A highly diastereoselective and enantioselective synthesis of polysubstituted pyrrolidines via an organocatalytic dynamic kinetic resolution cascade

Article abstract of DOI:10.1021/ol4006129

Highly functionalized pyrrolidine and piperidine analogues, with up to three stereogenic centers, were synthesized in good yield (50-95%), excellent dr (single isomer), and high ee (>90%) using a Cinchona alkaloid-derived carbamate organocatalyst. High stereoselective synergy was achieved by combining a reversible aza-Henry reaction with a dynamic kinetic resolution (DKR)-driven aza-Michael cyclization. Whereas both reactions proceed with moderate enantioselectivities (50-60% for each step), high enantioselectivities are obtained for the overall products devoid of dr sacrifice.

Full text of DOI:10.1021/ol4006129

ORGANIC
LETTERS
2013
Vol. 15, No. 8
1958–1961
A Highly Diastereoselective and
Enantioselective Synthesis of
Polysubstituted Pyrrolidines via an
Organocatalytic Dynamic Kinetic
Resolution Cascade
Tao Cheng, Sixuan Meng, and Yong Huang*
Key Laboratory of Chemical Genomics, School of Chemical Biology and Biotechnology,
Shenzhen Graduate School of Peking University, Shenzhen, China
huangyong@pkusz.edu.cn
Received March 7, 2013
ABSTRACT
Highly functionalized pyrrolidine and piperidine analogues, with up to three stereogenic centers, were synthesized in good yield (50À95%), excellent
dr (single isomer), and high ee (>90%) using a Cinchona alkaloid-derived carbamate organocatalyst. High stereoselective synergy was achieved by
combining a reversible aza-Henry reaction with a dynamic kinetic resolution (DKR)-driven aza-Michael cyclization. Whereas both reactions proceed
with moderate enantioselectivities (50À60% for each step), high enantioselectivities are obtained for the overall products devoid of dr sacrifice.
Dynamic kinetic asymmetric transformations (DYKATs)
have now emerged as powerful tools for the construction of
stereogenic centers.1 Using this strategy, products of high
optical purity can be synthesized via reaction cascades that no
longer necessitate overwhelmingly selective steps. In addition,
ee enrichment of the overall products does not come at the
cost of sacrificing either yields or diastereoselectivities. Since
the discovery of organocatalytic cascade reactions,2 DYKAT
has enjoyed remarkable advances that have led to a number
of efficient organocatalytic domino processes3 and synergistic
catalysis cascades, a process in which two distinct catalytic
mechanisms are merged into a single reaction.2a In particular,
cascades that take advantage of a first reversible step and
a second dynamic kinetic resolution step (DKR) have led
to one-pot syntheses of chiral cyclic compounds with partic-
ularly high ee and dr in excellent yield. For example, Wang
reported a cascade featuring a reversible thio-Michael
and a DKR Michael reaction for his synthesis of substitued
thiochromans.4 Cordova, Jørgensen, Wang and others
ꢀ
(3) For recent examples of organocatalytic cascade reactions, see: (a)
Zhang, X.; Song, X.; Li, H.; Zhang, S.; Chen, X.; Yu, X.; Wang, W.
Angew. Chem., Int. Ed. 2012, 51, 7282. (b) Takizawa, S.; Nguyen, T. M.;
Grossmann, A.; Enders, D.; Sasai, H. Angew. Chem., Int. Ed. 2012, 51,
5423. (c) Rajkumar, S.; Shankland, K.; Brown, G. D.; Cobb, J. A. Chem.
Sci. 2012, 3, 584. (d) Maciver, E. E.; Knipe, P. C.; Cridland, A. P.;
Thompson, A. L.; Smith, M. D. Chem. Sci. 2012, 3, 537. (e) Albrecht, L.;
Ransborg, L. K.; Lauridsen, V.; Overgaard, M.; Zweifel, T.; Jørgensen,
K. A. Angew. Chem., Int. Ed. 2011, 50, 12496. (f) Celebi-Oelcuem, N.;
Lam, Y.-H.; Richmond, E.; Ling, K. B.; Smith, A. D.; Houk, K. N.
Angew. Chem., Int. Ed. 2011, 50, 11478. (g) Ozboya, K. E.; Rovis, T.
Chem. Sci. 2011, 2, 1835. (h) Cai, Q.; Zheng, C.; Zhang, J.-W.; You, S.-L.
Angew. Chem., Int. Ed. 2011, 50, 8665. (i) Tan, B.; Candeias, N. R.;
Barbas, C. F., III. Nat. Chem. 2011, 3, 473. (j) Zhang, S.-L.; Xie, H.-X.;
Zhu, J.; Li, H.; Zhang, X.-S. Nat. Commun. 2011, 2, 1214.
(1) (a) Parsons, A. T.; Johnson, J. S. Dynamic Kinetic Asymmetric
Transformations Involving CarbonÀCarbon Bond Cleavage, in Asym-
metric Synthesis II: More Methods and Applications; Christmann, M.,
€
Brase, S., Eds.; Wiley-VCH: Weinheim, Germany, 2012. (b) Steinreiber, J.;
Faber, K.; Griengl, H. Chem.;Eur. J. 2008, 14, 8060. (c) Trost, B. M.;
Fandrick, D. R. Aldrichimica Acta 2007, 40, 59.
(2) (a) Allen, A. E.; MacMillan, D. W. C. Chem. Sci. 2012, 3, 633. (b)
Patil, N. T.; Shinde, V. S.; Gajula, B. Org. Biomol. Chem. 2012, 10, 211.
(c) Pellissier, H. Adv. Synth. Catal. 2012, 354, 237. (d) Albrecht, L.;
Jiang, H.; Jørgensen, K. A. Angew. Chem., Int. Ed. 2011, 50, 8492. (e)
ꢀ
ꢀ
Ruiz, M.; Lopez-Alvarado, P.; Giorgi, G.; Menendez, J. C. Chem. Soc.
Rev. 2011, 40, 3445. (f) Grondal, C.; Jeanty, M.; Enders, D. Nat. Chem.
2010, 2, 167.
(4) Wang, J.; Xie, H.-X.; Li, H.; Zu, L.-S; Wang, W. Angew. Chem.,
Int. Ed. 2008, 47, 4177.
r
10.1021/ol4006129
Published on Web 04/08/2013
2013 American Chemical Society

disclosed a series of “organo-metal cooperative catalysis”
utilizing reversible Michael reactions and transition metal
catalyzed diastereoselective carbocyclization to access five-
membered carbo- and heterocycles.5 Very recently, Zhao
reported a chiral cyclohexane synthesis that combined a
reversible Henry reaction and a subsequent selective Michael
cyclization.6 In this report, we describe the highly diastereo-
and enantioselective synthesis of polysubstituted pyrrolidines
using a parallel DYKAT strategy. In this cascade, a reversible
aza-Henry reaction was combined with an aza-Michael
cyclization to yield N-containing heterocycles. Among recent
organocatalytic chiral syntheses of pyrrolidine analogues,
most reactions have relied on aldehydes as a key component.7
Processes free of aldehydes primarily reside in the Lewis acid
and the organometallic paradigm.8 Noteworthy, most stra-
tegies only permit access to pyrrolidines with specific sub-
stitution patterns, as restricted by their individual reaction
mechanisms. Our aza-Henry/aza-Michael cascade (Table 1)
offers a solution to the 2,3,5-trisubstituted pyrrolidine scaf-
fold. One advantage of this method is that it does not require
the common usage of a gem-diester like substrates.7cÀg
The NO2 substrate 1a was readily available in two steps
following literature procedures.3c The racemic reaction
proceeded smoothly at rt with DBU as the catalyst, yield-
ing the desired trisubstituted pyrrolidine in quantitative
yield. In order for the second step aza-Michael cyclization
to proceed, the N;H acidity of the aza-Henry product
had to be strong enough to be deprotonated by DBU.
Aldimines other than Ts-NdC stalled at the initial aza-
Henry stage, with no heterocyclization typically observed, as
documented previously.9 A single diastereomer (trans-4aa)
was isolated. NMR experiments revealed that the initial aza-
Henry reaction was modestly diastereoselective, giving a
mixture of trans/cis products in a ca. 2:1 ratio. It was
observed that only the trans aza-Henry adduct trans-3aa
underwent subsequent cyclization.
The need to employ NTs aldimines posted a challenging
task for the development of an asymmetric synthesis.
There is no successful organocatalytic aza-Henry reac-
tion involving this quite reactive imine functionality.10
Dual functional H-bond/base catalysts were examined
for asymmetric induction.11 Moderate yields and en-
antioselectivities were observed with the popularly used
TakemotoÀJacobsen amine/thiourea.12 Attempts to
improve ee through thiourea modification were fruit-
less. The corresponding amine/squaramide, developed
by Rawal et al.,13 afforded a nearly racemic product
(Figure 1).
Table 1. Cascade Design and Initial Condition Screeninga
(7) For the synthesis of chiral pyrrolidine derivatives using asym-
metric organocatalysis, see: (a) Jui, N. T.; Garber, J. A. O.; Finelli, F. G.;
Macmillan, D. W. C. J. Am. Chem. Soc. 2012, 134, 11400. (b) Kumar, I.;
Mir, N. A.; Gupta, V. K.; Rajnikant Chem. Commun. 2012, 48, 6975.
ꢀ
(c) Lin, S.-Z.; Deiana, L.; Zhao, G.-L.; Sun, J. L.; Cordova, A. Angew.
Chem., Int. Ed. 2011, 50, 7642. (d) He, L.; Chen, X.-H.; Wang, D.-N.;
Luo, S.-W.; Zhang, W.-Q.; Yu, J.; Ren, L.; Gong, L.-Z. J. Am. Chem.
Soc. 2011, 133, 13504. (e) Shi, F.; Luo, S.-W.; Tao, Z.-L.; He, L.; Yu, J.;
Tu, S.-J.; Gong, L.-Z. Org. Lett. 2011, 13, 4680. (f) Li, H.; Zu, L.-S.; Xie,
H. X.; Wang, J.; Wang, W. Chem. Commun. 2008, 5636. (g) Vicario,
J. L.; Reboredo, S.; Badia, S.; Carrillo, L. Angew. Chem., Int. Ed.
2007, 46, 5168. (h) Xu, X.-N.; Furukawa, T.; Okino, T.; Miyabe, H.;
Takemoto, Y. Chem.;Eur. J. 2006, 12, 466.
(8) (a) Wang, M.; Wang, Z.; Shi, Y.-H.; Shi, X.-X.; Fossey, J. S.;
Deng, W.-P. Angew. Chem., Int. Ed. 2011, 50, 4897. (b) Yamashita, Y.;
Imaizumi, T.; Kobayashi, S. Angew. Chem., Int. Ed. 2011, 50, 4893. (c)
Oura, I.; Shimizu, K.; Ogata, K.; Fukuzawa, S. Org. Lett. 2010, 12, 1752.
(d) Kuwano, P.; Kashiwabara, M.; Ohsumi, M.; Kusano, H. J. Am.
Chem. Soc. 2008, 130, 808. (e) Fukuzawa, S.; Oki, H. Org. Lett. 2008, 10,
1747. (f) Saito, S.; Tsubogo, T.; Kobayashi, S. J. Am. Chem. Soc. 2007,
129, 5364. (g) Zeng, W.; Chen, G.-Y.; Zhou, Y.-G.; Li, Y.-X. J. Am.
Chem. Soc. 2007, 129, 750.
entry
solvent
additive
yield (%)b
ee (%)c
1
2
3
4
5
6
7
DCM
À
À
À
À
À
50
33
23
46
26
72
>99
83
91
65
91
25
91
80d
toluene
THF
Et2O
(9) (a) Gu, Q.; You, S.-L. Chem. Sci. 2011, 2, 1519. (b) Scherrer,
R. A.; Donovan, S. F. Anal. Chem. 2009, 81, 2768. (c) Gimbert, C.;
CH3CN
toluene
toluene
˚
5 A M.S.
~
ꢀ
Moreno-Manas, M.; Perez, E.; Vallribera, A. Tetrahedron 2007, 63,
8305.
(10) Generally, <10% ee was observed using N-Ts aldimines; see:
˚
5 A M.S.
a Reactions were carried out on a 0.1 mmol scale using 10 mol %
catalyst at rt; 1a/2a = 1:1.5 at 0.2 M. b Isolated yields after flash column
chromatography. c Ee’s were determined by chiral HPLC analysis.
d After stirring at rt for 1 d, heated to 100 °C for 8 h.
ꢀ
ꢀ
(a) Gomez-Bengoa, E.; Linden, A.; Lopez, R.; Mugica-Mendiola, I.;
Oiarbide, M.; Palomo, C. J. Am. Chem. Soc. 2008, 130, 7955. (b) Wang,
C.; Zhou, Z.; Tang, C. Org. Lett. 2008, 10, 1707. N-Ts aldimines was
only demonstrated successfully in transition metal catalyzed aza-Henry
reactions; see: (c) Zhou, H.; Peng, D.; Qin, B.; Hou, Z.; Liu, X.; Feng, X.
J. Org. Chem. 2007, 72, 10302.
(11) For detailed catalyst and condition screening, see Supporting
Information for details.
(12) (a) Takemoto, Y. Chem. Pharm. Bull. 2010, 58, 593. (b) Connon,
S. J. Chem. Commun. 2008, 2499. (c) Taylor, M. S.; Jacobsen, E. N.
Angew. Chem., Int. Ed. 2006, 45, 1520.
(13) For recent reviews on chiral squaramides as organocatalysts,
see: (a) Aleman, J.; Parra, A.; Jiang, H.; Jørgensen, K. A. Chem.;Eur.
J. 2011, 17, 6890. (b) Storer, R. I.; Aciro, C.; Jones, L. H. Chem. Soc. Rev.
2011, 40, 2330. For pioneering references by Rawal et al., see: (c)
Konishi, H.; Lam, T. Y.; Malerich, J. P.; Rawal, V. H. Org. Lett. 2010,
12, 2028. (d) Zhu, Y.; Malerich, J. P.; Rawal, V. H. Angew. Chem., Int.
Ed. 2010, 49, 153. (e) Malerich, J. P.; Hagihar, K.; Rawal, V. H. J. Am.
Chem. Soc. 2008, 130, 14416.
(5) For reviews on “organo-metal cooperative catalysis”, see: (a) Shao,
Z.; Zhang, H. Chem. Soc. Rev. 2009, 38, 2745. (b) Du, Z.; Shao, Z. Chem.
Soc. Rev. 2013, 42, 1337. For recent examples of DYKATs in this
paradigm, see: (c) Deiana, L.; Afewerki, S.; Palo-Nieto, C.; Verho, O.;
ꢀ
Johnston, E. V.; Cordova, A. Sci. Rep. 2012, 2, 851. (d) Sun, W.; Zhu, G.;
Hong, L.; Wang, R. Chem.;Eur. J. 2011, 17, 13958. (e) Zhao, G.-L.;
Ullah, F.; Deiana, L.; Lin, S.; Zhang, Q.; Sun, J.; Ibrahem, I.; Dziedzic, P.;
ꢀ
Cordova, A. Chem.;Eur. J. 2010, 16, 1585. (f) Jensen, K. L.; Franke,
ꢀ
P. T.; Arroniz, C.; Kobbelgaard, S.; Jørgensen, K. A. Chem.;Eur. J.
2010, 16, 1750.
(6) Dai, Q.; Arman, H.; Zhao, J. C.-G. Chem.;Eur. J. 2013, 19,
1666.
Org. Lett., Vol. 15, No. 8, 2013
1959

Table 2. Substrate Scopea
Figure 1. TakemotoÀJacobsen amine/thiourea and Rawal
amine/squaramide chiral bifunctional catalyst.
Surprisingly, switching to a single point amine/amide
catalyst 5 allowed excellent cascade stereoselectivity.14
Interestingly, stronger H-bond analogues, such as sulphon-
amides, were much less selective, which indicated that the
H-bond strength of the catalyst ought to be finely tuned
within a narrow range for optimal results. The enantio-
selectivity of this cascade was in direct inverse proportion
to the polarity of solvents. In order to improve the overall
conversion of this cascade, several additives were screened
˚
and 5 A molecular sieves were found to significantly
a The absolute stereochemistry of the products was derived from the
single X-ray diffraction of trans-4aa. b Stirred at rt for 24 h and heated at
100 °C for 8 h. Results for rt 48 h: 30% yield, >99% ee. c Reaction time: 2 d.
accelerate the aza-Michael cyclization without eroding
the product ee. The desired trans-4aa was isolated in
72% yield with 91% ee. Heating to 100 °C accelerated
the cyclization, and a quantitative yield of trans-4aa could
be achieved with 80% ee. Both the relative and absolute
configuration of trans-4aa was unambiguously determined
by single-crystal X-ray diffraction.15
Nitroalkenes bearing various esters and ketones readily
engaged in the cyclization cascade. Both pyrrolidines and
piperidines were synthesized in good-to-excellent yields
and selectivities. It was evident that, for groups which
allowed faster cyclization, lower ee’s were observed. This is
in consistent with our reversible-DKR cascade model.
The DKR aza-Michael cyclization step was confirmed
by an independent experiment using racemic trans-3aa
(Scheme 1). Trans-3aa did not cyclize in the absence of
a base. Pyrrolidine trans-4aa was isolated in 50% ee in 80%
yield after 24 h, revealing a modestly selective kinetic
resolution. With aza-Henry reversibility omitted, this result
correlatedtoanSfactor of 3. Standing pure cis-3aa in DCM
solution led to the emergence of starting materials, a process
which could be accelerated by the addition of organic bases.
The substrate scope was examined, and the results are
summarized in Table 2. The substrate tolerance for aldi-
mines was particularly broad. Both aromatic and aliphatic
aldimineswithvarious substitutions were well suitedtothis
cascade, and high enantioselectivities (>90%) were ob-
tained uniformly, regardless of electronics and the sub-
stitution pattern. Heterocyclic aldimines derived from
pyridines were particularly selective (>99% ee). The rate
of the aza-Michael cyclization for this substrate, on the
other hand, proved very slow, and only a 30% isolated
yield wasobtained after 2 days at rt. The overall yield could
be improved by heating at 100 °C upon completion of the
initial step, while still maintaining 90% enantioselectivity.
This observation is worth noting, as relatively loose
H-bonding catalysis rarely shows high selectivity at such
elevated temperatures. This might be the result of ampli-
fied rate differences between the rate of reversibility of the
aza-Henry vs the aza-Michael cyclization. The conversion
to pyrrolidine was noticeably slower for aliphatic sub-
strates, and the reaction time increased to 2 days.
Scheme 1. DKR aza-Michael Cyclization Using Racemic 3aa
(14) For literature examples employing amide derived Cinchona
alkaloids as chiral organocatalysts, see: (a) Liu, X.-H.; Lin, L.-L.; Feng,
X.-M. Chem. Commun. 2009, 6145. (b) Seitz, T.; Baudoux, J.; Bekolo,
H.; Cahard, D.; Plaquevent, J.-C.; Lasne, M.-C.; Rouden, J. Tetrahe-
dron 2006, 62, 6155. (c) Brunner, H.; Baur, M. A. Eur. J. Org. Chem.
Real time NMR experiments showed that the first aza-
Henry reaction proceeded rather quickly (completed with-
in 4 h), generating a 2.3:1 mixture of trans-3aa and cis-3aa
(Table 3). The rate determining step was evidently the
cyclization step, as the primary aza-Henry adduct could be
isolated by column chromatography on SiO2.
ꢀ
2003, 2854. (d) Baur, M. A.; Riahi, A.; Henin, F.; Muzart, J. Tetra-
hedron: Asymmetry 2003, 14, 2755. (e) Drees, M.; Kleiber, L.; Weimer,
M.; Strassner, T. Eur. J. Org. Chem. 2002, 2405. (f) Brunner, H.;
Schmidt, P. Eur. J. Org. Chem. 2000, 2119.
(15) CCDC 905204, http://www.ccdc.cam.ac.uk. See Supporting
Information for details.
Under standard reaction conditions, the trans/cis ratio of
the thermodynamic equilibrium of the aza-Henry adduct was
1960
Org. Lett., Vol. 15, No. 8, 2013

The appending side chain could be engaged in additional
cyclization reactions with the pyrrolidine nitrogen. Inter-
estingly, attempts to tether the aromatic ring and pyrroli-
dine via dehydrogenative annulation resulted in a fully
aromatized compound due to in situ oxidation.16
Table 3. Real Time Conversion, dr, and ee for the Intermediates
and Product
Scheme 2. Product Derivatization
entry
time (h)
trans/cis (3aa)
yield (%)
ee (%)
1
4
2.3
2.5
2.7
3.1
2.3
10
91
91
91
91
80
2
8
24
3
16
24
6
40
4
5a
72
quant.
a The aza-Michael cyclization was carried at 100 °C.
In summary, we have developed a highly diastereo- and
enantioselective organocatalytic cascade that provides
heavily substituted pyrrolidines with up to three stereo-
genic centers. This domino reaction relies on a reversible
aza-Henry reaction and a subsequent DKR aza-Michael
cyclization, both catalyzed by an amine/amide Cinchona
alkaloid derivative, to ensure overall stereoselectivity. The
unique features of this reaction and the further utility of
amide derived H-bond catalysts are being investigated in
detail and will be reported in due course.
2.3:1 (trans-3aa vs cis-3aa). A stand alone aza-Henry experi-
ment between nitropropane and N-Ts benzaldimine using
catalyst 5 yielded similar levels of dr and ee. While trans-3aa
was formed with modest ee (65%), the optical purity of the cis
isomer was merely 20% ee initially. Ee’s of trans-3aa, cis-3aa,
and the product were monitored over time. Ee’s of both
intermediate isomers gradually decreased over time, accom-
panied by a steady accumulation of the cascade product
trans-4aa of constantly high optical purity. Eventually, both
cis-3aa and trans-3aa became racemic. Factoring a 3:1 kinetic
resolution efficiency (S = 3), the initial moderate 65%
enantioselectivity for trans-3aa (er = 6.6:1) was enhanced
by the subsequent DKR step to a theoretical enantiomeric
ratio of 6.6 Â 3 = 19.8/1, compared to the 91% observed ee,
for the pyrrolidine product trans-4aa. Based on the above
analysis, the low ee of cis-3aa is direct evidence that the cis-to-
trans isomerization is primarily due to reaction reversibility
(CÀC bond cleavage), not NO2 group epimerization under
basic conditions. Due to the modest selectivity of the kinetic
resolution step, direct epimerization of cis-3aa of 20% ee
would result in a theoretical ee of 83% (er = 3.7 Â 3 = 11.1)
for the final product.
Acknowledgment. This work is financially supported
by the National Basic Research Program of China
(2012CB722602), grants of Shenzhen special funds
for the development of biomedicine, Internet, new en-
ergy and new material industries (JC201104210111A
and JC201104210112A), Shenzhen innovation funds
(GJHZ20120614144733420), and the Shenzhen Peacock
Program (KQTD201103). The Shenzhen municipal devel-
opment and reform commission is thanked for a public
service platform program.
Supporting Information Available. Experimental pro-
cedures, characterization data, and NMR spectra for all
new compounds, and X-ray data. This material is avail-
able free of charge via the Internet athttp://pubs.acs.org.
(16) (a) Ackermann, L.; Wang, L.; Lygin, A. V. Chem. Sci. 2012, 3,
177. (b) Morimoto, K.; Hirano, K.; Satoh, T.; Miura, M. Org. Lett.
2010, 12, 2068. (c) Ueura, K.; Satoh, T.; Miura, M. Org. Lett. 2007, 9,
1407.
The rich functionalities surrounding this scaffold al-
lowed a number of chemical manipulations (Scheme 2).
For example, an isoquinoline fused pyrrolidine skele-
ton could be accessed through a three-step sequence.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 8, 2013
1961