Computational Investigation of the Aza-Cope Rearrangement Leading to Angularly Substituted 1-Azabicyclic Rings

Abstract

A computational study of the aza-Cope rearrangement leading to angularly substituted 1-azabicyclic ring systems is presented. The calculations estimate the probability of the proton transfer between reaction intermediates and protic solvents, explain the experimentally observed reaction selectivity, and suggest new potentially more efficient systems for further in vitro and in silico investigations.

Angularly substituted 1-azabicyclic ring systems such as octahydroindole derivatives are essential scaffolds presented in a plethora of natural products and biologically active compounds. Some examples are shown in Figure [1]: strychnine,[1] tuboxenine,[2] and mesemembrine.[3] Despite the apparent interest in compounds containing the aforementioned structural motif only scarce amount of synthetic approaches have been reported.[4] [5] [6]

In 2005 Aron and Overman reported the synthesis of angularly substituted 1-azabicyclic ring systems via aza-Cope rearrangements.[7] The same group has developed a dynamic kinetic resolution process in further studies devoted to synthesis of angularly substituted 1-azabicycles.[4] [8] Scheme [1] depicts the mechanism proposed by Overman and co-workers on the example of the octahydroindole 7 synthesis. The aminoketal 6 in the presence of trifluoroacetic acid (TFA) leads to the formation of the iminium ion intermediate 4. The intermediate 4 undergoes aza-Cope sigmatropic rearrangement leading to the intermediate 5. The isotopic labeling experiments of Overman and co-workers led to the formation of the product d₃-7, suggesting the equilibrium between the intermediates 1–4 essential for the observed dynamic kinetic resolution.

Despite the synthetical attractiveness of this transformation,[9] [10] [11] no theoretical studies of the reaction mechanism are presented in the literature to the best of our knowledge. We believe that a deeper understanding of the reaction mechanism based on theory can be highly beneficial for further experimental investigations. Thus, we have decided to perform quantum chemical calculations at the density functional level of theory (DFT: B3LYP-D3BJ/def2-TZVP//B3LYP-D3BJ/def2-SVP).[12–24]

In the first stage of our computational study, we sought to evaluate how reversible the proton transfer events responsible for the equilibrium between the intermediates 1–4 are. Our initial hypothesis was that protic solvents assist at the proton transfer step. Thus, one molecule of solvent (water) was taken explicitly. The starting point (the reference 0.0 kcal/mol) for these calculations was the intermediate 4 (Scheme [2]). Three pathways are compared: 4 → 1, 4 → 2, and 4 → 3 in accordance with the proton transfers suggested by isotopic labeling experiments from the Overman group. The computed respective transition states TS_4-1 , TS_4-2 , and TS_4-3 involving one molecule of water explicitly are depicted in Scheme [2].

The computed Gibbs free energy barriers of these three steps strictly deny the feasibility of these events even at the elevated (120 °C in accordance with the experimental conditions) temperature: ΔG ^‡ > 40 kcal/mol. Moreover, all three proton transfer events are computed to be endergonic: ΔG > 10 kcal/mol. One can draw two conclusions: 1) the aza-Cope-rearrangement-ready species 4 is thermodynamically the most stable isomer facilitating the reaction; 2) the proton transfer events are not promoted just by the protic solvent and thus, most probably are mediated by the acid present in the mixture (CF₃COOH).

The next stage of the computational study was to investigate kinetic and thermodynamic properties of the central part of the reaction mechanism, the aza-Cope sigmatropic rearrangement event 4 → 5. Scheme [3]A presents the computed Gibbs free energy barriers corresponding to two alternative transition state structures: boatlike transition state TS_boat_4-5 (on the top) and chairlike TS_chair_4-5 (on the bottom). The DFT optimized structures of both transition states (the most stable conformations) are shown in Scheme [3]B. The calculations suggest that reaction preferably proceeds via TS_chair_4-5, ΔG ^‡(via TS_chair_4-5 ) = 24.5 kcal/mol, while the TS_boat_4-5 is computed to be 8.2 kcal/mol less stabilized. The computed endergonicity of reaction 4 → 5 is in good agreement with the experimentally confirmed reaction reversibility.[7]

The synthesis of the azabicyclic compounds described by Overman and co-workers selectively led to the cis product 7 (Scheme [1]). However, the possibility of obtaining trans products would be of clear synthetic interest due to the presence of this respective scaffold in several natural products,[25] e.g., alkaloid tuboxenine (Figure [1]). Thus, as the next challenge of our computational study, we analyzed the energetic probability of trans product formation in the considered reaction. Scheme [4]A presents the computed activation ΔG ^‡ and the reaction ΔG Gibbs free energies for the trans pathway (left, leading to the intermediate trans-5) and the cis pathway (right, with the formation of the expected (based on the experimental evidence) intermediate cis-5) of the aza-Cope rearrangement 4 → 5. The cis pathway is computed to be both kinetically (ΔΔG ^‡ = ΔG ^‡(4 → trans-5) – ΔG ^‡(4 → cis-5) = 36.1 – 24.5 = 11.6 kcal/mol and thermodynamically (ΔΔG = ΔG(4 → trans-5) – ΔG(4 → cis-5) = 22.8 – 10.5 = 12.3 kcal/mol) more favorable.

These computed energetic properties agree with the experimental evidence of the reaction cis product selectivity. Having these results in hand, we have decided to analyze in silico whether we can favor the formation of the trans product replacing the bridgehead hydrogen in intermediate 4 (Scheme [4]B, right) to various substituents R with different electronic and steric properties (intermediates 4_R as shown in Scheme [4]B, left). Figure [2] depicts the computed Gibbs free energy profiles for the trans (left) and cis pathways (right) varying the substituent R. The ‘palette’ of tested substituents R is depicted in the upper right corner of Figure [2]. As one can see, there is no favorable to the reaction (i.e., energy decreasing) influence of the substituent R on either thermodynamic or kinetic probability of the trans pathway. Indeed, the case of R = H (unsubstituted system, green color) shows the lowest barrier of 36.1 kcal/mol and the 22.8 kcal/mol endergonic reaction, while all other tested substituents R led to even higher barriers and more endergonic reactions. The situation is, however, different for the cis pathway (Figure [2], right), where the substituent R can both increase and decrease the kinetic barrier as well as the thermodynamic price. The highest barrier and the most endergonic reaction are computed for the R = tBu for both trans and cis pathways. This influence of the tBu substituent that is strongly unfavorable for the reaction (energy increasing) is most probably related to the steric factors. Our calculations predict that methoxy substituent (R = OMe) would lead to the cis product in the most efficient way. Indeed, in the case of R = OMe, the computed Gibbs free energy barrier is ΔG ^‡(4-OMe → cis-5-OMe) = 23.1 kcal/mol, which is 1.3 kcal/mol lower as compared to the R = H reaction, and the reaction Gibbs free energy ΔG(4-OMe → cis-5-OMe) = 9.4 kcal/mol, i.e., 1.1 kcal/mol less endergonic reaction than R = H counterpart. Thus, based on the computed energies the methoxy-substituted system R = OMe (and also R = COOH, COOMe as shown in Figure [2]) should be more efficient as compared to the unsubstituted case of R = H, and we hope that our results will encourage further experimental applications.

In conclusion, we have performed a quantum chemical investigation of the reaction leading to angularly substituted 1-azabicyclic systems, namely octahydroindole derivatives, via aza-Cope rearrangements experimentally explored by Overman and co-workers. Taken together, our results rationalize the experimental observations and suggest new, potentially more efficient reactions for subsequent experimental studies.

(1) Our calculations involving one explicit molecule of protic solvent (water) deny the possibility of proton transfer events. Thus, the mechanism of the experimentally observed proton transfers must include additional mediators, and trifluoracetic acid seems to be the most plausible. (2) In agreement with the experimental evidence of the reaction reversibility the aza-Cope rearrangement in the core of the reaction mechanism is endergonic. The chairlike transition state is energetically preferred as compared to the boatlike counterpart. (3) The experimentally observed cis selectivity is caused by both kinetic and thermodynamic unfavourability of the respective alternative trans pathway. Our in silico reaction-scope modeling with the varied bridgehead substituent deny the possibility of trans reaction for the explored systems. At the same time, it predicts higher efficiency for the methoxy-substituted cis product formation compared to the unsubstituted counterpart. To the best of our knowledge this methoxy-substituted system was not reported.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

Calculations were partially performed at the Vienna Scientific Cluster (VSC). We thank Prof. Leticia González and Prof. Nuno Maulide for generous support, fruitful discussions, and resources.

• References and Notes

• 1 Knight SD, Overman LE, Pairaudeau G. J. Am. Chem. Soc. 1995; 117: 5776
  CrossrefSearch in Google Scholar
• 2 Ziegler FE, Kloek JA, Zoretic PA. J. Am. Chem. Soc. 1969; 91: 4943
  CrossrefSearch in Google Scholar
• 3 Smith MT, Crouch NR, Gericke N, Hirst M. J. Ethnopharmacol 1996; 50: 119
  CrossrefPubMedSearch in Google Scholar
• 4 Ito T, Overman LE, Wang J. J. Am. Chem. Soc. 2010; 132: 3272
  CrossrefPubMedSearch in Google Scholar
• 5 Reggelin M, Junker B, Heinrich T, Slavik S, Bühle P. J. Am. Chem. Soc. 2006; 128: 4023
  CrossrefPubMedSearch in Google Scholar
• 6 Shi S, Szostak M. Org. Lett. 2015; 17: 5144
  CrossrefPubMedSearch in Google Scholar
• 7 Aron ZD, Overman LE. Org. Lett. 2005; 7: 913
  CrossrefPubMedSearch in Google Scholar
• 8 Aron ZD, Ito T, May TL, Overman LE, Wang J. J. Org. Chem. 2013; 78: 9929
  CrossrefPubMedSearch in Google Scholar
• 9 Liu W.-B, Okamoto N, Alexy EJ, Tran K, Stoltz BM. J. Am. Chem. Soc. 2016; 138: 5234
  CrossrefPubMedSearch in Google Scholar
• 10 Liu J, Cao CG, Sun HB, Zhang X, Niu D. J. Am. Chem. Soc. 2016; 138: 13103
  CrossrefPubMedSearch in Google Scholar
• 11 Mou ZD, Zhang X, Niu D. Green Synth. Catal. 2021; 2: 70
  CrossrefSearch in Google Scholar
• 12 Computational DetailsThe conformational space of all molecules has been initially searched using meta-dynamics simulations based on tight-binding quantum chemical calculations as implemented in the software package conformer-rotamer ensemble sampling tool (CREST). The structures located with CREST have then been subjected to a B3LYP-D3BJ/def2-SVP geometry optimization. The nature of all stationary points (minima and transition states) was verified through the computation of harmonic vibrational frequencies. The thermal corrections to the Gibbs free energies were combined with the single point energies calculated at the B3LYP-D3BJ/def2-TZVP level of theory to yield Gibbs free energies (‘G ₂₉₈’) at 298.15 K. The DFT calculations have been performed with the Gaussian 16 program package. The SMD model with parameters of methanol was applied to consider implicit solvation effects in both, the geometries, and energies. All energies are reported in kcal/mol. The energy profiles were constructed using the most stable conformation (the global minimum) of each intermediate and transition state. Free energies in solution have been corrected to a reference state of 1 mol/L at 298.15 K through the addition of RTln(24.46) = +7.925 kJ/mol to the gas phase (1 atm) free energies.
• 13 Becke A. J. Chem. Phys. 1993; 98: 5648
  CrossrefSearch in Google Scholar
• 14 Cancès E, Mennucci B, Tomasi J. J. Chem. Phys. 1997; 107: 3032
  CrossrefSearch in Google Scholar
• 15 Marenich AV, Cramer CJ, Truhlar DG. J. Phys. Chem. B 2009; 113: 6378
  Search in Google Scholar
• 16 Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani G, Barone V, Petersson GA, Nakatsuji H, Li X, Caricato M, Marenich AV, Bloino J, Janesko BG, Gomperts R, Mennucci B, Hratchian HP, Ortiz JV, Izmaylov AF, Sonnenberg JL, Williams-Young D, Ding F, Lipparini F, Egidi F, Goings J, Peng B, Petrone A, Henderson T, Ranasinghe D, Zakrzewski VG, Gao J, Rega N, Zheng G, Liang W, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Throssell K, Montgomery JA. Jr, Peralta JE, Ogliaro F, Bearpark MJ, Heyd JJ, Brothers EN, Kudin KN, Staroverov VN, Keith TA, Kobayashi R, Normand J, Raghavachari K, Rendell AP, Burant JC, Iyengar SS, Tomasi J, Cossi M, Millam JM, Klene M, Adamo C, Cammi R, Ochterski JW, Martin RL, Morokuma K, Farkas O, Foresman JB, Fox DJ. Gaussian 16, Revision C.01. Gaussian Inc; Wallingford CT: 2016
  Search in Google Scholar
• 17 Lee C, Yang W, Parr RG. Phys. Rev. B: Condens. Matter Mater. Phys. 1988; 37: 785
  CrossrefPubMedSearch in Google Scholar
• 18 Vosko SH, Wilk L, Nusair M. Can. J. Phys. 1980; 58: 1200
  CrossrefSearch in Google Scholar
• 19 Stephens PJ, Devlin FJ, Chabalowski CF, Frisch MJ. J. Phys. Chem. 1994; 98: 11623
  CrossrefSearch in Google Scholar
• 20 Grimme S, Antony J, Ehrlich S, Krieg H. J. Chem. Phys. 2010; 132: 154104
  Search in Google Scholar
• 21 Weigend F, Ahlrichs R. Phys. Chem. Chem. Phys. 2005; 7: 3297
  Search in Google Scholar
• 22 Grimme S, Ehrlich S, Goerigk L. J. Comput. Chem. 2011; 32: 1456
  CrossrefPubMedSearch in Google Scholar
• 23 Pracht P, Bohle F, Grimme S. Phys. Chem. Chem. Phys. 2020; 22: 7169
  CrossrefPubMedSearch in Google Scholar
• 24 Grimme S. J. Chem. Theory Comput. 2019; 15: 2847
  CrossrefPubMedSearch in Google Scholar
• 25 Brown HC, Salunkhe AM. Tetrahedron Lett. 1993; 34: 1265
  CrossrefSearch in Google Scholar

Corresponding Author

Publication History

Received: 30 July 2022

Accepted after revision: 18 September 2022

Accepted Manuscript online:
18 September 2022

Article published online:
28 October 2022

© 2022. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References and Notes

• 1 Knight SD, Overman LE, Pairaudeau G. J. Am. Chem. Soc. 1995; 117: 5776
  CrossrefSearch in Google Scholar
• 2 Ziegler FE, Kloek JA, Zoretic PA. J. Am. Chem. Soc. 1969; 91: 4943
  CrossrefSearch in Google Scholar
• 3 Smith MT, Crouch NR, Gericke N, Hirst M. J. Ethnopharmacol 1996; 50: 119
  CrossrefPubMedSearch in Google Scholar
• 4 Ito T, Overman LE, Wang J. J. Am. Chem. Soc. 2010; 132: 3272
  CrossrefPubMedSearch in Google Scholar
• 5 Reggelin M, Junker B, Heinrich T, Slavik S, Bühle P. J. Am. Chem. Soc. 2006; 128: 4023
  CrossrefPubMedSearch in Google Scholar
• 6 Shi S, Szostak M. Org. Lett. 2015; 17: 5144
  CrossrefPubMedSearch in Google Scholar
• 7 Aron ZD, Overman LE. Org. Lett. 2005; 7: 913
  CrossrefPubMedSearch in Google Scholar
• 8 Aron ZD, Ito T, May TL, Overman LE, Wang J. J. Org. Chem. 2013; 78: 9929
  CrossrefPubMedSearch in Google Scholar
• 9 Liu W.-B, Okamoto N, Alexy EJ, Tran K, Stoltz BM. J. Am. Chem. Soc. 2016; 138: 5234
  CrossrefPubMedSearch in Google Scholar
• 10 Liu J, Cao CG, Sun HB, Zhang X, Niu D. J. Am. Chem. Soc. 2016; 138: 13103
  CrossrefPubMedSearch in Google Scholar
• 11 Mou ZD, Zhang X, Niu D. Green Synth. Catal. 2021; 2: 70
  CrossrefSearch in Google Scholar
• 12 Computational DetailsThe conformational space of all molecules has been initially searched using meta-dynamics simulations based on tight-binding quantum chemical calculations as implemented in the software package conformer-rotamer ensemble sampling tool (CREST). The structures located with CREST have then been subjected to a B3LYP-D3BJ/def2-SVP geometry optimization. The nature of all stationary points (minima and transition states) was verified through the computation of harmonic vibrational frequencies. The thermal corrections to the Gibbs free energies were combined with the single point energies calculated at the B3LYP-D3BJ/def2-TZVP level of theory to yield Gibbs free energies (‘G ₂₉₈’) at 298.15 K. The DFT calculations have been performed with the Gaussian 16 program package. The SMD model with parameters of methanol was applied to consider implicit solvation effects in both, the geometries, and energies. All energies are reported in kcal/mol. The energy profiles were constructed using the most stable conformation (the global minimum) of each intermediate and transition state. Free energies in solution have been corrected to a reference state of 1 mol/L at 298.15 K through the addition of RTln(24.46) = +7.925 kJ/mol to the gas phase (1 atm) free energies.
• 13 Becke A. J. Chem. Phys. 1993; 98: 5648
  CrossrefSearch in Google Scholar
• 14 Cancès E, Mennucci B, Tomasi J. J. Chem. Phys. 1997; 107: 3032
  CrossrefSearch in Google Scholar
• 15 Marenich AV, Cramer CJ, Truhlar DG. J. Phys. Chem. B 2009; 113: 6378
  Search in Google Scholar
• 16 Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani G, Barone V, Petersson GA, Nakatsuji H, Li X, Caricato M, Marenich AV, Bloino J, Janesko BG, Gomperts R, Mennucci B, Hratchian HP, Ortiz JV, Izmaylov AF, Sonnenberg JL, Williams-Young D, Ding F, Lipparini F, Egidi F, Goings J, Peng B, Petrone A, Henderson T, Ranasinghe D, Zakrzewski VG, Gao J, Rega N, Zheng G, Liang W, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Throssell K, Montgomery JA. Jr, Peralta JE, Ogliaro F, Bearpark MJ, Heyd JJ, Brothers EN, Kudin KN, Staroverov VN, Keith TA, Kobayashi R, Normand J, Raghavachari K, Rendell AP, Burant JC, Iyengar SS, Tomasi J, Cossi M, Millam JM, Klene M, Adamo C, Cammi R, Ochterski JW, Martin RL, Morokuma K, Farkas O, Foresman JB, Fox DJ. Gaussian 16, Revision C.01. Gaussian Inc; Wallingford CT: 2016
  Search in Google Scholar
• 17 Lee C, Yang W, Parr RG. Phys. Rev. B: Condens. Matter Mater. Phys. 1988; 37: 785
  CrossrefPubMedSearch in Google Scholar
• 18 Vosko SH, Wilk L, Nusair M. Can. J. Phys. 1980; 58: 1200
  CrossrefSearch in Google Scholar
• 19 Stephens PJ, Devlin FJ, Chabalowski CF, Frisch MJ. J. Phys. Chem. 1994; 98: 11623
  CrossrefSearch in Google Scholar
• 20 Grimme S, Antony J, Ehrlich S, Krieg H. J. Chem. Phys. 2010; 132: 154104
  Search in Google Scholar
• 21 Weigend F, Ahlrichs R. Phys. Chem. Chem. Phys. 2005; 7: 3297
  Search in Google Scholar
• 22 Grimme S, Ehrlich S, Goerigk L. J. Comput. Chem. 2011; 32: 1456
  CrossrefPubMedSearch in Google Scholar
• 23 Pracht P, Bohle F, Grimme S. Phys. Chem. Chem. Phys. 2020; 22: 7169
  CrossrefPubMedSearch in Google Scholar
• 24 Grimme S. J. Chem. Theory Comput. 2019; 15: 2847
  CrossrefPubMedSearch in Google Scholar
• 25 Brown HC, Salunkhe AM. Tetrahedron Lett. 1993; 34: 1265
  CrossrefSearch in Google Scholar

• Supplementary Material