Enantioselective [3 + 2]-cycloadditions catalyzed by a protected, multifunctional phosphine-containing α-amino acid

Article abstract of DOI:10.1021/ja0734243

Catalytic asymmetric [3 + 2]-cycloaddition reactions between α-allenic esters and enones are presented. We have found that a simple phosphine-containing protected α-amino acid derivative is capable of promoting such cycloadditions in high yields with significant levels of regioselectivity and enantioselectivity. Furthermore, employing chiral racemic γ-substituted allenoates in the cycloaddition with chalcone substrates results in a "deracemization" reaction furnishing cyclopentenes in high yields with up to 93% ee. Copyright

Full text of DOI:10.1021/ja0734243

Published on Web 08/21/2007
Enantioselective [3 + 2]-Cycloadditions Catalyzed by a Protected,
Multifunctional Phosphine-Containing r-Amino Acid
Bryan J. Cowen and Scott J. Miller*
Department of Chemistry, Yale UniVersity, P.O. Box 208107, New HaVen, Connecticut 06520-8107
Received May 14, 2007; E-mail: scott.miller@yale.edu
Table 1. Amino Acid Catalyst Survey for [3 + 2]-Cycloadditionsa
R-Amino acids are receiving explosive attention as catalysts for
organic reactions.1 The catalysis exhibited by proline is perhaps
most versatile,2 although there is now a growing list of reactions
that may be catalyzed by other amino acids, including both
proteinogenic and nonproteinogenic variants. Concerning the latter,
“unnatural” amino acids have also proven important as ligands for
metal-catalyzed reactions. In this spirit, the introduction of diphe-
nylphosphinylalanine (1) by Gilbertson marked a significant
advance both conceptually3 and in the development of highly
selective metal-based catalysts.4 The dual capacity of phosphines
to serve as either ligands or as catalysts in the absence of metals
stimulated a study of 1 in this latter context. We wondered if
incorporation of 1 into peptides would create a functionality-rich,
chiral environment for the Lewis basic P atom that might be tuned
as a function of peptide structures (i.e., 1-P).5,6
yield
R+γ (%)b
ee R
(%)d
entry
catalyst
temp/time
R:
γc
1
2
3
4
5
6
7
8
1a
1b, R1 ) Cbz, R2 ) H
23 °C/3 h
23 °C/3 h
1c, R1 ) CO2Me, R2 ) H 23 °C/3 h
93
88
96
86
40
60
95
16
94:6
95:5
93:7
93:7
69
68
66
51
1d, R1 ) Ac, R2dH
23 °C/3 h
23 °C/3 h
23 °C/3 h
-25 °C/30 h
-40 °C/24 h
1e, R1 ) Ts, R2 ) H
88:12 rac
1f, R1 ) Boc, R2 ) Me
93:7
>99:1
>99:1
rac
80
84
1a
1a
a Reactions were run in PhCH3 with 1.5 equiv of allenic ester and 10
mol % of 1. b Isolated yield after silica gel chromatography. c Determined
from the 1H NMR spectrum of purified product. d All ee’s were measured
using chiral HPLC.
In this context, we examined the formal catalytic [3 + 2]-cy-
cloaddition of allenoate esters and enones pioneered by Lu and
co-workers (Scheme 1).7 Allene 2a and enone 3 combine to give
cycloadduct 4 in the presence of substoichiometric quantities of
triphenyl phosphine (PPh3). From a mechanistic point of view, the
reaction is often represented as proceeding through a zwitterionic
dipole en route to a formal [3 + 2]-cycloaddition.8 Notably, signi-
ficant progress has been reported with chiral catalysts, including
important contributions by Zhang,9 Fu,10 and Wallace.11 We noted
previously that this coupling reaction may be diverted to give a
conjugate addition product when phosphine catalysts are exchanged
for amines.12 We now report that R-amino acids of type 1 lead to
substantial enantiocontrol over the phosphine-catalyzed pathway
(Scheme 1).
difference. Catalyst 1d delivers the product with less selectivity
(R:γ ) 93:7; 4R exhibiting 51% ee; entry 4). Replacement of the
Boc group in 1a with the tosyl group (catalyst 1e) leads to a total
loss of enantioselectivity (entry 5). Likewise, replacement of the
NH function in catalyst 1a with an N-Me group (catalyst 1f; entry
6) leads to racemic products, suggesting that subtle H-bonding
effects may be important in the transition state. Notably, each cata-
lyst in the series provides reasonable reaction rates, although the
urethanes are the most active, delivering >88% yield of 4 within
3 h (23 °C).
Lowering the reaction temperature leads to an increase in
selectivity. For example, when the reaction is run at -25 °C,
regioselectivity improves, with essentially exclusive formation of
4R (R:γ >99:1); the major product under these conditions exhibits
80% ee (entry 7). At -25 °C, 30 h is required for the reaction to
reach completion (95% isolated yield). Further lowering of the
temperature leads to a modest increase in selectivity (84% ee; entry
8), but the rate is slow under these conditions (entry 8).
Scheme 1
An examination of the substrate scope reveals that catalyst 1a
tolerates a range of substrates (Table 2). Methoxy-substituted case
5 affords 6 with slightly enhanced selectivity (entry 1, 84% ee).
Heteroatom-substituted cases, along with heteroaromatic substrates,
also afford products with excellent regioselectivity, although
enantiomeric ratios are somewhat lower. For example, heterocycle
8 is formed from 7 with an R:γ ratio of 94:6 and 65% ee (entry 2).
Furanoid enone 9 leads to spirocycle 10 with total regiocontrol and
76% ee (entry 3). Pyrrole derivative 11 exhibits a similar reaction
profile, with 12 formed with >99:1 R:γ selectivity, and 71% ee
(entry 4). Acyclic substrates also participate, with somewhat lower
selectivity. R-Methylated enone 13 gives 14 with 85:15 regiose-
lectivity and 70% ee (entry 5). Chalcone (15) leads to compound
Our initial experiments established that indeed simple derivatives
of 1 are effective catalysts for the cycloaddition of 2b and 3, deli-
vering the cycloadducts 4R and 4γ in excellent yield and regiose-
lectivity (eq 1, Table 1). For example, with catalyst 1a, allenoate
2b and tetralone 3 undergo cycloaddition to form 4R and 4γ in a
combined yield of 93% (94:6 regioisomeric ratio), with the major
adduct 4R exhibiting 69% ee (entry 1). When the urethane substi-
tuent is swapped from t-Bu to either benzyl (catalyst 1b) or methyl
(catalyst 1c), the results are similar (entries 2 and 3). Exchange of
the urethane to acetamide (1d), however, produces a more dramatic
9
10988
J. AM. CHEM. SOC. 2007, 129, 10988-10989
10.1021/ja0734243 CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Substrate Scope for [3 + 2]-Cycloadditionsa
Table 3. Substrate Scope for [3 + 2]-Cycloadditions Employing
γ-Substituted Allenic Ester (()-17a
a Reactions were run with 1.5 equiv of allenic ester. b Isolated yield after
silica gel chromatography. c All ee’s were measured using chiral HPLC.
the major enantiomer that is formed. Ensemble 24 may assume
transition state organization via the illustrated H-bond and requires
a All data are the average of two experiments. Reactions were run at
-25 °C in PhCH3 with 1.5 equiv of allenic ester and 10 mol % of 1a.
1
b Isolated yield after silica gel chromatography. c Determined from the H
NMR spectrum of purified product. d All ee’s were measured using chiral
HPLC. e Reaction was run at 4 °C. f Reaction was run with 5 equiv of enone.
g Reaction was run at 23 °C with 20 mol % of 1a. h See Supporting
Information for the determination of absolute stereochemistry. i Benzyl
allenoates were selected for their ease of handling and optimal performance.
that the dipolarophile approach the zwitterion from the π-face
opposite one of the Ph rings of the catalyst. Of course, alternatives
are possible, and detailed studies of mechanism are now ongoing.
16 with 82% ee, although regioselectivity is reduced to a nearly
statistical level (entry 6).
Chalcone, however, provided an opportunity to observe a unique
“deracemization” reaction upon cycloaddition with racemic γ-sub-
stituted allene 17. In these cases, we were particularly interested
in whether or not racemic allene substrates could be subjected to
“dynamic kinetic asymmetric transformations,” as the initial adducts
of catalyst-allene reaction are intermediates such as 18, in which
the element of planar chirality has been erased (eq 2).
Acknowledgment. We are grateful to the National Science
Foundation (CHE-0639069) and Merck & Co. for support.
Supporting Information Available: Experimental procedures and
characterization. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) For example, see: (a) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed.
2004, 43, 5138. (b) Jarvo, E. R.; Miller, S. J. Tetrahedron 2002, 58, 2481.
(2) Movassaghi, M.; Jacobsen, E. N. Science 2002, 298, 1904.
(3) Agarkov, A.; Greenfield, S.; Xie, D.; Pawlick, R.; Starkey, G.; Gilbertson,
S. R. Biopolymers: Pept. Sci. 2006, 84, 48.
(4) Gilbertson, S. R.; Collibee, S. E.; Agarkov, A. J. Am. Chem. Soc. 2000,
122, 6522.
(5) Miller, S. J. Acc. Chem. Res. 2004, 37, 601.
(6) For examples of bifunctional chiral phosphines, see: (a) Shi, M.; Chen,
L.-H.; Li, C.-Q. J. Am. Chem. Soc. 2005, 127, 3790. (b) Matsui, K.;
Takizawa, S.; Sasai, H. Synlett 2006, 5, 761.
Indeed, we were pleased to find that when (()-17 is exposed to
chalcone in the presence of a full equivalent of catalyst 1a, a 94%
yield of cyclopentene 19 is formed within 7 h and that the product
exhibits 91% ee (Table 3, entry 1). Furthermore, when a substo-
ichiometric amount of the catalyst is employed, a similar result is
observed, although the reaction rate is diminished (entry 2, 38%
yield within 24 h, 93% ee). These effects are maintained through
a set of chalcones. For example, p-chloro-substituted chalcone 20
may be converted to 21 in 96% yield with the product exhibiting
87% ee (entry 3). p-Methoxy-substituted chalcone 22 delivers its
corresponding cycloadduct 23 in 89% yield, with 90% ee (entry
4). In each of these cases, these highly substituted cycloadducts
are formed as single regio- and diastereomers. These examples
constitute unique cases of allenoate deracemizations via chiral
phosphine-catalyzed [3 + 2]-cycloaddition.13
(7) (a) Zhang, C.; Lu, X. J. Org. Chem. 1995, 60, 2906. (b) Du, Y.; Lu, X.;
Yu, Y. J. Org. Chem. 2002, 67, 8901.
(8) For examples of phosphine-catalyzed allenoate annulations, see: (a) Zhu,
X.-F.; Henry, C. E.; Wang, J.; Dudding, T.; Kwon, O. Org. Lett. 2005, 7,
1387. (b) Zhu, X.-F.; Schaffner, A.-P.; Li, R. C.; Kwon, O. Org. Lett.
2005, 7, 2977. (c) Zhu, X.-F.; Lan, J.; Kwon, O. J. Am. Chem. Soc. 2003,
125, 4716.
(9) Zhu, G.; Chen, Z.; Jiang, Q.; Xiao, D.; Cao, P.; Zhang, X. J. Am. Chem.
Soc. 1997, 119, 3836.
(10) Wilson, J. E.; Fu, G. C. Angew. Chem., Int. Ed. 2006, 45, 1426.
(11) Wallace, D. J.; Sidda, R. L.; Reamer, R. A. J. Org. Chem. 2007, 72, 1051.
(12) Evans, C. A.; Miller, S. J. J. Am. Chem. Soc. 2003, 125, 12394.
(13) For a diastereoselective example employing an achiral phosphine, see:
Ung, A. T.; Schafer, K.; Lindsay, K. B.; Pyne, S. G.; Amornraksa, K.;
Wouters, R.; Van der Linden, I.; Biesmans, I.; Lesage, A. S. J.; Skelton,
B. W.; White, A. H. J. Org. Chem. 2002, 67, 227.
(14) Xia, Y.; Liang, Y.; Chen, Y.; Wang, M.; Jiao, L.; Huang, F.; Liu, S.; Li,
Y.; Yu, Z.-X. J. Am. Chem. Soc. 2007, 129, 3470.
(15) (a) Zhu, X.-F.; Henry, C.-E.; Kwon, O. J. Am. Chem. Soc. 2007, 129,
6722. (b) Mercier, E.; Fonovic, B.; Henry, C.; Kwon, O.; Dudding, T.
Tetrahedron Lett. 2007, 48, 3617. (c) Dudding, T.; Kwon, O.; Mercier,
E. Org. Lett. 2006, 8, 3643.
Our current thinking on the basis of asymmetric induction is
stimulated by recent observations of Yu14 and Kwon15 concerning
mechanism. Transition state 24 is consistent with the identity of
JA0734243
9
J. AM. CHEM. SOC. VOL. 129, NO. 36, 2007 10989