Enantioselective synthesis of carbon-11 labeled L-alanine using phase transfer catalysis of Schiff bases

Article abstract of DOI:10.1016/j.tet.2016.08.069

Radiolabeled amino acids are an important class of compounds that can be used for Positron Emission Tomography (PET) imaging of the amino acid transporter status of various diseases e.g., cancer. Current radiochemistry techniques do not offer synthesis ap

Full text of DOI:10.1016/j.tet.2016.08.069

Tetrahedron 72 (2016) 6551e6557
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Enantioselective synthesis of carbon-11 labeled -alanine using
L
phase transfer catalysis of Schiff bases
*
ꢀ
Ulrike Filp , Aleksandra Pekosak, Alex J. Poot, Albert D. Windhorst
Radionuclide Center, Radiology and Nuclear Medicine, VUmc Amsterdam, De Boelelaan 1085c, 1081HV Amsterdam, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 May 2016
Received in revised form 12 August 2016
Accepted 23 August 2016
Available online 27 August 2016
Radiolabeled amino acids are an important class of compounds that can be used for Positron Emission
Tomography (PET) imaging of the amino acid transporter status of various diseases e.g., cancer. Current
radiochemistry techniques do not offer synthesis approaches that are generally applicable and result in
high yields and enantiomeric purity. Here, the radiosynthesis of L
-[11C]alanine is described employing an
enantioselective alkylation of a Schiff base glycine precursor with [11C]methyl iodide. By conducting
a comprehensive reaction conditions optimization and a strategic analysis of several phase-transfer
Keywords:
CeC coupling
Amino acids
Enantioselective synthesis
Radiolabeling
catalysts that facilitate enantioselective alkylation, the radiosynthesis of L
-[11C]alanine was achieved in
good radiochemical conversion, short reaction times and above 90% enantiomeric excess. This new
methodology is broadly applicable and could also be used for the radiolabeling of other amino acids with
carbon-11.
Ó 2016 Elsevier Ltd. All rights reserved.
Positron Emission Tomography (PET)
1. Introduction
neurological disorders as well, which is proven by the application of
L
-[11C]DOPA14,15 as important radiolabeled neurotransmitter.
Since current amino acid based PET tracers show good results,
Positron Emission Tomography (PET)1 is a non-invasive tech-
nique, often applied as diagnostic tool in today’s healthcare, that
allows the visualization of cellular processes in vivo in real time to
study diseases like cancer or neurological disorders. Two important
classes of compounds in molecular imaging are amino acids
and peptides, which are used to establish new biological targets
and tools for a better disease diagnosis and treatment strategies.2e4
As diagnostic agents for oncology imaging, radiolabeled amino
acids often have improved sensitivity and specificity over other
there is need for improved and general methods for the radiosyn-
thesis of this class of PET tracers. Optimally, for amino acid based
PET tracers it is desired that the native structure of the amino acid is
not changed, hence properties of amino acids by exchanging a car-
bon-12 atom for a carbon-11 is beneficial for guaranteeing real
natural behavior. Thereby, carbon-11 labeled amino acids can be
tracked into the metabolic pathways in which they are involved.
Current precursor molecules for the synthesis of radiolabeled
amino acids and amino acid derivatives already contain the chi-
rality of the desired radiolabeled product and this involves chal-
lenging precursor synthesis. Furthermore, using chiral starting
materials have the uncertainty if chirality is maintained during
radiosynthesis, since these reactions often require harsh condi-
tions. Nevertheless, amino acids are chiral molecules and conse-
quently a synthesis is needed that results in an enantiomeric pure
product to avoid chiral separation and a 50% loss of the final radi-
olabeled product resulting in a low yield. Final challenges in the
synthesis of radiolabeled amino acids with carbon-11 is the syn-
thesis time, since carbon-11 is a short-lived radioisotope with
a half-life of 20.4 min and therefore reactions are required to pro-
ceed within minutes instead of hours that are reported for non-
radioactive synthesis.16
PET tracers for oncology imaging, like 2-[18F]fluoro-2-deoxy-
D-
glucose (FDG).5 Amino acid uptake is increased to support the rapid
growth and proliferation of tumor cells, by which amino acids are
used as nutrients or for protein synthesis.6,7 Various studies have
shown that amino acid transporters are elevated on tumor tissue.8
Making use of this knowledge, important tracers like [11C]methi-
onine9 and O-(2-[18F]fluoroethyl)- -tyrosine10 are used routinely
L
in clinical settings for tumor diagnosis. Furthermore, multiple
amino acids and amino acid analogs, e.g., [11C]glutamine11 or
derivatives [18F](2S,4S)-4-(3-Fluoropropyl)glutamine12 and 3-(1-
[
18F]fluoromethyl)- -alanine13 are under preclinical development.
L
Next to oncologic disorders, amino acids can be used to study
With respect to the asymmetric synthesis of amino acids, chiral
alkylation of Schiff base glycine derivatives using Phase-Transfer
* Corresponding author. Fax: þ31 20 44 49121; e-mail address: u.filp@vumc.nl
(U. Filp).
http://dx.doi.org/10.1016/j.tet.2016.08.069
0040-4020/Ó 2016 Elsevier Ltd. All rights reserved.

6552
U. Filp et al. / Tetrahedron 72 (2016) 6551e6557
Catalysis (PTC) (Scheme 1)17 is an ideal and general applicable
method to synthesize natural and unnatural amino acids. Though
rarely reported, the synthesis of alanine has been described uti-
lizing methyl iodide as alkylating reagent (O’Donnell:18 ee (enan-
tiomeric excess) not reported; Corey:19 ee of 97%). Due to the
small size of methyl iodide compared to more bulky alkylating
agents ee is mostly lower or is not evaluated at all. Nowadays,
asymmetric synthesis is possible for many amino acids and small
peptides.17,20,21
on a reverse-phase analytical column. The deprotected reaction
mixture check was performed using a chiral column to determine
in which D/L
-[11C]alanine 3 was separated and allowed the calcu-
lation of the enantiomeric excess of the final product.
To explore the reactivity of [11C]MeI towards precursor 1, the
procedure as was described by Kato et al., was investigated.28,29
Schiff base 1 was suspended in DMSO and in the presence of
TBAF-solution (Tetrabutylammonium fluoride, 1 M in THF) as
a base, [11C]MeI was added to the reaction mixture by direct dis-
tillation. Alkylation of 1 with [11C]MeI according to the published
procedure was successful and alkylation yields exceeded 80%
The radiosynthesis of [11C]alanine has first been reported in the
late 1970’s when Langstrom et al. described a 48% ee yield of [11C]
alanine utilizing an asymmetric synthesis procedure.22 Neverthe-
less, it took more than 10 years to develop a synthesis that yielded
ꢁ
€
(Fig. 1B). Deprotection of alkylated intermediate 2 was to yield
[
D/L-
11C]alanine 3, proved to be straight forward and high yielding
80% ee of
L
-[11C]alanine.23,24 Alternatively other strategies have
when 6 M solution of HCl was added to the reaction mixture and
heated shortly. As anticipated for this part of the study, no enan-
tiomeric selectivity was obtained in the alkylation reactions, which
was also demonstrated by the obtained chiral HPLC chromatograms
for [11C]alanine (Fig. 1C). Besides the use of TBAF to synthesize
amino acids, also inorganic alkali-metal bases are often described
in the chiral synthesis of amino acids by alkylation. Therefore, next
to the use of TBAF as organic base, inorganic alkali-metal bases
were investigated as well to evaluate the suitability of these bases
emerged as well, which make use of Nickel-complexes and upon
varying the alkylating agent, many amino acids are possible.25e27
However, the synthesis of these Nickel-reagents is considered
cumbersome and the release of the unprotected amino acid by
hydrolysis of the complex is tedious. Another drawback of the use
of Ni-complexes in radiochemistry is that it only allows the syn-
thesis of single radiolabeled amino acids, whereas the methodology
that we developed should allow translation towards peptide radi-
olabeling with carbon-11.
In this paper we have adopted the use of PTC for the chiral
radiosynthesis of L
-[11C]alanine to demonstrate the use of this
method for the enantioselective radiolabeling of amino acids and as
potential strategy for PET tracer development. An improved
asymmetric synthesis of L
-[11C]alanine by an enantioselective al-
kylation of a Schiff base glycine precursor with [11C]methyl iodide
([11C]MeI) is here described. We focused our radiolabeling ap-
proach on asymmetric synthesis (Scheme 1) with highly special-
ized chiral catalysts, as it uses an accessible precursor, low amounts
of catalyst and we could implement [11C]MeI as our first alkylating
agent. With more sophisticated alkylating agents this methodology
is applicable as well to acquire other amino acids. Ultimately, future
research with the methodology presented in this paper to syn-
thesize L
-[11C]alanine, should allow the synthesis of radiolabeled
peptides with carbon-11 as PET tracers.
2. Results and discussion
Initial focus of this study was the radiosynthesis of racemic D/L-
[
11C]alanine to study the reactivity of carbon-11 labeled alkylating
reagents towards the Schiff base (1) and the required reaction
conditions for these reactions. As a precursor for the synthesis of
[
11C]alanine, glycine derivative 1 was used, which was modified as a
Schiff base at the N-terminus as a biphenyl imine to activate the a-
carbon of glycine for alkylation. Furthermore, the C-terminal car-
boxylic acid was protected as a tert-butyl ester during the alkylation
reactions. To thoroughly study the radiochemical conversion of
the alkylation reaction of precursor 1 with [11C]MeI and the fol-
lowing deprotection under acidic conditions and the enantiomeric
excess of the final product, analysis was performed with High
Performance Liquid Chromatography (HPLC) of both reactions in-
dependently. The analysis of the alkylation reaction was performed
Fig. 1. (A) Radiochemical synthesis of [11C]alanine by alkylation of precursor 1 with
[
11C]MeI and its acidic deprotection; (B) HPLC profiles of the UV and radioactive signal
of the crude alkylation mixture of 1 with [11C]MeI; (C) Analysis of deprotected D/L-[11C]
alanine by chiral HPLC.
Scheme 1. Proposed mechanism of an asymmetric alkylation of a Schiff’s base for amino acid synthesis.

U. Filp et al. / Tetrahedron 72 (2016) 6551e6557
6553
to deprotonate 1 and perform alkylation reactions of Schiff bases
performed at 0e10 ꢁC for 5e10 min with generally 7
cursor 1 and 10 mol % of catalyst.
mmol of pre-
with [11C]MeI and synthesize [11C]alanine. Unfortunately, when
using aqueous solutions of NaOH no alkylation was observed to 2 at
room temperature. Likewise, only low radiochemical conversions
were observed while using CsOHꢀH2O as a solid base for the al-
kylation of 1 with [11C]MeI. An explanation for the low reactivity of
precursor 1 towards [11C]MeI is the poor solubility of alkali metal
hydroxides in organic solvents, whereas TBAF is soluble in organic
solvents and can act as a base more easily. A general observation
from all alkylation reactions that were investigated and analysed by
HPLC in this study (Fig. 1 B), was the formation of very high con-
centrations of benzophenone in the reaction mixture coming from
precursor 1. It can be concluded that precursor 1 is clearly unstable
during the alkylation reactions, yielding high concentration of
benzophenone and unreactive glycine tert-butyl ester. Despite
these high concentrations of benzophenone, the investigated al-
kylation reactions with [11C]MeI proved to be successful. This can
be explained by the stoichiometry of Schiff base 1 to [11C]MeI in
radiochemical alkylation reactions. [11C]MeI is present in nano-
molar concentrations, meaning that, despite the high concentra-
tions of benzophenone in the reaction, there is still a large excess of
Next to TBAF as base, many studies in organic chemistry have
performed enantioselective alkylations with alkali metal hydrox-
ides aqueous solution as base.31 The main advantage of using
aqueous solutions of metal hydroxides lays in the more accurate
amount of base added to the reaction mixture as otherwise possible
with hygroscopic alkali hydroxides. Therefore, we initially exam-
ined various amounts of aqueous alkaline bases in the asymmetric
alkylation reactions, which was added to the reaction mixture
containing 1 and [11C]MeI in toluene as organic solvent. In general,
the observed alkylation conversion to obtain compound 2 was low
and never exceeded 50%. Despite the low conversion, we carefully
analyzed the ee of the obtained product by chiral HPLC (Fig. 3) and
discovered that the application of PTC 4 induced the chiral alkyl-
ation of Schiff bases with [11C]MeI resulting in moderate to high
ee of L
-[11C]alanine. As the conversion rates of the alkylation re-
actions, using these conditions, was unpredictable and too low,
a more reactive alkylation reagent, [11C]MeOTf ([11C]methyl tri-
fluoromethanesulfonate) instead of [11C]MeI, was investigated to
enhance the alkylation reaction. Unfortunately, only low conver-
sions were observed and [11C]MeOTf was not further used as al-
kylation reagent in this study. To increase the reaction yields and
maintain the high ee’s of the reactions (Fig. 3A), mixtures of
aqueous CsOH solution and organic bases like TBAF, TBAOH (Tet-
ramethyl ammoniumhydroxide) or TBAHSO4 (Tetrabutylammo-
nium hydrogensulfate) were used, indeed resulting in high
conversions of 90%, which is in accordance with previous findings.
Unfortunately however, the ee of all reactions dropped to un-
desirable and low rates.
precursor 1, which is present in mmoles, available for alkylation.
Since this instability did not hamper the radioalkylation reactions
investigated, no further attention was paid to this observation.
2.1. Chiral alkylation reactions with phase-transfer catalyst 4
These encouraging initial results formed the basis to move for-
ward to the asymmetric synthesis of [11C]alanine to selectively
obtain the D- or L-enantiomer. To achieve this, the optimization of
Since alkali bases proved to be optimal thus far with respect to
enantioselectivity of the alkylation reaction, other Cesium bases
were examined. Aqueous solutions of 1 M and 10 M concentrations
of Cs2CO3 were evaluated for the alkylation reaction, but no con-
version was observed in any of the attempts. Furthermore, semi-
organic base 1 M aqueous Cesium acetate was used in the alkyl-
ation, however, no reaction between 1 and [11C]MeI was observed.
Presumably this is due to lower basicity of the used Cesium salts
compared to CsOHꢀH2O.
catalyst, temperature, solvent and time were taken into account
and modified to achieve near quantitative radiochemical yield and
as high as possible enantiomeric excess. Executing the reaction
conditions as were described in organic literature was a first set-off
point in this study.30 The initial challenge in the application of
phase-transfer catalysis reactions in radiochemistry is that in or-
ganic chemistry mixtures are reacted for several hours before
work-up, which is not possible working with carbon-11, where the
maximum time of reaction and analysis of the product is 3 half-
lives. This study was set out with the aim of the enantioselective
To further explore the use of CsOH and its application for chiral
alkylation reactions, CsOH was used as a dry powder (CsOH solid
base). This led to a significant increase in the conversions of the
reaction of over 80%, when an excess of base was used compared to
precursor 1 (Fig. 3B). Furthermore, the stereoselectivity of the al-
kylation reaction was influenced dramatically when using CsOH
synthesis of L
-[11C]alanine and the first set of experiments con-
centrated on phase-transfer catalyst 4 (Fig. 2). Other PTCs have
been investigated in this study as well after optimization of the
chiral alkylation with PTC 4, to determine the influence of the
catalyst on the ee of the product (Fig. 2). The reactions were
Fig. 2. Chiral phase-transfer catalysts explored for the radiosynthesis of D/L-[11C]alanine with (A) quaternary ammonium based catalysts and (B) the cinchonidinium based catalysts.

6554
U. Filp et al. / Tetrahedron 72 (2016) 6551e6557
Fig. 3. (A) Enantiomeric excess of L-[11C]alanine 3 with different concentrations of aqueous CsOH solutions. (B) HPLC conversions for the alkylation reaction of Schiff base 1 with
11C]MeI in the presence of different bases. (C) Enantiomeric excess of L-[11C]alanine 3 in the presence of different bases. (D) Conversion rates of asymmetric alkylation to compound
2 and ee of the formed product in the presence of different solvent mixtures. (E) Analysis of D/L-[11C]alanine by chiral HPLC after deprotection of the enantiomeric product.
[
solid base and the ee of the obtained [11C]alanine was increased to
over 80% when using 40 equiv of CsOH (Fig. 3C). A drawback, of
using CsOH as a solid base, is the hygroscopicity and its commercial
availability only as CsOHꢀH2O. Therefore, one other alkali metal
that is less hygroscopic than CsOH hydroxide was investigated.17,20
Fig. 3B depicts the results for various concentrations of solid KOH in
comparison to CsOH, when the reactions were performed at 5 ꢁC.
In comparison to the use of CsOH, the obtained results with KOH
were not as desired since the conversion rates of the alkylation
reaction dropped and became unreliable. Therefore, it was con-
cluded that only the strongest metal hydroxide base is preferred
over the other available metal hydroxides, meaning that CsOH is
superior over KOH. These results led us to believe, that the use of
solid CsOH base is most promising and we were able to achieve
both high conversion and ee. An additional benefit for using solid
bases instead of aqueous solutions was the rate of hydrolysis of
Schiff base precursor 1, which was much less compared to aqueous
solutions and after completion of the alkylation reaction we had
more intact precursor available. To increase the stability of the re-
action even further during the alkylation reaction, CsOHꢀH2O was
azeotropically dried with MeCN beforehand. Unfortunately, the use
of azeotropically dried CsOH resulted in a reduction in ee ratios of
the obtained [11C]alanine to cca. 70% and therefore these prepara-
tions were discontinued.
Finally the impact of the solvent on the alkylation reaction was
assessed as well as the ee of the formed product, therefore various
solvent mixtures have been used. In a few papers the use of tolu-
ene32 was described, which yielded satisfactory results already if
CsOH as a solid base (w20 equiv) was used, however other describe
mixtures of toluene/dichloromethane33 resulting in an improve-
ment in both conversion yield and ee of the obtained product
(Fig. 3D). The best result was the use of a small amount of 5% of
dichloromethane in toluene. An explanation for this improved al-
kylation reaction can be found in the fact that it improves the
solubility of the catalyst in the reaction mixture.
2.2. Screening of multiple phase-transfer catalysts
With the currently available optimal conditions of the chiral
synthesis of L
-[11C]alanine, other chiral PTCs have been evaluated to
further improve the ee of the obtained products. Therefore classical
PTCs, termed first generation cinchonidonium salts 7 and 8 were
investigated, as well as dimeric cinchonidinium salt 9 and 10
(Fig. 2B). To complete the series of PTCs quaternary ammonium
based compounds were used, namely compound 5, which is an
stereoisomer of 4 and should result in the formation of D
-[11C]ala-
nine and catalyst 6 which should, because of its bulkier character,
improve the ee of the reaction (Fig. 2A).
All depicted PTCs in Fig. 2 have been applied for the chiral al-
kylation of Schiff base precursor 1 and its alkylation with [11C]MeI
using the optimal conditions that were investigated previously
when applying PTC 4. Results of these alkylation reactions are
summarized in Table 1.
PTC 5, which is the stereoisomer of PTC 4 should yield D
-[11C]
alanine predominantly and was evaluated to prove that the theory
of using inverse catalysts for this kind of alkylation reactions is also
valid in radiochemistry resulting in the formation of the opposite
enantiomer (Table 1). Nevertheless, alkylation results and the ob-
tained ee of the product, when using PTC 5, was lower than for PTC
4. To find and explanation for the less optimal performance of PTC 5
in comparison to its counterpart PTC 4, the optical and chemical
purity of both catalysts was further investigated. Unfortunately,
Table 1
Results from the alkylation reactions (n>3, decay corrected) and the ee of the ob-
tained product using PTC 5 to 10 with the optimal conditions obtained for catalyst 4
Entry
Applied PTC
Alkylation conversion (%)
ee of [11C]alanine (%)
1
2
3
4
5
6
7
4
5
6
7
8
9
10
96.9ꢂ1.1
47.5ꢂ9.4
6.6ꢂ0.5
Trace
72.7ꢂ5.5
73.8ꢂ9.8
59.7ꢂ27.1
88.2ꢂ2.0
58.3ꢂ9.1
80.6ꢂ8.0
ND
D-ala
90.4ꢂ2.7
Thus far, our optimization study showed that with CsOH as solid
base in a mixture of toluene/dichloromethane with 10 mol % of PTC
4 yielded the best results with respect to the obtained radio-
68.2ꢂ3.2
52.5ꢂ11.3
chemical conversion and the ee of L
-[11C]alanine.

U. Filp et al. / Tetrahedron 72 (2016) 6551e6557
6555
despite multiple attempts to investigate the optical rotation of the
catalyst and the analysis of the purity by HPLC, no full explanation
for the observed performance of the catalyst could be found.
Therefore, to optimize the performance of PTC 5 and thus the
that demonstrates a successful enantioselective radiosynthesis
with readily available [11C]MeI as reagent in radiochemistry, the
smallest alkylating agent possible, where in organic chemistry
most do not even establish reactions with methyl iodide. Taking
into account all these results, the best results for the radiosynthesis
of [11C]alanine by the alkylation of Schiff base 1 were obtained
using catalyst 4 with respect to yield and ee. All in all, a good in-
corporation was achieved when the alkylation reaction proceeded
at 5 ꢁC for 5e10 min in a mixture of toluene/dichloromethane in the
presence of 30 equiv of CsOH as base and 0.1 equiv of PTC 4,
respectively.
radiosynthesis of
D
-[11C]alanine, the reaction conditions for the
alkylation reactions with this catalyst were slightly modified to
yield the best possible conditions. Despite changes in temperature
ranging from 5 to 45 ꢁC and the change towards other alkaline
bases such as NaOH and KOH, the only real improvement when
using PTC 5 was increasing the amount of CsOH to 100 equiv
compared to the precursor. Using these conditions the conversion
of the alkylation reaction of precursor 1 with [11C]MeI was im-
To conclude, L
-[11C]alanine was synthesized within 50 min in
proved to 97%, but the ee of the obtained product,
never exceeded 66%.
D
-[11C]alanine,
high radiochemical yield of 20% (decay corrected) calculated from
the end of bombardment (EOB). The specific activity was >50 GBq/
The final chiral quaternary ammonium salt that was in-
vestigated in this study was PTC 6. Due to the bulky character of this
PTC 6 it was expected that this could positively influence the
enantioselectivity of the alkylation reaction. Unfortunately, with
PTC 6 the conversion of the alkylation reaction was less than 5%,
even when high concentrations of CsOH were used in the alkylation
reaction. As a result of the low conversion in the alkylation reaction,
the enantiomeric ratios could not be reliably determined and the
use of PTC 6 in this study was discontinued.
mmol at the end of the synthesis (EOS) and the highest ee achieved
was >90%. Both chemical and radiochemical purities were >95%.
3. Conclusion
We have synthesized L
-[11C]alanine via a new, general applica-
ble, radiochemistry method utilizing a PTC catalyzed enantiose-
lective alkylation with [11C]MeI, followed by acidic deprotection.
These findings provide new insights in the suitability of phase-
transfer catalyzed reactions in radiochemistry and the potential
to synthesize novel amino acid PET tracers using this methodology.
Finally, this radiosynthesis strategy would allow the synthesis of
carbon-11 radiolabeled peptides as well.
Next to the quaternary ammonium salts as PTC, first generation
cinchonidinium based PTCs have been investigated as chiral aux-
iliary for the alkylation of Schiff base 1 with [11C]MeI. When ex-
ploring PTCs 7 and 8,34e36 two catalysts that only differ in the
hydroxyl-position, PTC 8 outperformed 7 and good results were
achieved. Also when PTC 8 was applied for the chiral alkylation
reactions, CsOH in an excess of 20e40 equiv is optimal in a mixture
of toluene and dichloromethane. When using PTC 8 and a temper-
ature of 5 ꢁC, an alkylation conversion of 73% could be achieved
with an ee of the reaction of 90%, which was very satisfactory. More
recently, dimer PTCs have been described that are based on the first
generation cinchonidonium based catalysts. Both PTC 933,37 and 10
have been investigated and it was expected that the more volu-
minous character of both PTCs would again increase the stereo-
selectivity of the alkylation reaction. Unfortunately, neither PTC 9
nor 10 did yield any reasonable conversion in the alkylation re-
action nor a satisfactory enantiomeric excess. PTC 9 only resulted in
traces of the alkylated product, even with the adaption of the re-
action conditions in which the amounts of base and the tempera-
ture were changed. In the end, PTC 10, which has been described in
the literature as highly promising for organic chemistry33 proved to
be the more successful PTC of the dimer cinchonidinium based
PTCs. When using PTC 10 with optimized conditions, that were
earlier described for PTC 4, a conversion could be achieved of 61%
with an ee of 53%. Nevertheless, both conversion and ee of the
product were lower than obtained in the presence of PTC 4 or 8.
To summarize all achieved results within this study, phase-
transfer catalysis allows the enantioselective synthesis of carbon-
11 labeled amino acids as potential PET tracers, which was shown
in this study for the radiosynthesis of [11C]alanine. Despite these
achievements, it should be noted that there are still uncertainties
about the mode of action, origin of the enantioselectivity
and enolate binding. A quantum-mechanical analysis provided by
Cook gave insights into the lowest energy enantiomeric allylation
transition states of cinchonidinium-derived catalysts, which pro-
4. Materials and methods
4.1. General
N-(diphenylmethylene)glycine tert-butyl ester 1 was purchased
from ABCR (Karlsruhe, Germany) and catalysts were purchased
from Sigma Aldrich (Zwijndrecht, The Netherlands) and Wako Pure
Chemical Industries (Osaka, Japan). All commercially available
chemicals were used without further purification. The non-
radioactive reference compounds were synthesized according to
reported methods and were used to verify the identity of radio-
labeled compounds. 1H and 13C nuclear magnetic resonance (NMR)
spectra were obtained using a Bruker AC 250 and AC 400 (Billerica,
USA) and chemical shifts (
the solvent (7.27 ppm for CDCl3, 3.31 ppm for MeOD, 2.50 ppm for
DMSO-d6) and tetramethylsilane as an internal standard (
d) were defined relative to the signal of
d¼0).
High resolution mass spectra (HRMS) were carried out using
a Bruker micrOTOF-Q instrument in positive or negative ion mode
(capillary potential of 4500 V). Flash chromatography purifications
were performed on a Buchi system operated by SepacoreControl
system. Analytical HPLC systems used were equipped with: a Wa-
ters 600E pump, a manual Rheodyne injector (20e100 mL loop),
a Waters PDA and GinaStar software from Raytest (Straubenhardt,
Germany). The radioactive profile was monitored with a Raytest 2.5
inch radioactivity detector (Raytest, Germany). Analytical HPLC
columns used in this study were a Grace Smart C-18 5
mm,
4.6ꢀ250 mm with a mixture consisting of acetonitrile/4 mM so-
dium formateþ4% DMF (70/30, v/v) at a constant flow rate of 1 mL/
min for the analysis of 2, UV monitoring at 254 nm. Enantiomeric
purity of the amino acid was determined using an analytical
vides us with
L
-[11C]alanine.38 Catalyst 4 and 5 are conformational
Reprosil chiral-aa (8
mm; 4.6ꢀ250 mm) from Dr. Maisch GmbH
rigid homochiral quaternary ammonium bromides and have been
shown to have a substantially higher catalytic activity then their
heterochiral diastereomers.31,39 With this theory it could be
explained why PTC 4 was most successful in our study, due to their
favorable transition states. Nevertheless, in radiochemistry the
alkylating agent is present in a very small amount compared to
both the Schiff base and the catalyst. Notably, this is the first report
(Ammerbuch, Germany) at 214 nm. The product was eluted with
methanol/water (70/30, v/v) at a flow rate of 1 mL/min or as stated
otherwise. Radiochemical conversions are based on the AUC of the
radioactivity profile of HPLC analysis. The [11C]MeI synthesis was
performed before every radiochemical reactions on an in-house
built synthesis device and according to procedures described in
literature.

6556
U. Filp et al. / Tetrahedron 72 (2016) 6551e6557
4.2. Synthesis of reference compounds and non-
commercially available PTCs
5.37e5.28 (m, 4H), 5.19e5.08 (m, 4H), 5.02 (d, 2H, J¼12 Hz), 4.47
(dd, 2H, J¼20, 8 Hz), 4.06e3.99 (m, 6H), 3.77e3.73 (m, 2H),
3.66e3.60 (m, 2H), 3.45e3.38 (m, 2H), 2.80e2.73 (m, 2H),
2.36e2.31 (m, 2H), 2.15e2.06 (m, 4H), 1.91e1.84 (m, 2H), 1.53e1.45
4.2.1.
butyl ester (300 mg, 2.07 mmol) was suspended in dichloro-
methane (4 mL) and treated with benzophenone imine (381 L,
L
-N-(diphenyl)alanine tert-butyl ester (2a). L-Alanine tert-
(m, 2H) ppm; 13C NMR (100 MHz, DMSO-d6)
d 150.8, 148.5, 141.7,
m
138.4, 136.0, 134.7, 134.4, 131.7, 130.2, 128.9, 128.0, 126.4, 125.5,
124.1, 121.3, 120.1, 118.1, 117.1, 69.7, 68.4, 63.5, 59.3, 51.3, 26.4, 24.7,
21.2 ppm; HRMS (ESI) calculated C52H60N4O2: 386.7379 ([MþH]2þ),
found 386.2387 ([MþH]2þ).
1.82 mmol). The reaction was stirred at room temperature over-
night. The precipitate was filtered and the filtrate evaporated to
dryness. The crude product was purified using flash chromatog-
raphy (SepacoreÒ flash system) with 4% ethyl acetate in hexane.
The collected fraction was evaporated to give
L-N-(diphenyl)ala-
5. Radiochemistry
nine tert-butyl ester as a white solid (Yield: 472 mg, 1.60 mmol,
77%).
5.1. Radiochemical procedure for the production of [11C]MeI
1H NMR (250 MHz, CDCl3)
d 7.58 (m, 2H), 7.41e7.36 (m, 3H),
7.30e7.24 (m, 3H), 7.14 (m, 2H), 3.98 (q, 1H, J¼8 Hz), 1.37 (s, 9H),
Cyclotron produced [11C]CO2 was carried in a stream of helium
and trapped in 0.1 mL of a 0.1 M lithium aluminium hydride solu-
tion in THF in a glass reaction vessel at room temperature. After
trapping, the gas flow was increased to 20 mL/min and the THF
evaporated at 130 ꢁC. After evaporation to dryness 0.2 mL of 56%
hydriodic acid was added and the [11C]MeI was distilled from the
reaction vessel under a stream of helium (flow 20 mL/min) to the
second reaction vessel for the alkylation reaction.40
1.35 (d, 3H, J¼8 Hz) ppm; 13C NMR (100 MHz, CDCl3)
d 172.1, 169.3,
139.8, 136.6, 132.4, 130.4, 128.7, 80.7, 61.3, 28.1, 19.2 ppm; HRMS
(ESI) calculated C20H23NO2: 310.1807 ([MþH]þ), found 310.1797
([MþH]þ).
4.2.2.
D
-N-(diphenyl)alanine tert-butyl ester (2b). Compound 2b
was synthesized analogous to the method used for the synthesis of
compound 2a. 2b was obtained in a yield of 365 mg (1.24 mmol,
60%) as colorless oil.
5.2. Alkylation procedure of Schiff base 1
1H NMR (400 MHz, CDCl3)
d 7.58 (m, 2H), 7.41e7.36 (m, 3H),
7.30e7.24 (m, 3H), 7.14 (m, 2H), 3.99 (q, 1H, J¼8 Hz), 1.37 (s, 9H),
[
11C]MeI is distilled into a closed reaction vessel containing the
1.34 (d, 3H, J¼8 Hz) ppm; 13C NMR (100 MHz, CDCl3)
d
172.1, 169.3,
Schiff base 1, phase transfer catalyst and base, in a mixture of tol-
uene/dichloromethane (19/1, v/v). The color changed instantly to
yellow, [11C]MeI was trapped in the second reaction vessel prior to
heating or cooling of the reaction mixture. At set timepoints,
samples were taken from the reaction mixture for analysis by
radioHPLC to determine the alkylation conversion rates. For
deprotection, 0.1 mL of 6 M HCl was added to the reaction mixture
prior to heating to 100 ꢁC for 2 min. After cooling to room tem-
perature, a sample was taken for analysis on chiral radioHPLC to
139.8, 136.6, 132.4, 130.4, 128.7, 80.7, 61.3, 28.1, 19.2 ppm; HRMS
(ESI) calculated C20H23NO2: 310.1807 ([MþH]þ), found 310.1812
([MþH]þ).
4.2.3.
a,a
0-Biscinchonidinium-m-xylene dibromide (pre-PTC 9*). The
synthesis was performed according to described literature pro-
cedures.37 (ꢃ)-Cinchonidine (2.00 g, 6.79 mmol) and dibromo-m-
xylene (0.88 g, 3.33 mmol) were dissolved in EtOH (5 mL), DMF
(6 mL) and chloroform (2 mL) and refluxed at 100 ꢁC for 4 h. When
all dibromo-m-xylene was consumed according to TLC, the reaction
mixture was cooled to room temperature, diluted with MeOH and
precipitated in cold diethyl ether. The desired product was obtained
in a yield of 72% as a pink solid (3.38 g, 4.88 mmol).
determine the enantiomeric excess of L
-[11C]alanine.
5.3. Optimized procedure for the alkylation with [11C]MeI to
obtain
-[11C]alanine
L
1H NMR (400 MHz, DMSO-d6)
d
9.02 (d, 2H, J¼2.5 Hz), 8.37 (d,
In a reaction vessel, Schiff base 1 (1.89 mg, 7
(0.5 mg, 0.8
mol) and CsOHꢀH2O (30 mg, 200
pended in a mixture of toluene/dichloromethane (300
m
mol), catalyst 4
mol) are sus-
L, 19:1, v/v)
2H, J¼10 Hz), 8.15 (d, 3H, J¼7.5 Hz), 7.94 (d, 2H, J¼5 Hz), 7.91e7.84
(m, 4H), 7.81e7.75 (m, 3H), 6.79 (d, 2H, J¼2.5 Hz), 6.61 (s, 2H),
5.79e5.65 (m, 2H), 5.32 (d, 2H, J¼12.5 Hz), 5.22e5.15 (m, 4H), 5.00
(d, 2H, J¼10 Hz), 4.38e4.30 (m, 2H), 4.01e3.94 (m, 2H), 3.83e3.78
(m, 2H), 3.60e3.51 (m, 2H), 3.18 (d, 2H, J¼5 Hz), 2.20e2.10 (m,
3H), 2.07e2.03 (m, 3H), 1.87e1.83 (m, 2H), 1.37e1.30 (m, 2H) ppm;
m
m
m
and the color changes instantly to yellow. After distilling [11C]MeI in
the reaction vial, the mixture is cooled to 5 ꢁC and stirred for 5 min.
A sample is taken for analysis on Grace Smart RP18 column (ace-
tonitrile/sodiumformate 4 mMþ4% DMF 70/30, v/v) with a re-
tention time (tR) of 8.3 min for product 2. The deprotection is
initialized by the addition of 0.1 mL of 6 M HCl solution and heating
to 100 ꢁC for 1.5 min. A second sample is taken for analysis on chiral
13C NMR (100 MHz, DMSO-d6)
d 155.5, 155.4, 152.8, 150.4, 144.1,
143.6, 140.5, 135.1, 134.7, 133.8, 132.6, 129.5, 128.9, 125.4, 121.6,
72.9, 69.4, 67.5, 64.4, 55.8, 42.1, 31.0, 29.4, 26.3 ppm; HRMS (ESI)
calculated
C
46H52N4O2: 346.7084 ([MþH]2þ), found 346.2017
radioHPLC to determine the enantiomeric excess of
L
-[11C]alanine
([MþH]2þ).
with HPLC column Reprosil chiral-aa (methanol/water 70/30, v/v)
with a tR of 4.5 min for
alanine.
L
-[11C]alanine and 10.3 min for -[11C]
D
4.2.4.
9).
a
,a
0-Bis[O(9)-allylcinchonidinium]-m-xylene dibromide (PTC
0-Biscinchonidinium-m-xylene dibromide 9* (300 mg,
a
,
a
0.43 mmol) was suspended in dichloromethane (5 mL) and allyl
bromide (1 mL, 11.56 mmol) and 50% aqueous KOH (2 mL,
17.60 mmol) were added at room temperature. The suspension was
stirred vigorously for 3 h until the reactants were consumed
according to TLC. The mixture was diluted with water and extracted
with dichloromethane. The combined organic extracts were evap-
orated to yield 305 mg of 9 as orange solid (0.39 mmol, 91%).
Acknowledgements
This study received funding from the European Union, Marie
Curie actions RADIOMI (FP7-PEOPLE-2012-ITN) under project ref-
erence no. 316882. We gratefully acknowledge the Cyclotron BV
that provides us with [11C]CO2 on a daily basis.
1H NMR (400 MHz, DMSO-d6)
d
9.05 (d, 2H, J¼4 Hz), 8.32 (d, 2H,
Supplementary data
J¼8 Hz), 8.17 (d, 2H, J¼8 Hz), 8.08 (s, 1H), 7.96e7.94 (m, 2H),
7.91e7.87 (m, 2H), 7.82 (t, 3H, J¼8 Hz), 7.73 (d, 2H, J¼4 Hz), 6.51 (s,
2H), 6.23e6.13 (m, 2H), 5.79e5.70 (m, 2H), 5.51 (d, 2H, J¼20 Hz),
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tet.2016.08.069.

U. Filp et al. / Tetrahedron 72 (2016) 6551e6557
6557
20. Hashimoto, T.; Maruoka, K. Chem. Rev. 2007, 107, 5656.
21. Maruoka, K.; Tayama, E.; Ooi, T. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5824.
22. Langstrom, B.; Stridsberg, B. Int. J. Appl. Radiat. Isot. 1978, 30, 151.
References and notes
ꢁ
€
1. Lundqvist, H.; Lubberink, M.; Tolmachev, V. Eur. J. Phys. 1998, 19, 537.
2. Jager, P. L.; Vaalburg, W.; Pruim, J.; de Vries, E. G.; Langen, K. J.; Piers, D. a J. Nucl.
Med. 2001, 42, 432.
3. Fani, M.; Maecke, H. R. Eur. J. Nucl. Med. Mol. Imaging 2012, 39, S11.
4. Huang, C.; McConathy, J. J. Nucl. Med. 2013, 54, 1007.
5. Lee, T. S.; Ahn, S. H.; Lim, S. M. Nucl. Med. Biol. 2009, 36, 681.
6. Wu, G. Amino Acids 2009, 37, 1.
7. Vaalburg, W.; Coenen, H. H.; Crouzel, C.; Elsinga, P. H.; Langstrom, B.; Lemaire,
C.; Meyer, G. J. Int. J. Rad. Appl. Instrum. B 1992, 19, 227.
8. Fuchs, B. C.; Bode, B. P. Semin. Cancer Biol. 2005, 15, 254.
9. Glaudemans, A. W. J. M.; Enting, R. H.; Heesters, M. a a M.; Dierckx, R. a. J. O.;
Van Rheenen, R. W. J.; Walenkamp, A. M. E.; Slart, R. H. J. a Eur. J. Nucl. Med. Mol.
Imaging 2013, 40, 615.
10. Langen, K. J.; Hamacher, K.; Weckesser, M.; Floeth, F.; Stoffels, G.; Bauer, D.;
Coenen, H. H.; Pauleit, D. Nucl. Med. Biol. 2006, 33, 287.
11. Qu, W.; Oya, S.; Lieberman, B. P.; Ploessl, K.; Wang, L.; Wise, D. R.; Divgi, C. R.;
Chodosh, L. P.; Thompson, C. B.; Kung, H. F. J. Nucl. Med. 2012, 53, 98.
12. Wu, Z.; Zha, Z.; Li, G.; Lieberman, B. P.; Choi, S. R.; Plossl, K.; Kung, H. F. Mol.
ꢁ
€
23. Antoni, G.; Langstrom, B. Acta Chem. Scand. 1986, 40, 152.
ꢁ
€
24. Fasth, K.; Anonti, G.; Langstrom, B. Acta Chem. Scand. 1990, 44, 527.
25. Belokon, Y. N.; Bulychev, A. G.; Vitt, S. V.; Struchkov, Y. T. J. Am. Chem. Soc. 1985,
107, 4252.
ꢁ
€
26. Fasth, K.; Langstrom, B. Acta Chem. Scand. 1990, 44, 720.
27. Popkov, A.; Elsinga, P. H. Curr. Org. Chem. 2013, 17, 2127.
28. Kato, K.; Tsuji, A. B.; Saga, T.; Zhang, M.-R. Bioorg. Med. Chem. Lett. 2011, 21,
2437.
ꢁ
€
29. Suzuki, C.; Kato, K.; Tsuji, A. B.; Kikuchi, T.; Zhang, M.-R.; Arano, Y.; Saga, T.
Bioorg. Med. Chem. Lett. 2013, 23, 4567.
30. Shirakawa, S.; Yamamoto, K.; Kitamura, M.; Ooi, T.; Maruoka, K. Angew. Chem.,
Int. Ed. Engl. 2005, 44, 625.
31. Ooi, T.; Uematsu, Y.; Kameda, M.; Maruoka, K. Tetrahedron 2006, 62, 11425.
32. Kitamura, M.; Shirakawa, S.; Maruoka, K. Angew. Chem., Int. Ed. Engl. 2005, 44,
1549.
33. Park, H.; Jeong, B.; Yoo, S.; Lee, J.; Park, M.; Lee, J.; Kim, M.; Jew, S. Angew. Chem.,
Int. Ed. Engl. 2002, 6, 3036.
34. Donnell, M. J. O.; Wip, S.; Huffman, J. C. Tetrahedron 1994, 50, 4507.
35. Lygo, B.; Crosby, J.; Lowdon, T. R.; Peterson, J. A.; Wainwright, P. G. Tetrahedron
2001, 57, 2403.
36. Lygo, B.; Andrews, B. I. Acc. Chem. Res. 2004, 37, 518.
37. Jew, S.; Jeong, B.; Yoo, M.; Huh, H.; Park, H. Chem. Commun. 2001, 1244.
38. Cook, T. C.; Andrus, M. B.; Ess, D. H. Org. Lett. 2012, 14, 5836.
39. Petrova, G. P.; Li, H. B.; Maruoka, K.; Morokuma, K. J. Phys. Chem. B 2014, 118,
5154.
€
Pharmacol. 2014, 11, 3852.
13. Wang, L.; Zha, Z.; Qu, W.; Qiao, H.; Lieberman, B. P.; Plossl, K.; Kung, H. F. Nucl.
Med. Biol. 2012, 39, 933.
14. Torstenson, R.; Hartvig, P.; Langstrom, B.; Westerberg, G.; Tedroff, J. Ann. Neurol.
€
ꢁ
€
1997, 41, 334.
ꢁ
€
15. Torstenson, R.; Tedroff, J.; Langstrom, B. J. Cereb. Blood Flow. Metab. 1999, 19,
1142.
16. Scott, P. J. H. Angew. Chem., Int. Ed. Engl. 2009, 48, 6001.
17. Ooi, T.; Maruoka, K. Angew. Chem., Int. Ed. Engl. 2007, 46, 4222.
18. Donnell, M. J. O.; Boniece, J. M.; Earp, S. E. Tetrahedron Lett. 1978, 2641.
19. Corey, E. J.; Xu, F.; Noe, M. C. J. Am. Chem. Soc. 1997, 119, 12414.
ꢁ
€
ꢁ
40. Langstrom, B.; Antoni, G.; Gullberg, P.; Halldin, C.; Malmborg, P.; Nagren, K.;
€
~
Rimland, A.; Svard, H.; Langstra, B.; Malmborg, F. J. Nucl. Med. 1987, 1037.