), 6.22 (t, J = 2.3 Hz, 1 H), 4.37 (s, 1 H), 3.66 (s, 6 H), 2.08 (s, 3 H), 1.76 (s, 3 H), 1.74 (s, 3 H). ¹³C NMR (100 MHz, CDCl₃): δ = 160.0, 139.1, 135.3, 128.9, 128.7, 127.8, 127.5, 126.7, 105.8, 99.2, 69.6, 68.3, 55.2, 22.5, 20.3, 15.0. IR (neat): ν_max = 2927, 1594, 1454, 1425, 1345, 1202, 1152, 1063, 842, 737, 697. HRMS (ESI): m/z calcd for C₂₁H₂₅O₃ [M + H]: 325.1798; found: 325.1791. Typical Procedure for the Synthesis of 3 To a stirred solution of above-obtained 2 (1.0 equiv) in dry CH₂Cl₂ (3.7 mL) was added Sc(OTf)₃ (0.1 equiv) at 0 °C, and stirring was continued at the same temperature. After completing the starting material, the reaction was quenched with water or saturated aq NaHCO₃ solution and extracted with CH₂Cl₂. The organic layer was washed with aq NaCl solution, dried over Na₂SO₄, filtered, and concentrated under reduced pressure. The crude product was purified by using silica gel column chromatography (EtOAc/hexanes) to give 3. Characterization Data of 3a 118 mg, 96% yield. ¹H NMR (400 MHz, CDCl₃): δ = 7.44 (d, J = 7.6 Hz, 2 H), 7.3 (t, J = 7.6 Hz, 2 H), 7.2 (t, J = 7.6 Hz, 1 H), 6.34 (d, J = 2.0 Hz, 1 H), 6.15 (d, J = 2.0 Hz, 1 H), 5.93 (s, 1 H), 3.73 (s, 3 H), 3.58 (s, 3 H), 2.1 (br s, 1 H), 1.92 (s, 3 H), 1.32 (s, 3 H), 1.30 (s, 3 H). ¹³C NMR (125 MHz, CDCl₃): δ = 159.9, 155.7, 153.2, 143.2, 142.4, 134.2, 131.5, 128.3, 126.9, 125.7, 98.5, 95.2, 69.4, 55.4, 55.1, 50.3, 21.8, 21.2, 9.9. IR (neat): ν_max = 3395, 2956, 2925, 2855, 1595, 1454, 1349, 1203, 1152, 1090, 699. HRMS (ESI): m/z calcd for C₂₁H₂₃O₂ [M – OH]: 307.1698; found: 307.1692.
• 17 Majetich G, Shimkus JM. J. Nat. Prod. 2010; 73: 284

Publication History

Received: 14 September 2020

Accepted after revision: 15 October 2020

Article published online:
13 November 2020