Synthesis of 3,4,5-Trisubstituted Isoxazoles through Gold-Catalyzed Cascade Cyclization-Oxidative Alkynylation and Cyclization-Fluorination of 2-Alkynone O-Methyloximes

Abstract

Gold-catalyzed tandem cyclization-oxidative alkynylation and cyclization-fluorination reactions of 2-alkynone O-methyloximes are described. The reactions proceed smoothly at room temperature in the presence of 10 mol % of $(PPh_3)AuNTf_2$, 2.5 equivalents of selectfluor, and 2 equivalents of $K_3PO_4$. 2-Alkynone O-methyloximes undergo intramolecular oxyauration/cyclization and ensuing oxidative cross-coupling and fluorination process to afford the corresponding 3,4,5-trisubstituted isoxazoles in a cascade manner.

Introduction

Isoxazole1 is an important scaffold frequently found in many natural products and biologically active compounds such as analgesics,2 antibiotics,3 antidepressants,4 and anticancer agent.5 It has also been used as a synthetic intermediate, building block and chiral ligand in organic synthesis. 5 The prevailing methods for the synthesis of isoxazoles mostly involve the condensation between a hydroxylamine and a 1,3-dicarbonyl compound,6 the condensation between a hydroxylamine and α,β-unsaturated carbonyl compound,7 and the [3+2] cycloaddition reaction between an alkyne and a nitrile oxide.8 These methods usually require harsh conditions and suffer from poor chemo- and regio-selectivities. Furthermore, they are unsuitable for the selective synthesis of fully substituted isoxazoles. So far, only a few methodologies for the selective synthesis of fully substituted isoxazoles have been reported. Larock and co-workers reported a stepwise synthesis of trisubstituted isoxazoles that featured the iodocyclization of 2-alkynone O-methyloximes and a subsequent palladium-catalyzed coupling reaction.9 Harrity and co-workers demonstrated a stepwise reaction involving [3+2] cycloaddition reaction of nitrile oxides and alkynylboronates to provide isoxazole boronic esters, followed by Suzuki coupling reaction.10 Denmark and co-workers also reported a [3+2] cycloaddition reaction between alkynyldimethylsilyl ethers and nitrile oxides to produce isoxazolylsilanols, the cross-coupling reactions of which afforded 3,4,5-trisubstituted isoxazoles.11 In these three cases, extra functionalization at the 4-position, such as iodination, borylation, or silylation, is required and is followed by crosscoupling reaction with a genuine substituent in a stepwise manner. Considering “atom economy”12 to prevent wasting atoms and “pot economy”13 to target the inherent efficiency by avoiding intermediate isolation, the direct synthesis of trisubstituted isoxazoles in one sequence is highly desirable and challenging. However, cycloaddition reactions with internal alkynes to afford trisubstituted isoxazoles are limited to a handful of highly activated alkynes (e.g., acetylene dicarboxylate and related electron-deficient acetylenes) and suffer from poor regioselectivities14 as mentioned previously. Facile one-pot domino reactions have rarely been reported. 15 The lack of direct synthetic methods for the fully substituted isoxazoles has become more acute due to increasing needs and applications in the pharmaceutical industry.

In recent years, gold catalysts have drawn significant attention due to their applicability in many organic transformations, which originates from their promising properties such as air- and moisture-stability, a broad range of functional group compatibilities, superior π-acidity, fine-tunable reactivity with various ligands, and mild reaction conditions. 16 Interestingly, gold rarely changes oxidation state during the reaction. Hence, for a long time, gold-catalyzed cross-coupling reactions were not considered feasible although many organic transformations are catalyzed by gold. However, more recently, it was discovered that gold was able to undergo Au(I)/Au(III) redox cycles in the presence of an external oxidant such as Selectfluor, which enabled crosscoupling reactions.17 Therefore, to develop novel direct synthetic methods for the fully substituted isoxazoles, we envisioned cyclization of 2-alkynone O-methyloxime 1a catalyzed by π-acidic gold, leading to the cationic gold intermediate 2; the subsequent oxidative alkynylation might provide trisubstituted isoxazole 4 in a cascade manner (Scheme 1). This “pot economic” transformation involves gold-catalyzed cyclization reaction combined with oxidative cross-coupling in one-pot, which allows direct access to the highly functionalized isoxazoles.

Scheme 1.Strategy for cascade cyclization–oxidative alkynylation.

 

Experimental

General Remarks. Reagents were purchased in Sigma-Aldrich Ltd. Alfa Aesar, Strem chemicals Inc. or Acros organics, and used without further purification. All solvents were distilled before used under dry Ar over appropriate drying agents (sodium benzophenone ketyl, metal hydride). Reactions were performed in an oven-dried glassware under positive Ar pressure with magnetic stirring. TLC was performed on 0.25 mm E. Merck silica gel 60 F254 plates and visualized under UV light (254 nm) or by staining with p-anisaldehyde or cerium ammonium molybdate (CAM). Flash chromatography was performed on E. Merch 230−400 mesh silica gel 60. NMR spectra were recorded on Varian Unity 400 instruments at 24 ℃. Chemical shifts are expressed in ppm relative to TMS (1H, 0 ppm), or CDCl3 (1H, 7.26 ppm; 13C, 77.2 ppm); coupling constants are expressed in Hz. High resolution mass spectra electrospray ionization (HRMS-ESI) were obtained on an Agilent technologies 6220 TOF LC/MS spectrometer.

Synthesis of (Z)-4-Phenylbut-3-yn-2-one O-methyloxime (1a). In an oven-dried roundbottom flask, NH2OCH3·HCl (263 mg, 3.20 mmol) and NaOAc (258 mg, 3.20 mmol) were dissolved in anhydrous EtOH (10.5 mL) under Ar atmosphere. 4-Phenyl-3-butyn-2-one (310 μL, 2.10 mmol) was added and the resulting suspension was stirred at rt. After 1.5 h, EtOH was evaporated. Then, the residue was diluted with H2O (20 mL) and extracted with EtOAc (3 × 20 mL). The combined organic extracts were dried over anhydrous MgSO4, filtered, and evaporated. Column chromatography (60:1 hexane/EtOAc) afforded 1a (191 mg, 52%) as a colorless liquid. TLC: Rf 0.14 (14:1 hexane/EtOAc); IR (KBr film) 2936, 2223, 1599, 1490, 1375, 1313, 1170, 1061, 757, 690 cm−1; 1H-NMR (400 MHz, CDCl3) δ 7.54−7.51 (m, 2H), 7.40−7.32 (m, 3H), 3.97 (s, 3H), 2.13 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 137.7, 132.3, 129.6, 128.6, 121.9, 99.3, 81.2, 62.5, 20.8; LRMS (ESI) m/z (rel int): 174 ([M + H]+, 100), 128 ([M − NOCH3 ]+, 46); HRMS (ESI) m/z calcd for C11H11NO 173.0841, found 173.0840.

Representative Procedure for the Synthesis of 3-Methyl- 5-phenyl-4-phenylethynylisoxazole (4a). To a stirring solution of K3PO4 (170 mg, 800 μmol), PPh3AuNTf2 (29.6 mg, 40.0 μmol) and Selectfluor (354 mg, 1.00 mmol) in MeCN (4 mL) were added (Z)-4-phenylbut-3-yn-2-one O-methyloxime (1a) (69.3 mg, 400 μmol), phenylacetylene (3a) (88.0 μL, 800 μmol) and H2O (72.0 μL) sequentially under Ar atmosphere. The resulting suspension was stirred at rt for 17 h. After the completion of the reaction, the reaction mixture was diluted with H2O (20 mL) and extracted with EtOAc (3 × 20 mL). The combined organic extracts were dried over anhydrous MgSO4, filtered, and evaporated. Purification by column chromatography (35:1 hexane/EtOAc) afforded 4a(52 mg, 50%) as a yellow solid, 5 (12.6 mg, 20%) as a yellow solid, 6 (12.9 mg, 18%) as a yellow liquid, and 7a (43.6 mg, 27%) as a white solid.

4a: mp 67−69 ℃ (from EtOAc); TLC: Rf 0.20 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.20−8.17 (m, 2H), 7.57−7.55 (m, 2H), 7.53−7.47 (m, 3H), 7.40−7.39 (m, 3H), 2.44 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 168.4, 162.0, 131.7, 130.8, 129.1, 129.0, 128.7, 127.5, 126.4, 122.9, 99.3, 96.3, 78.9, 10.8; LRMS (ESI) m/z (rel int): 260 ([M + H]+, 84), 219 ([M – CH2CN]+, 100); HRMS (ESI) m/z calcd for C18H13NO 259.0997, found 259.0997.

5: mp 76−78 ℃ (from EtOAc) (Lit.,25 65−68 ℃); TLC: Rf 0.13 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 7.77−7.75 (m, 2H), 7.48−7.42 (m, 3H), 6.36 (s, 1H), 2.36 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 169.9, 160.6, 130.2, 129.1, 127.8, 126.0, 100.4, 11.7; LRMS (ESI) m/z (rel int): 160 ([M + H]+, 100); HRMS (ESI) m/z calcd for C10H9NO 159.0684, found 159.0682

6: TLC: Rf 0.26 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 7.82−7.79 (m, 2H), 7.51−7.47 (m, 2H), 7.45 −7.41 (m, 1H), 2.37 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 152.3 (d, JC-F = 9.4 Hz), 152.1 (d, JC-F = 7.1 Hz), 142.5 (d, JC-F = 253.9 Hz), 130.1, 129.2, 126.2 (d, JC-F = 5.0 Hz), 125.4 (d, JC-F = 3.4 Hz), 8.9 (d,JC-F = 2.9 Hz); 1F-NMR (376 MHz, CDCl3) δ –180.5 (s, 1F); LRMS (ESI) m/z (rel int): 178 ([M + H]+, 2), 105 ([M – NCCH3CF]+, 100). HRMS (ESI) m/z calcd for C10H8FNO 177.0590, found 177.0590.

7a: mp 85−87 ℃ (from EtOAc) (Lit.,26 82−85 ℃); TLC: Rf 0.44 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 7.55−7.52 (m, 4H), 7.40−7.32 (m, 6H); 13C-NMR (100 MHz, CDCl3) δ 132.7, 129.4, 128.6, 122.0, 81.8, 74.1; LRMS (ESI) m/z (rel int): 405 ([2M + H]+, 8). 203 ([M + H]+, 28); HRMS (ESI) m/z calcd for C16H10 202.0783, found 202.0784.

Synthesis of 4-(2-Chlorophenylethynyl)-3-methyl-5-phenylisoxazole (4b). 1-Chloro-2-ethynylbenzene (3b) was used. The suspension was stirred for 25 h at rt. Column chromatography (50:1 hexane/EtOAc) afforded 4b (55.0 mg, 47%) as a yellow solid, 5 (13.8 mg, 22%) as a yellow solid, 6 (16.5 mg, 23%) as a yellow liquid, and 7b (47 mg, 22%) as a white solid.

4b: mp 71−73 ℃ (from EtOAc); TLC: Rf 0.15 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.27−8.25 (m, 2H), 7.59−7.57 (m, 1H), 7.53−7.46 (m, 4H), 7.34−7.26 (m, 2H), 2.47 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 168.8, 162.1, 135.9, 133.3, 130.9, 129.9, 129.7, 129.0, 127.3, 126.8, 126.6, 123.0, 99.1, 93.2, 84.1, 10.8; LRMS (ESI) m/z (rel int): 294 ([M + H]+, 100), 253 ([M – CH2CN]+, 53); HRMS (ESI) m/z calcd for C18H12ClNO 293.0607, found 293.0608.

7b: mp 142−144 ℃ (from EtOAc) (Lit.,27 138−140 ℃); TLC: Rf 0.39 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 7.58 (ddd, 2H, J = 8.0, 1.6, 0.4 Hz), 7.42 (ddd, 2H, J = 8.0, 1.6, 0.4 Hz), 7.31 (td, 2H, J = 8.0, 1.6 Hz), 7.24 (td, 2H, J = 8.0, 1.6 Hz); 13C-NMR (100 MHz, CDCl3) δ 137.2, 134.6, 130.5, 129.7, 126.8, 122.0, 79.6, 78.6; LRMS (ESI) m/z (rel int): 271 ([M + H]+, 100); HRMS (ESI) m/z calcd for C16H8Cl2 270.0003, found 270.0004.

Synthesis of 4-(3-Chlorophenylethynyl)-3-methyl-5-phenylisoxazole (4c). 1-Chloro-3-ethynylbenzene (3c) was used. The suspension was stirred for 17 h at rt. Column chromatography (40:1 hexane/EtOAc) afforded 4c (40.0 mg, 34%) as a yellow solid, 5 (19.0 mg, 30%) as a yellow solid, 6 (15.0 mg, 21%) as a yellow liquid, and 7c (23.0 mg, 11%) as a white solid.

4c: mp 81−83 ℃ (from EtOAc); TLC: Rf 0.25 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.16−8.14 (m, 2H), 7.54−7.48 (m, 4H), 7.34 (dt, 1H, J = 7.6, 2.0 Hz), 7.38−7.30 (m, 2H), 2.43 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 168.9, 162.0, 134.6, 131.5, 131.0, 130.0, 129.8, 129.2, 129.1, 127.4, 126.5, 124.6, 98.8, 94.7, 80.1, 10.8; LRMS (ESI) m/z (rel int): 294 ([M + H]+, 100), 253 ([M – CH2CN]+, 53); HRMS (ESI) m/z calcd for C18H12ClNO 293.0607, found 293.0610

7c: mp 78−80 ℃ (from EtOAc) (Lit.,26 73 ℃); TLC: Rf 0.5 (14:1 hexane/EtOAc); IR (KBr film) 3066, 1583, 1557, 1470, 1455, 1401, 1261, 1164, 1093, 1076, 907, 890, 850, 793, 776, 675, 522 cm−1; 1H-NMR (400 MHz, CDCl3) δ 7.51 (m, 2H), 7.41 (d, 2H, J = 8.0 Hz), 7.38−7.32 (m, 2H), 7.28 (t, 2H, J = 8.0 Hz); 13C-NMR (100 MHz, CDCl3) δ 134.6, 132.5, 130.9, 129.9, 129.4, 123.5, 80.8, 74.9.

Synthesis of 4-(3,5-Difluorophenylethynyl)-3-methyl-5-phenylisoxazole (4d). K3PO4 (85.0 mg, 400 μmol), PPh3AuNTf2 (14.8 mg, 20.0 μmol), Selectfluor (177 mg, 500 μmol), MeCN (2 mL), (Z)-4-phenylbut-3-yn-2-one O-methyloxime (1a) (34.6 mg, 200 μmol), 1-ethynyl-3,5-difluorobenzene (3d) (48.0 μL, 400 μmol) and H2O (36.0 μL) were used. The suspension was stirred for 22 h at rt. Column chromatography (40:1 hexane/EtOAc) afforded 4d (26.8 mg, 45%) as a yellow solid, 5 (6.8 mg, 21%) as a yellow solid, 6 (9.2 mg, 26%) as a yellow liquid, and 7d (17.0 mg, 16%) as a white solid.

4d: mp 99−101 ℃ (from EtOAc); TLC: Rf 0.24 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.14−8.12 (m, 2H), 7.54−7.48 (m, 3H), 7.07−7.04 (m, 2H), 6.86 (tt, 1H, J = 8.8, 2.0 Hz), 2.43 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 169.3, 163.0 (dd, JC-F = 243.5, 11.7 Hz), 132.3, 131.1, 129.6, 129.2, 128.6, 127.3, 126.5, 114.7 (dd, JC-F = 19.0, 7.2 Hz), 105.1 (t, JC-F = 25.6 Hz), 81.0, 62.5, 20.8; LRMS (ESI) m/z (rel int): 296 ([M + H]+, 100), 255 ([M – CH2CN]+, 35); HRMS (ESI) m/z calcd for C18H11F2NO 295.0809, found 295.0809.

7d: mp 87−89 ℃ (from EtOAc); TLC: Rf 0.62 (14:1 hexane/EtOAc); IR (KBr film) 3100, 1610, 1583, 1456, 1402, 1094, 1077, 960, 890, 794, 681 cm−1; 1H-NMR (400 MHz, CDCl3) δ 7.07−7.02 (m, 4H), 6.90−6.84 (tt, 2H, J = 8.8, 2.0 Hz); 13C-NMR (100 MHz, CDCl3) δ 162.9 (dd, JC-F = 250, 13.5 Hz), 124.1 (t, JC-F = 11.9 Hz), 115.8 (dd, JC-F = 19.1, 8.0 Hz), 106.2 (t, JC-F = 25.4 Hz), 80.2 (d, JC-F = 4.0 Hz), 75.2; 19F-NMR (376 MHz, CDCl3) δ −108.7 (t, 4F, J= 8.3 Hz).

Synthesis of 4-(3-Fluorophenylethynyl)-3-methyl-5-phenylisoxazole (4e). K3PO4 (85.0 mg, 400 μmol), PPh3AuNTf2 (14.8 mg, 20.0 μmol), Selectfluor (177 mg, 500 μmol), MeCN (2 mL), (Z)-4-phenylbut-3-yn-2-one O-methyloxime (1a)(34.6 mg, 200 μmol), 1-ethynyl-3-fluorobenzene (3e) (46.0 μL, 400 μmol) and H2O (36.0 μL) were used. The suspension was stirred for 20 h at 60 ℃. Column chromatography (40:1 hexane/EtOAc) afforded 4e (23.0 mg, 41%) as a yellow solid, 5 (8.7 mg, 27%) as a yellow solid, 6 (9.1 mg, 26%) as a yellow liquid, and 7e (18.2 mg, 19%) as a white solid.

4e: mp 79−81 ℃ (from EtOAc); TLC: Rf 0.22 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.17−8.14 (m, 2H), 7.54−7.47 (m, 3H), 7.39−7.32 (m, 2H), 7.27−7.23 (m, 1H), 7.12−7.07 (m, 1H), 2.44 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 168.9, 162.7 (d, JC-F = 246.0 Hz), 162.0, 131.0, 130.4 (d, JC-F = 8.7 Hz), 129.1, 127.6 (d, JC-F = 2.9 Hz), 127.4, 126.5, 124.7 (d, JC-F = 9.4 Hz), 118.5 (d, JC-F = 22.9 Hz), 116.4 (d, JC-F = 21.6 Hz), 98.9, 94.9, 79.9, 10.8. LRMS (ESI) m/z (rel int): 278 ([M + H]+, 100), 237 ([M – CH2CN]+, 62); HRMS (ESI) m/z calcd for C18H12FNO 277.0903, found 277.0903.

7e: mp 122−124 ℃ (from EtOAc) (Lit.,28 122−123 ℃); TLC: Rf 0.40 (14:1 hexane/EtOAc); IR (KBr film) 3030, 1604, 1583, 1482, 1461, 1428, 1265, 1130, 1078, 933, 868, 782 cm−1; 1H-NMR (400 MHz, CDCl3) δ 7.33−7.30 (m, 4H), 7.21−7.20 (m, 2H), 7.12−7.07 (m, 2H); 13C-NMR (100 MHz, CDCl3) δ 162.5 (d, JC-F = 240.8 Hz), 130.4 (d, JC-F = 8.7 Hz), 128.7 (d, JC-F = 3.6 Hz), 123.6 (d, JC-F= 9.4 Hz), 119.5 (d,JC-F = 23.0 Hz), 117.1 (d, JC-F = 21.6 Hz), 80.8 (d, JC-F = 2.8 Hz), 74.6; 19F-NMR (376 MHz, CDCl3) δ –112.2− –112.3 (m, 2F).

Synthesis of 4-(2,4-Difluorophenylethynyl)-3-methyl-5-phenylisoxazole (4f). K3PO4 (85.0 mg, 400 μmol), PPh3-AuNTf2 (14.8 mg, 20 μmol), Selectfluor (177 mg, 500 μmol), MeCN (2 mL), (Z)-4-phenylbut-3-yn-2-one O-methyloxime (1a) (34.6 mg, 200 μmol), 1-ethynyl-2,4-difluorobenzene (3f) (55.0 μL, 400 μmol) and H2O (36.0 μL) were used. The suspension was stirred for 24 h at 60 ℃. Column chromatography (40:1 hexane/EtOAc) afforded 4f (17.0 mg, 30%) as a yellow solid, 5 (2.7 mg, 8%) as a yellow solid, 6 (8.4 mg, 24%) as a yellow liquid, and 7f (9.1 mg, 8%) as a yellow solid.

4f: mp 87−89 ℃ (from EtOAc); TLC: Rf 0.36 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.21−8.18 (m, 2H), 7.54−7.48 (m, 4H), 6.95−6.90 (m, 2H), 2.44 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 168.9, 164.5 (d, JC-F = 11.2 Hz), 163.3 (dd,JC-F = 240.0, 19.0 Hz), 161.9, 134.1 (dd, JC-F = 9.9, 2.8 Hz), 131.0, 129.1, 127.3, 126.5, 118.8, 112.0 (dd, JC-F = 21.8, 3.5 Hz), 104.7 (t, JC-F = 25.0 Hz), 98.9, 88.7, 83.9, 10.8; LRMS (ESI) m/z (rel int): 296 ([M + H]+, 100), 255 ([M – CH2CN]+, 63); HRMS (ESI) m/z calcd for C18H11F2NO 295.0809, found 295.0811.

7f: mp 168−170 ℃ (from EtOAc) (Lit.,29 187−188 ℃); TLC: Rf 0.48 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 7.53−7.48 (m, 2H), 6.91−6.84 (m, 4H); 13C-NMR (100 MHz, CDCl3) δ 164.6 (dd, JC-F = 205.0, 11.0 Hz), 163.7 (dd, JC-F = 202.0, 9.1 Hz), 135.5 (dd, JC-F = 7.9, 1.8 Hz), 112.2 (dd, JC-F = 17.6, 3.0 Hz), 107.0 (dd, JC-F = 12.8, 3.7 Hz), 104.8 (t, JC-F = 20.6 Hz), 78.3, 75.1; LRMS (ESI) m/z (rel int): 275 ([M + H]+, 100); HRMS (ESI) JC-F calcd for C16H6F4 274.0406, found 274.0404.

Synthesis of 4-(2-Fluorophenylethynyl)-3-methyl-5-phenylisoxazole (4g). 1-Ethynyl-2-fluorobenzene (3g) was used. The suspension was stirred for 20 h at rt. Column chromatography (50:1 hexane/EtOAc) afforded 4g (40.3 mg, 36%) as a yellow solid, 5 (11.9 mg, 19%) as a yellow solid, 6 (3.7 mg, 5%) as a yellow liquid, and 7g (49.9 mg, 26%) as a yellow solid.

4g: mp 68−70 ℃ (from EtOAc); TLC: Rf 0.33 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.23−8.20 (m, 2H), 7.55−7.46 (m, 4H), 7.40−7.34 (m, 1H), 7.17 (td, 1H, J = 7.6, 1.2 Hz), 7.16−7.13 (m, 1H), 2.45 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 168.8, 162.9 (d, JC-F = 250.3 Hz), 162.0, 133.2, 130.9, 130.7 (d, JC-F = 7.6 Hz), 129.1, 127.4, 126.5, 124.3 (d, JC-F = 3.1 Hz), 115.9 (d, JC-F = 20.6 Hz), 111.7 (d, JC-F = 16.0 Hz), 99.1, 89.9, 84.1 (d, JC-F = 3.8 Hz), 10.8; LRMS (ESI) m/z (rel int): 278 ([M + H]+, 100), 236 ([M – CH3CN]+, 80); HRMS (ESI) m/z calcd for C18H12FNO 277.0903, found 277.0904.

7g: mp 144−146 ℃ (from EtOAc); TLC: Rf 0.42 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 7.52 (td, 2H, J = 7.6, 2.0 Hz), 7.39−7.34 (m, 2H), 7.13 (td, 2H, J = 7.6, 1.2 Hz), 7.11−7.08 (m, 2H); 13C-NMR (100 MHz, CDCl3) δ 164.0 (d, JC-F = 251.8 Hz), 134.5, 131.3 (d, JC-F = 7.6 Hz), 124.3 (d, JC-F = 3.1 Hz), 115.9 (d, JC-F = 20.6 Hz), 110.7 (d, JC-F = 15.3 Hz), 78.6, 76.1; LRMS (ESI) m/z (rel int): 477 ([2M + H]+, 6), 239 ([M + H]+, 100). HRMS (ESI) m/z calcd for C16H8F2 238.0594, found 238.0591.

Synthesis of 4-(3,5-Bis(trifluoromethyl)phenylethynyl)-3-methyl-5-phenylisoxazole (4h). 1-Ethynyl-3,5-bis(trifluoromethyl) benzene (3h) was used. The suspension was stirred for 18 h at rt. Column chromatography (50:1 hexane/EtOAc) afforded 4h (48.2 mg, 30%) as a white solid, 5 (12.4 mg, 19%) as a yellow solid, 6 (10 mg, 14%) as a yellow liquid, and 7h (36.2 mg, 10%) as a brown solid.

4h: mp 124−126 ℃ (from EtOAc); TLC: Rf 0.19 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.14−8.11 (m, 2H), 7.95 (s, 2H), 7.87 (s, 1H), 7.57−7.48 (m, 3H), 2.46 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 169.8, 161.9, 132.3 (q, JC-F = 33.0 Hz), 131.5 (d, JC-F= 3.5 Hz), 131.3, 129.3, 127.1, 126.6, 125.3, 123.1 (q,JC-F= 270.0 Hz), 122.3 (m), 98.1, 92.9, 82.6, 10.8; LRMS (ESI) m/z (rel int): 396 ([M + H]+, 91); HRMS (ESI) m/z calcd for C20H11F6NO 395.0745, found 395.0746.

7h: mp 146−148 ℃ (from EtOAc) (Lit.,30 158.7−161.0 ℃); TLC: Rf 0.55 (14:1 hexane/EtOAc); IR (KBr film) 3095, 2219, 1615, 1463, 1281, 1262, 1167, 1107, 1022, 899, 801, 699, 684 cm−1; 1H-NMR (400 MHz, CDCl3) δ 7.55−7.52 (m, 4H), 7.40−7.32 (m, 2H); 13C-NMR (100 MHz, CDCl3) δ 132.5 (q, JC-F = 33.7 Hz), 132.6 (d, JC-F = 2.9 Hz), 123.8, 123.3 (t, JC-F = 3.6 Hz), 121.6, 118.8, 79.9, 76.3; 19F-NMR (376 MHz, CDCl3): δ –63.3 (s, 12F).

Synthesis of 4-(2-(3-Methyl-5-phenylisoxazol-4-yl)ethynyl) benzonitrile (4i). 4-Ethynylbenzonitrile (3i) was used. The suspension was stirred for 40 h at rt. Column chromatography (40:1 hexane/EtOAc) afforded 4i (36.7 mg, 32%) as a yellow solid, 5 (10 mg, 16%) as a yellow solid, 6 (15.4 mg, 22%) as a yellow liquid.

4i: mp 138−140 ℃ (from EtOAc); TLC: Rf 0.16 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.15−8.13 (m, 2H), 7.70−7.67 (m, 2H), 7.64−7.61 (m, 2H), 7.55−7.50 (m, 3H), 2.44 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 169.5, 161.8, 132.4, 132.1, 131.2, 129.2, 127.8, 127.2, 126.6, 118.5, 112.3, 98.5, 94.4, 83.4, 10.8; LRMS (ESI) m/z (rel int): 285 ([M + H]+, 92), 244 ([M – CH2CN]+, 32); HRMS (ESI) m/z calcd for C19H12N2O 284.0950, found 284.0950.

Synthesis of 3-Methyl-4-(4-nitrophenylethynyl)-5-phenylisoxazole (4j). 1-Ethynyl-4-nitrobenzene (3j) was used. The suspension was stirred for 19 h at rt. Column chromatography (40:1 hexane/EtOAc) afforded 4j (44 mg, 36%) as a white solid, 5 (10 mg, 16%) as a yellow solid, 6 (16.7 mg, 24%) as a yellow liquid.

4j: mp 144−146 ℃ (from EtOAc); TLC: Rf 0.15 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.27 (dt, 2H, J = 6.8, 2.0 Hz), 8.16−8.13 (m, 2H), 7.69 (dt, 2H, J = 6.8, 2.0 Hz), 7.54−7.52 (m, 3H), 2.46 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 169.7, 161.8, 147.5, 132.4, 131.3, 129.7, 129.2, 127.2, 126.6, 124.0, 98.4, 94.2, 84.3, 10.8; LRMS (ESI) m/z (rel int): 305 ([M + H]+, 77), 264 ([M – CH2CN]+, 100); HRMS (ESI) m/z calcd for C18H12F2O3 304.0848, found 304.0848

Synthesis of 4-(3,4-Dichlorophenylethynyl)-3-methyl-5-phenylisoxazole (4k). 1,2-dichloro-4-ethynylbenzene (3k) was used. The suspension was stirred for 17 h at 60 ℃. Column chromatography (40:1 hexane/EtOAc) afforded 4k (45.0 mg, 34%) as a yellow solid, 5 (9.7 mg, 15%) as a yellow solid, 6 (34.5 mg, 48%) as a yellow liquid, and 7k (38.0 mg, 14%) as a yellow solid.

4k: mp 135−137 ℃ (from EtOAc); TLC: Rf 0.30 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.14−8.12 (m, 2H), 7.63 (d, 1H, J = 2.0 Hz), 7.54?7.49 (m, 3H), 7.47 (d, 1H, J = 8.4 Hz), 7.37 (dd, 1H, J = 8.4, 2.0 Hz), 2.43 (s, 3H); 13C-NMR (100 MHz, CDCl3) δ 168.9, 161.7, 133.3, 133.0, 132.8, 131.6, 130.9, 130.6, 129.0, 127.1, 126.3, 122.6, 98.5, 93.6, 80.8, 10.6; LRMS (ESI) m/z (rel int): 328 ([M + H]+, 100); HRMS (ESI) m/z calcd for C18H11Cl2NO 327.0218, found 327.0218.

7k: mp 194−196 ℃ (from EtOAc); TLC: Rf 0.50 (14:1 hexane/EtOAc); IR (KBr film) 3405, 2306, 1655, 1261, 1026, 999, 827, 766, 669 cm−1; 1H-NMR (400 MHz, CDCl3) δ 7.61 (d, 2H, J = 2.0 Hz), 7.43 (d, 2H, J = 8.4 Hz), 7.35 (dd, 2H, J = 8.4, 2.0 Hz); 13C-NMR (500 MHz, CD3OD) δ 131.2, 130.9, 128.7, 95.0, 68.0, 66.1, 65.8, 53.6.

Synthesis of 3-Methyl-4-(4-methylphenylethynyl)-5-phenylisoxazole (4l). 1-Ethynyl-4-methylbenzene (3l) and K3PO4 were used. The suspension was stirred for 19 h at rt. Column chromatography (40:1 hexane/EtOAc) afforded of 4l (26.8 mg, 25%) as a yellow solid, 5 (32.2 mg, 51%) as a yellow solid, 6 (10.0 mg, 14%) as a yellow liquid, and 7l (36.7 mg, 20%) as a yellow solid.

4l: mp 73−75 ℃ (from EtOAc); TLC: Rf 0.20 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 8.19−8.17 (m, 2H), 7.52−7.47 (m, 3H), 7.45 (d, 2H, J = 8.0 Hz), 7.20 (d, 2H, J = 8.0 Hz), 2.43 (s, 3H), 2.40 (s, 3H); 13C-NMR(100 MHz, CDCl3) δ 168.2, 162.0, 139.2, 131.6, 130.7, 129.5, 129.0, 127.6, 126.4, 119.9, 99.5, 96.5, 78.2, 21.8, 10.8; LRMS (ESI) m/z (rel int): 274 ([M + H]+, 89), 233 ([M – CH2CN]+, 100); HRMS (ESI) m/z calcd for C19H15NO 273.1154, found 273.1158.

7l: mp 140−142 ℃ (from EtOAc) (Lit.,27 137−138 ℃); TLC: Rf 0.45 (14:1 hexane/EtOAc); 1H-NMR (400 MHz, CDCl3) δ 7.43−7.41 (m, 4H), 7.15−7.13 (m, 4H), 2.37 (s, 6H); 13C-NMR (100 MHz, CDCl3) δ 139.7, 132.6, 129.4, 119.0, 81.8, 73.7, 21.8; LRMS (ESI) m/z (rel int): 231 ([M + H]+, 100); HRMS (ESI) m/z calcd for C18H14 230.1096, found 230.1096.

 

Results and Discussion

Recently, α,β-acetylenic oximes of hydroxylamine have been cyclized to the corresponding isoxazoles in the presence of gold catalyst.18 Thus, we initially thought that the α,β-acetylenic oximes of hydroxylamine could serve as substrates for a one-pot cyclization–oxidative cross-coupling reaction. However, when 4-phenylbut-3-yn-2-one oxime (1b) was used as a substrate, it provided mainly isoxazole 5 from a cyclization–protodeauration and a trace of oxidative cross-coupling product 4a (Scheme 2). (Z)-4-Phenylbut-3-yn-2-one oxime was equilibrated with (E) at room temperature and cyclized into the corresponding isoxazole even without gold catalyst. This uncatalyzed cyclization was rapid and was a potential problem for one-pot cyclization– oxidative cross-coupling reaction because it could not generate gold-complexed intermediate 2. Therefore, we considered O-methyloximes as substrates, which were not cyclized spontaneously at room temperature and could be readily synthesized from commercially available ynones and methoxyamine hydrochloride in the presence of sodium acetate at room temperature using ethanol as the solvent. A mixture of E-Z isomers sometimes resulted and the desired Z isomer could be separated by column chromatography.

Scheme 2.Gold-catalyzed cyclization–oxidative alkynylation and cyclization-fluorination of oxime 1b.

With the desired (Z)-4-phenylbut-3-yn-2-one O-methyloxime (1a) in hand, we began to investigate various gold catalysts and combinations with silver salts: (PPh3)AuCl, (PCy3)AuCl, (PhO)3PAuCl, AuCl, AuCl3, (PPh3)AuNTf2, (IPr)AuCl (IPr = N,N'-bis(2,6-diisopropylphenyl)imidazol-2-ylidene), along with AgOTf, AgNTf2, or AgSbF6 (Table 1). To our delight, most Au catalysts produced cyclization–oxidative cross-coupling product 4a except (IPr)AuCl (Table 1, entry 18). Under the standard reaction conditions, (PPh3)AuCl afforded 4a in 29% yield as well as cyclized isoxazole 5, fluorinated isoxazole 6, and alkyne dimer 7 (Table 1, entry 1). Notably, no homocoupling product of 1a was observed. The addition of AgOTf or AgSbF6 made the reaction slower at room temperature and increased the formation of cyclized isoxazole 5 (entries 2 and 3). The gold(I) complex (PEt3)AuCl afforded 4a in 35% yield with increased formation of 5 (entry 6). Among the gold(I) complexes 8–11 bearing bulky biphenyl-based phosphine ligands, which have been shown to be excellent ligands for Pd-catalyzed reactions,19 only 10 yielded comparable results and produced 4a and 5 in 30% and 40% yields, respectively (entry 12). After extensive optimization studies, (PPh3)-AuNTf2 (10 mol %) was identified as the best catalyst for one-pot cyclization–oxidative cross-coupling reaction (entry 19). Under this condition, cross-coupling product 4a and fluorinated product 6 were isolated in 50% and 18% yield, respectively. Furthermore, when 2 equivalents of alkyne 3a were used, a better result was obtained than when 1.5 or 3 equivalents of alkyne 3a were used (entry 19 vs. entries 20 and 21).

Table 1.Optimization of gold-catalyzed cyclization−oxidative alkynylation and cyclization−fluorination of (Z)-2-alkynone O-methyloxime 1aa

The role of H2O in gold-catalyzed reactions is still unclear and under investigation. Recently, it has been reported that the addition of H2O enhances the solubility of the reagents such as base and Selectfluor, which are poorly soluble in CH3CN.20 Indeed, when the reaction was carried out in the presence of 4 Å molecular sieves without the addition of H2O, 4a was obtained in decreased yield (35%, Scheme 3). Using MeOH instead of H2O did not improve the yield of 4a (33%, Scheme 3).

Scheme 3.Gold-catalyzed cyclization–oxidative alkynylation and cyclization–fluorination in the absence of H2O.

The base effect has also been extensively studied as shown in Table 2. Increasing the amount of potassium phosphate tribasic did not improve the yield of 4a (3 equiv., entry 2). Interestingly, the reactions of carbonate bases provided comparable results, except that of Cs2CO3, which was sluggish even at 60 ℃ and afforded 5 as a sole product in 28% yield (entries 4–9). Meanwhile, usage of amine bases such as Et3N and 2,6-lutidine increased the formation of 5 instead of 4a (entries 12 and 13).

Table 2.aReaction conditions: 1a (400 μmol), 3a (800 μmol), Au-catalyst (40.0 μmol), Selectfluor (1.00 mmol), Base (800 μmol), H2O (4.00 mmol), CH3CN (4 mL). bIsolated yields. cIsolated yields based on 3a. d3 equivalents of base were used. eThe reaction was carried out at 60 ℃ and 72% of starting 1a was recovered.

Selectfluor has been known as the best external oxidant for gold-catalyzed cross-coupling reactions.21 When the reaction was carried out in the presence of 1.5 equivalents of Selectfluor, 4a was isolated in 15% yield and 61% of starting (Z)-2-alkynone O-methyloxime 1a was recovered after 2 days (Table 3, entry 1). The best yield was observed with 2.5 equivalents of Selectfluor (Table 1, entry 19). Using more than 2.5 equivalents of Selectfluor decreased the yield of 4a from 50% to 48% (Table 3, entry 3). Meanwhile, when the amount of Selectfluor was increased from 1.5 to 3.0 equivalents, the formation of fluorinated isoxazole 6 increased from 3% to 20%. As expected, other oxidants such as PhI(OAc)2, oxone, and t-BuOOH did not generate fluorinated isoxazole 6 (entries 4–7), neither did N-fluorobenzene-sulfonimide (NFSI), even carrying fluorine source (entries 8 and 9). The reactions of these oxidants were very sluggish in the formation of 4a even at 80 ℃. Interestingly, oxone afforded cyclized isoxazole 5 in 71% yield as a major product at 60 ℃ (entry 6).

Table 3.aReaction conditions: 1a (400 μmol), 3a (800 μmol), (PPh3)AuNTf2 (40.0 μmol), oxidant, K3PO4 (800 μmol), H2O (4.00 mmol), CH3CN (4 mL). bIsolated yields. cIsolated yields based on 3a. d% of recovered 1a. e80 ℃. f60 ℃

Current cyclization–oxidative alkynylation and cyclization– fluorination reactions showed profound concentration effect and solvent effect on the yield of the desired product. Upon increasing the reaction concentration from 0.1 to 0.5 M (Table 1, entry 19 vs. Table 4, entry 1), the yield of 4a decreased from 50% to 8%, and 30% of starting (Z)-2-alkynone O-methyloxime 1a was recovered. In contrast, upon lowering the reaction concentration from 0.1 to 0.02 M (Table 1, entry 19 vs. Table 4, entry 2), 4a was isolated in 40% yield after the extended reaction time (93 h). Alternative solvent such as DMF was less effective even at 80 ℃, leading to the formation of cyclized isoxazole 5 in 43% as a major product and 4a in 16% (entry 3). In less polar solvent 1,4-dioxane, the yield was far below the predictions: only 2% yield of 4a was obtained at 60 ℃ (entry 4). In particular, no conversion was observed in the relatively non-polar solvents, DCE and toluene (entries 5 and 6).

Table 4.aReaction conditions: 1a (400 μmol), 3a (800 μmol), (PPh3)AuNTf2 (40.0 μmol), Selectfluor (1.00 mmol), K3PO4 (800 μmol), H2O (4.00 mmol), Solvent. bIsolated yields. cIsolated yields based on 3a. d% of recovered 1a. e80 aReaction conditions: 1a (400 μmol), 3a (800 μmol), (PPh3)AuNTf2 (40.0 μmol), Selectfluor (1.00 mmol), K3PO4 (800 μmol), H2O (4.00 mmol), Solvent. bIsolated yields. cIsolated yields based on 3a. d% of recovered 1a. e80°. f60 ℃. f60 ℃

Under the standard reaction conditions, the reaction scope with respect to various alkyne coupling partners was probed as shown in Table 5. The reaction proceeded smoothly with alkynes bearing electron-withdrawing groups on the phenyl ring (entries 1–10). The reaction with alkyne 3b, which has Cl on the phenyl ring, led to 4b in 47% yield with the formation of 5, 6 and 7 at ambient temperature (entry 1). On the other hand, the reactions with alkynes bearing electrondonating groups on the phenyl ring were sluggish and did not provide the coupling products (entries 12 and 13), except for the reaction of alkyne 3l (entry 11). Instead, the reactions with these electron-rich alkynes afforded cycloisomerized isoxazole 5 as a major product. Current observations are consistent with those of the Sonogashira cross-coupling process: The reaction proceeds more rapidly with more electron-poor alkynes.22 K3PO4 is supposed to abstract the acetylenic proton of the terminal alkyne, thus forming a gold-acetylide species (vide infra).23 It should be pointed out that the employed bases are usually not basic enough to deprotonate the alkyne in order to generate the anionic nucleophile leading to the gold-acetylide species. Therefore, electron-withdrawing substituents on the phenyl ring that make the alkyne proton more acidic would lead to easier abstraction of the acetylenic proton.

Table 5.aReaction conditions: 1a (400 μmol), 3b–n (800 μmol), (PPh3)AuNTf2 (40.0 μmol), Selectfluor (1.00 mmol), K3PO4 (800 μmol), H2O (4.00 mmol), CH3CN (4 mL). bIsolated yields. cIsolated yields based on 3b–n. d60 ℃. e45% of staring 1a was recovered. f80% of staring 1a was recovered.

To discriminate the cyclization–oxidative alkynylation mechanism from the alkynylation through direct C-H activation of in situ formed isoxazole 5, the isoxazole 5 was prepared independently and treated with (PPh3)AuNTf2 (10 mol %) in the presence of Selectfluor, K3PO4, and H2O (Scheme 4). This experiment led to recovery of the starting material 5 and formation of alkyne dimer 7 with no detectable trace of the alkynylation product 4a and fluorinated isoxazole 6, which suggests that the alkynylation mechanism does not operate through direct C-H activation of isoxazole 5. Therefore, the product 4a would be formed through oxidative coupling before the protodeauration of the gold(I) complex.

Scheme 4.Gold-catalyzed alkynylation of isoxazole 5.

On the basis of previous literature and our observation, two plausible mechanistic pathways can be suggested for the gold-catalyzed cyclization–oxidative alkynylation of (Z)-2-alkynone O-methyloxime (Scheme 5). Initial coordination of the gold(I) catalyst A to the terminal alkyne 3a would form a π-coordinated alkyne adduct B, which transforms into the σ-acetylide gold(I) complex C in the presence of base. Then, oxidation by Selectfluor would afford the cationic gold(III) intermediate D, which reacts with the second molecule of C and gives the homo-coupled product 7. This mechanism has been well documented recently.24 On the other hand, the coordination of gold(III) intermediate D to the alkyne of O-methyloxime 1a and subsequent cyclization would lead to the gold(III) complex E bearing both isoxazole and alkyne substituents; E readily decomposes to yield cross-coupled product 4a and the cationic gold(I) catalyst A after reductive elimination. In the other possible mechanistic pathway, the gold(I)-catalyzed cyclization of O-methyloxime 1a may occur before the formation of σ-acetylide gold(I) complex C. The gold(I) complex G, which is formed after the demethylation of the cyclized gold(I) complex F, may give the proto-deauration product 5. Otherwise, Selectfluor would oxidize the gold(I) complex G to the cationic gold(III) intermediate H, the reductive elimination of which would afford the fluorinated isoxazole 6, or the fluoride-acetylide ligand exchange of which would give the gold(III) complex E. As mentioned previously, E undergoes reductive elimination to provide cross-coupled product 4a and to regenerate the cationic gold(I) catalyst A. This suggested mechanism involves a redox Au(I)/Au(III) catalytic cycle mediated by the external oxidant Selectfluor.

Scheme 5.Plausible mechanism for gold-catalyzed cyclization–oxidative alkynylation of (Z)-2-alkynone O-methyloxime.

 

Conclusion

In conclusion, we have developed a novel one-pot method to synthesize trisubstituted isoxazoles from (Z)-2-alkynone O-methyloxime. This gold-catalyzed cascade reaction proceeds under mild conditions (room temperature) via the intermolecular addition of the oxime oxygen atom to the tethered triple bond and subsequent alkynylation or fluorination. Mechanistically, gold-catalyzed cycloisomerization has been combined with the gold-catalyzed oxidative cross-coupling or fluorination process. The current synthetic method is straightforward and represents a typical pot-economical synthesis in that no intermediate purification operations are needed. This is the first example to synthesize various alkynyl isoxazoles from structurally simple O-methyloxime in a tandem manner, and offers the efficient construction of functionalized isoxazoles.