Metal-free DBU promoted regioselective synthesis of isoxazoles and isoxazolines

Abstract

A new simple and efficient metal-free 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) promoted regioselective synthesis of 3,5-disubstituted isoxazoles and isoxazolines from aldoximes has been described. This method allows the reaction to proceed efficiently on aldoximes containing unprotected phenolic hydroxyl groups. Furthermore, with the use of higher equivalents of N-chlorosuccinimide, chloro-substituted isoxazoles and isoxazolines were obtained as the only products via tandem one-pot 1,3-dipolar cycloaddition followed by regioselective chlorination.

---

Isoxazoles and isoxazolines are five-membered nitrogen containing heterocycles commonly found in a variety of natural products and drugs and are known to exhibit a wide range of pharmacological activities. The structures of biologically active isoxazoles/isoxazolines A–F are shown in Fig. 1.¹

Fig. 1 Examples of bioactive isoxazoles and isoxazolines.

Although numerous methods have been reported for the synthesis of isoxazoles,² the cycloaddition of alkynes with nitrile oxides is probably the most direct route to access these heterocycles. Uncatalyzed thermal cycloaddition reaction of nitrile oxide with alkyne is neither chemo- nor regioselective and, as a consequence, leads to formation of multiple products. Mostly the regioselectivity in this reaction has been achieved using Cu catalysts.³ There also exists few metal-free protocols for synthesis of isoxazoles; however most of these reports involve use of hypervalent iodine.^2g,4 Many hypervalent iodine reagents are explosive and have chemical reactivity issues.⁵ Apart from the use of hypervalent iodine, only two other metal-free conditions (Et₃N,⁶ NaOCl/Et₃N ⁷) have been reported. However, the former reaction was not suitable for phenolic hydroxyl containing aldoximes; and later condition has been reported for only one example. The detailed substrate scope has never been established. In this context, we thought of exploiting this reaction to establish an efficient and elegant metal-free protocol for synthesis of isoxaxoles and isoxazolines with broad substrate scope. Our efforts identified 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as an excellent reagent which promotes 1,3-dipolar cycloaddition of nitrile oxides with alkynes/alkenes without need of any metal catalyst. DBU has been reported to promote synthesis of variety of heterocyclic scaffolds such as isoquinolinones⁸ and benzothiazoles.⁹ Herein, we report DBU-promoted synthesis of 3,5-disubstituted isoxazoles. Furthermore, we also report an interesting DBU-promoted tandem one-pot 1,3-dipolar cycloaddition reaction followed by regioselective chlorination (Fig. 2).

Fig. 2 DBU promoted synthesis of isoxazoles and isoxazolines.

Our investigations were started with the reaction of chloro-aldoxime 2a with phenylacetylene 3a in presence of 1 equiv. DBU in chloroform, which produced desired 3,5-disubstituted isoxazole 4a in 10 min in 95% yield (Fig. 3). Next, we elaborated this method for direct synthesis of isoxazole 4a from aldoxime 1a. Under similar reaction conditions, when the reaction of 4-hydroxyphenyl aldoxime 2a with phenylacetylene 3a in presence of 1.2 equiv. NCS was performed, multiple products were formed. Further, optimization studies indicated that DMF¹⁰ is the best solvent for this reaction and sequential addition is required to get desired product in good yield (Fig. 3). The sequential addition (1a + NCS + DMF, rt, 1 h followed by addition of DBU and 3a, rt, 10 min) was followed, to get product 4a in good yield, without formation of side-products. However, when we performed this reaction (from 1a) in presence of Et₃N (in place of DBU) under similar reaction conditions, multiple products were formed; which clearly indicated the superiority of our protocol over earlier Et₃N (ref. 6) method. The addition of DBU is slightly exothermic and quickly converts aldoximes to nitrile oxides which subsequently reacts with respective alkynes/alkenes to produce desired heterocycles. It is noteworthy to mention that the direct synthesis of hydroxy-substituted phenyl isoxazole 4a from corresponding nitrile oxide has never been reported. The limitation of earlier methods for hydroxy-substituted phenyl nitrile oxides has also been mentioned.^3a However, the present DBU-promoted synthesis provided this product in excellent yield.

Fig. 3 DBU-promoted synthesis of 4-hydroxy substituted isozaxole 4a. Reagents and conditions: (a) 1 equiv. DBU, CHCl₃, rt, 10 min, 95%; (b) 1 equiv. DBU, 1.2 equiv. NCS, DMF, 1 h, 80%.

Next, the substrate scope of the reaction was investigated for variety of aldoximes and phenylacetylenes (Fig. 4). The optimized reaction condition involving the use of 1 equiv. of DBU and 1.2 equiv. of NCS was employed. Phenyl aldoxime along with those substituted with alkyl electron-donating groups produced isoxazoles in excellent yields (examples 4b, 4d, 4e, and 4i). In case of aldoximes substituted with other electron-donating groups such as OH, OMe (examples 4a, 4c, 4f, 4g and 4j), desired products were formed in 65–80% yield. Interestingly, in latter reactions, additionally a ring-chlorinated isoxazole products were obtained as side products in 10–20% yield. Heterocyclic aldoximes also participated well in this reaction, producing desired products 4k and 4l in good yields. It is noteworthy to mention that the pyridine linked isoxazole 4l prepared herein, has a close structural similarity to that of nicotine receptor modulator (structure D, Fig. 1). Aliphatic oximes (example 4m) and phenyl aldoxime with electron-withdrawing group (example 4n) does not participated in this reaction. Among alkynes, all substituted phenylacetylenes participated well in this reaction, whereas aliphatic alkynes such as propargyl alcohol (example 4j) and N-propargylated isatin (example 4h) produced lower yields.

Fig. 4 Scope of the reaction for synthesis of isoxazoles. Reagents and conditions: aldoxime (1 equiv.), alkyne (1.2 equiv.), DBU (1 equiv.), NCS (1.2 equiv.), DMF, 1–8 h, 50–88%. Reaction time: 1 h for 4a, 4f, 4g and 4j; and 8 h for other examples.

Further, we examined the scope of this reaction for synthesis of various substituted isoxazolines using similar reaction conditions (Fig. 5). Phenyl, p-tolyl and m-tolyl aldoximes gave good yields, irrespective of the location of alkyl substituent (examples 6f, 6g, 6i, 6k, 6l, 6n and 6p). Similarly, the electron-rich styrenes (examples 6h, 6i and 6p) and halo-substituted styrenes gave good yields. Among heteroaryl aldoximes, the pyridine aldoxime produced corresponding isoxazoline 6m in good yield, whereas indole aldoxime produced corresponding isoxazoline 6q, comparatively in a lesser yield.

Fig. 5 Scope of the reaction for synthesis of isoxazolines. Reagents and conditions: aldoxime (1 equiv.), styrene (1.2 equiv.), DBU (1 equiv.), NCS (1.2 equiv.), DMF, 1–8 h, 63–85%. Reaction time: 1 h for 6b and 6d; and 8 h for other examples.

Next, we sought to explore the observed finding of one-pot 1,3-dipolar cycloaddition reaction followed by ring chlorination to produce chlorinated isoxazoles and isozaxolines (Fig. 6). For this, we thought to use higher equivalents of NCS than the required amount, to get increased yields of chlorinated products. To our surmise, when aldoxime 1a was treated with phenylacetylene 3a in presence of 2.2 equivalents of NCS and 1 equiv. of DBU, exclusively the chloro-substituted product 7a was obtained (88% yield). Similarly, we prepared other chlorinated isoxazoles 7b, 7c, 7f and 7g using higher equivalents of NCS. The tandem one-pot 1,3-dipolar cycloaddition followed by regioselective chlorination also worked well for synthesis of chlorinated isoxazolines (Fig. 6; examples 7d, 7e and 7h). In case of furan-2-aldoxime, the isoxazole product 7g was formed in good yield (90%), however lower yield (60%) was obtained for isoxazoline product 7h. In case of indole-3-aldoxime and pyridine-2-aldoxime, ring chlorination product was not formed even after using higher equivalents of NCS. It is noteworthy to mention that the furyl compound 7g has a structural similarity with reported COX-1 inhibitor (structure B, Fig. 1).

Fig. 6 One-pot synthesis of chloro-substituted isoxazoles and isoxazolines. Reagents and conditions: aldoxime (1 equiv.), styrene or alkyne (1.2 equiv.), DBU (1 equiv.), NCS (2.2 equiv.), DMF, 8 h, 60–90%.

Next, we sought to investigate the mechanistic sequence of 1,3-dipolar cycloaddition and ring chlorination (Fig. 7) reactions using LCMS analysis. For this, we first performed reaction of aldoxime 1a with phenylacetylene 3a under optimized reaction conditions (1.2 equiv. NCS). The LCMS chromatogram after 30 min reaction time (Fig. 7a) indicated formation of desired product 4a (t_R = 9.4 min). However, with the use of 2.2 equiv. NCS, the LCMS chromatogram after 40 min indicated formation of chlorinated isoxazole 7a (t_R = 10.5 min) along with product 4a (t_R = 9.4 min). Further, when this reaction was monitored after 8 h, the LCMS chromatogram showed solely the formation of chlorinated product 7a. This study indicated that the chlorination reaction must be occurring after formation of isoxazole skeleton. In order to further understand the other possibility of first chlorination and then 1,3-dipolar cycloaddition, we treated aldoxime 1a with 2.2 equiv. of NCS and stirred for 8 h, which indicated formation of ring-chlorinated aldoxime only in 15%, which further indicates that in one-pot tandem protocol, chlorination occurs after 1,3-dipolar cycloaddition reaction.

Fig. 7 LC chromatograms of reaction of 1a with 3a with the change in NCS equivalents. MS spectras are provided in ESI.‡ The products 4a and 7a appears at t_R 9.4 and 10.5 min, respectively.

In conclusion, we have developed a new metal-free DBU-promoted synthesis of isoxazole and isoxazolines from aldoximes. Additionally, an interesting one-pot protocol for tandem 1,3-dipolar cycloaddition followed by regioselective ring chlorination has been reported. This methodology was applied to one-pot synthesis of 4-chloro-furyl-isoxazole and pyridine-isoxazole scaffold which are reported to possess potent biological activities.

Experimental section

General procedure for preparation of isoxazoles 4a–4l, isoxazolines 6a–6q and chlorinated isoxazole/isoxazoline products 7a–7h

To the stirred solution of aldoximes (100 mg, 1 equiv.) in DMF (3 ml) was added N-chlorosuccinimide (1.2 or 2.2 equiv.) at room temperature and reaction was stirred for 0.5–1 h. Then, DBU (1 equiv.) and styrenes or alkynes (1.2 equiv.) were added and reaction was further stirred for 1–8 h. After completion of the reaction (confirmed by TLC), chilled water (20 ml) was added and product was extracted with EtOAc (3 × 10 ml). The organic layer was collected, dried on anhydrous sodium sulphate and solvent was evaporated on rotary evaporator to get the crude product. The crude product was purified by silica gel (#100–200) column chromatography using 2–20% EtOAc: hexane to get pure isoxazoles 4a–4l, isoxazolines 6a–6q or chlorinated isoxazole/isoxazoline products 7a–7h in 50–90% yield. The spectral data of representative compounds 4a, 6a and 7a is shown here. Spectral data of other compounds 4b–4l, 6b–6q and 7b–7h is provided in ESI.‡

4-(5-Phenylisoxazol-3-yl) phenol⁶ (4a). White solid; yield 80%; m. p. 187–189 °C; ¹H NMR (400 MHz, CDCl₃, ppm): δ 7.83 (dd, J = 4, 8 Hz, 2H), 7.76 (d, J = 8 Hz, 2H), 7.47 (m, 3H), 6.95 (d, J = 8 Hz, 2H), 6.78 (s, 1H); ¹³C NMR (CDCl₃ + CD₃OD, 125 MHz): δ 170.0, 162.9, 158.8, 130.2, 128.9, 128.2, 127.2, 125.7, 120.0, 115.7, 97.3; IR (CHCl₃): ν_max 3433, 2925, 1631, 1450, 1353, 1095, 1017 cm⁻¹; ESI-MS: m/z 238.08 [M + H]⁺; HRMS: m/z 238.0864 calcd for C₁₅H₁₁NO₂ + H⁺ (238.0864).

3,5-Diphenyl-4,5-dihydroisoxazole^4a (6a). White solid; yield 85%; m. p. 72–74 °C; ¹H NMR (400 MHz, CDCl₃, ppm): δ 7.70 (dd, J = 4.0, 8.0 Hz, 2H), 7.30–7.42 (m, 8H), 5.75 (dd, J = 11.0, 8.3 Hz, 1H), 3.79 (dd, J = 16.6, 11.0 Hz, 1H), 3.35 (dd, J = 16.6, 8.3 Hz, 1H); ¹³C NMR (CDCl₃, 100 MHz): δ 156.1, 140.9, 130.2, 129.4, 128.8, 128.2, 126.7, 125.9, 82.6, 43.2; IR (CHCl₃): ν_max 3433, 2920, 1730, 1446, 1352, 1120, 893, 751, 686 cm⁻¹; ESI-MS: m/z 224.10 [M + H]⁺; HRMS: m/z 224.1084 calcd for C₁₅H₁₃NO + H⁺ (224.1070).

2-Chloro-4-(5-phenylisoxazole-3-yl)phenol (7a). White solid; yield 88%; m. p. 193–196 °C; ¹H NMR (400 MHz, CDCl₃, ppm): δ 7.89 (d, J = 4.1, Hz, 1H), 7.82–7.84 (m, 2H), 7.68–7.71 (m, 1H), 7.47–7.50 (m, 3H), 7.13 (d, J = 8 Hz, 1H), 6.77 (s, 1H), 5.78 (s, 1H); ¹³C NMR (CDCl₃ + CD₃OD, 125 MHz): δ 170.3, 161.9, 154.4, 130.3, 128.9, 128.3, 127.1, 126.3, 125.7, 121.0, 116.7, 97.2; IR (CHCl₃): ν_max 3346, 2922, 2851, 1594, 1422, 1019 cm⁻¹; ESI-MS: m/z 272.04 [M + H]⁺; HRMS: m/z 272.0481 calcd for C₁₅H₁₀ClNO₂ + H⁺ (272.0473).

Acknowledgements

Authors thank analytical department, IIIM for NMR and MS analysis of our compounds. This work was supported by CSIR 12th FYP grant # BSC-0205. SM is thankful to CSIR for the award of research fellowship. Authors are grateful to Dr Ajai P. Gupta and Manoj Kushwaha for LCMS analysis.

References and notes

1. For biological activities of isoxazoles and isoxazolines, see: (a) Y. Al-Abed, D. Dabideen, B. Aljabari, A. Valster, D. Messmer, M. Ochani, M. Tanovic, K. Ochani, M. Bacher, F. Nicoletti, C. Metz, V. A. Pavlov, E. J. Miller and K. J. Tracey, J. Biol. Chem., 2005, 280, 36541–36544; (b) R. Kakarla, J. Liu, D. Naduthambi, W. Chang, R. T. Mosley, D. Bao, H. M. Micolochick Steuer, M. Keilman, S. Bansal, A. M. Lam, W. Seibel, S. Neilson, P. A. Furman and M. J. Sofia, J. Med. Chem., 2014, 57, 2136–2160; (c) A. P. Kozikowski, Acc. Chem. Res., 1984, 17, 410–416; (d) V. Nair and T. D. Suja, Tetrahedron, 2007, 63, 12247–12275; (e) K. Takenaka, T. Nagano, S. Takizawa and H. Sasai, Tetrahedron: Asymmetry, 2010, 21, 379–381; (f) S. Tapadar, R. He, D. N. Luchini, D. D. Billadeau and A. P. Kozikowski, Bioorg. Med. Chem. Lett., 2009, 19, 3023–3026; (g) K. A. El Sayed, P. Bartyzel, X. Shen, T. L. Perry, J. K. Zjawiony and M. T. Hamann, Tetrahedron, 2000, 56, 949–953; (h) M. Gopalakrishnan, J. Ji, C.-H. Lee, T. Li and K. B. Sippy, WO2009/149135A1 (Abbott Laboratories), 2009; (i) P. Proksch, A. Putz, S. Ortlepp, J. Kjer and M. Bayer, Phytochem. Rev., 2010, 9, 475–489; (j) V. Jager and P. A. Colinas, in Synthetic applications of 1,3-dipolar cycloaddition chemistry toward heterocycles and natural products, ed. A. Padwa and W. H. Pearson, Wiley, Hoboken, NJ, 2002, pp. 361–472.
2. For synthetic reports on isoxazoles, see: (a) M. P. Bourbeau and J. T. Rider, Org. Lett., 2006, 8, 3679–3680; (b) L. Cecchi, F. De Sarlo and F. Machetti, Eur. J. Org. Chem., 2006, 2006, 4852–4860; (c) C.-y. Chen, T. Andreani and H. Li, Org. Lett., 2011, 13, 6300–6303; (d) J. A. Crossley and D. L. Browne, J. Org. Chem., 2010, 75, 5414–5416; (e) E. Gayon, O. Quinonero, S. Lemouzy, E. Vrancken and J.-M. Campagne, Org. Lett., 2011, 13, 6418–6421; (f) O. Jackowski, T. Lecourt and L. Micouin, Org. Lett., 2011, 13, 5664–5667; (g) S. Minakata, S. Okumura, T. Nagamachi and Y. Takeda, Org. Lett., 2011, 13, 2966–2969; (h) M. S. Mohamed Ahmed, K. Kobayashi and A. Mori, Org. Lett., 2005, 7, 4487–4489; (i) C. Praveen, A. Kalyanasundaram and P. T. Perumal, Synlett, 2010, 777–781; (j) S.-R. Sheng, X.-L. Liu, Q. Xu and C.-S. Song, Synthesis, 2003, 2763–2764; (k) B. J. Stokes, C. V. Vogel, L. K. Urnezis, M. Pan and T. G. Driver, Org. Lett., 2010, 12, 2884–2887; (l) S. Tang, J. He, Y. Sun, L. He and X. She, Org. Lett., 2009, 11, 3982–3985; (m) M. Ueda, A. Sato, Y. Ikeda, T. Miyoshi, T. Naito and O. Miyata, Org. Lett., 2010, 12, 2594–2597; (n) J. P. Waldo and R. C. Larock, J. Org. Chem., 2007, 72, 9643–9647; (o) B. Willy, F. Rominger and T. J. J. Müller, Synthesis, 2008, 293–303.
3. (a) S. B. Bharate, A. K. Padala, B. A. Dar, R. R. Yadav, B. Singh and R. A. Vishwakarma, Tetrahedron Lett., 2013, 54, 3558–3561; (b) T. V. Hansen, P. Wu and V. V. Fokin, J. Org. Chem., 2005, 70, 7761–7764; (c) F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtsev, L. Noodleman, K. B. Sharpless and V. V. Fokin, J. Am. Chem. Soc., 2004, 127, 210–216.
4. (a) A. Yoshimura, K. R. Middleton, A. D. Todora, B. J. Kastern, S. R. Koski, A. V. Maskaev and V. V. Zhdankin, Org. Lett., 2013, 15, 4010–4013; (b) A. Yoshimura, C. Zhu, K. R. Middleton, A. D. Todora, B. J. Kastern, A. V. Maskaev and V. V. Zhdankin, Chem. Commun., 2013, 49, 4800–4802; (c) B. A. Mendelsohn, S. Lee, S. Kim, F. Teyssier, V. S. Aulakh and M. A. Ciufolini, Org. Lett., 2009, 11, 1539–1542; (d) A. M. Jawalekar, E. Reubsaet, F. P. J. T. Rutjes and F. L. van Delft, Chem. Commun., 2011, 47, 3198–3200.
5. R. D. Richardson, J. M. Zayed, S. Altermann, D. Smith and T. Wirth, Angew. Chem., Int. Ed., 2007, 46, 6529–6532.
6. M. A. Weidner-Wells, T. C. Henninger, S. A. Fraga-Spano, C. M. Boggs, M. Matheis, D. M. Ritchie, D. C. Argentieri, M. P. Wachter and D. J. Hlasta, Bioorg. Med. Chem. Lett., 2004, 14, 4307–4311.
7. D. E. Kizer, R. B. Miller and M. J. Kurth, Tetrahedron Lett., 1999, 40, 3535–3538.
8. W. Chen, J. Cui, Y. Zhu, X. Hu and W. Mo, J. Org. Chem., 2012, 77, 1585–1591.
9. F. Wang, S. Cai, Z. Wang and C. Xi, Org. Lett., 2011, 13, 3202–3205.
10. K. E. Larsen and K. B. G. Torssell, Tetrahedron, 1984, 40, 2985–2988.

---

Footnotes

† IIIM Publication number IIIM/1723/2014.

‡ Electronic supplementary information (ESI) available: NMR spectra of all compounds. See DOI: 10.1039/c4ra14694h

---