Letter pubs.acs.org/OrgLett
Asymmetric Synthesis of Carbocyclic Propellanes Lisa M. Schneider,† Volker M. Schmiedel,† Tommaso Pecchioli,† Dieter Lentz,† Christian Merten,‡ and Mathias Christmann*,† †
Institut für Chemie und Biochemie, Freie Universität Berlin, Takustraße 3, 14195 Berlin, Germany Ruhr-Universität Bochum, Organische Chemie 2, Universitätsstraße 150, 44801 Bochum, Germany
‡
S Supporting Information *
ABSTRACT: A modular synthesis of functionalized carbocyclic propellanes was developed. Formation of the first of two quaternary bridgehead centers has been achieved by desymmetrization of prostereogenic ketones by either Hajos−Parrish−Eder−Sauer−Wiechert-type processes or Werner’s catalytic asymmetric Wittig reaction. The obtained bicyclic enones were subjected to conjugate additions upon which the remaining ring was formed by olefin metathesis. All bridges are amenable to further derivatization, which renders those compounds useful as central units in fragment-based drug discovery or as ligand scaffolds.
P
followed by diastereoselective addition of alkenyl-cuprates to obtain bicyclic dienes (B → C). Ring-closing metathesis8 then leads to the propellane core structure (C → D). All three bridges of the final structures can be further functionalized. Our study commenced with the synthesis of the desymmetrization precursors from readily available cycloalkane-1,3-diones. Reductive alkylation of 4 and 5 with 4‑pentenal9 was followed by Michael addition with methyl vinyl ketone to afford triketones 6 and 7 (Scheme 2a) in good
ropeller-shaped molecules with three rings sharing a central C−C bond are commonly known as propellanes, a term coined by Ginsburg in 1966.1 The first examples of their synthesis were reported as early as 1934.2 Secondary metabolites bearing a purely carbocyclic propellane substructure are rarely found in nature, whereas heterocyclic propellanes are more common.3 Recently, small-ring propellanes have been enjoying renewed interest as “spring-loaded” electrophiles in strain-release reactions to access nonclassical bioisosteres, such as [1.1.1]bicyclopentanes.4 Larger propellanes, on the other hand, have received much less attention in drug design,5 probably due to the lack of general methods for their preparation.6 The main challenge in the asymmetric synthesis of these polycyclic systems lies in the stereoselective construction of the adjacent quaternary bridgehead carbons. Herein, we describe the first general approach to the synthesis of chiral carbocyclic propellanes (Scheme 1). It is based on the catalytic desymmetrization7 of prochiral triketones (A → B)
Scheme 2. Preparation of Prochiral Triketones
Scheme 1. General Strategy for the Synthesis of Carbocyclic Propellanes
yields over two steps. Triketones with shorter alkenyl and alkynyl side chains were prepared according to literature protocols (see the Supporting Information).10 Brominated triketones 10 and 11 were obtained from the known 2‑allylcycloalkane-1,3-diones 8 and 9 by propargylation followed by conversion to the 1‑bromoalkynes using NBS/ AgNO3 (Scheme 2b). For the 6‑membered dione, hydration of Received: March 22, 2017 Published: April 26, 2017 © 2017 American Chemical Society
2310
DOI: 10.1021/acs.orglett.7b00836 Org. Lett. 2017, 19, 2310−2313
Letter
Organic Letters the triple bond afforded trione 11 under the bromination conditions. In contrast, under the same reaction conditions, the cyclopentanedione derivative decomposed. Thus, hydration of the 1-bromoalkyne was achieved in a separate step using catalytic amounts of PPh3AuNTf211 to give the brominated triketone 10. The prochiral triketones were then transformed into bicyclic enones by either asymmetric Robinson annulations or via catalytic asymmetric Wittig reactions (Scheme 3). In the case
Scheme 4. Conjugate Addition
Scheme 3. Desymmetrization of Prochiral Triketones
a Performed on gram scale. bStarting material synthesized from 14 in three steps (see the Supporting Information). cStarting material obtained by olefin isomerization of allyl-substituted Hajos−Parrish ketone (see the Supporting Information). dStarting material literature known (see the Supporting Information).
only the cis-isomers being obtained. Unfortunately, cyclopentenone derivative 17 failed to react and gave only decomposition products after workup. The addition of 3‑butenyl cuprates to the bicyclic enones was more sluggish and afforded the higher homologues 26−29 in moderate yields as single diastereomers (Scheme 4, conditions B). Diene 29 was obtained from the allyl-substituted Hajos−Parrish ketone by Pd-catalyzed16 isomerization of the allyl group prior to the addition of butenyl cuprate. To complete the synthesis of the propellanes, dienes 18−29 were subjected to 1 mol % of Umicore M71SIPr in refluxing toluene (Scheme 5, conditions A). The ring closures leading to 5- and 6-membered carbocycles proceeded smoothly to give the propellanes 30−34 in excellent yields. When the higher homologue 22 was subjected to the reaction conditions, an inseparable mixture of the desired 7-membered carbocycle 35 as well as the 6-membered derivative as minor product was obtained due to prior Ru-catalyzed olefin isomerization,17 which could be effectively suppressed by addition of catalytic amounts of 1,4-benzoquinone (BQ). Under the optimized reaction conditions B, propellanes 35−39, containing a 7‑membered ring, were obtained in good yields. Attempts to synthesize propellanes containing a 4-membered ring by metathesis reaction of bis-vinyl derivative 24 failed. Instead, the diene underwent formal intramolecular hydrovinylation18 under the reaction conditions to give regioisomeric propellanes
of the 1,3-cyclopentanedione 6, the Robinson annulation was performed using 30 mol % of (S)-proline, giving Hajos− Parrish ketone derivative 12 in good yield and excellent enantioselectivity (Scheme 3a). For the synthesis of the homologous Wieland−Miescher derivatives 13−15, tertleucine−diamine catalysts developed by Luo afforded comparably high enantioselectivities (Scheme 3b).12 For the synthesis of iso-Hajos−Parrish ketone derivatives13 16 and 17, brominated triketones 10 and 11 were subjected to an intramolecular catalytic asymmetric Wittig reaction introduced by Werner14 (Scheme 3c). To our delight, enones 16 and 17 were obtained with high enantiomeric excess (92− 96% ee) in 50−58% yield. The purity of the starting materials turned out to be crucial to avoid premature debromination of the triketone. With the bicyclic enones 12−17 in hand, we turned our attention to the conjugate addition starting with vinyl nucleophiles (Scheme 4, conditions A). It was found that higher order cyanocuprates introduced by Lipshutz15 were effective in generating the corresponding dienes in consistent yields. With 3 equiv of the cuprate, diene 18 was obtained in 45% yield. The yield could not be improved further by using additional equivalents of nucleophile. By subjecting enones 12−17 to these reaction conditions, dienes 19−25 were obtained in moderate to excellent yields. In all cases, the reaction proceeded with complete diastereoselectivity, with 2311
DOI: 10.1021/acs.orglett.7b00836 Org. Lett. 2017, 19, 2310−2313
Letter
Organic Letters
(Scheme 6). To this end, propellane 30 was treated with L‑Selectride at low temperature, which led to reduction of the more accessible keto group in the 6-membered ring to give 42 with hydride attack coming exclusively from the side of the unsaturated bridge (see SI for X-ray data of 43). Alternatively, selective acetalization of the same keto functionality gave 44. Reduction of the remaining carbonyl group under Luche conditions20 gave alcohol 45 in good diastereoselectivity. Functionalization of the unsaturated bridge of 45 via hydroboration followed by oxidative workup gave diol 46, whereas epoxidation yielded 47 (relative configuration based on NOE experiments). The stereochemical outcome of both reactions can be rationalized by hydroxyl-mediated delivery of the electrophilic reagents attacking the electron-rich double bond. The absolute configuration of all compounds in Scheme 6 was assigned after X-ray crystallographic analysis of the bisnitrobenzoate 48. Accordingly, the absolute configuration of all propellanes synthesized via (S)-proline catalyzed Robinson annulation is analogous to the X-ray (1R,6S), whereas the propellanes synthesized using Luo’s conditions all have the opposite absolute configuration. To determine the absolute configuration of propellanes 31 and 36, built up via asymmetric catalytic Wittig reaction, both molecules were subjected to IR and VCD spectroscopy. The resulting spectra were compared with calculated spectra, obtained after structure optimization of 31 and 36 at the B3LYP/6-311++G(2d,p)/IEFPCM(CHCl3) level of theory (Figure 1). Some small regions of the experimental IR and VCD spectra were removed due to artifacts arising from total absorbance, namely the carbonyl stretching region (1760− 1710 cm−1) and a solvent band (940−870 cm−1). For 31, the remaining spectral range features several strong VCD bands, which can be correlated with the predicted spectral pattern as indicated by the band assignments. Thus, the comparison confirms the absolute configuration of 31 to be (1S,5R). In the case of 36, the propellane can adopt two conformations which are predicted to be almost equally populated (ΔG298K = 0.15 kcal/mol). Taking into account both conformers, the agreement of simulated spectra with the experimental ones again confirms the absolute configuration of 31 to be (1S,5R). In summary, we have developed the first general approach for the asymmetric synthesis of carbocyclic propellanes. Desymmetrization of the prochiral triketones using either
Scheme 5. Ring-Closing Metathesis
a
Performed on gram scale. bStarting from 24 using conditions A (10 mol % cat.), separable by flash chromatography. cStarting from 25: PtCl2 (5 mol %), PhMe, 80 °C.
40a and 40b in 71 and 16% yield, respectively. Platinumcatalyzed enyne−cycloisomerization19 of propargyl derivative 25 gave the corresponding propellane 41, containing a conjugated diene, in good yield. To demonstrate the utility of our approach toward functionalizations of the propellane core, selective transformations of each of the three bridges were investigated Scheme 6. Functionalization of the Propellane Core
2312
DOI: 10.1021/acs.orglett.7b00836 Org. Lett. 2017, 19, 2310−2313
Letter
Organic Letters
Berlin) for experimental support and Ulrike Warzok (FU Berlin) for her help. C.M. thanks the FCI for a Liebig Fellowship and the Deutsche Forschungsgemeinschaft (DFG) for support through the Cluster of Excellence RESOLV (“Ruhr Explores Solvation”, EXC 1069).
■
Figure 1. Optimized geometries for 31 (a) and 36 (b) and comparison of experimental VCD (top) and IR (bottom) spectra with predicted spectra for 31 (c) and 36 (d). The numbers indicate some characteristic band assignments.
amine-catalyzed Robinson annulations or catalytic asymmetric Wittig reactions gave the bicyclic enones in high enantiomeric excesses. Synthesis of the propellane core was completed through diastereoselective conjugate addition of unsaturated alkyl chains to the enones, followed by ring-closing metathesis. Each of the bridges of propellane 30 could be further functionalized. The absolute configuration of all obtained propellanes was determined either by X-ray crystallography or combined theoretical/experimental IR and VCD spectroscopy. The utility of propellanes in the synthesis of natural products and their use as ligands for drug discovery are currently under investigation.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b00836. X-ray crystallographic data for 21 (CIF) X-ray crystallographic data for 43 (CIF) X-ray crystallographic data for 48 (CIF) Experimental procedures, stereochemistry assignments, and characterization data of the synthesized compounds (PDF)
■
REFERENCES
(1) Altman, J.; Babad, E.; Itzchaki, J.; Ginsburg, D. Tetrahedron 1966, 22, 279−304. (2) Diels, O.; Friedrichsen, W. Liebigs Ann. Chem. 1934, 513, 145− 155. (3) Pihko, A. J.; Koskinen, A. M. P. Tetrahedron 2005, 61, 8769− 8807. (4) Gianatassio, R.; Lopchuk, J. M.; Wang, J.; Pan, C. M.; Malins, L. R.; Prieto, L.; Brandt, T. A.; Collins, M. R.; Gallego, G. M.; Sach, N. W.; Spangler, J. E.; Zhu, H.; Zhu, J.; Baran, P. S. Science 2016, 351, 241−246. (5) Torres-Gómez, H.; Lehmkuhl, K.; Schepmann, D.; Wünsch, B. Eur. J. Med. Chem. 2013, 70, 78−87. (6) (a) Dilmac, A. M.; Spuling, E.; de Meijere, A.; Bräse, S. Angew. Chem., Int. Ed. 2016, DOI: 10.1002/anie.201603951. (b) Nakajima, R.; Yamamoto, N.; Hirayama, S.; Iwai, T.; Saitoh, A.; Nagumo, Y.; Fujii, H.; Nagase, H. Bioorg. Med. Chem. 2015, 23, 6271−6279. (7) Zeng, X. P.; Cao, Z. Y.; Wang, Y. H.; Zhou, F.; Zhou, J. Chem. Rev. 2016, 116, 7330−7396. (8) Kotha, S.; Aswar, V. R. Org. Lett. 2016, 18, 1808−1811. (9) Ramachary, D. B.; Kishor, M. J. Org. Chem. 2007, 72, 5056− 5068. (10) (a) Lacoste, E.; Vaique, E.; Berlande, M.; Pianet, I.; Vincent, J.M.; Landais, Y. Eur. J. Org. Chem. 2007, 2007, 167−177. (b) Prakash, C.; Mohanakrishnan, A. K. Eur. J. Org. Chem. 2008, 2008, 1535− 1543. (c) Fagnoni, M.; Schmoldt, P.; Kirschberg, T.; Mattay, J. Tetrahedron 1998, 54, 6427−6444. (11) Xie, L.; Wu, Y.; Yi, W.; Zhu, L.; Xiang, J.; He, W. J. Org. Chem. 2013, 78, 9190−9195. (12) (a) Xu, C.; Zhang, L.; Zhou, P.; Luo, S.; Cheng, J.-P. Synthesis 2013, 45, 1939−1945. (b) Zhou, P.; Zhang, L.; Luo, S.; Cheng, J.-P. J. Org. Chem. 2012, 77, 2526−2530. (13) (a) Eagan, J. M.; Hori, M.; Wu, J.; Kanyiva, K. S.; Snyder, S. A. Angew. Chem., Int. Ed. 2015, 54, 7842−7846; Angew. Chem. 2015, 127, 7953−7957. (14) Werner, T.; Hoffmann, M.; Deshmukh, S. Eur. J. Org. Chem. 2014, 2014, 6630−6633. (15) Lipshutz, B. H.; Parker, D. A.; Nguyen, S. L.; McCarthy, K. E.; Barton, J. C.; Whitney, S. E.; Kotsuki, H. Tetrahedron 1986, 42, 2873−2879. (16) (a) Gauthier, D.; Lindhardt, A. T.; Olsen, E. P. K.; Overgaard, J.; Skrydstrup, T. J. Am. Chem. Soc. 2010, 132, 7998−8009. (b) Larionov, E.; Li, H.; Mazet, C. Chem. Commun. 2014, 50, 9816−9826. (17) De Bo, G.; Markó, I. E. Eur. J. Org. Chem. 2011, 2011, 1859− 1869. (18) (a) Terada, Y.; Arisawa, M.; Nishida, A. Angew. Chem., Int. Ed. 2004, 43, 4063−4067; Angew. Chem. 2004, 116, 4155−4159. (b) Yamamoto, Y. Chem. Rev. 2012, 112, 4736−4769. (19) (a) Schelwies, M.; Farwick, A.; Rominger, F.; Helmchen, G. J. Org. Chem. 2010, 75, 7917−7919. (b) Fürstner, A.; Szillat, H.; Gabor, B.; Mynott, R. J. Am. Chem. Soc. 1998, 120, 8305−8314. (20) Krief, A.; Surleraux, D. Synlett 1991, 1991, 273−275.
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Christian Merten: 0000-0002-4106-1905 Mathias Christmann: 0000-0001-9313-2392 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank Umicore for the generous donation of metathesis catalysts. V.M.S. thanks the Beilstein Institut and the Dahlem Research School for Ph.D. scholarships. We thank Luise Schefzig, Isabelle Heing genannt Becker, and Caitlin Puro (FU 2313
DOI: 10.1021/acs.orglett.7b00836 Org. Lett. 2017, 19, 2310−2313
