N-prolinylanthranilamide pseudopeptides as bifunctional organocatalysts for asymmetric aldol reactions

Article abstract of DOI:10.1021/ol202284n

Proline anthranilamide-based pseudopeptides were shown to be effective organocatalysts for enantioselective direct aldol reactions of a selection of aldehydes with various ketones with excellent yield, enantioselectivity up to 99% and anti to syn diastere

Full text of DOI:10.1021/ol202284n

ORGANIC
LETTERS
2011
Vol. 13, No. 20
5548–5551
N-Prolinylanthranilamide Pseudopeptides
as Bifunctional Organocatalysts for
Asymmetric Aldol Reactions
Anthony J. Pearson* and Santanu Panda
Department of Chemistry, Case Western Reserve University, Cleveland, Ohio 44106,
United States
ajp4@cwru.edu
Received August 23, 2011
ABSTRACT
Proline anthranilamide-based pseudopeptides were shown to be effective organocatalysts for enantioselective direct aldol reactions of a
selection of aldehydes with various ketones with excellent yield, enantioselectivity up to 99% and anti to syn diastereoselectivity up to 25:1.
Over the past decade or so, interest in organocatalysis
has continued to grow, and much of this interest has
focused on asymmetric aldol reactions,1 the earliest reports
of which used L-proline as the catalyst for intramolecular
reactions,2 later extended to intermolecular processes by
List and co-workers.3 Proline derivatives have since been
popular targets as small molecule organocatalysts. We
recently reported the use of N-prolinylanthranilic acids 1
and 2, which are very effective for aldol reactions of cyclic
ketones with aromatic aldehydes; in most cases, catalyst 2
gives >95% ee.4 During that study, it was found that
methyl substitution ortho to the carboxylic acid (as in 2)
led to superior performance (compared to 1), but methyl
substitution ortho to the anilide (not shown here) did not.
However, for aldol reactions of acetone with, for example,
p-nitrobenzaldehyde, catalyst 2 (80% ee) is only marginally
better than proline itself (70À75% ee). Using the anthra-
nilic acid unit as a convenient platform for structural
variation, we are currently seeking to develop catalysts that
perform well over a broader spectrum of aldol reactions.
One possible structural modification is to extend the an-
thranilic carboxyl terminus by attaching an additional
amino acid to create a pseudotripeptide. We anticipated
that such a structure would show a preferred conformation
that results from the intramolecular hydrogen bonding
shown in Figure 1A, which has been characterized for
anthranilamide oligomers.5 Thus, one can engineer a re-
verse turn that might lead to a minimal binding pocket for
the aldol partners, one of which is temporarily bound
covalently as a prolinyl enamine (Figure 1B). While small
peptides have previously been used as catalysts for aldol6
and other7 reactions, we are not aware of this particular
structural motif being employed. A selection of catalyst
structures that were chosen for evaluation is summarized in
Figure 1C; in addition, we screened the methyl ester of
catalyst 5 (numbered 13 in this report) to assess the role of
(5) (a) Hamuro, Y.; Geib, S. J.; Hamilton, A. D. J. Am. Chem. Soc.
1996, 118, 7529. See also:(b) Huang, B.; Parquette, J. R. J. Am. Chem.
Soc. 2001, 123, 2689. (c) Prabhakaran, P.; Kale, S. S; Puranik, V. G.;
Rajamohanan, P. R.; Chetina, O.; Howard, J. A. K.; Hofmann, H.-J.;
Sanjayan, G. J. J. Am. Chem. Soc. 2008, 130, 17743.
(6) (a) Kofoed, J.; Nielsen, J.; Reymond, J.-L. Bioorg. Med. Chem.
Lett. 2003, 13, 2445. (b) Kattiger, P.; Kovasy, R.; Revell, J. D.; Ivan, S.;
Wennemers, H. Org. Lett. 2005, 7, 1101. (c) Zou, W.; Ibrahem, I.;
Dziedzic, P.; Sunden, H.; Cordova, A. Chem. Commun. 2005, 4946. (d)
Da, C.-S.; Che, L.-P.; Guo, Q. -P.; Wu, F.-C.; Ma, X.; Jia, Y.-N. J. Org.
Chem. 2009, 74, 2541. (e) Davie, E. A. C.; Mennen, S. M.; Xu, Y.; Miller,
S. J. Chem. Rev. 2007, 107, 5759.
(1) Recent reviews: (a) Trost, B. M.; Brindle, C. S. Chem. Soc. Rev.
2010, 39, 1600. (b) Zlotin, S. G.; Kucherenko, A. S.; Beletskaya, I. P.
Russ. Chem. Rev. 2009, 78, 737.
(2) (a) Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39, 1615. (b)
Eder, U.; Sauer, G.; Weichert, R. Angew. Chem., Int. Ed. 1971, 10, 496.
(3) List, B.; Lerner, R. A.; Barbas, C. F., III J. Am. Chem. Soc. 2000,
122, 2395.
ꢀ
ꢀ
(7) (a) Oku, J.; Ito, N.; Inoue, S. Makromol. Chem. 1979, 180, 1089.
ꢀ
(b) Julia, S.; Masana, J.; Vega, J. C. Angew. Chem., Int. Ed. 1980, 19, 929.
(4) Pearson, A. J.; Panda, S. Tetrahedron 2011, 67, 3969.
r
10.1021/ol202284n
Published on Web 09/29/2011
2011 American Chemical Society

the carboxylic acid group as an activator of the electrophilic
aldehyde partner.
Upon irradiation of NH(A) NOE was observed for the CH
(2.6%) and one each of 2Â CH2 (1.1 and 0.8%) proline
residue protons. Enhancement (0.6%) of the aromatic
H ortho to the anilide NH(A) was also observed, but
irradiation of the aromatic proton had no effect on
NH(A), from which we conclude that the NH is not syn
˚
to the aromatic residue (the internuclear distance is 3.7 A
in the anti conformation shown). NOE (1.5%) was ob-
served between NH(A) and NH(B), and between NH(B)
and the isopropyl CH3 as well as the aromatic CH3,
indicating the proximity of these groups.
Figure 1. (A) Intramolecular hydrogen bonding for N-acylan-
thranilamides (dashed line).5 (B) Proposed binding pocket model
for anthranilic acid derived pseudotripeptides. (C) Catalyst struc-
tures investigated.
Synthesis of catalysts 3À13 began with reaction of the
selected amino benzyl ester (methyl ester for 13) with the
corresponding isatoic anhydride8 followed by amide
coupling9 with N-Cbz-L-proline then global deprotection
(details can be found in the Supporting Information). Each
amide coupling product was shown to be optically pure by
chiral HPLC analysis. Catalyst 6 was chosen for NMR
studies to test for intramolecular hydrogen bonding and to
secure data on conformation. The 1H NMR chemical shift
in CDCl3 for the anilide NH was observed at δ10.73 ppm,
which agrees with Hamilton’s data that supports the
proposed intramolecular hydrogen bonding depicted in
Figure 1,5a (see details in Supporting Information). In
contrast, the valinyl amide NH was observed at a higher
field (δ6.53 in chloroform). In addition, the anilide NH
showed a much lower sensitivity to solvent than the valinyl
NH (in d6-DMSO: δ10.88 [Δδ = 0.15] vs, 7.8 [Δδ = 1.3],
respectively). We observed small dependence of the anilide
NH chemical shift on temperature in 20% DMSO in
chloroform, also consistent with this hydrogen bonding
motif.10 A linear shift to a lower field was observed as the
temperature was decreased, with a temperature coefficient
(Δδ/ΔT) of 4.6 ppb/K for the anilide NH(A) versus
10.2 ppb/K for the valinyl NH(B) (see Figure 2 for NH
labeling).5a,10 These observations support the proposed in-
tramolecular hydrogen bonding for NH(A) but not NH(B),
as depicted in Figure 1A. NOE difference experiments in
CDCl3 are consistent with the conformation shown in
Figure 2 (full details are in the Supporting Information).
Figure 2. Selected NOE difference results for catalyst 6 in CDCl3.
NOE signals were not observed between the aromatic
proton ortho to NH(A) and any proton in the proline
nucleus. Assuming this conformation persists over the
entire aldol reaction profile, we expect significant connec-
tivity between enamine, aldehyde, and carboxylic acid in
the key transition state.
The aldol reaction between p-nitrobenzaldehyde and
acetone is a commonly used benchmark to study organo-
catalytic aldol reactions. Optimized reaction conditions
that were developed in our earlier work4 were used for the
present study, and the results are presented in Table 1.
Catalyst 2 was superior to catalyst 1, which is consistent
with our earlier observation.4 Modification of the acid side
chain started with the introduction of glycine, which
appears to be effective when comparing catalyst 3 vs 1.
Introducing a 6-methyl group, as in catalyst 4, markedly
improves the ee vs catalyst 3, possibly due to conforma-
tional bias that further favors the desired trans amide
structure. Modification of catalyst structure by introdu-
cing a second stereogenic center (catalysts 5 and 6) further
improved the yield and ee. However, the change compared
with 3 or 4 was rather modest, so we examined D-valine as a
subunit (catalysts 7 and 8) to test for possible mismatching
of chirality. A negligible change in enantioselectivity re-
sulted from this modification (entries 5 vs 7, and 6 vs 8).
Introduction of a serine unit for possible reinforced hydro-
gen bonding was not effective (entry 9). Also, introduction
of a phenylalanine unit that might show a pi-stacking
interaction with the aromatic aldehyde did not lead to
any improvement (entry 10). The fact that both isomers of
valine provide better catalysts might be due to a conforma-
tional biasing that results from alpha substitution, related
(8) Varnavas, A.; Lassiani, L.; Valenta, V.; Berti, F.; Tontini, A.;
Mennuni, L.; Makovec, F. Eur. J. Med. Chem. 2004, 39, 85.
ꢀ ꢀ
(9) Quelever, G.; Burlet, S.; Garino, C.; Pietrancosta, N.; Laras, Y.;
Kraus, J. L. J. Comb. Chem. 2004, 6, 695.
(10) Medda, A. K.; Park, C. M.; Jeon, A.; Kim, H.; Sohn, J. H.; Lee,
H.-S. Org. Lett. 2011, 13, 3486.
Org. Lett., Vol. 13, No. 20, 2011
5549

On the other hand, the aldol reaction between o-nitroben-
zaldehyde and cyclopentanone afforded anti/syn 5:1, 98%
ee (Table 2, entry 10), which compares very favorably with
Table 1. Optimization of Organocatalytic Aldol Reaction of p-
Nitrobenzaldehyde and Acetone
Table 2. Organocatalytic Aldol Reaction of Selected Aromatic
Aldehydes and Ketonesa
entrya
catalyst
yield (%)b
ee (%)c
1
1
50
50
55
58
55
55
60
58
45
52
55
58
28
60
80
69
81
77
87
78
85
68
70
74
83
63
2
2
3
3
entry
R1,R2
H,H
R3
yield (%)b
drc
ee (%)d
4
4
1
4-NO2
55
65
85
50
38
76
70
84
78
70
68
52
NA
NA
NA
NA
NA
9:1
87
92
90
91
91
99
97
92
99
98
93
88
5
5
2
H,H
2-NO2
6
6
3
H,H
2,4-(NO2)2
2,4-(Cl)2
4-Br
7
7
4
H,H
8
8
5
H,H
9
9
6
-(CH2)3-
-(CH2)3-
-(CH2)3-
-(CH2)2-
-(CH2)2-
-CH2SCH2-
4-NO2
10
11
12
13
10
11
12
13
7
2-NO2
25:1
12:1
1.2:1
5:1
8
4-pyridyle
4-NO2
9
10
11f
12f
2-NO2
a Unless specified otherwise, the concentration of aldehyde is 0.13 M,
and v/v of acetone/DMSO is (1/4), 10 mol % of catalyst, 5 mol % of
TFA, 550 mol % of water and the reactions were run at 4 °C. b Isolated
yield after separation from unreacted starting materials and catalyst.
c Determined by HPLC on CHIRALPAK AD-H column for isolated
products after separation from unreacted starting materials and catalyst.
4-NO2
10:1
9:1
-OC(CH3)2O- 4-NO2
a All reactions were run under the general conditions noted for
Table 1. b Isolated yield after separation from unreacted starting materi-
als and catalyst. c Diastereomer ratios (anti/syn) were estimated from the
1H NMR spectrum of the crude product mixture. d Determined by
HPLC on CHIRALPAK AD-H column for isolated products after
separation from unreacted starting materials and catalysts. e Pyridine-4-
al was the aldehyde substrate. f Reaction time was 5 days.
to the well-known ThorpeÀIngold effect.11 However, in-
troduction of a cyclopentane group as a conformational
lock did not give significant improvement; again, 6-methyl
substitution improves the ee (catalyst 12 vs 11). The role of
the acid group was also evaluated using the methyl ester
(13) of catalyst 5 (entry 13), which reacts slowly with
reduced enantioselectivity compared with its acid counter-
part 5 (entry 3).
Direct aldol reactions using various aldehydes and
ketones were explored with catalyst 6 (Table 2). In most
cases, the reaction afforded anti product with high yield
and high diastereo- and enantioselectivity. The aldol reac-
tion between o-nitrobenzaldehyde and acetone afforded
92% ee and good yield, an improvement over the reaction
with p-nitrobenzaldehyde, as anticipated for a more steri-
cally hindered aldehyde. High enantioselectivity was also
observed for 2,4-dinitrobenzaldehyde with an excellent
yield (entry 3). Using p-bromo- and 2,4-dichlorobenzalde-
hyde, the aldol product was obtained with comparable
yield and good enantioselectivity (entries 4 and 5). The
aldol reaction of cyclohexanone and p-nitrobenzaldehyde
afforded 99% ee with good yield and diastereoselectivity,
and with o-nitrobenzaldehyde afforded 25:1 dr, 97% ee
and good yield, the sterically hindered aldehyde giving
better diastereoslectivity. Aldol reactions with cyclopentanone
afforded excellent enantioselectivity but moderate diastereos-
electivity, as observed for other prolinamide catalysts.1,12
other catalysts reported in the literature.12 The aldol
reaction between dihydro-2H-thiopyran-4(3H)-one (a use-
ful surrogate for 3-pentanone) and p-nitrobenzaldehyde
gave the anti product with somewhat variable diastereos-
electivity but uniformly high ee for the anti product (entry
11). Using the acetonide of 1,3-dihydroxyacetone as the
nucleophile, the anti product was obtained with good ee
and diastereoselectivity (entry 12). Thus, with catalyst 6,
the enantioselectivity of the acetone aldol reactions were
greatly improved while not compromising the aldol reac-
tions of cyclic ketones.
Aldol reactions between acetone and substituted isatin
derivatives were next explored (Table 3). The resulting
3-substituted-3-hydroxyindolin-2-ones are desirable tar-
gets due to related structural features found in natural
products and drug candidates.13 Although this reaction
has been reported in the literature,14 only in very few cases
were high ee’s obtained.15 The aldol reaction of isatin and
(13) (a) Peddibhotla, S. Curr. Bioact. Compd. 2009, 5, 20. (b) Galliford,
C. V.; Scheidt, K. A. Angew. Chem., Int. Ed. 2007, 46, 8748. (c) Marti, C.;
Carreira, E. M. Eur. J. Org. Chem. 2003, 2209. (d) Zhou, F.; Liu, Y. L.;
Zhou, J. Adv. Synth. Catal. 2010, 352, 1381.
(14) (a) Luppi, G.; Cozzi, P. G.; Monari, M.; Kaptein, B.; Broxterman,
Q. B.; Tomasini, C. J. Org. Chem. 2005, 70, 7418. (b) Angelici, G.; Correa,
R. J.; Garden, S. J.; Tomasini, C. Tetrahedron Lett. 2009, 50, 814.
(15) (a) Nakamura, S.; Hara, N.; Nakashima, H.; Kubo, K.; Shibata,
N.; Toru, T. Chem.;Eur. J. 2008, 14, 8079. (b) Malkov, A. V.;
(11) Beesley, R. M.; Ingold, C. K.; Thorpe, J. F. J. Chem. Soc., Trans.
1915, 107, 1080.
(12) (a) Lei, M.; Shi, L.; Li, G.; Chen, S.; Fang, W.; Ge, Z.; Cheng, T.;
Li, R. Tetrahedron 2007, 63, 7892. (b) Giacalone, F.; Gruttadauria, M.;
Meo, P. L.; Riela, S.; Noto, R. Adv. Synth. Catal. 2008, 350, 2747.
ꢀꢁ
ꢀ
Kabeshov, M. A.; Bella, M.; Kysilka, O.; Malyshev, D. A.; Pluhackova,
ꢁ
K.; Kocovsky, P. Org. Lett. 2007, 9, 5473. (c) Guo, Q.; Bhanushali, M.;
Zhao, C.-G. Angew. Chem., Int. Ed. 2010, 49, 9460.
5550
Org. Lett., Vol. 13, No. 20, 2011

acetone catalyzed by L-proline has been reported to favor
the (S) enantiomer,16 whereas prolinamides generally af-
ford the (R) enantiomer as the major product.14a,15a,15c
With catalyst 6, up to 84% ee of the (R) enantiomer and
100% conversion were obtained. This result was improved
to 97% ee after a single recrystallization. Interestingly,
while catalyst 2 works well for the aldol reaction between
acetone and p-nitrobenzaldehyde, it gives very poor en-
antioselectivity for the reaction between 5-bromoisatin
and acetone (Table 3, entry 2). The considerable improve-
ment using catalyst 6, while not compromising other aldol
reactions, provides a compelling reason for investigating
structural variations based on the anthranilamide scaffold.
Catalysts substituted with a 6-methyl group again afforded
higher enantioselectivity than the unsubstituted analogues
(entries 1, 3, 5 vs 2, 4, 6), the effect being even greater than
in the earlier aldol reactions. Catalyst 6 was chosen for
further optimization. Lowering the catalyst loading to
5 mol % improved the enantiomeric excess without com-
promising the yield of the product (entry 7), while the ee
was further improved by reducing the amount of acetone
used in the reaction (entry 8).These optimized conditions
were used to explore the reactions between acetone and
other substituted isatin derivatives. Good enantioselecti-
vity was obtained using isatin itself as an aldol acceptor
(entry 12). 5-Substituted isatin derivatives reacted with
somewhat better enantioselectivity than 4- and 6-substi-
tuted derivatives (entries 8À11 vs 13 and 14). Reaction of
4,6-dibromoisatin with acetone yielded (S)-convolutamy-
dine A in quantitative yield and reasonable enantiomeric
excess, which was considerably improved by crystal-
lization (entry 15). This compound exhibits potent in-
hibitory activity toward the differentiation of HL-60
leukemia cells.17
modifications that might also lead to enhanced selectivity,
suchasvariation of the substituent at C-6 ofthe anthranilic
acid moiety.
Table 3. Organocatalytic Aldol Reactions of Acetone and Isatin
Derivativesa
entry
catalyst
R1,R2,R3
conversion (%)b
ee (%)c
1
1
H,Br,H
H,Br,H
H,Br,H
H,Br,H
H,Br,H
H,Br,H
H,Br,H
H,Br,H
H,F,H
85
5
2
2
100
18
3
5
100
32
4
6
100
77
5
11
12
6
100
35
6
100
76
7d
8e
100
82
6
100 (62)f
100 (50)f
100 (58)f
100 (50)f
90 (48)f
100 (60)f
90 (52)f
100 (65)f
84 (97)g
80 (92)g
82 (95)g
81 (95)g
70 (89)g
74 (90)g
69 (85)g
76 (91)g
9e
6
10e
11e
12e
13e
14e
15e
6
H,Cl,H
H,CH3,H
H,H,H
H,H,Br
Br,H,H
Br,H,Br
6
6
6
6
6
a Unless specified otherwise, the concentration of the isatin is 0.16 M,
and v/v of acetone/DMSO is (1/5), 10 mol % of catalyst, 5 mol % of
b
TFA, 550 mol % of water and the reactions were run at 4 °C.
%
Conversion was determined from the crude 1H NMR spectrum. c De-
termined by HPLC on CHIRALPAK AD-H column for isolated
products after separation from unreacted starting materials and catalyst.
d 5 mol % of catalyst was used. e 5 mol % of catalyst was used,
concentration of the isatin was 0.18 M, and v/v of acetone/DMSO is
(1/10), 5 mol % of TFA, 550 mol % of water and the reactions were run
at 4 °C. f Isolated yield after recrystallization. g Enantiomeric excess
determined after recrystallization.
In conclusion, a series of N-prolinylanthranilamide
pseudopeptides were designed as catalysts for enantiose-
lective aldol reactions. The anthranilic acid core represents
a very convenient template for structural engineering, in
the present case modification of the carboxylic acid by
incorporation of an amino acid residue, which was found
to be beneficial for a number of important aldol reactions.
Future studies will focus on addressing other structural
Acknowledgment. We are grateful to Dr. Dale Ray of
the Department of Chemistry, Case Western Reserve
University for assistance with the 2D-NMR experiments.
Supporting Information Available. Spectroscopic data,
experimental details for the preparation of catalysts,
NOE data and HPLC data. This material is avail-
able free of charge via the Internet at http://pubs.acs.
org.
(16) Correa, R. J.; Garden, S. J.; Angelici, G.; Tomasini, C. Eur. J.
Org. Chem. 2008, 736.
(17) Kamano, Y.; Zhang, H.-P.; Ichihara, Y.; Kizu, H.; Komiyama,
K.; Pettit, G. R. Tetrahedron Lett. 1995, 36, 2783.
Org. Lett., Vol. 13, No. 20, 2011
5551