Abstract
The first enantioselective aminolysis of aromatic trans-2,3-epoxy sulfonamides has been accomplished, which was efficiently catalyzed by a Gd-N,N′-dioxide complex. Under the directing effect of the sulfonamide-moiety the ring-opening reaction proceeded selectively at the C-3 position in highly enantioselective manner furnishing various Ts- and SES-protected 3-amino-3-phenylpropan-2-olamines as products.
Keywords: Ring opening, Kinetic resolution, Regioselective, Amino alcohols, Gadolinium-catalysis
In 2012 our group developed the first asymmetric epoxidation of alkenyl sulfonamides catalyzed by Hf-BHA (bishydroxamic acid) system.[1] In 2014 our group discovered that W-salts were able to catalyze highly regioselective ring-opening of 2,3-epoxy sulfonamides with various amines, alcohols, phenol and halides as nucleophiles.[2] The combination of asymmetric epoxidation and regioselective aminolysis of enantioenriched amino epoxides provides a direct access to achieve chiral 1,3-diaminoalkan-2-ols. However, the Hf-catalyzed enantioselective epoxidation of alkenyl sulfonamides is only applicable to terminal and aliphatic olefins. As 3-amino-3-phenylpropan-2-olamine is a characteristic structural unit present in a large number of biologically active compounds and drug candidates to treat conditions ameliorated by monoamine reuptake[3], it is highly desirable to develop a method to synthesize these compounds in highly enantioselective manner. Recently, our group reported the first regio- and enantioselective aminolysis of 2,3-epoxy and 3,4-epoxy alcohols, which were promoted by W-BHA and Ni-BINAM (1,1′-binaphthyl-2,2′-diamine) catalytic systems, respectively (Scheme 1, equations 1 and 2).[4,5] As a continuation of our research in this field, we report the first enantioselective aminolysis of trans-aromatic 2,3-epoxy sulfonamides with a variety of amines as nucleophiles and a Gd-N,N′-dioxide as catalyst furnishing diverse Ts (tosyl) and SES [(2-trimethylsilyl)-ethanesulfonyl]-protected 3-amino-3-phenyl-propan-2-olamines as products in complete regiocontrol and high enantioselectivities (Scheme 1, equation 3).
Scheme 1.
Enantioselective aminolysis of epoxy allylic alcohols (Equation 1), epoxy homoallylic alcohols (Equation 2) and 2,3-epoxy sulfonamides (Equation 3)
For optimization of the reaction conditions, we used aniline (1a) and racemic trans-2,3-epoxy cinnamyl sulfonamide 2a as standard substrates. Initially, we employed our W-BHA and Ni-BINAM catalysts for this ring-opening reaction. Unfortunately, both reactions failed to provide the product in highly enantioselective manner. (Table 1, entries 1 and 2). Then we studied this reaction using Hf-BHA as catalyst, which proved to be good catalyst for asymmetric epoxidation of alkenyl sulfonamides. In this case the product was also obtained in low enantiomeric excess (Entry 3). Furthermore, we tested [Co(salen)] and [Cr(salen)Cl] complexes, which were successful catalysts for the kinetic resolution of terminal and aromatic epoxides as well as the desymmetrization of meso-epoxides.[6] However, the reaction using Cr-catalyst gave the product only with moderate enantioselectivity, while only traces of product were formed in the case of the Co-catalyst (Entries 4 and 5). Moreover, several other metal salts were studied using the BINAM-derivative 4 as ligand giving no significantly improved results (Entries 6–8). We also screened some other privileged chiral ligands 7–10 for this kinetic resolution reaction, but in the most cases the products were obtained either in racemic form or in very low enantioselectivities (Entries 9–11). Only in the case of a pipecolinic acid derived N,N′-dioxide 10 as ligand with La(OTf)3 the reaction provided the product with excellent yield and good asymmetric induction (Entry 12).[7] Encouraged by this result, we then investigated scandium, yttrium, cerium, samarium and ytterbium triflate for this ring-opening reaction (Entries 13–17). The best result with respect to both yield and enantioselectivity was obtained in the case of Sm(OTf)3 (Entry 16). Subsequently, a brief solvent screening was undertaken giving no better outcome (Entries 18–22). The results obtained indicated that the level of facial selectivity of this ring-opening reaction is significantly influenced by the ion size of the lanthanides tested. Therefore, we then studied several lanthanide triflates with similar ion radius as Sm(OTf)3 (Entries 23–26).[8] Finally, the optimum result was achieved in the case of Gd(OTf)3 as catalyst (Entry 25).[9]
Table 1.
Metals, ligands and solvents screening for the ring-opening reaction of trans-2,3-epoxy cinnamyl sulfonamide with aniline as nucleophile [a]
| |||||
|---|---|---|---|---|---|
| Entry | Metal | Ligand | Solvent | Yield [%][b] | ee [%] [c] |
| 1 | W(OEt)6 | 4 | Chloroform | 78 | 6 |
| 2 | Ni(ClO4)2•6H2O | 5 | Chloroform | 71 | 16 |
| 3 | Hf(OTf)4 | 6 | Chloroform | 57 | 36 |
| 4 | [d] | [d] | Chloroform | traces | n.d. |
| 5 | [e] | [e] | Chloroform | 94 | 57 |
| 6 | CrCl3•3THF | 5 | Chloroform | 61 | 15 |
| 7 | Co(ClO4)2•6H2O | 5 | Chloroform | 81 | 36 |
| 8 | Mn(ClO4)2•6H2O | 5 | Chloroform | 65 | 56 |
| 9 | Co(ClO4)2•6H2O | 7 | Chloroform | 50 | 8 |
| 10 | Co(ClO4)2•6H2O | 8 | Chloroform | 84 | 0 |
| 11 | Co(ClO4)2•6H2O | 9 | Chloroform | 67 | 0 |
| 12 | La(OTf)3 | 10 | Chloroform | 95 | 78 |
| 13 | Sc(OTf)3 | 10 | Chloroform | 22 | 95 |
| 14 | Y(OTf)3 | 10 | Chloroform | 92 | 60 |
| 15 | Ce(OTf)3 | 10 | Chloroform | 94 | 83 |
| 16 | Sm(OTf)3 | 10 | Chloroform | 97 | 94 |
| 17 | Yb(OTf)3 | 10 | Chloroform | 92 | 60 |
| 18 | Sm(OTf)3 | 10 | EtOAc | 42 | 55 |
| 19 | Sm(OTf)3 | 10 | THF | 45 | 38 |
| 20 | Sm(OTf)3 | 10 | MeCN | 40 | 58 |
| 21 | Sm(OTf)3 | 10 | t-BuOH | 67 | 60 |
| 22 | Sm(OTf)3 | 10 | Toluene | 47 | 46 |
| 23 | Pr(OTf)3 | 10 | Chloroform | 95 | 86 |
| 24 | Eu(OTf)3 | 10 | Chloroform | 96 | 91 |
| 25 | Gd(OTf)3 | 10 | Chloroform | 98 | 96 |
| 26 | Te(OTf)3 | 10 | Chloroform | 97 | 80 |
Unless otherwise specified, reactions were performed on a 0.25 mmol scale of aniline (1a) using 2.3 equiv. racemic trans-2,3-epoxy cinnamyl sulfonamide 2a, 10 mol% metal salts and 12 mol% ligand at r.t. in 8.0 mL solvent.
Yields of the isolated product after flash chromatography were based on aniline.
All enantiomeric excesses were determined by HPLC-analysis on chiral stationary phase.
[Co(salen)] was used as the catalyst.
[Cr(salen)Cl] was used as the catalyst.
After establishing the best reaction conditions we started to evaluate the substrate spectrum of this reaction (Table 2). We first reacted diverse primary and secondary aromatic amines with the epoxide 2a. Generally, the reactions proceeded smoothly at room temperature affording the products 3a–m in excellent yields and enantiomeric excesses. Subsequently, we investigated the substrate scope further by employing diverse aliphatic amines as nucleophiles, which turned out to be less reactive, assumably due to the higher basicity of aliphatic amines decreasing the Lewis acidity of the Gd-complex. Thus the reactions were conducted at 55 °C furnishing the products 3n-t in high yields and enantioselectivities. Furthermore, we studied the substrate scope by varying the structure of 2,3-epoxy sulfonamides. In the cases of substituted trans-2,3-epoxy cinnamyl sulfonamides both reactions provided the products 3u and 3v in high yields and enantioselectivities. Substitution at the 2-position of the 2,3-epoxy sulfonamide showed detrimental effect to the asymmetric induction giving the product 3w in moderate enantioselectivity. Remarkably, all the reactions mentioned above proceeded with complete regioselectivities in favor of the formation of C-3-regioisomers. When terminal and aliphatic trans-2,3-epoxy sulfonamides were used as substrates, the products 3x and 3y were obtained in low enantiomeric excesses, albeit with complete regioselectivities. In the case of the cis-epoxy sulfonamide as substrate the reaction proceeded with not only low facial selectivity but also low regioselectivity. Moreover, we also studied the use of SES-protected 2,3-epoxy amines as precursors for this Gd-catalyzed reaction, since SES-group is known as simply removable protective group.[10] To our delight, almost same selectivity was found in comparison to the Ts-protected epoxy amines. In addition, 2-nosyl-protected 2,3-epoxy amine also turned out to be competent substrate.
Table 2.
Kinetic resolution by ring-opening reactions of trans-2,3-epoxy sulfonamides with various amines as nucleophiles.[11],[a–e]
|
Unless otherwise specified, reactions were performed on a 0.25 mmol scale of amines 1 using 2.3 equiv. racemic trans-2,3-epoxy sulfonamides (2), 10 mol% Gd(OTf)3 and 12 mol% N,N′-dioxide 10 at room temperature in 8.0 mL chloroform.
Unless otherwise specified, the products were obtained with complete regioselectivities C3:C2>99:1, which were determined by 1H-NMR-spectroscopy and HPLC-analysis on chiral stationary phase.
Yields of the isolated products after flash chromatography were based on the amines used.
All enantiomeric excesses were determined by HPLC-analysis on chiral stationary phase.
Reaction temperature: r.t. for 3a–h, 3j, 3l, 3r, 3u, 3v and 3x; 55 °C for 3i, 3k, 3m, 3n–q, 3s, 3t, 3w, and 3y-3cc.
ent-N,N′-dioxide was used as ligand.
The enantiomeric excesses were determined by HPLC-analysis on chiral stationary phase on the corresponding acetylated derivatives.
The enantiomeric excess was determined by HPLC-analysis on chiral stationary phase on the corresponding trimethylsilylated derivative.
Reactions were performed using 20 mol% Gd(OTf)3 and 24 mol% N,N′-dioxide 10.
Depicted is only the relative configuration. The absolute stereochemistry is not determined.
SES: (2-trimethylsilyl)-ethanesulfonyl.
Ns: 2-nitrobenzenesulfonyl.
As demonstrated in Scheme 2, enantioenriched chiral amines were also employed as precursors in this Gd-catalyzed kinetic resolution. Interestingly, both (R)- and (S)-α-methylbenzylamine favored the same enantiomer of the epoxide in the ring opening process affording two different diastereomers as products both in high diastereomeric ratios (Scheme 2, equations 1 and 2). Additionally, in the absence of the chiral ligand the reaction proceeded with a low diastereoselectivity in favor of the enantiomer, which was disfavored in the reaction employing the chiral N,N′-dioxide ligand (Equation 3). These results indicated that the ligand-effect is predominant over the substrate effect in this Gd-catalyzed ring opening reaction with α-methylbenzylamines as nucleophiles.
Scheme 2.
Kinetic resolution by ring-opening reactions of trans-2,3-epoxy cinnamyl sulfonamide 2a with enantioenriched α-methylbenzylamines.[11]
Since this ring-opening process is a kinetic resolution of racemic 2,3-epoxy sulfonamides, we selected two reactions using an aliphatic and an aromatic amine, respectively, to determine the s-factor. In both cases the recovered 2,3-epoxy sulfonamide was obtained in good enantioselectivities and the s-factors were determined to be 47.1 and 264, respectively (Scheme 3)[12].
Scheme 3.
Determination of the s-factors of the kinetic resolution of racemic trans-2,3-epoxy cinnamyl sulfonamide with aniline as nucleophile.
To study the directing effect of the sulfonamide group, first, we performed the Gd-catalyzed reaction using unfunctionalized trans-1-phenyl propylene oxide as substrate. In this case the product was afforded in low yield and enantioselectivity (Table 3, entry 1). Furthermore, we conducted a series of control reactions using epoxy sulfone, Boc-protected epoxy amine and Boc-protected epoxy sulfonamide as substrates. The reactions employing the epoxy sulfone and the Boc-protected epoxy amine as substrates afforded similar results in comparison to trans-1-phenyl propylene oxide (Entries 3 and 4), while in the case of Boc-protected epoxy sulfonamide no desired product was formed (Entry 5). The results revealed that the presence of NH-sulfonyl moiety as anchoring group is crucial for both the reactivity and the facial selectivity of this ring-opening reaction.
Table 3.
Enantioselective ring-opening reactions of trans-2,3-epoxy cinnamyl sulfonamide and its analogues with aniline as nucleophile [11],[a]
| |||||
|---|---|---|---|---|---|
| Entry | R | n | t (°C) | Yield [%][b] | ee [%] [c] |
| 1 | Me | 0 | 55 | 21 | 14 |
| 2 | NHTs | 1 | r.t. | 98 | 96 |
| 3 | SO2pTolyl | 2 | 55 | 52 | 26 |
| 4 | NHBoc | 1 | 55 | 23 | 30 |
| 5 | BocNTs | 1 | 55 | 0 | - |
Unless otherwise specified, reactions were performed on a 0.25 mmol scale of aniline (1a) using 2.3 equiv. racemic epoxides, 10 mol% Gd(OTf)3 and 12 mol% N,N′-dioxide 10 at room temperature or 55 °C in 8.0 mL chloroform.
Yields of the isolated products after flash chromatography based on aniline.
All enantiomeric excesses were determined by HPLC-analysis on chiral stationary phase.
In summary, we developed an unprecedented enantioselective aminolysis of aromatic trans-2,3-epoxy amines derivatives with sulfonamide-moiety as directing group. This kinetic resolution was efficiently promoted by the Gd-N,N′-dioxide catalytic system. Various aromatic and aliphatic amines were successfully employed as nucleophiles for the ring opening of aromatic trans-2,3-epoxy sulfonamides furnishing diverse Ts- and SES-protected 3-amino-3-phenylpro-pan-2-olamines as products in complete regiocontrol and high enantioselectivities. Furthermore, the successful of chiral amines in this Gd-catalyzed aminolysis promised the use of this method to prepare structurally complex amino alcohols bearing multiple stereogenic centers.
Supplementary Material
Acknowledgments
Japan Society for the Promotion of Science (JSP-ACT-C) and the National Institutes of Health (NIH, 2R01GM068433) are greatly appreciated for providing financial support. C. W. thanks the Alexander von Humboldt Foundation for his postdoctoral fellowship. We would also like to thank Dr. Alexander Filatov for X-ray crystal structure analysis.
References
- 1.Olivares-Romero JL, Li Z, Yamamoto H. J Am Chem Soc. 2012;134:5440–5443. doi: 10.1021/ja211880s. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.a) Wang C, Yamamoto H. J Am Chem Soc. 2014;136:6888–6891. doi: 10.1021/ja5029809. [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Wang C, Yamamoto H. Org Lett. 2014;16:5937–5939. doi: 10.1021/ol503091n. [DOI] [PubMed] [Google Scholar]
- 3.a) Kim CY, Mahaney PE, Trybulski EJ, Zhang P, Terefenko EA, McComas CC, Marella MA, Coghlan RD, Heffernan GD, Cohn ST, Vu AT, Sabatucci JP, Ye F. 20050222148 A1. Patent US. 2005; b) Mahaney PE, Heffeman GD, Coghlan RD, Cohn ST, Kim CY, Jenkins DJ, Douglas J, Marella MA, Anthony M, Mccomas CC, Sabatucci JP, Terefenko EA. 20070072897 A1. Patent US. 2007; c) Vu AT, Terefenko EA, Moore WJ, Cohn ST, Zhang P, Mahaney PE, Trybulski EJ. 2008042362 A1. Patent WO. 2008; d) Vu AT, Cohn ST, Terefenko EA, Moore WJ, Zhang P, Mahaney PE, Trybulski EJ, Goljer I, Dooley R, Bray JA, Johnston GH, Leiter J, Deecher DC. Bioorg Med Chem Lett. 2009;19:2464–2467. doi: 10.1016/j.bmcl.2009.03.054. [DOI] [PubMed] [Google Scholar]; e) Vu AT, Cohn ST, Zhang P, Kim CY, Mahaney PE, Bray JA, Johnston GH, Koury EJ, Cosmi SA, Deecher DC, Smith VA, Harrison JE, Leventhal L, Whiteside GT, Kennedy JD, Trybulski EJ. J Med Chem. 2010;53:2051–2062. doi: 10.1021/jm901559e. [DOI] [PubMed] [Google Scholar]
- 4.Wang C, Yamamoto H. Angew Chem Int Ed. 2014;53:13920–13923. doi: 10.1002/anie.201408732. [DOI] [PMC free article] [PubMed] [Google Scholar]; Angew Chem. 2014;126:14140–14143. [Google Scholar]
- 5.Wang C, Yamamoto H. J Am Chem Soc. 2015;137:4308–4311. doi: 10.1021/jacs.5b01005. [DOI] [PubMed] [Google Scholar]
- 6.a) Martínez LE, Leighton JL, Carsten DH, Jacobsen EN. J Am Chem Soc. 1995;117:5897–5898. [Google Scholar]; b) Larrow JF, Schaus SE, Jacobsen EN. J Am Chem Soc. 1996;118:7420–7421. [Google Scholar]; c) Tokunaga M, Larrow JF, Kakiuchi F, Jacobsen EN. Science. 1997;277:936–938. doi: 10.1126/science.277.5328.936. [DOI] [PubMed] [Google Scholar]; d) Ready JM, Jacobsen EN. J Am Chem Soc. 1999;121:6086–6087. [Google Scholar]; e) Schaus SE, Brandes BD, Larrow JF, Tokunaga M, Hansen KB, Gould AE, Furrow ME, Jacobsen EN. J Am Chem Soc. 2002;124:1307–1315. doi: 10.1021/ja016737l. [DOI] [PubMed] [Google Scholar]; f) Bandini M, Cozzi PG, Melchiorre P, Umani-Ronchi A. Angew Chem Int Ed. 2004;43:84–87. doi: 10.1002/anie.200352073. [DOI] [PubMed] [Google Scholar]; Angew Chem. 2004;116:86–89. [Google Scholar]; g) Bartoli G, Bosco M, Carlone A, Locatelli M, Massaccesi M, Melchiorre P, Sambri L. Org Lett. 2004;6:2173–2176. doi: 10.1021/ol049372t. [DOI] [PubMed] [Google Scholar]; h) White DE, Tadross PM, Lu Z, Jacobsen EN. Tetrahedron. 2014;70:4165–4180. doi: 10.1016/j.tet.2014.03.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.For reviews on the use of pipecolinic acid or proline derived N,N′-dioxides as chiral ligands in the asymmetric catalysis, see: Liu X, Lin L, Feng X. Acc Chem Res. 2011;44:574–587. doi: 10.1021/ar200015s.Liu X, Lin L, Feng X. Org Chem Front. 2014;1:298–302.
- 8.Greenwood NN, Earnshaw A. Chemistry of Elements. 2. Butterworth-Heinemann; Oxford: 1997. pp. 1227–1249. [Google Scholar]
- 9.For an example on the N,N′-dioxide-Gd catalysis, see: Yao Q, Wang Z, Zhang Y, Liu X, Lin L, Feng X. J Org Chem. 2015;80 doi: 10.1021/acs.joc.5b00649.
- 10.a) Weinreb SM, Demko DM, Lessen TA. Tetrahedron Lett. 1986;27:2099–2102. [Google Scholar]; b) Garigipati RS, Tschaen DM, Weinreb SM. J Am Chem Soc. 1990;112:3475–3482. [Google Scholar]
- 11.For details on the determination of the absolute stereochemistry of the amino alcohols, see Supporting Information: Page 21–22.
- 12.The selectivity factor s was calculated as s= ln[(1−C)/(1−ee)]/ln[(1−C)(1+ee)], where C is the conversion.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.