Synthesis of optically pure 4-fluoro-glutamines as potential metabolic imaging agents for tumors

Article abstract of DOI:10.1021/ja109203d

A versatile synthetic route to prepare all four stereoisomeric 4-fluoro-glutamines was developed by exploiting a Passerini three-component reaction. The skeleton of 4-substituted glutamine derivatives was efficiently constructed. Subsequent four-step reactions, highlighted by a "neutralized" TASF fluorination, provided the desired products with high yields and excellent optical purity. The optically pure fluorine-18 labeled 4-fluoroglutamines were also successfully prepared using either a 18-crown-6/KHCO₃ or K[222]/K₂CO₃ catalysis system. Preliminary cell uptake and inhibition studies using the 9L tumor cells ana SF188_BC1-XL tumor cells (a glutamine addicted tumor derived from glioblastoma) provided strong evidence for their potential application in conjunction with positron emission tomography (PET) for in vivo imaging of tumors, which use glutamine as an alternative energy source.

Full text of DOI:10.1021/ja109203d

ARTICLE
pubs.acs.org/JACS
Synthesis of Optically Pure 4-Fluoro-Glutamines as Potential
Metabolic Imaging Agents for Tumors
Wenchao Qu,*,† Zhihao Zha,† Karl Ploessl,† Brian P. Lieberman,† Lin Zhu,† David R. Wise,§,
Craig B. Thompson,§, and Hank F. Kung*,†,‡
Departments of †Radiology, ‡Pharmacology, and §Cancer Biology, and Abramson Cancer Centre, School of Medicine,
University of Pennsylvania, Philadelphia, Pennsylvania 19104, United States
S Supporting Information
b
ABSTRACT: A versatile synthetic route to prepare all four
stereoisomeric 4-fluoro-glutamines was developed by exploit-
ing a Passerini three-component reaction. The skeleton of
4-substituted glutamine derivatives was efficiently construc-
ted. Subsequent four-step reactions, highlighted by a “neu-
tralized” TASF fluorination, provided the desired products
with high yields and excellent optical purity. The optically
pure fluorine-18 labeled 4-fluoroglutamines were also suc-
cessfully prepared using either a 18-crown-6/KHCO3 or
K[222]/K2CO3 catalysis system. Preliminary cell uptake and inhibition studies using the 9L tumor cells and SF188Bcl-xL tumor
cells (a glutamine addicted tumor derived from glioblastoma) provided strong evidence for their potential application in conjunction
with positron emission tomography (PET) for in vivo imaging of tumors, which use glutamine as an alternative energy source.
’ INTRODUCTION
1 and 2 are L-glutamine (natural amino acid) analogues, whereas 3
and 4 are D-glutamine (unnatural amino acid) analogues. They may
serve as potential diagnostic tracers for positron emission tomo-
graphy (PET) in order to assess glutamine utilization in various
types of tumor cells (Figure 1).
Amino acids (AAs) are essential chemicals for biological
system. In addition to serving as the basic building blocks for
peptides and proteins, they also perform various critical physiolo-
gical functions as metabolic intermediates and neurotransmitters.
Fluorine substituted AAs not only play important roles in peptide-
based drugs and protein engineering,1-8 but have also been
particularly useful in positron emission tomography (PET) imaging
for the diagnosis of cancer.9-14
It is now generally accepted that genetic alterations, such as the
up-regulation of the PI3K/Akt/mTor pathway (independent of
hypoxia), can stimulate aerobic glycolysis and support tumor proli-
feration (Warburg effect). Recently, several reports have suggested
that glutamine (Gln) can serve as an alternative source of metabolic
energy for tumor cells.15-18 Overexpression of the oncogene c-Myc
(Myc) is essential for the coordinated expression of genes necessary
for cells to use glutamine catabolism that exceed the cellular
requirement for protein and nucleotide biosynthesis.19,20 “Glutami-
nolysis,” especially in myc-overexpressing cells, plays a significant
role in tumor growth and metabolism. This may represent an attrac-
tive new strategy for developing therapeutic and diagnostic agents
for cancer therapy.18 Therefore, Gln and its analogues may poten-
tially be useful as metabolic indicators for studying the increased
utilization of Gln in tumors. There is a significant interest in synthe-
sizing all four stereospecific18Flabeled4-fluoro-glutamine (4-FGln)
isomers: [18F]4-FGln, 1 (2S,4R); [18F]4-FGln, 2 (2S,4S); [18F]4-
FGln, 3(2R,4S), and[18F]4-FGln, 4(2R,4R) (Figure1) (Cautions:
18F, T1/2 = 110 min, 511 KeV, radioactive material). Among them,
To confirm the successful preparation and identity of these four
18F labeled stereoisomers of 4-FGln (see Figure 1), a practical
synthesis of the nonradioactive molecules is needed. Many reports
on synthesizing various stereospecific fluorinated R-amino acids
have appeared recently in the literature.21-37 A review of synthetic
methods for preparing 4-fluorinated glutamic acid and glutamine
analogues was published by Dave et al. in 2003.38 Preparation of
diastereomeric mixtures of 4-FGln39 was reported and 18F labeled
4-FGln was also mentioned.40 However, to the best of our knowl-
edge, there is no specific report on the synthesis of optically pure
4-FGln and the 18F labeled counterparts. The lack of literature
reports on this topic is not surprising since there are formidable
challenges facing the synthesis of such densely functionalized
glutamine analogues. The presence of electron-withdrawing fluorine
and carboxylic groups often induces undesirable epimerization at the
C2 and C4 positions leading to the formation of four possible
stereoisomers at different stages of preparation. Previously reported
methods usually introduced a fluorine atom in an early stage of the
whole synthesis, but we prefer a synthetic pathway that adds the
fluorine atom at the very end (or as late as feasible) of the synthesis.
Unlike a “normal” strategy for synthesizing fluorine substituted AAs,
Received: October 28, 2010
Published: December 29, 2010
r
2010 American Chemical Society
1122
dx.doi.org/10.1021/ja109203d J. Am. Chem. Soc. 2011, 133, 1122–1133
|

Journal of the American Chemical Society
ARTICLE
Figure 1. Structures of all four 4-fluoro-glutamine isomers: [18F]4-FGln, 1 (2S,4R); [18F]4-FGln, 2 (2S,4S); [18F]4-FGln, 3 (2R,4S), and [18F] 4-FGln
4 (2R,4R).
thiourea for selective removal of the chloroacetyl group resulted in a
relatively low yield of the desired alcohol 15.55 Fortunately, adding a
stoichiometric amount of NaHCO3 in this reaction dramatically
improved the reaction yield to 91%.56 The tosylation reaction of 15
gave tosylate 16 (the precursor for fluorination) in 96% yield
(Scheme 2).
Fluorination Reaction. To explore the fluorination method
leading to the desired fluorinated intermediate 19, several precursors,
15-18, were prepared and tested. The results of the fluorination
reactions using different precursors and reagents are summarized in
Table 1. Neither the reactions of alcohol 15 with (diethylamino)-
sulfur trifluoride (DAST)57 (entries 1 and 2) nor the reactions
between tosylate 16 and potassium fluoride (KF) or tetrabutylam-
monium fluoride (TBAF) (entries 3-5) provided any desired pro-
duct 19.58 Tetrabutylammonium difluorotriphenylsilicate (TBAT)59
and tetrabutylammonium difluorotriphenyl stannate60 (entries6-8)
gave only low yields despite a large excess of reagents used. As these
approaches did not give satisfactory results, we investigated the use of
the more active LG to replace the tosylate group. The preparation of
triflate 17 by triflation of 15 was unsuccessful. The failure of this
reaction may be due to the instability of this triflate. The preparation
and separation of nosylate 18 was achieved later,61 although a slow
decomposition of 18 at room temperature was observed. Never-
theless, the reaction of TBAT with 18 did not afford any desired
fluoride 19.
Figure 2. Retrosynthetic analysis for the preparation of 4-fluoro-L-
glutamine via a Passerini three-component reaction.
the preparation of 18F labeled glutamine has to be simple and fast
with a high yield incorporation of 18F very close to the final step, due
to the short physical half-life of 18F (T1/2 = 110 min). Such stringent
requirements have placed additional obstacles beyond the problem
of introducing a fluorine atom into a multifunctionalized glutamine
backbone.
In this paper, we report the following: (1) stereospecific synthesis
of 4-FGlns 1 (2S,4R), 2 (2S,4S), 3 (2R,4S), and 4 (2R,4R) in a con-
vergent manner; (2) analysis and confirmation of the structures of all
4-FGln stereoisomers; (3) development of practical methods for
the preparation of all four isomers, [18F]4-FGln, 1 (2S,4R); [18F]4-
FGln, 2 (2S,4S); [18F]4-FGln, 3 (2R,4S), and [18F]4-FGln, 4
(2R,4R); (4) investigation of their ability to penetrate tumor cells
as indicators of glutamine metabolism.
’ RESULTS AND DISCUSSION
Alternatively, a direct fluorination method (a combination of
perfluoro-1-butanesulfonyl fluoride(PBSF)-NR3(HF)3-NR3 (R=Et
or diisopropylethyl) with alcohol (15) reported by Yin was tested
(entry 9).62 Combinations of different bases (Et3N or DIPEA),
solvents (THF or MeCN), and substrate addition methods (one shot
addition or slow addition via syringe pump) were tested but did not
show any improvement (yield e20%). Instead of the desired pro-
duct, a large amount of intramolecular cyclized lactone 20 and the
dimer 21 were produced in these reactions (Figure 3). A plausible
explanation for this phenomenon is proposed in the Scheme 3.
Interestingly, a similar phenomenon was also observed in the
synthesis of (2S,4S)- and (2S,4R)-5-fluoroleucine.31 Similarly, the
bimolecular cyclization could be the reason for the formation of dimer
21. Finally, the use of another hypervalent silicon fluorination agent,
tris(dimethylamino) sulfonium difluorotrimethylsilicate (TASF),
gave significantly improved yields (entry 10).63 The yield was
further improved (to 60-70%) (entry 11) by adding TASF in
two portions at a 12-h interval.
To our pleasant surprise, two diastereomeric 4-OH isomers of 15,
22 (2S,4S) and23 (2S,4R) (1:1 ratio based on HPLC analysis) could
be separated by flash chromatography. The tosylation reaction
provided 24 (2S,4S) and 25 (2S,4R) in excellent yields with no
observed epimerization. Thus, the sequence of reactions provided
ready access to the optically pure precursors for replacing 4-OTs with
4-F (Scheme 4).
Optimization of the Fluorination Reaction and Synthesis
of the Final Product. After the exploration of the reactivity of
TASF for the fluorination reaction, we turned our attention to
Retrosynthetic Analysis and Intermediate Synthesis. We
hypothesized that the carbon-4 position functionalized glutamine
skeleton could be assembled by aldehyde 5, isocyanide 6, and acid
7 via a Passerini three-component reaction (Figure 2)41,42 This step
also enabled the introduction of the amide protected with a 2,4,6-
trimethoxybenzyl (Tmob) protecting group (PG). The Tmob group
proved to be compatible in a “global” acidic deprotection condition at
the end of the synthesis.43,44 Initially, the other isocyanide, 2,4-dime-
thoxybenzyl (dmob) isocyanide, was introduced to form the dmob
protecting group for the amide.45 In the final deprotection stage, it was
found that the removal of the dmob group by TFA was very sluggish.
Following our design, we first conducted the synthesis of two
intermediates 5 and 6 (Scheme 1). Starting from a partially protected
L-aspartic acid derivative 8, esterification using tert-butyl tricloro-
acetimidate46 under the catalysis of BF3 afforded the fully protected
intermediate 9 with a 97% yield.47,48 Removal of the benzyl group by
catalytic hydrogenation followed by a mixed anhydride preparation
and a subsequent sodium borohydride reduction gave homoserine 11
in 87% overall yield.49 Subsequent oxidation of 11 with Dess-Martin
periodinane (DMP)50 provided aldehyde 5 in 86% yield.51,52 The
isocyanide 6 was prepared from 2,4,6-trimethoxybenzylamine 12.43,44
After conversion to formamide 13 by simply reacting with ethyl
formate,53 treatment with triphosgene gave 6 in 68% yield over two
steps.54
Next, the Passerini three-component reaction was carried out.
To our delight, the 4-acyloxy substituted Gln derivative 14 was
successfully formed under mild reaction conditions in an efficient,
atom-economical manner (95% yield). Simply treating 14 with
1123
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
Scheme 1. Synthesis of Intermediates 5 and 6
Scheme 2. Synthesis of 4-Hydroxy and 4-Tosyl Substituted Gln
Table 1. Exploration Results of Fluorination Reaction
entry
conditions
substrate
yield (%)a
1
2
DAST, CH2Cl2, -78 °C to rt, 5 h
DAST, pyridine, toluene, rt 30 min
to 80 °C 45 min
KF, TBABb, DMF, 80 °C, 6 h
TBAF, THF, rt, 20 min
TBAF, MeI, THF,
15
15
N/A
N/A
3
4
5
16
16
16
N/A
N/A
N/A
synthesizing two optically pure 4-FGlns, 1 (2S,4R) and 2 (2S,4S).
Subsequent fluorination of 24 provided a 78% yield with a full
inversion of stereochemistry at the C4-position and an epimer-
ization at the C2-position (4-FGln derivative 26 (2S,4R) and 27
(2R,4R), 1:1 ratio) (Scheme 4). The TASF fluorination condi-
tions selected above, which may be good enough for maintaining
the SN2 fluorination reaction and fully inverse the configuration
at C4 position, did not maintain the stereospecificity at the C2
position. The basicity of the reagent TASF may be responsible
for this observed epimerization at the C2 position.
-78 °C to rt, 12 h
10-15 equiv TBAT, MeCN,
60 °C, 12-24 h
6
7
16
16
26-40
3 eq [Bu4N][Ph3SnF2],
19
MeCN, 60 °C, 24 h
10 equiv TBAT, MeCN, 60 °C, 3 h
PBSF, rt, Et3N/Et3N(HF)3
8
9
18
16
N/A
14-20
c
or DIPEA/DIPEA(HF)3
We reasoned that moderating the basicity of the fluorination
reagents might alleviate the above-mentioned epimerization. As
reported previously, we employed a combination of TBAF and
2-mesitylenesulfonic acid to reduce epimerization at the C2
position in the synthesis of stereospecific 5-fluoroleucine.31
Although directly following this method did not provide us with
successful fluorination reactions, we returned to this idea with
our agent, TASF. To reduce the basicity of TASF further, a
10
11
5 equiv TASF, CH2Cl2, 0 °C
to rt, 12 h
16
16
40
2 equiv TASF, CH2Cl2, 0 °C to rt,
12 h, 2 more equiv TASF, 12 h at rt
60-70
a Isolated yields. b TBAB: tetrabutylammonium bromide. c MeCN or toluene
as solvent, substrate was added with one portion or slow addition.
method, fine crystals of 26 were obtained and the structure
was determined by X-ray crystallography (Figure 4), which
assigned the absolute configuration of 26. Finally, a global
deprotection was accomplished by reacting the fully protected
4-FGln 26 with TFA/dimethyl sulfide (DMS) for 2.5 h.64
Further purification using a Dowex 50WX8-200 (Hþ form) resin
column, followed by recrystallization from EtOH/H2O, pro-
vided the final product, 4-FGln (2S,4R), 1, in 71% yield.
milder acidic reagent Et3N (HF)3 was chosen. TASF was care-
3
fully titrated with Et3N (HF)3 to adjust the pH value of the
3
reaction to 7-8. This “neutralized” TASF was tested for
fluorination of the tosylate 24 (Scheme 5). After adding a more
polar solvent (THF/CH2Cl2 mixture) and increasing the reac-
tion temperature, the conversion of the tosylate to fluoride 26
was achieved in a respectable yield (77%) with no epimerization
at the C-2 position (Scheme 4). Using a slow evaporation
1124
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
reactants contributed to the difference in reactivity. Interestingly,
we observed an entirely reversed reactivity for the fluorination
reaction. Tosylate 24, a stereoisomer with a threo-configuration,
showed better reactivity compared with erythro-tosylate 25. In
the case shown in Scheme 6, the relative bulky arylamine is the
nucleophile whereas the small-sized bromide played the role of
leaving group. In our example (Scheme 5), however, the size of
the two groups is reversed. The nucleophile is a small fluoride
and the bulky para-toluenesulfonyloxy group acts as the leaving
group. Such a difference could account for the observed reactivity
of the fluorination reaction in our study.
Figure 3. Structures of two byproduct in entry 9.
Scheme 3. Plausible Mechanism for the Formation of
Cyclized Byproducts 20 and 21
After successful synthesis of the desired 4-FGln products 1 and
2, the other two 4-FGln stereoisomers, 3 and 4, were also
synthesized following the same synthetic strategy starting with
partially protected unnatural D-aspartic acid 32 (Figure 5). The
reaction yields and spectra of all of the D-isomer intermediates
(33-36, 27, and 29) and the final products 3 (2R,4S) and 4
(2R,4R) were comparable with the analogue L-isomers with
reversed optical rotation values. The X-ray crystal structures of
fluoride 29 and 4-FGln 3 further confirmed the configuration
assignments of these molecules (see Supporting Information).
Development of Radiofluorination Methods. To prepare
the desired [18F]4-FGln at a “no-carrier-added” level, we tested a
SN2 nucleophilic substitution reaction of “naked” [18F]fluoride
with tosylate 16, in the presence of Kryptofix 222 (K[222]) and
K2CO3 as the catalyst in CH3CN at 80 °C for 10 min (the most
commonly used 18F fluorination condition). It was found that,
instead of [18F]26 and [18F]28, the two expected fluoride
intermediates, all four possible isomers [18F]26-29 were
formed in the reaction mixture (Scheme 7). While there was
no starting material 16 left, a significant amount of elimination
byproduct 37 was observed (determined by LCMS). The high
basicity of both K[222] and K2CO3 may be responsible for the
epimerization at both of the C2 positions.
To determine the purity of the final product, a specific chiral
HPLC method, which combines 1.0 mM CuSO4 aqueous solution
as the mobile phase and reverse chiral column as the stationary phase
with a control of column temperature (adjusted at 10 °C), was
selected. This method clearly differentiated the four 4-FGln stereo-
isomers and the HPLC profiles demonstrated that the purity of
4-FGln 1 (2S,4R) was 98.8%. A slow evaporating recrystallization
method provided excellent crystals of 1 and the X-ray crystal-
lographic analysis (Figure 4) data further supported the structure
assignment. The optically purified 4-FGln 1 (2S,4R) has never been
presented before. The results reported in here firmly establish the
configurations of each of the 4-FGln isomers, which will facilitate
future applications of this series of biologically interesting and
medically useful compounds.
We also accomplished the synthesis and assignment of 4-FGln
2 (2S,4S) (Scheme 5). Following the exact reaction conditions
for synthesizing the fluoride intermediate 26, the desired fluori-
nated glutamine derivative 28 was only obtained in 18% yield. By
adjusting the pH value of TASF to 8-9, instead of 7-8
mentioned above, the fluorination yield reached 50%. However,
under this higher pH condition, there was an increased epimer-
ization byproduct 29 (ratio of 28/29: 80/20), which made this
reaction less attractive. After increasing the amount of “neutra-
lized” TASF from 5 to 8 equiv, slightly decreasing the reaction
temperature, as well as prolonging the reaction time from 10 to
24 h, the yield of this reaction was improved to 30%. Following
the same deprotection and purification methods, the final
product 2 was obtained in 67% yield. Chiral HPLC profiles
showed that the purity of this product was 95%. Further attempts
to recrystallize the product in order to remove a small amount of
unwanted byproduct were not successful.
To avoid this epimerization during the 18F labeling reaction, a
neutral phase transfer catalyst, tetrabutylammonium bicarbonate
(TBABC), was tested as the radiolabeling catalyst. Surprisingly,
similar to that observed for its fluorine-19 counterpart (Table 1,
entries 3-5), the TBABC catalyzed reaction only afforded small
amounts of labeled product using either CH3CN or tert-butanol
as the solvent.67 Alternatively, a combination of a mild basic
agent potassium bicarbonate (KHCO3) and a neutral phase
transfer catalyst, 18-Crown-6, was tested for this “hot” labeling
reaction. Using the tosylate precursor 24, we found that the
desired [18F]26, a (2S,4R) isomer, was successfully formed as the
major product. Subsequently, the desired final product, [18F]1,
was finally prepared by treating intermediate [18F]26 with TFA/
anisole at 60 °C for 5 min (Scheme 7). A direct identification of
[18F]1 can be performed successfully by coeluting it with 4-FGln
1 (2S,4R), a “cold” standard under the established chiral HPLC
method. We were delighted to find that this method is also
capable of separating and identifying all four radiolabeled and
“cold” 4-FGln isomers 1, 2, 3, and 4 (Figure 6). Upon the basis of
this chiral HPLC analysis, we found that the KHCO3/18-Crown-
6 catalyzed radiolabeling method could prepare [18F]1 with high
radiochemical and stereoisomeric purity (>95%).
The different reactivity between tosylate 24 and 25 is not
surprising. Krasnov and his colleagues investigated the nucleo-
philic substitution of two stereoisomers of 4-bromo substituted
glutamate derivatives (Scheme 6).65,66 First, they found that this
type of reaction follows a “rigorous” SN2 mechanism, that is, a full
inversion at the C4 position. Second, the bromide 31 (erythro)
showed higher reactivity compared to the other stereoisomer, 30
(threo). The ratio of reaction rates (k2 to k1), S, ranges from 3 to 5
despite of the change of different solvents and different nucleo-
philes. They proposed a transition state model and concluded
that the steric effect and the nonbonding interactions between
the nucleophiles as well as the bulky functional groups of the
Radiolabeling with [18F]fluoride starting from the four stereo-
specific precursors (24, 25, 35, and 36) proved to be very tricky. As
demonstrated above, the activation of [18F] fluoride by K[222] and
K2CO3, a commonly used method to produce “naked” fluoride,
readily led to epimerization. The basicity of K[222] and K2CO3
promoted the epimerization of the C2 position (Scheme 7).
1125
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
Scheme 4. Synthesis of Fluorination Precursors 24 and 25 as Well as Exploring Synthesis of Optically Pure 4-FGln 1
Scheme 5. Synthesis of Optically Pure 4-FGlns 1 and 2
To circumvent the problem, a less basic catalyst system, 18-crown-6
and KHCO3, was employed. The improved reaction conditions led
to the successful conversion of two tosylate precursors 24 (2S,4S)
and 35 (2R,4R) to the corresponding final products of [18F]4-FGln,
[18F]1 (2S,4R), and [18F]3 (2R,4S), which contain only an inverted
C4 position from that of the starting configuration. The starting
precursors, 24 (2S,4S) and 35 (2R,4R) both have a similar threo-
configuration for the 4-OTs and 2-NHBoc groups.
24 (2S,4S) and 35 (2R,4R), under K[222]/K2CO3 conditions, the
results were more predictable. For precursor 24 (2S,4S), the labeling
reaction led to a “predictable” and “repeatable” epimerization (1:1
ratio) at the C2 position and produced two final 18F labeled 4-FGlns,
[18F]1 (2S,4R) and [18F]4 (2R,4R). Similarly, when precursor 35
(2R,4R) was treated with K[222]/K2CO3 and [18F] fluoride, it led
to the formation of [18F]2 (2S,4S) and [18F]3 (2R,4S) (1:1 ratio).
To obtain each of the isomers selectively, the labeling intermediates
containing all of the protecting groups were purified by HPLC and it
was found that all of the isomers could be readily separated
(Scheme 8 and Figure 7). Thus, all four isomers can be “predictably”
prepared for further biological study.
One possible explanation for the above radio-fluorination
results is presented in Scheme 9: tosylate precursors 25 and 35 can
easily convert to each other under both reaction conditions (due to a
base-induced epimerization at the C2 position). Because of the better
reactivity of the threo-stereoisomer, 35, compared with the erythro-
isomer, 25, thefluorinated product, [18F]29, can be formed in a faster
manner relative to [18F]28 under both of the radiolabeling condi-
tions. However, the further epimerization of the fluorinated product
from [18F]29 to [18F]28 could not be promoted fast enough in a
15 min reaction period at 70 °C under the 18-crown-6/KHCO3
conditions (a relative weak base). On the contrary, the K[222]/
K2CO3 (a stronger base) combined radiolabeling process could
When the threo-configuration was switched to the erythro-config-
uration, such as for 25 (2S,4R) and 36 (2R,4S), the reactivity for 18F
labeling changed significantly and unexpectedly (Scheme 8). Using
18-crown-6/KHCO3 as the catalysts for 18F fluorination of 25
(2S,4R) and 36 (2R,4S) led to the formation of [18F]4FGln, 3
(2R,4S) and [18F]4FGln, 1 (2S,4R) as the major products, respec-
tively. It is important to note that the 18F fluorination reaction in the
presence of 18-crown-6 and KHCO3 apparently inverted the C2 and
C4 (SN2) positions simultaneously (double inversion). These
findings are highly unexpected. Clearly, only the precursors with
erythro-configuration yielded a “double-inversion” radiolabeling, that
is, 25 (2S,4R) led to the formation of [18F]4FGln, 3 (2R,4S), while
36 (2R,4S) resulted in [18F]4FGln, 1 (2S,4R) (Scheme 8).
To obtain the two other stereoisomers of [18F]4-FGln, [18F]2
and [18F]4, an alternative process had to be pursued. When we
conducted 18F fluorination of both threo-configuration precursors,
1126
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
Figure 5. Structures of starting material D-aspartic acid derivative 32
and two sets of intermediates 33 and 34, 35 and 36.
especially during the preparation of [18F]4-FGln. To avoid this side
reaction, a careful control of the reaction temperature and basicity is
essential. The methods described in this paper for the synthesis
of stereoisomers of “hot” and “cold” 4-FGln put forward a rela-
tively simple strategy to obtain these otherwise hard to produce
isomers. Although further improvement of the radiolabeling yield
may be necessary, it is apparent that the threo-precursor, 24, can be
directly radiolabeled under 18-crown-6/KHCO3 conditions to give
the desired L-glutamine isomer [18F]4-FGln, 1 (2S,4R) as a major
product.
Biological Evaluation of [18F]4-FGln ([18F]1, [18F]2, [18F]3,
and [18F]4). The cell uptake of the four [18F]4-FGln isomers was
evaluated using the 9L and SF-188 tumor cell lines. The cell uptake
results (% dose/100 μg protein) were compared to those of [3H]-L-
glutamine. The 9L tumor cells are commonly used for studying
tumor metabolism and amino acid uptake.10,68,69 The transformed
SF188-Bcl-xL tumor cells with up-regulation of c-myc gene were
reported to have a significant “addiction” to the L-glutamine;15,16
therefore, they were also used to test the hypothesis that the new
[18F]4-FGln isomers are suitable glutamine mimetics for imaging
the “glutaminolysis” in tumor cells.
Figure 4. X-ray crystallography structures of 26 (a) and 1 (b), ORTEP
drawing of the title compounds with 30% probability thermal ellipsoids.
Scheme 6. Nucleophilic Substitution of Bromide with Aryl-
amine As Nucleophile: A Kinetics Study65,66
There were distinctive differences among all the isomers
(Figure 8); while [18F]FDG displayed the expected high uptake
and retention in the 9L cells. The two isomers with 2S config-
uration, the natural L-amino acid of Gln derivatives, displayed the
highest uptake. The [18F]4-FGln, 1 (2S,4R), showed a high uptake
in 9L cells and the uptake continued nonstop for 2 h, the maximum
time point studied. The [18F]4-FGln, 2 (2S,4S), also displayed
excellent uptake similar to that of 1 (2S,4R). However, it appeared
that the uptake value dropped precipitously at 2 h (Figure 8). The
mechanisms responsible for the efflux at this later time point are
unclear at this stage. The unnatural D-Gln (containing the D-amino
acid), [18F]4-FGln isomers, 3 (2R,4S), and 4 (2R,4R), all showed a
drastically lower cell uptake as compared to the [18F]4-FGln, 1
(2S,4R), and 2 (2S,4S), as well as the [3H]-L-Gln, suggesting that the
configuration of the amino group at the C-2 position is critical for
tumor cell uptake, and that the 2S configuration of Gln (commonly
known as the L-isomer) is essential for transportation across the cell
membrane.
convert [18F]29 to [18F]28 fast enough to form a roughly 1:1 mix-
ture of [18F]29 and [18F]28 (Figure 7) after a 10 min reaction period
at 80 °C.
There are several challenging issues associated with the prepara-
tion of 18F labeled 4-fluoroglutamine. First, there are two optical
centers (the C2 and the C4 positions) in this molecule that can,
theoretically, lead to the possible formation of four stereoisomers (1,
2, 3, and4) during the synthetic process. Second, the R-amino group
is adjacent to the carboxylic acid, and therefore, as in other naturally
occurring R-amino acids, it is easily epimerized under basic condi-
tions. Next, the glutamine contains a labile amide group, which can
easily hydrolyze to form glutamic acid. The glutamine and also
glutamic acid are readily converted, or cyclized, to form pyroglutamic
acid under either acidic or basic conditions. As such, 4-fluoropyr-
oglutamic acid is a common impurity in the preparation of 4-FGln.
In addition, when the 4-FGln derivatives are synthesized by the SN2
fluorination reaction, it often produces the elimination byproduct,
The tumor cell uptake of [18F]4-FGln is a highly selective
process; the two 18F labeled native L-Glns (2S) displayed a higher
uptake than that of the 18F labeled unnatural D-Glns (2R). It is likely
that the [18F]4-FGln, 1 (2S,4R), and [18F]4-FGln, 2 (2S,4S), enter
tumor cells through an active transport system. Once inside the cells,
just as in the case of naturally occurred L-Gln (2S), the glutaminase
may strip off the amino group at the C5 position. Subsequently the
C2 amino group would also be removed to give the key intermediate,
2-ketoglutaric acid. It may also be possible that the Gln could be
effluxed and be transported out of the cells. We noticed that in
comparison with [18F]4-FGln, 1 (2S,4R), [18F]4-FGln, 2 (2S,4S) or
1127
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
Figure 6. HPLC profiles of all four 18F labeled 4-FGln and the “cold” standards. Stationary phase: Chiral Chirex3126 (D-penicillamine) (150 ꢀ 4.6
mm); Mobile phase: 1 mM CuSO4, 1 mL/min, 10 °C. Traces A-D are the radio traces of the four 18F labeled isomers of 4-FGln; trace E is the UV
spectra of the “cold” compounds 1, 2, 3, and 4 under identical HPLC conditions.
Scheme 7. Exploration of Radio-Fluorination Reaction: Preparation of [18F]1
its metabolites may have a better ability to exit the 9L cells (Figure 8).
We do not know if this is directly related to the transporter on the
membrane or due to other metabolic processes. Preliminary meta-
bolic studies of the cell uptake of the two L-Gln analogues, [18F]1
(2S,4R) and [18F]2 (2S,4S) suggested that the glutamines were
taken up in the 9L cells by glutamine transporters70 and predomi-
nantly converted to the corresponding [18F]4-fluoro-glutamic acid
(>60% at 30 min, 1 and 2 h after incubation) (data not shown). The
most likely enzyme responsible would be L-glutaminase.19 More
detailed studies will be needed to explore the mechanisms responsible
for intracellular metabolism and trapping inside the cells. Such studies
will be very important if these tracers are to be used for in vivo PET
imaging of tumor growth.
Using SF188-Bcl-xL cells (derived from human glioblastoma
cells), it was demonstrated that, as part of reprogramming
of cellular metabolism, the cells showed glutamine addiction
(Glutaminolysis).15,71 The results suggest that the transcriptional
regulatory properties of the oncogene Myc coordinate the expression
of genes necessary for cells to engage in glutamine catabolism for
protein and nucleotide biosynthesis. Myc-dependent glutaminolysis
1128
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
Figure 7. HPLC profiles of radio labeled intermediates from precursors 35 and 24 under K[222]/K2CO3 conditions (traces A and B) with “cold” 4-FGln
derivatives, 26, 27, 28, and 29 as reference (trace C). HPLC conditions, Chiralpak OD column; mobile phase, hexanes/EtOH 98.5/1.5; flow rate, 1.2 mL/min.
Scheme 8. Exploration of Radio-Fluorination Reaction: Preparation of All Four Isomers: [18F]1, [18F]2, [18F]3, and [18F]4
redirected mitochondrial metabolism to depend on glutamine
catabolism to sustain cellular viability and TCA cycle leading
to the production of oxaloacetate (anaplerosis). The ability of
Myc-expressing SF188-Bcl-xL cells to promote glutaminolysis does
not depend on concomitant activation of PI3K/Akt/mTOR path-
way, which activates “aerobic glycolysis” (Warburg effect).
1129
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
Figure 8. Cell uptake study results of all four [18F]4-FGln isomers ([18F]1, [18F]2, [18F]3, [18F]4), H-3 labeled Gln, and [18F]FDG. All radiotracers
were evaluated in the 9L tumor cell line.
Scheme 9. Tentative Explanation for the Formation of [18F]29 as a Major Product with Tosylate 25 (2S,4R) as the Precursor in
the Presence of 18-Crown-6 and KHCO3 and the Formation of a 1 to 1 Ratio of [18F]28 and [18F]29 under the K[222]/ K2CO3
Conditions
The stimulation of mitochondrial glutamine metabolism resulted in
reduced glucose carbon entering the TCA cycle and a decreased
contribution of glucose to the mitochondrial-dependent synthesis
of phospholipids. As such, SF188-Bcl-xL tumor cells will serve as a
suitable model for testing the glutamine uptake and retention from
which an estimation of glutamine utilization may be feasible.
To demonstrate that the transformed SF188-Bcl-xL tumor
cells can preferentially take up glutamine, we have studied the cell
previously, the SF188-Bcl-xL tumor cells use glutamine and glucose as
a source of metabolic energy.15,71 Notably, the uptake of glutamine
(10% dose/100 μg of protein) into the SF188-Bcl-xL tumor cells was
twice as high as that of 18F-FDG (5% dose/100 μg of protein) after
60 min incubation, suggesting that they are undergoing “glutamino-
lysis” (Figure 9).
Under the same in vitro incubation condition, the uptake of
two [18F]4-FGln, 1 (2S,4R), and 2 (2S,4S), into SF188-Bcl-xL
cells demonstrated a dramatically higher uptake than that of 3H-
glutamine and 18F-FDG (Figure 9). It is noted that [18F]4-FGln, 1
displayed an uptake of 22% dose/100 μg of protein at 60 min,
while the other isomer, [18F]4-FGln, 2 (2S,4S), showed a faster
kinetics, and the uptake peaked (21% dose/100 μg of protein) at
30 min after incubation. Preliminary metabolism analysis indicated
3
uptake of 18F-FDG and H-glutamine as tracers simultaneously in
phosphate buffered saline (PBS). Results presented in Figure 9
demonstrated a clear and prominent uptake of 3H-glutamine in these
cells (10-12% dose/100 μg of protein after 60 min incubation). The
uptake of glutamine and FDG in the cells suggest that the tumor cells
utilize both glucose (18F-FDG) and 3H-glutamine. As demonstrated
1130
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
Figure 9. Results of uptake of two [18F]4-FGln, 1 (2S,4R), and 2 (2S,4S), into SF188-Bcl-xL tumor cells: ([18F]1 in solid blue), ([18F]2 in dashed blue),
3H-Gln (in black), and [18F]FDG (in red).
Figure 10. In vitro cell uptake inhibition studies of [18F]4-FGln, 1 (2S,4R) conducted in 9L cells.
that the [18F]4-FGln, 1 (2S,4R) and 2 (2S,4S), both converted to
the corresponding glutamic acids as found in the 9L cells, but
[18F]4-FGln, 2 (2S,4S) showed a higher defluorination (data not
shown). The results suggest that the [18F]4-FGln, 1 (2S,4R) and 2
(2S,4S) may be better than 3H-glutamine with a higher propensity
as the alternative energy source. The differences in cell uptake
between the 4S versus 4R isomer of [18F]4-FGln were highly
significant. Despite some previous reports on 4-FGln, such dis-
parities on biological properties between stereoisomers have never
been predicted or mentioned before. The synthesis and radiolabel-
ing method reported in this paper will make it possible to
investigate the novel observation further. The in vitro tumor cell
uptake studies suggest that the [18F]4-FGln, 1 (2S,4R) will be a
preferred isomer for in vivo imaging. Detailed in vivo imaging
studies are warranted to fully characterize the relationship of tumor
uptake and glutaminolysis.
Among all the [18F]4-FGln isomers, [18F]4-FGln, 1 (2S,4R)
displayed the highest uptake by the 9L tumor cells. Because of
1131
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
this observation, we only used this [18F]4-FGln tracer in our
further investigations. In vitro cell uptake inhibition studies were
conducted in 9L cells in order to test the specificity of this
radiotracer (Figure 10). Two natural amino acids, L-Gln and
L-serine, served as the positive control inhibitors for the Gln
transporters. L-Gln inhibits the system N transporter and L-serine
inhibits the ASC transporter. Three different concentrations
were tested for each inhibitor ranging from 0.5 to 5 mM. After
30 min of coincubation of L-Gln and [18F]4-FGln, 1 (2S,4R), the
system N transporter was clearly inhibited. The results showed
nearly 90% inhibition (0.49% dose/100 μg protein inhibited vs
control 3.56% dose/100 μg protein). L-Serine also showed con-
siderable inhibition (1.41% inhibited vs control 3.56% dose/
100 μg protein) of the ASC transporter. MeAIB, which inhibits
the system A transporter, served as the negative control. At a con-
centration of 5 mM, there was little or no effect on the system A
transport system (3.69% inhibited vs control 3.56% dose/100 μg
protein). The results clearly demonstrate a dose-dependent
response to L-Gln and L-serine, with MeAIB showing no inhibi-
tion of the system A transport system.
’ ACKNOWLEDGMENT
We thank Stand Up 2 Cancer (SU2C) for financial support.
Authors thank Dr. Chaitanya Divgi for helpful discussions. Authors
also want to thank Dr. Carita Huang for her editorial assistance.
’ REFERENCES
(1) Jaeckel, C.; Koksch, B. Eur. J. Org. Chem. 2005, 4483.
(2) Meng, H.; Kalsani, V.; Kumar, K. ACS Symp. Ser. 2007, 949, 487.
(3) Jaeckel, C.; Salwiczek, M.; Koksch, B. Angew. Chem., Int. Ed.
2006, 45, 4198.
(4) Meng, H.; Kumar, K. J. Am. Chem. Soc. 2007, 129, 15615.
(5) Zheng, H.; Comeforo, K.; Gao, J. J. Am. Chem. Soc. 2009, 131, 18.
(6) Chiu, H.-P.; Kokona, B.; Fairman, R.; Cheng, R. P. J. Am. Chem.
Soc. 2009, 131, 13192.
(7) O'Hagan, D. Chem. Soc. Rev. 2008, 37, 308.
(8) Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc.
Rev. 2008, 37, 320.
(9) Mercer, J. R. J. Pharm. Pharm. Sci. 2007, 10, 180.
(10) McConathy, J.; Goodman Mark, M. Cancer Metastasis Rev.
2008, 27, 555.
Additional biological evaluation of the most promising Gln
analogue [18F]4-FGln, 1 (2S,4R) isomer in normal and tumor
bearing mice by PET imaging and biodistribution demonstrated that
the Gln tracer indeed localized in the tumor tissue in vivo (data not
shown). These results, which will be reported in other publications,
further support our findings that the new L-Gln analogue, [18F]4-
FGln, 1 (2S,4R), is metabolized by tumor tissue.
(11) Schuster, D. M.; Votaw, J. R.; Nieh, P. T.; Yu, W.; Nye, J. A.;
Master, V.; Bowman, F. D.; Issa, M. M.; Goodman, M. M. J. Nucl. Med.
2007, 48, 56.
(12) Laverman, P.; Boerman, O. C.; Corstens, F. H. M.; Oyen,
W. J. G. Eur. J. Nucl. Med. Mol. Imaging 2002, 29, 681.
(13) Jager, P. L.; Vaalburg, W.; Pruim, J.; de Vries, E. G.; Langen,
K. J.; Piers, D. A. J. Nucl. Med. 2001, 42, 432.
(14) Plathow, C.; Weber, W. A. J. Nucl. Med. 2008, 49 (Suppl. 2), 43S.
(15) Wise, D. R.; DeBerardinis, R. J.; Mancuso, A.; Sayed, N.; Zhang,
X.-Y.; Pfeiffer, H. K.; Nissim, I.; Daikhin, E.; Yudkoff, M.; McMahon,
S. B.; Thompson, C. B. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 18782.
(16) DeBerardinis, R. J.; Lum, J. J.; Hatzivassiliou, G.; Thompson,
C. B. Cell Metab. 2008, 7, 11.
(17) Vander Heiden, M. G.; Cantley, L. C.; Thompson, C. B. Science
2009, 324, 1029.
(18) Thompson, C. B. N. Eng. J. Med. 2009, 360, 813.
(19) Gao, P.; Tchernyshyov, I.; Chang, T. C.; Lee, Y. S.; Kita, K.;
Ochi, T.; Zeller, K. I.; De Marzo, A. M.; Van Eyk, J. E.; Mendell, J. T.;
Dang, C. V. Nature 2009, 458, 762.
’ CONCLUSIONS
In summary, a new synthetic pathway to 4-FGln derivatives
was developed. The synthesis of all four optically pure 4-FGln was
accomplished using a Passerini three-component reaction and
“neutralized” TASF nucleophilic fluorination reaction as key steps.
In addition, tosylates 24, 25, 35, and 36, which are suitable
precursors for 18F radio-fluorination reactions, were synthesized
on grams scale. It is reasonable to postulate that this synthetic
method could be used to synthesize various functional group
substituted (at the 4-position) and optically enriched Gln or
glutamic acid analogues, since fluoro-substituted functionalization,
one of the hardest functionalization processes, has been per-
formed. In addition, we successfully accomplished the syntheses
of four stereoisomeric 18F 4-FGlns, [18F]1, [18F]2, [18F]3, and
[18F]4. In vitro cell uptake studies (9L tumor cells) suggest that
uptake is highly specific for two of the L-Gln analogues, [18F]1
(2S,4R) and [18F]2 (2S,4S), and it is likely associated with the
native L-Gln transporters. Between them, [18F]1 (2S,4R) was
selected as the leading probe for further biological evaluation. We
hope that the successful development of this new metabolic probe
for the detection of glutaminolysis will contribute to the advance-
ment of cancer diagnosis and therapy.
(20) Dang, C. V. Cancer Res. 2010, 70, 859.
(21) Qiu, X.-L.; Meng, W.-D.; Qing, F.-L. Tetrahedron 2004, 60, 6711.
(22) Smits, R.; Cadicamo, C. D.; Burger, K.; Koksch, B. Chem. Soc.
Rev. 2008, 37, 1727.
(23) Pigza, J. A.; Quach, T.; Molinski, T. F. J. Org. Chem. 2009, 74, 5510.
(24) Hu, Z.; Han, W. Tetrahedron Lett. 2008, 49, 901.
(25) Padmakshan, D.; Bennett, S. A.; Otting, G.; Easton, C. J. Synlett
2007, 1083.
(26) Li, G.; van der Donk, W. A. Org. Lett. 2007, 9, 41.
(27) Suzuki, A.; Mae, M.; Amii, H.; Uneyama, K. J. Org. Chem. 2004,
69, 5132.
(28) Hoveyda, H. R.; Pinault, J.-F. Org. Lett. 2006, 8, 5849.
(29) Chorghade, M. S.; Mohapatra, D. K.; Sahoo, G.; Gurjar, M. K.;
Mandlecha, M. V.; Bhoite, N.; Moghe, S.; Raines, R. T. J. Fluorine Chem.
2008, 129, 781.
(30) Charrier, J.-D.; Hadfield, D. S.; Hitchcock, P. B.; Young, D. W.
Org. Biomol. Chem. 2004, 2, 797.
’ ASSOCIATED CONTENT
(31) Charrier, J.-D.; Hadfield, D. S.; Hitchcock, P. B.; Young, D. W.
Org. Biomol. Chem. 2004, 2, 474.
(32) Charrier, J.-D.; Hitchcock, P. B.; Young, D. W. Org. Biomol.
Chem. 2004, 2, 1310.
(33) Qiu, X.-L.; Meng, W.-D.; Qing, F.-L. Tetrahedron 2004, 60,
5201.
S
Supporting Information. Full experimental procedures
b
and data, 1H/13CNMR spectra of new compounds, HPLC
profiles and reported X-ray structural data. This material is
available free of charge via the Internet at http://pubs.acs.org.
(34) Yajima, T.; Nagano, H. Org. Lett. 2007, 9, 2513.
(35) Mykhailiuk, P. K.; Afonin, S.; Chernega, A. N.; Rusanov, E. B.;
Platonov, M. O.; Dubinina, G. G.; Berditsch, M.; Ulrich, A. S.; Komarov,
I. V. Angew. Chem., Int. Ed. 2006, 45, 5659.
’ AUTHOR INFORMATION
Corresponding Author
wenchao@mail.med.upenn.edu; kunghf@sunmac.spect.upenn.edu
1132
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133

Journal of the American Chemical Society
ARTICLE
(36) Mykhailiuk, P. K.; Afonin, S.; Palamarchuk, G. V.; Shishkin,
O. V.; Ulrich, A. S.; Komarov, I. V. Angew. Chem., Int. Ed. 2008, 47, 5765.
(37) Shendage, D. M.; Froehlich, R.; Bergander, K.; Haufe, G. Eur. J.
Org. Chem. 2005, 719.
(38) Dave, R.; Badet, B.; Meffre, P. Amino Acids 2003, 24, 245.
(39) Tolmana, V.; Sedmera, P. J. Fluorine Chem. 2000, 101, 5.
(40) (a) Dinkelborg, L.; Friebe, M.; Krasikowa, R. N.; Belokon, Y.;
Kuznetsova, O. F.; Graham, K.; Lehmann, L.; Berndt, M. EP 2007/
009518, WO/2008/052788, 2008. (b) Krasikova, R. N.; Kuznetsova,
O. F.; Fedorova, O. S.; Belokon, Y. N.; Maleev, V. I.; Mu, L.; Ametamey,
S.; Schubiger, P. A.; Friebe, M.; Berndt, M.; Koglin, N.; Mueller, A.;
Graham, K.; Lehmann, L.; Dinkelborg, L. M. J. Med. Chem. 2010, DOI:
10.1021/jm101068q. (c) Friebe, M.; Schmitt-willich, H.; Berndt, M.;
Dinkelborg, L.; Koglin, N.; Graham, K. U.S. Patent Application
20100290991-A1.
(41) Domling, A. Chem. Rev. 2006, 106, 17.
(42) Hulme, C.; Gore, V. Curr. Med. Chem. 2003, 10, 51.
(43) Hudson, D. USP 4 935 536, 1990.
(44) Weygand, F.; Steglich, W.; Bjarnason, J. Chem. Ber 1968, 101, 3642.
(45) Pietta, P. G.; Biondi, P. A.; Brenna, O. J. Org. Chem. 1976, 41,
703.
(46) Armstrong, A.; Brackenridge, I.; Jackson, R. F. W.; Kirk, J. M.
Tetrahedron Lett. 1988, 29, 2483.
(47) Bergmeier, S. C.; Cobas, A. A.; Rapoport, H. J. Org. Chem. 1993,
58, 2369.
(48) Adlington, R. M.; Baldwin, J. E.; Catterick, D.; Pritchard, G. J. J.
Chem. Soc., Perkin Trans. 1 1999, 855.
(49) Ramsamy, K.; Olsen, R. K.; Emery, T. Synthesis 1982, 42.
(50) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.
(51) Werner, R. M.; Shokek, O.; Davis, J. T. J. Org. Chem. 1997, 62,
8243.
(52) Wernic, D.; DiMaio, J.; Adams, J. J. Org. Chem. 1989, 54, 4224.
(53) Pigge, F. C.; Coniglio, J. J.; Fang, S. Organometallics 2002, 21,
4505.
(54) Costa, S. P. G.; Maia, H. L. S.; Pereira-Lima, S. M. M. A. Org.
Biomol. Chem. 2003, 1, 1475.
(55) Masaki, M.; Kitahara, T.; Kurita, H.; Ohta, M. J. Am. Chem. Soc.
1968, 90, 4508.
(56) Naruto, M.; Ohno, K.; Naruse, N.; Takeuchi, H. Tetrahedron
Lett. 1979, 251.
(57) Middleton, W. J. J. Org. Chem. 1975, 40, 574.
(58) Konas, D. W.; Coward, J. K. J. Org. Chem. 2001, 66, 8831.
(59) Pilcher, A. S.; Ammon, H. L.; DeShong, P. J. Am. Chem. Soc.
1995, 117, 5166.
(60) Gingras, M. Tetrahedron Lett. 1991, 32, 7381.
(61) Chang, C.-Y.; Yang, T.-K. Tetrahedron: Asymmetry 2003, 14,
2239.
(62) Yin, J.; Zarkowsky, D. S.; Thomas, D. W.; Zhao, M. M.;
Huffman, M. A. Org. Lett. 2004, 6, 1465.
(63) RajanBabu, T. V.; Middleton, W. J.; Tortorelli, V. J.; e-EROS
Encyclopedia of Reagents for Organic Synthesis; John Wiley & Sons: New
York, 2001
(64) Greene, T. W.; Wuts, P. G. M. , 3rd ed.; John Wiley & Sons,
Inc.: New York City, 1999.
(65) Krasnov, V. P.; Koroleva, M. A. Russ. Chem. Bull. 1995, 631.
(66) Krasnov, V. P.; Koroleva, M. A.; Evstigneeva, N. G.; Nizova,
I. A. Russ. Chem. Bull. 1995, 635.
(67) Kim, D. W.; Ahn, D. S.; Oh, Y. H.; Lee, S.; Kil, H. S.; Oh, S. J.;
Lee, S. J.; Kim, J. S.; Ryu, J. S.; Moon, D. H.; Chi, D. Y. J. Am. Chem. Soc.
2006, 128, 16394.
(68) Yu, W.; Williams, L.; Camp, V.; Olson, J.; Goodman, M. Bioorg.
Med. Chem. Lett. 2010, 20, 2140.
(69) Wang, L.; Qu, W.; Lieberman, B.; Ploessl, K.; Kung, H. Bioorg.
Med. Chem. Lett. 2010, 20, 3482.
(70) McGivan, J. D.; Bungard, C. I. Front. Biosci. 2007, 12, 874.
(71) DeBerardinis, R. J.; Mancuso, A.; Daikhin, E.; Nissim, I.;
Yudkoff, M.; Wehrli, S.; Thompson, C. B. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 19345.
1133
dx.doi.org/10.1021/ja109203d |J. Am. Chem. Soc. 2011, 133, 1122–1133