Asymmetric Michael addition of α-nitro-ketones using catalytic peptides

Article abstract of DOI:10.1016/j.tetlet.2007.01.073

Peptide-based catalysts have been developed that promote the asymmetric Michael addition of nitroalkanes. The most effective peptides contain a β-turn structural element as well as a basic histidine and an arylsulfonamide-protected arginine. Excellent yields with enantioselectivities of up to 74% ee have been observed.

Full text of DOI:10.1016/j.tetlet.2007.01.073

Tetrahedron Letters 48 (2007) 1993–1997
Asymmetric Michael addition of a-nitro-ketones
using catalytic peptides
Brian R. Linton,a,* Michael H. Reutershan,a Christopher M. Aderman,a
Elizabeth A. Richardson,a Kristen R. Brownell,a Charles W. Ashley,a
Catherine A. Evansb and Scott J. Millerb,c
aDepartment of Chemistry, Bowdoin College, Brunswick, ME 04011, United States
bDepartment of Chemistry, Merkert Chemistry Center, Boston College, Chestnut Hill, MA 02467, United States
cDepartment of Chemistry, Yale University, New Haven, CT 06520, United States
Received 22 November 2006; revised 11 January 2007; accepted 15 January 2007
Available online 18 January 2007
Abstract—Peptide-based catalysts have been developed that promote the asymmetric Michael addition of nitroalkanes. The most
effective peptides contain a b-turn structural element as well as a basic histidine and an arylsulfonamide-protected arginine. Excellent
yields with enantioselectivities of up to 74% ee have been observed.
Ó 2007 Elsevier Ltd. All rights reserved.
Nitroalkanes are valuable synthetic building blocks for
the formation of new carbon–carbon bonds, as well as
precursors to numerous other functional groups.1 The
electron-withdrawing nature of the nitro group facili-
tates removal of the alpha proton under mild condi-
tions, and the resulting conjugate base can serve as a
nucleophile in a variety of transformations, such as
nitroaldol (Henry) reactions and Michael additions.
Reduction of the nitro groups to amines or hydrolysis
to carbonyl derivatives can provide access to a wide
variety of synthetic products.
We sought to employ small peptides as catalysts in
these reactions, with the hypothesis that hydrogen-
bonding amino acid side chains could aid in transition
state organization.8,9 Peptides have proven to be effec-
tive catalysts for a range of chemical reactions, and
provide quick access to libraries that can be tailored
to individual synthetic transformations.10 A related
peptide-catalyzed conjugate addition of azide ion11
likely takes advantage of hydrogen-bonding in the
transition state.12
The search for peptide-based catalysts capable of pro-
moting the Michael addition of nitroalkanes began with
a targeted design strategy. Pentapeptides with a propen-
sity for b-turn nucleation were utilized to provide the
structural rigidity necessary to control facial selectiv-
ity.13 A histidine residue was incorporated to serve as
a mild base to deprotonate the nitroalkane. Additional
hydrogen-bonding functionality was included to provide
potential binding sites that might lead to transition state
stabilization. A series of peptides were screened using
the Michael addition shown in Eq. 1 (R1 = Ph, R2 =
R3 = Me). The catalysts were prepared through Fmoc-
solid-phase peptide synthesis, and were screened directly
after cleavage from the resin, without further purifica-
tion. Systematic variations were introduced in an effort
to explore their impact on selectivity and to probe mech-
anistic issues. The most selective peptide 1 was found to
Hydrogen-bonding interactions have previously been
shown to modulate the regiochemistry of nitroalkane
alkylation.2 There is a growing body of literature that
suggests that these interactions can also orchestrate effi-
cient enantioselectivity in a variety of catalytic reac-
tions.3 Enantioselectivity in the reactions of
nitroalkane nucleophiles has been demonstrated using
hydrogen-bonding receptors4 and Cinchona alkaloids.5
High levels of selectivity have also been observed with
proline derivatives,6 and a variety of metal-based
catalysts.7
Keywords: Asymmetric catalysis; Michael addition; Hydrogen-
bonding.
*
Corresponding author. Tel.: +1 207 725 3616; fax: +1 207 725 3017;
e-mail: blinton@bowdoin.edu
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.01.073

1994
B. R. Linton et al. / Tetrahedron Letters 48 (2007) 1993–1997
include a D-Pro-Aib turn element, a benzyl-histidine,
and an Pbf-protected arginine.
However, other modifications indicated the subtlety of
transition state organization required for maximal effect.
Deprotection of the arginine side chain to produce a
peptide with a cationic arginine guanidinium (entry 5;
as the BF4 salt) produced only racemic products. While
one would expect the deprotected functional group to
exhibit stronger binding to the intermediate nitronate
than the neutral guanidine of peptide 1,2 it is possible
that strong complexation with the side chain competes
with effects of other parts of the peptide. Other more
subtle variations of the protecting group on the arginine
side chain also led to an elimination of selectivity.
Replacement of the (Pbf) with the analogous (Pmc)
showed similar activity, but nitro and bis(Boc)-protected
arginines both produced racemic products (entries 6, 7).
It is currently unclear if this is due to changes in the
hydrogen-bonding capacity of the side-chain guanidine
or if the aromatic sulfonamide plays a role in selectivity.
Elongation of the side chain by one methylene using
homoarginine (entry 8) produced racemic products as
well.
O
O
O
O
R3
NO2
R1
R1
R3
R2 NO2
R2
O
Me
Me
Me
NH2
O O
Me
N
S
Me
Me
N
N
H
N
H
O
O
HN
O
Me
O
O
NH
O
Bn
N
N
NH
Me
Bn
1
MeO
ð1Þ
Analysis of select members of the peptide library pro-
vided insight into those aspects of the catalyst that are
important for selectivity and the mechanistic role played
by each residue (Table 1). The first two peptides differed
only in the stereochemistry of the proline residue (entries
1, 2), and while both show selectivity in the Michael
addition, each favors a different stereoisomer.14,15 Fur-
ther modification of the turn elements only reduced
selectivity. A significant structural change was intro-
duced by transposing the arginine and histidine resi-
dues.16 This exchange of residues led to an increase in
selectivity for the D-proline peptide 1, but produced
racemic products for the L-isomer (entry 3). Also of note
was the observation that the direction of the selectivity
for peptide 1 was reversed in comparison with entry 2,
where the residues were transposed. This suggests a
direct structural role for one or both residues in the
organization of the transition state.
Other hydrogen-bonding side chains were employed in
an attempt to clarify the role of hydrogen bonds in selec-
tivity. Urea and thiourea are analogs of the arginine
side-chain guanidine, but should differ in their hydro-
gen-bonding strength.17,18 Their incorporation into pep-
tides with the same spacing as peptide
1 was
accomplished through modification of an ornithine side
chain with an isocyanate and isothiocyanate, respec-
tively. The urea peptide (entry 9) showed no selectivity,
while the thiourea peptide 2 (entry 10) produced prod-
ucts with 14% ee. Although this is no advance on the
24% ee realized with peptide 1, the greater selectivity
of thiourea over urea is consistent with stronger associ-
ation to oxyanions.17 This result may indicate that
hydrogen bond strength plays an important role in the
transition state.
The role of the arginine residue was explored through
either its deletion, or variation of its position in the cat-
alyst sequence. Replacement of the arginine with a phen-
ylalanine produced only racemic products as expected.
O
Me
S
Me
N
N
N
N
H
O
H
H
HN
O
O
NH
O
Bn
Table 1. Selected peptide catalysts and selectivity observed for the
N
Michael addition shown in Eq. 1 (R1 = Ph, R2 = R3 = Me)
N
NH
Me
Bn
2
Entry Catalyst
ee
12
O
1
2
Ac-His(Bn)-L-ProAib-Arg(Pbf)-Gly-OMe
MeO
Ac-His(Bn)-D-ProAib-Arg(Pbf)-Gly-OMe
Oct-Arg(Pbf)-L-Pro-Aib-His(Bn)-Phe-OM
Oct-Arg(Pbf)-D-Pro-Aib-His(Bn)-Phe-OMe (1)
Oct-Arg(H+)-D-Pro-Aib-His(Bn)-Phe-OMe
Oct-Arg(NO2)-D-Pro-Aib-His(Bn)-Phe-OMe
Oct-Arg(Boc)2-D-Pro-Aib-His(Bn)-Phe-OMe
Oct-hArg(Pmc)-D-Pro-Aib-His(Bn)-Phe-OMe
Oct-Orn(CONHEt)-D-Pro-Aib-His(Bn)-Phe-OMe
ꢀ6
3
4
5
6
7
8
9
0
Histidine was included in the peptide design to initiate
the chemical reaction, but it appears to play a role in
determining selectivity as well. To test the role of histi-
dine, a peptide was created that was identical to peptide
1 except that the histidine was replaced by an alanine.
Without the basic imidazole only starting materials were
recovered. Addition of equal amounts (0.02 mol equiv)
of this control peptide and N-methyl imidazole (NMI)
led to complete conversion to products, but with no
observed selectivity (entry 11). This suggests that the
histidine residue does not merely deprotonate the
nitroalkane, but also plays a role in the selectivity of
the subsequent nucleophilic addition. It seems likely that
24
0
0
0
0
0
10
11
12
13
14
Oct-Orn(CSNHEt)-D-Pro-Aib-His(Bn)-Phe-OMe (2) 14
Oct-Arg(Pbf)-D-Pro-Aib-Ala-Phe-OMe + NMI
Oct-Arg(Pbf)-D-Pro-Aib-His(Trt)-Phe-OMe
Boc-His(p-Me)-D-ProAib-Arg(Pbf)-Gly-OMe
Boc-Ala(3-pyridyl)-D-ProAib-Arg(Pbf)-Gly-OMe
0
0
ꢀ4
ꢀ6
All reactions were conducted at 4 °C. Selectivity was determined by
chiral HPLC.

B. R. Linton et al. / Tetrahedron Letters 48 (2007) 1993–1997
1995
this is accomplished through association with the inter-
mediate nitro-enolate anion. The benzyl side chain on
the histidine is not thought to play a significant role
since peptides that contain a p-methyl-histidine or even
a 3-pyridylalanine showed selectivities similar to their
s-benzyl-histidine analogs, although the increased bulk
of a trityl histidine led to racemic products (entries 12–
14).19
greater selectivity, but further increasing the steric bulk
provided a less selective reaction (entry 3). A cyclic
nitroketone analog, 2-nitroindanone, showed nearly
racemic products in comparison with the acyclic version,
suggesting that the ideal conformation is s-trans.
Replacement of the phenyl group with the cyclohexyl
analog (entry 4) or an ethyl ester (entry 5) resulted in
complete loss of selectivity. The peptide catalyst was,
however, capable of performing selective reactions of
t-butyl 2-nitropropionate with phenyl vinyl ketone as
the electrophile (entry 6). Phenyl vinyl ketone was the
most effective electrophile with nitroesters, but reaction
with nitroketones produced only racemic products. Phen-
ylketone substrates demonstrated reduced synthetic
yields with increasing alkyl chain length, while more
reactive nitroesters and cyclohexylketone showed excel-
lent yields.
The selectivity of peptide 1 can be modified by optimiz-
ing reaction conditions. Lowering substrate concentra-
tions to 100 mM or below indeed had a positive effect,
but the expense of reactivity, so 100 mM was used for
all subsequent experiments. Toluene was the best solvent
with a self-consistent comparison producing 40% ee,
while benzene (30% ee), diethyl ether (18% ee) and hex-
anes (racemic with peptide precipitation) showed re-
duced selectivity. The peptides are minimally soluble in
toluene, thus the catalysts were introduced into the reac-
tion medium as dichloromethane solutions. The final
concentration did not exceed 3% v/v dichloromethane
in toluene. Maximal selectivities were observed at 4 °C.
Electronic effects were investigated by comparing 2-
nitropropiophenone with para-methoxy and para-nitro
derivatives at 250 mM substrate. (At lower concentra-
tions, the nitrophenylketone failed to react.) Under
these conditions, the methoxyphenylketone produced
10% ee, the phenylketone 24% ee, and the nitrophenyl-
ketone 40% ee. These results suggest that it is the
nucleophilic attack that is responsible for selectivity.
The selectivity and synthetic flexibility of peptide 1 was
further probed by exploring substrates with differing
functional groups and steric bulk (Table 2). Nitroketone
derivatives were synthesized from acyl imidazoles and
nitronate salts,20 while nitroester derivatives were ob-
tained by nitration of the alpha-bromoesters.21 All as-
pects of the substrates were found to result in changes
in selectivity. An increase in the size of the alpha-side
chain (R2) from methyl to ethyl (entries 1, 2) showed
Kinetic analysis for several substrates all showed that
the peptide-catalyzed reaction was >10 times faster than
with N-methyl imidazole (Fig. 1). This peptide-catalyzed
rate acceleration could result from various effects. The
deprotonated anion could be stabilized by interactions
Table 2. Selectivity and isolated yields observed with various nitrocarbonyls using peptide 1 (2 mol %) as a catalyst
Entry
Substrate
Electrophile
Yield (%)
ee (%)
52
O
O
1
82
NO2
NO2
NO2
NO2
NO2
NO2
O
O
O
O
O
O
O
2
3
4
5
6
64
29
99
96
85
74
60
0
O
O
0
O
O
O
50
Ph
All reactions were conducted at 4 °C. Selectivity was determined by chiral HPLC.

1996
B. R. Linton et al. / Tetrahedron Letters 48 (2007) 1993–1997
H
100
90
80
70
60
50
40
30
20
10
0
O
δ
H
H
N
N
R1
Pbf
O
NO2
N
O
N
δ
δ
O
O
R2
H
O
N
C7H15
NO2
O
Phe
N
H
O
HH
O
N
O
Bn
N
H
H
O
N
N
O
O
NO2
200
Figure 2. A possible assignment of side chain roles.
0
50
100
time (hrs)
150
of catalytic peptides for other important carbon-bond
forming reactions.
Figure 1. Kinetic comparison of select substrates with either peptide 1
(closed symbols) or N-methylimidazole (open symbols). Substrates
include ethyl ester (d, s), cyclohexyl ketone (m, n), and phenyl
ketone (j, h). All reactions were performed at 4 °C.
Acknowledgments
This work was supported by an award from Research
Corporation. B.R.L. would like to thank Bowdoin Col-
lege for a Porter Fellowship for Advanced Study and
Research, and thanks to the National Science Founda-
tion (MRI-0116416) for support of this work (CHE-
0639069).
with the peptide, but the results above suggest that
nucleophilic attack is rate determining. Two possibilities
that are consistent with rate acceleration are direct acti-
vation of the electrophile, or assembly of the transition
state on the peptide catalyst. Additionally, the rate of
reaction for nitroesters (entry 5) and cyclohexylketones
(entry 4) is faster than the analogous phenylketone (en-
try 1), suggesting that the greater selectivity of entry 1
may be correlated with reduced starting material
reactivity.
Supplementary data
Experimental procedures and product characterizations
for all new compounds synthesized. Supplementary data
associated with this article can be found, in the online
version, at doi:10.1016/j.tetlet.2007.01.073.
Synthetic and kinetic data suggest a mechanistic role for
peptide side chains in catalytic selectivity. Both arginine
and histidine must be part of the peptide and their trans-
position changes the direction of the observed selectiv-
ity, suggesting interactions with both side chains in the
transition state. Greatest selectivity is observed for the
D-Pro-Aib turn element, which promotes a b-turn and
orients the amide NH oriented towards the same face
as the histidine and arginine side chains.22 Rate acceler-
ations are consistent with the preorganization of nucleo-
phile and electrophile on the peptide. The low
selectivity of the cyclic nitroindanone and reduced yields
for increased alkyl chain length (R2) suggest that the
nucleophile could interact with the peptide optimally
when the carbonyl and nitro groups are in an s-trans
configuration. One possibility is the accumulation of
hydrogen-bonding interactions between the peptide
and anionic intermediate (Fig. 2), and our ability to fully
understand and control these interactions will provide
the key to maximizing selectivity in this and other
reactions.
References and notes
1. Ono, N. The Nitro Group in Organic Synthesis; Wiley-
VCH: New York, NY, 2001; Rosini, G.; Ballini, R.
Synthesis 1988, 833–847; Barrett, A. G. M.; Graboski, G.
Chem. Rev. 1986, 86, 751–762; Seebach, D.; Colvin, E. W.;
Lehr, F.; Weller, T. Chimia 1979, 33, 1–18.
2. Linton, B. R.; Goodman, M. S.; Hamilton, A. D. Chem.
Eur. J. 2000, 6, 2449–2455.
3. For a review, see: Taylor, M. S.; Jacobsen, E. N. Angew.
Chem. Int. Ed. 2006, 45, 1520–1543.
´
´
´
4. Chinchilla, R.; Najera, C.; Sanchez-Agullo, P. Tetra-
hedron: Asymmetry 1994, 5, 1393–1402; Okino, T.;
Nakamura, S.; Takemoto, Y. Org. Lett. 2004, 6, 625–627.
5. (a) Wynberg, H.; Heider, R. Tetrahedron Lett. 1975, 46,
4057–4060; (b) Latvala, A.; Stanchev, S.; Linden, A.;
Hesse, M. Tetrahedron: Asymmetry 1993, 4, 173–176; (c)
´
´
Vakulya, B.; Varga, S.; Csampai, A.; Soos, T. Org. Lett.
2005, 7, 1967–1969.
6. Yamaguchi, M.; Igarashi, Y.; Reddy, R. S.; Shiraishi, T.;
Hirama, M. Tetrahedron 1997, 53, 11223–11236; Halland,
N.; Hazell, R. G.; Jørgensen, K. A. J. Org. Chem. 2002,
67, 8331–8338.
7. Keller, E.; Veldman, N.; Spek, A. L.; Feringa, B. L.
Tetrahedron: Asymmetry 1997, 8, 3403–3413; Funabashi,
K.; Saida, Y.; Kanai, M.; Arai, T.; Sasai, H.; Shibasaki,
M. Tetrahedron Lett. 1998, 39, 7557–7558; Itoh, K.;
Kanemasa, S. J. Am. Chem. Soc. 2002, 124, 13394–13395.
8. Schreiner, P. R. Chem. Soc. Rev. 2003, 32, 289–296.
9. Miller, S. J. Acc. Chem. Res. 2004, 37, 601–610.
In summary, catalytic peptides have been developed that
promote the asymmetric Michael addition of nitrocar-
bonyl compounds. Methodologically, this approach
provides a mild, asymmetric method for the creation
of new carbon–carbon bonds that may be complemen-
tary to other organocatalytic and metal-based
approaches. These results also demonstrate a mecha-
nism-driven catalytic design predicated on bifunctional
catalysis. These studies set the stage for further design

B. R. Linton et al. / Tetrahedron Letters 48 (2007) 1993–1997
1997
10. For a collection of articles on this theme, see: Biopolymers
(Pept. Sci.) 2006, 84, special Issue (Toniolo, C.; Miller,
S. J., Guest Eds.).
17. Linton, B. R.; Goodman, M. S.; Fan, E.; van Arman,
S. A.; Hamilton, A. D. J. Org. Chem. 2001, 66, 7313–
7319.
11. Guerin, D. J.; Miller, S. J. J. Am. Chem. Soc. 2002, 124,
2134–2136.
12. Harriman, D. J.; Deslongchamps, G. J. J. Comp.-Aided
Mol. Des. 2004, 18, 303–308.
13. Haque, T. S.; Little, J. C.; Gellman, S. H. J. Am. Chem.
Soc. 1996, 118, 6975–6985; Venkatraman, J.; Shan-
karamma, S. C.; Balaram, P. Chem. Rev. 2001, 101,
3131–3152.
18. Urea and thiourea have been used in a variety of
asymmetric catalysts (see Refs. 3 and 5c), including those
that promote Michael additions to nitroalkenes: Cao,
C.-L.; Ye, M.-C.; Sun, X.-L.; Tong, Y. Org. Lett. 2006, 8,
2901–2904; Okino, T.; Hoashi, Y.; Takemoto, Y. J. Am.
Chem. Soc. 2003, 125, 12672–12673; Asymmetric Strecker
reactions: Vachal, P.; Jacobsen, E. N. J. Am. Chem. Soc.
2002, 124, 10012–10014.
14. For early illustrations of proline showing this effect, see:
Jarvo, E. R.; Copeland, G. T.; Papaioannou, N.; Boni-
tatebus, P. J., Jr.; Miller, S. J. J. Am. Chem. Soc. 1999,
121, 11638–11643.
15. The absolute configuration of the major enantiomer has
not been established so far.
19. For synthetic reasons the histidines in entries 13 and 14
were included at the N-termini, and are quite similar to
entry 2.
20. Baker, D. C.; Putt, S. R. Synthesis 1978, 478–479.
21. Kornblum, N.; Blackwood, R. K.; Powers, J. W. J. Am.
Chem. Soc. 1957, 79, 2507–2509.
16. The N-terminal octanoyl chain was introduced to increase
solubility in organic solvents.
22. Vasbinder, M. M.; Jarvo, E. R.; Miller, S. J. Angew.
Chem. Int. Ed. 2001, 40, 2824–2827.