Design, synthesis, and evaluation of a new generation of modular nucleophilic glycine equivalents for the efficient synthesis of sterically constrained α-amino acids

Article abstract of DOI:10.1021/jo0616198

A new generation of modular achiral glycine equivalents have been evaluated with respect to their synthetic utility for the production of tailor-made, sterically constrained α-amino acids, which proved to be the most efficient approach developed to date for the synthesis of symmetrical α,α-disubstituted-α-amino acids. Among the new series of achiral glycine equivalents, one was found to be a superior glycine derivative for the Michael additions with various (R)- or (S)-N-(E-enoyl)-4-phenyl-1,3- oxazolidin-2-ones representing a general and practical synthesis of sterically constrained β-substituted pyroglutamic acids. In particular, the application of these complexes allowed for the preparation of several β-substituted pyroglutamic acids which include electron-releasing and sterically demanding substituents in the structure thus increasing the synthetic efficiency and expanding the generality of these Michael addition reactions.

Full text of DOI:10.1021/jo0616198

Design, Synthesis, and Evaluation of a New Generation of Modular
Nucleophilic Glycine Equivalents for the Efficient Synthesis of
Sterically Constrained r-Amino Acids
Trevor K. Ellis,† Hisanori Ueki,† Takeshi Yamada,†,‡ Yasufumi Ohfune,‡ and
Vadim A. Soloshonok*,†
Department of Chemistry and Biochemistry, UniVersity of Oklahoma, Norman, Oklahoma 73019, and
Graduate School of Science, Department of Material Science, Osaka City UniVersity, Sugimoto, Sumiyoshi,
Osaka 558-8585, Japan
Vadim@ou.edu
ReceiVed August 3, 2006
A new generation of modular achiral glycine equivalents have been evaluated with respect to their synthetic
utility for the production of tailor-made, sterically constrained R-amino acids, which proved to be the
most efficient approach developed to date for the synthesis of symmetrical R,R-disubstituted-R-amino
acids. Among the new series of achiral glycine equivalents, one was found to be a superior glycine
derivative for the Michael additions with various (R)- or (S)-N-(E-enoyl)-4-phenyl-1,3-oxazolidin-2-
ones representing a general and practical synthesis of sterically constrained â-substituted pyroglutamic
acids. In particular, the application of these complexes allowed for the preparation of several â-substituted
pyroglutamic acids which include electron-releasing and sterically demanding substituents in the structure
thus increasing the synthetic efficiency and expanding the generality of these Michael addition reactions.
Introduction
resistance in proteins6 has also proven to be essential in the
modern era of de novo designed peptides.7
The synthesis of tailor-made1 amino acids has been at the
forefront of synthetic organic chemistry research for some time.2
This is due, in most part, to the breadth of their application,
which ranges from food additives to sophisticated pharmaceu-
ticals.3 This is especially true for sterically constrained or
polyfunctional unnatural amino acids due to the profound
influence their physical and chemical properties may impart on
vital biologically active systems.4 Their ability to influence the
three-dimensional shape,5 biological activity,4 and degradative
It has been these pharmacological and medicinally relevant
applications that have provided the motivation to devise a sound
synthetic methodology that is general, environmentally benign,
and industrially attractive for the synthesis of these profoundly
influential compounds. To date, there are numerous approaches
to the symmetric and asymmetric synthesis of sterically
constrained amino acids,8 although each seems to be plagued
by limitations. Therefore, after evaluating the previously
published methodologies, the most straightforward, practical,
and general approach seems to be the homologation of nucleo-
philic glycine equivalents. Although there are many deficiencies
† University of Oklahoma.
‡ Osaka City University.
10.1021/jo0616198 CCC: $33.50 © 2006 American Chemical Society
Published on Web 10/04/2006
8572
J. Org. Chem. 2006, 71, 8572-8578

New Modular Nucleophilic Glycine EquiValents
SCHEME 1
that need to be addressed with respect to this synthetic approach,
it seems that most involve the insufficient design of the glycine
equivalents themselves. The majority of the commercially
available glycine equivalents may be characterized as having
unacceptable properties such as chemical instability, poor
industrial scalability, or poor economic feasibility.
Results and Discussion
Therefore, to overcome these shortcomings, our laboratory
has recently developed a new series of modified glycine
equivalents based on a modular skeleton9 that has provided
superior chemical reactivity, stability, solubility, and cost
efficiency, even compared to their Ni(II)-containing counterparts
previously introduced by our laboratory10 and others.11 The
modular design concept of the glycine equivalents has opened
(1) The rapidly growing list of amino acids isolated from various natural
sources makes the terms unnatural, unusual, noncoded, or nonproteinogenic
amino acids, which are most frequently used in the literature, dependent
on the success of specific scientific achievements. For instance, amino acids
containing the most xenobiotic element fluorine have been shown to be
synthesized by microorganisms. Moreover, recent spectacular developments
in the generation of bacteria with an expanded genetic code make all above-
mentioned definitions rather obsolete. Therefore, the time-independent term
tailor-made, meaning rationally designed/synthesized amino acids with
presupposed physical, chemical, 3D-structural, and biological features, in
the absence of a better definition, seems to be a more appropriate use as a
common name for such amino acids. (a) Kukhar’, V. P.; Soloshonok, V.
A., Eds.; Fluorine-Containing Amino Acids. Synthesis and Properties; John
Wiley and Sons Ltd.: Chichester, 1994. (b) Mehl, R. A.; Anderson, J. C.;
Santoro, S. W.; Wang, L.; Martin, A. B.; King, D. S.; Horn, D. M.; Schultz,
P. G. J. Am. Chem. Soc. 2003, 125, 935.
various avenues of opportunity for a whole host of reaction
conditions, given the advantage of virtually complete substrate
adaptability.
There are several modules that must be combined to assemble
the desired Ni(II) complexes; however, a simple, general, and
reliable synthetic route has been devised for this purpose
(Scheme 1). The first step is the coupling to the proper
“phenone” module with the corresponding “acid” module to
form the intermediate “acetamide” module. This is accomplished
by the slow addition of a 2-aminophenone 1/acetonitrile solution
to slurry of bromoacetyl bromide 2, potassium carbonate, and
acetonitrile. This procedure is usually accomplished with yields
greater than 90% and nearly ideal chemical purity. Next, the
ligand 6a-f or 7a-f will be produced from the substitution of
the bromine atom from the acetamide module intermediate 3
or 4 with virtually any secondary amine 5a-f. This reaction is
also accomplished in an acetonitrile solution with potassium
carbonate, to liberate the hydrobromic salt of the amine
produced; however, the application of heat is generally utilized
to accelerate the reaction rate. It should also be noted that the
previous two reactions may be conducted via a one-pot reaction;
however, we have found it more convenient to store large
amounts of the acetamide intermediates 3 or 4 for their eventual
conversion to their corresponding ligands 6a-f or 7a-f as
needed. The final reaction of the glycine equivalent synthesis
is the Schiff base formation, between the ligand 6a-f or 7a-f
and glycine 8, coupled with the simultaneous Ni(II) complex-
ation accomplished in methanol with heat to accelerate the
reaction. Potassium hydroxide is utilized to catalyze the imine
formation as well as to neutralize the corresponding acid formed
from the reaction.
(2) (a) Najera, C. Synlett 2002, 1388. (b) Park, K. H.; Kurth, M. J.
Tetrahedron 2002, 58, 8629.
(3) (a) Breuer, M.; Ditrich, K.; Habicher, T.; Hauer, B.; KeBeler, M.;
Sturmer, R.; Zelinski, T. Angew. Chem., Int. Ed. 2004, 43, 788. (b) Wang,
L.; Brock, A.; Herberich, B.; Schultz, P. G. Science 2001, 292, 498. (c)
Gallos, J. K.; Sarli, V. C.; Massen, Z. S.; Varvogli, A. C.; Papadoyanni, C.
Z.; Papasyrou, S. D.; Argyropoulos, N. G. Tetrahedron 2005, 61, 565. (d)
Wang, M.-X.; Lin, S.-J.; Liu, J.; Zheng, Q.-Y. AdV. Synth. Catal. 2004,
346, 439.
(4) (a) Zhang, X.; Ni, W.; Van der Donk, W. A. J. Org. Chem. 2005,
70, 6685. (b) O’Donnell, M. J.; Alsina, J.; Scott, W. L. Tetrahedron Lett.
2003, 44, 8403.
(5) (a) Karle, I. L.; Kaul, R.; Rao, R. B.; Raghothama, S.; Balaram, P.
J. Am. Chem. Soc. 1997, 119, 12048. (b) Marshall, B. R.; Hodgkin, E. E.;
Langs, D. A.; Smith, G. D.; Zabrocki, J.; Leplawy, M. T. Proc. Natl. Acad.
Sci. U.S.A. 1990, 31, 129. (c) Toniolo, C.; Crisma, M.; Bonora, G. M.;
Benedetti, E.; Di Blasio, B.; Pavone, V.; Pedone, C.; Santini, A. Biopolymers
1991, 31, 129. (d) Huston, S. E.; Marshall, G. R. Biopolymers 1994, 34,
75. (e) Aleman, C. Biopolymers 1994, 34, 841. (f) Yokum T. S.; Gauthier,
T. J.; Hammer, R. P.; McLaughlin, M. L. J. Am. Chem. Soc. 1997, 119,
1167. (g) Rossi, P.; Felluga, F.; Tecilla, P.; Formaggio, F.; Crisma, M.;
Toniolo, C.; Scrimin, P. J. Am. Chem. Soc. 1999, 121, 6948.
(6) (a) Peptide Secondary Structure Mimetics. Tetrahedron Symposia-
in-Print; Kahn, M., Ed.; Tetrahedron 1993, 49, No. 50 and references therein.
(b) Machetti, F.; Cordero, F. M.; De Sarlo, F.; Papini, A. M.; Alcaro, M.
C.; Brandi, A. Eur. J. Org. Chem. 2004, 2928. (c) Giannis, A.; Kolter, T.
Angew. Chem., Int. Ed. Engl. 1993, 32, 1244.
(7) Special Issue. Protein Design. Degrado, W. F., Guest Ed. Chem. ReV.
2001, 101.
Now that the ease and cost efficiency for the synthesis of
the new generation of glycine equivalents has been established,
the versatility, chemical stability, and chemical reactivity need
to be evaluated. Therefore, it was determined that these glycine
equivalents 9a-f or 10a-f should first be subjected to extreme
conditions to confirm the stability of the complex as well as to
ensure that the methylene moiety, introduced into the ligand
6a-f or 7a-f via the addition of bromoacetyl bromide 2, is
nonenolizable and therefore would not interfere with the overall
atom economy of the process or the recyclable nature of the
ligand 6a-f or 7a-f. Therefore, to evaluate these possibilities
(8) Special Issue. Asymmetric Synthesis of Novel Sterically Constrained
Amino Acids. Tetrahedron Symposia-in-Print #88; Hruby, V. J., Soloshonok,
V. A., Guest Eds. Tetrahedron 2001, 57, No. 30 and references therein.
(9) Ellis, T. K.; Ueki, H.; Soloshonok, V. A. Tetrahedron Lett. 2005,
46, 941.
(10) (a) Soloshonok, V. A.; Cai, C.; Hruby, V. J. Org. Lett. 2000, 2,
747. (b) Ueki, H.; Ellis, T. K.; Martin, C. H.; Soloshonok, V. A. Eur. J.
Org. Chem. 2003, 10, 1954.
(11) Belokon, Y. N.; Bespalova, N. B.; Churkina, T. D.; Cisarova, I.;
Ezernitskaya, M. G.; Harutyunyan, S. R.; Hrdina, R.; Kagan, H. B.;
Kocovsky, P.; Kochetkov, K. A.; Larionov, O. V.; Lyssenko, K. A.; North,
M.; Polasek, M.; Peregudov, A. S.; Prisyazhnyulk, V. V.; Vyskocil, S. J.
Am. Chem. Soc. 2003, 125, 12860.
J. Org. Chem, Vol. 71, No. 22, 2006 8573

Ellis et al.
SCHEME 2
to the production of R,R-disubstituted-R-amino acids, the
flexibility of the modular design did not seem to have a major
impact on the outcome of the reactions. However, as our focus
was diverted toward the synthesis of another structurally and
chemically unique group of amino acids, â-substituted pyro-
glutamic acids, the adaptability of these complexes proved to
be extremely advantageous. The reaction conditions for the
Michael addition reactions, employed for the synthesis of the
â-substituted pyroglutamic acids, are undoubtedly milder;
however, the level of intricacy is multiplied as two stereogenic
centers will be established by the formation of one bond.
Therefore, the homogeneity of the enolate formed on the glycine
structure, provided by the Ni(II)-complexed glycine equivalent,
as well as the steric constraints surrounding the site of addition
becomes of paramount importance to aid in the organization of
the transition state thereby providing the potential for increased
yields as well as diastereoselectivities. Given the complex nature
of these reactions compounded by the necessity of milder bases,
these reactions provide an excellent platform to compare the
reactivity of various derivatives within the series of the new
glycine equivalents. This is especially true given the simple
workup procedures utilized, as established from previous efforts
as the crude reaction mixtures were merely poured over a
solution of icy 5% acetic acid to quench the reactions followed
by simple filtration.
The initial reactivity comparison studies were conducted to
determine the effect of the alkyl group of the amino moiety of
the complexes with respect to their differences in lipophilicity,
as well as their electronic and steric properties.13 The two (S)-
N-(E-enoyl)-4-phenyl-1,3-oxazolidin-2-ones 16a,b14 chosen for
these experiments included a para-methoxyphenyl or an iso-
propyl substituent in the â position to limit the rate, by virtue
of their corresponding electronic and steric contributions, of the
corresponding reactions to increase the accuracy while deter-
mining the relative reactivity of each of the complexes. Each
of the following reactions discussed in this section was
conducted under a set of standard conditions at ambient
temperature, which included the use of commercial-grade DMF
as the solvent and 15 mol % of DBU as the catalyst (Scheme
3). The initial experiments conducted, involving the application
of the piperidine-derived Ni(II) complex 9d, were impressive,
as in both examples explored, the corresponding products 17d
and 18d were obtained in high chemical yield (>95 and >99%,
respectively) and high diastereoselectivity (>98% de)15 follow-
ing the complete conversion from the starting material 9d in
20-25 min (entries 1 and 2). Although the reaction rate was
decreased from the previous example, the complete conversion
of the dibenzylamine-containing complex 9c to the product 17c
was realized for the application of the aromatic-substituted
Michael acceptor 16a, in 2.5 h (entry 3). However, after 26 h,
the reaction of the Michael acceptor bearing the iso-propyl group
16b was limited to 88% conversion to the appropriate product
18c (entry 4). The decrease in the reaction rates of the
dibenzylated Ni(II) complex 9c could be accounted for by the
as well as any unforeseen complications, homologation reactions
of the complexes 9b-d were attempted in the presence of
sodium tert-butoxide and benzyl bromide (Scheme 2). The initial
experiment conducted, with the dibutylated complex 9b as the
glycine equivalent, was found to be successful with 3.5 equiv
of the base and the electrophile because the corresponding R,R-
dibenzylated glycine-containing complex was found in greater
than 90% yield with no other observable products. It was later
discovered that the excess of the base and benzyl bromide could
be decreased to 2.5 equiv each with similar results, contrary to
the results obtained with the application of their picolinic acid-
containing predecessors (Scheme 2 (table); entry 1).12 These
reactions were then repeated using various alkyl groups on the
amino function of the Ni(II) complexes 9 to ensure some
generality. It was found that the complexes incorporating either
a piperidine 9d or dibenzylamine 9c moiety yielded similar
results as both reactions resulted in greater than 90% yield and
greater than 99% conversion to the desired products 11b or 11c
(entries 2 and 3). These experiments were complimented by
two more reactions that were conducted to demonstrate that
these results were reproducible with various other alkyl halides.
Therefore, it was found that the dibutylated Ni(II) complex 9b
would also react with allyl or cinnamyl bromide in the presence
of the sodium tert-butoxide with success similar to the earlier
example generating the corresponding products 12a or 13a in
high chemical yield of 92 or 88%, respectively (entries 4 or 5).
It was also found that the introduction of two unactivated alkyl
groups could be accomplished; however, the application of the
more reactive alkyl iodide was necessary for the reaction to
reach completion, such as the synthesis of 14a via homologation
with methyl iodide (entry 6). Although the direct alkylation of
the complex 9b with propargyl bromide was unsuccessful due
to the production of various byproducts, seemingly from the
reactivity of the alkyl halide, slightly milder phase-transfer
conditions were more appropriate for this process; however,
longer reaction times or more concentrated aqueous bases are
necessary to complete the reaction which only proceeded to 33%
conversion to the bisalkylated product 15a in 24 h while 67%
of the monoalkylated product 15b remained (entry 7).
(13) Soloshonok, V. A.; Ueki, H.; Ellis, T. K.; Yamada, T.; Ohfune, Y.
Tetrahedron Lett. 2005, 46, 1107.
(14) For the synthesis of the chiral Michael acceptors utilized in this
manuscript, see: Soloshonok, V. A.; Ueki, H.; Jiang, C.; Cai, C.; Hruby,
V. J. HelV. Chim. Acta 2002, 85, 3616.
Although these alkylation experiments were useful for
examining the stability of this new series of methylene-activated
glycine derivatives as well as for demonstrating an efficient route
(15) The diastereoselectivity described for the products of the described
reactions was assigned by 1H NMR analysis of the crude reaction mixtures
as well as by the optical rotations of the purified products on the basis of
our laboratory’s previous investigations of similar systems (see ref 16 and
17).
(12) (a) Ellis, T. K.; Martin, C. H.; Ueki, H.; Soloshonok, V. A.
Tetrahedron Lett. 2003, 44, 1063. (b) Ellis, T. K.; Martin, C. H.; Tsai, G.
M.; Ueki, H.; Soloshonok, V. A. J. Org. Chem. 2003, 68, 6208. (c) Ellis,
T. K.; Hochla, V. M.; Soloshonok, V. A. J. Org. Chem. 2003, 68, 4973.
8574 J. Org. Chem., Vol. 71, No. 22, 2006

New Modular Nucleophilic Glycine EquiValents
SCHEME 3
SCHEME 4
the iso-propyl-derived acceptor 16b achieved a dismal 60%
conversion to 18f within 26 h (entry 10).
increase in the free rotation of the benzyl groups coupled with
the enhanced size of the substituents.
After analyzing the information obtained by these relative
reactivity experiments, it was not difficult to identify the most
likely candidate for further investigation. The piperidine moiety
proved to be the most compatible group to incorporate into the
Ni(II) complex for enhanced reactivity under the conditions for
these Michael addition reactions. Their reactivity far surpasses
that of the previously published glycine equivalents including
the picolinic acid-derived Ni(II) complexes explored within our
laboratory.16 Therefore, the generality and limitations for the
application of this new glycine equivalent, with respect to this
type of chiral Michael addition reactions, were of great interest.
The first two experiments that were investigated were fairly
straightforward as both of the Michael acceptors were not as
electron-rich or bulky as the previous two that were employed
for the reactivity study. As expected, the crotonyl- and cinnamic
acid-derived Michael acceptors 16c and 16d were successfully
incorporated into the glycine equivalent to produce the corre-
sponding â-substituted pyroglutamic acid precursors 19c and
19d in high chemical yield, 86% and >99%, respectively, with
a diastereomeric purity greater than 98% in both cases (Scheme
4 (table), entries 1 and 2). The Michael acceptor with the
2-methoxyphenyl moiety incorporated into the structure 16e was
explored as an interesting example which includes two key
features that would decrease the reactivity of the acceptor due
to the electron-donating nature of the substituent as well as to
its location on the phenyl ring which would also serve to
increase the steric constraint around the active site. Although
the electronic and steric factors contributed by the Michael
acceptor 16e did slow the reaction rate, the reaction did proceed
to completion, providing the appropriate product 19e in high
chemical yield without compromising the stereochemical out-
come (entry 3). As evidenced from the outcome from the
previous reactions, the rate of these 1,4 addition reactions is
effected by both the electronic and steric bulk contributed by
the Michael acceptor; however, it was rather difficult to
With these preliminary results in mind, the remaining
reactions were conducted utilizing complexes that contained
amino groups which were expected to also demonstrate optimal
reactivity; therefore, the results of the following experiment were
initially unexpected. The incorporation of a dimethylamino
group into the Ni(II) complex framework 9a was expected to
increase the reactivity by eliminating the steric considerations
while not seeming to effect the overall electronic nature of the
complex. Although the reaction with the aromatic-containing
Michael acceptor 16a proceeded, it took 2 h for the reaction to
progress to completion providing the corresponding product 17a
in excellent chemical yield (99%, entry 5), whereas the reaction
with the sterically challenged iso-propylated acceptor 16b did
not achieve complete conversion (62%) even after 2 h (entry
6). However, the unexpected outcomes from the reactions could
be easily explained by observing the reaction as it progressed
because it was evident that the Ni(II) complex 9a was not
completely soluble at the standard concentration utilized for
these experiments, 0.1 g in 2 mL, intrinsically decreasing the
reaction rate. Although the application of the morpholyl-derived
Ni(II) complex 9e proved synthetically useful as reactions with
both of the Michael acceptors 16a,b proceeded to completion
and provided acceptable chemical yields of the products 17e
and 18e, >95 and >79%, respectively, the rates of the reactions
were drastically slower, taking 2.25 h for the para-methoxy-
phenyl-derived acceptor 16a and 18 h for the iso-propyl-
containing acceptor 16b (entries 7 and 9), when compared with
their cyclic counterpart, the piperidyl-derived complex 9d
(entries 1 and 2). There seems to be little to no rationale that
the effect could arise from steric effects; however, the introduc-
tion of the oxygen atom into the ring could introduce a number
of electronic differences. In the last example investigated within
this series, the indolyl group incorporated into the structure of
the glycine equivalent 9f is cyclic; however, it may be the overall
size of this group that could account for the dismal reactivity,
as neither reaction proceeded to completion given the time
allotted for the reaction. The reaction which included the
Michael acceptor bearing the aromatic substituent 16a obtained
83% conversion to 17f in 2.5 h (entry 9), and the reaction with
(16) (a) Soloshonok, V. A.; Cai, C.; Yamada, T.; Ueki, H.; Ohfune, H.;
Hruby, V. J. J. Am. Chem. Soc. 2005, 127, 15296. (b) Ref 10a. (c)
Soloshonok, V. A.; Cai, C.; Hruby, V. J. Tetrahedron Lett. 2000, 41, 9645.
(d) Soloshonok, V. A.; Cai, C.; Hruby, V. J. Angew. Chem., Int. Ed. 2000,
39, 2172. (e) Soloshonok, V. A.; Cai, C.; Hruby, V. J. Tetrahedron:
Asymmetry 1999, 10, 4265.
J. Org. Chem, Vol. 71, No. 22, 2006 8575

Ellis et al.
SCHEME 5
determine the effect of each factor independently. From the
following experiment, conducted with a Michael acceptor-
containing a o-trifluoromethylphenyl group, introduced to retain
the steric effects from the previous example while reversing
the electronic contribution of the acceptor, 16f, it was evident
that the steric factor seemed to play a large role in determining
the rate of the reaction. Although the reaction proceeded to
completion providing the appropriate product 19f in excellent
yield (>98%) and diastereomeric purity (>98% de), 1 h was
necessary to complete the reaction (entry 4). The following
experiment involved the application of an electron-poor Michael
acceptor which included a 2,6-difluorophenyl group incorporated
into the structure 16g. Again, the reaction proceeded to
completion providing the corresponding product 19g in a
diastereomeric form (>98% de) and very high chemical yield
(>99%); however, only 4 min was required for this experiment
to be completed (entry 5). Although the electronic donation of
the 2-methoxyphenyl-containing Michael acceptor 16e did not
prevent the progression of the earlier reported reaction, it was
found that the electronic contributions of a N-benzyl-indolyl
moiety into the skeleton of the reactive site of the Michael
acceptor proved too much as the reaction between the complex
9d and the Michael acceptor 16h progressed sluggishly only
providing 16% conversion to the corresponding product 19h in
20 h (entry 6). However, by employing the electron-withdrawing
capabilities of a tosyl group, the electron-donating character of
the indolyl moiety could be sufficiently reduced to allow the
Michael addition reaction to proceed as evidenced by the
completion (>99% conversion) of the reaction between the
piperidine-containing Ni(II) complex 9d and the corresponding
Michael acceptor 16i in 30 min (entry 7). Although the
piperidine-derived complex 9d has proven to be quite useful to
this point, it does seem that limitations do exist, as no products
could be obtained from the reaction conducted with the N-trityl-
imidazolyl-substituted Michael acceptor 16j even after 24 h.
This, however, is not so unexpected given the shear bulk of the
N-trityl-imidazolyl group coupled with the electron-releasing
nature of the substituent. Presumably, because of the massive
steric demands of the group, similar results could be obtained
with the incorporation of a tert-butyl group into the Michael
acceptor 16k as no products were observed within the 24 h
reaction time.
the stereochemical outcome. Therefore, the following reactions
will involve the study of the acetophenone-derived, piperidine-
containing complex 10d (Scheme 5). As expected from previous
experiences, the Michael addition reactions involving this new
complex 10d, crotonyl, and cinnamic acid-derived chiral
Michael acceptors 16c,d were limited only by the time necessary
to conduct the first TLC analysis (Scheme 5 (table); entries 1
and 2). The reactions both proceeded to completion, in >99%
conversion, in under 2 min providing the corresponding products
20c,d in greater than 98% yield. However, the first bit of
promising information came with the conclusion of the following
reaction, as the Michael acceptor bearing the iso-propyl group
16b reacted with the acetophenone-derived complex 10d (entry
3) at a faster rate than the benzophenone complex utilized earlier
(Scheme 3 (table); entry 4) providing complete consumption
of the starting material 10d in 15 min. The reaction of the
Michael acceptor which included the o-methoxyphenyl sub-
stituent 16e for this reaction also provided exciting results as
the reaction rate was cut from 1.75 to 0.5 h merely by the
application of the acetophenone complex 10d, whereas the
complete chemical conversion (>99%) and yield of the products
20e (94%) were not effected by the modification of the Ni(II)
complex (Scheme 5 (table); entry 4). The presence of an
electron-withdrawing p-trifluoromethylphenyl in the Michael
acceptor 16f did not break with the trend for this set of reactions,
as the reaction proceeded approximately six times faster with
the acetophenone-derived complex 10d (20 min) compared to
the more bulky benzophenone-containing complex 9d (entry
5). Revisiting the application of the N-Ts-indolyl-containing
Michael acceptor 16i with the new improved glycine equivalent
10d also proved useful as the complete conversion of the starting
material 10d to products 20i was observed in approximately
10 min, whereas in the previous case, nearly 30 min (entry 6)
was necessary to obtain similar results.
Although the superior reactivity demonstrated by the piperi-
dine-derived Ni(II) complex 9d has far surpassed expectations,
it seems that there remains some room for improvement. As
discussed earlier, the transition state of these Michael addition
reactions is extremely crowded, therefore one may assume that
relieving some of the possible steric interactions could lead to
favorable increases in reactivity. So far in this study, only the
substituents of the amino function of the complexes have been
evaluated for this purpose; however, previous investigations
within our laboratory17,18 have demonstrated that utilizing an
o-aminoacetophenone module rather than the 2-aminobenzophe-
none one in the construction of the Ni(II) complexes can lead
to increased reactivity without risking any adverse effects on
With the accomplishment of enhanced reactivity, compared
to previously published examples, added to the list of advantages
which already includes enhanced solubility, chemical stability,
and cost efficiency, it is easy to foresee the general utility and
synthetic applicability of this series of glycine equivalents for
the methodology based on their homologation. However, there
are more items that must be addressed such as the disassembly
(17) (a) Soloshonok, V. A.; Ueki, H.; Tiwari, R.; Cai, C.; Hruby, V. J.
J. Org. Chem. 2004, 69, 4984. (b) Ref 9a. (c) Cai, C.; Soloshonok, V. A.;
Hruby, V. J. J. Org. Chem. 2001, 66, 1339. (d) Soloshonok, V. A.; Cai, C.;
Hruby, V. J. J. Org. Chem. 2000, 65, 6688. (e) Soloshonok, V. A.; Cai, C.;
Hruby, V. J. Tetrahedron Lett. 2000, 41, 135.
(18) Cai, M.; Cai, C.; Mayorov, A. V.; Xiong, C.; Cabello, C. M.;
Soloshonok, V. A.; Swift, J. R.; Trivedi, D.; Hruby, V. J. J. Pept. Res.
2004, 63, 116.
8576 J. Org. Chem., Vol. 71, No. 22, 2006

New Modular Nucleophilic Glycine EquiValents
SCHEME 6
of the complexes as well as the recovery of the target amino
acids. Although the chemical stability that has been referred to
throughout this manuscript might lead one to believe that the
disassembly of the Ni(II) complexes might be difficult, it turns
out to be rather simple and straightforward assuming the
appropriate conditions are employed, as the Ni(II) complex will
disassemble under sufficiently concentrated aqueous acids.
Therefore, 3 N hydrochloric acid will suffice to disassemble
the complexes which are typically in a solution of warm
methanol for purposes of solubility. Once the complex has been
broken down into its pieces (the target amino acid, metal ions,
and the ligand), the solution is treated with ammonium
hydroxide to quench the hydrochloric salt of the organic soluble
ligand which can be recovered by extraction into methylene
chloride and reused to form a new lot of Ni(II) complexes of
glycine in nearly quantitative yield, extinguishing any concern
about the poor atom economy of this methodology. The addition
of the ammonium hydroxide solution plays an additional role
in the case of the disassembled Michael adducts as it catalyzes
the ring closure and release of the chiral auxiliary 23 which
can be reused to form a new portion of chiral Michael acceptors.
The resulting aqueous solution of the amino acids and Ni(II)
ions may be subjected to ion-exchange chromatography for
purposes of isolation of the free amino acid in 82-94% yield
(Scheme 6).
Experimental Section
Condensation of 2-Aminobenzophenone 1 and Bromoacetyl
bromide 2 Yielding N-(2-Benzoyl-phenyl)-2-bromo-acetamide
3 and 4. A solution of bromoacetyl bromide 2 (46.46 g, 230.28
mmol) in acetonitrile (2 mL/1 g of bromoacetyl bromide) was
slowly added to a slurry of 2-aminobenzophenone 1 (20.18 g,
102.32 mmol) and potassium carbonate (70.71 g, 511.61 mmol) in
240 mL of acetonitrile. The reaction was stirred at ambient
temperature (room-temperature water bath) for 1 h, and upon
completion (monitored by TLC), the acetonitrile was evaporated
under vacuum. Water (200 mL) was then added to the crude mixture
and extracted with dichloromethane (200 mL) three times. The
organic portions were combined, dried, and concentrated under
vacuum to afford the corresponding R-bromoamide product in 98%
yield and greater than 99% chemical purity.
Alkylation of Secondary Amines with N-(2-Benzyoly/acetyl-
phenyl)-2-bromo-acetamide, Yielding the Corresponding N-(2-
Benzoyl/acetyl-phenyl)-2-dialkylamino-acetamide 6a-f or 7d.
General Procedure. To a slurry of N-(2-benzoyl/acetyl-phenyl)-
2-bromo-acetamide 3 and 4 (1 equiv) and potassium carbonate (1.2
equiv) in acetonitrile (10 mL/1 g of N-(2-benzoyl-phenyl)-2-bromo-
acetamide) was added the corresponding secondary amine 5a-f
(1.1 equiv). The reaction was allowed to proceed for 2 h at 60-70
°C (monitored by TLC) before the reaction mixture was concen-
trated under vacuum. Water was added to the viscous liquid,
followed by extraction with dichloromethane. The organic portions
were combined, dried with magnesium sulfate, and concentrated
in a vacuum to afford the corresponding N-(2-benzoyl/acetyl-
phenyl)-2-dialkylamino-acetamide 6a-f or 7d in nearly quantitative
yield and high chemical purity >99%.
Synthesis of the Ni(II) Complexes of Glycine Schiff Bases with
N-(2-Benzyoly/acetyl-phenyl)-2-dialkylamino-acetamides 9a-f
and 10d. General Procedure. A solution of potassium hydroxide
(9 equiv) in methanol (7 mL/1 g of KOH) was added to a suspension
of N-(2-benzyoly/acetyl-phenyl)-2-dialkylamino-acetamides 6a-f
and 7d (1 equiv), glycine (5 equiv), and nickel nitrate hexahydrate
(2 equiv) in methanol (10 mL/1 g of 6a-f and 7d) at 60-70 °C.
Upon complete consumption of the N-(2-benzyoly/acetyl-phenyl)-
2-dialkylamino-acetamides 6a-f and 7d, monitored by TLC, the
reaction mixture was poured over a slurry of ice and 5% acetic
acid. After the complete precipitation, products 9a-f and 10d were
filtered and dried, in an low-temperature oven (50 °C) overnight.
The product was obtained in high chemical yield (99%) and high
chemical purity without further purification.
In summary, a new series of achiral glycine equivalents have
been evaluated with respect to their synthetic utility for the
production of tailor-made, sterically demanding, and function-
alized amino acids. Among the new series of achiral glycine
equivalents, the piperidine-derived, acetophenone-containing
complex 10d was found to be a superior glycine derivative for
the Michael additions with various (R)- or (S)-N-(E-enoyl)-4-
phenyl-1,3-oxazolidin-2-ones 16a-l representing a general and
practical synthesis of sterically constrained â-substituted pyro-
glutamic acids. In particular, the application of these complexes
allowed for the preparation of several â-substituted pyroglutamic
acids which include electron-releasing and sterically demanding
substituents in the structure thus increasing the synthetic
efficiency and expanding the generality of these Michael
addition reactions.
J. Org. Chem, Vol. 71, No. 22, 2006 8577

Ellis et al.
Dialkylation of Ni(II) Complexes 9b-d with Activated Alkyl
Bromides Yielding Complexes 11a-c and 12a-14a. General
Procedure. To a solution of sodium tert-butoxide (2.5 equiv) in
DMF (10 mL/1 g of complex 9b-d) were added complex 9b-d
(1 equiv) and the corresponding alkylating reagent (2.5 equiv). The
reaction was stirred at ambient temperature (room-temperature water
bath) for 15 min. Upon completion (monitored by TLC), the
reaction mixture was poured into ice water, and the resulting
precipitate was filtered and washed with water to afford the products
11a-c and 12a-14a in yields ranging from 92 to 96% and in
greater than 99% purity.
Phase-Transfer Homologation of a Ni(II) Complex of the
Glycine Schiff Base with N-(2-Benzoyl-phenyl)-2-dibutylamino-
acetamide 9b with Propargyl Bromide. General Procedure. To
a solution of 9b in CH2Cl2 (1 mL/g) at room temperature were
added tetrapropylammonium bromide (0.25 equiv), 30% sodium
hydroxide solution (1 mL/mL of CH2Cl2), and propargyl bromide
(3.5 equiv). The resultant mixture was rigorously stirred overnight
at room temperature. To the resultant slurry mixture were added
additional water and CH2Cl2, and the water was extracted several
times with CH2Cl2. The organic layer was dryed with MgSO4,
filtered, and then evaporated in a vacuum to yield a crystalline
compound. This compound was washed first with water and then
with hexane and then dried completely to yield the final products.
Michael Addition of the Oxazolidinone-Derived Amides of
Unsaturated Acids 16a,b and Nucleophilic Glycine Equivalents
9a,c-f. General Procedure. To a flask containing 9a,c-f (0.10
g) were added 3-((E)-3-phenylacryloyl)oxazolidin-2-one (1.05
equiv) and 1.5 mL of DMF and DBU (15 mol %). The reaction
mixture was stirred at room temperature and monitored by TLC.
After the disappearance of the starting glycine equivalent by TLC,
the reaction mixture was poured into a beaker containing 100 mL
of ice water. After the ice had melted, the corresponding product
17a,c-f or 18a,c-f was filtered from the aqueous solution and
dried in an oven to afford the appropriate product in high chemical
yields.
Michael Addition of the Oxazolidinone-Derived Amides of
Unsaturated Acids 16c-k and Nucleophilic Glycine Equivalent
9d. General Procedure. To a flask containing 9d (0.10 g) were
added 3-((E)-3-phenylacryloyl)oxazolidin-2-one 16c-k (1.05 equiv)
and 1.5 mL of DMF and DBU (15 mol %). The reaction mixture
was stirred at room temperature and monitored by TLC. After the
disappearance of the starting glycine equivalent by TLC, the reaction
mixture was poured into a beaker containing 100 mL of ice water.
After the ice had melted, the corresponding product 19c-k was
filtered from the aqueous solution and dried in an oven to afford
the appropriate product in high chemical yields.
Michael Addition of the Oxazolidinone-Derived Amides of
Unsaturated Acids 16b-f,i,l-q and the Nucleophilic Glycine
Equivalent 10d. General Procedure. To a flask containing 10d
(0.10 g) were added 3-((E)-3-phenylacryloyl)oxazolidin-2-one 16b-
f,i,l-q (1.05 equiv) and 1.5 mL of DMF and DBU (15 mol %).
The reaction mixture was stirred at room temperature and monitored
by TLC. After the disappearance of the starting glycine equivalent
by TLC, the reaction mixture was poured into a beaker containing
100 mL of ice water. After the ice had melted, the corresponding
product 20b-f,i,l-q was filtered from the aqueous solution and
dried in an oven to afford the appropriate product in high chemical
yields.
Decomposition of the Ni(II) Complex of the R,R-Diallylglycine
Schiff Base with N-(2-Benzoyl-phenyl)-2-dibutylamino-aceta-
mide 12a Yielding R,R-Diallylglycine 21 and N-(2-Benzoyl-
phenyl)-2-dibutylamino-acetamide 6b. To a solution of 22 mL
(14.5 mL/1 g of complex 12a) of MeOH and 11 mL (7.25 mL/1 g
of complex 12a) of 3 N HCl at 70 °C was added 1.471 g (2.527
mmol) of complex 12a. The solution was stirred for 30 min and
then evaporated. The acid 21 and Ni(II) were extracted in 50 mL
of DI water from ligand 6b and CH2Cl2 and then evaporated.
Following the evaporation, the crystalline compounds were dis-
solved in the minimum amount of DI water and placed on an ion-
exchange column using Dowex 50 × 2-100 resin. The column
was first washed with DI water until neutral, followed by 8%
aqueous ammonium hydroxide (500 mL) to elute acid 21. This
solution was evaporated to afford 0.3843 g (2.476 mmol, 98% yield)
of acid 21. The Ni(II) was eluted with concentrated HCl after the
column was returned to neutral with DI water. Ligand 6b was
recovered by the evaporation of the organic layer from the
aforementioned separation.
Decomposition of Complexes 20c-e,l-m; Isolation of (2S,3S)-
3-Alkyl- and (2S,3R)-3-Arylpyroglutamic Acids 22c-e,l-m;
Recovery of Ligand 10d; and Starting Chiral Auxiliary (S)-23.
General Procedure. A solution of pure complex 22c-e,l-m (25
mmol) in MeOH (50 mL) was slowly added to a stirring solution
of aqueous 3 N HCl in MeOH (90 mL, ratio 1:1, acid/MeOH) at
70 °C. Upon the disappearance of the red color, the reaction mixture
was evaporated in a vacuum until dryness. Water (85 mL) was
added, and the resultant mixture was treated with an excess of
concentrated ammonium hydroxide and extracted with methylene
chloride. The methylene chloride extracts were dried over magne-
sium sulfate and evaporated in a vacuum to afford 10.59 g of a 1:1
mixture (99%) of ligand 10d and chiral auxiliary (S)-23. The
aqueous solution was evaporated in a vacuum, dissolved in a
minimum amount of water, and loaded on a cation-exchange resin
Dowex 50 × 2 100 column. The column was washed with water,
and the acidic fraction was collected to give the pyroglutamic acid
22c-e,l-m. An analytically pure sample of the product was
obtained by crystallization of the compound from THF/n-hexane.
Acknowledgment. This work was supported by the Depart-
ment of Chemistry and Biochemistry, University of Oklahoma,
as well as by the U.S. Department of Education (GAANN
Fellowship for T.K.E.). The authors gratefully acknowledge
generous financial support from Central Glass Company (Tokyo,
Japan) and Ajinomoto Company (Tokyo, Japan).
Supporting Information Available: Full characterization and
1H NMR spectra of all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO0616198
8578 J. Org. Chem., Vol. 71, No. 22, 2006