Synthesis 2010(15): 2652-2669  
DOI: 10.1055/s-0029-1218821
FEATUREARTICLE
© Georg Thieme Verlag Stuttgart ˙ New York

Enantioselective Friedel-Crafts Alkylation of Furans and Indoles with Simple α,β-Unsaturated Ketones Catalyzed by Oxazaborolidinone

Shinya Adachi, Fumi Tanaka, Kazuya Watanabe, Atsushi Watada, Toshiro Harada*
Department of Chemistry and Materials Technology, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
Fax: +81(75)7247580; e-Mail: harada@chem.kit.ac.jp;

Further Information

Publication History

Received 17 May 2010
Publication Date:
18 June 2010 (online)

Biographical Sketches

Shinya Adachi was born in Kobe, Japan in 1980. He graduated from Kyoto Institute of Technology in 2003. He received his Ph.D. under the guidance of Prof. Harada­ at Kyoto Institute of Technology in 2010. He is now a postdoctoral fellow in the group of Prof. Mukund P. Sibi at North Dakota State University and working in the field of asymmetric catalysis.
Fumi Tanaka was born in Kochi, Japan and received her B.Sc. in 2007 and M.Sc. in 2009 from Kyoto Institute of Technology. She is now a Research Associate at Yushiro Chemical Industry, Co., Ltd. Kanagawa, Japan.
Kazuya Watanabe was born in Osaka, Japan in 1986 and received his B.Sc. in 2009 from Kyoto Institute of Technology. Currently, he is working as a master course student with Professor Harada at Kyoto Institute of Technology focusing on the development of asymmetric reactions catalyzed by oxazaborolidi­nones.
Atsushi Watada was born in Kobe, Japan in 1987 and received his B.Sc. in 2010 from Kyoto Institute of Technology. Currently, he is working as a master course student with Professor Harada at Kyoto Institute of Technology focusing on the development of asymmetric reactions catalyzed by oxazaborolidi­nones.
Toshiro Harada was born in Kyoto, Japan, in 1952. He graduated from Kyoto University in 1975. He received his Ph.D. from Kyoto University (1980) under the guidance of Professor Zenichi­ Yoshida. Then, he was appointed as a Research Associate in Kyoto Institute of Technology in 1981. He spent the year 1982-1983 as a postdoctoral fellow with Professor W. Clark Still at Columbia University. He was promoted to Full Professor in 2002 in the Department of Chemistry and Materials Technology, Kyoto Institute of Technology. His interests are in the development of chiral Lewis acid catalyzed enantioselective reactions and convergent carbon-carbon bond-forming reactions utilizing carbenoid intermediates.

Abstract

allo-Threonine-derived oxazaborolidinone (OXB) catalyzes the Friedel-Crafts alkylation of furans and indoles with simple acyclic α,β-unsaturated ketones to give the products in high yields and with high enantioselectivities. With 5-10 mol% of the OXB catalyst, enantioselectivities of up to 94% ee could be achieved for a variety of substrates. The use of N,N-dimethylaniline (2.5-10 mol%) as an additive is found to be essential for enantioselectivity. The effects of the additive are discussed in terms of the retardation of a proton-catalyzed racemic pathway, which deteriorates the enantioselectivity of the Friedel-Crafts reactions.

Recently, much attention has been focused on the catalytic enantioselective Friedel-Crafts (F-C) alkylation reaction as an atom-economical method for forming a carbon-carbon bond between electron-rich aromatics and electron-deficient alkenes in a stereodefined manner (Scheme  [¹] ). [¹] Electron-rich heteroaromatics, such as indoles and pyrroles, are ubiquitous constituents in pharmacologically important molecules. The enantioselective F-C alkylation of the heteroaromatics provides a straightforward and efficient means to incorporate these frameworks enantioselectively into the target bioactive molecules or their precursors.

In 2001, Jørgensen and co-workers reported the enantio­selective F-C alkylation catalyzed by a Cu(box)(OTf)2 complex. [²] It was shown that β,γ-unsaturated α-keto esters 1a, activated by the catalyst, react smoothly with indoles and other electron-rich aromatics to give the F-C alkylation products of high enantioselectivities. Following the first demonstration of the catalytic enantioselective F-C alkylation, a number of chiral Lewis acid catalyzed reactions of indoles and pyrroles have been developed for other bidentate chelating substrates, such as alkylidene malonates 2, [³] α,β-unsaturated thioesters 1b, [4] acyl phosphonates 1c, [5] acyl imidazoles 1d, [6] acyl pyridine oxides 1e, [7] α′-hydroxy ketones 1f, [8] and nitroalkenes 3. [9]

The strategy for using such bidentate chelating substrates is that these should be coordinated by a Lewis acidic metal, such as Cu(II), Sc(III), and Pd(II), to form a rigid activated complex 7a-c, in general with an s-cis-anti conformation, ideal for the enantioselective attack of electron-rich aromatics with the aid of well-designed chiral ligands (Scheme  [²] ). In addition, an electron-withdrawing group adjacent to the alkenoyl moiety notably enhances their electrophilic reactivity even prior to the coordination by the metal atoms.

Scheme 1

Although the F-C alkylation products derived from some of these bidentate substrates can be transformed to the corresponding aldehydes, ketones (for 1f), and esters (for 1b-f and 2), the employment of α,β-unsaturated aldehydes 4, ketones 5, and esters 6 would provide more straightforward entry into the adducts bearing such conventional functionalities. However, for such monodentate carbonyl compounds, the control of coordination modes, lone pair discrimination (8a,b vs 8c,d) as well as s-cis, s-trans control (8a,c vs 8b,d), is more difficult and a higher Lewis acidity is required for activation. Therefore, the use of simple nonchelating α,β-unsaturated carbonyl compounds as electron-deficient alkenes represents a synthetic challenge and has been less studied.

Scheme 2

For the enantioselective F-C alkylation reaction with α,β-unsaturated aldehydes, an organocatalytic approach has been revealed to be very effective. The first contribution was presented by MacMillan and co-workers, who demonstrated that α,β-unsaturated aldehydes are enantioselectively activated by chiral amine catalysts in the form of iminium salts, undergoing the enantioselective addition of pyrroles, [¹0a] indoles, [¹0b] and anilines. [¹0c] [¹¹] Other chiral amine catalysts has been also reported. [¹²]

The first example of the F-C alkylation with α,β-unsaturated ketones has been reported by Umani-Ronchi et al. [¹³] It was shown that chiral Al(III)-salen complex catalyzes the reaction of indoles affording the corresponding adducts with enantiomeric excesses in the 80% range for most of the substrates. Recently, improved enantioselectivity has been reported by Pedro et al. by using a chiral Zr(IV)-BINOL catalyst. [¹4] The catalyst system is specifically effective for alk-1-enyl aryl ketones. At the 20 mol% catalyst loading, the reaction of pyrroles and indoles afforded the products in high yields and high enantioselectivities. Very recently, highly enantioselective reactions for chalcones (ArCH=CHCOAr′) catalyzed by a chiral Sc(III)-N,N′-dioxide complex (2-10 mol%) has been reported by Feng and co-workers. [¹5] Organocatalytic examples of the F-C alkylation of indoles and pyrroles with enones via ketiminine salts has also been reported. [¹6]

Furans also represent an important class of electron-rich five-membered heterocycles that are broadly found as structural motifs of many natural products and pharmaceutically important substances. [¹7] Moreover, they can be employed as useful intermediates in synthetic chemistry. [¹8-²0] Hence, the synthesis and transformation of furans have attracted significant attention. [²¹]

While furans are less nucleophilic than indoles and pyrroles, [²²] they do undergo the F-C alkylation with α,β-unsaturated ketones by the catalysis of Brønsted acids or Lewis acids to give racemic products. [²³] However, the catalytic enantioselective version of the furan F-C alkylation is much less developed than that of indoles and pyrroles. In their study on the copper-catalyzed enantioselective F-C alkylation, Jørgensen and co-workers reported a few examples of the reaction of 2-methylfuran with β,γ-unsaturated α-keto esters. [²b] MacMillan et al. reported an organocatalytic enantioselective cascade process in which 2-methylfuran was added to the β-position of an α,β-unsaturated aldehyde, via an iminium mechanism, followed by electrophilic chlorination of the resulting enamine intermediate. [¹¹a] Very recently, the F-C reaction of furans with highly reactive ethanetricarboxylates has been described, but with limited results. [³f] The enantioselective reaction of highly nucleophilic 2-methoxyfuran with α,β-unsaturated 2-acyl imidazoles [²4] and nitroalkenes [²5] has been reported. To date, however, the scope of the reaction with respect to both furan derivatives and electron-deficient alkenes remains to be expanded.

We recently reported that allo-threonine-derived oxazaborolidinones (OXB) 9 are efficient catalysts for the enantioselective Michael reaction [²6] and Diels-Alder reaction [²7] of acyclic α,β-unsaturated ketones. [²8] The high enantioselectivities and the absolute stereochemical course of the reactions with unsaturated ketones are rationalized in terms of the common activated complex model 10 in which the substrate coordinates to the boron atom in an s-cis-anti fashion, allowing to be attacked by the nucleophile on the open Si (front) face.

Figure 1 Structures of 9 and 10

The present study was aimed at developing the reaction between the less electrophilic α,β-unsaturated ketones and the less nucleophilic heteroaromatics such as furans in order to expand the scope of the synthetically useful, catalytic enantioselective F-C alkylation. For this purpose, attention was focused on the characteristic features of the OXB catalysts for the enantioselective activation of ketonic carbonyl groups.

We herein report the full details of the development of an OXB-catalyzed enantioselective F-C alkylations of furans and indoles with simple α,β-unsaturated ketones. [²9] Application to the enantioselective synthesis of γ-keto carboxylic acids and a tetrahydropyrano[3,4-b]indole is also described. The effect of an amine additive is discussed in terms of the retardation of a proton-catalyzed racemic pathway, which deteriorates the enantioselectivity of the F-C alkylation reaction.

F-C Alkylation of Furans

We recently reported that furan (11a) undergoes enantio­selective Diels-Alder reaction with alkyl vinyl ketones in the presence of OXB 9a (10 mol%) at -78 ˚C in toluene (Equation  [¹] ). [²7b] Under similar conditions, the reaction of 2-methylfuran (11b) with vinyl ketone 12a did not afford the Diels-Alder adduct, but the F-C product 13ba in 8% yield (Equation  [²] ). The observation prompted us to examine the possibility of asymmetric induction at the β-position of F-C products in the reaction with crotyl ketone 12b.

Equation 1

Equation 2

Initial examination of the reaction of 12b with 11b (5 equiv) in the presence of OXB 9a (10 mol%) resulted in the nonenantioselective formation of the corresponding F-C product 13bb irrespective of the solvents employed (Table  [¹] , entries 1-3). Further examination revealed the remarkable effect of amine additives. In the presence of an amine (5-10 mol%), 13bb was obtained in an enantiomerically enriched form (entries 4-16). Of various aniline and pyridine derivatives examined, N,N-dimethylaniline (5 mol%) was superior, exhibiting the best enantioselectivity (69% ee) and product yield (96%) at -40 ˚C in Et2O (entry 7). When the amount of N,N-dimethylaniline was reduced to 2.5 mol%, the enantioselectivity was decreased considerably (entry 17). On the other hand, increasing the amount of the additive to 7.5 and 10 mol% brought about an improved enantioselectivity of 77% and 80% ee, respectively (entries 18 and 19). Further increase (20 mol%), however, resulted in a decrease in the product yield at -40 ˚C (entry 21) and led to a decline in the enantioselectivity at -20 ˚C (entry 22). Although, with 10 mol% of N,N-dimethylaniline 90% ee was obtained at -78 ˚C, the reaction was very sluggish (entry 20).

Table 1 Enantioselective F-C Alkylation of Furan 11b with Enone 12b Catalyzed by OXB 9a a
Entry Solvent Additive (mol%) Yield (%) ee (%)
 1b,c toluene - 69  1
 2b EtCN - 48  4
 3 Et2O - 97  2
 4 toluene PhNMe2 (5) 97 44
 5 CH2Cl2 PhNMe2 (5) 91 20
 6 EtCN PhNMe2 (5) 84 55
 7 Et2O PhNMe2 (5) 96 69
 8 t-BuOMe PhNMe2 (5) 50 57
 9 Et2O 4-MeC6H4NMe2 (5) 10 64
10 Et2O 4-BrC6H4NMe2 (5) 73 44
11 Et2O PhNEt2 (5)  2 30
12 EtCN PhN(i-Pr)2 (5) 26 14
13 EtCN 2,6-lutidine (5) 72 32
14 Et2O 2,6-lutidine (5)  7 22
15 EtCN 2,4,6-collidine (5) 21 27
16 Et2O DTBPd (10) 12 22
17 Et2O PhNMe2 (2.5) 77  6
18 Et2O PhNMe2 (7.5) 98 77
19 Et2O PhNMe2 (10) 92 80
20e Et2O PhNMe2 (10)  2 90
21 Et2O PhNMe2 (20) 23 82
22f Et2O PhNMe2 (20) 82 75

a Unless otherwise noted, reactions were carried out by using 12b (2.0 mmol, 0.3 M), 11b (5.0 equiv), and OXB 9a (10 mol%) at -40 ˚C for 18 h. b The reaction was carried out for 2 h. c The reaction was carried out at 0 ˚C.
d 2,6-Di-tert-butylpyridine.
e The reaction was carried out at -78 ˚C.
f The reaction was carried out at -20 ˚C.

Equation 3

The scope of the OXB-catalyzed F-C reaction of furans was examined in Et2O at -40 ˚C in the presence of N,N-dimethylaniline (Equation  [³] , Table  [²] ). In addition to methyl­ derivative 11b, the reaction worked well with other 2-substituted furans 11c-e (Table  [²] , entries 1-4). The reactions of the 2-ethyl, 2-butyl, and 2-benzyl derivatives with enone 12b afforded the corresponding F-C products with 77-85% ee and in high yields. Of these, benzyl derivative 11e was somewhat less reactive and 20 mol% of the catalyst was required. 2,3-Dimethylfuran (11f) exhibited high enantioselectivity (91% ee) in the reaction with enone 12b (entry 5). On the other hand, the parent furan (11a) hardly reacted with 12b (entry 6).

Table 2 Enantioselective F-C Alkylation of Furans 11 with Enones 12 Catalyzed by OXB 9a a
Entry Furan R³ R4 Enone R¹ R² Product Yield (%) ee (%)b
1 11b Me H 12b Me Et 13bb 92 80
2 11c Et H 12b Me Et 13cb 95 77
3 11d Bu H 12b Me Et 13db 96 79
4c 11e Bn H 12b Me Et 13eb 62 85
5 11f Me Me 12b Me Et 13fb 98 91
6 11a H H 12b Me Et 13ab trace nd
7 11b Me H 12c Me Pr 13bc 79 81
8 11b Me H 12d Me Bu 13bd 92 86
9 11f Me Me 12d Me Bu 13fd 89 93
10c 11f Me Me 12e Bu Me 13fe 85 87
11d 11b Me H 12f CO2Et Me 13bf 81 85
12 11f Me Me 12f CO2Et Me 13ff 99 89
13e 11b Me H 12g Ph Me 13bg  0 -

a Unless otherwise noted, reactions were carried out by using 12 (2.0 mmol), 11 (5.0 equiv), PhNMe2 (10 mol%), and OXB 9a (10 mol%) in Et2O (3.1 mL) at -40 ˚C for 20 h.
b Determined by HPLC on a chiral stationary phase; nd = not determined. c The reaction was carried out with 20 mol% of OXB 9a and PhNMe2.
d The reaction was carried out for 44 h.
e Enone 12g was recovered in 98% yield.

Aliphatic enones other than 12b, such as crotyl alkyl ketones 12c,d and hex-1-enyl methyl ketone (12e) also serves as an acceptor, affording the F-C alkylation products of furans 11b and 11f with 81-93% ee and in high yields (entries 7-10). Enone 12f bearing an ethoxycarbonyl group at the β-position was also a good substrate (entries 11 and 12). The reaction of furan derivatives 11b and 11f gave the corresponding product 13bf and 13ff in 85% ee and 89% ee, respectively, in high yields. On the other hand, attempted reaction with benzalacetone (12g) resulted in the recovery of the enone (entry 13).

Enantioselective Synthesis of γ-Keto Carboxylic Acids­

The enantioselective conjugate addition of a hydroxycarbonyl anion equivalent to α,β-unsaturated ketones is one of the most straightforward methods for the preparation of enantioenriched γ-keto carboxylic acids (Scheme  [³] ). [³0] The use of cyanide as a hydroxycarbonyl anion equivalent might be less attractive for this purpose because of possible racemization in the transformation of the cyano group to the carboxy group. [³¹] Furans can be oxidatively transformed to carboxylic acids under relatively moderate conditions [³²] and could serve as a hydroxycarbonyl anion equivalent. In order to utilize the present OXB-catalyzed enantioselective F-C alkylation of furans in the preparation of enantioenriched γ-keto carboxylic acids, oxidative transformation of the alkylation product 13 was examined (Equation  [4] ).

Scheme 3

Equation 4

Table 3 Oxidation of 13 with RuCl3˙H2O and NaIO4 a
Entry Furan F-C product 13
Yield (%) eeb (%)
R4 R¹ R²
1 13bc H Me Pr 14a 96 82
2 13fb Me Me Et 14b 81 92
3 13fd Me Me Bu 14c 89 93
4 13fe Me Bu Me 14d 85 88

a The reactions of 13 were carried out with RuCl3˙H2O (1 mol%) and NaIO4 (15 equiv) in a mixture of CH2Cl2, MeCN, and H2O at r.t. for 4 h.
b Enantioselectivities were determined by HPLC on a chiral stationary phase after conversion into the corresponding benzylamide.

Oxidation of 13bc (81% ee) with RuCl3˙H2O (1 mol%) and NaIO4 (15 equiv) in CH2Cl2, MeCN, and H2O [³²] went to completion in four hours to give the γ-keto acid 14a (R¹ = Me, R² = Pr) in 96% yield (Table  [³] , entry 1). The enantiopurity of 14a was determined to be 82% ee by a chiral phase HPLC analysis of benzylamide derivative, establishing that no racemization occurs during the oxidation. The absolute structure of 14a was determined to be S by comparison of the optical rotation sign with literature data. [³0] The F-C alkylation products of 2,3-dimethylfuran were obtained in higher enantioselectivity than those of 2-methylfuran. Under similar conditions, oxidation of these products 13 afforded γ-keto acids 14b-d of high ee (88-93% ee) in high yields (entries 2-4).

F-C Alkylation of Indoles

Catalysis by OXB was examined also for the F-C alkylation of indoles with α,β-unsaturated ketones because of the synthetic importance of the transformation. Attempts were made to optimize the conditions for the reaction of indole (15a) with enone 12b in the presence of catalyst 9a (10 mol%) (Equation  [5] , Table  [4] ).

Table 4 Enantioselective F-C Alkylation of Indole with Hex-4-en-3-one Catalyzed by OXB 9a a
Entry Temp (˚C) Methodb Additive (mol%) Yield (%) ee (%)
 1c -40 A - 83 68
 2c -85 A - 90 78
 3c,d -85 A - 90 87
 4 -85 A PhNMe2 (2.5) 92 87
 5 -85 A PhNMe2 (5) 56 87
 6e -85 A PhNMe2 (7.5) 27 82
 7 -85 B - 96 77
 8 -40 B - 85 60
 9 -85 B PhNMe2˙HCl (2.5) 92 86
10 -85 B AcOH (2.5) 47 29
11 -85 B CF3CO2H (2.5) 61 25

a Unless otherwise noted, reactions were carried out by using 12b (2 mmol), 15a (1 equiv), and OXB 9a (10 mol%) in Et2O (3.1 mL) at -85 ˚C for 20 h. b OXB 9a was prepared by treatment of 18 with PhBCl2 followed by removal of HCl in vacuo (Method A) or by treatment with allyltri­methylsilane (4 equiv) for 14-17 h at r.t. (Method B).
c By-product 17 was obtained in 3% yield.
d A mixture of 15a and 9a was stirred for 24 h at -85 ˚C before the addition of 12b.
e The reaction was carried out for 32 h.

Equation 5

In contrast to the reaction of furans, F-C product 16ab was obtained in relatively high enantioselectivity (68% ee) even without an amine additive at -40 ˚C (Table  [4] , entry 1). The reaction of a stoichiometric amount of 15a went to completion in 20 h at -85 ˚C to give 16ab in 78% ee (entry 2), demonstrating a higher reactivity of indoles compared to furans. [²²] When a mixture of 15a and catalyst 9a was stirred for 24 hours before the addition of 12b, the enantioselectivity was improved to 87% ee (entry 3). In the presence of 2.5 mol% of N,N-dimethylaniline, high enantioselectivity comparable to entry 3 was obtained (entry 4). Unlike the reaction of furans, further increase of the additive resulted in a decrease in product yields (entries 5 and 6). In the reactions without the amine additive (entries 1-3), indole dimer 17 was formed as a by-product in 3% yield while the formation of 17 was not detected in the presence of the additive (entries 4-6).

OXB catalyst 9a was prepared prior to use by treatment of N-tosylamino acid 18 with dichlorophenylborane followed by removal of hydrogen chloride in vacuo (Method A) (Equation  [6] ). Hydrogen chloride, contaminated with the catalyst, might be responsible for the formation of indole dimer 17. [³³] Thus, 17 might be formed by a pathway involving protonation of indole, the attack of another indole to the resulting iminium intermediate, and subsequent proton migration leading to dimer salt 17˙HCl (Scheme  [4] ). Judging from the yields of 17 in entries 1-3, the residual hydrogen chloride is estimated to be about 3 mol%, well consistent with the minimum amount (2.5 mol%) of the amine additive required to suppress the formation of 17.

Equation 6

To avoid contamination by hydrogen chloride, OXB 9a, prepared according to Equation  [6] , was treated with allyl­trimethylsilane (4 equiv) at room temperature for 14-17 hours and used after concentration in vacuo (Method B). Under these conditions, the formation of dimer 17 was not detected, confirming the complete removal of hydrogen chloride (entry 7). F-C alkylation product 16ab was obtained in high yield, but with lower enantioselectivity (77% ee). The enantioselectivity was comparable to that in the reaction with Method A in the absence of the amine additive (entry 2), suggesting that the residual hydrogen chloride is not influential to the lower enantioselectivity.

Scheme 4

Equation 7

In the presence of hydrogen chloride, N,N-dimethyl­aniline forms a hydrochloride salt. Therefore, the effect of the salt was then examined. When the reaction was carried out with catalyst 9a prepared by Method B in the presence of N,N-dimethylaniline hydrochloride (2.5 mol%), 16ab was obtained in 86% ee (entry 9). The result shows that N,N-dimethylaniline exerted the advantageous effect in the form of the hydrochloride salt. High enantioselectivity was also obtained in the absence of N,N-dimethylaniline when catalyst 9a (Method A) was used after treatment with indole (entry 3). Under these conditions, hydrochloride salt of indole dimer (17˙HCl) might serve as an additive instead of N,N-dimethylaniline hydrochloride. Examination of other Brønsted acids such as acetic acid and trifluoroacetic acid resulted in lowering of the yield and the enantioselectivity (entries 10 and 11).

The scope of the OXB-catalyzed F-C alkylation was explored with respect to indole architecture in the reaction with hex-4-en-3-one (12b) under the optimized conditions using 10 mol% of 9a (Method A) in the presence of 2.5 mol% of N,N-dimethylaniline (Equation  [7] ). Indole derivatives 15b,c bearing electron-donating substituents at the 5-position also gave the corresponding adducts 16 enantioselectively in high yields (Table  [5] , entries 3 and 4). While a chlorine and bromine substituent at the 5 position lowered conversion, a good level of enantioselectivity was retained for 15d,e (entries 5 and 6). In the organocatalytic F-C alkylation of indoles, it has been observed that substitution on the nitrogen atom had a detrimental effect on both reactivity and enantioselectivity. [¹6a] The OXB-catalyzed reaction has no such limitation, being applicable to N-substituted indoles 15f-h (entries 7-9). Of these, N-methyl derivative 15f exhibited enhanced selectivity of 93% ee in comparison with parent indole. Sterically hindered 2-methyl derivatives 15i-k also underwent the F-C alkylation reaction without lowering the reactivity and selectivity (entries 10-12).

To further investigate the scope of the reaction, the reactions of indole (15a) and 2-methylindole (15i) were carried out with a variety of α,β-unsaturated ketones. The reaction of 15a with crotyl butyl ketone (12d) proceeded with high enantioselectivity (92% ee) and in good yield (entry 13). In comparison with crotyl ketones 12b,d, enones 12e,g, bearing a butyl and a phenyl group at the β position, were less reactive. The F-C alkylation of 15a resulted in low or poor product yield (entries 14 and 17). However, the F-C reaction of 15i proceeded efficiently with such less reactive enones. Not only 12e but also benzalacetone derivatives 12g,h gave the corresponding F-C alkylation products 16 in high yields and high selectivities (entries 15, 18, and 19). Functionalized enones, such as ethoxycarbonyl derivative 12f, (benzyloxy)methyl derivative 12i, and, di(methoxy)methyl derivative 12j, could be used as acceptors of the present F-C alkylation (entries 16, 20, and 21). Of these, 12f and 12j exhibited superior enantioselectivities. OXB 9a did not exhibit high selectivity in the reaction with phenyl ketone 12k and cyclic enone 12l (entries 22 and 23). Such structural limitation of enones was observed also in the OXB-catalyzed Michael reaction [²6] and Diels-Alder reaction. [²7]

One of the characteristic features of the present indole F-C alkylation reaction is that it can be carried out at relatively low catalyst loadings. A 10 mol% of catalyst 9a is enough to obtain the results shown in Table  [5] . Further reduction of the amount of the catalyst is possible by the elongation of the reaction time. Thus, when the reaction of 15a with 12b was carried out at a 5 mol% catalyst loading for 48 hours, 16ab was obtained in high yield without lowering enantioselectivity (entry 2).

Catalysis by OXB could be applied successfully to intramolecular F-C alkylation [6a] [³4] as exemplified in enan­tioselective synthesis of tetrahydropyranoindole 20 (Equation  [8] ). Although indolyl enone 19 was not fully soluble in cold Et2O, its treatment with 9a (20 mol%) in the presence of N,N-dimethylaniline for five days gave 20 with 90% ee.

Equation 8

The absolute stereochemistry of indole F-C products 16ab, [¹6a] 16af, [¹6a ] and 16ik [¹³a] were determined to be S by the comparison of their optical rotation signs with literature data. The selectivity for S-enantiomers, observed commonly in the F-C alkylation of furans and indoles, is consistent with activated complex 10, thus ascertaining further the validity of the model.

Table 5 Enantioselective F-C Alkylation of Indoles 15 with Enones 12 Catalyzed by OXB 9a a (continued)
Entry Indole R³ R4 R5 Enone R¹ R² Product Yield (%) ee (%)b
 1 15a H H H 12b Me Et 16ab 92 87
 2c 15a H H H 12b Me Et 16ab 90 87
 3 15b H H MeO 12b Me Et 16bb 85 89
 4 15c H H Me 12b Me Et 16cb 73 83
 5d 15d H H Cl 12b Me Et 16db 55 85
 6d 15e H H Br 12b Me Et 16eb 23 85
 7 15f Me H H 12b Me Et 16fb 93 93
 8 15g allyl H H 12b Me Et 16gb 93 85
 9 15h Bn H H 12b Me Et 16hb 52 75
10 15i H Me H 12b Me Et 16ib 91 87
11 15j Me Me H 12b Me Et 16jb 83 81
12 15k H Me MeO 12b Me Et 16kb 73 75
13 15a H H H 12d Me Bu 16ad 81 92
14 15a H H H 12e Bu Me 16ae 19 72
15d 15i H Me H 12e Bu Me 16ie 90 82
16 15a H H H 12f EtO2C Me 16af 96 94
17d 15a H H H 12g Ph Me 16ag 2 nd
18d 15i H Me H 12g Ph Me 16ig 87 84
19d 15i H Me H 12h 4-FC6H4 Me 16ih 81 81
20e 15i H Me H 12i BnOCH2 Me 16ii 70 75
21e 15i H Me H 12j (MeO)2CH C5H11 16ij 85 92
22 15i H Me H 12k Me Ph 16ik 52 37

23e 15i H Me H 12l cyclohex-2-enone
16il 47 55

a Unless otherwise noted, reactions were carried out by using 12 (2 mmol), 15 (1 equiv), PhNMe2 (2.5 mol%), and 9a (10 mol%) in Et2O (3.1 mL) at -85 ˚C for 20 h.
b Determined by HPLC on a chiral stationary phase; nd = not determined. c The reaction was carried out with 9a (5 mol%) without the addition of PhNMe2 for 48 h. The catalyst prepared by Method A was used after treatment with 15a for 24 h at -85 ˚C.
d The reaction was carried out for 96 h.
e The reaction was carried out for 48 h.

Effect of Amine Additive

According to the generally accepted mechanism for the Lewis-acid catalyzed F-C alkylation of electron-rich heteroaromatics, [8b] [9b] [³5] the OXB-catalyzed reaction proceeds by initial complexation of enone 12 with catalyst 9a followed by the attack of a heteroaromatic (ArH) to the resulting activated complex 21 to form zwitterionic intermediate 22 (Scheme  [5] ). Then, a proton attached to the heteroaromatic ring of intermediate 22 needs to undergo 1,3-migration to an enolate carbon to furnish F-C alkylation product 13 or 16 with regeneration of the catalyst. The proton migration may proceed either by a concerted mechanism or by a stepwise mechanism involving deprotonation/protonation sequence. As a proton-catalyzed racemic background reaction is likely to degrade the overall enantioselectivity of the OXB-catalyzed reaction (vide infra­), it is essential to gain mechanistic information on the proton migration step.

Scheme 5

For this purpose, reactions were carried out by using 1-methyl-3-deuteroindole (15f-d; 90%-d). The reaction of 15f-d with 12b in the presence of 9a (Method A) and N,N-dimethylaniline afforded F-C product 16fb-d in 86% yield and with 85% ee (Equation  [9] ). In accord with the reaction pathway in Scheme  [5] , ¹H NMR analysis of the product showed that a deuterium was incorporated exclusively at the α position in 84%-d. Compound 16fb-d was obtained as a 87:13 mixture diastereomers. The major isomer was assigned as syn-16fb-d based on a smaller vicinal coupling constant (J = 6.2 Hz) relative to the minor anti-isomer (J = 8.0 Hz) for the α proton of 16fb (Figure  [²] ).

Equation 9

Figure 2  syn/anti-Vicinal couplings in 16fb

A crossover experiment was then examined with a mixture of 15f-d (0.5 equiv) and 1-allyl derivative 15g (0.5 equiv) under similar conditions for 20 hours (Equation  [¹0] ). The reaction afforded H/D scrambled products 16fb-d and 16gb-d enriched with 72% and 22% deuterium, respectively, as determined by ¹H NMR analysis (Table  [6] , entry 1). Diastereoselective formations of syn-isomers were observed for both products. In another experiment, reaction was quenched before completion after four hours (entry 2). Under these conditions, H/D scrambling was observed for products 16fb-d (65%-d) and 16gb-d (58%-d) [³6] but not for the recovered indoles 15f-d (90%-d) and 15g (0%-d), excluding the possibility of a prior H/D scrambling between the substrates. In addition, scrambling between the F-C products was ruled out by the absence of d 2-products as confirmed by mass spectrometric analyses. Extensive H/D scrambling was also observed when a crossover reaction was carried out with 9a prepared by Method B in the absence of the amine additive (entry 3).

Equation 10

Table 6 Crossover Experiments with 15f-d and 15g a
Entry Product Yield (%) ee (%) d (%)b syn/anti b
1 16fb-d 47 91 72 85:15
16gb-d 48 84 22 86:14
2c 16fb-d 32 87 65 86:14
16gb-d 10 85 58 84:16
3d 16fb-d 41 78 57 87:13
16gb-d 27 81 43 88:12

a Unless otherwise noted, reactions were carried out with 12b, 15f-d (0.5 equiv), 15g (0.5 equiv), PhNMe2 (2.5 mol%), and 9a (10 mol%, Method A) in Et2O at -85 ˚C for 20 h.
b Determined by ¹H NMR spectroscopy.
c Reaction was carried out for 4 h.
d Reaction was performed with 9a prepared by Method B in the absence of PhNMe2.

The crossover experiments clearly showed that the 1,3-proton migration in 22 did not proceed by a concerted mechanism. The result is compatible with a stepwise mechanism involving deprotonation of 22 by solvent Et2O to form an ion pair of enol boronate 23 and Et2O+H (Scheme  [6] ). Subsequent protonation of 23 leading to the F-C products 13 or 16 (pathway a) would occur, at least partially, in the solvent cage via an initially produced intimate ion pair. Hence, the observed extensive H/D scrambling is difficult to be rationalized by such direct pathway. It is more probable that 23 undergoes protonation at the carbonyl oxygen atom to form relatively stable zwitterion 24, [³7] which furnishes the F-C products with H/D scrambling through a bimolecular mechanism (pathway b). [³8] The selective formation of the syn-diastereomers in Equations  [9] and  [¹0] is well consistent with the deuteration at the less hindered side, anti to the indolyl group of 24.

Scheme 6

In the OXB-catalyzed F-C alkylations of indoles and furans, N,N-dimethylaniline as an additive is indispensable to obtain high enantioselectivity. In the reaction of indole, the enantioselectivities were about 10% lower without the amine additive irrespective of Method A and B for the preparation of OXB 9a (Table  [4] , entries 2 and 7). The effect of the additive was more prominent in the furan F-C alkylations. In the absence of the additive, 9a prepared by Method A, exhibited no enantioselectivity in the reaction of 2-methylfuran with 12b (Table  [¹] , entry 3). Under similar conditions, the use of 9a, prepared by Method B and not contaminated with hydrogen chloride, also resulted in the formation of racemic (3% ee) product 13ab in 61% yield. It is very likely that the racemic product was produced through a proton catalyzed pathway via intermediate 25 which is formed by protonation of enone 12 with 24 (Scheme  [6] ). In the indole F-C alkylations without the amine additive, the proton-catalyzed racemic pathway proceeded in competition with the OXB-catalyzed enan­tio­selective pathway, degrading the overall selectivity of the reaction.

The optimal amounts of the amine additive for the F-C alkylation of indoles and furans are different, being 2.5 and 10 mol%, respectively. As described above, OXB 9a prepared by Method A was contaminated with ca. 3 mol% of hydrogen chloride. Accordingly, in the reaction of indoles, the amine exerted beneficial effect in the form of the hydrochloride salt. On the other hand, in the reaction of furans with 10 mol% of N,N-dimethylaniline, a major portion of the additive remained as a free base. The poor selectivity (6% ee) observed with 2.5 mol% additive (Table  [¹] , entry 17) suggested that the hydrochloride salt is not much effective for furans. Indeed, when the reaction of 2-methylfuran (11b) with enone 12b was carried out in the presence of an increased amount of N,N-dimethyl­aniline hydrochloride (35 mol%) with catalyst 9a (10 mol%) prepared by Method B (-40 ˚C, 20 h), F-C product 13ab was obtained in 85% yield and in 19% ee. The low selectivity, while being slightly better than that obtained in the absence of the additive, shows that the effect of the amine salt is rather small for furans.

The role of N,N-dimethylaniline hydrochloride in the reaction of indoles could be understood as a catalyst for the conversion of zwitterion intermediate 24 into the F-C product 16 [Scheme  [7] (a)]. Thus, 24 would react with the hydrochloride salt by a protonation (at the enolate carbon)/deprotonation (of the hydroxy group) sequence to give 16 with regeneration of catalyst 9a and the amine salt. [³9] The catalytic protonation/deprotonation sequence would prevent the accumulation of 24, which could induce the proton catalyzed racemic pathway via 25. The effect of free N,N-dimethylaniline, predominant for furans, could also be rationalized in terms of a catalyst for proton transfer [Scheme  [7] (b)]. For the free amine, deprotonation 24 precedes protonation of the resulting enol boronate 23.

Scheme 7

In the reactions of indole (15a) with 13b, the product yields were decreased considerably by the increase of the amine additive (5-7.5 mol%) (Table  [4] , entries 5 and 6). The results imply that the free amine retards the OXB-catalyzed reaction as well, presumably owing to the reversible formation an amine-OXB complex. [40] In the furan F-C alkylation, 10 mol% of N,N-dimethylaniline efficiently retarded the proton-catalyzed pathway (Table  [¹] , entry 19) although, under these conditions, the free amine might partially deactivate the OXB catalyst by complexation. Further increase of the amine additive led to a considerable decrease in reaction rate by the almost exhaustive poisoning of the catalyst (entry 21).

Owing to the lower nucleophilic reactivity, [²²] higher temperature (-40 ˚C) was required in the reaction of furans. Under these conditions, the proton-catalyzed racemic pathway might become predominant. This could be the reason why increased amount of the amine additive was necessary to achieve high enantioselectivity. An increase in the portion of the racemic pathway was observed also in the reaction of indoles at the higher temperature. Thus, when the reaction of 15a with 12b was carried out at -40 ˚C, 16ab was obtained only in moderate enantioselectivity (Table  [4] , entries 1 and 8).

The OXB catalysis is demonstrated to be effective in the enantioselective F-C alkylation of electron-rich hetero­aromatics. The enantioselective F-C alkylation of furans with a monodentate α,β-unsaturated ketones is realized for the first time by OXB catalysis. In addition, the F-C products of furans are successfully converted into the corresponding γ-keto carboxylic acids without the loss of enantiopurity. The catalyst system is also applied to the F-C alkylation of indoles, expanding the scope of substrates. The observed absolute structures of the F-C products are consistent with the activated complex model 10 proposed previously for relevant Michael reaction and Diels-Alder reaction. The use of N,N-dimethylaniline (2.5 mol% for indoles and 10 mol% for furans) as an additive is found to be essential to obtain high enantioselectivity. The effect of the additives is discussed in terms of the retardation of a proton-catalyzed racemic pathway, which deteriorates the enantioselectivity of the F-C alkylation reaction.

¹H (500 MHz) and ¹³C NMR (126 MHz) spectra were recorded on a Bruker DRX-500 spectrometer. Spectra were referenced to residual CHCl3 (δ = 7.26) for ¹H resonances and CDCl3 (δ = 77.0) for ¹³C resonances. MS measurements were conducted with a JEOL JMS-700 instrument. Optical rotations were measured with Horiba SEPA-300 Polarimeter. HPLC analyses were performed on a Beckman Gold system. All reactions were performed under an inert atmosphere of dry argon. Flash chromatography was carried out with silica gel C-300 of Wako Pure Chemical Industries, Ltd. CH2Cl2 and EtCN were dried and distilled over CaH2. Et2O and toluene were distilled from sodium benzophenone ketyl. The following compounds were prepared according to a literature procedure; 11d, [] 11e, [] 12c,k, [] 12e, [44] 12f, [45] 12h, [46] 12i, [47] 12j, [48] 15g, [5a] 15h, [49] 15f-d, [50] 18, [²6a] and N,N-dimethylaniline hydrochloride. [] Dichlorophenylborane and N,N-dimethylaniline were used after distillation. Furans 11a,b,f were dried and distilled over sodium before use. Furans 11c-e were passed through basic alumina and distilled before use.

F-C Alkylation of Furans; ( S )-5-(5-Methylfuran-2-yl)hexan-3-one (13bb); Typical Procedure (Table 2, Entry 1)

OXB 9a (0.20 mmol) was prepared as follows (Method A). To a solution of O-benzoyl-N-tosyl-l-allo-threonine (18; 75.5 mg, 0.20 mmol) in anhyd CH2Cl2 (2.0 mL) under argon at r.t. was added dichlorophenylborane (28.5 µL, 0.220 mmol). After stirring for 30 min, the mixture was concentrated in vacuo. To a mixture of the resulting 9a in anhyd Et2O (3.1 mL) at -40 ˚C were added 2-methylfuran (11b; 821 mg, 10 mmol), N,N-dimethylaniline (24.2 mg, 0.20 mmol), and hex-4-en-3-one (12b; 196 mg, 2 mmol). After stirring at -40 ˚C for 18 h, the reaction mixture was quenched with sat. aq NaHCO3 (10 mL) and filtered. The filtrate was extracted with Et2O (3 × 20 mL), the combined organic layers were dried (Na2SO4), and concentrated in vacuo. Purification of the residue by flash chromatography (SiO2, gradient elution with 1-2% Et2O in hexane) gave 332 mg (92%) of 13bb (80% ee); [α]D ²8 +4.98 (c 1.33, CHCl3).

¹H NMR (500 MHz, CDCl3): δ = 1.03 (3 H, t, J = 7.3 Hz), 1.21 (3 H, d, J = 6.9 Hz), 2.23 (3 H, s), 2.38 (2 H, q, J = 7.3 Hz), 2.51 (1 H, dd, J = 8.0, 16.2 Hz), 2.79 (1 H, dd, J = 6.0, 16.2 Hz), 3.34 (1 H, sext, J = 7.0 Hz), 5.81-5.84 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.7, 13.5, 19.0, 29.1, 36.4, 48.2, 104.3, 105.7, 150.4, 157.3, 210.2.

HRMS (EI): m/z calcd for C11H16O2: 180.1150; found: 180.1157.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 0.5% i-PrOH in hexane); t R: 6.0 min (major S-enantiomer) and 6.7 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

1-(5-Methylfuran-2-yl)octan-3-one (13ba)

¹H NMR (500 MHz, CDCl3): δ = 0.88 (3 H, t, J = 7.3 Hz), 1.21-1.33 (4 H, m), 1.56 (2 H, m), 2.23 (3 H, s), 2.39 (2 H, t, J = 7.5 Hz), 2.72 (2 H, t, J = 7.1 Hz), 2.84 (2 H, t, J = 7.7 Hz), 5.81-5.84 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.4, 13.9, 22.3, 22.4, 23.5, 31.3, 40.9, 42.8, 105.7, 105.9, 150.4, 152.8, 209.9.

HRMS (EI): m/z calcd for C13H20O2: 208.14633; found: 208.1466.

( S )-5-(5-Ethylfuran-2-yl)hexan-3-one (13cb)

[α]D ¹8 +3.60 (c 2.50, CHCl3) (77% ee).

¹H NMR (500 MHz, C6D6): δ = 0.99 (3 H, t, J = 7.3 Hz), 1.17 (3 H, t, J = 7.6 Hz), 1.28 (3 H, d, J = 6.9), 1.99 (2 H, q, J = 7.3 Hz), 2.26 (1 H, dd, J = 8.1, 16.4 Hz), 2.57 (2 H, q, J = 7.5 Hz), 2.63 (1 H, dd, J = 5.8, 16.4 Hz), 3.36 (1 H, sext, J = 6.9 Hz), 5.90-5.93 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 8.2, 12.7, 19.5, 22.1, 29.7, 36.5, 48.5, 105.0, 105.1, 156.5, 158.1, 208.1.

HRMS (EI): m/z calcd for C12H18O2: 194.1307; found: 194.1313.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 0.5% i-PrOH in hexane); t R: 5.5 min (major S-enantiomer) and 6.2 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(5-Butylfuran-2-yl)hexan-3-one (13db)

[α]D ²7 +3.59 (c 1.00, CHCl3) (79% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.92 (3 H, t, J = 7.4 Hz), 1.04 (3 H, t, J = 7.3 Hz), 1.22 (3 H, d, J = 6.9 Hz), 1.35 (2 H, sext, J = 7.5 Hz), 1.56-1.61 (2 H, m), 2.38 (2 H, q, J = 7.3 Hz), 2.50 (1 H, dd, J = 7.9, 16.1 Hz), 2.55 (2 H, t, J = 7.5 Hz), 2.79 (1 H, dd, J = 6.1, 16.1 Hz), 3.35 (1 H, sext, J = 7.2 Hz), 5.82-5.84 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.7, 13.8, 18.9, 22.3, 27.7, 29.1, 30.2, 36.5, 48.2, 104.1, 104.8, 154.9, 157.1, 210.2.

HRMS (EI): m/z calcd for C14H22O2: 222.1620; found: 222.1623.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 0.5% i-PrOH in hexane); t R: 5.0 min (major S-enantiomer) and 5.4 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(5-Benzylfuran-2-yl)hexan-3-one (13eb)

[α]D ²7 +4.39 (c 1.00, CHCl3) (85% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.00 (3 H, t, J = 7.3 Hz), 1.22 (3 H, d, J = 6.9 Hz), 2.32 (2 H, q, J = 7.3 Hz), 2.49 (1 H, dd, J = 7.8, 16.1 Hz), 2.78 (1 H, dd, J = 6.2, 16.1 Hz), 3.35 (1 H, sext, J = 6.9 Hz), 5.85-5.87 (2 H, m), 7.20-7.31 (5 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.7, 18.9, 29.1, 34.5, 36.4, 48.1, 104.4, 106.6, 126.4, 128.4, 128.6, 138.3, 152.8, 158.0, 210.1.

HRMS (EI): m/z calcd for C17H20O2 256.1463; found: 256.1472.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min 1% i-PrOH in hexane); t R: 6.2 min (major S-enantiomer) and 6.9 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(4,5-Dimethylfuran-2-yl)hexan-3-one (13fb)

[α]D ²7 +5.30 (c = 2.00, CHCl3) (91% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.04 (3 H, t, J = 7.3 Hz), 1.20 (3 H, d, J = 6.9 Hz), 1.87 (3 H, s), 2.14 (3 H, s), 2.38 (2 H, q, J = 7.3 Hz), 2.49 (1 H, dd, J = 7.9, 16.2 Hz), 2.78 (1 H, dd, J = 6.0, 16.2 Hz), 3.30 (1 H, sext, J = 7.0 Hz), 5.74 (1 H, s).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.7, 9.8, 11.2, 19.0, 29.0, 36.4, 48.2, 106.9, 114.1, 145.4, 156.0, 210.2.

HRMS (EI): m/z calcd for C12H18O2: 194.1307; found: 194.1306.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 0.1% i-PrOH in hexane); t R: 11.3 min (major S-enantiomer) and 14.5 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-2-(5-Methylfuran-2-yl)heptan-4-one (13bc)

[α]D ²9 +3.30 (c 1.03, CHCl3) (81% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.89 (3 H, t, J = 7.4 Hz), 1.22 (3 H, d, J = 6.9 Hz), 1.59 (2 H, sext, J = 7.4 Hz), 2.23 (3 H, s), 2.34 (2 H, t, J = 7.3 Hz), 2.50 (1 H, dd, J = 8.1, 16.3 Hz), 2.79 (1 H, dd, J = 5.9, 16.3 Hz), 3.33 (1 H, sext, J = 7.0 Hz), 5.82-5.83 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.5, 13.7, 17.1, 19.0, 28.9, 45.2, 48.5, 104.3, 105.7, 150.3, 157.3, 209.8.

HRMS (EI): m/z calcd for C12H18O2: 194.1307; found: 194.1309.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 0.1% i-PrOH in hexane); t R: 12.3 min (major S-enantiomer) and 13.7 min (minor R-enantiomer). The absolute configuration of was determined by converting it into (S)-2-methyl-4-oxoheptanoic acid.

( S )-2-(5-Methylfuran-2-yl)octan-4-one (13bd)

[α]D ²0 +1.59 (c 1.00, CHCl3) (86% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.89 (3 H, t, J = 7.4 Hz), 1.22 (3 H, d, J = 6.9 Hz), 1.29 (2 H, m), 1.54 (2 H, m), 2.23 (3 H, s), 2.36 (2 H, t, J = 7.5 Hz), 2.51 (1 H, dd, J = 8.0, 16.3 Hz), 2.79 (1 H, dd, J = 5.9, 16.3 Hz), 3.34 (1 H, sext, J = 7.0 Hz), 5.82-5.83 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.5, 13.8, 19.0, 22.3, 25.8, 29.0, 43.1, 48.5, 104.3, 105.7, 150.3, 157.3, 209.9.

HRMS (EI): m/z calcd for C13H20O2 208.1463; found: 208.1469.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 0.5% i-PrOH in hexane); t R: 12.4 min (major S-enantiomer) and 13.7 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-2-(4,5-Dimethylfuran-2-yl)octan-4-one (13fd)

[α]D ²9 +6.4 (c 1.22, CHCl3) (93% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.89 (3 H, t, J = 7.3 Hz), 1.20 (3 H, d, J = 6.9 Hz), 1.28 (2 H, m), 1.54 (2 H, m), 1.87 (3 H, s), 2.14 (3 H, s), 2.35 (2 H, t, J = 7.5 Hz), 2.48 (1 H, dd, J = 8.0, 16.2 Hz), 2.77 (1 H, dd, J = 6.0, 16.2 Hz), 3.29 (1 H, sext, J = 6.9 Hz), 5.73 (1 H, s).

¹³C NMR (125.8 MHz, CDCl3): δ = 9.8, 11.2, 13.8, 19.0, 22.3, 25.8, 28.9, 43.0, 48.5, 107.0, 114.1, 145.4, 156.0, 209.9.

HRMS (EI): m/z calcd for C14H22O2: 222.1620; found: 222.1621.

The ee value was determined by HPLC analysis using a Chiralpak AS-H column (1 mL/min, 0.1% i-PrOH in hexane); t R: 12.8 min (major S-enantiomer) and 17.5 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-4-(4,5-Dimethylfuran-2-yl)octan-2-one (13fe)

[α]D ²0 +2.60 (c = 2.00, CHCl3) (87% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.86 (3 H, t, J = 7.0 Hz), 1.17-1.33 (4 H, m), 1.45-1.62 (2 H, m), 1.87 (3 H, s), 2.07 (3 H, s), 2.13 (3 H, s), 2.58 (1 H, dd, J = 6.6, 16.2 Hz), 2.74 (1 H, dd, J = 7.5, 16.2 Hz), 3.12 (1 H, quint, J = 7.1 Hz), 5.74 (1 H, s).

¹³C NMR (125.8 MHz, CDCl3): δ = 9.9, 11.2, 13.9, 22.5, 29.3, 30.3, 33.7, 34.5, 48.1, 108.2, 114.0, 145.4, 154.4, 207.9.

HRMS (EI): m/z calcd for C14H22O2: 222.1620; found: 222.1626.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 0.1% i-PrOH in hexane); t R: 8.8 min (major S-enantiomer) and 10.5 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

Ethyl ( S )-2-(5-Methylfuran-2-yl)-4-oxopentanoate (13bf)

[α]D ³0 +47.7 (c = 2.08, CHCl3) (85% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.23 (3 H, t, J = 7.1 Hz), 2.18 (3 H, s), 2.23 (3 H, s), 2.80 (1 H, dd, J = 4.5, 18.0 Hz), 3.29 (1 H, dd, J = 10.0, 18.0 Hz), 4.12-4.20 (3 H, m), 5.86 (1 H, m), 6.01 (1 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.5, 14.0, 29.9, 40.2, 44.1, 61.3, 106.3, 107.4, 149.1, 151.6, 171.2, 205.8.

HRMS (FAB/m-NBA): m/z calcd for C12H17O4 (M + H+): 225.1127; found: 225.1132.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 1% i-PrOH in hexane); t R: 12.1 min (major S-enantiomer) and 13.4 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

Ethyl ( S )-2-(4,5-Dimethylfuran-2-yl)-4-oxopentanoate (13ff)

[α]D ²5 +62.9 (c = 2.00, CHCl3) (89% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.24 (3 H, t, J = 7.2 Hz), 1.87 (3 H, s), 2.14 (3 H, s), 2.18 (3 H, s), 2.79 (1 H, dd, J = 4.6, 18.0 Hz), 3.29 (1 H, dd, J = 9.9, 18.0 Hz), 4.09-4.20 (3 H, m), 5.90 (1 H, s).

¹³C NMR (125.8 MHz, CDCl3): δ = 9.7, 11.2, 14.0, 29.9, 40.2, 44.2, 61.2, 109.9, 114.7, 146.8, 147.8, 171.3, 205.9.

HRMS (EI): m/z calcd for C13H18O4: 238.1205; found: 238.1198.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 6% i-PrOH in hexane); t R: 6.7 min (minor R-enantiomer) and 7.2 min (major S-enantiomer). The absolute stereochemistry was assumed by analogy.

Enantioselective Synthesis of γ-Keto Carboxylic Acids; ( S )-2-Methyl-4-oxoheptanoic Acid (14a); [³0 ] Typical Procedure (Table 3, Entry 1)

RuCl3˙H2O (1.7 mg, 0.0083 mmol) was added to a mixture of NaIO4 (2.66 g, 12.4 mmol) in CH2Cl2 (11 mL), MeCN (0.64 mL), and H2O (11 mL) and the mixture was stirred for 1 h at r.t. To the mixture was added a CH2Cl2 (4.8 mL) solution of 13bc (161 mg, 0.829 mmol) (82% ee). After stirring for 4 h at r.t., the reaction mixture was acidified to pH 1 with aq 1 N NaHSO4. The mixture was filtered through a pad of Celite. The organic layer was separated and extracted with aq 5% K2CO3 (3 × 10 mL). The combined aqueous layers were acidified to pH 1 with solid NaHSO4 and then extracted with CH2Cl2 (3 × 20 mL). The combined organic layers were dried (Na2SO4) and concentrated in vacuo to give 126 mg (96%) of (S)-14a. The analytical sample was prepared after purification by flash chromatography (SiO2, gradient elution with 10-50% EtOAc in hexane); [α]D ²9 -11.2 (c 0.48, CHCl3) {Lit. [³0] [α]D ²5 -14.6 (c 0.48, CHCl3)} (92% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.90 (3 H, t, J = 7.5 Hz), 1.21 (3 H, d, J = 7.2 Hz), 1.60 (2 H, sext, J = 7.4 Hz), 2.38 (2 H, m), 2.46 (1 H, dd, J = 5.4, 17.8 Hz), 2.89 (1 H, dd, J = 8.1, 17.8 Hz), 2.97 (1 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13,6, 16.9, 17.2, 34.5, 44.8, 45.4, 181.7, 209.0.

(S)-2-Methyl-4-oxoheptanoic Acid Benzylamide: To determine the enantiopurity of (S)-14a, it was converted into the corresponding benzylamide. To a solution of the acid (22.0 mg, 0.139 mmol) in THF (3.3 mL) at 0 ˚C was added benzylamine (26.3 mg, 0.245 mmol), 1-hydroxybenzotriazole hydrate (33.1 mg, 0.245 mmol), and 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (31.4 mg, 0.164 mmol). The mixture was stirred at 0 ˚C for 3 h and at r.t. for 17 h. After concentration, the residue was partitioned between EtOAc (20 mL) and aq 1 N HCl (10 mL). The organic layer was separated, washed with sat. aq NaHCO3 (10 mL) and concentrated in vacuo. Purification of the residue by flash chromatography (SiO2, gradient elution with 15-30% EtOAc in hexane) gave 27.8 mg (81%) of (S)-2-methyl-4-oxoheptanoic acid benzylamide (82% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.89 (3 H, t, J = 7.4 Hz), 1.16 (3 H, d, J = 7.0 Hz), 1.57 (2 H, sext, J = 7.4 Hz), 2.36 (2 H, t, J = 7.5 Hz), 2.40 (1 H, dd, J = 4.3, 17.9 Hz), 2.79 (1 H, m), 2.94 (1 H, dd, J = 9.1, 17.9 Hz), 4.40 (2 H, m), 6.14 (1 H, br), 7.20-7.33 (5 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.6, 17.2, 17.9, 35.8, 43.4, 44.9, 46.5, 127.3, 127.6, 128.6, 138.4, 175.5, 210.2.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 5% i-PrOH in hexane); t R: 18.0 min (major S-enantiomer) and 21.3 min (minor R-enantiomer).

( S )-2-Methyl-4-oxohexanoic Acid (14b) []

[α]D ²9 -11.8 (c 1.05, CHCl3).

¹H NMR (500 MHz, CDCl3): δ = 1.05 (3 H, t, J = 7.3 Hz), 1.21 (3 H, d, J = 7.2 Hz), 2.38-2.49 (3 H, m), 2.87 (1 H, dd, J = 8.1, 17.7 Hz), 2.98 (1 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.6, 16.9, 34.5, 36.0, 45.0, 181.6, 209.4.

HRMS (EI): m/z calcd for C7H12O3: 144.0786; found: 144.0793.

(S)-2-Methyl-4-oxohexanoic Acid Benzylamide: By a procedure similar to that described for 14a, the acid 14b was converted into (S)-2-methyl-4-oxohexanoic acid benzylamide (80% yield, 92% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.02 (3 H, t, J = 7.3 Hz), 1.44 (3 H, d, J = 7.0 Hz), 2.39-2.44 (3 H, m), 2.79 (1 H, m), 2.94 (1 H, dd, J = 9.1, 17.8 Hz), 4.40 (2 H, m), 6.14 (1 H, br), 7.19-7.32 (5 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.6, 17.9, 35.9, 36.1, 43.5, 46.1, 127.3, 127.6, 128.6, 138.4, 175.5, 210.6.

The ee value was determined by HPLC analysis using a Chiralcel AS-H column (1 mL/min, 5% i-PrOH in hexane); t R: 25.4 min (major S-enantiomer) and 31.6 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-2-Methyl-4-oxooctanoic Acid (14c)

[α]D ²9 -11.2 (c 1.02, CHCl3).

¹H NMR (500 MHz, CDCl3): δ = 0.89 (3 H, t, J = 7.4 Hz), 1.20 (3 H, d, J = 7.2 Hz), 1.30 (2 H, sext, J = 7.3 Hz), 1.52-1.58 (2 H, m), 2.37-2.43 (2 H, m), 2.47 (1 H, dd, J = 5.4, 17.7 Hz), 2.86 (1 H, dd, J = 8.0, 17.7 Hz), 2.97 (1 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.8, 16.8, 22.2, 25.8, 34.5, 42.6, 45.4, 181.8, 209.0.

HRMS (EI): m/z calcd for C9H16O3: 172.1100; found: 172.1103.

(S)-2-Methyl-4-oxooctanoic Acid Benzylamide: By a procedure similar to that described for 14a, the acid 14c was converted into (S)-2-methyl-4-oxooctanoic acid benzylamide (80% yield, 93% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.89 (3 H, t, J = 7.4 Hz), 1.44 (3 H, d, J = 7.0 Hz), 1.28 (2 H, m), 1.51 (2 H, m), 2.37-2.44 (3 H, m), 2.79 (1 H, m), 2.96 (1 H, dd, J = 9.1, 17.9 Hz), 4.41 (2 H, m), 6.07 (1 H, br), 7.24-7.33 (5 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.8, 17.9, 22.3, 25.8, 35.8, 42.8, 43.5, 46.5, 127.3, 127.6, 128.6, 138.4, 175.4, 210.3.

The ee value was determined by HPLC analysis using a Chiralcel OB-H column (1 mL/min, 5% i-PrOH in hexane); retention times: 14.3 min (major S-enantiomer) and 19.3 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-2-(2-Oxopropyl)hexanoic Acid (14d)

[α]D ²8 -18.9 (c 1.11, CHCl3).

¹H NMR (500 MHz, CDCl3): δ = 0.88 (3 H, t, J = 6.9 Hz), 1.27-1.32 (4 H, m), 1.51 (1 H, m), 1.65 (1 H, m), 2.15 (3 H, s), 2.51 (1 H, m), 2.84-2.90 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.8, 22.5, 29.0, 30.0, 31.3, 39.9, 44.6, 181.1, 206.9.

HRMS (EI): m/z calcd for C9H16O3: 172.1100; found: 172.1103.

(S)-2-(2-Oxopropyl)hexanoic Acid Benzylamide: By a procedure similar to that described for 14a, the acid 14d was converted into (S)-2-(2-oxopropyl)hexanoic acid benzylamide (54% yield, 88% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.87 (3 H, t, J = 7.0 Hz), 1.23-1.41 (5 H, m), 1.62 (1 H, m), 2.14 (3 H, s), 2.48 (1 H, dd, J = 3.6, 18.1 Hz), 2.61 (1 H, m), 2.99 (1 H, dd, J = 9.7, 18.2 Hz), 4.43 (2 H, m), 6.00 (1 H, br), 7.24-7.36 (5 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.9, 22.6, 29.6, 30.2, 32.3, 41.8, 43.6, 46.1, 127.4, 127.7, 128.6, 138.3, 174.9, 207.9.

The ee value was determined by HPLC analysis using a Chiralpak AS-H column (1 mL/min, 4% i-PrOH in hexane); t R: 20.9 min (major S-enantiomer) and 23.3 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

F-C Alkylation of Indoles; ( S )-5-(1 H -Indol-3-yl)hexan-3-one (16ab); [¹6a ] Typical Procedure (Table 5, Entry 1)

To a mixture of OXB 9a (0.20 mmol), prepared by Method A, in anhyd Et2O (3.1 mL) at -85 ˚C were added indole (15a; 234 mg, 2.0 mmol), N,N-dimethylaniline (6.09 mg, 0.050 mmol), and hex-4-en-3-one (12b; 196 mg, 2.0 mmol). After stirring at -85 ˚C for 20 h, the reaction mixture was quenched by the addition of sat. aq NaHCO3 (10 mL) and filtered. The filtrate was extracted with EtOAc (3 × 20 mL), dried (Na2SO4), and concentrated in vacuo. Purification of the residue by flash chromatography (SiO2, 8-12% EtOAc in hexane) gave 396 mg (92%) of (S)-16ab (87% ee); [α]D ²³ +8.37 (c 3.01, CHCl3) {Lit. [¹6a] [α]D r.t. -8.5 (c 0.93, CHCl3) for R-enantiomer (96% ee)}.

¹H NMR (500 MHz, CDCl3): δ = 1.02 (3 H, t, J = 7.3 Hz), 1.40 (3 H, d, J = 6.9 Hz), 2.38 (2 H, q, J = 7.3 Hz), 2.71 (1 H, dd, J = 8.3, 15.9 Hz), 2.94 (1 H, dd, J = 6.0, 15.9 Hz), 3.68 (1 H, m), 6.96 (1 H, d, J = 2.3 Hz), 7.14 (1 H, m), 7.21 (1 H, m), 7.35 (1 H, d, J = 8.1 Hz), 7.67 (1 H, d, J = 7.9 Hz), 8.09 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.6, 21.2, 27.0, 36.4, 50.2, 111.2, 119.09, 119.12, 120.1, 121.0, 121.9, 126.2, 136.5, 211.4.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 3% i-PrOH in hexane); t R: 33.6 min (major S-enantiomer) and 36.5 min (minor R-enantiomer).

( S )-5-(5-Methoxy-1 H -indol-3-yl)hexan-3-one (16bb)

[α]D ²5 -1.03 (c 3.08, CHCl3) (89% ee).

¹H NMR (500 MHz, C6D6): δ = 0.96 (3 H, t, J = 7.3 Hz), 1.46 (3 H, d, J = 6.9 Hz), 2.02 (2 H, m), 2.48 (1 H, dd, J = 7.9, 15.8 Hz), 2.77 (1 H, dd, J = 6.3, 15.8 Hz), 3.68 (3 H, s), 3.81 (1 H, sext, J = 7.1 Hz), 6.67 (1 H, d, J = 2.3 Hz), 7.11-7.15 (2 H, m), 7.34 (1 H, d, J = 1.4 Hz), 7.42 (1 H, br s).

¹³C NMR (125.8 MHz, C6D6): δ = 8.2, 21.9, 27.8, 36.6, 50.6, 55.9, 101.9, 112.7, 112.8, 121.2, 121.5, 127.7, 132.6, 154.9, 210.2.

HRMS (EI): m/z calcd for C15H19NO2: 245.1416; found: 245.1423.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 2.5% i-PrOH in hexane); t R: 78.7 min (major S-enantiomer) and 84.1 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(5-Methyl-1 H -indol-3-yl)hexan-3-one (16cb)

[α]D ³0 +1.12 (c 3.03, CHCl3) (83% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.05 (3 H, t, J = 7.3 Hz), 1.41 (3 H, d, J = 6.9 Hz), 2.41 (2 H, q, J = 7.3 Hz), 2.51 (3 H, s), 2.72 (1 H, dd, J = 8.4, 15.9 Hz), 2.94 (1 H, dd, J = 5.9, 15.8 Hz), 3.66 (1 H, m), 6.91 (1 H, d, J = 2.3 Hz), 7.05 (1 H, m), 7.25 (1 H, m), 7.46 (1 H, m), 8.04 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.7, 21.3, 21.6, 27.1, 36.5, 50.3, 111.0, 118.8, 120.4, 120.5, 123.6, 126.5, 128.4, 134.9, 211.6.

HRMS (EI): m/z calcd for C15H19NO: 229.1467; found: 229.1463.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 5.0%, i-PrOH in hexane); t R: 25.5 min (major S-enantiomer) and 34.3 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(5-Chloro-1 H -indol-3-yl)hexan-3-one (16db)

[α]D ³¹ +3.48 (c 3.04, CHCl3) (85% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.01 (3 H, t, J = 7.2 Hz), 1.36 (3H, d, J = 6.9 Hz), 2.37 (2 H, q, J = 7.3 Hz), 2.68 (1 H, dd, J = 8.0, 16.0 Hz), 2.87 (1 H, dd, J = 6.2, 16.0 Hz), 3.59 (1 H, sext, J = 7.0 Hz), 6.97 (1 H, d, J = 2.3 Hz), 7.13 (1 H, dd, J = 1.9, 8.6 Hz), 7.24 (1 H, d, J = 8.6 Hz), 7.60 (1 H, d, J = 1.6 Hz), 8.14 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.6, 21.2, 26.9, 36.4, 50.0, 112.2, 118.6, 120.8, 121.6, 122.2, 124.9, 127.3, 134.8, 211.0.

HRMS (EI): m/z calcd for C14H16ClNO: 249.0920; found: 249.0930.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 4%, i-PrOH in hexane); t R: 33.6 min (major S-enantiomer) and 37.8 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(5-Bromo-1 H -indol-3-yl)hexan-3-one (16eb)

[α]D ²8 +2.00 (c 3.00, CHCl3) (85% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.01 (3 H, t, J = 7.3 Hz), 1.35 (3 H, d, J = 6.9 Hz), 2.37 (2 H, q, J = 7.3 Hz), 2.68 (1 H, dd, J = 8.0, 16.0 Hz), 2.86 (1 H, dd, J = 6.2, 16.0 Hz), 3.58 (1 H, m), 6.95 (1 H, d, J = 2.3 Hz), 7.20 (1 H, d, J = 8.5 Hz), 7.25 (1 H, dd, J = 1.8, 8.6 Hz), 7.76 (1 H, d, J = 1.7 Hz), 8.20 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.6, 21.2, 26.9, 36.4, 50.0, 112.4, 112.7, 120.7, 121.5, 121.6, 124.7, 128.0, 135.1, 211.1.

HRMS (EI): m/z calcd for C14H16BrNO: 293.0415; found: 293.0406.

The ee value was determined by HPLC analysis using a Chiralpak AS-H (1 mL/min, 7.0%, i-PrOH in hexane); t R: 29.9 min (minor R-enantiomer) and 45.0 min (major S-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(1-Methyl-1 H -indol-3-yl)hexan-3-one (16fb)

[α]D ²8 +12.7 (c 3.00, CHCl3) (93% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.02 (3 H, t, J = 7.3 Hz), 1.39 (3 H, d, J = 6.9 Hz), 2.38 (2 H, q, J = 7.3 Hz), 2.69 (1 H, dd, J = 8.3, 15.9 Hz), 2.92 (1 H, dd, J = 6.0, 15.9 Hz), 3.66 (1 H, m), 3.74 (3 H, s), 6.84 (1 H, s), 7.12 (1 H, m), 7.24 (1 H, m), 7.30 (1 H, d, J = 8.2 Hz), 7.65 (1 H, d, J = 7.9 Hz).

¹H NMR (500 MHz, C6D6): δ = 0.99 (3 H, t, J = 7.3 Hz), 1.49 (3 H, d, J = 6.9 Hz), 2.03 (2 H, m), 2.50 (1 H, dd, J = 8.0, 15.9 Hz), 2.80 (1 H, dd, J = 6.2, 15.8 Hz), 3.06 (3 H, s), 3.84 (1 H, sext, J = 6.9 Hz), 6.53 (1 H, s), 7.14 (1 H, d, J = 8.1 Hz), 7.30 (1 H, m), 7.36 (1 H, m), 7.80 (1 H, d, J = 7.9 Hz).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.7, 21.4, 27.0, 32.5, 36.4, 50.4, 109.3, 118.6, 119.2, 119.6, 121.5, 125.0, 126.6, 137.1, 211.1.

HRMS (EI): m/z calcd for C15H19NO: 229.1467; found: 229.1462.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 2% i-PrOH in hexane); t R: 17.5 min (major S-enantiomer) and 20.1 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(1-Allyl-1 H -indol-3-yl)hexan-3-one (16gb)

[α]D ²8 +10.93 (c 3.00, CHCl3) (85% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.02 (3 H, t, J = 7.3 Hz), 1.40 (3 H, d, J = 6.9 Hz), 2.37 (2 H, q, J = 7.3 Hz), 2.70 (1 H, dd, J = 8.4, 15.8 Hz), 2.91 (1 H, dd, J = 5.8, 15.8 Hz), 3.66 (1 H, sext, J = 7.0 Hz), 4.68 (2 H, d, J = 5.4 Hz), 5.09 (1 H, m), 5.20 (1 H, m), 5.99 (1 H, m), 6.88 (1 H, s), 7.12 (1 H, m), 7.22 (1 H, t, J = 8.1 Hz), 7.30 (1 H, d, J = 8.2 Hz), 7.66 (1 H, d, J = 7.9 Hz).

¹H NMR (500 MHz, C6D6): δ = 0.99 (3 H, t, J = 7.3 Hz), 1.48 (3 H, d, J = 6.9 Hz), 2.02 (2 H, q, J = 7.4 Hz), 2.50 (1 H, dd, J = 8.1, 15.8 Hz), 2.79 (1 H, dd, J = 6.0, 15.8 Hz), 3.84 (1 H, sext, J = 6.9 Hz), 4.16 (2 H, m), 4.85 (1 H, m), 4.95 (1 H, m), 5.65 (1 H, m), 6.68 (1 H, s), 7.22 (1 H, d, J = 8.1 Hz), 7.29 (1 H, m), 7.34 (1 H, m), 7.79 (1 H, d, J = 7.8 Hz).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.6, 21.3, 27.0, 36.4, 48.6, 50.3, 109.6, 117.1, 118.8, 119.3, 120.1, 121.6, 123.9, 126.8, 133.5, 136.6, 211.1.

HRMS (EI): m/z calcd for C17H21NO: 255.1623; found: 255.1617.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 3.0%, i-PrOH in hexane); t R: 12.0 min (major S-enantiomer) and 16.2 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(1-Benzyl-1 H -indol-3-yl)hexan-3-one (16hb)

[α]D ³¹ +10.49 (c 3.01, CHCl3) (75% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.12 (3 H, t, J = 7.3 Hz), 1.53 (3 H, d, J = 6.9 Hz), 2.41 (2 H, q, J = 7.3 Hz), 2.80 (1 H, dd, J = 8.3, 15.8 Hz), 3.03 (1 H, dd, J = 5.9, 15.8 Hz), 3.81 (1 H, sext, 8.1 Hz), 5.30 (2 H, s), 7.02 (1 H, m), 7.18-7.20 (2 H, m), 7.25-7.31 (2 H, m), 7.34-7.39 (4 H, m), 7.81 (1 H, d, J = 7.6 Hz).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.46, 21.1, 26.9, 36.2, 49.5, 50.1, 109.6, 118.8, 119.2, 120.2, 121.6, 124.2, 126.5, 126.8, 127.3, 128.5, 136.7, 137.5, 210.7.

HRMS (EI): m/z calcd for C21H23NO: 305.1780; found: 305.1780.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 0.5%, i-PrOH in hexane); t R: 25.2 min (major S-enantiomer) and 28.0 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(2-Methyl-1 H -indol-3-yl)hexan-3-one (16ib)

[α]D ²4 +20.9 (c 3.03, CHCl3) (87% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.97 (3 H, t, J = 7.3 Hz), 1.48 (3 H, d, J = 7.1 Hz), 2.23 (1 H, m), 2.35 (1 H, m), 2.39 (3 H, s), 2.86 (1 H, dd, J = 6.8, 15.8 Hz), 3.06 (1 H, dd, J = 7.8, 15.8 Hz), 3.65 (1 H, sext, J = 7.2 Hz), 7.11-7.17 (2 H, m), 7.24 (1 H, m), 7.69 (1 H, m), 7.98 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.5, 11.7, 21.0, 27.1, 36.5, 49.1, 110.5, 114.7, 118.6, 118.8, 120.4, 126.8, 130.4, 135.4, 211.7.

HRMS (EI): m/z calcd for C15H19NO: 229.1467; found: 229.1458.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 3.0% i-PrOH in hexane); t R: 20.9 min (major S-enantiomer) and 28.1 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(1,2-Dimethyl-1 H -indol-3-yl)hexan-3-one (16jb)

[α]D ²8 +60.11 (c 2.99, CHCl3) (81% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.96 (3 H, t, J = 7.3 Hz), 1.46 (3 H, d, J = 7.1 Hz), 2.18-2.26 (1 H, m), 2.31-2.39 (1 H, m), 2.42 (3 H, s), 2.86 (1 H, dd, J = 6.8, 16.0 Hz), 3.05 (1 H, dd, J = 7.7, 16.0 Hz), 3.61-3.69 (4 H, m), 7.09 (1 H, t, J = 7.2 Hz), 7.18 (1 H, t, J = 7.4 Hz), 7.28 (1 H, d, J = 8.1 Hz), 7.68 (1 H, d, J = 7.9 Hz).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.5, 10.3, 21.3, 27.4, 29.3, 36.5, 49.3, 108.8, 114.4, 118.3, 119.0, 120.1, 125.9, 132.2, 136.8, 211.2.

HRMS (EI): m/z calcd for C16H21NO: 243.1623; found: 243.1627.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 1.0%, i-PrOH in hexane); retention times: 17.5 min (minor R-enantiomer) and 19.9 min (major S-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-(5-Methoxy-2-methyl-1 H -indol-3-yl)hexan-3-one (16kb)

[α]D ³0 +41.07 (c 2.99, CDCl3) (75% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.94 (3 H, t, J = 7.3 Hz), 1.43 (3 H, d, J = 7.1 Hz), 2.22 (1 H, m), 2.32 (1 H, m), 2.36 (3 H, s), 2.80 (1 H, dd, J = 6.8, 15.7 Hz), 3.01 (1 H, dd, J = 7.7, 15.7 Hz), 3.58 (1 H, sext, J = 7.2 Hz), 3.89 (3 H, s), 6.78 (1 H, dd, J = 2.2, 8.7 Hz), 7.12-7.14 (2 H, m), 7.76 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.5, 12.0, 20.8, 27.1, 36.6, 48.9, 56.1, 102.2, 109.5, 110.9, 114.7, 127.4, 130.7, 131.5, 153.3, 211.4.

HRMS (EI): m/z calcd for C16H21NO2: 259.1572; found: 259.1573.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 3.0%, i-PrOH in hexane); t R: 24.1 min (major S-enantiomer) and 27.4 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-2-(1 H -Indol-3-yl)octan-4-one (16ad)

[α]D ³¹ +2.38 (c 3.02, CHCl3) (92% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.90 (3 H, t, J = 7.4 Hz), 1.29 (2 H, sext, J = 7.5 Hz), 1.42 (3 H, d, J = 6.9 Hz), 1.55 (2 H, m), 2.38 (2 H, t, J = 7.5 Hz), 2.73 (1 H, dd, J = 8.3, 16.0 Hz), 2.95 (1 H, dd, J = 5.9, 16.0 Hz), 3.70 (1 H, m), 6.95 (1 H, d, J = 2.3 Hz), 7.16 (1 H, m), 7.22 (1 H, m), 7.35 (1 H, d, J = 8.1 Hz), 7.69 (1 H, d, J = 7.9 Hz), 8.16 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.7, 21.1, 22.2, 25.7, 26.9, 43.0, 50.5, 111.3, 119.0, 119.1, 120.2, 120.9, 121.8, 126.2, 136.5, 211.1.

HRMS (EI): m/z calcd for C16H21NO 243.1623; found: 243.1622.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 3% i-PrOH in hexane); t R: 29.1 min (major S-enantiomer) and 33.5 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-4-(1 H -Indol-3-yl)octan-2-one (16ae) []

[α]D ²8 +13.44 (c 1.22, CHCl3) (72% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.83 (3 H, t, J = 7.2 Hz), 1.21-1.31 (4 H, m), 1.67-1.82 (2 H, m), 2.03 (3 H, s), 2.81 (1 H, dd, J = 6.8, 15.8 Hz), 2.90 (1 H, dd, J = 7.6, 15.8 Hz), 3.47 (1 H, sext, J = 7.1 Hz), 6.96 (1 H, d, J = 2.4 Hz), 7.12 (1 H, m), 7.19 (1 H, m), 7.35 (1 H, d, J = 8.1 Hz), 7.67 (1 H, d, J = 7.9 Hz), 8.06 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 14.0, 22.6, 29.8, 30.4, 32.8, 35.5, 50.2, 111.3, 119.0, 119.1, 119.3, 121.2, 121.8, 126.5, 136.5, 209.0.

HRMS (EI): m/z calcd for C16H21NO: 243.1623; found: 243.1630.

The ee value was determined by HPLC analysis using a Chiralpak AD-H (1 mL/min, 1.5%, i-PrOH in hexane); t R: 59.7 min (major S-enantiomer) and 63.7 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-4-(2-Methyl-1 H -indol-3-yl)octan-2-one (16ie)

[α]D ²6 +30.2 (c 3.01, CHCl3) (82% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.85 (3 H, t, J = 7.2 Hz), 1.17-1.35 (4 H, m), 1.76 (1 H, m), 1.93 (1 H, m), 1.98 (3 H, s), 2.38 (3 H, s), 2.88 (1 H, dd, J = 6.0, 15.8 Hz), 3.11 (1 H, dd, J = 8.5, 15.8 Hz), 3.42 (1 H, m), 7.09-7.15 (2 H, m), 7.24 (1 H, m), 7.66 (1 H, d, J = 7.5 Hz), 8.00 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 11.8, 13.9, 22.5, 30.0, 30.6, 32.6, 34.8, 49.2, 110.4, 113.0, 118.6, 118.8, 120.3, 127.0, 131.4, 135.5, 209.3.

HRMS (EI): m/z calcd for C17H23NO: 257.1780; found: 257.1784.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 5% i-PrOH in hexane); t R: 18.4 min (minor R-enantiomer) and 24.1 min (major S-enantiomer). The absolute stereochemistry was assumed by analogy.

Ethyl ( S )-2-(1 H -Indol-3-yl)-4-oxopentanoate (16af) [¹6a]

[α]D ²6 +121.9 (c 3.00, CHCl3) (94% ee) {Lit. [¹6a] [α]D r.t. -106.4 (c 1.04, CHCl3) for R-enantiomer (95% ee)}.

¹H NMR (500 MHz, CDCl3): δ = 1.20 (3 H, t, J = 7.1 Hz), 2.19 (3 H, s), 2.86 (1 H, dd, J = 4.3, 18.0 Hz), 3.53 (1 H, dd, J = 10.3, 18.0 Hz), 4.09 (1 H, m), 4.19 (1 H, m), 4.42 (1 H, dd, J = 4.3, 10.3 Hz), 7.03 (1 H, d, J = 2.2 Hz), 7.16 (1 H, t, J = 7.9 Hz), 7.21 (1 H, m), 7.33 (1 H, d, J = 8.2 Hz), 7.74 (1 H, d, J = 7.9 Hz), 8.47 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 13.9, 29.8, 37.8, 46.1, 60.9, 111.3, 112.4, 119.0, 119.5, 122.1, 122.2, 126.0, 136.2, 173.8, 207.1.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 10% i-PrOH in hexane); t R: 19.4 min (major S-enantiomer) and 27.0 min (minor R-enantiomer).

4-(1 H -Indol-3-yl)-4-phenylbutan-2-one (16ag) [¹6a]

¹H NMR (500 MHz, CDCl3): δ = 2.09 (3 H, s), 3.18 (1 H, dd, J = 7.8, 16.0 Hz), 3.27 (1 H, dd, J = 7.4, 16.0 Hz), 4.85 (1 H, t, J = 7.6 Hz), 6.98 (1 H, m), 7.04 (1 H, m), 7.15-7.20 (2 H, m), 7.28 (2 H, m), 7.32-7.33 (3 H, m), 7.45 (1H, d, J = 7.9 Hz), 8.06 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 30.4, 38.4, 50.3, 111.6, 118.8, 119.4, 121.4, 122.2, 126.4, 126.5, 127.7, 128.5, 136.6, 144.0, 207.7.

( R )-4-(2-Methyl-1 H -indol-3-yl)-4-phenylbutan-2-one (16ig) [54]

[α]D ²8 +17.1 (c 3.04, CHCl3) (84% ee).

¹H NMR (500 MHz, CDCl3): δ = 2.07 (3 H, s), 2.38 (3 H, s), 3.42 (1 H, dd, J = 6.4, 16.2 Hz), 3.52 (1 H, dd, J = 8.5, 16.2 Hz), 4.96 (1 H, m), 7.11 (1 H, m), 7.17 (1 H, m), 7.22-7.25 (2 H, m), 7.31-7.34 (2 H, m), 7.40-7.42 (2 H, m), 7.58 (1 H, m), 7.98 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 11.8, 30.5, 36.7, 48.1, 110.5, 112.8, 118.8, 119.0, 120.5, 125.8, 127.16, 127.23, 128.2, 131.7, 135.3, 144.0, 208.1.

The ee value was determined by HPLC analysis using a Chiralpak AD-H column (1 mL/min, 4% i-PrOH in hexane); t R: 37.3 min (major S-enantiomer) and 44.0 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( R )-4-(4-Fluorophenyl)-4-(2-methyl-1 H -indol-3-yl)butan-2-one (16ih)

[α]D ²7 +15.80 (c 3.03, CHCl3) (81% ee).

¹H NMR (500 MHz, CDCl3): δ = 2.05 (3 H, s), 2.39 (3 H, s), 3.35 (1 H, dd, J = 6.5, 16.3 Hz), 3.45 (1 H, dd, J = 8.2, 16.3 Hz), 4.88 (1 H, t, J = 7.4 Hz), 6.93-6.97 (2 H, m), 7.06 (1 H, m), 7.12 (1 H, m), 7.24 (1 H, d, J = 8.0 Hz), 7.27-7.30 (2 H, m), 7.47 (1 H, d, J = 7.9 Hz), 7.97 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 11.9, 30.6, 36.0, 48.3, 110.5, 112.8, 114.9 (d, J = 21.1 Hz), 118.8, 119.2, 120.8, 127.1, 128.7 (d, J = 7.9 Hz), 131.7, 135.4, 139.7 (d, J = 3.4 Hz), 161.1 (d, J = 244 Hz), 207.8.

HRMS (EI): m/z calcd for C19H18FNO: 295.1372; found: 295.1367.

The ee value was determined by HPLC analysis using a Chiralpak AD-H (1 mL/min, 4.0%, i-PrOH in hexane); t R: 36.4 min (major R-enantiomer) and 41.3 min (minor S-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-5-Benzyloxy-4-(2-methyl-1 H -indol-3-yl)pentan-2-one (16ii)

[α]D ³0 +17.39 (c 3.15, CHCl3) (75% ee).

¹H NMR (500 MHz, CDCl3): δ = 2.03 (3 H, s), 2.39 (3 H, s), 3.04 (1 H, dd, J = 8.0, 16.4 Hz), 3.15 (1 H, dd, J = 5.3, 16.5 Hz), 3.70 (1 H, m), 3.81-3.90 (2 H, m), 4.54 (2 H, s), 7.05-7.12 (2 H, m), 7.24 (1 H, d, J = 7.9 Hz), 7.29-7.38 (5 H, m), 7.58 (1 H, d, J = 7.8 Hz), 7.89 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 12.0, 30.5, 32.9, 45.7, 72.9, 73.1, 110.5, 110.7, 118.8, 118.9, 120.6, 127.2, 127.46, 127.54, 128.3, 132.0, 135.4, 138.3, 208.4.

HRMS (EI): m/z calcd for C21H23NO2: 321.1729; found: 321.1738.

The ee value was determined by HPLC analysis using a Chiralpak AD-H (1 mL/min, 6.0%, i-PrOH in hexane); t R: 33.8 min (major S-enantiomer) and 36.4 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-1,1-Dimethoxy-2-(2-methyl-1 H -indol-3-yl)nonan-4-one (16ij)

[α]D ²8 +44.60 (c 1.00, CHCl3) (92% ee).

¹H NMR (500 MHz, CDCl3): δ = 0.79 (3 H, t, J = 7.3 Hz), 1.05-1.18 (4 H, m), 1.41 (2 H, m), 2.18 (1 H, dt, J = 7.3, 16.7 Hz), 2.30 (1 H, dt, J = 7.5, 16.7 Hz), 2.42 (3 H, s), 2.97-3.09 (2 H, m), 3.23 (3 H, s), 3.38 (3 H, s), 3.79 (1 H, m), 4.67 (1 H, d, J = 6.6 Hz), 7.03-7.09 (2 H, m), 7.22 (1 H, m), 7.63 (1 H, d, J = 7.5 Hz), 7.83 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 12.1, 13.8, 22.3, 23.2, 31.2, 36.0, 42.7, 43.4, 54.2, 54.8, 106.9, 110.0, 110.4, 119.0, 119.1, 120.6, 127.4, 132.6, 135.4, 210.3.

HRMS (EI): m/z calcd for C20H29NO3: 331.2147; found: 331.2139.

The ee value was determined by HPLC analysis using a Chiralpak AD-H (1 mL/min, 8.0%, i-PrOH in hexane); t R: 9.7 min (major S-enantiomer) and 17.3 min (minor R-enantiomer). The absolute stereochemistry was assumed by analogy.

( S )-3-(2-Methyl-1 H -indol-3-yl)-1-phenylbutan-1-one (16ik) [¹³a]

[α]D ²8 +23.5 (c 2.50, CHCl3) (37% ee) {Lit.¹³a [α]D -57.4 (c 1.00, CHCl3) for R-enantiomer (84% ee)}.

¹H NMR (500 MHz, CDCl3): δ = 1.53 (3 H, d, J = 7.1 Hz), 2.38 (3 H, s), 3.40 (1 H, dd, J = 7.5, 16.2 Hz), 3.56 (1 H, dd, J = 6.5, 16.2 Hz), 3.78 (1 H, sext, J = 7.1 Hz), 7.11 (2 H, m), 7.25 (1 H, m), 7.40 (2 H, m), 7.51 (1 H, m), 7.71 (1 H, m), 7.75 (1 H, br s), 7.90 (2 H, m).

¹³C NMR (125.8 MHz, CDCl3): δ = 12.0, 21.0, 27.3, 45.6, 110.5, 115.4, 118.9, 119.0, 120.6, 127.0, 128.0, 128.4, 130.3, 132.7, 135.5, 137.2, 200.0.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 6% i-PrOH in hexane); t R: 40.2 min (minor R-enantiomer) and 46.8 min (major S-enantiomer).

3-(2-Methyl-1 H -indol-3-yl)cyclohexanone (16il) [55]

[α]D ³¹ +11.08 (c 2.02, CHCl3) (55% ee).

¹H NMR (500 MHz, CDCl3): δ = 1.80 (1 H, m), 2.01 (1 H, m), 2.22 (1 H, m), 2.31-2.38 (4 H, m), 2.47-2.53 (3 H, m), 2.99 (1 H, t, J = 13.7 Hz), 3.21 (1 H, tt, J = 3.8, 12.9 Hz), 7.07 (1 H, t, J = 7.1 Hz), 7.12 (1 H, t, J = 7.8 Hz), 7.29 (1 H, d, J = 8.0 Hz), 7.66 (1 H, d, J = 7.8 Hz), 7.76 (1 H, br s).

¹³C NMR (125.8 MHz, CDCl3): δ = 12.1, 26.1, 31.5, 37.3, 41.5, 48.2, 110.6, 113.9, 118.9, 119.1, 121.0, 126.9, 130.1, 135.4, 211.7.

HRMS (EI): m/z calcd for C15H17NO: 227.1310; found: 227.1315.

The ee value was determined by HPLC analysis using a Chiralcel OD column (1 mL/min, 3.0%, i-PrOH in hexane); t R: 65.0 min (minor), and 70.0 min (major).

2,3-Dihydro-1 H ,1′ H -[2,3′]biindolyl (17) [³³]

¹H NMR (500 MHz, CDCl3): δ = 3.23 (1 H, dd, J = 8.3, 15.6 Hz), 3.49 (1 H, dd, J = 9.1, 15.6 Hz), 5.27 (1 H, t, J = 8.6 Hz), 6.69 (1 H, d, J = 7.7 Hz), 6.79 (1 H, t, J = 7.4 Hz), 7.09-7.18 (4 H, m), 7.22 (1 H, m), 7.36 (1 H, d, J = 8.2 Hz), 7.61 (1 H, d, J = 7.9 Hz), 7.99 (1 H, br s)

¹³C NMR (125.8 MHz, CDCl3): δ = 37.7, 56.4, 109.2, 111.3, 118.8, 119.4, 119.5, 119.6, 121.7, 122.3, 124.7, 125.8, 127.5, 128.9, 136.8, 150.9.

Preparation of OXB 9a by Method B

To a solution of 18 (75.5 mg, 0.20 mmol) in CH2Cl2 (2 mL) under argon r.t. was added dichlorophenylborane (28.5 µL, 0.220 mmol). The mixture was stirred for 30 min. To the resulting mixture was added allyltrimethylsilane (91.4 mg, 0.80 mmol) at r.t. After stirring for 14-17 h, the mixture was concentrated in vacuo to give 9a.

6-(1-Methyl-1 H -indol-2-ylmethoxy)hex-4-en-3-one (19)

To a solution of 2-allyloxymethyl-1-methyl-1H-indole [³4b] (1.00 g, 5.0 mmol) and hex-4-en-3-one (1.47 g, 15.0 mmol) in CH2Cl2 at r.t. was added 2nd generation Grubbs catalyst (84.9 mg, 0.1 mmol). The reaction mixture was heated to reflux for 2 h. The mixture was concentrated in vacuo and the residue was purified by flash chromatography (SiO2, 10-50% EtOAc in hexane) to give 0.86 g (67%) of 19.

¹H NMR (500 MHz, CDCl3): δ = 1.09 (3 H, t, J = 7.3 Hz), 2.55 (2 H, q, J = 7.3 Hz), 3.79 (3 H, s), 4.15 (2 H, dd, J = 2.0, 4.5 Hz), 4.73 (2 H, s), 6.33 (1 H, td, J = 2.0, 16.0 Hz), 6.49 (1 H, s), 6.78 (1 H, td, J = 4.5, 16.0 Hz), 7.11 (1 H, t, J = 7.1 Hz), 7.24 (1 H, m), 7.33 (1 H, d, J = 8.2 Hz), 7.59 (1 H, d, J = 7.9 Hz).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.93, 29.9, 33.7, 64.7, 67.8, 103.4, 109.2, 119.6, 120.8, 122.1, 127.0, 129.2, 134.9, 138.2, 141.4, 200.6.

HRMS (EI): m/z calcd for C16H19NO2: 257.1416; found: 257.1417.

( R )-1-(9-Methyl-1,3,4,9-tetrahydropyrano[3,4- b ]indol-4-yl)butan-2-one (20)

To a mixture of OXB 9a (0.10 mmol), prepared by Method A, in anhyd Et2O (4.7 mL) at -85 ˚C were added N,N-dimethylaniline (4.6 mg, 0.038 mmol), and 19 (128.7 mg, 0.50 mmol). After stirring at -85 ˚C for 7 d, the reaction mixture was quenched with sat. aq NaHCO3 (10 mL) and filtered. The filtrate was extracted with EtOAc (3 × 20 mL), and the combined organic extracts were dried (Na2SO4) and concentrated in vacuo. Purification of the residue by flash chromatography (SiO2, 2-3% EtOAc in toluene) gave 77 mg (60%) of 20 (90% ee); [α]D ²6 -24.6 (c 1.53, CHCl3).

¹H NMR (500 MHz, CDCl3): δ = 1.08 (3 H, t, J = 7.3 Hz), 2.43 (2 H, q, J = 7.3 Hz), 2.86 (1 H, dd, J = 9.4, 17.6 Hz), 2.93 (1 H, dd, J = 4.2, 17.6 Hz), 3.53 (1 H, m), 3.58 (3 H, m), 3.85 (1 H, dd, J = 3.4, 11.4 Hz), 3.98 (1 H, dd, J = 1.8, 11.4 Hz), 4.78 (1 H, dd, J = 1.2, 14.6 Hz), 4.87 (1 H, d, J = 14.6 Hz), 7.10 (1 H, t, J = 7.1 Hz), 7.20 (1 H, m), 7.29 (1 H, d, J = 8.2 Hz), 7.47 (1 H, d, J = 7.9 Hz).

¹³C NMR (125.8 MHz, CDCl3): δ = 7.8, 28.5, 29.4, 36.7, 45.5, 63.1, 69.7, 108.8, 109.4, 118.1, 119.2, 121.2, 125.7, 133.0, 136.9, 211.1.

HRMS (EI): m/z calcd for C16H19NO2: 257.1416; found: 257.1421.

The ee value was determined by HPLC analysis using a Chiralcel AS-H column (1 mL/min, 3% i-PrOH in hexane); t R: 20.6 min (major) and 31.0 min (minor). The absolute stereochemistry was assumed by analogy.

Crossover Experiment; ( S )-4-Deutero-5-(1-methyl-1 H -indol-3-yl)hexan-3-one (16fb- d ) and ( S )-4-Deutero-5-(1-allyl-1 H -indol-3-yl)hexan-3-one (16gb- d ); Typical Procedure (Table 6, entry 1)

To a mixture of OXB 9a (0.10 mmol), prepared by Method A, in anhyd Et2O (1.6 mL) at -85 ˚C were added 1-allylindole (15g; 78.6 mg, 0.500 mmol), 3-deutero-1-methylindole (15f-d; 66.1 mg, 0.500 mmol), N,N-dimethylaniline (3.05 mg, 0.0250 mmol), and hex-4-en-3-one (12b; 98 mg, 1.00 mmol). After stirring at -85 ˚C for 20 h, the reaction mixture was quenched with sat. aq NaHCO3 (10 mL) and filtered. The filtrate was extracted three times with EtOAc (3 × 20 mL), dried (Na2SO4), and concentrated in vacuo. Purification of the residue by flash chromatography (SiO2, 5-12% EtOAc in hexane) gave 108 mg (47%) of 16fb-d (91% ee, 72%-d, syn/anti = 85:15) and 123 mg (48%) of 16gb-d (84% ee, 22%-d, syn/anti = 86:14).

( S )-16fb- d

¹H NMR (500 MHz, C6D6): δ = 0.99 (3 H, t, J = 7.3 Hz), 1.48 (3 H, m), 2.02 (2 H, q, J = 7.3 Hz), 2.46-2.52 (0.39 H for H syn , m), 2.77-2.82 (0.89 H for H anti , m), 3.06 (3 H, s), 3.84 (1 H, m), 6.52 (1 H, s), 7.14 (1 H, d, J = 8.1 Hz), 7.30 (1 H, m), 7.36 (1 H, m), 7.80 (1 H, d, J = 7.8 Hz).

MS (EI): m/z (%) = 231 (M - d 1 + +1 and M+ + 2, 13.0), 230 (M - d 1 + and M+ + 1, 81.6), 229 (M+, 34.2), 158 (100).

( S )-16gb- d

¹H NMR (500 MHz, C6D6): δ = 0.96 (3 H, t, J = 7.3 Hz), 1.45 (3 H, m), 2.00 (2 H, q, J = 7.3 Hz), 2.45-2.50 (0.81 H for H syn , m), 2.74-2.78 (0.97 H for H anti , m), 3.81 (1 H, m), 4.15 (2 H, m), 4.83 (1 H, m), 4.93 (1 H, m), 5.63 (1 H, m), 6.66 (1 H, s), 7.19 (1 H, d, J = 8.1 Hz), 7.24-7.76 (2 H, m), 7.77 (1 H, d, J = 7.8 Hz).

MS (EI): m/z (%) = 257 (M - d 1 + + 1 and M+ + 2, 6.4), 256 (M - d 1 + and M+ + 1, 41.8), 255 (M+, 90.1), 184 (100).

Acknowledgment

This work was supported partially by KAKENHI (17550101).

36

As 15f-d is more reactive than 15g, the deuterium contents of the products were influenced by the conversion of the reaction.

38

Alternatively, protonation of 23 at the enolate oxygen atom may afford the enol form of product 13 or 16, which undergoes tautomerization to the keto form with H/D scrambling. However, the effect of the amine additive is difficult to be rationalized by this pathway.

36

As 15f-d is more reactive than 15g, the deuterium contents of the products were influenced by the conversion of the reaction.

38

Alternatively, protonation of 23 at the enolate oxygen atom may afford the enol form of product 13 or 16, which undergoes tautomerization to the keto form with H/D scrambling. However, the effect of the amine additive is difficult to be rationalized by this pathway.

Scheme 1

Scheme 2

Figure 1 Structures of 9 and 10

Scheme 3

Scheme 4

Scheme 5

Figure 2  syn/anti-Vicinal couplings in 16fb

Scheme 6

Scheme 7