Construction of Vicinal Stereocenters via Asymmetric Cyanosilylation

Abstract

Asymmetric cyanosilylation serves as an important tool to convert easily available ketones into cyanohydrins of diverse reactivity. Whereas a large library of organocatalysts and transition-metal catalysts have been identified for monoketones, cyanosilylation of more-complex substrates, particularly those giving enantioenriched vicinal stereocenters, is underexplored in comparison. Here, a pair of recently published kinetic resolution and desymmetrization methods are highlighted for their success in constructing complex vicinal stereocenters by cyanosilylation using tailored aluminum and magnesium catalysts, respectively.

1 Introduction

2 Kinetic Resolution of α-Branched Ketones

3 Desymmetrization of 1,3-Diketones

4 Conclusion

Introduction

The degrees of saturation and three-dimensionality are two molecular parameters of increasing importance in drug discovery, as they have been found to be closely related to the clinical success and promiscuity of drug candidates.[1] [2] As carbon atoms often serve as the backbones of bioactive compounds, tetrasubstituted stereocenters (carbons with four different substituents) add arguably the most structural diversity to a molecule. Furthermore, these tetrasubstituted carbons, when accompanied by an additional and neighboring stereocenter, are particularly privileged motifs in natural products and drugs, and have been prominent targets of asymmetric catalysis. Nevertheless, this task is often associated with a pair of daunting challenges. First, the crowded nature of the vicinal and highly substituted stereocenters requires the synthetic method to be tolerant of strong steric hindrance. Secondly, a stereoselective construction of the pair of chiral carbons entails not only a good control of the absolute configuration but also good control of the relative configuration of the two stereocenters. Thus, the selected chiral catalysts are expected to impose excellent stereocontrol to access one of four possible isomers, a much more complex undertaking compared with the synthesis of a single stereocenter.

Decades of the development of methods to forge vicinal stereocenters, continuously at the forefront of asymmetric catalysis, have produced a wide range of approaches, including enantioselective difunctionalization of olefins and substitution reactions of prochiral nucleophiles and electrophiles. These methods and their application in the synthesis of natural products have been summarized in several informative reviews.[3] On the other hand, this short account aims to highlight a pair of recently disclosed approaches to the construction of neighboring stereocenters of high structural diversity and complexity. These two methods both feature the use of asymmetric cyanosilylation of ketones, a transformation that has evolved into a practical and reliable tool to access chiral tertiary alcohols over the past two decades (Scheme [1]A).[4]

The rapid development and wide application of asymmetric cyanosilylation in synthesis can be attributed to three major advantages. First, as the reactants of cyanosilylation, ketones are among the most easily accessible substrates in organic synthesis. Secondly, silyl cyanides are inert toward carbonyl groups in the absence of catalysts, thereby eliminating any background reactions that may erode the stereoselectivity. Thirdly, the cyanosilylation not only creates a chiral tertiary alcohol, but also introduces an additional nitrile group that adds greatly to the synthetic value of the product. These advantages have allowed asymmetric cyanosilylation to generate enantioenriched cyanohydrins with a diversity of substituents that have been further derivatized to highly functionalized molecules. Generally, Lewis acids with chiral ligands are often employed to complex with carbonyls and to exert enantiocontrol of the cyanation, whereas organocatalysts can activate silyl cyanides to create a chiral nucleophile. Nevertheless, asymmetric cyanosilylation has rarely been applied to generate more-complex tertiary alcohols with a neighboring stereocenter, a family of chiral moieties prevalent in drugs and natural products (Scheme [1]B).

2 Kinetic Resolution of α-Branched Ketones

A straightforward approach to generate vicinal stereocenters via cyanosilylation is the resolution of a racemic mixture of α-chiral ketones. The existing α-stereocenter is expected to influence the rate of cyanosilylation mediated by a chiral catalyst and to create a match–mismatch scenario. When the rate difference between enantiomeric ketones is large enough, an effective resolution can be reached. Although the resolution method has a yield cap of enantiopure isomers (i.e., 50%), both unreacted α-branched ketones and complex chiral cyanohydrins can be obtained in a single reaction, adding greatly to its synthetic applicability.

The resolution strategy was validated by Zhou and co-workers in 2021 with a tailored pair of catalysts (salen)AlCl 3 and phosphorane 4 (Scheme [2]A).[5] The phosphorane in the combination was previously demonstrated by the authors’ laboratory to activate the salen complex by replacing the chloride anion and, together, they make up a chiral pocket for the complexation of ketones.[6] On the other hand, the addition of hexamethylphosphoramide (HMPA) proved beneficial to the selectivity of the resolution, supposedly by complexing with the silyl cyanide as a Lewis base and thus enhancing its size for better enantiocontrol (Scheme [2]B).

The customized cyanosilylation conditions can resolve a diverse panel of α-substituted ketones, indicative of a high application potential to access assorted structures (Scheme [3]A). Generally, excellent yields, diastereoselectivity, and enantioselectivity of cyanohydrins can be obtained with slightly more than half an equivalent of the silyl cyanide (i.e., 56 mol% to ketone), while the resolution can provide enantioenriched unreacted ketones in equal efficiency and selectivity with slightly more cyanides (i.e., 60 mol% to ketone). Vicinal stereocenters containing a tertiary carbon with a variety of (hetero)aryl and alkyl substituents can be forged. The scope of the cyanosilylation extends beyond α-tertiary methyl ketones, with ethyl ketone (5), α,α-diaryl (6), α,α-dialkyl (7), and α-quaternary ketones (8) all accommodated, albeit with slightly diminished yields and/or selectivities.

The synthetic application of these vicinal stereocenters from resolution was also greatly diversified by the designer silyl cyanide 2. The chlorinated cyanide not only outcompetes simple trimethylsilyl cyanide in terms of resolution selectivity, but also brings along abundant reactivity for derivatization to the cyanohydrin products with its chloromethyl substituents. Thus, the methylene motif can serve as a nucleophile and help convert the nitrile group into chloromethyl (9) or methyl (10) ketones. After a nitrile reduction, the chloride can also be substituted, and a methylated amino alcohol 11 can be obtained from the cyanohydrin.

3 Desymmetrization of 1,3-Diketones

In principle, the enantiodiscrimination between enantiomeric ketones should involve a similar mode of chiral induction to the differentiation between enantiotropic carbonyls in a prochiral diketones. When such a diketone is α,α-disubstituted, a desymmetric cyanosilylation would forge a pair of quaternary and tertiary stereocenters. Given that the prochiral substrates are easily accessible through substitution, and the cyanosilylation products have multiple functional groups (including the remaining ketone, a tertiary alcohol, and a nitrile), the desymmetrization can swiftly increase molecular complexity in a selective manner.

In 2022, our group successfully realized an asymmetric cyanosilylation using a magnesium catalyst 13 with a tetradentate chiral ligand (Scheme [4]).[7] The catalyst built upon a novel family of dinuclear zinc complexes that we previously devised for the reductive desymmetrization of malonic esters.[8] Although the magnesium catalyst shares a similar ligand scaffold with its zinc counterparts, titration and control experiments are consistent with a bifunctional and mononuclear complex with a hydrogen-bond donor (HBD), instead of a dinuclear magnesium catalyst. It is proposed that 1,3-diketones can form a ‘two-point chelation’, with two carbonyls complexing with the HBD and the Lewis acidic magnesium center, respectively. As determined from the stereochemical assignment, the magnesium catalyst has an excellent facial selectivity of addition and a slightly lower site selectivity for carbonyls.

The magnesium catalyst was effective in cyanosilylating various 1,3-diketones and provided access to one of four possible stereoisomeric cyanohydrins with good enantio- and diastereoselectivity (Scheme [5]A). A plethora of quaternary carbons substituted with diversely shaped aryl and alkyl groups can be accommodated in the desymmetrization to give complex vicinal stereocenters (e.g., 14–17). Particularly, for challenging substrates in which two substituents of the diketone are of relatively similar sizes (e.g., 17, benzyl vs ethyl, compared with the case of benzyl vs methyl), excellent enantiopurity of the major diastereomer can still be obtained, although the diastereoselectivity decreases. On the other hand, these highly functionalized chiral structures provide ample opportunities for derivatization to molecules of even higher complexity (Scheme [5]B). It has been demonstrated that fused ring (18), polysubstituted heterocycle (19), polyol (20), and amino diol (21) products can be rapidly generated from the desymmetrization product by simple operations, including hydrolysis, reduction, or cyclization.

4 Conclusion

In summary, two strategies for the construction of vicinal stereocenters by using cyanosilylation, namely kinetic resolution and desymmetrization, are demonstrated here. It is worth noting that an asymmetric cyanosilylation through dynamic kinetic resolution should be able to accomplish the same task by epimerizing the α-stereocenter of ketones and could therefore be an important future direction. These transformations can rapidly build up molecular complexity and diversity from easily available starting materials by introducing both a nitrile and a tertiary alcohol motif. It can be foreseen that the continuing development of new chiral catalysts for asymmetric cyanosilylation could help streamline the construction of complex molecules.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

The University of Hong Kong and the State Key Laboratory of Synthetic Chemistry are gratefully acknowledged.

• References

• 1 Lovering F, Bikker J, Humblet C. J. Med. Chem. 2009; 52: 6752
  CrossrefPubMedSearch in Google Scholar
• 2 Lovering F. MedChemComm 2013; 4: 515
  CrossrefSearch in Google Scholar

For selected reviews, see:
• 3a Long R, Huang J, Gong J, Yang Z. Nat. Prod. Rep. 2015; 32: 1584
  CrossrefPubMedSearch in Google Scholar
• 3b Büschleb M, Dorich S, Hanessian S, Tao D, Schenthal KB, Overman LE. Angew. Chem. Int. Ed. 2016; 55: 4156
  CrossrefPubMedSearch in Google Scholar
• 3c Zhou F, Zhu L, Pan B.-W, Shi Y, Liu Y.-L, Zhou J. Chem. Sci. 2020; 11: 9341
  CrossrefPubMedSearch in Google Scholar

For selected reviews of cyanosilylation of ketones, see:
• 4a Brunel J.-M, Holmes IP. Angew. Chem. Int. Ed. 2004; 43: 2752
  CrossrefPubMedSearch in Google Scholar
• 4b North M, Usanov DL, Young C. Chem. Rev. 2008; 108: 5146
  CrossrefPubMedSearch in Google Scholar
• 4c Kurono N, Ohkuma T. ACS Catal. 2016; 6: 989
  CrossrefSearch in Google Scholar
• 4d Pahar S, Kundu G, Sen SS. ACS Omega 2020; 5: 25477
  CrossrefPubMedSearch in Google Scholar
• 4e Wu W.-B, Yu J.-S, Zhou J. ACS Catal. 2020; 10: 7668
  CrossrefSearch in Google Scholar
• 5 Wu W.-B, Yu X, Yu J.-S, Wang X, Wang W.-G, Zhou J. CCS Chem. 2022; 4: 2140
  CrossrefPubMedSearch in Google Scholar

For selected examples, see:
• 6a Zeng X.-P, Cao Z.-Y, Wang X, Chen L, Zhou F, Zhu F, Wang C.-H, Zhou J. J. Am. Chem. Soc. 2016; 138: 416
  CrossrefPubMedSearch in Google Scholar
• 6b Zeng X.-P, Zhou J. J. Am. Chem. Soc. 2016; 138: 8730
  CrossrefPubMedSearch in Google Scholar
• 7 Xu P, Shen C, Xu A, Low K.-H, Huang Z. Angew. Chem. Int. Ed. 2022; 61: e202208443
  CrossrefPubMedSearch in Google Scholar

For selected examples, see:
• 8a Xu P, Huang Z. Nat. Chem. 2021; 13: 634
  CrossrefPubMedSearch in Google Scholar
• 8b Zheng Y, Zhang S, Low K.-H, Zi W, Huang Z. J. Am. Chem. Soc. 2022; 144: 1951
  CrossrefPubMedSearch in Google Scholar
• 8c Xu P, Liu S, Huang Z. J. Am. Chem. Soc. 2022; 144: 6918
  CrossrefPubMedSearch in Google Scholar
• 8d Liu H, Lau HM, Xu P, Chan TH, Huang Z. Nat. Commun. 2022; 13: 4759
  CrossrefPubMedSearch in Google Scholar

Corresponding Author

Publication History

Received: 09 August 2023

Accepted after revision: 25 August 2023

Accepted Manuscript online:
25 August 2023

Article published online:
05 October 2023

© 2023. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References

• 1 Lovering F, Bikker J, Humblet C. J. Med. Chem. 2009; 52: 6752
  CrossrefPubMedSearch in Google Scholar
• 2 Lovering F. MedChemComm 2013; 4: 515
  CrossrefSearch in Google Scholar

For selected reviews, see:
• 3a Long R, Huang J, Gong J, Yang Z. Nat. Prod. Rep. 2015; 32: 1584
  CrossrefPubMedSearch in Google Scholar
• 3b Büschleb M, Dorich S, Hanessian S, Tao D, Schenthal KB, Overman LE. Angew. Chem. Int. Ed. 2016; 55: 4156
  CrossrefPubMedSearch in Google Scholar
• 3c Zhou F, Zhu L, Pan B.-W, Shi Y, Liu Y.-L, Zhou J. Chem. Sci. 2020; 11: 9341
  CrossrefPubMedSearch in Google Scholar

For selected reviews of cyanosilylation of ketones, see:
• 4a Brunel J.-M, Holmes IP. Angew. Chem. Int. Ed. 2004; 43: 2752
  CrossrefPubMedSearch in Google Scholar
• 4b North M, Usanov DL, Young C. Chem. Rev. 2008; 108: 5146
  CrossrefPubMedSearch in Google Scholar
• 4c Kurono N, Ohkuma T. ACS Catal. 2016; 6: 989
  CrossrefSearch in Google Scholar
• 4d Pahar S, Kundu G, Sen SS. ACS Omega 2020; 5: 25477
  CrossrefPubMedSearch in Google Scholar
• 4e Wu W.-B, Yu J.-S, Zhou J. ACS Catal. 2020; 10: 7668
  CrossrefSearch in Google Scholar
• 5 Wu W.-B, Yu X, Yu J.-S, Wang X, Wang W.-G, Zhou J. CCS Chem. 2022; 4: 2140
  CrossrefPubMedSearch in Google Scholar

For selected examples, see:
• 6a Zeng X.-P, Cao Z.-Y, Wang X, Chen L, Zhou F, Zhu F, Wang C.-H, Zhou J. J. Am. Chem. Soc. 2016; 138: 416
  CrossrefPubMedSearch in Google Scholar
• 6b Zeng X.-P, Zhou J. J. Am. Chem. Soc. 2016; 138: 8730
  CrossrefPubMedSearch in Google Scholar
• 7 Xu P, Shen C, Xu A, Low K.-H, Huang Z. Angew. Chem. Int. Ed. 2022; 61: e202208443
  CrossrefPubMedSearch in Google Scholar

For selected examples, see:
• 8a Xu P, Huang Z. Nat. Chem. 2021; 13: 634
  CrossrefPubMedSearch in Google Scholar
• 8b Zheng Y, Zhang S, Low K.-H, Zi W, Huang Z. J. Am. Chem. Soc. 2022; 144: 1951
  CrossrefPubMedSearch in Google Scholar
• 8c Xu P, Liu S, Huang Z. J. Am. Chem. Soc. 2022; 144: 6918
  CrossrefPubMedSearch in Google Scholar
• 8d Liu H, Lau HM, Xu P, Chan TH, Huang Z. Nat. Commun. 2022; 13: 4759
  CrossrefPubMedSearch in Google Scholar