AmIno Acid Homologation by the Blaise Reaction: A new entry into nitrogen heterocycles

Article abstract of DOI:10.1021/jo900324d

(Chemical Equation Presented) A general strategy for the amino acid homologation via Blaise reaction and subsequent reduction is presented. This strategy involves the preparation of protected α-amino nitriles from the corresponding amino acids, followed by the zinc-mediated condensation of tert-butyl bromoacetate, to give the imidazolidones after iminozincate cyclization. Reduction gave the saturated imidazolidinones with cis or trans stereochemistry, depending on the reduction conditions. This strategy was applied to nonfunctionalized amino acids and to functionalized amino acids such as serine and aspartic acid. Additionally, acidic hydrolysis of cis or trans imidazolidinones to the corresponding chiral 4-aminopyrrolidones is described.

Full text of DOI:10.1021/jo900324d

Amino Acid Homologation by the Blaise Reaction: A New Entry
into Nitrogen Heterocycles
Cam Thuy Hoang, Francelin Bouille`re, Sine Johannesen, Ana¨ıs Zulauf, Ce´cilia Panel,
Annie Pouilhe`s, Didier Gori, Vale´rie Alezra,* and Cyrille Kouklovsky*
Laboratoire de Chimie des Proce´de´s et Substances Naturelles, Institut de Chimie Mole´culaire et des
Mate´riaux d’Orsay (UMR CNRS no. 8182), Baˆtiment 410, UniVersite´ Paris-Sud, F-91405 Orsay, France
cykouklo@icmo.u-psud.fr; Valezra@icmo.u-psud.fr
ReceiVed February 16, 2009
A general strategy for the amino acid homologation via Blaise reaction and subsequent reduction is
presented. This strategy involves the preparation of protected R-amino nitriles from the corresponding
amino acids, followed by the zinc-mediated condensation of tert-butyl bromoacetate, to give the
imidazolidones after iminozincate cyclization. Reduction gave the saturated imidazolidinones with cis or
trans stereochemistry, depending on the reduction conditions. This strategy was applied to nonfunction-
alized amino acids and to functionalized amino acids such as serine and aspartic acid. Additionally,
acidic hydrolysis of cis or trans imidazolidinones to the corresponding chiral 4-aminopyrrolidones is
described.
Introduction
(Figure 1). Moreover, 1,2-diamines are often used as ligands
for transition metals or as organic catalysts.7
The design and synthesis of new amino acid derivatives is a
continuous challenge in organic and bioorganic chemistry.
Within the ever-growing families of nonproteinogenic amino
acids, ꢀ-amino acids2 and γ-amino acids3 have emerged as
powerful tools in peptide science, as their corresponding peptides
possess interesting conformational properties. More recently, a
new class of functionalized amino acids, the ꢀ,γ-diamino acids
(of syn or anti configuration), has been the subject of closer
scrutiny. These compounds, which possess the 1,2-diamine
structural motif, may be found in several natural products such
as pseudotheonamide A14 or aminocarnitine,5 and are valuable
and flexible building blocks, as they can be precursors for both
ꢀ- or γ-peptides6 or various families of heterocyclic compounds
ꢀ,γ-Diamino acids are structurally related to statines (γ-amino
ꢀ-hydroxy acids), which can be found in several biologically
relevant natural products. Early work by Schostarez and co-
workers showed that the replacement of the hydroxyl group in
statine 1, a component of the antihypertensive peptide pepstatin,
by an amino group resulted in an enhancement of biological
activity (Figure 2).8 This set the stage for the interest in the
synthesis of various ꢀ,γ-diamino acids.
Several other approaches to the stereoselective synthesis of
ꢀ,γ-diamino acids have been reported, including intramolecular
conjugate addition of amines onto R,ꢀ-unsaturated esters,9
enolate condensation onto aminoacetals,10 or stereoselective
imine reduction.11 As we were interested in ꢀ,γ-diamino acids,
(6) (a) Simpson, G. L.; Gordon, A. H.; Lindsay, D. M.; Promsawan, N.;
Crump, M. P.; Mulholland, K.; Hayter, B. R.; Gallagher, T. J. Am. Chem. Soc.
2006, 128, 10638. (b) Cheng, R. P.; Gellman, S. H.; De Grado, W. F. Chem.
ReV. 2001, 101, 3219. (c) Wang, X.; Espinosa, J. F.; Gellman, S. H. J. Am.
Chem. Soc. 2000, 122, 4821.
(7) For a review on 1,2-diamines, see: Lucet, D.; Le Gall, T.; Mioskowski,
C. Angew. Chem., Int. Ed. 1998, 37, 2580.
(8) (a) Schostarez, H. J. J. Org. Chem. 1988, 53, 3628. (b) Thaisrivongs, S.;
Schostarez, H. J.; Pals, D. T.; Turner, S. R. J. Med. Chem. 1997, 30, 3762. See
also: (c) Arrowsmith, R. J.; Carter, K.; Dann, J. G.; Davies, D. E.; Harris, C. J.;
Morton, J. A.; Lister, P.; Bobinson, J. A.; Williams, D. J. J. Chem. Soc., Chem.
Commun. 1986, 755.
(1) Erasmus student from Åarhus University, Denmark.
(2) Juaristi, E.; Soloshonok, V. A. EnantioselectiVe Synthesis of ꢀ-Aminoacids,
2nd ed.; Wiley-Intersience: New York, 2005.
(3) (a) Seebach, D.; Kock, D. F.; Gla¨tti, A. Biopolymers 2006, 84, 23. (b)
Baldauf, C.; Gu¨nther, R.; Hofmann, H. J. Biopolymers 2005, 80, 675. (c) Seebach,
D.; Beck, A. K.; Bierbaum, D. J. Chem. BiodiVersity 2004, 1, 1111. (d) Hill,
D. J.; Mio, M. J.; Prince, R. B.; Hugues, T. S.; Moore, J. S. Chem. ReV. 2001,
101, 3893. (e) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem. ReV. 2001,
101, 3219.
(4) Shiori, T.; Irako, N. Chem. Lett. 2002, 130.
(5) (a) Shinagawa, S.; Kanamaru, T.; Harada, S.; Asai, M.; Okazaki, H.
J. Med. Chem. 1987, 30, 1458. (b) Calvisi, G.; Dell’Uomo, N.; De Angelis, F.;
Dejas, R.; Giannessi, F.; Tinti, M. O. Eur. J. Org. Chem. 2003, 4501.
(9) (a) Yoo, D.; Kwon, S.; Kim, Y. G. Tetrahedron: Asymmetry 2005, 16,
3762. (b) Doherty, A. M.; Kornberg, B. E.; Reily, M. D. J. Org. Chem. 1993,
58, 795; see also ref 4.
10.1021/jo900324d CCC: $40.75
Published on Web 05/07/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 4177–4187 4177

Hoang et al.
FIGURE 1. ꢀ,γ-Diamino acids and their synthetic derivatives.
carbonate) resulted in the fomation of an enamino ester instead
of the classical ꢀ-ketoester.16 This made the Blaise reaction a
convenient method for the synthesis of nitrogen-containing
compounds from nitriles.17
It appeared that the Blaise reaction could also be used as an
entry into ꢀ-amino acids by reduction of the subsequent enamino
ester. This strategy was applied to the synthesis of ꢀ-amino-γ-
hydroxyesters from cyanohydrins,18 ꢀ-lactams,19 and meth-
ylphenidate analogs20 and to the preparation of carbohydrate-
derived ꢀ-amino acids.21 In all cases, the enamino ester was
reduced by sodium cyanoborohydride under acidic conditions.22
As organozinc reagents are known to be tolerant of many
functional groups and of adjacent stereogenic centers, we
thought it would be possible to apply the Blaise reaction-enamino
ester reduction synthetic sequence to chiral R-amino nitriles
obtained from R-amino acids for the synthesis of stereochemi-
cally defined ꢀ,γ-diamino acids and derivatives thereof, ac-
cording to the sequence in Figure 4.
We have recently demonstrated the viability of this synthetic
route by the synthesis of chiral 4-aminopyrrolidones23 and its
application to the synthesis of the antipsychotic compound
nemonapride.24 In this article, we would like to present a full
account of our studies, including the transformation of func-
tionalized R-amino acids.
FIGURE 2. Statine 1 and its deoxyamino analogue.
we planned to prepare these compounds in a practical and
selective fashion by homologation of R-amino acids by the
Blaise reaction.
The Blaise reaction, first reported in 1901,12 involves the
condensation of the zinc enolate of a haloester (generally an
R-bromoester) onto a nitrile, giving a ꢀ-ketoester after acidic
hydrolysis (Figure 3).13 Despite being close to the Reformatsky
reaction with carbonyl compounds, the Blaise reaction has
received less attention because of low yields and side reactions,
especially self-condensation of the starting haloester. Several
improvements have been made in order to improve the yields
and to minimize side reactions: slow addition of the bromoester
and use of tert-butyl esters in place of methyl or ethyl esters
resulted in an increase in yields and purities. The activation of
zinc metal was also found to be beneficial. Typical experimental
procedures describe zinc dust preactivation by washing with
acid or in situ sonochemical or chemical activation with strong
Bronsted acids.14 These methods allow the obtention of ꢀ-ke-
toesters in good and reproducible yields. More recently, a
practical and highly efficient method for organozinc reagent
formation by decarboxylation of a malonate monoester in the
presence of zinc salts and subsequent condensation (the “de-
carboxylative Blaise reaction”) has been disclosed, making the
Blaise reaction an efficient synthetic tool.15
Results and Discussion
As depicted in Figure 4, the synthesis of ꢀ,γ-diamino acids
according to our strategy started with the synthesis of conve-
niently protected R-amino nitriles from R-amino acids. L-Valine
3 was chosen as model for early studies. Protection of L-valine
as its Cbz carbamate or as its N-phthalimide, followed by
The synthetic versatility of the Blaise reaction was further
improved when Kishi and co-workers discovered that hydrolysis
of the reaction mixture with a basic solution (50% potassium
(16) Hannick, S. M.; Kishi, Y. J. Org. Chem. 1983, 48, 3833.
(17) See, inter alia: Erian, A. W. J. Prakt. Chem. 1999, 341, 147.
(18) Syed, J.; Forster, S.; Effenberger, F. Tetrahedron: Asymmetry 1998, 9,
805.
(19) (a) Mauduit, M.; Kouklovsky, C.; Langlois, Y.; Riche, C. Org. Lett.
2000, 2, 1053. (b) Mauduit, M.; Kouklovsky, C.; Langlois, Y. Eur. J. Org. Chem.
2000, 1595.
(10) (a) Giri, N.; Petrini, M.; Profeta, R. J. Org. Chem. 2004, 69, 7303. (b)
Potier, P.; Christine, C.; Ikhiri, K.; Ahond, A.; Al Mourabit, A.; Poupat, C.
Tetrahedron 2000, 56, 1837. (c) Merino, P.; Franco, S.; Merchan, F. L.; Tejero,
T. J. Org. Chem. 2000, 65, 5575. (d) McDonald, S. J. F.; Clarke, G. D. E.;
Dowle, M. D.; Harrison, L. A.; Hodgson, S. T.; Inglis, G. G. A.; Johnson, M. R.;
Shah, P.; Upton, R. J.; Walls, S. B. J. Org. Chem. 1999, 64, 5166.
(11) Martin-Martinez, M.; Bartolome´-Nebreda, J. M.; Gomez-Monterrey, I.;
Gonzalez-Muniz, R.; Garcia-Lopez, M. T.; Ballaz, S.; Fortuno, A.; Del Rio, J.;
Herranz, R. J. Med. Chem. 1997, 40, 3402.
(12) (a) Blaise, E. C.R. Hebd. Se´ances Acad. Sci. 1901, 132, 478. (b) Kagan,
H. B.; Suen, Y. H. Bull. Soc. Chim. Fr. 1966, 1819.
(13) Reviews: (a) Rao, H. S. P.; Rafi, S.; Padmavathy, K. Tetrahedron 2008,
64, 9037. (b) Ocampo, R.; Dolbier, W. R., Jr. Tetrahedron 2004, 60, 9325.
(14) Shin, H.; Choi, B. S.; Lee, K. K.; Choi, H.-W.; Chang, J. H.; Lee, K. W.;
Nam, D. H.; Kim, N.-S. Synthesis 2004, 2629.
(20) Deutsch, H. M.; Ye, X.; Shi, Q.; Liu, Z.; Schweri, M. M. Eur. J. Med.
Chem. 2001, 36, 303.
(21) Dondoni, A.; Massi, A.; Minghini, E. Synlett 2006, 539.
(22) Borch, R. F.; Berstein, M. D.; Dupont Durst, H. J. Am. Chem. Soc.
1971, 93, 2897. For recent applications of the cyanoborohydride reduction of
enamino esters, see: (a) Alladoum, J.; Dechoux, L. Tetrahedron Lett. 2005, 46,
8203. (b) Gurjar, M. K.; Karmakar, S.; Mohapatra, D. K.; Phalgune, U. D.
Tetrahedron Lett. 2002, 43, 1897. (c) Furstero, S.; Pina, B.; de la Torre, M. G.;
Navarro, A.; de Arellano, C. R.; Simon, A. Org. Lett. 1999, 1, 977.
(23) Hoang, C. T.; Alezra, V.; Guillot, R.; Kouklovsky, C. Org. Lett. 2007,
9, 2521.
(15) Lee, J. H.; Choi, B. S.; Chang, J. H.; Lee, H. B.; Yoon, J.-Y.; Lee, J.;
Shin, H. J. Org. Chem. 2007, 72, 10261.
(24) Hoang, C. T.; Nguyen, V. H.; Alezra, V.; Kouklovsky, C. J. Org. Chem.
2008, 73, 1162.
4178 J. Org. Chem. Vol. 74, No. 11, 2009

Aminoacid Homologation by the Blaise Reaction
FIGURE 3. Blaise reaction and its outcome.
FIGURE 4. Strategy for the synthesis of ꢀ,γ-diamino acids from R-amino acids by the Blaise reaction.
SCHEME 1. Initial Attempts for the Blaise Reaction with r-Amino Nitriles
primary amide formation and phosphorus oxychloride-mediated
dehydration, gave nitriles 6a,b in good yields and without
racemization. These nitriles were used for the Blaise reaction
(Scheme 1): slow addition of methyl bromoacetate to a
suspension of zinc dust (preactivated by washing with acid,
followed by water and ethanol, then drying under vacuum) in
refluxing THF containing nitrile 6a resulted in the formation
of enamino ester 7, albeit in low yield. None of the in situ zinc
activation methods allowed better yields. Replacement of methyl
bromoacetate for tert-butyl bromoacetate gave only a slightly
better yield (30%).
was not stable under the Blaise reaction conditions, the
organozinc reagent reacting with the imide function rather than
the nitrile function.
Enaminoester 7 was then submitted to reduction according
to the previously described conditions.22 The corresponding ꢀ,γ-
diamino ester 8 was obtained as a 2:1 mixture of diastereomers,
which were separated but not assigned.
Although this synthetic route gave rise to the expected ꢀ,γ-
diamino acid 8 (in its protected form) in a short synthetic
sequence, it suffered from low yields in the condensation
reaction and low stereoselectivity in the reduction step. Since
we thought that these problems arose from the presence of an
N-H group in the substrate, we decided to add a second stable
protecting group on the amine nitrogen before undertaking the
These experiments suggested that R-amino nitriles having a
free N-H group were not suitable substrates for the Blaise
reaction. Unfortunately, the N-phthalimido R-amino nitrile 6b
J. Org. Chem. Vol. 74, No. 11, 2009 4179

Hoang et al.
TABLE 1. Benzylation of r-Amino Nitrile 6a
entry
base (equiv)
X (equiv)
solvent
conditionsa
time
12 h
yield (%)
ee (%)b
sm (%)
17
ee (%)b
99
1
2
3
4
5
6
NaH (3)
NaH (1)
NaH (0.8)
NaH (1)
KH (1)
Br (3)
Br (3)
I (3)
I (10)
I (10)
I (10)
THF
THF
THF
THF
THF
DMF
A
A
A
B
B
B
86
74
67
94
94
84
9
21
62
78
87
99
12 h
1 h 30 min
1 h 30 min
10 min
NaH (1)
10 min
a A: addition of benzylating reagent onto the anion; B: addition of the anion onto the benzylating reagent. b Determined by HPLC on chiral column.
SCHEME 2. Blaise Reaction on Diprotected Amino Nitrile 9
Blaise reaction. Therefore, N-Cbz amino nitrile 6a was benzyla-
ted (NaH, benzyl bromide, THF) to give the amino nitrile 9 in
high yield (Table 1).25 However, determination of enantiomeric
excess by HPLC on a chiral column indicated that extensive
racemization occurred, the diprotected amino nitrile being
obtained with only 9% ee (Table 1, entry 1). Reducing the
amount of base led to an incomplete reaction, and the benzylated
product was obtained in 21% ee (entry 2). Switching to the more
reactive benzyl iodide gave 9 with increased ee, together with
enantiomerically pure remaining starting material (entry 3).
Therefore, we concluded that racemization occurred on product
9, probably due to the presence of the unreacted amide anion.
It was then important to perform a fast alkylation reaction: using
a large excess of benzyl iodide and performing inverse addition
of the anion onto the benzylating reagent dramatically improved
the ee’s (entries 4 and 5). Finally, using DMF as solvent at
room temperature gave the enantiomerically pure benzylated
amino nitrile in a good yield (entry 6).
With the enantiomerically pure R-amino nitrile 9 in hand,
the Blaise reaction was attempted using tert-butyl bromoacetate.
Optimization studies showed that the best method for activation
of zinc dust was with 1,2-dibromoethane. Addition of a few
drops of tert-butyl bromoacetate just after zinc activation ensured
that the reaction had started by the appearance of a green color.16
Satisfactory yields were obtained this time, thus demonstrating
the need for a double nitrogen protection. However, two
products were obtained, which could be separated by chroma-
tography: the enamino ester 10 and the imidazolidinone 11, the
latter arising from a cyclization of the intermediate iminozincate
onto the carbamate protecting group (Scheme 2). It is remarkable
that the sole iminozinc reagent reacts on the carbamate and not
the organozinc reagent. Treating the crude product with sodium
hydride in THF led to complete conversion into the imidazo-
lidinone, which was then isolated in 76% yield.26
HPLC analysis of both products showed complete conserva-
tion of stereochemical integrity, thus demonstrating that neither
the Blaise reaction nor the cyclization induced racemization.
The Blaise reaction was then performed with R-amino nitriles
derived from other R-amino acids: L-alanine, L-phenylalanine,
L-leucine, L-proline, and L-serine (Scheme 3). Amino nitriles
15-17 were prepared according to the described procedure,23
and compounds 18 and 19 according to known procedures.27,28
Blaise reaction followed by cyclization gave cleanly the
imidazolidinones 20-24 in good yields and in enantiomerically
pure form. In some cases, complete cyclization to the imida-
zolidinone occurred, without adding sodium hydride to the crude
mixture. All imidazolidinones were obtained as Z isomers, as
determined by 1H NMR NOE experiments, except for proline-
derived compound 23, which was obtained as a mixture of Z
and E isomers. With L-serine-derived R-amino nitrile 19, it was
possible to isolate in good yield the noncyclized enamino ester
25 by omitting the sodium hydride mediated reaction.
Another interesting substrate would be the R-amino nitrile
29 derived from L-aspartic acid, in which a second electrophilic
group is present. Indeed, the intermediate iminozincate could
cyclize in a 5-exo-trig fashion, either on the carbamate protecting
group or on the ester carboxyl group. Accordingly, the known29
R-amino nitrile 29 was prepared in four steps from L-aspartic
acid methyl ester 26 (Scheme 4). As we faced difficulties in
introducing a benzyl group on the amine nitrogen, the Blaise
(26) Control experiments showed that cyclization occurred before potassium
carbonate workup. Adding sodium hydride before this workup did not improve
the yields.
(27) Cobb, A. J. A.; Shaw, D. M.; Longbottom, D. A.; Gold, J. B.; Ley,
S. V. Org. Biomol. Chem. 2005, 3, 84.
(25) All compounds described from now on were synthesized in both racemic
and enantiomerically pure form. The enantiomeric purity of each intermediate
was checked by HPLC on chiral column.
(28) Cossu, S.; Conti, S.; Giacomelli, G.; Falorni, M. Synthesis 1994, 1429.
(29) Ressler, C.; Nagarajan, G. R.; Kirissawa, M.; Kashelikar, D. V. J. Org.
Chem. 1971, 36, 3960.
4180 J. Org. Chem. Vol. 74, No. 11, 2009

Aminoacid Homologation by the Blaise Reaction
SCHEME 3. Blaise Reaction with Amino Acid Derived Amino Nitriles
SCHEME 4. Blaise Reaction with L-Aspartic Acid Derived r-Amino Nitrile
SCHEME 5. Stereoselective Reduction of Imidazolidinone
11 with Sodium Cyanoborohydride
reaction was attempted on the monoprotected substrate. Fortu-
nately, the reaction proceeded cleanly to give exclusively the
lactam 30 resulting from cyclization of the iminozincate onto
the ester, no cyclization onto the carbamate to give 31 being
observed.30 Compound 30, which was obtained as a single Z
isomer and in enantiomerically pure form, is actually (as its
enantiomer) a fragment of the macrocyclic peptide microscle-
rodermin E,31 which possesses interesting cytotoxic properties.
In view of the rigid, cyclic structure of imidazolidinones, their
reduction might provide better stereoinduction than that of
acyclic enamino esters, as observed in the early studies. Indeed,
reduction of imidazolidinone 11 with sodium cyanoborohydride
at -78 °C and under acidic conditions gave the imidazolidinone
32, together with a small amount of transesterified methyl ester
33 (Scheme 5). Both compounds were obtained as single
isomers. Treatment of tert-butyl ester 32 with acidic methanol
gave the methyl ester, identical to compound 33, thus proving
(30) In contrast, Blaise reaction with the L-glutamic acid derived amino nitrile
led to a mixture of products arizing from both possible cyclizations. Zulauf, A.;
Alezra, V.; Kouklovsky, C. Unpublished results.
(31) (a) Isolation: Schmidt, E. W.; Faulkner, D. J. Tetrahedron 1998, 54,
3043. (b) Total synthesis: Zhu, J.; Ma, D. Angew. Chem., Int. Ed. 2003, 42,
5348.
that both products have the same configuration. The cis
relationship between the two ring substituents was determined
J. Org. Chem. Vol. 74, No. 11, 2009 4181

Hoang et al.
TABLE 2. Sodium Cyanoborohydride Reduction of Imidazolidinones 20-23
entry
substrate
R1
Me
iBu
Bn
yield (%)
cis:transa
1
2
3
4
20
21
22
23
74
55
55
67
1.6:1
1:2
1:1.5
1:2
-(CH2)3-
a
1
Ratio determined by H NMR.
SCHEME 6. Determination of Absolute Configuration of Enaminoester 38
on the basis of NOE experiments in 1H NMR and later
confirmed by X-ray analysis.23
stereogenic center was accomplished with the Mosher protocol
(Scheme 6): 32,33 the major diasteromer 38a was acylated with
both enantiomers of MTPA-Cl. Since analysis of the resulting
diastereomeric amides 39a,b was hampered by the presence of
rotamers, the carbamate protecting group was removed by
hydrogenation. This reaction was accompanied by the opening
Although reduction of L-valine-derived imidazolidinone oc-
curred with complete stereoselectivity, other substrates gave
disappointing results, with very low selectivity (Table 2).
Furthermore, the cis/trans selectivity was inverted for L-leucine,
L-phenylalanine, and L-proline derived substrates (entries 2-4;
for a method for stereomer assignement, vide infra). Attempts
to improve selectivity using other reducing reagents such as
sodium triacetoxyborohydride, zinc borohydride, DIBAL-H, or
L-Selectride were unsuccessful, no reaction occurring. Reduction
of lactam 30 with sodium cyanoborohydride led to low
conversion and stereoselectivity.
Reduction of bicyclic enamino ester 24 derived from L-serine
was thwarted by concomitant reduction of the oxazolidine ring.
Alternatively, reduction of the acyclic enamino ester 25 gave a
separable 3:1 mixture of diastereomers under the same condi-
tions. Determination of the absolute configuration of the new
1
of the oxazolidine ring, to give N-isopropylamines 40a,b. H
NMR analysis showed differences in chemical shifts of +0.022
for the ester group and of -0.111 for the primary hydroxyl
group, thus indicating the S configuration for the new stereogenic
center.
The stereoselectivity of this reduction could be explained
using a Felkin-Anh model, in which the intermediate imminium
ion adopts a conformation with the iminium CdN bond
perpendicular to the polar and bulky N-Cbz group. Attack of
(32) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512.
(33) The enantiomeric purity of each of stereoisomers of 38 was determined
by HPLC after trifluoroacetylation (see Supporting Information).
4182 J. Org. Chem. Vol. 74, No. 11, 2009

Aminoacid Homologation by the Blaise Reaction
SCHEME 7. Model for Stereoinduction in the Reduction of Enaminoester 25
SCHEME 8. Hydrogenation of Lactam 30
SCHEME 9. Birch Reduction of Imidazolidinone 11
the hydride anti to this group, results in the formation of the
(R,S) product 38a as the major diastereomer (Scheme 7).34
As reduction with hydride reagents gave unsufficient stereo-
selectivity, other reduction conditions were tested. Hydro-
genation35-38 of imidazolidinone over palladium or platinum
catalysts resulted only in migration of the alkene into the
imidazolidinone ring, thus removing the stereogenic center in
the imidazolone substrate. Hydrogenation of compound 25 led
to extensive degradation including reductive opening of the
oxazolidine ring. However, lactam 30, obtained from L-aspartic
acid, proved to be a good substrate for hydrogenation (Scheme
8). Although a short (20 min) hydrogenation over palladium
catalyst removed selectively the carbamate protecting group,
overnight reaction gave the saturated lactam 42 as a single
diastereomer.39 The trans configuration was determined after
extensive NMR studies and can be explained by the coordination
of the free amine group (after in situ hydrogenolysis of the
benzyl carabamate) to the catalyst, thus directing hydrogen
attack on the si face. This exceptionally high selectivity for
reduction allows the stereoselective preparation of a new type
of γ-amino acid, which could be used as building block for
γ-peptide synthesis or as chiral organic catalyst.
1,4-Reduction of R,ꢀ-unsaturated carbonyl or carboxylic
derivatives under dissolving metal conditions (the Birch reduc-
tion) is a popular method for stereoselective reduction or enolate
formation and has been widely used in synthesis.40 However,
the Birch reduction of enamino esters has received little
attention: to the best of our knowledge, only a single experiment
was reported, in which a cyclic enamino ester was reduced with
lithium in liquid ammonia (stereochemistry not determined).41
As we were looking for accurate conditions for imidazolidinone
reduction, we attempted Birch reduction (Scheme 9). Treatment
of imidazolidinone 11 with excess sodium in THF/NH3 gave a
mixture of three products after analysis of the crude product by
1H NMR: the reduced ester 43, the corresponding carboxylic
acid 44, and the reduced alcohol 45. In each product, the double
bond has been cleanly reduced. The presence of carboxylic acid
is due to partial saponification of the tert-butyl ester. As drying
of the ammonia prior reaction improved the ratio of products,
we assume that the presence of the primary alcohol is due to
traces of moisture in the reaction mixture. For the sake of
isolation, the crude mixture was saponified (2.5 N NaOH, THF,
reflux) to give the carboxylic acid 44 and the alcohol 45 in good
overall yield. Moreover, each compound was obtained as a
single diastereomer and was enantiomerically pure. Oxidation
of the alcohol 45 (PDC, DMF) returned the acid 44, thus proving
that both compounds have the same configuration. The Birch
reduction of imidazolidinone is therefore totally stereoselective.
(34) For a discussion about the stereoselectivity of the reduction of related
ketones, see: Liang, X.; Andersch, J.; Bols, M. J. Chem Soc., Perkin Trans. 1
2001, 2136, and references therein.
(35) Heterogeneous hydrogenation of enamino esters: (a) Hashiguschi, S.;
Natsugari, H.; Ochiai, M. J. Chem. Soc., Perkin Trans. 1 1988, 2345. (b) Hiyama,
T.; Nishide, K.; Kobayashi, K. Chem. Lett. 1984, 361.
(36) Rhodium-catalyzed hydrogenation of (Z)-enamino esters: (a) Yan, Y.;
Zhang, X. Tetrahedron Lett. 2006, 47, 1567. (b) Kubryk, M.; Hansen, K. B.
Tetrahedron: Asymmetry 2006, 17, 205. (c) Dai, Q.; Yang, W.; Zhang, X. Org.
Lett. 2005, 7, 5343. (d) Hu, X.-P.; Zheng, Z. Org. Lett. 2005, 7, 419. (e) Wu,
H.-P.; Hoge, G. Org. Lett. 2004, 6, 3645. (f) Hsiao, Y.; Rivera, N. R.; Rosner,
T.; Krska, S. W.; Njolito, E.; Wang, F.; Sun, Y.; Armstrong, J. D.; Grabowski,
E. J. J.; Tillyer, R. D.; Spindler, F.; Malan, C. J. Am. Chem. Soc. 2004, 126,
9918. (g) Holz, J.; Monsees, A.; Jiao, H.; You, J.; Komarov, I. V.; Fischer, C.;
¨
Drauz, K.; B; orner, A. J. Org. Chem. 2003, 68, 1701. (h) Pena, D.; Minaard,
A. J.; de Vries, J. G.; Feringa, B. L. J. Am. Chem. Soc. 2002, 124, 14552. (i)
Tang, W.; Zhang, X. Org. Lett. 2002, 4, 4159. (j) Lee, S.-G.; Zhang, Y. J. Org.
Lett. 2002, 4, 2429.
(37) Rhodium-catalyzed hydrogenation of (E)-enamino esters: (a) Tang, W.;
Wang, W.; Chi, Y.; Zhang, X. Angew. Chem., Int. Ed. 2003, 42, 3509. (b)
Yasutake, M.; Gridnev, I. D.; Higashi, N.; Imamoto, T. Org. Lett. 2001, 3, 1701.
(c) Zhu, G.; Chen, Z.; Zhang, X. J. Org. Chem. 1999, 64, 6907. (d) Lubell,
W. D.; Kitamura, M.; Noyori, R. Tetrahedron: Asymmetry 1991, 2, 543.
(38) Ruthenium-catalyzed hydrogenation of enamino esters: Wu, J.; Chen,
X.; Guo, R.; Yueng, C.-H.; Chan, A. S. C. J. Org. Chem. 2003, 68, 2490.
(39) The enantiomeric purity of lactam 42 was determined after conversion
into its Fmoc carbamate (see Supporting Information).
J. Org. Chem. Vol. 74, No. 11, 2009 4183

Hoang et al.
TABLE 3. Birch Reduction of Imidazolidinones 11, 20-23
entry
substrate
acid (yield %)
alcohol (yield %)
dea (%)
1
2
3
4
11
20
21
22
44 (52)
46 (66)
48 (47)
50 (50)
45 (27)
47 (14)
49 (15)
51 (18)
>99
>99
>99
>99
a
1
Ratio determined by H NMR.
1
SCHEME 10. Chemical Shits in H NMR for Both Isomers of Valine-Derived Imidazolidinones
The same behavior was observed with Birch reduction of
other imidazolidinones (Table 3), except for substrate 24, for
which hydrolysis of the oxazolidine ring was observed. A
mixture of carboxylic acid and alcohol, both obtained as single
diastereomers, was obtained with all substrates except for
proline-derived imidazolidinone, which gave only the carboxylic
acid. However, the latter compound proved to be very difficult
to extract from the reaction mixture and had to be converted to
the methyl ester to be isolated.
the cis isomers and 2.9-3.3 ppm for the trans isomers, whereas
chemical shifts for Hb are 4.0-4.1 ppm for the cis isomers and
3.6-3.7 for the trans isomer. For proline derivatives, the
chemical shifts were sensibly higher due to the conformation
of the bicyclic molecule. This consistency in the chemical shift
allows easy diastereomer assignement.
The origin of the trans selectivity for the Birch reduction
could be explained according to the generally admitted mech-
anism:40b first electron transfer gives rise to a sp3 dianionic
radical in which the R substituent and the enolate are in an anti
configuration due to steric repulsion and to stabilization of the
SOMO orbital with the parallel neigbouring C-C bond (Scheme
11). The second electron transfer affords a trianionic species
that is reprotonated with retention of configuration to give the
trans isomer. Deprotonation of the nitrogen atom prevents
ꢀ-elimination of the enolate. Therefore, we assume that the Birch
reduction proceeds under kinetic control, the preferred config-
uration of the carbon atom bearing the anion being the source
of the stereoinduction.
Although this reduction was found to be general for all
imidazolidinones mentioned above, attempts for the reduction
of lactam 30 under the same conditions only led to degradation.
Determination of the relative configuration was made on the
1
basis of H NMR analysis with comparison with cyanoboro-
hydride reduction products. Thus, the tert-butyl ester of the cis-
disubstituted valine-derived imidazolidinone 32 (for which
configuration was secured by X-ray analysis) was removed by
saponification to give the free acid 53 (Scheme 10). Comparison
of 1H NMR spectra with acid 44 (obtained after Birch reduction)
showed differences in the chemical shifts for the protons next
to the nitrogen atoms, thus indicating that 53 and 44 were
diastereomers. Therefore, the trans relative configuration was
assigned to the Birch reduction products.
In conclusion, we have demonstrated that enamino esters can
be reduced under dissolving metal conditions in good yields
and with a high degree of stereoselectivity. This reaction proved
to be general and useful for the synthesis of ꢀ,γ-diamino acids.
Hydrolysis of the substituted imidazolidinones was ac-
complished under strongly acidic conditions:43,44 cis-disubsti-
tuted imidazolidinone 32 was hydrolyzed in 3 N HCl at 90 °C
This difference in chemical shifts was found to be consistent
for all products:42 Ha has a chemical shift of 3.4-3.7 ppm for
(40) Reviews: (a) Venturello, P.; Barbero, M. In Science of Synthesis; Georg
Thieme Verlag: Stuttgart, 2005; 8b, p 881. (b) Pradhan, S. K. Tetrahedron 1986,
42, 6351. (c) Caine, D. Org. React. (New York) 1976, 23, 1.
(41) Adams, A. D.; Schlessinger, R. H.; Tata, J. R.; Venit, J. J. J. Org. Chem.
1986, 51, 3070.
(43) Acidic hydrolysis of cyclic ureas: (a) Shimizu, M.; Kamei, M.; Fujisawa,
T. Tetrahedron Lett. 1995, 36, 8607. (b) Misiti, D.; Santaniello, M.; Zappia, G.
Synth. Commun. 1992, 22, 883. (c) Dunn, P. J.; Ha¨ner, R.; Rapoport, H. J. Org.
Chem. 1990, 55, 5017.
(42) For a table of chemical shifts, see Supporting Information.
4184 J. Org. Chem. Vol. 74, No. 11, 2009

Aminoacid Homologation by the Blaise Reaction
TABLE 4. Hydrolysis of trans-Imidazolidinones
entry
substrate
R1
substrate relative configuration
conditions
product
yield (%)
1
2
3
4
5
32
46
48
52
37
iPr
Me
iBu
-(CH2)3-
-(CH2)3-
cis
trans
trans
trans
3 N HCl, 90 °C, 12 h
3 N HCl, 140 °C, 120 h
3 N HCl, 140 °C, 120 h
3 N HCl, 90 °C, 12 h
3 N HCl, 90 °C, 12 h
54a
54b
54c
54d
54d
99a
99
99
99
99
cis/trans (2:1)
a Isolated as the free base after purification on Dowex-H+ column.
SCHEME 11. Hypothesis for the Stereoselectivity of the Birch Reduction
SCHEME 12. Hydrolysis of cis-Imidazolidinone 32
for the synthesis of new, amino acid derived heterocyclic
structures. We have shown that the Blaise reaction is compatible
with chiral, functionalized amino nitriles obtained from amino
acid, and we have introduced the Birch reduction for the
stereoselective reduction of the corresponding cyclic enamino
ester. Deprotection to ꢀ,γ-diamino acids as well as their use in
peptide synthesis is currently under investigation in our
laboratory.
for 12 h (Scheme 12). However, instead of the expected ꢀ,γ-
diamino acid, the 4-aminopyrrolidone 54a was obtained, result-
ing fom lactamization. This compound was obtained neutral after
purification on an ion-exchange resin (Dowex H+). Although
this lactam was very resistant toward acidic or basic hydrolysis,
it could be easily reduced to the corresponding 3-aminopyrro-
lidine.24
Hydrolysis of the trans-disubstituted imidazolidinones (result-
ing from Birch reduction) was carried out under similar
conditions (Table 4), except for the alanine- and leucine-derived
compounds, which proved to be more resistant (entries 2 and
3). In this case, it was necessary to use harsher conditions (140
°C, 5 days) to ensure complete conversion. Nevertheless, high
yields of the corresponding pyrrolidones were obtained. In all
cases the hydrolysis reaction proceeded without racemization
or epimerization. For instance, hydrolysis of a cis/trans mixture
of proline-derived imidazolidinone gave the same stereochemical
ratio of pyrrolidones (entry 5). Ring opening of these lactams
to ꢀ,γ-diamino acids is still under investigation. Since it is
known that N-Boc lactams are very prone to basic hydrolysis,45
a change in nitrogen protecting group could provide a solution.
In conclusion, we have demonstrated that the Blaise
reaction-stereoselective reduction sequence is a valuable tool
Experimental Section
General Methods. See Supporting Information.
General Procedure for the Synthesis of Monoprotected r-Ami-
no Nitriles. A solution of N-Cbz L-amino acid (prepared from 50
mmol of L-amino acid according to standard procedure) in dry THF
(100 mL) was cooled to 0 °C, and Et3N (7 mL, 49.8 mmol) was
added. After 15 min at 0 °C, ethyl chloroformate (4.8 mL, 50.2
mmol) was added dropwise, and the clear solution stirred for 30
min, before addition of 30% aqueous ammonia solution (15 mL).
After 30 min at 0 °C, the mixture was diluted with H2O (60 mL)
and extracted with Et2O (2 × 100 mL). The combined organic layers
were dried (MgSO4), filtered, and concentrated under reduced
pressure. The solid residue was used in the next step without any
further purification.
A solution of the crude amide in pyridine (100 mL) was cooled
to 0 °C, and phosphorus oxychloride (4.42 mL, 47.3 mmol) was
added dropwise. The mixture was warmed to room temperature
and stirred for additional 4 h. The reaction mixture was quenched
with a 2 M aqueous HCl solution (100 mL) and extracted with
Et2O (2 × 100 mL). The combined organic layers were washed
with saturated CuSO4 aqueous solution, water, and brine, then dried
(MgSO4), filtered, and concentrated under reduced pressure to give
the pure R-amino nitrile.
(S)-N-Benzyloxycarbonyl-2-amino-isopentanenitrile (6a). Ob-
tained in 45% overall yield from L-valine (5.22 g, 22.5 mmol);
yellowish solid; mp 53 °C; H NMR (360 MHz, CDCl3) δ (ppm)
1.06 (d, J ) 7.0 Hz, 3H), 1.09 (d, J ) 7.0 Hz, 3H), 2.04 (t, J ) 8.8
(44) Basic hydrolysis of cyclic ureas: (a) Jones, R. C. F.; Schofield, J.
J. Chem. Soc., Perkin Trans. 1 1990, 375. (b) Hoffmann, K.; Melville, D. B.;
du Vigneaud, V. J. Biol. Chem. 1941, 137, 207.
1
(45) Flynn, D. L.; Zelle, R. E.; Grieco, P. A. J. Org. Chem. 1983, 48, 2424.
J. Org. Chem. Vol. 74, No. 11, 2009 4185

Hoang et al.
Hz, 1H), 4.46-4.56 (m, 1H), 5.15 (s, 2H), 5.64 (d, J ) 8.8 Hz
1H), 7.30-7.44 (m, 5H); 13C NMR (90 MHz, CDCl3) δ (ppm) 17.8,
18.3, 31.5, 48.8, 67.4, 117.6, 128.0, 128.2, 128.4, 135.5, 155.3;
HRMS (electrospray) (M + Na) calculated for C13H16N2O2Na:
255.1104, found 255.1103; IR (thin film, CH2Cl2) ν (cm-1) 3322.5,
MHz, CDCl3) δ (ppm) 0.87 (d, J ) 7.0, 3H), 0.95 (d, J ) 7.0,
3H), 1.45 (s, 9H), 2.00-2.15 (m, 1H), 3.92-3.97 (m, 1H), 3.97
(d, J ) 15.2, 1H), 4.75 (s, 1H), 5.02 (d, J ) 15.2, 1H), 7.17-7.35
(m, 5H), 9.34 (s, 1H); 13C NMR (50 MHz, CDCl3) δ (ppm) 16.1,
17.1, 28.2, 30.1, 44.5, 63.5, 79.9, 88.6, 127.7, 127.9, 128.7, 136.0,
153.4, 156.9, 167.8; HRMS (electrospray) (M + Na) calculated
353.1836, found 353.1868; IR (thin film, CH2Cl2) ν (cm-1) 3377.9,
1740.3, 1677.4, 1633.8. Anal. Calcd for C19H26N2O3: C, 69.06; H,
7.93, N, 8.48. Found: C, 68.82; H, 7.95; N, 8.33; [R]20D ) -81 (c
0.81, CH2Cl2); HPLC analysis: ee > 99%; Chiralpak AD column
(hexane/ethanol 85:15, 1 mL/min, 254 nm); retention times of
racemic mixture: 6.8 min (R) and 8.7 min (S).
2244.8, 1702.3; [R]20 ) -55 (c 1.13, CH2Cl2).
D
General Procedure for the Synthesis of N,N-Diprotected r-Ami-
no Nitriles. Preparation of Benzyl Iodide. Benzyl bromide (7.2 mL,
60 mmol) was added to a solution of sodium iodide (18 g, 120
mmol) in acetone (80 mL). The mixture was stirred for 24 h in the
dark at room temperature, then quenched with water (50 mL) and
extracted with Et2O (2 × 100 mL). The combined organic layers
were dried (MgSO4), filtered, and concentrated under reduced
pressure to afford the pure product as yellow oil. Yield: quantitative
(7.5 mL, 60 mmol).
Sodium hydride (152 mg, 6.3 mmol) was washed three times
with n-pentane, and then DMF (10 mL) was added. The suspension
was cooled to 0 °C, and the amino nitrile (6.0 mmol) dissolved in
dry DMF (8 mL) was added. The mixture was stirred for 30 min
and then transferred by cannula into a flask containing benzyl iodide
(7.5 mL, 60 mmol) in dry DMF (15 mL). The mixture was stirred
for additional 10 min at room temperature. The reaction was
quenched with saturated aqueous NH4Cl solution and extracted with
Et2O (2 × 30 mL). The combined organic layers were dried
(MgSO4), filtered, and concentrated under reduced pressure. The
residue was purified by flash chromatography (pentane/Et2O 8/2)
to give the diprotected amino nitrile.
(S,Z)-tert-Butyl 2-(3-(benzyloxycarbonylamino)-5-oxopyrrolidin-
2-ylidene)-acetate (30). Prepared according to procedure B. Yield:
74% (281 mg, 0.81 mmol); yellowish solid; mp 126 °C; 1H NMR
(250 MHz, CDCl3) δ (ppm) 1.20 (s, 9H), 2.38 (dd, J ) 10.4 Hz,
J ) 17.4 Hz, 1H), 2.85 (dd, J ) 4.9 Hz, J ) 17.4 Hz, 1H), 4.95
(m, 1H), 5.12 (s, 3H), 5.60 (m, 1H), 7.20-7.40 (m, 5H), 9.75 (br
s, 1H); 13C NMR (90 MHz, CDCl3) δ (ppm) 28.1, 35.8, 48.9, 67.3,
80.6, 92.6, 128.1, 128.3, 128.5, 135.7, 151.1, 155.7, 167.3, 173.8;
HRMS (electrospray) (M + Na) calculated for C18H22N2O5Na:
369.1426, found 369.1421; IR (thin film, CH2Cl2) ν (cm-1) 3424,
3214, 1719, 1701, 1265, 1154; [R]20 ) -72.4 (c 0.99, CH2Cl2).
D
General Procedure for the Reduction of Enaminoesters with
Sodium Cyanoborohydride. Preparation of dry HCl 2 N (2 mL).
Acetyl chloride (0.3 mL, 4.7 mmol) was added dropwise to MeOH
(2 mL) at 0 °C, and the solution was stirred for15 min at 0 °C.
Enaminoester (0.50 mmol) was dissolved in CH2Cl2/MeOH (9
mL, 2/1 mixture), and a small amount of bromocresol green was
added. The mixture was cooled to -78 °C, and dry HCl (1.6 mL)
solution was added. To the resulting yellow solution was added
sodium cyanoborohydride (47 mg, 0.75 mmol). After 15 min at
-78 °C, the mixture was warmed to room temperature and stirred
for additional 1 h 30. The reaction was quenched with 2.5 M NaOH
aqueous solution to give a deep blue solution and extracted with
CH2Cl2 (2 × 20 mL). The organic layers were washed with
saturated brine, then dried over MgSO4, filtered, and concentrated
under reduced pressure. The residue was purified by flash
chromatography.
(S)-N-Benzyl-N-Benzyloxycarbonyl-2-amino-isopentanenitrile (9).
1
Yield: 84% (1.61 g, 5 mmol); colorless oil; H NMR (400 MHz,
CDCl3, 330 K) δ (ppm) 0.80 (d, J ) 6.8 Hz, 3H), 1.05 (d, J ) 5.9,
3H), 2.07-2.16 (m, 1H), 4.40 (d, J ) 15.7, 1H), 4.54 (d, J ) 9.8,
1H), 4.71 (d, J ) 15.7, 1H), 5.17 (s, 2H), 7.22-7.28 (m, 10H);
13C NMR (90 MHz, CDCl3) δ (ppm) 18.3, 19.2, 30.7, 49.7, 55.3,
68.0, 117.1, 127.5, 127.8, 128.0, 128.3, 135.4, 136.6, 155.5; HRMS
(electrospray) (M + Na) calculated 345.1573, found 345.1584; IR
(CH2Cl2) ν (cm-1) 2241.3, 1705.7. Anal. Calcd for C20H22N2O2:
C, 74.51; H, 6.88; N, 8.69. Found: C, 74.24; H, 6.83; N, 8.52;
[R]20 ) -48 (c 1.15, CH2Cl2); HPLC analysis: ee ) 99%; (S,S)
D
Whelk-01 column (hexane/ethanol 99:1, 1 mL/min, 254 nm);
retention times of racemic mixture: 15.7 min (S) and 17.5 min (R).
General Procedure for the Blaise Reaction of r-Amino Ni-
triles. Procedure A. To a stirred suspension of Zn dust (503 mg,
7.7 mmol) in refluxing THF (10 mL) were added 1,2-dibromoethane
(0.3 mL, 3.5 mmol) and few drops of tert-butyl bromoacetate. To
the resulting greenish suspension was then rapidly added a solution
of amino nitrile (1.1 mmol) in THF (3 mL). A solution of tert-
butyl bromoacetate (0.61 mL, 4.2 mmol) in dry THF (3 mL) was
added dropwise in 10 min, and the mixture was stirred for additional
1 h at reflux. After cooling to room temperature, the reaction was
quenched with 50% K2CO3 aqueous solution (10 mL). After stirring
for 15 min, the resulting mixture was filtered through Celite, and
the precipitate was washed with Et2O. The aqueous layer was
extracted with Et2O (2 × 20 mL), and the combined extracts were
dried (MgSO4), filtered, and concentrated under reduced pressure.
The crude material was dissolved in dry THF (10 mL) and cooled
to 0 °C before addition of sodium hydride (32 mg, 1.3 mmol). The
mixture was warmed up to room temperature and stirred for
additional 2 h. The reaction was quenched by introduction of
saturated NH4Cl aqueous solution, and the mixture was extracted
with Et2O (2 × 20 mL). The combined organic layers were dried
over MgSO4, filtered, and concentrated under reduced pressure. The
crude material was purified by flash chromatography (heptane/ethyl
acetate 8:2) to give pure 2-imidazolidinone.
(4R,5S)-5-(1-Methylethyl)-1-phenylmethyl-imidazolidine-2-one-
4-acetic Acid tert-Butyl Ester (32). A single diastereomer was
observed by H NMR of the crude material. Purification by flash
1
chromatography (heptane/AcOEt 1:1) gave 32 (cis relative config-
uration). Yield: 84% (139 mg, 0.42 mmol); white solid; mp 90 °C;
1H NMR (360 MHz, CDCl3) δ (ppm) 0.97 (d, J ) 7.0, 6H), 1.41
(s, 9H), 1.82-1.94 (m, 1H), 2.49-2.55 (m, 2H), 3.38 (dd, J )
3.5, 7.9, 1H), 3.92-4.02 (m, 1H), 3.95 (d, J ) 15.8, 1H), 4.99 (d,
J ) 15.8, 1H), 5.18 (s, 1H), 7.18-7.31 (m, 5H); 13C NMR (90
MHz, CDCl3) δ (ppm) 17.5, 20.5, 27.8, 28.0, 36.0, 46.0, 51.1, 60.9,
81.3, 127.0, 127.5, 128.3, 137.1, 162.4, 170.6; HRMS (electrospray)
(M + H) calculated for C19H29N2O3: 333.2173, found 333.2181;
IR (thin film, CH2Cl2) ν (cm-1) 3234.9, 2973.6, 1699.4. Anal. Calcd
for C19H28N2O3: C, 68.65; H, 8.49; N, 8.43; O, 14.44. Found: C,
68.62; H, 8.48; N, 8.28; O, 14.59; [R]20D ) +14 (c 1.08, CH2Cl2);
HPLC analysis: ee > 99%; (S,S) Whelk-01 column (hexane/ethanol
100:3, 1 mL/min, 225 nm); retention times of racemic mixture:
33.4 min (4S, 5R) and 39.0 min (4R, 5S).
(S,Z)-tert-Butyl 2-(3-aimno-5-oxopyrrolidin-2-ylidene) Acetate
(41). A solution of enamino ester 30 (0.44 mmol) in dry methanol
(2 mL) was hydrogenated at 1 bar for 25 min in the presence of
10% Pd/C (85.5 mg, 0.79 mmol). Filtration of the catalyst through
a Celite pad and concentration under reduced pressure gave pure
compound 41 as a colorless oil. Yield: quantitative (93 mg, 0.44
mmol); 1H NMR (250 MHz, CDCl3) δ (ppm) 1.23 (s, 9H), 1.7 (br
s, 2H), 2.21 (dd, J ) 5.5 Hz, J ) 17.7 Hz, 1H), 2.79 (dd, J ) 8.9
Hz, J ) 17.7 Hz, 1H), 4.10 (dd, J ) 5.5 Hz, J ) 8.9 Hz, 1H), 5.10
(s, 1H), 9.70 (br s, 1H); 13C NMR (90 MHz, CDCl3) δ (ppm) 28.1,
38.6, 50.2, 80.4, 91.7, 160.6, 167.6, 174.5; HRMS (electrospray)
(M + H) calculated for C10H17N2O3: 213.1236; found 213.1234;
Procedure B. This procedure is identical to the previous one,
except that the crude material is not treated with sodium hydride
but directly purified by flash chromatography (pentane/Et2O 7:3).
(S)-1-Benzyl-4-(Z)-tert-butyloxycarbonylidene-5-(1-methylethyl)-
imidazolidine-2-one (11). Prepared according to procedure A. Yield:
72% (261 mg, 0.8 mmol); white solid; mp 105 °C; 1H NMR (250
4186 J. Org. Chem. Vol. 74, No. 11, 2009

Aminoacid Homologation by the Blaise Reaction
IR (thin film, CH2Cl2) ν (cm-1) 3347, 1748, 1682, 1265; [R]20
-34.7 (c 1.02, CH2Cl2).
)
) 7.0, 3H), 0.78 (d, J ) 7.0, 3H), 1.85-2.02 (m, 1H), 2.32 (d, J
) 6.3, 2H), 2.93 (t, J ) 3.5, 1H), 3.69-3.77 (m, 1H), 3.83 (d, J )
15.5, 1H), 4.80 (d, J ) 15.5, 1H), 6.76 (s, 1H), 7.14-7.30 (m,
5H), 10.32 (br s, 1H); 13C NMR (62.5 MHz, CDCl3) δ (ppm) 14.8,
17.4, 27.8, 41.9, 44.5, 46.7, 64.0, 127.5, 127.8, 128.7, 136.5, 161.9,
D
((2R,3S)-3-Amino-5-oxopyrrolidin-2-yl)-2-acetic Acid tert-Bu-
tyl Ester (42). A solution of enamino ester 30 (0.25 mmol) in dry
methanol (3 mL) was hydrogenated at 1 bar for 15 h in the presence
of 10% Pd/C (53 mg, 0.5 mmol). Filtration of the catalyst through
a Celite pad and concentration under reduced pressure gave pure
compound 42 as a yellowish solid. Yield: 68% (36 mg, 0.17 mmol);
mp 67 °C; 1H NMR (250 MHz, CDCl3) δ (ppm) 1.44 (s, 9H), 1.96
(br s, 2H), 2.10 (dd, J ) 6.1 Hz, J ) 17.1 Hz, 1H), 2.35 (dd, J )
9.2 Hz, J ) 16.5 Hz, 1H), 2.62 (dd, J ) 4.9 Hz, J ) 16.5 Hz, 1H),
2.65 (dd, J ) 7.9 Hz, J ) 17.1 Hz, 1H), 3.32 (m, 1H), 3.57 (dt, J
) 4.3 Hz, J ) 9.2 Hz, 1H), 6.51 (br.s, 1H); 13C NMR (90 MHz,
CDCl3) δ (ppm) 27.9, 39.5, 40.0, 53.1, 59.5, 81.4, 170.3, 175.3;
HRMS (electrospray) (M + Na) calculated for C10H19N2O3Na:
237.1186; found 237.1210; IR (thin film, CH2Cl2) ν (cm-1) 3423,
174.8; HRMS (electrospray) (M
+ Na) calculated for
C15H20N2O3Na: 299.1366, found 299.1370; IR (thin film, CH2Cl2)
ν (cm-1) 3333.3, 1704.2, 1651.9; [R]20 ) -65 (c 0.45, CH2Cl2).
D
General Procedure for the Cleavage of the Imidazolidines-2-
ones. A solution of the imidazolidinone in 3 M HCl solution (60
mL for 1 mmol of substrate) was heated in an open flask at the
appropriate temperature (see Table 4). After cooling to room
temperature, the mixture was concentrated under reduced pressure.
(4R,5S)-4-amino-5-(1-methylethyl)-1-phenylmethyl-pyrrolidin-
2-one (54a). Obtained after purification on Dowex 50X4 resin (10
g, 200-400 mesh), Yield: 99% (96 mg, 0.41 mmol); yellow oil;
1H NMR (250 MHz, D2O) δ (ppm) 0.59 (d, J ) 7.0, 3H), 0.82 (d,
J ) 7.0, 3H), 1.99-2.12 (m, 1H), 2.40 (dd, J ) 1.9, 19.0, 1H),
2.95 (dd, J ) 8.8, 19.0, 1H), 3.43 (dd, J ) 1.3, 3.2, 1H), 3.77 (dd,
J ) 1.9, 8.8, 1H), 4.16 (d, J ) 15.8, 1H), 4.66 (d, J ) 15.8, 1H),
7.21-7.36 (m, 5H); 13C NMR (62.5 MHz, D2O) δ (ppm) 13.7,
17.1, 27.3, 36.6, 44.0, 44.8, 68.6, 128.1, 128.5, 128.9, 135.2, 173.7;
HRMS (electrospray) (M + Na) calculated for C13H20N2ONa:
255.1468, found 255.1468; IR (MeOH) ν (cm-1) 3414.6, 1671.1;
3054, 2984, 1700, 1265, 1154; [R]20 ) +57.2 (c 1.02, CH2Cl2).
D
General Procedure for the Birch Reduction of Enaminoesters.
NH3 gas (25 mL) was passed through a tube containing NaOH
pellets and condensed into a cooled (-40 °C) glass tube containing
some pieces of sodium metal, giving a deep blue solution. The
cooling bath was removed, and dried ammonia gas was transferred
into a cooled (-40 °C) trinecked flask containing a solution of
enamino ester (0.5 mmol) in dry THF (5 mL). Sodium pieces (69
mg, 3 mmol) were added, and the deep blue solution was stirred
for an additional 30 min at -40 °C. The reaction was quenched by
addition of solid NH4Cl (268 mg, 5 mmol) and warmed to room
temperature to remove NH3 gas. Aqueous 10% NaOH solution (25
mL) was added, and the mixture stirred at 80 °C for 12 h. After
cooling to room temperature, the solution was acidified to pH 1
with 2 M HCl solution and extracted with AcOEt (2 × 20 mL).
The combined organic layers were dried over MgSO4, filtered, and
concentrated under reduced pressure. The residue was purified by
flash chromatography (AcOEt/AcOH 99:1) to give two products,
the carboxylic acid (less polar product) and the primary alcohol
(more polar compound).
[R]20 ) -26 (c 1.0, D2O).
D
Acknowledgment. This research was supported by the CNRS
(doctoral grant to C.T.H.) and Ministe`re de la Recherche et de
l’Enseignement Supe´rieur (doctoral grant to F.B). We thank
Felix Bellande for preparing compound 19.
Supporting Information Available: Full characterization
data for compounds 12-14, 15-17, 20-25, 29, 34-40, 45-53,
54b-d. Table of NMR data for imidazolidinones 29, 31, 33,
41, 43, 45, 47, 49, 50; copies of NMR spectra for all compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(4S,5S)-5-(1-Methylethyl)-1-phenylmethyl-imidazolidine-2-one-
4-acetic Acid (44). Yield: 52% (84 mg, 0.3 mmol); yellow solid;
mp 130-131 °C; 1H NMR (250 MHz, CDCl3) δ (ppm) 0.76 (d, J
JO900324D
J. Org. Chem. Vol. 74, No. 11, 2009 4187