Asymmetric formation of quaternary centers through aza-annulation of chiral β-enamino amides with acrylate derivatives

Article abstract of DOI:10.1021/ja9729591

The stereoselective formation of six-membered nitrogen heterocycles that contain an asymmetric quaternary carbon center was achieved through aza- annulation of β-enamino amide substrates with activated acrylate derivatives. Condensation of a racemic β-keto amide with an optically active primary amine, either (R)-α-methylbenzylamine or α-amino esters, generated the corresponding optically active tetrasubstituted secondary enamine, in which the enamine tautomer was stabilized through conjugation with an amide carbonyl. Treatment of the intermediate enamine with acryloyl chloride, acrylic anhydride, or sodium acrylate/ethyl chloroformate resulted in aza- annulation to give the corresponding δ-lactam with high diastereoselectivity (>96% de). For the variety of different β-enamino amide substrate classes examined in this reaction, the optimum method for activation of the acrylate derivative was the use of EtO₂CCl. When aza-annulation was performed with an α-acetamido-substituted acrylate derivative, the stereoselective formation of the quaternary carbon center was accompanied by poor selectivity for generation of the center α to the lactam carbonyl. Crystallographic analysis of one α-amido aza-annulation product was performed to identify the key structural features of these molecules.

Full text of DOI:10.1021/ja9729591

VOLUME 120, NUMBER 11
MARCH 25, 1998
© Copyright 1998 by the
American Chemical Society
Asymmetric Formation of Quaternary Centers through
Aza-Annulation of Chiral â-Enamino Amides with Acrylate
Derivatives
Petr Benovsky,† Gregory A. Stephenson,‡ and John R. Stille*,‡
Contribution from the Department of Chemistry, Michigan State UniVersity, East Lansing, Michigan
48824-1322, and Eli Lilly and Company, Lilly Corporate Center, Indianapolis, Indiana 46285
ReceiVed August 22, 1997
Abstract: The stereoselective formation of six-membered nitrogen heterocycles that contain an asymmetric
quaternary carbon center was achieved through aza-annulation of â-enamino amide substrates with activated
acrylate derivatives. Condensation of a racemic â-keto amide with an optically active primary amine, either
(R)-R-methylbenzylamine or R-amino esters, generated the corresponding optically active tetrasubstituted
secondary enamine, in which the enamine tautomer was stabilized through conjugation with an amide carbonyl.
Treatment of the intermediate enamine with acryloyl chloride, acrylic anhydride, or sodium acrylate/ethyl
chloroformate resulted in aza-annulation to give the corresponding δ-lactam with high diastereoselectivity
(>96% de). For the variety of different â-enamino amide substrate classes examined in this reaction, the
optimum method for activation of the acrylate derivative was the use of EtO2CCl. When aza-annulation was
performed with an R-acetamido-substituted acrylate derivative, the stereoselective formation of the quaternary
carbon center was accompanied by poor selectivity for generation of the center R to the lactam carbonyl.
Crystallographic analysis of one R-amido aza-annulation product was performed to identify the key structural
features of these molecules.
Introduction
bradykinin.4 Intramolecular hydrogen bonding has also been
observed spectroscopically for 2b.5
Conformationally constrained δ-lactam peptide analogues
have been valuable tools for the exploration of active conforma-
tion and properties of several biologically active peptides.1 Early
work in this area established the potential use of six-membered
nitrogen heterocycles for conformational control of peptide
backbones,2 and spectroscopic studies on 1 revealed the presence
of intramolecular hydrogen bonding in solution.3 More recently,
the preparation of 2a was reported, which is a cis-Gly-Pro
dipeptide mimetic with the conformational features found in
peptide type-VI turns, and this dipeptide-like fragment has been
incorporated successfully into analogues of cis-Gly6-Pro7-
Our approach to the efficient construction of conformationally
restricted molecules related in structure to dipeptides and
homologated dipeptides has incorporated the aza-annulation
† Michigan State University.
‡ Eli Lilly and Company.
S0002-7863(97)02959-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/05/1998

2494 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
BenoVsky et al.
Table 1. Effect of Chiral Amine and Temperature on Asymmetric
Scheme 1. General Strategy for Asymmetric
Aza-Annulation Reactions
Inductiona
amine 7
R1
Me
R2
Ph
T (°C)
8:9b
yield (%)c
7a
7b
7b
7b
7c
66
66
>97:3
21:79
7:93
85
63
68
77
43
Ph
Ph
Ph
CO2Me
CO2Et
CO2Et
CO2Et
iPr
0
-33
66
2:98
57:43
a See eq 1. b Determined by 1H NMR of the crude reaction mixture.
c Yield of the diastereomeric mixture after chromatography.
reaction as a route to δ-lactam products.6 In previous studies,
initial condensation of a racemic â-keto ester 3 with an optically
active primary amine led to formation of a chiral â-enamino
ester as a single geometric isomer 4 (Scheme 1).7 Treatment
of this optically active (Z)-enamine with an activated acrylic
acid derivative then resulted in formation of the corresponding
δ-lactam 5 through formal conjugate addition to form the
carbon-carbon bond and N-acylation to generate the amide
functionality. Heterocycle construction with this aza-annulation
process resulted in the stereoselective generation of a quaternary
center at C-5 with 1,4-asymmetric induction from the chiral
amine.
The utility of chiral R-amino ester derivatives in the asym-
metric aza-annulation reaction has been greatly enhanced by
the use of â-enamino amide derivatives. Herein, the reaction
of activated acrylic acid derivatives with â-enamino amides for
the general construction of six-membered nitrogen heterocycles
is reported for the first time. Reaction with this class of
substrates is dependent on the nature of the substrate, and has
resulted in the rapid construction of conformationally restricted
tri- and tetrapeptide analogues and a variety of other molecules
with high asymmetric induction.
Variation in the source of chirality, from R-methylbenzy-
lamine to R-amino acid esters, gave poorer selectivity for the
aza-annulation reaction with â-enamino ester substrate 6 (eq
1).7 Although 7a led to a >97:3 stereoselective formation of
Results and Discussion
Modification of Ester Substrates. Our first approach to the
construction of conformationally restricted peptide analogues
incorporated the synthesis and extension of the bicyclic amino
acid 12 derived from ester 11. Synthesis of 11 was ac-
complished through the stepwise condensation of the â-keto
ester 10 with 7a, which gave the corresponding â-enamino ester,
and subsequent aza-annulation with sodium acrylate/EtO2CCl
gave diastereomer 11 in 70% yield. Deprotection of the
carboxylate gave the corresponding acid 12 without evidence
for enamine reduction, even after extended exposure to the
reaction conditions.
8a with the quaternary center at C-5, use of the ethyl ester of
phenylglycine (7b) resulted in a significantly lower (21:79) ratio
of diastereomers 8 and 9 under the same reaction conditions
(Table 1). Deviation from a phenyl substituent at R1 or R2, the
use of the methyl ester of valine, led to poor diastereoselective
δ-lactam formation (57:43). Although these substrates showed
a great deal of promise for asymmetric construction of hetero-
cycles, the scope of this method for the general synthesis of
extended peptide structures with high enantiomeric purity was
greatly limited.
The utility of the aza-annulation reaction for the general
synthesis of extended peptide structures with high enantiomeric
purity was limited by the inability to efficiently modify fragment
12. Initial attempts to extend the peptide sequence through the
coupling of 12 with glycine ethyl ester proved somewhat
problematic. A number of standard peptide coupling reagents
were used for this purpose, but the steric hindrance around the
activated carboxylic acid intermediate prevented efficient
coupling at this site. The best yield for this reaction was
achieved with the use of 1,3-dicyclohexylcarbodiimide (DCC),
but gave only 15% of the desired product. Instead, formation
of the stable N-acylurea adduct from the reaction of the acid
with DCC was observed as the major product of the reaction.8
Alternative methods used to prepare the amide, by reactions
with either (COCl)2/pyridine/glycine ethyl ester or ethyl chlo-
roformate/glycine ethyl ester in THF, either were completely
unsuccessful or gave only a trace of the product.
(1) For reviews on conformationally restricted peptidomimetics, see: (a)
Davies, J. S. Amino Acids Peptides 1991, 22, 145. (b) Liskamp, R. M. J.
Recl. TraV. Chim. Pays-Bas 1994, 113, 1.
(2) (a) Freidinger, R. M.; Veber, D. F.; Perlow, D. S.; Brooke, J. R.;
Saperstein, R. Science 1980, 210, 656. (b) Freidinger, R. M.; Veber, D. F.;
Hirschmann, R.; Paege, L. M. Int. J. Peptide Protein Res. 1980, 16, 464.
(c) Freidinger, R. M.; Perlow, D. S.; Veber, D. F. J. Org. Chem. 1982, 47,
104.
(3) (a) Kemp, D. S.; Sun, E. T. Tetrahedron Lett. 1982, 23, 3759. (b)
Kemp, D. S.; McNamara, P. Tetrahedron Lett. 1982, 23, 3761. (c) Kemp,
D. S.; McNamara, P. E. J. Org. Chem. 1984, 49, 2286.
(4) Gramberg, D.; Robinson, J. A. Tetrahedron Lett. 1994, 35, 861.
(5) (a) Dumas, J.-P.; Germanas, J. P. Tetrahedron Lett. 1994, 35, 1493.
(b) Kim, K.; Dumas, J.-P.; Germanas, J. P. J. Org. Chem. 1996, 61, 3138.
(6) (a) Paulvannan, K.; Stille, J. R. Tetrahedron Lett. 1993, 34, 8197.
(b) Beholz, L. G.; Benovsky, P.; Ward, D. L.; Barta, N. S.; Stille, J. R. J.
Org. Chem. 1997, 62, 1033. (c) For a review on the subject of aza-
annulation, see: Stille, J. R.; Barta, N. S. In Studies in Natural Products
Chemistry: StereoselectiVe Synthesis; Atta-ur-Rahman, Ed.; New York,
1996; Vol. 18, pp 315-389.
Interestingly, the use of DPPA as the reagent led to the
formation of 13a in 10% yield, but a significant amount of an
alternative coupling product was formed. This unexpected
product was isolated in 53% yield, and has been assigned the
structure of the disubstituted urea 14 (Scheme 2). Generation
of 14, through Curtius rearrangement of the intermediate acyl
azide, followed by reaction with glycine ethyl ester, has
(7) (a) Barta, N. S.; Brode, A.; Stille, J. R. J. Am. Chem. Soc. 1994,
116, 6201. (b) See also: Cave´, C.; Desmae¨le, D.; d’Angelo, Jean; Riche,
C.; Chiaroni, A. J. Org. Chem. 1996, 61, 4361.
(8) Holmberg, K.; Hansen, B. Acta Chem. Scand. B 1979, 33, 410.

Aza-Annulation of â-Enamino Amides
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2495
Table 2. Effect of Chiral Amine and Acrylate Derivative on the
Scheme 2. Conversion of 10 to Amides with Extended
Peptide Chains
Asymmetric Aza-Annulation Reactiona
amine
R1
R2
methodb
17:18c
yield (%)d
7a
Me
Ph
A
B
C
A
B
C
A
A
>98:2
>98:2
>98:2
2:>98
2:>98
2:>98
>98:2
95:5e
99
86
67
96
80
49
90
46
7b
Ph
CO2Et
7c
7d
CO2Me
CO2Et
iPr
Bn
a See Scheme 3. b Reaction conditions: (i) 7 or 7‚HCl/NaHCO3,
toluene, reflux; (ii) (method A) sodium acrylate, ClCO2Et, THF, reflux;
(method B) sodium acrylate/acryloyl chloride, THF, reflux; (method
1
C) acryloyl chloride, THF, reflux. c Determined by H NMR of the
crude reaction mixture. d Yield of the diastereomeric mixture after
chromatography. e Assignment of the major diastereomer as 17 was
1
made on the basis of H NMR comparison to 17c.
acrylate derivatives were examined. Acryloyl chloride, acrylic
anhydride, and mixed anhydride reagents were employed for
the aza-annulation with asymmetric â-enamino amides. To
evaluate the effectiveness of each reagent, aza-annulation of the
R-methylbenzylamine enamine 16a was performed with these
three different activated acrylate derivatives, and yields and
diastereoselectivity ratios were obtained.
Efficient formation of δ-lactam products was obtained for
aza-annulation of enamines derived from a variety of â-keto
amides and (R)-methylbenzylamine 7a (Table 2). A significant
dependence of the aza-annulation reaction outcome on the type
of reagent employed was observed. Although acryloyl chloride
produced efficient heterocycle formation with â-enamino ester
substrates,7 similar reaction with the â-enamino amides gave
significantly lower yields. The use of acrylic acid anhydride,
generated in situ by the reaction of sodium acrylate with acryloyl
chloride, resulted in somewhat improved yields. The mixed
anhydride, formed by the combination of sodium acrylate with
EtO2CCl, proved to be the optimum reagent for aza-annulation
with â-enamino amide substrates. The increased efficiency in
product formation with anhydride reagents paralleled that found
for the analogous ester substrates; however, the decreased
diastereoselectivity observed for reaction of ester substrates with
anhydride reagents was not observed for the corresponding
amides.7 In each case, the diastereoselective formation of the
quaternary carbon from the â-enamino amide substrates was
high (>98:2), and was independent of the acrylate derivative
used for aza-annulation.12
A similar reactivity was observed for the â-enamino amide
derived from condensation of phenylglycine ethyl ester (7b) with
15. The aza-annulation of 16b was significantly more efficient
when acrylic acid anhydride was used instead of acryloyl
chloride, and a further increase in yield was obtained through
the use of the mixed anhydride. In each case, stereoselective
formation of 18b occurred to the extent of >98:2. Use of the
valine-derived substrate 16c, which resulted in poor diastereo-
selectivity in the case of the â-enamino ester substrate (57:43),
gave excellent stereoselective formation of 17c (>98:2) in high
Scheme 3. Asymmetric Aza-Annulation of â-Enamino
Amide Intermediates with Acrylate Derivativesa
a Reaction conditions: (i) 7 or 7‚HCl/NaHCO3, toluene, reflux; (ii)
(method A) sodium acrylate, ClCO2Et, THF, reflux; (method B) sodium
acrylate/acryloyl chloride, THF, reflux; (method C) acryloyl chloride,
THF, reflux.
precedent in related work.9 The difficulties encountered in
conversion of 12 to 13a served to reinforce the need to examine
methods for aza-annulation with â-keto amide substrates.
Cyclic Amide Substrates. Investigation of â-keto amide
substrates in the asymmetric aza-annulation reaction was
initiated with a study of aza-annulation reagents (Scheme 3).
On the basis of our studies with achiral imine10 and â-enamino
carbonyl substrates,6,7 as well as the use of this methodology
in the synthesis of natural products,11 three different classes of
(11) (a) Paulvannan, K.; Schwarz, J. B.; Stille, J. R. Tetrahedron Lett.
1993, 34, 215. (b) Paulvannan, K.; Stille, J. R. Tetrahedron Lett. 1993, 34,
6673. (c) Paulvannan, K.; Stille, J. R. J. Org. Chem. 1994, 59, 1613. (d)
Cook, G. R.; Beholz, L. G.; Stille, J. R. Tetrahedron Lett. 1994, 35, 1669.
(e) Cook, G. R.; Beholz, L. G.; Stille, J. R. J. Org. Chem. 1994, 59, 3575.
(12) Prior to purification, the crude ratio of diastereomers was determined
by 1H NMR. The major isomer was assigned by correlation to the previously
reported analogous aza-annulation7 and Michael addition reactions. For an
excellent review on asymmetric Michael addition reactions, see: d’Angelo,
J.; Desmae¨le, D.; Dumas, F.; Guingant, A. Tetrahedron: Asymmetry 1992,
3, 459.
(9) (a) Arrieta, A.; Palomo, C. Tetrahedron Lett. 1981, 22, 1729. (b)
Chorev, M.; MacDonald, S. A.; Goodman, M. J. Org. Chem. 1984, 49,
821. (c) Tam, T. F.; Coles, P. Synthesis 1988, 383. (d) Neville, C. F.;
Grundon, M. F.; Ramachandran, V. N.; Reisch, J. J. Chem. Soc., Perkin
Trans. 1 1991, 259. (e) Nishi, N.; Tsunemi, M.; Nakamura, K.; Tokura, S.
Makromol. Chem. 1991, 192, 1811.
(10) Paulvannan, K.; Stille, J. R. J. Org. Chem. 1992, 57, 5319.

2496 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
BenoVsky et al.
Scheme 5. Asymmetric Aza-Annulation of Acyclic
Scheme 4. Asymmetric Aza-Annulation of
Cyclopentanone-Derived â-Enamino Amide Intermediates
with Acrylate Derivativesa
â-Enamino Amide Intermediates with Acrylate Derivatives
a Yields represent the two-step conversion of 19 to product.
b Determined by 1H NMR of the crude reaction mixture. c Ratio of the
diastereomeric mixture after chromatography.
Scheme 6. Aza-Annulation with Threonine Derivativesa
yield (90%) for the â-enamino amide. Interestingly, even the
phenylalanine-derived 7d was effective at asymmetric induction
(95:5), but the yield of the aza-annulation reaction was low with
this source of asymmetry.
The five-membered ring substrate 19 showed different
reactivity and stability patterns than the analogous six-membered
ring substrate (Scheme 4). The â-keto amide 19 was generated
from the corresponding â-keto ester species, and was used
without extensive purification for subsequent formation of
enamines 20 and 22. Both 20 and 22 were more sluggish in
their formation, and these enamines were isolated in low yield
for use in the aza-annulation process. Aza-annulation of 20
led to generation of 21 in only moderate yield, but the product
was obtained with high diastereoselectivity (>98:2). Treatment
of 22 with the acrylate mixed anhydride led to a more efficient
aza-annulation than that of 20, but the amino acid source of
asymmetry led to only an 84:16 crude ratio of diastereomers;
after purification of 23, a 75:25 ratio of diastereomers was
obtained.
Acyclic Amide Substrates. The aza-annulation with acyclic
â-keto amide substrates resulted in ring formation with a high
degree of diastereoselectivity (Scheme 5). Condensation of 24
with 7a efficiently generated the corresponding enamine as a
single geometric isomer, as determined by 1H NMR of the crude
reaction mixture, and subsequent aza-annulation with the
acrylate mixed anhydride generated 25 with high diastereose-
lectivity (>98:2). However, all attempts to isolate 25 resulted
in partial hydrolysis of the mixture to 26. To obtain an accurate
yield for the carbon-carbon bond formation process, crude 25
was treated with p-TsOH in THF/H2O to promote complete
hydrolysis of the enamide functionality. For the three-step
process of enamine formation, aza-annulation reaction, and
hydrolysis, an 82% yield of 26 was achieved, and the product
was obtained with a >98:2 diastereomer ratio.
a Reaction conditions: (i) (a) EtOH, HCl, reflux; (b) (1) 2,4,6-
trichlorophenol, DCC (60%); (2) BnNH2 (80%); (ii) PCC/Celite (1:1),
CH2Cl2 (a) 91% (i-ii); (b) 56% based on recovered starting material);
(iii) 7a, toluene, reflux; (iv) sodium acrylate/EtO2CCl, THF, 25 °C;
(v) acryloyl chloride, THF reflux; (vi) hydrolysis (see the text). Overall
aza-annulation yields; (iii, iV, Vi) (a) 45%; (b) 75%; (iii, V, Vi) a: 30%;
b: 65%.
tion of 24 with 7b generated a single isomeric enamine
intermediate, and aza-annulation gave δ-lactam 27 with >98:2
diastereoselectivity. Again, facile hydrolysis was observed for
this disubstituted terminal enamide, which precluded efficient
isolation of 27. However, intentional hydrolysis and subsequent
isolation gave a 71% yield of 28 for the three-step process
without loss of stereochemical integrity.
Compounds 29a and 29b, protected derivatives of threonine,
were the source of the next â-keto carboxylate species studied
(Scheme 6). In this case, the â-keto ester was prepared in
addition to the amide for comparative analysis. The corre-
sponding amino-protected amino acid derivatives, 30a and 30b,
were prepared through standard esterification and peptide
coupling strategies, respectively. While oxidation of 30a
Reaction of 24 with 7b followed by aza-annulation led to
results similar to those obtained for 7a (Scheme 5). Condensa-

Aza-Annulation of â-Enamino Amides
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2497
proceeded smoothly to give 31a (91% yield for the two-step
process from 29a), oxidation of 30b was more problematic.
Oxidation of 30b proceeded best with PCC/Celite (1:1), but the
reaction was very sensitive and an unoptimized 56% yield of
31b, based on recovered starting material, could be obtained.
Condensation of the â-keto ester 31a with 7a, followed by
treatment with the acrylate mixed anhydride, gave the aza-
annulation reaction product 32a. As observed for 25 and 27,
this terminal enamide was sensitive toward hydrolysis condi-
tions, and even with aqueous NaHCO3 workup, hydrolysis
occurred completely to give the acyclic product. Overall, the
threestep condensation/aza-annulation/hydrolysis procedure pro-
vided 33a in 45% yield. Conversion of 31a to 33a was also
performed with acryloyl chloride as the aza-annulation reagent;
however, reaction under these conditions gave only a 30%
overall yield of 33. The major byproduct of the reaction was
found to be the corresponding acrylamide 34. Independent of
the reagent used for this reaction, the chiral R-amino acid was
formed with >98:2 stereoselectivity.
Scheme 7. Introduction of an R-Amido Substituent in the
δ-Lactam Product
The three-step condensation/aza-annulation/hydrolysis pro-
cedure produced much better results for the analogous â-keto
amide substrate. With the use of aza-annulation conditions that
employed the generation of the mixed anhydride reagent
(sodium acrylate/EtO2CCl), a 75% yield of 33b was obtained.
In this case, complete hydrolysis of 32 to 33 was effected simply
by attempted silica gel chromatographic purification of the
product. Aza-annulation of 31 with acryloyl chloride as the
reagent produced 33b in 65% yield, which paralleled observa-
tions for the ester derivative.
Several important features of the aza-annulation reaction of
acyclic â-keto carboxylate derivatives are of significance. The
acyclic substrates were much more sensitive toward hydrolysis.
However, after intentional hydrolysis of the reaction mixture,
R-chiral â-keto amide products of carbon-carbon bond forma-
tion were obtained in good yield with high diastereoselective
bond formation. The 1,4-asymmetric induction that occurred
during the aza-annulation process appeared as a 1,6-relationship
in the product of hydrolysis. In addition, the high diastereo-
selectivity obtained in these reactions resulted, in part, from
the important formation of a single geometric isomer of the
intermediate â-enamino carboxylate derivative.
r-Amino Lactam Products. The use of sodium 2-aceta-
midoacrylate (35) as the aza-annulation reagent led to the
formation of 2-acetamido δ-lactam products 36 and 37 (Scheme
7).7,13 Treatment of 16a with the mixed anhydride of 35,
generated in situ by the reaction of 35 with ClCO2Et, resulted
in formation of equal amounts of two diastereomeric R-aceta-
mido δ-lactam products in 67% yield. Separation of the
diastereomers allowed for characterization of each product, and
treatment of 37 with NaH led to epimerization R to the lactam
carbonyl that resulted in a 45:55 ratio of stereoisomers 36 and
37. On the basis of this information, >98:2 diastereoselective
formation of the quaternary center occurred, with no bias for
selective generation of the stereoisomers at the R position.
Figure 1. ORTEP representation of 36.
Crystallization of diastereomer 36 allowed for the X-ray
crystallographic characterization of the molecule, and the three-
dimensional features of 36 are illustrated in Figure 1. With
the stereochemical configuration of the chiral amine known,
the generation of the quaternary carbon center resulted through
carbon-carbon bond formation from the least hindered face of
the enamino amide as shown by the favored orientation shown
for 16 (Scheme 3, R2 (Ph) > R1 (Me)). Interestingly, although
36 had greater potential than 37 for intramolecular hydrogen
bonding, with both the benzyl amide and the acetamido
substituents cis relative to each other, intramolecular hydrogen
bonding was not observed in the solid state. In the solid state,
(13) For previous reports of the use of 2-amidoacrylic acid derivatives
in aza-annulation, see: (a) Thyagarajan, B. S.; Gopalakrishnan, P. V.
Tetrahedron 1964, 20, 1051. (b) Meyer, H.; Bossert, F.; Horstmann, H.
Justus Liebigs Ann. Chem. 1978, 1483. (a) Danieli, B.; Lesma, G.;
Palmisano, G. J. Chem. Soc., Chem. Commun. 1980, 109. (c) Danieli, B.;
Lesma, G.; Palmisano, G. Gazz. Chim. Ital. 1981, 111, 257. (c) McCabe,
R. W.; Young, D. W.; Davies, G. M J. Chem. Soc., Chem. Commun. 1981,
395. (d) Chiba, T.; Takahashi, T. Chem. Pharm. Bull. 1985, 33, 2731. (e)
Chiba, T.; Takahashi, T.; Kaneko, C. Chem. Pharm. Bull. 1985, 33, 4002.
(f) Capps, N. K.; Davies, G. M.; Loakes, D.; McCabe, R. W.; Young, D.
W. J. Chem. Soc., Perkin Trans. 1 1991, 3077. (g) Hua, D. H.; Park, J.-G.;
Katsuhira, T.; Bharathi, S. N. J. Org. Chem. 1993, 58, 2144. (h) Kolar, P.;
Tisler, M. J. Heterocycl. Chem. 1993, 30, 1253.

2498 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
BenoVsky et al.
Scheme 8. Asymmetric Aza-Annulation of â-Enamino
Amino Acid Intermediates with Acrylate Derivatives
centers. Condensation with chiral primary amines led to
formation of â-keto carboxylic acid derivatives, to generate a
single geometric isomer of the intermediate â-enamino amides.
The â-enamino amides were formed from a variety of chiral
primary amines, including R-methylbenzylamine and esters of
R-amino acids, and gave high diastereoselective formation of
quaternary carbon centers. Mixed anhydrides of acrylate
derivatives were the most effective reagents for the aza-
annulation reaction of â-enamino amides to form δ-lactams.
Formation of extended peptide sequences was demonstrated at
both the 2-position of the lactam and the carboxylate site of
the â-keto amide substrate. Aza-annulation resulted in the rapid
and efficient construction of δ-lactam products that can contain
a high degree of complexity and have an asymmetrically
substituted quaternary carbon.14 Products derived from non-
cyclohexanone substrates, the cyclopentanone and acyclic
compounds, were sensitive toward hydrolysis, but the resultant
products of “Michael addition” were formed in good yield with
high asymmetric induction. In addition, the products of the aza-
annulation reaction were â-amino acids, which are important
species in the study of biologically active systems.15
Experimental Section
General Methods. All reactions were carried out by performing
standard inert atmosphere techniques to exclude moisture and oxygen,
and reactions were performed under an atmosphere of nitrogen or argon.
Acryloyl chloride was purchased from Fluka and used without
purification. Compounds 10,16 15,17 and 1917 were prepared according
to established procedures. Compound 24 was prepared by alkylation
of benzylacetoacetamide with MeI in EtOH/EtONa. Azeotropic
removal of H2O was assisted by the use of 4-Å molecular sieves.18
Isolated compounds were crystallized from a minimun amount of the
eluent; those compounds for which melting points were not reported
were oils.
the cis-acetamido substituent is oriented in a pseudoequatorial
position directed away from the benzyl amide substituent. This
could be the result of more favorable intramolecular interactions
that are favored during lattice packing.
Although intramolecular hydrogen bonds were not present,
several intermolecular hydrogen bonding interactions were
observed. The lactam carbonyl [O(2)] and the R-NH [N(19)]
of the acetamido substituent of one molecule form a comple-
mentary/reciprocal intermolecular hydrogen bonding relationship
with the R-NH [N(19)] of the acetamido substituent and the
lactam carbonyl [O(2)] of a second molecule, respectively (2.132
Å each). The NH [N(23)] of the benzyl amide was also found
to interact (2.106 Å) with the carbonyl [O(20)] of the R-aceta-
mido substituent of a third molecule. Although solid-state
intramolecular hydrogen bonding was not observed in this case,
this does not provide evidence against possible intramolecular
hydrogen bonding in solution, which has been observed in
similar molecules.3
General Procedure for the Preparation of â-Ketoamides: Method
A (15). A mixture of the necessary â-keto ester (1 equiv), amine (2
equiv), and DMAP (0.3 equiv) in toluene was heated at reflux for 24
h. Method B (15). A mixture of the necessary â-keto ester (1 equiv)
and amine (1.5 equiv) in xylene was heated at reflux for 24 h. Method
C (39a, 39b). A mixture of 38 (10 mmol) and the appropriate amino
acid (10 mmol) in xylene was heated at reflux for 2 h. Each reaction
mixture was then cooled and concentrated to a residue, and the crude
product was purified with flash column chromatography (eluent as
indicated).
â-Keto Amino Acid Substrates. An important direction of
this investigation, one which is key to the potential use of this
methodology for the synthesis of conformationally restricted
peptide derivatives, was the demonstrated ability to extend the
amino acid sequence at the amide position of the substrate. The
necessary substrates, 39a and 39b, were prepared by the reaction
of the corresponding amino acid ester with 38 (Scheme 8).
Subsequent condensation of 39 with 7a, followed by aza-
annulation with sodium acrylate/EtO2CCl, gave 13a as antici-
pated. Interestingly, the reaction of the glycine derivative a
(74%) was significantly more efficient than that of the valine
derivative b (50%). This difference in reaction efficiency
became even more pronounced when aza-annulation was
performed with 35/EtO2CCl. In this case, lactam formation
resulted in a good yield of 40a in the two step-process from
39a, but produced only minor amounts of the desired aza-
annulation products from 39b. In the case of substrate b,
epimerization of the amino acid residue was not observed under
either of the aza-annulation reaction conditions.
Data for 15:17 80:20; hexanes/EtOAc, 2.50 g, 10.8 mmol, 96% yield;
mp 85-86 °C; H NMR (300 MHz, CDCl3) δ 1.54-1.78 (m, 2 H),
1
1.80-2.00 (m, 2 H), 2.04 (m, 1 H), 2.20 (m, 1 H), 2.22-2.44 (m, 2
H), 3.14 (dd, J ) 5.5, 10.5 Hz, 1 H), 4.36-4.46 (m, 2 H), 5.63 (br s,
1 H), 7.15-7.31 (m, 5 H); 13C NMR (75 MHz, CDCl3) (mixture of
tautomers) δ 21.8, 22.5, 22.6, 24.3, 27.3, 29.2, 31.7, 42.2, 43.1, 43.3,
55.7, 96.8, 127.3, 127.5, 127.6, 127.7, 128.6, 128.7, 138.1, 138.2, 168.9,
170.5, 172.3, 210.6; IR (CHCl3) 1640, 1605, 1530 cm-1; HRMS calcd
for C14H17NO2 m/z 231.1259, obsd m/z 231.1268.
(14) For reviews on methods for the formation of quaternary carbons,
see: (a) Martin, S. F. Tetrahedron 1980, 36, 419. (b) Fuji, K. Chem. ReV.
1993, 93, 2037.
(15) For the importance of â-amino acids, see: (a) Lubell, W. D.;
Kitamura, M.; Noyori, R. Tetrahedron: Asymmetry 1991, 2, 543. (b) Juaristi,
E.; Quintana, D.; Escalante, J. Aldrichim. Acta 1994, 27, 3.
(16) (a) Taber, J. F.; Amedio, J. C., Jr.; Patel, Y. K. J. Org. Chem. 1985,
50, 3618. (b) Hatakeyama, S.; Satoh, K.; Sakurai, K.; Takano, S.
Tetrahedron Lett. 1987, 28, 2713.
(17) (a) Cossy, J., Thellend, A., Synthesis 1989, 753. (b) Labelle, M.;
Gravel, D. J. Chem. Soc., Chem. Commun. 1985, 105.
(18) Dehydration of condensation reactions was performed with the use
of a modified Dean-Stark apparatus in which the cooled distillate was
passed through 4-Å molecular sieves prior to return of the solvent to the
reaction mixture. Barta, N. S.; Paulvannan, K.; Schwarz, J. B.; Stille, J. R.
Synth. Commun. 1994, 24, 583.
Summary. The aza-annulation of chiral tetrasubstituted
â-enamino amides with activated acrylate derivatives resulted
in the efficient formation of quaternary asymmetric stereogenic

Aza-Annulation of â-Enamino Amides
J. Am. Chem. Soc., Vol. 120, No. 11, 1998 2499
Data for 39b: 80:20; hexanes:EtOAc, 0.26 g, 1.02 mmol, 75% yield;
1H NMR (300 MHz, CDCl3) (mixture of tautomers) δ 0.84-0.92 (m,
12 H), 1.57-1.80 (m, 6 H), 1.82-2.02 (m, 3 H), 2.02-2.22 (m, 6 H),
2.22-2.48 (m, 3 H), 3.18 (dd, J ) 5.4, 10.5 Hz, 1 H), 3.66 (s, 3 H),
3.69 (s, 3 H), 4.44-4.54 (m, 2 H), 5.82 (d, J ) 8.1 Hz, 1 H), 7.34 (d,
J ) 8.1 Hz, 1 H), 7.50 (d, J ) 8.1 Hz, 1 H); 13C NMR (75 MHz,
CDCl3) (mixture of tautomers) δ 17.4, 17.6, 18.6, 18.8, 21.6, 22.20,
22.24, 23.8, 24.0, 27.0, 27.1, 29.0, 30.7, 30.9, 31.1, 31.4, 41.87, 41.94,
51.8, 51.9, 55.4, 55.6, 56.3, 56.8, 56.9, 77.2, 96.6, 125.7, 128.6, 168.8,
168.9, 170.5, 171.9, 171.9, 172.1, 172.3, 209.5, 209.9; IR (neat) 1744,
1642, 1526 cm-1; HRMS calcd for C13H21NO4 m/z 255.1471, obsd m/z
255.1472.
31a. The 2-(benzyloxycarbonyl)amino derivative 29 was prepared
from rac-threonine.19 A mixture of 29 (1.13 g, 4.46 mmol) and
concentrated HCl (4 mL) was heated to reflux in EtOH (50 mL) for 5
h in a flask equipped for removal of H2O.18 After the reaction was
complete, solvent was removed, and the crude viscous product was
treated with a mixture of PCC (1.73 g, 8.03 mmol) and Celite (1:1
w/w) in CH2Cl2 (100 mL) at 0 °C. The mixture was stirred at 0 °C for
3 h, warmed gradually to room temperature, and stirred for an additional
10 h. The mixture was then filtered through a mixture of neutral
alumina/silica gel/Celite, 4:4:1. The filtrate was concentrated, and the
crude product (91% yield, two steps) was used in the aza-annulation
step without further purification.
J ) 3.5, 5.3 Hz, 1 H), 6.18 (br t, J ) 5.2 Hz, 1 H), 6.33 (q, J ) 7.1
Hz, 1 H), 7.06-7.29 (m, 10 H); 13C NMR (75 MHz, CDCl3) δ 14.1,
18.0, 24.3, 30.2, 30.4, 35.3, 44.0, 46.7, 49.4, 112.8, 125.2, 126.4, 127.6,
128.5, 128.8, 133.7, 137.6, 141.2, 169.4, 173.0; IR (CHCl3) 1665, 1634,
1512 cm -1; HRMS calcd for C25H28N2O2 m/z 388.2151, obsd m/z
388.2147.
Data for 21: 65:35; hexanes/EtOAc, 0.045 g, 0.12 mmol, 50% yield;
mp 57-58 °C; [R]25D ) -38.2° (c ) 0.4, CHCl3); 1H NMR (300 MHz,
CDCl3) δ 1.51 (d, J ) 7.2 Hz, 3 H), 1.60-1.94 (m, 2 H), 2.07-2.37
(m, 3 H), 2.57-2.78 (m, 3 H), 4.40 (dd, J ) 5.5, 14.7 Hz, 1 H), 4.46
(dd, J ) 5.7, 14.6 Hz, 1 H), 4.74 (t, 2.2 Hz, 1 H), 6.22 (q, J ) 7.2 Hz,
1 H), 6.24 (br t, J ) 5.5 Hz, 1 H), 7.10-7.30 (m, 10 H); 13C NMR (75
MHz, CDCl3) δ 14.4, 29.2, 29.3, 30.8, 36.3, 43.9, 49.8, 55.9, 110.2,
126.0, 126.9, 127.5, 127.7, 128.5, 128.9, 137.9, 139.5, 140.2, 169.5,
172.4; IR (CHCl3) 1669, 1628, 1509 cm-1; HRMS calcd for C24H26N2O2
m/z 374.1994, obsd m/z 374.2000.
1
Data for 33b: [R]23 ) +29.1° (c ) 0.35, EtOH); H NMR (300
D
MHz, CDCl3) δ 1.42 (d, J ) 6.9 Hz, 3 H), 1.84-2.14 (m, 2 H), 2.17
(s, 3 H), 2.47-2.66 (m, 2 H), 4.20-4.40 (m, 2 H), 4.96-5.16 (m, 3
H), 5.79 (d, J ) 7.5 Hz, 1 H), 6.93 (br s, 1 H), 7.05 (bt, J ) 5.4 Hz,
1 H), 7.13-7.40 (m, 15 H); 13C NMR (75 MHz, CDCl3) δ 21.6, 25.1,
28.5, 30.6, 44.0, 48.8, 67.1, 71.5, 126.1, 127.3, 127.5, 127.5, 128.1,
128.2, 128.5, 128.6, 128.6, 136.1, 137.4, 143.0, 154.9, 166.5, 170.5,
205.4; IR (CHCl3) 1715, 1669, 1534, 1497 cm-1; HRMS calcd for
C30H33N3O5 m/z 515.2420, obsd m/z 515.2402.
General Procedure For Aza-Annulation of â-Ketoamides and
â-Ketoesters. The primary amine or primary amine salt (0.5-5 mmol,
1.1 equiv) was taken up in toluene (0.05 M relative to the amine), and
the â-ketoamide or â-ketoester (1.0 equiv) was added at room
temperature. In the case of amine hydrochloride salt, NaHCO3 (1.5
equiv) was added, and for condensation that involved â-ketoesters, 0.02
mL of Et2O/BF3 (0.3 equiv) was added. The flask was fitted with a
modified Dean-Stark trap filled with 4-Å molecular sieves,18 and the
mixture was heated at reflux until the reaction was complete as
determined by NMR analysis (10-18 h). Solvent was removed under
reduced pressure. A solution of acrylate derivative was then added to
the intermediate enamine at room temperature, and the reaction mixture
was stirred at room temperature for 12-18 h (method A, mixed
anhydride of acrylic acid (freshly prepared from combination of sodium
acrylate (1.3 equiv) and ethyl chloroformate (1.3 equiv) for 1 h in dry
THF at -78 °C (0.05 M solution); method B, acrylic acid anhydride
(freshly prepared from combination of sodium acrylate or the 2-aceta-
mido acrylate derivative at -78 °C (1.3 equiv) and acryloyl chloride
(1.3 equiv) for 1 h in dry THF (0.05 M solution); method C, acryloyl
chloride (1.3 equiv) in dry THF (0.05 M solution)).7 Reactions were
quenched by the addition of H2O (for mixed anhydrides or acrylic
anhydride) or saturated aqueous NaHCO3 (acryloyl chloride), and the
mixture was extracted four times with either Et2O or EtOAc. The
combined organic fractions were dried (Na2SO4) and filtered, and the
solvent was evaporated under reduced pressure. The crude product
was purified by flash column chromatography (silica, 230-400 mesh,
eluent as indicated).
Data for 36: Rf ) 0.27; 90:5:5; Et2O/MeOH/petroleum ether, 0.15
g, 0.31 mmol, 67% yield; [R]24 ) -322.8° (c ) 0.36, CHCl3); mp
D
1
119-120 °C (sealed); H NMR (300 MHz, CDCl3) δ 1.45-1.72 (m,
3 H), 1.56 (d, J ) 7.1 Hz, 3 H), 1.80-1.97 (m, 3 H), 1.93 (s, 3 H),
2.03-2.15 (m, 3 H), 2.41 (dd, J ) 5.6, 13.2 Hz, 1 H), 4.10 (dd, J )
4.9, 14.5 Hz, 1 H), 4.28 (td, J ) 12.0, 5.9 Hz, 1 H), 4.42 (dd, J ) 6.2,
14.5 Hz, 1 H), 5.39 (t, J ) 3.9 Hz, 1 H), 5.55 (q, J ) 7.1 Hz, 1 H),
6.01 (br t, J ) 5.1 Hz 1 H), 6.53 (d, J ) 5.8 Hz, 1 H), 7.06-7.29 (m,
10 H); 13C NMR (75 MHz, CDCl3) δ 17.1, 18.0, 23.2, 24.0, 34.8, 36.5,
44.1, 47.8, 48.8, 55.4, 120.6, 126.3, 127.2, 127.7, 127.8, 128.6, 128.8,
135.7, 137.8, 141.3, 169.9, 170.4, 173.7; IR (CHCl3) 1665, 1509 cm-1
;
HRMS calcd for C27H31N3O3 m/z 445.2366, obsd m/z 445.2376.
Data for 37: Rf ) 0.22; 90:5:5; Et2O/MeOH/petroleum ether, 0.15
g, 0.31 mmol, 67% yield; [R]25 ) -134.9° (c ) 0.75, CHCl3); mp
D
1
116-117 °C (sealed); H NMR (300 MHz, CDCl3) δ 1.30-1.60 (m,
3 H), 1.48 (d, J ) 6.8 Hz, 3 H), 1.80-2.15 (m, 4 H), 1.92 (s, 3 H),
2.75 (dd, J ) 6.6, 13.2 Hz, 1 H), 3.97 (td, J ) 6.6, 14.4 Hz, 1 H), 4.36
(dd, J ) 4.2, 10.2 Hz, 1 H), 4.43 (dd, J ) 5.7, 16.2 Hz, 1 H), 5.04 (dd,
J ) 4.4, 5.6 Hz, 1 H), 6.14 (q, J ) 6.8 Hz, 1 H), 6.28 (t, J ) 6.0, 1 H),
6.48 (d, J ) 6.6 Hz, 1 H), 7.05-7.30 (m, 10 H); 13C NMR (75 MHz,
CDCl3) δ 14.5, 17.7, 23.2, 24.3, 35.5, 35.9, 44.2, 46.3, 50.3, 51.4, 112.9,
125.5, 126.5, 127.8, 128.6, 128.9, 133.6, 137.8, 141.0, 168.4, 170.3,
172.9; IR (CHCl3) 1669, 1640, 1511 cm-1; HRMS calcd for C27H31N3O3
m/z 445.2366, obsd m/z 445.2327.
Data for 13a: 65:35; hexanes:EtOAc, 0.10 g, 0.26 mmol, 74% yield;
1
98:2 ratio of diastereomers; H NMR (300 MHz, CDCl3) δ 1.27 (t, J
) 7.2 Hz, 3 H), 1.36-1.54 (m, 2 H), 1.54-1.71 (m, 2 H), 1.76 (d, J
) 6.9 Hz, 3 H), 1.92 (m, 1 H), 2.08-2.22 (m, 2 H)), 2.46 (m, 1 H),
2.52-2.70 (m, 2 H), 3.92 (dd, J ) 4.5, 18.3 Hz, 1 H)), 4.10 (dd, J )
5.4, 18.3 Hz, 1 H)), 4.20 (q, J ) 7.2 Hz, 2 H), 5.12 (dd, J ) 3.0, 5.1
Hz, 1 H), 6.51 (dd, J ) 12.3, 6.9 Hz, 1 H), 6.52 (br s, 1 H), 7.16-7.34
(m, 5 H), 13C NMR (75 MHz, CDCl3) δ 14.0, 14.5, 17.8, 24.3, 30.2,
30.3, 35.4, 41.7, 46.6, 49.4, 61.6, 113.1, 125.2, 126.4, 128.5, 133.3,
Data for 11: 80:20; hexanes/EtOAc, 1.98 g, 5.08 mmol, 70% yield;
D
1
[R]20 ) -93.5° (c ) 1.7, CHCl3); H NMR (300 MHz, CDCl3) δ
1.24-1.56 (m, 2 H), 1.51 (d, J ) 7.1 Hz, 3 H), 1.69 (ddd, J ) 6.4,
12.6, 12.8 Hz, 1 H), 1.76-1.92 (m, 2 H), 2.03 (m, 1 H), 2.16 (m, 1
H), 2.27 (ddd, J ) 1.9, 6.4, 13.1 Hz, 1 H), 2.41 (m, 1 H), 2.58 (ddd,
J ) 2.0, 6.4, 18.0 Hz, 1 H), 4.91 (dd, J ) 5.3, 3.0 Hz, 1 H), 5.02 (d,
J ) 13.0 Hz, 1 H), 5.09 (d, J ) 13.0, 1 H), 6.20 (q, J ) 7.2 Hz, 1 H),
7.04-7.30 (m, 10 H); 13C NMR (75 MHz, CDCl3) δ 14.6, 18.5, 24.4,
30.4, 31.0, 35.4, 46.7, 50.6, 67.1, 112.3, 125.6, 126.3, 128.3, 128.4,
128.5, 128.6, 133.6, 135.5, 142.4, 168.7, 174.2; IR (CHCl3) 1728, 1669,
1638, 1497 cm-1; HRMS calcd for C25H27NO3 m/z 389.1991, obsd m/z
389.2190.
Data for 17a: 65:35; hexanes/EtOAc, 0.31 g, 0.80 mmol, 99% yield;
mp 126-127 °C; [R]23D ) -149.5° (c ) 0.31, CH2Cl2); 1H NMR (300
MHz, CDCl3) δ 1.28-1.44 (m, 2 H), 1.39 (d, J ) 7.10 Hz, 3 H), 1.47-
1.65 (m, 2 H), 1.85 (m, 1 H), 2.10 (m, 1 H), 2.38-2.64 (m, 3 H), 4.32
(dd, J ) 5.4, 14.6 Hz, 1 H), 4.41 (dd, J ) 6.0, 14.6 Hz, 1 H), 4.96 (dd,
141.3, 169.36, 169.44, 173.6; IR (neat) 1748, 1660, 1632, 1507 cm-1
;
HRMS calcd for C22H28N2O4 m/z 384.2049, obsd m/z 384.2049.
Hydrogenation of 11. To a solution of 11 (1.65 g, 1.27 mmol) in
40 mL of EtOH was added 10% Pd/C (0.15 g). The reaction vessel
was flushed three times with H2, and the reaction was placed under a
balloon of H2. The reaction mixture was stirred for 4 h, filtered through
a pad of Celite, and concentrated under reduced pressure to give 12:
0.37 g, 1.24 mmol, 97% yield; [R]24D ) -116.1° (c ) 1.73, THF); mp
1
133 °C dec; H NMR (300 MHz, DMSO-d6) δ 1.30-1.58 (m, 3 H),
1.50 (m, 1 H), 1.63 (d, J ) 7.1 Hz, 3 H), 1.82 (m, 1 H), 2.01 (m, 1 H),
2.08-2.22 (m, 2 H), 2.38 (m, 1 H), 2.57 (m, 1 H), 4.90 (dd, J ) 2.6,
4.7 Hz, 1 H), 6.12 (q, J ) 7.1 Hz, 1 H), 7.12-7.32 (m, 5 H), 12.78 (br
s, 1 H); 13C NMR (75 MHz, DMSO-d6) δ 14.8, 18.3, 24.0, 30.2, 34.7,
(19) Greenstein J. P., Winitz M. Chemistry of the Amino Acids; John
Wiley & Sons: New York, 1961; Vol. 2, p 895.

2500 J. Am. Chem. Soc., Vol. 120, No. 11, 1998
BenoVsky et al.
45.9, 50.1, 55.0, 110.4, 125.4, 128.3, 129.4, 134.6, 142.5, 167.8, 175.6;
IR (KBr) 1725, 1601 cm-1; HRMS calcd for C18H21NO3 m/z 299.1522,
obsd m/z 299.1531.
2.49-2.79 (m, 3 H), 3.72 (dd, J ) 6.0, 18.0 Hz, 1 H), 3.80 (dd, J )
6.0, 18.0 Hz, 1 H), 4.09 (q, J ) 7.2 Hz, 2 H), 4.87 (dd, J ) 3.5, 4.7
Hz, 1 H), 5.00 (s, 1 H), 5.56 (t, J ) 5.4 Hz, 1 H), 6.25 (q, J ) 6.9 Hz,
1 H), 7.09-7.27 (m, 5 H); 13C NMR (75 MHz, CDCl3) δ 14.1, 15.4,
17.8, 24.6, 29.3, 30.2, 33.8, 41.8, 50.0, 52.4, 61.2, 113.5, 125.4, 126.5,
128.5, 135.6, 141.7, 156.8, 170.2, 171.2; IR (CHCl3) 1746, 1636, 1617,
1559 cm-1; HRMS (FAB) M + 1 calcd for C22H29N3O4 m/z 299.1522,
obsd m/z 299.1531.
General Procedure for Hydrolysis of Aza-Annulation Enamides.
After workup of the aza-annulation reaction, the crude enamide product
was taken up in a mixture of H2O (5.0 mL) and THF (25 mL), and
p-TsOH (0.03 g) was added. The mixture was stirred at room
temperature for 24 h, washed with an excess of saturated aqueous
NaHCO3, and extracted with 15 mL of Et2O three times, and the organic
layers were combined and dried (Na2SO4). Products were crystallized
from EtOAc/hexanes (1:1).
Crystal Structure of 36. A single crystal of 0.1 × 0.2 × 0.2 mm
dimensions was mounted on a glass capillary. Data were collected
using copper radiation, at 293 K, on a conventional Siemens P3 X-ray
diffractometer. Three standard reflections were measured every 97
reflections; no crystal decay was detected. Lorentz and polarization
corrections were applied to the data; no corrections were made for
absorption. The structures were solved by direct methods using the
SHELX86,20 and the remaining atoms were located in succeeding
difference Fourier synthesis. The hydrogen atoms were generated at
idealized positions and added to the structure factor calculations, but
not refined. The structure was refined in full-matrix least-squares on
F2 using SHELXTL,20 where the function minimized was ∑w(||Fo | -
2
|Fc2||)2, and the weight w is defined per the Killean & Lawrence
method21 with terms of 0.020 and 1.0. Atomic scattering factors and
the values for ∆f′ and ∆f′′ were taken from the International Tables
for X-ray Crystallography.22 Anomalous dispersion effects were
included in Fc.23 The final difference Fourier map showed a maximum
residual electron density of 0.78 e/Å3. The residual electron density is
attributed to a disordered diethyl ether molecule’s presence in the lattice.
The final indices were R1 ) 0.1002, wR2 ) 0.2688, where I > 2∂(I),
and R1 ) 0.1144, wR2 ) 0.2951 for all data. The ratio of data to
parameters is 2061:322, and the goodness-of-fit on F2 is 1.129. The
complete structural report is included in the Supporting Information
and has been submitted to the Cambridge Structural Data Centre.
Peptide Coupling with DPPA. To the carboxylic acid 12 (0.40 g,
1.37 mmol) in dry THF (35 mL) were added DPPA and freshly prepared
ethylglycine (0.16 g, 1.50 mmol) (generated by treatment of the
corresponding hydrochloride salt with Ba(OH)2 in CHCl3) under Ar at
0 °C. The mixture was stirred for 30 min, Et3N (0.22 g, 1.60 mmol)
was added, and the solution was gradually warmed to room temperature.
After the solution was stirred for an additional 12 h, the mixture was
diluted with 30 mL of EtOAc and washed sequentially with 20 mL of
5% HCl, 2 × 30 mL of H2O, 25 mL of saturated aqueous NaHCO3,
and 25 mL of brine. The organic layer was dried (Na2SO4), and then
purified by flash column chromatography (hexanes:EtOAc gradient:
65:35 to 50:50 to 0:100) to give 13 (0.077 g, 0.20 mmol, 15% yield)
and 14 (0.28 g, 0.70 mmol, 53% yield).
Data for 26: 0.30 g, 0.80 mmol, 82% yield; [R]20D ) +58.81 or (c
) 0.7, CHCl3); mp 159-160 °C; 1H NMR (300 MHz, CDCl3) δ 1.32
(s, 3 H), 1.38 (d, J ) 6.9 Hz, 3 H), 1.88-2.22 (m, 4 H), 2.11 (s, 3 H),
4.28 (dd, J ) 5.6, 14.8 Hz, 1 H), 4.34 (dd, J ) 5.5, 14.8 Hz, 1 H),
4.98 (m, 1 H), 5.75 (br d, J ) 7.5 Hz, 1 H), 6.69 (br t, J ) 4.9 Hz, 1
H), 7.11-7.29 (m, 10 H); 13C NMR (75 MHz, CDCl3) δ 19.7, 21.8,
26.5, 31.4, 31.8, 43.7, 48.8, 59.0, 126.1, 127.3, 127.5, 127.6, 128.60,
128.61, 138.0, 143.0, 171.2, 208.8; IR (CHCl3) 1717, 1628, 1541 cm-1
;
HRMS calcd for C23H28N2O3 m/z 380.2100, obsd m/z 380.2107.
1
Data for 28: 0.30 g, 0.68 mmol, 71% yield; H NMR (300 MHz,
CDCl3) δ 1.14 (t, J ) 7.1 Hz, 3 H), 1.33 (s, 3 H), 2.02-2.22 (m, 4 H),
2.11 (s, 3 H), 4.01-4.22 (m, 2 H), 4.32 (d, J ) 5.8 Hz, 2 H), 5.44 (d,
J ) 7.1 Hz, 1 H), 6.41 (br d, J ) 7.1 Hz, 1 H), 6.52 (br t, J ) 5.8 Hz,
1 H), 7.10-7.30 (m, 10 H); 13C NMR (75 MHz, CDCl3) δ 14.0, 19.6,
26.6, 31.2, 31.5, 43.9, 56.5, 59.0, 61.9, 127.2, 127.6, 127.7, 128.5, 128.7,
128.9, 136.5, 137.9, 170.8, 170.8, 171.2, 208.9; IR (CHCl3) 1746, 1709,
1646, 1532 cm-1; HRMS calcd for C25H30N2O5 m/z 438.2155, obsd
m/z 438.2157.
Acknowledgment. Support from the National Institutes of
Health (Grant GM44163) is gratefully acknowledged. Spectral
product characterization was performed on NMR instrumenta-
tion purchased in part with funds from NIH Grant 1-S10-
RR04750 and from NSF Grant CHE-88-00770. The invaluable
intellectual input and guidance of Professor William Reusch
(Michigan State University) is greatly appreciated. The authors
express their thanks to Leland O. Weigel and James E. Audia
(Eli Lilly and Company) for stimulating and valuable conversa-
tion.
Data for 14: hexanes:EtOAc ) 65:35 f 50:50 f EtOAc gradient,
1
0.28 g, 0.70 mmol, 53% yield; mp 85-86 °C; H NMR (300 MHz,
CDCl3) δ 1.19 (t, J ) 7.2 Hz, 3 H), 1.29 (m, 1 H), 1.42-1.52 (m, 2
H), 1.53 (d, J ) 6.9 Hz, 3 H), 1.59-1.87 (m, 3 H), 1.97 (m, 1 H),
(20) Sheldrick, G. M. shelxs86, program for crystal structure determi-
nation, University of Go¨ttingen, Federal Republic of Germany, 1986.
(21) Killean, R. C. G.; Lawrence, J. L. Acta Crystallogr. B 1969, 25,
1750.
(22) International Tables for X-ray Crystallography; Table 2.2B, Kynoch
Press: Birmingham, England, 1974 (present distributor: Kluwer Academic
Publishers: Dordrecht); Vol. 4, p 99, and Table 2.3.1, Vol. 4, p 149.
(23) Ibers, J. A.; Hamilton, W. C. Acta Crystallogr. 1964, 17, 781.
Supporting Information Available: Physical data for
selected compounds and copies of crystallographic data (11
pages). See any current masthead page for ordering information
and Web access instructions.
JA9729591