Kinetic resolution of alcohols catalyzed by tripeptides containing the N-alkylimidazole substructure [6]

Full text of DOI:10.1021/ja973892k

J. Am. Chem. Soc. 1998, 120, 1629-1630
Scheme 1
1629
Kinetic Resolution of Alcohols Catalyzed by
Tripeptides Containing the N-Alkylimidazole
Substructure
Scott J. Miller,* Gregory T. Copeland, Nikolaos Papaioannou,
Thomas E. Horstmann, and Elizabeth M. Ruel
Department of Chemistry, Eugene F. Merkert
Chemistry Center, Boston College
Chestnut Hill, Massachusetts 02167-3860
ReceiVed NoVember 13, 1997
Peptide-substrate interactions frequently account for the
specificity of enzymes. The combined effect of intermolecular
hydrogen bonding, hydrophobic effects, electrostatics, and solvent
reorganization is at the heart of the precise enzyme-substrate
associations which lead to selectivity.1 The application of these
concepts to the design of small molecule catalysts promises to
make the development of low-molecular-weight enzyme-like
systems possible.2 In this context, we are investigating synthetic
peptides containing nonproteinogenic amino acids that impart
catalytic activity with the ultimate goal of developing selective
peptide-based catalysts for asymmetric synthesis. Herein we
report the design and synthesis of new, functional peptides that
catalyze the kinetic resolution of certain secondary alcohols.3,4
Our initial design focused on peptide 1 (Scheme 1), which
contains 3-(1-imidazolyl)-(S)-alanine (IA) as the N-terminal amino
acid.5,6 Within IA is the N-methylimidazole (NMI) substructure,
which is capable of catalyzing the acylation of secondary alcohols
by acetic anhydride through a nucleophilic mechanism.7 We felt
that incorporation of IA into short, folded peptides would allow
for the formation of an acyl imidazolium intermediate (e.g., 1-Ac)8
in proximity to the chiral environment created by the peptide
backbone. In particular, we introduced IA into a predisposed
â-turn structure defined by the proline-R-aminoisobutyric acid
framework.9 Incorporation of the C-terminal (R)-R-methylben-
zylamide was intended to create the possibility for π-stacking of
the charged acylimidazolium ion with the phenyl group of the
catalyst (1-Ac).4e,10
devoid of a second functional group (eq 1, Table 1). When 1
equiv of trans-2-(N-acetylamino)cyclohexan-1-ol (2), 1 equiv of
aromatic alcohol 3, and one equiv of acetic anhydride were treated
with 0.05 equiv of NMI, the corresponding acetate esters 2-Ac
and 3-Ac were each formed in equal quantities. However, when
the identical experiment was conducted with â-turn catalyst 1 in
place of NMI, 2-Ac and 3-Ac were observed in a 6:1 ratio. As
a control, the reaction was conducted in the presence of tripeptide
4 which lacks the alkyl imidazole substructure. In this experiment,
no products were detected (<2% by 400 MHz NMR spectros-
copy), indicating that the NMI substructure is crucial for catalysis.
One possible explanation for the preferential acylation of 2 relative
to 3 in the presence of 1 could be the existence of a favorable
transition state hydrogen bond between the amide of 2 and the
peptide backbone of 1. While alternative explanations cannot
be ruled out at this time, the results substantively demonstrate
the capacity of peptide architecture to perturb reaction selectivi-
ties.11
Preliminary experiments were designed to demonstrate that
substrate interactions with the peptide backbone were kinetically
significant. Accordingly, we performed competition experiments
between alcohols substituted with amides (which could participate
in intermolecular transition state hydrogen bonding) and alcohols
Table 1. Competitive Acylation Experiments between Alcohols 2
and 3
(1) For a recent compendium of pertinent reviews, see: Gellman, S. H.,
Ed.; Chem. ReV. 1997, 97 (Chemical Reviews Thematic Issue on Molecular
Recognition), 1231-1734.
(2) (a) Breslow, R. Acc. Chem. Res. 1995, 28, 146-153. (b) Murakami,
Y.; Kikuchi, J.; Hiseada, Y.; Hayashida, O. Chem. ReV. 1996, 96, 721-758.
(3) The kinetic resolution of racemic alcohols is a powerful approach to
the preparation of optically pure compounds. For a review of enzymatic
approaches, see: Wong, C.-H.; Whitesides, G. M. Enzymes in Synthetic
Organic Chemistry; Elsevier Science Ltd.: Oxford, 1994; Chapter 2.
(4) For previously reported chiral nucleophilic catalysts which effect kinetic
resolution of racemic alcohols, see: (a) Vedejs, E.; Chen, X. J. Am. Chem.
Soc. 1996, 118, 1809-1810. (b) Vedejs, E.; Daugulis, O.; Diver, S. T. J.
Org. Chem. 1996, 61, 430-431. (c) Ruble, J. C.; Fu, G. C. J. Org. Chem.
1996, 61, 7230-7231. (d) Ruble, J. C.; Latham, H. A.; Fu, G. C. J. Am.
Chem. Soc. 1997, 119, 1492-1493. (e) Kawabata, T.; Nagato, M.; Takasu,
K.; Fuji, K. J. Am. Chem. Soc. 1997, 119, 3169-3170.
We then turned our attention to issues of enantioselective
catalysis (eq 2). Treatment of racemic trans-2-(N-acetylamino)-
cyclohexan-1-ol (2, 10 equiv) with 1 equiv of acetic anhydride
in the presence of 0.05 equiv of peptide 1 resulted in the formation
of the corresponding amidoacetate 2-Ac in 95% yield (relative
to Ac2O); the product exhibited an experimentally reproducible
enantiomeric excess of 48% (kfast/kslow (S) ) 3.0).12 In contrast,
the naphthyl-substituted alcohol 3 did not exhibit detectable
enantioselectivity (S ) 1) under analogous conditions and at
(5) For a synthesis of IA, see: (a) Tohodo, K.; Hamada, Y.; Shiori, T.
Synlett 1994, 247-249. (b) Arnold, L. D.; Kalantar, T. H.; Vederas, J. C. J.
Am. Chem. Soc. 1985, 107, 7105-7109.
(6) Assembly of peptides followed conventional solution phase techniques.
See the Supporting Information for details.
(7) (a) Guibe-Jampel, E.; Bram, G.; Vilkas, M. Bull. Soc. Chim. Fr. 1973,
1021-1027. (b) Ho¨fle, G.; Steglich, W.; Vorbru¨ggen, H. Angew. Chem., Int.
Ed. Engl. 1978, 17, 569-583. (c) A general base-type mechanism cannot be
excluded. See: Pandit, N. K.; Connors, K. A. J. Pharm. Sci. 1982, 71, 485-
491.
(8) Rybachenko, V. I.; Chervinskii, A. Y.; Kapkan, L. M.; Semenova, R.
G.; Titov, E. V. Zh. Org. Khim. 1976, 12, 240-241.
(9) Ravi, A.; Balaram, P. Tetrahedron 1984, 40, 2577-2583.
(10) Ma, J. C.; Dougherty, D. A. Chem. ReV. 1997, 97, 1303-1324.
(11) Hydrogen bonding has been implicated as a factor in determining the
stereochemical course of stoichiometric acyl transfer reactions involving chiral
acyl halides. See: Ishihara, K.; Kubota, M.; Yamamoto, H. Synlett 1994, 611-
614.
S0002-7863(97)03892-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/06/1998

1630 J. Am. Chem. Soc., Vol. 120, No. 7, 1998
Communications to the Editor
Table 2. Influence of Solvent on the Kinetic Resolution of
Hydroxyamide 2 Catalyzed by 1
comparable conversion (eq 3). Thus, substrate 2, which was
a All reactions were performed with 10 equiv of racemic substrate,
1 equiv of acetic anhydride, and 0.05 equiv of catalyst 1. Reactions
proceeded to 90-100% yield based on Ac2O. Yields were measured
following silica gel chromatography. Absolute configurations were
assigned by analogy to authentic materials. b A minimum amount of
CHCl3 was introduced initially to solubilize catalyst and starting alcohol.
c Determined by chiral HPLC or GC analysis. d Calculated according
to the method of Kagan (ref 12).
found to undergo preferential acylation in the competition
experiment employing peptide 1, exhibited an appreciable re-
sponse to the chiral environment of the peptide. Substrate 3 was
inert to chirality transfer.
Optimization studies revealed that solvents which favor hy-
drogen bonding lead to selectivity enhancement (Table 2). For
example, optimum selectivities were obtained in nonpolar solvents
that are not Lewis basic (entry 3 84% enantiomeric excess (ee)
in toluene, S ) 12.6; entry 4 12% ee in CH3CN, S ) 1.3). Protic
media proved to have a deleterious effect on reaction selectivity
as well, with a chloroform-t-BuOH medium leading to a
completely nonselective reaction (entry 5).13 It is important to
note that these data do not distinguish between the significance
of intramolecular or intermolecular hydrogen bonds. Neverthe-
less, they do reflect the potential for substantial kfast/kslow values
to be obtained with relatively simple peptide systems.
catalyst than 1 under the optimized conditions, affording 2-Ac
with 53% ee (S ) 3.5; cf. 84% ee, S ) 12.6 with 1). Additional
studies are now underway to identify (i) the specific catalyst-
substrate interactions that dictate selectivity and (ii) their relation-
ship to catalytically active peptide conformations.16
Experiments employing different substrates have provided
further data concerning catalyst-substrate interactions (Figure 1).
When the optimized conditions were employed (Table 2, entry
3), cyclohexene derivative 5 exhibited a kfast/kslow ratio of 3.0 under
the influence of peptide 1 (entry 1); cycloheptane derivative 6
also exhibited an appreciable kfast/kslow ratio of 4.6 under the same
reaction conditions (entry 2). In contrast, acetate ester 7 (entry
3), isosteric with acetamide 2 but less Lewis basic and incapable
of acting as a hydrogen bond donor through the ester, showed a
greatly diminished kfast/kslow ratio of 1.4 under the reaction
conditions. These data also point to transition state involvement
of the amide of the substrate with the asymmetric functional group
array provided by the catalyst backbone.
In summary, we have established that short peptides containing
the synthetic amino acid IA are asymmetric catalysts for the
kinetic resolution of functionalized racemic alcohols. The present
data suggest that alcohols containing an amide functional group
are substrates for the peptide catalyst and that transition state
hydrogen bonding may play a role in enantiomer differentiation.
Further studies toward the development of more general and
highly selective peptide-based asymmetric catalysts are in progress.
In addition, these studies should afford significant information
on peptide-substrate interactions which are of general interest
in the context of biological catalysis.
Acknowledgment. We acknowledge the National Science Foundation
for financial support (CHE-9612563). Acknowledgment is also made to
the donors of the Petroleum Research Fund administered by the American
Chemical Society for partial support of this research. This research is
also supported by a Research Innovation Award sponsored by Research
Corporation. G.T.C. is grateful to the Department of Education for a
GAANN fellowship. S.J.M. and E.R. are grateful to Boston College for
a Research Incentive Grant and an Undergraduate Research Fellowship,
respectively.
Supporting Information Available: Characterization data for all
compounds and experimental details for their preparation (9 pages). See
any current masthead page for ordering information and Web access
instructions.
Figure 1. Influence of substrate structure on resolutions catalyzed by 1.
We have now begun to investigate the consequences of peptide
backbone conformation on reaction selectivities in detail. Gell-
man and co-workers have recently delineated the consequences
of stereogenic center configuration on the propensity of given
peptide sequences to form â-turns in a nonpolar medium.14 We
synthesized the epimeric â-turn catalyst 8 to evaluate the
significance of the absolute configuration of the R-methylbenz-
amide moiety. Of note is the fact that whereas peptide 1 exists
as essentially one conformation in CDCl3 solution at 23 °C (>90%
by 400 MHz 1H NMR),15 peptide 8 exists as 3:1 conformational
mixture at the same temperature. Catalyst 8 was then studied
for its ability to effect kinetic resolution of 2 under the optimized
conditions. Peptide 8 proved to be a substantially less-selective
JA973892K
(15) Detailed conformational analyses employing NMR techniques are in
progress. Peptides 1, 4, and 8 each exhibited IR stretches consistent with both
CdO‚‚‚H-N hydrogen bonds (<3400 cm-1), and non-hydrogen-bonded N-H
groups (>3400 cm-1). At 0.002 M/CH2Cl2, 1, 3427, 3365 cm-1; 4, 3421,
3358 cm-1; 8, 3421, 3364 cm-1). For an interpretation of N-H stretches in
CH2Cl2, see: (a) Gellman, S. H.; Dado, G. P.; Liang, G.-B.; Adams, B. R. J.
Am. Chem. Soc. 1991, 113, 1164-1173. (b) Liang, G.-B.; Desper, J. M.;
Gellman, S. H. J. Am. Chem. Soc. 1993, 115, 925-938. (c) Gardner, R. R.;
Liang, G.-B.; Gellman, S. H. J. Am. Chem. Soc. 1995, 117, 3280-3281. (d)
Rao, C. P.; Nagaraj, R.; Rao, C. N. R.; Balaram, P. Biochemistry 1980, 19,
425-431.
(12) S values were calculated according to the method of Kagan. See:
Kagan, H. B.; Fiaud, J. C. Top. Stereochem. 1988, 18, 249-330.
(13) Running the reaction to higher levels of conversion under these
conditions leads to a decrease in selectivity. At 33% conversion, the product
exhibits 60% ee (S ) 5.3).
(14) For a detailed discussion of the influence of backbone stereochemical
configuration on â-turn conformation, see: Haque, T. S.; Little, J. C.; Gellman,
S. H. J. Am. Chem. Soc. 1996, 118, 6975-6985.
(16) Use of the unsubstituted pyrrolidine-derived amide of N-BOC-IA as
a catalyst for the resolution of 2 afforded 2-Ac without detectable enantio-
selectivity. Other derivatives of IA as well as other nucleophile-loaded amino
acids are currently under study.