Optically active N-alkyl aziridines via stereospecific reductive cyclization of α-mesylated acetamides

Abstract

An efficient method for the synthesis of optically active N-alkyl aziridines has been realized for the first time by stereospecific reductive cyclization of optically active α-mesylated acetamides. A series of optically active N-alkyl aziridines are prepared in moderate to good yields and excellent ees. The employment of (9-BBN)₂ as the reagent is crucial for the success of the transformation.

---

Optically active aziridine moieties exist in a number of biologically important alkaloids¹ and therapeutic agents such as mitomycin A,^2a,b madurastatin A1,^2e ficellomycin,^2f and azicemicin B^2c,d (Fig. 1). They also belong to key structural motifs in agrochemicals and dyes.² Structurally, the ring strain possessed by aziridine makes them highly reactive species. For example, they can undergo ring-opening reactions in the presence of nucleophiles,³ and they can also be employed for cycloaddition reactions.⁴ Optically active aziridines have become highly valuable building blocks for the synthesis of nitrogen-containing compounds.

Fig. 1 Therapeutic agents containing optically active aziridine moieties.

The synthesis of optically active aziridines has thus gained significant interest. Common methods for the stereoselective synthesis of aziridines include Wenker cyclization or Mitsunobu reaction of optically active amino alcohols,⁵ reaction of imines with carbene compounds (aza-Darzens reaction)⁶ or sulfur ylides (Corey–Chaykovsky aziridination),⁷ or reaction of alkenes with nitrenes in situ generated from various nitrene precursors in the presence of a catalyst.⁸ However, most of these methods applied well to the synthesis of aziridines with N-acyl or -sulphonyl functionalities in order to achieve a good yield. So far, only a few methods are available for the synthesis of optically active N-alkyl aziridines. Lindsley and coworkers⁹ described a three-step protocol for the synthesis of terminal N-alkyl aziridines (Fig. 2a). An enantioselective aziridination was reported by Buchwald and coworkers from allylic hydroxylamine ester through a CuH-catalyzed regioselective intramolecular hydroamination¹⁰ (Fig. 2b). Nevertheless, a facile and straightforward synthetic methods for optically active N-alkyl aziridines from readily accessible starting material remains highly desirable. We have recently reported a stereospecific nucleophilic substitution of optically active α-aryl-α-mesylated acetamides with arylboronic acids as nuclephiles.¹¹ The readily availability of optically active α-aryl-α-mesylated acetamides from mandelic acid derivatives or by asymmetric hydrogenation¹² makes them attractive feedstocks for preparing other valuable optically active intermediates. Herein we report a facile preparation of optically active N-alkyl aziridines by stereospecific reductive cyclization of α-mesylated acetamides using (9-BBN)₂ as a reducing reagent (Fig. 2c).

Fig. 2 The construction of aziridines.

Optically active (R)-2-(benzylamino)-2-oxo-1-phenylethyl methanesulfonate (1a) was chosen as the model substrate for study of reductive cyclization (Table 1). The reaction was initially carried out in toluene with (9-BBN)₂ as the reducing reagent and K₂CO₃ as the base. A decent yield (25%) of 3a with an excellent ee (97%) was achieved at 60 °C for 8 h (entry 1). However, no 3a was formed with other reducing reagents including BH₃, LiAlH₄, and DIBAL-H (entries 2–4). We thus employed (9-BBN)₂ as the reducing reagent for the rest of study. When K₃PO₄ was used as the base, the yield of 3a increased to 62% without loss of ee (97%, entry 5). Further screening of the base including inorganic and organic bases did not further improve the yield (entries 6–11). It is noteworthy that no product of 3a was achieved in the absence of base (entry 12). We further investigated the effect of the solvents. Polar solvents such as dioxane, THF, DMF or CH₃CN did not yield any product (entries 14–16 and 18). Nonpolar or less polar solvent such as DCE, cyclohexane, or mesitylene provided the desired product, albeit with lower yields (entries 13, 17 and 19). The absolute configuration of 3a was determined as S by comparing the optical rotation with reported data,¹³ indicating that 1a was cyclized reductively to form 3a through an S_N2-type process.

Table 1 Reductive cyclization of (R)-2-(benzylamino)-2-oxo-1-phenylethyl methanesulfonate (1a)

Entries^a	Base	Solvent	Reducing agent	Yield^b (%)	ee^c (%)
a Unless otherwise specified, all reactions were performed at 60 °C for 8 h in the selected solvent (5 mL) with 1a (0.5 mmol), 2a (1 mmol), base (2 equiv.). b Isolated yields. c Determined by chiral HPLC on a chiralcel column.
1	K2CO3	Toluene	(9-BBN)2	25	97
2	K2CO3	Toluene	BH3·THF	0	ND
3	K2CO3	Toluene	LiAlH4	0	ND
4	K2CO3	Toluene	DABAL-H	Trace	ND
5	K3PO4	Toluene	(9-BBN)2	62	97
6	TEA	Toluene	(9-BBN)2	14	97
7	KOtBu	Toluene	(9-BBN)2	0	ND
8	DABCO	Toluene	(9-BBN)2	0	ND
9	CsF	Toluene	(9-BBN)2	7	98
10	KOH	Toluene	(9-BBN)2	0	ND
11	Cs2CO3	Toluene	(9-BBN)2	40	93
12	None	Toluene	(9-BBN)2	0	ND
13	K3PO4	DCE	(9-BBN)2	27	91
14	K3PO4	Dioxane	(9-BBN)2	0	ND
15	K3PO4	THF	(9-BBN)2	0	ND
16	K3PO4	DMF	(9-BBN)2	0	ND
17	K3PO4	Cyclohexane	(9-BBN)2	22	94
18	K3PO4	CH3CN	(9-BBN)2	0	ND
19	K3PO4	Mesitylene	(9-BBN)2	20	97

---

We then looked into the substrate scope of this reaction. As depicted in Table 2, a series of optically active aziridines were formed in excellent enantioselectivities and moderate to good yields from optically active α-substituted-phenyl-α-mesylated acetamides with (9-BBN)₂ as the reducing reagent with a minimum loss of enantiomeric purity. Various optically active N-alkyl aziridines including N-cyclic alkyl (3c–e), N-noncyclic primary alkyl (3f–h) and N-noncyclic secondary aziridines (1i–g) were prepared in moderate to good yields and excellent ee's. Surprisingly, no desired aziridine product was formed when an N-aryl substrate was employed. Halogen-substituted benzylic mesylates (1k–o) at various positions were all applicable. However, mesylates with electron-donating substituents at aryl groups were generally too unstable to prepare. It should be noted that ortho-halo substituted benzylic mesylates such as 1k, 1o were also suitable for the transformation, providing the optically active aziridines 3k and 3o in excellent enantioselectivities and moderate yields.

Table 2 Formation of optically active aziridines 3a–o by reductive cyclization^a

a Unless otherwise specified, the reactions were carried in toluene under nitrogen at 60 °C for 8 h in the presence of (9-BBN)₂ (2 equiv.) and K₃PO₄ (2 equiv.), isolated yields, ee value of the substrate in parenthesis.

---

To shed light on the mechanism of this transformation, the following experiments were implemented (Scheme 1). Firstly, a deuterium-labeling experiment with 1c′ as the substrate proved no deprotonation occurring at the benzylic position (eqn (1)). Secondary, no deuterium was included in 3c with C₆D₆ as solvent, indicating that the hydrogen in the product is derived from (9-BBN)₂ rather than solvent (eqn (2)). Thirdly, only a trace amount of the target product 3c was isolated when the (9-BBN)₂ was reduced to 0.5 equiv., indicating that multiple equivalents of 9-BBN were needed for a complete transformation of 1c to the target product 3c (eqn (3)). Fourthly, no target product 3a was formed without base in the reaction, and only a large amount of 1a was recovered (eqn (4)), demonstrating that the base had played a crucial role in the reduction of amide. We believe that the addition of base not only enhanced the reducing ability of 9-BBN, but also promoted the formation of the aziridine ring. To understand why no aziridine was formed when DIBAL-H was used as the reducing reagent, we isolated a secondary amine as a major product from the reaction (eqn (5)), indicating that DIBAL-H was too active as the reducing agent for the transformation. When an N-phenyl substrate was employed, a demesylate product was isolated in 46% yield (eqn (6)), indicating a facile demesylation process with an N-aryl substrate. This result explained the limitation of the reductive cyclization in the preparation of N-aryl aziridines.

Scheme 1 Mechanistic investigation.

On the basis of the above observation, we proposed two reaction pathways for the reductive cyclization (Scheme 2). In pathway I, 9-BBN reacts with substrate 1a to generate imidate I, which further react with a 2^nd equiv. of 9-BBN to form II. In the presence of a base, intramolecular cyclization at a S_N2-type fashion occurs to form cyclic product III, which can break down to a cyclic iminium IV. Further reduction with a 3^rd equiv. of 9-BBN to form product 3a. Alternatively in pathway II, intermediate II can give rise to the formation of imine V under basic conditions, which can further reduced by 9-BBN to form VI. In the presence of a base, cyclization of VI through a S_N2-type process takes place to form product 3a.

Scheme 2 Proposed mechanism.

In summary, an efficient method for the synthesis of optically active N-alkyl aziridines has been realized for the first time by stereospecific reductive cyclization of optically active α-mesylated acetamides. With this method, a series of optically active N-alkyl aziridines have been prepared in moderate to good yields with excellent ees. Mechanistic investigation has indicated an intramolecular S_N2-type process under reductive conditions and the employment of (9-BBN)₂ has been proven to be crucial for the success of reaction. Because of the readily availability of optically active α-hydroxy acetamides, this method offers a facile and expedite preparation of optically active aziridines complementary to other existing methods. The methodology should find a number of applications in drug discovery and process chemistry.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We are grateful to the Strategic Priority Research Program of the Chinese Academy of Sciences XDB20000000, CAS (QYZDY-SSW-SLH029), NSFC (21725205, 21432007, 21572246), STCSM-18520712200, and K.C. Wong Education Foundation.

Notes and references

1. For selected reviews, see: (a) J. B. Sweeney, Chem. Soc. Rev., 2002, 31, 247; (b) I. D. G. Watson, L. Yu and A. K. Yudin, Acc. Chem. Res., 2006, 39, 194; (c) G. S. Singh, M. DQhooge and N. De Kimpe, Chem. Rev., 2007, 107, 2080; (d) S. Stanković, M. DQhooge, S. Catak, H. Eum, M. Waroquier, V. Van Speybroeck, N. De Kimpe and H.-J. Ha, Chem. Soc. Rev., 2012, 41, 643; (e) G. Callebaut, T. Meiresonne, N. De Kimpe and S. Mangelinckx, Chem. Rev., 2014, 114, 7954; (f) B. H. Rotstein, S. Zaretsky, V. Rai and A. K. Yudin, Chem. Rev., 2014, 114, 8323.
2. (a) T. Hata, T. Hoshi, K. Kanamori, A. Matsumae, Y. Sano, T. Shima and R. Sugawara, J. Antibiot., 1956, 9, 141; (b) T. Hata and R. Sugawara, J. Antibiot., 1956, 9, 147; (c) T. Tsuchida, H. Iinuma, N. Kinoshita, T. Ikeda, T. Sawa, M. Hamada and T. Takeuchi, J. Antibiot., 1995, 48, 217; (d) T. Tsuchida, R. Sawa, Y. Takahashi, H. Iinuma, T. Sawa, H. Naganawa and T. Takeuchi, J. Antibiot., 1995, 48, 1148; (e) K. I. Harada, K. Tomita, K. Fujii, K. Masuda, Y. Mikami, K. Yazawa and H. Komaki, J. Antibiot., 2004, 57, 125; (f) A. D. Argoudelis, F. Reusser, H. A. Whaley, L. Baczynskyj, S. A. Mizsak and R. L. Wnuk, J. Antibiot., 1976, 29, 1001.
3. (a) X. E. Hu, Nucleophilic ring opening of aziridines, Tetrahedron, 2004, 60, 2701–2743; (b) T. N. Wade, J. Org. Chem., 1980, 45, 5328–5333; (c) J. Wu, X.-L. Hou and L.-X. Dai, J. Chem. Soc., Perkin Trans. 1, 2001, 1314–1317; (d) L. Ma and J. Xu, Prog. Chem., 2004, 16(2), 220–235.
4. (a) M. R. Kuszpit, W. D. Wulff and J. J. Tepe, J. Org. Chem., 2011, 76, 2913–2919; (b) R. A. Craig II, N. R. O'Connor, A. F. G. Goldberg and B. M. Stoltz, Chem. – Eur. J., 2014, 20, 4806–4813; (c) M. Sengoden and T. Punniyamurthy, Angew. Chem., Int. Ed., 2013, 52, 572–575; (d) Y.-M. Shen, W.-L. Duan and M. Shi, Eur. J. Org. Chem., 2004, 3080–3089; (e) Y. Du, Y. Wu, A.-H. Liu and L.-N. He, J. Org. Chem., 2008, 73, 4709–4712.
5. (a) X. Y. Li, N. Chen and J. X. Xu, Synthesis, 2010, 3423–3428; (b) B. R. Buckley, A. P. Patel and K. G. U. Wijayantha, J. Org. Chem., 2013, 78, 1289–1292.
6. (a) S. E. Larson, G. L. Li, G. B. Rowland, D. Junge, R. C. Huang, H. L. Woodcock and J. C. Antilla, Org. Lett., 2011, 13, 2188–2191; (b) Q. Peng, D. Guo, J. Bie and J. Wang, Angew. Chem., Int. Ed., 2018, 57, 3767; (c) T. Akiyama, T. Suzuki and K. Mori, Org. Lett., 2009, 11, 2445; (d) C. De Fusco, T. Fuoco, G. Croce and A. Lattanzi, Org. Lett., 2012, 14, 4078; (e) B. M. Trost, T. Saget and C.-I. Hung, Angew. Chem., Int. Ed., 2017, 56, 2440.
7. (a) J. J. Li, Name Reactions in Heterocyclic Chemistry, Wiley, Hoboken, 2005, pp. 2–14; (b) A. Padwa and S. S. Murphree, ARKIVOC, 2006, 6–33; (c) Aziridines and Epoxides in Organic Synthesis, ed. A. Yudin, Wiley-VCH, Weinheim, 2006.
8. (a) M. M. Abu-Omar, Dalton Trans., 2011, 40, 3435–3444; (b) J. W. W. Chang, T. M. U. Ton and P. W. H. Chan, Chem. Rec., 2011, 11, 331–357; (c) D. Mansuy, J. P. Mahy, A. Dureault, G. Bedi and P. Battioni, J. Chem. Soc., Chem. Commun., 1984, 1161–1163; (d) D. A. Evans, M. M. Faul and M. T. Bilodeau, J. Org. Chem., 1991, 56, 6744–6746; (e) P. Müller, C. Baud and Y. Jacquier, Tetrahedron, 1996, 52, 1543–1548; (f) N. Jung and S. Brase, Angew. Chem., Int. Ed., 2012, 51, 5538; (g) D. A. Evans, M. M. Faul, M. T. Bilodeau, B. A. Anderson and D. M. Barnes, J. Am. Chem. Soc., 1993, 115, 5328; (h) Z. Li, R. W. Quan and E. N. Jacobsen, J. Am. Chem. Soc., 1995, 117, 5889; (i) X. Wang and K. Ding, Chem. – Eur. J., 2006, 12, 4568; (j) J. Tao, L.-M. Jin and X. P. Zhang, Beilstein J. Org. Chem., 2014, 10, 1282.
9. (a) O. O. Fadeyi, M. L. Schulte and C. W. Lindsley, Org. Lett., 2010, 12, 3276; (b) T. J. Senter, M. C. O'Reilly, K. M. Chong, G. A. Sulikowski and C. W. Lindsley, Tetrahedron Lett., 2015, 56, 1276.
10. H. Wang, J. C. Yang and S. L. Buchwald, J. Am. Chem. Soc., 2017, 139, 8428–8431.
11. D. Tian, C. Li, G. Gu, H. Peng, X. Zhang and W. Tang, Angew. Chem., Int. Ed., 2018, 57, 7176.
12. G. Gu, T. Yang, O. Yu, H. Qian, J. Wang, J. Wen, L. Dang and X. Zhang, Org. Lett., 2017, 19, 5920.
13. A. Kawamoto and M. Wills, Tetrahedron: Asymmetry, 2000, 11, 3257–3261.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8qo00774h

---