Enantioselective Formal [3 + 1 + 1] Cycloaddition Reaction by Ru(II

Mar 1, 2017 - A further increased reaction temperature (45 °C) cannot further improve the ... In turn, decreasing the reaction temperature to 0 °C n...
0 downloads 0 Views 1MB Size
Letter pubs.acs.org/OrgLett

Enantioselective Formal [3 + 1 + 1] Cycloaddition Reaction by Ru(II)/Iminium Cocatalysis for Construction of Multisubstituted Pyrrolidines Mingfeng Li, Rui Chu, Jianghui Chen, Xiang Wu, Yun Zhao, Shunying Liu,* and Wenhao Hu* Shanghai Engineering Research Center of Molecular Therapeutics and New Drug Development, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200062, China S Supporting Information *

ABSTRACT: A Ru(II)/iminium cocatalyzed asymmetric formal [3 + 1 + 1] cycloaddition reaction of diazoacetophenones, anilines, and enals is disclosed to construct multisubstituted pyrrolidines in one step with excellent diastereoselectivity and enantioselectivity. The reaction mechanism was postulated as a successful trapping of Ru(II)-associated ammonium ylides via a selective 1,4-addition to chiral amine activated enals followed by a tandem aza-aldol process. The control experiments and theoretical density functional theory investigation revealed that the reversible NaOAc-facilitated aza-aldol process led to the diastereomeric conversion to provide a stable product.

P

diastereo-, and enantioselectivity.3 Even though several examples of other assembly sequences by [4 + 1],4 [2 + 2 + 1],5 and [3 + 1 + 1]6 models were also, respectively, illustrated with a good diastereoselectivity to construct pyrrolidine frameworks, their asymmetric versions were not yet established. To the best of our knowledge, the only two diastereoselective examples via formal [3 + 1 + 1] cycloaddition model were established by our group,6 which was realized by a three-component tandem reaction of diazo compounds, anilines, and β,γ-unsaturated α-ketoesters (Scheme 2a). Mainly owing to the high activity of the ketoester

yrrolidines are unique and important core structures in many natural alkaloids, bioactive compounds, and pharmaceuticals.1 They are also widely used as building blocks or intermediates in synthetic chemistry to construct heterocyclic compounds.2 The highly selective synthesis of chiral pyrrolidines is increasingly attractive for synthetic organic and pharmaceutical chemists. Theoretically, there are four classes of assembly sequences to construct an N-containing five-membered ring of pyrrolidine, including [3 + 2], [4 + 1], [2 + 2 + 1], and [3 + 1 + 1] cycloaddition models (Scheme 1). For each model to construct a pyrrolidine framework, it is highly challenging and important to address an efficient enantioselective control of the desired process. Among the developed frameworks, the [3 + 2] cycloaddition reaction of azomethine ylides with olefins represents an outstanding tool and common method with high regio-,

Scheme 2. Formal [3 + 1 + 1] Model To Construct Pyrrolidine Framework

Scheme 1. Schematic Representation of Methodology Models for Construction of Pyrrolidine Framework

Received: January 10, 2017

© XXXX American Chemical Society

A

DOI: 10.1021/acs.orglett.7b00055 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters Table 1. Reaction Condition Optimizationa

group, the racemic background reaction is too strong to inhibit, and it is very difficult to achieve the enantioselective control. To address this long-lasting enantioselective control problem and achieve an alternative approach for the construction of chiral pyrrolidine frameworks, we envisioned changing a highly active ketoester group to a less active aldehyde group and then activating the aldehyde with a chiral iminium (Scheme 2b). The metal complex/iminium trapping ammonium ylide process with enals is readily followed by an intramolecular aza-aldol ringclosure reaction. It will give an asymmetric formal [3 + 1 + 1] cycloaddition model to rapidly construct chiral multisubstituted pyrrolidines from simple starting materials (Scheme 2b). The combination of amine catalysis with transition-metal cooperative catalysis7 is an emerging and exciting field aiming to achieve organic transformations that cannot be accomplished by individual catalysis. Compatible transition metals are commonly used to combine with enamines; however, they are rarely used with iminiums.8 We recently reported an iridium(I)/chiral iminium alyzed three-component reaction of diazoacetates, indoles, and enals to give functionalized indole derivatives in good yields and with an excellent enantioselectivity and a moderate diastereoselectivity.9 Based on the success of our previous studies for developing asymmetric multicomponent reactions (AMCRs) for the rapid construction of complex molecules,9,10 we decided to choose diarylprolinol silyl ethers as the chiral amine catalysts to activate the aldehyde group in enals to trap ammonium ylides. The initial exploration was carried out by reacting cinnamyl aldehyde 1a, 4-bromoaniline 2a, and diazoacetophenone 3a in CH2Cl2 at room temperature. The commonly used transition metal catalysts for the decomposition of diazo compounds, [Ir(COD)Cl]2, Rh2(OAc)4, [PdCl(η3-C3H5)]2,11 and [Ru(pcymene)Cl2]2 (5 mol %) were respectively employed with (S)-2(diphenyl((trimethylsilyl)oxy)methyl)pyrrolidine (S)-4a (20 mol %) as cocatalyst and p-nitrobenzoic acid (p-NBZA, 40 mol %) as additive (Table 1, entries 1−4). All these metal complexes gave the asymmetric formal [3 + 1 + 1] cycloaddition product 5a in good HPLC yields of >80% and with an excellent enantioselectivity, but unfortunately, very low isolated yields (13−22%) and a poor diastereoselectivity were obtained (entries 1−4). The starting materials 1a and 3a were completely consumed within 0.5 h along with N−H insertion product 6 derived from the reaction of 2a and 3a as the main side product. Most products were decomposed during the purification procedure by flash chromatography on silica gel. This indicates that the product is unstable in the acidic environment. Neutralization with saturated sodium bicarbonate aqueous solution had no improvement in the isolated yield even after the reaction was immediately completed. Consideration on the notably higher conversion rate and the economic cost, [Ru(pcymene)Cl2]2 was employed for the next exploration. Intriguingly, prolonging the reaction time from 0.5 to 48 h notably improved the diastereoselectivity from 50:50 to 88:12 but led to a lower yield of 12% (entry 5). The improved dr value indicates that the major isomer is possibly the product of thermodynamic control. To improve the stability of the product in the reaction and the purification procedure, various additives were investigated. Both benzoic and acetic acid were identified as an effective additive, evidenced by a similar diastereoselectivity, however, still in a poor isolated yield (Table 1, entries 6 and 7). Encouragingly, inorganic base sodium acetate, usually matched with chiral diarylprolinol silyl ethers12 to activate enal by iminium catalysis, gave a more

a

General reaction conditions: M/(S)-4a/p-NBZA/1a/2a/3a = 0.05:0.2:0.4:1:1.5:1.5. bIsolated yield of 5a. cDetermined by LC−MS analysis of the crude mixture. dDetermined by chiral HPLC analysis. e The amount of additive is 20 mol %.

clear reaction (entry 8). Good similar isolated yields were obtained in the reaction time of both 0.5 and 48 h (entry 8 vs 9). A notable improvement in diastereoselectivity was also observed when extended the reaction time from 0.5 to 48 h. Potassium acetate gave a relative lower yield and enantioselectivity (entry 10). The similar additive phenomenon was also observed by Wang et al.,12a but the mechanism is not yet clear. Amine (S)-4b with larger steric tert-butyldimethylsilyl (TBS) instead of (S)-4a showed an improvement in the reaction enantioselectivity (97% ee) with the same high diastereoselectivity (entry 11). MacMillan’s catalyst 4c, as a kind of versatile catalyst for activating enal by imnium fromation,13 deteriorated the dr value to 67:33 and gave no enantioselectivity (entry 12). Screening of solvents had no obvious improvement in the yield and the stereoselectivity (see Table S1). On the hypothesis of thermodynamic control of products, the reaction temperature should have a profound effect on the reaction. As expected, increasing the reaction temperature to 35 °C further improved the yield to 65%, diastereoselectivity to >95:5, and enantioselectivity to 98% in a shorter reaction time of 24 h (entry 13). A further increased reaction temperature (45 °C) cannot further improve the yield (entry 14). In turn, decreasing the reaction temperature to 0 °C nearly gave no diastereoselectivity (55:45 dr) and a lower yield of 30% (entry 15). These results demonstrate that the minor isomer of 5a is a kinetic control of product and converses into a thermodynamic control of product at last. The isolated minor isomer of 5a was monitored by HPLC under the standard conditions giving totally pure major isomer of 5a after 24 h (see Figure S1), which further confirms the diastereomeric conversion in this reaction which resulting in the high diastereoselectivity. Therefore, the optimal reaction conditions were established to give 5a in 65% yield and with high dr (>95:5) and excellent ee value (98%, entry 13). Ruthenium complexes have been well established as efficient catalysts for various transformations,14 also as effective catalyst for the decomposition of diazo compounds,15 while only a few B

DOI: 10.1021/acs.orglett.7b00055 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters

cocatalysis of the ruthenium complex and the chiral iminium via a possible favored transition state IIIa (Scheme 3a).

catalytic systems combining ruthenium complex and chiral amine catalyst have been reported.16 As far as we know, this is the first example about the combination of ruthenium complex and chiral amine catalyst by iminium activation to achieve enantioselective control. The screening of substrate scope shows that the reaction has a good tolerance to broad substrates in most cases, and the results are summarized in Table 2. In general, the reaction tolerates

Scheme 3. Proposed Paths for Diastereomeric Conversion

Table 2. Substrate Scope of the Reactiona

To get insight into the conversion mechanism of minor 5a to major 5a, a DFT calculation investigation and control experiment were conducted. The model based on DFT calculations for the optimized intermediates using m06/6-31g* gave a difference of 6.47 kcal/mol in favor of the major product of 5a (see Figure S3), which supports the major isomer of 5a is a thermodynamic product. For the conversion process, both the base-facilitated C− N and the acid-facilitated C−O bond cleavages in minor 5a will generate major 5a (Scheme 3b). A signal of ion fragment D is notably observed in the HR-MS spectra of 5a (see Figure S4), which indicates that product 5 is readily decomposed via active intermediate C and D (path b) to result in many side products. It might be responsible for the instability of the products and the improvement in the diastereoselectivity when prolonging the reaction time in an acidic environment (vide supra, Table 1, entries 1−7). In contrast, with sodium acetate as additive, the reversible base-facilitated aza-aldol pathway converses minor into major isomer of 5a via intermediates A and B (path a). It is possible that the base-catalyzed intramolecular addition to intermediate B proceeds more easily than the acid-catalyzed intermolecular addition to intermediate D. To further explore the reaction mechanism, several control experiments were conducted. A reaction of the N−H insertion product 6 and cinnamaldehyde under the standard conditions gave a desired product 5a but in a low yield of 12% and with >95:5 and 97% ee. This indicates that there is a competitive stepwise tandem process (see Scheme S1) vs a dominated threecomponent tandem process in this reaction. When p-NBZA was used instead of sodium acetate, no 5a was observed (see Scheme S2a). These results confirmed the mechanism of the basefacilitated diastereomeric conversion process. In the absence of the addictive (S)-4b or NaOAc (see Scheme S2b), the reaction gave only the N−H insertion side product together with total recovery of cinnamaldehyde 1a. LC-MS spectra of the reaction mixture solution also confirmed the generation of iminium cation II from 1 and (S)-4b (see Figure S5). On the basis of the above-mentioned investigations, a possible mechanism was proposed and outlined in Scheme 4. The ammonium ylide intermediate Ia or enolate Ib is generated in situ from the ruthenium-catalyzed diazo decomposition of 2 and 3. For the sodium acetate promoted iminium catalytic cycle, cinnamyl aldehyde 1 is activated by (S)-4b to form an iminium ion II, which effectively traps the ammonium ylide intermediate Ia or Ib to afford an enamine intermediate III. Hydrolysis of III yields intermediate IV and regenerates chiral amine 4b followed a reversible aza-aldol ring-closure reaction leading to the formation of the asymmetric formal [3 + 1 + 1] cycloaddition product 5.

a General reaction conditions: [Ru(p-cymene)Cl2]2/(S)-4b/NaOAc/ 1a/2a/3a = 0.05:0.2:0.2:1:1.5:1.5. b−dReaction conditions as performed in Table 1. NP: no product.

various substitutions on the aromatic or heteroaryl rings of enals 1 (entries 1−6), anilines 2 (entries 7−11), and diazo compounds 3 (entries 12−16). In most cases, moderate yields (up to 65%) were obtained with high dr ratio (up to >95:5) and excellent enantioselectivity (up to 98% ee). It was found that the diastereoselectiviy was highly depended on the diazo compounds (entry 16). The absolute configuration of the major isomer of 5a was determined as (2R,4S,5R) by single crystal X-ray crystallography, and that of the minor isomer of 5 was designated as (2S,4S,5R) by comparing its 2D NOESY NMR spectra with that of major 5c (Figure S2). Those of other products were then assigned by analogy. No diastereometric isomers at C-4 or C-5 were observed, indicating the reaction generates the two stereocenters with a very high diastereoselectivity by the C

DOI: 10.1021/acs.orglett.7b00055 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters

Scheidt, K. A. Angew. Chem., Int. Ed. 2007, 46, 8748. (c) Pearson, W. H. In Studies in Natural Product Chemistry; Atta-Ur-Rahman, Ed.; Elsevier: New York, 1998; Vol. 1, p 323. (d) Herzon, S. B.; Myers, A. G. J. Am. Chem. Soc. 2005, 127, 5342. (3) (a) Padwa, A.; Weingarten, M. D. Chem. Rev. 1996, 96, 223. (b) Hodgson, D. M.; Pierard, F. Y. T. M.; Stupple, P. A. Chem. Soc. Rev. 2001, 30, 50. (c) Yang, W.-L.; Tang, F.-F.; He, F.-S.; Li, C.-Y.; Yu, X.-X.; Deng, W.-P. Org. Lett. 2015, 17, 4822. (d) Arasasppan, H.; Thangamuthu, M. D. J. Heterocyclic Chem. 2016, 53, 313. (e) Sato, T.; Miyazaki, T.; Arai, T. J. Org. Chem. 2015, 80, 10346. (f) Liu, T.-L.; Li, Q.H.; He, Z.-L.; Zhang, J.-W.; Wang, C.-J. Chin. J. Catal. 2015, 36, 68. (4) (a) Rigby, J. H.; Qabar, M. J. Am. Chem. Soc. 1991, 113, 8975. (b) Rigby, J. H.; Laurent, S. J. Org. Chem. 1999, 64, 1766. (5) (a) Hong, D.; Zhu, Y.; Li, Y.; Lin, X.; Lu, P.; Wang, Y. Org. Lett. 2011, 13, 4668. (b) Göbel, A.; Imhof, W. Chem. Commun. 2001, 593. (6) (a) Zhu, Y.-G.; Zhai, C.-W.; Yue, Y.-L.; Yang, L.-P.; Hu, W.-H. Chem. Commun. 2009, 11, 1362. (b) Zhang, X.; Ji, J.-J.; Zhu, Y.-G.; Jing, C.-C.; Li, M.; Hu, W.-H. Org. Biomol. Chem. 2012, 10, 2133. (7) (a) Bertelsen, S.; Marigo, M.; Brandes, S.; Diner, P.; Jørgensen, K. A. J. Am. Chem. Soc. 2006, 128, 12973. (b) Deng, Y.; Kumar, S.; Wang, H. Chem. Commun. 2014, 50, 4272. (c) Inamdar, S. M.; Shinde, V. S.; Patil, N. T. Org. Biomol. Chem. 2015, 13, 8116. (8) (a) Ibrahem, I.; Breistein, P.; Córdova, A. Angew. Chem., Int. Ed. 2011, 50, 12036. (b) Ibrahem, I.; Ma, G.; Afewerki, S.; Córdova, A. Angew. Chem., Int. Ed. 2013, 52, 878. (c) Meazza, M.; Ceban, V.; Pitak, M. B.; Coles, S. J.; Rios, R. Chem. - Eur. J. 2014, 20, 16853. (d) Wei, Y.; Yoshikai, N. J. Am. Chem. Soc. 2013, 135, 3756. (e) Rueping, M.; Bui, L.; Dufour, J. ACS Catal. 2014, 4, 1021. (9) Li, M.-F.; Guo, X.; Jin, W.-F.; Zheng, Q.; Liu, S.-Y.; Hu, W.-H. Chem. Commun. 2016, 52, 2736. (10) (a) Zhang, D.; Zhou, J.; Xia, F.; Kang, Z.-H.; Hu, W.-H. Nat. Commun. 2015, 6, 5801. (b) Ma, X.-C.; Jiang, J.; Lv, S.-Y.; Yao, W.-F.; Yang, Y.; Liu, S.-Y.; Xia, F.; Hu, W.-H. Angew. Chem., Int. Ed. 2014, 53, 13136. (c) Jing, C.-C.; Xing, D.; Qian, Y.; Shi, T.-D.; Zhao, Y.; Hu, W.-H. Angew. Chem., Int. Ed. 2013, 52, 9289. (d) Qiu, L.; Guo, X.; Ma, C.-Q.; Qiu, H.; Liu, S.-Y.; Yang, L.-P.; Hu, W.-H. Chem. Commun. 2014, 50, 2196. (e) Tang, M.; Wu, Y.; Liu, Y.; Cai, M.-Q.; Xia, F.; Liu, S.-Y.; Hu, W.-H. Huaxue Xuebao 2016, 74, 54. (11) (a) Greenman, K. L.; Carter, D. S.; Van Vranken, D. L. Tetrahedron 2001, 57, 5219. (b) Chen, Z.-S.; Duan, X.-H.; Zhou, P.-X.; Ali, S.; Luo, J.-Y.; Liang, Y.-M. Angew. Chem., Int. Ed. 2012, 51, 1370. (c) Zhang, Z.-H.; Liu, Y.-Y.; Gong, M.-X.; Zhao, X.-K.; Zhang, Y.; Wang, J.-B. Angew. Chem., Int. Ed. 2010, 49, 1139. (d) Xiao, Q.; Zhang, Y.; Wang, J.-B. Acc. Chem. Res. 2013, 46, 236. (12) (a) Li, H.; Wang, J.; Xie, H.-X.; Zu, L.-S.; Jiang, W.; Duesler, E. N.; Wang, W. Org. Lett. 2007, 9, 965. (b) Brandau, S.; Maerten, E.; Jørgensen, K. A. J. Am. Chem. Soc. 2006, 128, 14986. (c) Marigo, M.; Bertelsen, S.; Landa, A.; Jørgensen, K. A. J. Am. Chem. Soc. 2006, 128, 5475. (13) Erkkila, A.; Majander, I.; Pihko, P. M. Chem. Rev. 2007, 107, 5416. (14) (a) Maas, G. Chem. Soc. Rev. 2004, 33, 183. (b) Trost, B. M.; Toste, F. D.; Pinkerton, A. B. Chem. Rev. 2001, 101, 2067. (c) Zhang, M.; Fang, X.-J.; Neumann, H.; Beller, M. J. Am. Chem. Soc. 2013, 135, 11384. (d) Inamoto, K.; Kadokawa, J.; Kondo, Y. Org. Lett. 2013, 15, 3962. (15) (a) Galardon, E.; Le Maux, P.; Simonneaux, G. J. Chem. Soc., Perkin Trans. 1 1997, 2455. (b) Galardon, E.; Le Maux, P.; Simonneaux, G. Tetrahedron 2000, 56, 615. (c) Deng, Q.-H.; Xu, H.-W.; Yuen, A. W.H.; Xu, Z.-J.; Che, C.-M. Org. Lett. 2008, 10, 1529. (16) (a) Ikeda, M.; Miyake, Y.; Nishibayashi, Y. Angew. Chem., Int. Ed. 2010, 49, 7289. (b) Nicewicz, D. A.; MacMillan, D. W. C. Science 2008, 322, 77.

Scheme 4. Proposed Reaction Mechanism

Sodium acetate accelerates the reversible aza-aldol reaction to promote the kinetic minor isomer to convert into thermodynamic isomer. Possibly due to the hydrogen bond formation in IV, the reversible aza-aldol process gave the highly diastereoselective products. In conclusion, an asymmetric extraordinary formal [3 + 1 + 1] cycloaddition reaction catalyzed by the combination of ruthenium and iminium catalysis was successfully developed from simple starting materials of diazoacetophenones, anilines, and enals. Ruthenium(II)-associated ammonium ylide intermediates were successfully trapped by the chiral secondary amine activated enals via selective 1,4-addtition following a tandem base-facilitated aza-aldol process. Sodium acetate was demonstrated to play a crucial role both in the iminium catalysis and in the aza-aldol process. The established reaction provides an alternative approach of asymmetric sequence assembly model to synthesize multisubstituted pyrrolidines with excellent diastereoand enantioselectivity.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b00055. Experimental procedures and full spectroscopic data for all new compounds (PDF) X-ray data for 5a (CIF)



AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Wenhao Hu: 0000-0002-1461-3671 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank the National Science Foundation of China (Nos. 21332003 and 21672066) and the Science and Technology Commission of Shanghai Municipality (No. 15ZR1411000) for financial support.



REFERENCES

(1) (a) Michael, J. P. Nat. Prod. Rep. 2008, 25, 139. (b) Vitaku, E. D.; Smith, T.; Njardarson, J. T. J. Med. Chem. 2014, 57, 10257. (c) Hensler, M. E.; Bernstein, G.; Nizet, V.; Nefzi, A. Bioorg. Med. Chem. Lett. 2006, 16, 5073. (2) (a) Donohoe, T. J.; Bataille, C. J. R.; Churchill, G. W. Annu. Rep. Prog. Chem., Sect. B: Org. Chem. 2006, 102, 98. (b) Galliford, C. V.; D

DOI: 10.1021/acs.orglett.7b00055 Org. Lett. XXXX, XXX, XXX−XXX