Abstract
A new asymmetric synthesis of bicyclic pyrazolidinones through an alkaloid-catalyzed formal [3 + 2] cycloaddition of in situ-generated ketenes and azomethine imines is described. The products were formed in good to excellent yields (52–99% for 17 examples), with good to excellent diastereoselectivity (dr 5:1 to 27:1 for 11 examples), and with excellent enantioselectivity in all cases (≥96% ee). This method represents the first unambiguous example of an enantioselective reaction between ketenes and a 1,3-dipole.
Graphical Abstract
The synthesis of bicyclic pyrazolidinones has recently attracted attention due to the presence of the pyrazolidinone structural motif in many biologically active complex molecules.1 Some examples include anti-Alzheimer’s disease molecule 1 and related bicyclic pyrazolidinone derivatives that exhibit herbicidal, pesticidal, and antibiotic activity (Scheme 1).1 However, the catalytic asymmetric synthesis of pyrazolidinones of structure 1 has proven an elusive target.2 Enantioselective access to pyrazolidinone motifs could be envisaged through a chiral catalyst-controlled 1,3-dipolar cycloaddition of an azomethine imine with a suitable dipolarophile reactant partner.3 A number of research groups have followed that strategy.4–8 In a seminal contribution, Fu’s group reported a Cu(I)/chiral phosphaferrocene oxazoline-catalyzed asymmetric synthesis of the closely related bicyclic pyrazolines from azomethine imines and alkynes.4 On the other hand Studer and co-workers showed that structurally novel pyrazolidinones 2 containing a tetrahydroisoquinoline moiety could be assembled through a chiral benzotetramisole-catalyzed 1,3-dipolar cycloaddition of azomethine imines with mixed anhydrides.6 Although very high enantioselectivity could be obtained with this method, it is notably limited to the tetrahydroisoquinoline-containing imines, and as such is unable to provide access to the bicyclic pyrazolidinones represented by 1 or related fused bicyclic 5-membered ring systems. It is also restricted to aryl substituents (from the anhydride) at the stereogenic center α to the imide carbonyl. This reaction is proposed to involve an ammonium enolate intermediate, most likely accessed through deprotonation of an acyl ammonium precursor. Ye and coworkers more recently reported the N-heterocyclic carbene-catalyzed reaction of α-chloroaldehydes with azomethine imines to give pyrazolidinones 2 with excellent enantioselectivity in most cases.7 Brière and co-workers demonstrated that a related and unexpected regioisomer of a pyrazolidinone (3) could be prepared through the (DHQ)2PHAL-catalyzed Knoevenagel-aza-Michael cyclocondensation reaction of Meldrum’s acid with various azomethine imines, albeit with just one example formed in ≥90% ee.8 A significant drawback of this method is its restriction to Meldrum’s acid, and as a result the lack of a versatile diastereoselective variant, and moreover the lack of consistently high enantioselectivity across a range of examples. It should be noted that in all of these prior literature examples of organonucleophile-catalyzed enantioselective [3 + 2] cycloadditions, ketenes were not demonstrated to be involved as a reactant partner.
Scheme 1.
Plan for Synthesis of Bicyclic Pyrazolidinones.
As part of our interest in the development of new methodologies for the asymmetric synthesis of 5-membered ring structures from ketenes, we proposed that the pyrazolidinone motif 1 could be straightforwardly assembled through a chiral nucleophile-catalyzed formal [3 + 2] cycloaddition of azomethine imines of type 4 with in situ-generated ketenes 5. Our results described herein represent the first unambiguous example of an enantioselective reaction between ketenes and a 1,3-dipole, and only the second example of an asymmetric [3 + 2] cycloaddition reaction involving ketenes (Scheme 1).9,10
We began our studies by investigating alkaloid and phosphine catalysts which we and others had previously found successful for promoting formal [2 + 2] cycloadditions of ketenes with various coupling partners.10–13 In stark contrast to phosphines, alkaloid catalysts were found to be very successful promoters under the reaction conditions tested (in CH2Cl2, −25 °C, in situ ketene generation, acyl chloride added slowly). Slow addition of the acyl chloride over 10 h to the reaction solution (containing catalyst, azomethine imine, and Hünig’s base) was found to be beneficial to the yield of bicyclic pyrazolidinone, and limited both ketene homodimerization and aldehyde formation.12,14
Initially Mequinidine and TMS-quinine were found to be effective catalysts for providing high enantiocontrol, albeit accompanied by poor diastereoselectivity (entries 1 and 2). A number of Lewis acid salt additives were investigated in an attempt to improve diastereoselectivity, but all attempts met with failure (see Supporting Information (SI)). Among the many other alkaloid catalysts investigated (DHQ)2PHAL was found to be optimal, providing the desired product in excellent yield, with excellent enantioselectivity, and, significantly, with improved diastereoselectivity (entry 6). A catalyst loading of as low as 2.5 mol % was found to be effective in terms of providing the desired product in good yield, dr, and ee (entry 7). Indeed, it was only when the reaction was run using a 1 mol % catalyst loading that a significant drop in yield (to ca. 45%) was observed, although with preservation of excellent enantioselectivity (94% ee).
We then went on to evaluate the substrate scope of the (DHQ)2PHAL-catalyzed methodology using 10 mol% of the catalyst to ensure complete conversion and limit formation of aldehyde byproduct.14 The method displayed excellent tolerance of substrate variation with respect to changes in ketene substituent; methyl, ethyl, n-propyl, n-hexyl and acetoxy-substituents were all found to proceed with excellent enantioselectivity. In most cases, high yields were obtained (typically 70–99%), but It appears that AcO-substituted pyrazolidinones are somewhat acid-sensitive and that the slightly lower isolated yields (52–69%) are due to decomposition during purification on silica. The method was compatible with a variety of substituents on the aryl ring of the imines – both electron donating and electron withdrawing were tolerated with excellent enantioselectivity obtained in all cases. Importantly, both enantiomers of bicyclic pyrazolidinone products can be accessed by employing the pseudoenantiomeric (DHQ)2PHAL 9 and (DHQD)2PHAL 10 (see entries 3–6 and 12–17). Remarkably, for all examples examined during the study of substrate scope, enantioselectivity lower than 96% ee was never observed.
Diastereoselectivity favoring formation of the trans (anti)-diastereomer was moderate to good (dr up to 6.5:1) in most cases when an alkyl substituent was present on the ketene (Table 2, entries 1–11). The major diastereomer of the products was determined to be the trans-isomer by X-ray crystallographic analysis of the major isomer of 1f (see SI). The major enantiomer from the (DHQD)2PHAL 10-catalyzed formation of 1f was determined to be the (2S,3R)-enantiomer by X-ray crystallography. By analogy, the products of all 10-catalyzed reactions were assigned the (2S,3R)-configuration, while products of all the 9-catalyzed reactions were assigned the (2R,3S)-configuration. The absolute stereochemical outcome is consistent with models previously proposed by the Calter and Lectka groups for related alkaloid-catalyzed reactions.10a,15 Examples obtained with moderate diastereoselectivity could be enriched to high diastereopurity through simple recrystallization; For example, 1f (dr 3:1) was recrystallized from CH2Cl2/pentane to provide 1f in good yield (67%) and with excellent diastereomeric purity (dr 37:1). Interestingly, when an OAc substituent was present on the ketene, excellent levels of diastereoselectivity were obtained (entries 12–17). The origin of this increase in diastereoselectivity is currently under investigation.
Table 2.
Scope of Alkaloid-Catalyzed [3 + 2] cycloaddition of Ketenes with Azomethine Imines.
| ||||||
|---|---|---|---|---|---|---|
| entry/1 | cat. | R1 | R2 | % yielda | drb | % eec |
| 1/1a | 9 | Me | Ph | 90 | 3:1 | 99 |
| 2/1b | 9 | Me | 4-MeC6H4 | 94 | 5:1 | 98 |
| 3/1c | 9 | Me | 2-MeC6H4 | 85 | 5:1 | 99 |
| 4/1d | 10 | Me | 2-MeC6H4 | 80 | 4:1 | 99 |
| 5/1e | 9 | Me | 4-MeOC6H4 | 99 | 6.5:1 | 99 |
| 6/1f | 10 | Me | 4-MeOC6H4 | 97 | 3:1 | 99 |
| 7/1g | 9 | Me | 4-FC6H4 | 95 | 4:1 | 98 |
| 8/1h | 10 | Me | 4-FC6H4 | 93 | 3:1 | 99 |
| 9/1i | 9 | Et | 4-MeOC6H4 | 74 | 4:1 | 98 |
| 10/1j | 9 | n-Pr | 4-MeOC6H4 | 81 | 5:1 | 99 |
| 11/1k | 9 | n-Hex | 4-MeOC6H4 | 84 | 6:1 | 99 |
| 12/1l | 9 | OAc | 4-MeOC6H4 | 58 | 27:1 | 99 |
| 13/1m | 10 | OAc | 4-MeOC6H4 | 55 | 18:1 | 96 |
| 14/1n | 9 | OAc | 4-MeC6H4 | 64 | 22:1 | 99 |
| 15/1o | 10 | OAc | 4-MeC6H4 | 52 | 19:1 | 99 |
| 16/1p | 9 | OAc | Ph | 69 | 16:1 | 99 |
| 17/1q | 10 | OAc | Ph | 64 | 12:1 | 96 |
Isolated yield for both diastereomers.
dr determined by 1H NMR or HPLC analysis of crudes.
ee determined by chiral HPLC or chiral GC analysis for major diastereomer. (R,S)-isomer is the major isomer from the 9-catalyzed reaction. (S,R)-isomer is the major isomer from the 10-catalyzed reaction.
Our proposed mechanism for the (DHQ)2PHAL-catalyzed formation of bicyclic pyrazolidinones involves addition of the alkaloid catalyst to the less sterically hindered side of the monosubstituted ketene to afford an ammonium enolate I in stereoselective fashion (in accord with previous mechanistic proposals).10,11 Addition of the ammonium enolate to the azomethine imine provides access to zwitterionic species II, which subsequently undergoes 5-exo-trig cyclization to afford pyrazolidinone 1, along with simultaneous regeneration of the alkaloid catalyst. The participation of a ketene rather than an acyl chloride in reaction with the alkaloid catalyst was supported by the following experimental observations: The use of pre-generated methylketene (rather than propionyl chloride) led to the formation of bicyclic pyrazolidinone 1e in virtually the same enantiomeric excess (98% ee) and diastereoselectivity (7:1) as when propionyl chloride and i-Pr2NEt were used. Significantly the major isomer was formed with the same sense of enantioselectivity and diastereoselectivity as was observed for the in situ-generated ketene case. Other mechanistic observations included: (1) Reaction of azomethine imine with propionyl chloride was examined under typical reaction conditions (−25 °C, in CH2Cl2, but with no catalyst), and resulted in no product formation, thus suggesting that formation of azomethine imide followed by enolization is an unlikely mechanistic pathway. (2) In an experiment where (DHQ)2PHAL and azomethine imine were dissolved in CD2Cl2, no evidence for interaction of (DHQ)2PHAL with azomethine imine was observed by 1H NMR spectroscopy. (3) In the absence of the alkaloid catalyst, a complex mixture of products resulted.
The formation of the major diastereomer of the product with trans-relative stereochemistry may be rationalized through a transition state in which the attacking ammonium enolate is oriented gauche to the C=N of the iminium electrophile (Scheme 3). This arrangement is presumably stabilized relative to competing antiperiplanar arrangements, due to Coulombic attraction between the negatively charged oxygen of the enolate and the positively charged nitrogen of the imine.16 When an OAc substituent was present on the ketene (R1 = OAc in Scheme 3), very good to excellent levels of diastereoselectivity were obtained (Table 2, entries 12–17). We speculate that a favorable interaction between lone pairs on the carbonyl oxygen of the OAc group and the positively charged imine further stabilizes the putative gauche arrangement in the transition state (Scheme 3) leading to a greater preference for formation of the trans-isomer.
Scheme 3.
Model for Diastereoselection.
In summary, we have developed an alkaloid-catalyzed asymmetric synthesis of bicyclic pyrazolidinones, which represents the first unambiguous enantioselective [3 + 2] of ketenes with a 1,3-dipole. Pyrazolidinones were formed from in situ-generated ketenes and azomethine imines, with excellent enantioselectivity in all cases (17 examples). Future work will involve applications of the reported methodology and the development of other new nucleophile-catalyzed enantioselective [3 + 2] cycloaddition reactions of ketenes.
Supplementary Material
Scheme 2.
Proposed Mechanism for Catalytic Asymmetric Synthesis of Bicyclic Pyrazolidinones.
Table 1.
Optimization of Alkaloid-Catalyzed [3 + 2] cycloaddition of Ketenes with Azomethine Imines.
| ||||||
|---|---|---|---|---|---|---|
| entry | cat. | temp (°C) | additive | % yielda | drb | % eec |
| 1 | 6 | −25 | none | 80 | 1:1 | 98 |
| 2 | 7 | −25 | none | 63 | 1.5:1 | 98 |
| 3 | 6 | −78 | none | (70) | 1:1 | |
| 4 | 7 | −25 | none | (40) | 1.5:1 | |
| 5 | 8 | −25 | none | (25) | 2:1 | |
| 6 | 9 | −25 | none | 90 | 3:1 | 99 |
| 7d | 9 | −25 | none | 85 | 3:1 | 99 |
Isolated yield for both diastereomers. Conversion as determined by GC-MS in parentheses.
dr determined by 1H NMR or HPLC analysis of crudes. (R,R)-isomer = major in most cases.
ee determined by chiral HPLC or chiral GC analysis for major diastereomer.
2.5 mol % of catalyst used.
Acknowledgments
Support has been provided by the National Science Foundation and the National Institutes of Health: Grant Nos. CHE-1213638, CHE-1463728, and R15GM107800 to N.J.K.
Footnotes
Notes
The authors declare no competing financial interest.
Experimental procedures, spectroscopic data, chromatograms for all new compounds, and cifs for 1f and 1p. This material is available free of charge via the Internet at http://pubs.acs.org.
References
- 1.(a) Hanessian S, McNaughton-Smith G, Lombart HG, Lubell WD. Tetrahedron. 1997;53:12789–12854. [Google Scholar]; (b) Fischer R, Bretschneider T, Gesing ERF, Feucht D, Kuck K-H, Loesel P, Malsam O, Arnold C, Auler T, Hills MJ, Kehne H. 2005016873. PCT Int Appl WO. Chem. Abstr. 2005, 142, 261530.; (c) Kosower EM, Radkowsky AE, Fairlamb AH, Croft SL, Nea RA. Eur J Med Chem. 1995;30:659–671. [Google Scholar]; (d) Kosower EM, Hershkowitz E. 94658. Isr Patent, ISXXAQ IL. Chem. Abstr. 1994, 122, 214077.
- 2.For a diastereoselective approach: Winterton SE, Ready JM. Org Lett. 2016;18:2608–2611. doi: 10.1021/acs.orglett.6b01104.
- 3.Hashimoto T, Maruoka K. Chem Rev. 2015;115:5366–5412. doi: 10.1021/cr5007182. [DOI] [PubMed] [Google Scholar]
- 4.Shintani R, Fu GC. J Am Chem Soc. 2003;125:10778–10779. doi: 10.1021/ja036922u. [DOI] [PubMed] [Google Scholar]
- 5.Na R, Jing C, Xu Q, Jiang H, Wu X, Shi J, Zhong J, Wang M, Benitez D, Tkatchouk E, Goddard WA, III, Guo H, Kwon O. J Am Chem Soc. 2011;133:13337–13348. doi: 10.1021/ja200231v. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hesping L, Biswas A, Daniliuc CG, Mück-Lichtenfeld C, Studer A. Chem Sci. 2015;6:1252. doi: 10.1039/c4sc02612h. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gao ZH, Chen XY, Cheng JT, Liao WL, Ye S. Chem Commun. 2015;51:9328–9331. doi: 10.1039/c5cc01238d. [DOI] [PubMed] [Google Scholar]
- 8.Pair E, Berini C, Noël R, Sanselme M, Levacher V, Brière JF. Chem Commun. 2014;50:10218–10221. doi: 10.1039/c4cc04852k. [DOI] [PubMed] [Google Scholar]
- 9.Shao PL, Chen XY, Ye S. Angew Chem Int Ed. 2010;49:8412–8416. doi: 10.1002/anie.201003532. [DOI] [PubMed] [Google Scholar]
- 10.(a) Orr RK, Calter MA. Tetrahedron. 2003;59:3545–3565. [Google Scholar]; (b) Paull DH, Weatherwax A, Lectka T. Chem Rev. 2009;65:6771–6803. [Google Scholar]; (c) Chen S, Salo EC, Kerrigan NJ. Lewis Base and Acid Catalysts. In: List B, editor. Science of Synthesis Reference Library, Asymmetric Organocatalysis. Vol. 1. Chapter 1.1.10. Thieme; Stuttgart: 2012. pp. 455–496. [Google Scholar]
- 11.Ibrahim AA, Nalla D, Van Raaphorst M, Kerrigan NJ. J Am Chem Soc. 2012;134:2942–2945. doi: 10.1021/ja211678m. [DOI] [PubMed] [Google Scholar]
- 12.Calter MA, Orr RK. Org Lett. 2003;5:4745–4748. doi: 10.1021/ol0359517. [DOI] [PubMed] [Google Scholar]
- 13.(a) Chen S, Salo EC, Wheeler KA, Kerrigan NJ. Org Lett. 2012;14:1784–1787. doi: 10.1021/ol3003783. [DOI] [PubMed] [Google Scholar]; (b) Chen S, Mondal M, Ibrahim AA, Wheeler KA, Kerrigan NJ. J Org Chem. 2014;79:4920–4929. doi: 10.1021/jo500486e. [DOI] [PubMed] [Google Scholar]; (c) Mondal M, Chen S, Othman N, Wheeler KA, Kerrigan NJ. J Org Chem. 2015;80:5789–5794. doi: 10.1021/acs.joc.5b00869. [DOI] [PubMed] [Google Scholar]
- 14.Background decomposition of azomethine imine to aldehyde was found to occur during shorter acyl chloride addition times. Formation of aldehyde also occurs in the absence of catalyst and with low catalyst loadings.
- 15.Taggi AE, Hafez AM, Wack H, Young B, Ferraris D, Lectka T. J Am Chem Soc. 2002;124:6626–6635. doi: 10.1021/ja0258226. [DOI] [PubMed] [Google Scholar]
- 16.Aggarwal VK, Harvey JN, Richardson J. J Am Chem Soc. 2002;124:5747–5756. doi: 10.1021/ja025633n. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.