Direct carbon-carbon bond formation via soft enolization: A biomimetic asymmetric mannich reaction of phenylacetate thioesters

Article abstract of DOI:10.1021/ol101152b

(Equation Presented). An asymmetric Mannich reaction of phenylacetate thioesters and sulfonylimines using cinchona alkaloid-based amino (thio)urea catalysts is reported that employs proximity-assisted soft enolization. This approach to enolization is based on the cooperative action of a carbonyl-activating hydrogen bonding (thio)urea moiety and an amine base contained within a single catalytic entity to facilitate intracomplex deprotonation. Significantly, this allows thioesters over a range of acidity to react efficiently, thereby opening the door to the development of a general mode of enolization-based organocatalysis of monocarboxylic acid derivatives.

Full text of DOI:10.1021/ol101152b

ORGANIC
LETTERS
2010
Vol. 12, No. 15
3376-3379
Direct Carbon-Carbon Bond Formation
via Soft Enolization: A Biomimetic
Asymmetric Mannich Reaction of
Phenylacetate Thioesters
Mark C. Kohler, Julianne M. Yost, Michelle R. Garnsey, and Don M. Coltart*
Department of Chemistry, Duke UniVersity, Durham, North Carolina 27708
don.coltart@duke.edu
Received May 23, 2010
ABSTRACT
An asymmetric Mannich reaction of phenylacetate thioesters and sulfonylimines using cinchona alkaloid-based amino (thio)urea catalysts is
reported that employs proximity-assisted soft enolization. This approach to enolization is based on the cooperative action of a carbonyl-
activating hydrogen bonding (thio)urea moiety and an amine base contained within a single catalytic entity to facilitate intracomplex deprotonation.
Significantly, this allows thioesters over a range of acidity to react efficiently, thereby opening the door to the development of a general mode
of enolization-based organocatalysis of monocarboxylic acid derivatives.
Soft enolization1,2 provides a mild and operationally simple
approach to the deprotonation of certain types of monocar-
bonyl compounds. In contrast to hard enolization, wherein
deprotonation is achieved irreversibly using a very strong
base such as LDA, soft enolization occurs when a relatively
weak amine base and a carbonyl activating component act
in concert to effect reversible deprotonation. We have been
investigating this mode of enolization with thioesters in direct
carbon-carbon bond formation using Mg2+ Lewis acids for
carbonyl activation.2 Our inspiration for studying thioesters
in this context stems from the way in which enolization
occurs in the enzyme citrate synthase.3 Thioester activation
in citrate synthase is achieved by hydrogen bonding rather
than Lewis-acid coordination (Scheme 1a). While a weaker
form of carbonyl activation, it is sufficient to allow depro-
tonation by a weakly basic carboxylate group.4 This is likely
due in large part to the proximity effects imparted to the
system as a result of the close spatial arrangement of the
(3) (a) Mulholland, A. J.; Lyne, P. D.; Karplus, M. J. Am. Chem. Soc.
2000, 122, 534–535. (b) Usher, K. C.; Remington, S. J.; Martin, D. P.;
Drueckhammer, D. G. Biochem. 1994, 33, 7753–7759. (c) Wiegand, G.;
Remington, S. J. Ann. ReV. Biophys. Chem. 1986, 15, 97–117.
(4) The effect of hydrogen bonding on thioester acidity has been shown
for acetyl-CoA dehydrogenase. See: Rudik, I.; Thorpe, C. Arch. Biochem.
Biophys. 2001, 392, 341–348.
(1) For pioneering examples of soft enolization, see: Rathke, M. W.;
Cowan, P. J. J. Org. Chem. 1985, 50, 2622–2624. Rathke, M. W.; Nowak,
M. J. Org. Chem. 1985, 50, 2624–2626. Tirpak, R. E.; Olsen, R. S.; Rathke,
M. W. J. Org. Chem. 1985, 50, 4877–4879
.
(5) For accounts of proximity accelerated intramolecular transformations
in general, see: (a) Menger, F. M. Acc. Chem. Res. 1985, 18, 128–134. (b)
Kirby, A. AdV. Phys. Org. Chem. 1980, 17, 183–278. (c) Page, M. I. Chem.
Soc. ReV. 1973, 2, 295–323. (d) Page, M.; Jencks, W. Proc. Natl. Acad.
Sci. U.S.A. 1971, 68, 1678–1683. (e) Capon, B. Q. ReV., Chem. Soc. 1964,
18, 45–111.
(2) (a) Yost, J. M.; Garnsey, M. R.; Kohler, M. C.; Coltart, D. M.
Synthesis 2008, 56–58. (b) Zhou, G.; Lim, D.; Coltart, D. M. Org. Lett.
2008, 10, 3809–3812. (c) Zhou, G.; Yost, J. M.; Coltart, D. M. Synthesis
2007, 478–482. (d) Yost, J. M.; Zhou, G.; Coltart, D. M. Org. Lett. 2006,
8, 1503–1506
.
10.1021/ol101152b  2010 American Chemical Society
Published on Web 07/07/2010

would provide the activating hydrogen bonding interaction.
This would be connected to an amine such that the thioester
R-protons would be positioned in proximity to the base in
the bound state, thus enabling intracomplex deprotonation
(3f4). An amine was chosen over a carboxylate as it would
be both more strongly basic in this application and would
lead to a relatively stable Zwitterionic ammonium enolate
(4),12,13 further facilitating deprotonation. Reaction of the
enolate with imine 5, followed by proton transfer, would then
liberate the addition product (7) and regenerate the catalyst.
While the present study was underway, two reports10a,b
describing intermolecular amine catalyzed deprotonation of
R-aryl S-trifluoroethyl thioesters (cf. 8f9f10, Scheme 2)
Scheme 1. (a) Citrate Synthase Reaction and (b) Proposed
Biomimetic Asymmetric Mannich Reaction
Scheme 2. Intermolecular Amine-Promoted Deprotonation
base and the thioester R-protons.5,6 As we have previously
suggested,2 use of a thioester also aides soft enolization in
general, because of its enhanced acidity7 relative to other
simple carboxylate derivatives.8 Viewed in this way, depro-
tonation in citrate synthase may be thought of as a form of
soft enolization that results from a productive combination
of favorable thioester acidity and appropriate spatial orienta-
tion of both activating hydrogen bonding functionality to its
carbonyl and a base to its R-protons, such that proximity-
assisted intracomplex deprotonation can occur. We antici-
pated that by mimicking this mode of enolization, particularly
with regard to intracomplex deprotonation, we would be able
to develop a catalytic asymmetric approach to enolate-based
R-alkylation of monocarboxylate-derived substrates, which
remains an unmet goal in the field of organic synthesis.
Herein, we report our initial success through the development
of a biomimetic catalytic asymmetric Mannich reaction9 of
phenylacetic acid thioesters10,11 and sulfonylimines using
cinchona alkaloid-based amino (thio)urea catalysts.
appeared. As part of these studies, the authors identified a
functional pKa barrier of between 16-17 units for enolate
precursors, above which simple equilibrium based deproto-
nation should not be feasible.10a This suggests that use of
simple S-aryl or S-alkyl thioester derivatives of phenyl acetic
acid (pKa g 16.9, cf. 11, Scheme 2), would not be practical
in the context of equilibrium-based deprotonation using an
amine catalyst alone. To overcome this limitation, more
strongly acidic R-aryl S-trifluoroethyl thioesters (pKa < 16.9,
cf. 12) were employed, which indeed enabled amine cata-
lyzed additions. Unfortunately, augmenting reactivity by
relying only on electronic activation of the thioester is
inherently limited. At some point, as the thiol component of
the thioester is made increasingly electron withdrawing,
competing reactions such as ketene formation and acyl
transfer will likely arise, precluding this approach as a general
enolization technique.
In contrast to this approach, our proposed method of
achieving the additional activation needed for enolization
uses a combination of hydrogen bonding and proximity
effects5b,d to enable intracomplex soft enolization (3f4;
Scheme 1b). Not only should this facilitate deprotonation
kinetically relative to intermolecular processes, but the
enolate that forms should also be more stable thermodynami-
cally, due to the cooperative internal stabilization of the
hydrogen bonding and resulting ammonium moieties. Con-
sequently, the ability to use enolate precursors less acidic
The basic design elements of the catalyst (1) we chose to
focus on initially are shown in Scheme 1b. A urea or thiourea
(6) For a review of metal-mediated complex induced proximity effects
in deprotonation, see: Whisler, M. C.; MacNeil, S.; Snieckus, V.; Beak, P.
Angew. Chem., Int. Ed. 2004, 43, 2206–2225
.
(7) Bordwell, F. G.; Fried, H. E. J. Org. Chem. 1991, 56, 4218–4223.
(8) In other studies thioesters were shown to be well-suited to soft
enolization, compared to oxoesters and amides (see ref 2d).
(9) For reviews of organocatalytic Mannich reactions see: Jingjun, M.;
Ning, L.; Qiuhua, W.; Xin, Z.; Xiaohuan, Z.; Chun, W. Prog. Chem. 2008,
20, 76–86. Ting, A.; Schaus, S. E. Eur. J. Org. Chem. 2007, 5797–5815.
(10) For examples of thioesters in organocatalysis, see: (a) Alonso, D. A.;
Kitagaki, S.; Utsumi, N.; Barbas, C. F., III Angew. Chem., Int. Ed. 2008,
47, 4588–4591. (b) Utsumi, N.; Kitagaki, S.; Barbas, C. F., III Org. Lett.
2008, 10, 3405–3408. For Malonic acid half-thioesters, see: (c) Lubkoll,
J.; Wennemers, H. Angew. Chem., Int. Ed. 2007, 46, 6841–6844. (d) Ricci,
A.; Petterson, D.; Bernardi, L.; Fini, F.; Fochi, M.; Herrera, R. P.; Sgarzani,
(12) Recent theoretical studies support the involvement of an enolate
intermediate. See: Yalalov, D. A.; Tsogoeva, S. B.; Shubina, T. E.;
V. AdV. Synth. Catal. 2007, 349, 1037–1040
.
(11) For a review of thioesters in metal-mediated Mannich reactions
Martynova, I. M.; Clark, T. Angew. Chem., Int. Ed 2008, 47, 6624–6628.
see: Benaglia, M.; Cinquini, M.; Cozzi, F. Eur. J. Org. Chem. 2000, 563–
(13) Whether enolate stabilization is the result of a strong hydrogen bond
or an ionic bond is uncertain.
572
.
Org. Lett., Vol. 12, No. 15, 2010
3377

than R-aryl S-trifluoroethyl thioesters should be possible,
thereby opening the door to the development of a general
mode of enolization-based organocatalysis of monocarboxyl-
ate derivatives.
enantiomeric ratio, with the aliphatic thioesters (20 and 12)
reacting enantioselectively (entries 4 and 5). Of these, the
simple S-ethyl thioester 20 gave the best asymmetric induc-
tion. Further examination revealed that catalyst loading could
be reduced to 5 mol %, and that the reaction was complete
within 3 h in the case of 12 (entry 6), but not 20.
The general structure of the catalyst (1) we required for
our work is well-known in the context of bifunctional
catalysts, which have been used to facilitate a variety of
transformations.14 While those reactions proceed in mecha-
nistically different ways not relying on proximity assisted
enolization, the catalysts employed provided us with a
starting point for our own studies. Thus, we began our
investigations into the development of a biomimetic Mannich
reaction by combining sulfonylimine 13 and S-phenyl
thioester 11 in toluene, in the presence of 10 mol % of known
cinchona-derived amino thiourea 1415 (Table 1, entry 1). We
To improve asymmetric induction we examined urea
catalyst 15 (Table 1). Although typically less effective than
their thiourea analogues17 due to reduced hydrogen bonding
capabilities,18 urea catalysts have been reported to possess
superior anion stabilizing properties in some cases.19 In
addition, Wennemers has shown that enolates formed by
decarboxylation of malonic acid half thioesters undergo
conjugate addition to nitroolefins with higher enantioselec-
tivity when a urea rather than a thiourea cinchona-based
catalyst is used.10c Indeed, the enantioselectivity of the
Mannich reaction between sulfonylimine 13 and both
thioester 12 and 20 did increase somewhat with the use of
urea catalyst 15 (entries 7 and 8). To further improve the
enantioselectivity we conducted a cursory examination of
the effect of the aniline component of the catalyst using 16
and 17 (entries 9-12). Interestingly, no difference was found
between catalyst 15 and 16, suggesting that the electronic
properties of the anilinyl group may be unimportant to
asymmetric induction. In contrast, the enantioselectivity
improved when the larger CF3 and methyl groups were
replaced by hydrogen (17), implying that the steric properties
of the aniline system are relevant in this regard.
Table 1. Survey of Conditions for the Biomimetic Mannich
Reaction
thioester
(R)
time
conv
%
entry
cat. mol % (h) syn:anti er (ꢀ:R)
Using catalyst 17, the scope of the reaction with various
sulfonylimines was tested (Table 2). Both electron rich and
1
2
3
4
5
6
7
8
11 (Ph)
18 (4-NO2-C6H4) 14
19 (4-OMe-C6H4) 14
20 (Et)
12 (CH2CF3)
12
20
12
20
12
20
12
14
10
10
10
10
10
5
10
5
10
5
12
12
12
12
12
3
12
3
12
3
93:7
95:5
93:7
91:9
96:4
96:4
92:8
96:4
91:9
92:8
92:8
97:3
44:56
45:55
50:50
78:22
65:35
66:34
82:18
71:29
81:19
71:29
86:14
81:19
86
74
71
54
98
95
48
93
49
98
51
88
14
14
14
15
15
16
16
17
17
Table 2. Biomimetic Mannich Reaction with Various
Sulfonylimines
9
10
11
12
10
5
12
3
sulfonylimine thioester mol %
syn: er yield
were pleased to find that the desired Mannich reaction did
proceed in very good yield.16 Unfortunately, while the syn:
anti selectivity was excellent, the reaction exhibited poor
enantioselectivity. Variation of the thioester across a range
of electron rich and deficient systems (entries 2-5) had little
impact on diastereoselectivity, but significantly affected the
entry
(R) 17 product anti (ꢀ:R)
(R1)
%
1
2
3
4
5
6
7
8
9
13 (Ph)
13
20 (Et)
12 (CH2CF3)
20
5
25
26
27
28
29
30
31
32
33
34
95:5 87:13 77
93:7 81:19 95
92:8 76:24 66
92:8 73:27 94
93:7 85:15 73
91:9 74:26 90
93:7 84:16 41
83:17 76:24 50
98:2 88:12 78
97:3 66:34 85
21 (furanyl)
21
22 (4-Cl-C6H4)
22
20
12
20
12
20
5
20
5
23 (4-OMe-C6H4) 20
23 12
24 (2-Me-C6H4) 20
12
20
5
20
5
(14) See for example: (a) Andres, J. M.; Manzano, R.; Pedrosa, R.
Chemistry 2008, 14, 5116–5119. (b) Tan, K. L.; Jacobsen, E. N. Angew.
Chem., Int. Ed. 2007, 46, 1315–1317. (c) Lubkoll, J.; Wennemers, H. Angew.
Chem., Int. Ed. 2007, 46, 6841–6844. (d) Song, J.; Wang, Y.; Deng, L.
J. Am. Chem. Soc. 2006, 128, 6048–6049. (e) Inokuma, T.; Hoashi, Y.;
Takemoto, Y. J. Am. Chem. Soc. 2006, 128, 9413–9419. (f) Okino, T.;
Hoashi, Y.; Furukawa, T.; Xu, X.; Takemoto, Y. J. Am. Chem. Soc. 2005,
127, 119–125. (g) Wang, J.; Li, H.; Yu, X.; Zu, L.; Wang, W. Org. Lett.
2005, 7, 4293–4296. (h) Wang, J.; Li, H.; Duan, W.; Zu, L.; Wang, W.
Org. Lett. 2005, 7, 4713–4716. (i) McCooey, S. H.; Connon, S. J. Angew.
Chem., Int. Ed. 2005, 44, 6367–6370. (j) Song, J.; Ye, J.; Dixon, D. J.;
Hynes, P. S. Chem. Commun. 2005, 4481–4483. (k) Okino, T.; Hoashi, Y.;
Takemoto, Y. J. Am. Chem. Soc. 2003, 125, 12672–12673.
10 24
electron deficient systems reacted effectively. While the
reactions with S-trifluoroethyl thioester 12 could be con-
ducted using less catalyst and were consistently more rapid
(17) Doyle, A. G.; Jacobsen, E. N. Chem. ReV. 2007, 107, 5713–5743.
(18) Fan, E.; Van Arman, S. A.; Kincaid, S.; Hamilton, A. D. J. Am.
Chem. Soc. 1993, 115, 369–370.
(15) Connon, S. J. Chem. Commun. 2008, 2499–2510. Vakulya, B.;
Varga, S.; Csa´mpai, A.; Soo´s, T. Org. Lett. 2005, 7, 1967–1969.
(16) A similar experiment with the analogous oxoester (phenyl acetic
acid phenyl ester) gave only recovered starting material after 36 h.
(19) See for example: Maher, D. J.; Connon, S. J. Tetrahedron Lett.
2004, 45, 1301–1305. Nam, K. C.; Kang, S. O.; Ko, S. W. Bull. Korean
Chem. Soc. 1999, 20, 953–956. Scheerder, J.; Engbersen, J. F. J.; Casnati,
A.; Ungaro, R.; Reinhoudt, D. N. J. Org. Chem. 1995, 60, 6448–6454.
3378
Org. Lett., Vol. 12, No. 15, 2010

and higher yielding, asymmetric induction was uniformly
better with S-ethyl thioester 20.20
As shown above, despite being significantly less acidic
than S-trifluoroethyl thioester 12, S-ethyl thioester 20 readily
underwent the desired Mannich addition using catalysts
14-17. While the pKa values of compounds 20 and 12 have
not been reported, their differential acidity can be estimated
from the known6,21 pKa of 11, 35 and 36 (Figure 1), along
Figure 1. (a) Acidity of various phenylacetic acid derivatives. (b)
H/D exchange rates (Et3N, CD3OD, toluene-d8).
with the rate of H/D-exchange of the related R-phenyl
propionate derivatives 37-39.22 These data indicate that
O-ethyl ester 36 is less acidic than O-phenyl ester 35 by 4
units, and that the half-life for H/D exchange of S-
trifluoroethyl thioester 37 (T1/2 ) 5.5 h) is considerably less
than that of the corresponding S-phenyl thioester (38; T1/2
) 47 h). Moreover, the pKa of 11 and 38 is expected to be
similar. On this basis, it is reasonable to assume that the
pKa difference between S-trifluoroethyl thioester 12 and
S-ethyl thioester 20 is at least approximately 4 units. As such,
the effective pKa barrier of the present catalytic method is
estimated to be g20.9 units (16.9 + 4), which is well above
the range of 16-17 that has been estimated10a for simple
intermolecular amine catalyzed enolization. Significantly, this
pKa barrier is not that far beyond the acidity range of simple
thioesters.
We next tested the importance of proximity effects in the
Mannich addition reaction (Figure 2). To do so, thioester
12 and sulfonylimine 13 were combined in toluene-d8 along
with: (1) catalyst 17, (2) 40, (3) 41, and (4) 41 plus 1,3-
diphenyl urea, and the reactions were monitored by 1H NMR.
40 is structurally similar to catalyst 17 but is incapable of
hydrogen bonding. The transformation employing 40 showed
only a trace of product after 350 min, by which time the
reaction employing 17 was essentially complete. Compound
41 provided a second control offering similar steric bulk
around the amine as found in 17, which would also not be
capable of hydrogen bonding to the thioester carbonyl. Again,
only a trace of product was formed with this base. The
combination of 41 and 1,3-diphenyl urea showed little to no
increase in reaction rate over 41 alone. The substantial
increase in the rate of the transformation with 17 compared
to the other systems analyzed supports the notion of
proximity-assisted deprotonation.
Figure 2. Evidence of proximity-assisted soft enolization (black
square ) 17; red dash ) 40; blue diamond ) 41; green triangle )
41 + 1,3-diphenyl urea).
In conclusion, we have developed the first organocatalytic
Mannich reaction based on proximity-assisted intracomplex
soft enolization of thioesters using simple derivatives of
known cinchona alkaloid-based catalysts. Significantly, the
cooperative mode of enolization that functions in this case
allows thioesters over a range of acidity to react efficiently.
The functional pKa barrier associated with this method has
been estimated to be g21 units, suggesting its potential in
terms of developing a general enolization-based organocata-
lytic strategy applicable to simple monocarboxylic acid
derivatives. Mechanistic studies of this transformation are
underway, as are studies on the use of different electrophiles
and nucleophiles and the exploration and development of
improved catalysts.
Acknowledgment. M.C.K. holds a Burroughs-Wellcome
fellowship (Department of Chemistry, Duke University);
M.R.G. is supported by the Pharmacological Sciences
Training Program (Duke University). This work was sup-
ported by Duke University and NCBC (2008-IDG-1010).
Supporting Information Available: Experimental pro-
cedures and analytical data for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(20) The lower yields for 20 resulted from incomplete conversion.
(21) Bordwell, F. G.; Fried, H. E. J. Org. Chem. 1981, 46, 4327–4331
.
(22) Um, P. J.; Drueckhammer, D. G. J. Am. Chem. Soc. 1998, 120,
5605–5610.
OL101152B
Org. Lett., Vol. 12, No. 15, 2010
3379