Brucine diol-copper-catalyzed asymmetric synthesis of endo -pyrrolidines: The mechanistic dichotomy of imino esters

Article abstract of DOI:10.1021/acs.orglett.5b00277

Enantio- and diastereodivergent approaches to pyrrolidines are described by using catalyst- and substrate-controlled reaction pathways. A concerted endo-selective [3 + 2]-cycloaddition pathway is developed for the reaction of methyl imino ester, whereas endo-pyrrolidines with an opposite absolute stereochemical outcome are prepared by using the stepwise reaction pathway of tert-butyl imino ester. The development of catalyst- and substrate-controlled stereodivergent approaches highlights the inherent substrate-catalyst interactions in the [3 + 2]-cycloaddition reactions of metalated azomethine ylides.

Full text of DOI:10.1021/acs.orglett.5b00277

Letter
pubs.acs.org/OrgLett
Brucine Diol−Copper-Catalyzed Asymmetric Synthesis of endo-
Pyrrolidines: The Mechanistic Dichotomy of Imino Esters
Jian-Yuan Li,† Hun Young Kim,‡ and Kyungsoo Oh*,‡
†Department of Chemistry and Chemical Biology, Indiana University Purdue University Indianapolis, Indianapolis, Indiana 46037,
United States
‡College of Pharmacy, Chung-Ang University, 84 Heukseok-ro, Dongjak, Seoul 156-756, Republic of Korea
S
* Supporting Information
ABSTRACT: Enantio- and diastereodivergent approaches to pyrrolidines are described by using catalyst- and substrate-
controlled reaction pathways. A concerted endo-selective [3 + 2]-cycloaddition pathway is developed for the reaction of methyl
imino ester, whereas endo-pyrrolidines with an opposite absolute stereochemical outcome are prepared by using the stepwise
reaction pathway of tert-butyl imino ester. The development of catalyst- and substrate-controlled stereodivergent approaches
highlights the inherent substrate−catalyst interactions in the [3 + 2]-cycloaddition reactions of metalated azomethine ylides.
Scheme 1. Enantiodivergent Pyrrolidine Syntheses: Stepwise
and Concerted Reaction Pathways
hiral pyrrolidines are one of the key structural motifs
1
Cpresent in many biologically important compounds as well
as in the core structures of fascinating organocatalysts.2 Among
the synthetic strategies developed for chiral pyrrolidines, the
catalytic asymmetric [3 + 2]-cycloaddition reactions between in
situ generated azomethine ylides and activated alkenes render a
facile access to pyrrolidines with a diverse array of functional
groups.3 Chiral metal catalysts as well as organocatalysts have
been investigated in which the absolute and relative stereo-
chemistries of products are primarily controlled by specific chiral
ligand sets. However, the relationship between the stereo-
chemical outcome and the reaction pathway of [3 + 2]-
cycloaddition reactions of imino esters with activated alkenes are
not well understood due to the lack of clear experimental
evidence for either concerted or stepwise reaction pathway.4
Recently, we disclosed a diastereodivergent conjugate addition
reaction of imino esters to nitroalkenes in the presence of brucine
diol (BD)−Cu complexes (Scheme 1).5 Our substrate-
controlled diastereoselective conjugate addition reactions shed
light on the mechanistic insight into the stepwise exo- and endo-
selective [3 + 2]-cycloaddition reaction pathways of imino esters
with nitroalkenes.6 With the aim of further studying the reaction
mechanisms of [3 + 2]-cycloaddition reactions, we investigated
the absolute and relative stereochemical outcomes of pyrroli-
dines from imino esters and nitroalkenes under the BD−Cu
catalysis. Herein, we present an endo-selective concerted [3 + 2]-
cycloaddition reaction between imino esters and nitroalkenes,
ultimately leading to the discovery of the reaction pathway-
controlled enantiodivergent and the substrate-controlled dia-
stereodivergent [3 + 2]-cycloaddition reactions of azomethine
ylides.
First, we reexamined the [3 + 2]-cycloaddition reactions
between methyl imino ester 1 and nitroalkene 2 using various
BD−Cu catalysts (Table 1). Since the use of BD−Cu catalysts
derived from CuOAc exclusively led to the formation of syn-
conjugate addition product,5 we opted for the use of other
copper salts that typically require amine bases for the generation
of active BD−Cu catalysts.7 The change of copper salts
dramatically changed the course of the reaction, leading to the
formation of a mixture of pyrrolidines instead of conjugate
Received: January 28, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b00277
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Optimization of Endo-Selective [3 + 2]-
Cycloaddition Reaction
Scheme 2. Scope of Endo-Selective [3 + 2]-Cycloaddition
Reaction
a
b
c
entry
1
metal/base
solvent
THF
endo/exo yield (%) ee (%)
CuCl/Et3N
NR
d
2
CuCl/Et3N
TCE
25:1
3:1
55
20
65
61
57
67
NR
65
64
94
85
97
72
50
28
77
82
66
3
CuI/Et3N
THF
4
CuOTf/Et3N
CuOTf/DBU
CuOTf/DBN
CuOTf/DBN
CuOTf/DBN
CuOTf/DBN
Cu(OTf)2/DBN
Cu(NTf2)2/DBN
CuOTf/DBN
CuOTf/DBN
THF
4:1
5
THF
10:1
13:1
9:1
6
THF
7
2-MeTHF
PhCH3
CHCl3
THF
8
9
20:1
7:1
77
80
80
26
84
10
11
THF
10:1
1:1
e
12
f
THF
13
THF
25:1
a
b
Determined by crude 1H NMR. Isolated yield of products after
c
column chromatography. Determined by HPLC using a chiral
column. 20 mol % of EtOH additive. Reaction at 0 °C. 20 mol %
of BD-CuOTf. NR = no reaction.
d
e
f
addition products (Table 1, entries 1−4). Thus, the use of CuCl
in 1,1,2-trichloroethane preferentially provided endo-3a in 72%
ee (Table 1, entry 2). However, no further improvement in the
reactivity and selectivity was made when other reaction
parameters were varied. Gratifyingly, when we employed
CuOTf in combination with DBN as base the formation of
endo-3a improved to 82% ee with 13:1 dr (Table 1, entry 6). No
significant solvent effect was found (Table 1, entries 7−9). The
oxidation state of copper did not have much influence on the
observed stereoselectivities (Table 1, entries 10 and 11);
however, the use of either Cu(NTf2)2 or elevated reaction
temperature showed a much faster reaction albeit with
diminished selectivities (entries 11 and 12). Finally, the catalyst
loading was investigated to improve the reactivity and selectivity,
and the use of 20 mol % of BD−CuOTf was identified as an
optimal condition, leading to the formation of endo-3a in 97%
yield with 84% ee (Table 1, entry 13).
concerted [3 + 2]-cycloaddition pathway. To further verify such a
mechanistic pathway, we independently prepared the Mannich
precursor anti-5 (vide infra). Upon subjecting the Mannich
precursor to a stoichiometric amount of BD−CuOTf catalysts, a
slow formation of endo-3a was observed (Scheme 3). The
Scheme 3. Pyrrolidines from Intramolecular Mannich
Reaction Pathway
The optimized endo-selective [3 + 2]-cycloaddition conditions
were further investigated using other imino esters and nitro-
alkenes (Scheme 2). A wide range of imino esters with different
electronic and steric effects provided the desired endo-3a−h in
high yields with good to excellent enantio- as well as
diastereoselectivities.8 The reaction was also applicable to
other nitroalkenes with slightly diminished enantioselectivities
(endo-3i−j); however a general trend in the substrate−selectivity
relationship was not found since a combination of different
substituents on the imino esters and the nitroalkenes provided
excellent selectivities (endo-3k−o). The relative and absolute
stereochemistry of endo-3 was confirmed to be (2R,3S,4R,5R) by
comparison of its HPLC retention times with authentic samples
(see the Supporting Information for details).
The stereochemical outcome of the endo-selective [3 + 2]-
cycloaddition reaction provided insightful mechanistic informa-
tion. Upon close inspection of varied reaction conditions in
Table 1 and Scheme 2, we observed the exclusive formation of
endo-3 and exo-3 regardless of the reaction conversion. The
absence of conjugate addition products as well as other
diastereomeric pyrrolidines suggested a strong possibility of a
inefficient intramolecular Mannich cyclization of anti-5 under
our asymmetric conditions strongly implied a less likely scenario
for the stepwise reaction pathway to endo-3a. In addition, the
intramolecular Mannich reaction of anti-5 under acidic
conditions resulted in a 5:1:1.5 mixture of three pyrrolidines.
This result further strengthened the possible formation of other
diastereomeric pyrrolidines through a stepwise reaction
mechanism. Under basic conditions, however, the stereoselective
intramolecular Mannich reaction was observed to provide endo-
3a.
The stereochemical outcome of the concerted [3 + 2]-
cycloaddition reactions depicted in Scheme 2 was opposite to
the endo-pyrrolidines obtained through a sequence of anti-
selective conjugate addition and Mannich cyclization. To further
B
DOI: 10.1021/acs.orglett.5b00277
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
investigate the stepwise approach to endo-pyrrolidines, we
examined the Mannich cyclization of chiral anti-6 (Scheme 4).
lectivities of endo-7 via the stepwise reaction sequence. Also, the
fact that the formation of 4-epi-endo-3 was not observed in our
concerted [3 + 2]-cycloaddition conditions indicated that the
epimerization did not occur under the Cu−BD-catalyzed
asymmetric conditions.
Scheme 4. Scope of Stepwise Approach to endo-7
The observed enantioselectivities of endo-7 from the stepwise
reaction sequence in Scheme 4 were lower than those of
conjugate addition products, anti-6. For example, rather
surprisingly, some conjugate addition products, anti-6p and
anti-6q, displayed a more significant loss of enantioselectivity
under the base-promoted intramolecular Mannich cyclization
conditions (Scheme 6). The erosion of enantioselectivity of endo-
Scheme 6. Reversibility of Mannich Precursors
7 suggested that the Mannich precursors, anti-6, underwent a
retro-Michael reaction. However, the fact that no other
diastereometic pyrrolidines were observed clearly illustrated
that the recombination occurred without a significant dissocia-
tion of nitroalkenes and imino esters. The reversible nature of the
Mannich precursors, anti-6, implies the difficulty associated with
the development of endo-selective catalytic asymmetric [3 + 2]-
cycloaddition reactions of imino esters and nitroalkenes. Thus, it
is conceivable that the observation of low diastereo- and
enantioselectivities of endo-pyrrolidines in literature could be
due to the reversible nature of Mannich intermediates like anti-
5/6 in a stepwise reaction pathway.10
While more work is needed to assert the reaction mechanisms,
the specific substrate-catalyst interaction is believed to be a
crucial diverging factor for the endo-selective concerted [3 + 2]-
cycloaddition and conjugate addition reactions. Thus, the
stereochemical outcome of the BD-Cu-catalyzed reactions
between methyl imino ester and nitroalkenes could be explained
by invoking the stereoselective generation of enolates (Scheme
7).11 The O-metalated enolate of tert-butyl glycine imines should
favor a (E)-geometry with a coordinating species (THF or protic
additives) at the copper center, whereas a (Z)-geometry could be
The removal of a benzophenone imine moiety followed by
condensation with aldehydes led to the stereoselective formation
of endo-7 in good to excellent yields (see the Supporting
Information for details). In all cases, the formation of minor 4-
epi-endo-7 was observed while no other diastereomeric
pyrrolidines were detected. While the observed diastereoselec-
tivity of endo-7 varied among substrates (dr >25:1 to 3:1), a
synthetically useful level of enantioselecivity was obtained using a
range of nitroalkenes and aldehydes (80−90% ee’s). The relative
and absolute stereochemistry of endo-7 was unambiguously
confirmed to be (2S,3R,4S,5S), a stereochemical outcome
opposite to that of the endo-3 from the concerted [3 + 2]-
cycloaddition reaction pathway.9
In an effort to identify the origin of 4-epi-endo-pyrrolidines, we
reacted enantioenriched endo-pyrrolidines under basic condi-
tions (Scheme 5). Based on the formation of endo- and 4-epi-
endo-pyrrolidines in comparable enantioselectivities, we con-
cluded that the 4-epi-endo-pyrrolidines likely originated from the
epimerization of endo-pyrrolidines under basic conditions in
protic solvent, MeOH, a possible cause for varied diastereose-
Scheme 7. Stereomodels for Divergent Reaction Pathways
Scheme 5. Epimerization of endo-Pyrrolidines
C
DOI: 10.1021/acs.orglett.5b00277
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(c) Kim, H. Y.; Kim, H.; Oh, K. Angew. Chem., Int. Ed. 2010, 49, 4476.
(d) Kim, H. Y.; Oh, K. Org. Lett. 2011, 13, 1306.
(8) While 10 mol % of BD−Cu(NTf2)2 could be employed, the
observed enantio- and diastereoselectivities were somewhat sensitive to
the nature of substrates (i.e., 3d, 95% yield, 7:1 dr, 76% ee).
(9) The opposite enantioselectivity was confirmed by converting endo-
7a to endo-3a and comparing it with products from the concerted
reaction pathway in Scheme 2.
(10) For endo-pyrrolidines with low ee’s and dr’s, see: (a) Yan, X.-X.;
Peng, Q.; Zhang, Y.; Zhang, K.; Hong, W.; Hou, X.-L.; Wu, Y.-D. Angew.
Chem., Int. Ed. 2006, 45, 1979. (b) Li, Q.; Ding, C.-H.; Hou, X.-L.; Dai,
L.-X. Org. Lett. 2010, 12, 1080. (c) Arai, T.; Yokoyama, N.; Mishiro, A.;
Sato, H. Angew. Chem., Int. Ed. 2010, 49, 7895. (d) Li, Q.; Ding, C.-H.;
Li, X.-H.; Weissensteiner, W.; Hou, X.-L. Synthesis 2012, 44, 265.
(e) Imae, K.; Konno, T.; Ogata, K.; Fukuzawa, S.-I. Org. Lett. 2012, 14,
4410. (f) Gonzalez-Esguevillas, M.; Adrio, J.; Carretero, J. C. Chem.
Commun. 2012, 48, 2149. (g) Conde, E.; Bello, D.; de Cozar, A.;
favored for the N,O-metalated azomethine ylides of methyl
glycine imine. The consequence of such a geometric difference in
the enolates should result in the divergent reaction pathways to
either concerted [3 + 2]-cycloaddition or conjugate addition
products.
In summary, we have developed enantiodivergent approaches
to endo-pyrrolidnes based on two distinctive reaction pathways.
While the stereodivergent approaches using the same chiral
source are just gathering momentum in the catalytic asymmetric
reactions,12 the implementation of such asymmetric strategies to
multiple synthetic transformations has been challenging due to
the nature of specific factors that effects the reversal of
stereoselectivity. By utilizing the catalyst- and substrate-
controlled reaction pathways of imino esters, it was possible
for the first time to demonstrate the reversal of enantioselectivity.
The extensions of reaction pathway-controlled stereodivergent
approaches to other catalytic reactions are currently underway in
our laboratory, and our results will be reported in due course.
San
́
chez, M.; Vaz
́
quez, M. A.; Cossío, F. P. Chem. Sci. 2012, 3, 1486.
era, C.; Sansano, J. M.; Larranaga, O.; de Cozar,
(h) Castello, L. M.; Naj
́
A.; Cossío, F. P. Org. Lett. 2013, 15, 2902. For a successful endo-
pyrrolidine formation, possibly via a concerted reaction pathway, see:
(i) Arai, T.; Mishiro, A.; Yokoyama, N.; Suzuki, K.; Sato, H. J. Am. Chem.
Soc. 2010, 132, 5338.
ASSOCIATED CONTENT
* Supporting Information
■
S
(11) For a stereoselective enolate formation of N-protected glycinates
using LDA, see: Davis, F. A.; Deng, J. Org. Lett. 2004, 6, 2789.
(12) For recent reviews, see: (a) Bartok, M. Chem. Rev. 2010, 110,
1663. (b) Escorihuela, J.; Burguete, M. I.; Luis, S. V. Chem. Soc. Rev.
2013, 42, 5595.
Experimental procedures and characterization data for all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: kyungsoooh@cau.ac.kr.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was made possible by generous support from
Chung-Ang University (CAU). We thank Dr. Karl Dria at
Indiana University Purdue University Indianapolis (IUPUI) for
his assistance with spectral analysis.
REFERENCES
■
(1) For reviews, see: (a) Felpin, F.-X.; Lebreton, J. Eur. J. Org. Chem.
2003, 19, 3693. (b) Galliford, C. V.; Scheidt, K. A. Angew. Chem., Int. Ed.
2007, 46, 8748.
(2) For recent reviews, see: (a) Mukherjee, S.; Yang, J. W.; Hoffman, S.;
List, B. Chem. Rev. 2007, 107, 5471. (b) MacMillan, D. W. C. Nature
2008, 455, 304. (c) Jiang, H.; Jørgensen, K. L. Angew. Chem., Int. Ed.
2011, 50, 8492.
(3) For recent reviews, see: (a) Stanley, L. M.; Sibi, M. P. Chem. Rev.
2008, 108, 2887. (b) Adrio, J.; Carretero, J. C. Chem. Commun. 2011, 47,
6784.
(4) (a) Ayerbe, M.; Arrieta, A.; Cossío, F. P. J. Org. Chem. 1998, 63,
1795. (b) Vivanco, S.; Lecea, B.; Arrieta, A.; Prieto, P.; Morao, I.; Linden,
A.; Cossío, F. P. J. Am. Chem. Soc. 2000, 122, 6078. (c) de Coz
́
ar, A.;
Cossío, F. P. Phys. Chem. Chem. Phys. 2011, 13, 10858.
(5) Kim, H. Y.; Li, J.-Y.; Kim, S.; Oh, K. J. Am. Chem. Soc. 2011, 133,
20750; For a stereochemical reassignment, see: J. Am. Chem. Soc. 2012,
134, 16918.
4a
́
(6) According to Cossıo’s pyrrolidine nomenclature, the notation
endo signifies the “syn” relationship between the R1 of imino ester and
the electron-withdrawing group (EWG) of alkenes. Consequently, the
exo implies the “anti” relationship between the R1 and the EWG. The
notation endo′ and exo′ indicates an “anti” relationship between the R1
and the ester moiety in pyrrolidines.
(7) (a) Kim, H. Y.; Shih, H.-J.; Knabe, W. E.; Oh, K. Angew. Chem., Int.
Ed. 2009, 48, 7420. (b) Kim, H. Y.; Oh, K. Org. Lett. 2009, 11, 5682.
D
DOI: 10.1021/acs.orglett.5b00277
Org. Lett. XXXX, XXX, XXX−XXX