Synthesis of the positron-emitting radiotracer [18F]-2-fluoro-2- deoxy-d-glucose from resin-bound perfluoroalkylsulfonates

Article abstract of DOI:10.1039/b816032e

A new approach to the synthesis of 2-fluoro-2-deoxy-d-glucose (FDG, [ ^19/18F]-3) is described, which employs supported perfluoroalkylsulfonate precursors 33-36, where the support consists of insoluble polystyrene resin beads. Treatment of these resins with [ ¹⁹F]fluoride ion afforded protected FDG [¹⁹F]-18 as the major product, and the identities of the main byproducts were determined. Acidic removal of the acetal protecting groups from [¹⁹F]-18 was shown to produce [¹⁹F]FDG. The method has been applied to the efficient radiosynthesis of the imaging agent [¹⁸F]FDG, and was shown to produce the radiochemical tracer in good radiochemical yield (average 73%, decay corrected). The Royal Society of Chemistry 2009.

Full text of DOI:10.1039/b816032e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis of the positron-emitting radiotracer [18F]-2-fluoro-2-deoxy-D-
glucose from resin-bound perfluoroalkylsulfonates†
Lynda J. Brown,a Nianchun Ma,a Denis R. Bouvet,b Sue Champion,b Alex M. Gibson,b Yulai Hu,a
Alex Jackson,b Imtiaz Khan,b Nicolas Millot,b Amy C. Topley,a Harry Wadsworth,b Duncan Wynnb and
Richard C. D. Brown*a
Received 12th September 2008, Accepted 29th October 2008
First published as an Advance Article on the web 8th December 2008
DOI: 10.1039/b816032e
A new approach to the synthesis of 2-fluoro-2-deoxy-D-glucose (FDG, [19/18F]-3) is described, which
employs supported perfluoroalkylsulfonate precursors 33–36, where the support consists of insoluble
polystyrene resin beads. Treatment of these resins with [19F]fluoride ion afforded protected FDG
[19F]-18 as the major product, and the identities of the main byproducts were determined. Acidic
removal of the acetal protecting groups from [19F]-18 was shown to produce [19F]FDG. The method has
been applied to the efficient radiosynthesis of the imaging agent [18F]FDG, and was shown to produce
the radiochemical tracer in good radiochemical yield (average 73%, decay corrected).
FDG precursor 2 thus obtained, under basic or acidic conditions
Introduction
produces [18F]FDG.7–13 Despite the very low chemical yields
Positron-emission-tomography (PET) is a powerful imaging tool,1
which can be exploited clinically to locate and assess abnormalities
observed for the nucleophilic fluoridation reactions of triflate 1
(less than 5% based on 1 in the stoichiometric reaction), the use
in oncology,2 neurology,3 and cardiology.4 Considerable additional
interest in PET and microPET also emanates from applica-
tions of these techniques in drug discovery and development.5
[18F]2-Fluoro-2-deoxy-D-glucose ([18F]FDG, [18F]-3) is the most
widely used radiochemical tracer in PET applications possessing
the positron-emitting radionuclide 18F (half-life ~ 110 min).6,7
[18F]FDG is used to measure glucose uptake into tissue, giving rise
to real-time images that can assist in the diagnosis, management
and study of diseases such as cancer. [18F]FDG was originally
synthesised in the late 1970’s by electrophilic fluorination of glucal
with 18F2 in relatively low yields and limited stereoselectivity.8
However, this early approach is not currently considered to be
suitable for routine radiosynthesis and subsequently, various im-
proved synthetic routes and modifications have been reported,7,9–13
which are based on nucleophilic fluorination using [18F]fluoride
ion.12
of massive excesses of triflate 1 and limiting [18F]fluoride ion in
the radiolabeling step ensures that radiochemical yields typically
in excess of 60% are realised.
The most important method is the one conventionally used
in PET centres throughout the world, which involves the re-
action of a large excess (100 000 fold) of tetra-O-aceyl-2-O-
trifluoromethanesulfonyl D-mannose (1) with [18F]fluoride ion
in the presence of the phase-transfer agent, Kryptofix [2.2.2]
(4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo[8.8.8]hexacosane, 4)
(Scheme 1).11,12 Cleavage of the acetate protecting groups from the
Scheme 1 Synthesis of [18F]FDG ([18F]-3) used in PET centres.
Reverse phase HPLC purification is impractical for commercial
applications due to the short half-life of 18F, which requires the
radiotracer to be synthesised and purified as rapidly as possible,
and ideally within one hour for clinical use. The methods routinely
used for the radiosynthesis of [18F]-FDG have the disadvantage
that the protected FDG is produced as a mixture, with a large
stoichiometric excess of 1 and other degradation products arising
from side reactions of the starting material during fluoridation.
This mixture, containing the excess triflate and any other by-
products, is submitted to the deprotection conditions to give the
desired [18F]FDG, D-glucose and a variety of poorly characterised
products.14 The presence of the by-products is not considered to be
detrimental to the patient or the efficacy of the injected product.
However, we considered it advantageous to develop a method
of synthesis that produced FDG containing reduced amounts of
aSchool of Chemistry, University of Southampton, Highfield, Southampton,
UK SO17 1BJ. E-mail: rcb1@soton.ac.uk; Fax: +44 (0)23 80593781;
Tel: +44 (0)23 80594108
bGE Healthcare, The Grove Centre, White Lion Road, Amersham, Bucks,
UK HP7 9LL
† Electronic supplementary information (ESI) available: General experi-
mental procedures; preparations and characterization data for 8, 9, 11;
copies of 1H spectra for 8, 9, 12, 14, 16, 17, 18, 25, 26, 27, 28, 29, 30,
31, 32, 38, crude [19F]fluoridation reaction mixtures from 1 and resin 35;
copies of 19F NMR spectra for 33, 34, 35, 36 and crude [19F]fluoridation
reaction mixture from resin 35; MAS 1H and 13C NMR spectra for 35;
HPLC chromatograms of [18F]-18, [18F]-2. See DOI: 10.1039/b816032e
564 | Org. Biomol. Chem., 2009, 7, 564–575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
chemical impurity, which could also be applied to the production
of other radiotracers containing 18F. In this paper we describe an
efficient process for the synthesis of FDG from novel resin-bound
perfluoroalkylsulfonate precursors. We also report our studies
on the purity profile of the protected product cleaved from the
resin using [19F]fluoride ion. In addition, we demonstrate that the
methodology is suitable for radiosynthesis of [18F]FDG with high
radiochemical yield and improved chemical purity.15
Results and discussion
Scheme 2 Synthesis of [19F]FDG using stoichiometric fluoride. Reagents
and conditions: (i) Tf2O, pyridine, CH2Cl2, -15 ◦C to rt; (ii) K19F, 4, MeCN,
80 ◦C, 10 min.
The objective of our research was to develop a platform technology
based on solid-supported precursors that would allow the pro-
duction of 18F-labeled compounds, namely [18F]FDG, by means
of a nucleophilic cleavage reaction using [18F]fluoride ion.15,16
Solid-phase chemistry is readily amenable to automation and
is used extensively for the efficient synthesis of oligopeptides,
oligonucleotides, and other organic molecules on automated
synthesisers.17 Preparation of labeled radiopharmaceuticals by
nucleophilic fluoridation-cleavage of precursors attached to a
solid-phase would be advantageous in terms of the purity of
the cleaved fluorinated products, whilst facilitating automation of
the process and thus offering further protection from radiation
exposure and increased reproducibility.18 The new technology
would produce 18F-labeled tracers quickly and with high specific
activity yet minimise time-consuming purification steps.
The targeted strategy involved immobilisation of FDG pre-
cursor on a solid-support through an highly activated sulfonate
leaving group linker that would allow specific cleavage of the
radiotracer into solution using [18F]fluoride ion.19 In principle,
unreacted precursor should remain attached to the resin permit-
ting its separation from the product by means of a simple filtration
process, thus avoiding the presence of the excess reactive triflate in
the deprotection step. Following removal of the protecting groups,
the 18F-containing tracer could be passed through disposable
cartridges to remove Kryptofix [2.2.2] (4) and any residual
[18F]fluoride ion, leaving pure product ready for administration.
Preliminary studies concentrated on direct translation of the
conventional FDG solution-phase chemistry onto the solid-phase
to produce a supported analogue of the triflate 1. However,
progress along these lines was blocked due to limitations imposed
by the presence of the base-sensitive acetyl protecting groups. In-
vestigation of the solution reaction of triflate 1 with stoichiometric
[19F]fluoride yielded the desired fluoro-sugar [19F]-2 in less than 5%
yield along with elimination product 7, b-D-glucose pentaacetate
(6) and some unreacted triflate 1 (Scheme 2).20 This problem can
be overcome in the radiosynthesis by simply employing a very
large excess of the triflate 1, although it would be highly desirable
to start from a more chemically efficient fluoridation. However,
more serious problems were associated with the preparation of
nonaflate 8, which was chosen as a representative model for
more functionalised longer chain perfluoroalkylsulfonate esters
(Scheme 3). Even though the triflate 1 can be prepared in
good yields (90%) using triflic anhydride,21 reaction of 4 with
nonaflic anhydride led mainly to decomposition of the sugar with
significantly reduced yields of the desired nonaflate 8 (36%).22,23
This result was not encouraging with respect to the synthesis of
more complex sugar–perfluoroalkylsulfonates.
Scheme 3 Reactions of b-D-mannose tetraacetate 5 with nonafluorobu-
tanesulfonyl chloride and nonafluorobutanesulfonic anhydride. Reagents
and conditions: (i) (C4F9SO2)2O, pyridine, CH2Cl2; (ii) C4F9SO2Cl, pyri-
dine, AgOTf, THF.
It was clear that challenges associated with preparation of a
highly reactive perfluoroalkylsulfonic anhydride linker that con-
tained appropriate functionality to ultimately allow attachment
to a solid-support were considerable, causing us to turn our
attention to the use of less reactive perfluoroalklysulfonyl halides
as sulfonate ester precursors. Accordingly, attempts were made to
synthesise the model perfluoroalkylsulfonate 8 from the nonaflic
fluoride and chloride in the presence of different bases.22 The
best result was a very disappointing 7% yield of nonaflate 8
when the reaction was carried out in CH2Cl2 in the presence of
DMAP. Furthermore, most of the reactions conducted under basic
conditions were complicated by migration of the acetyl group
from the anomeric position to the C2 hydroxyl, which once again
highlighted the limitations of the ester protecting group strategy.
In an effort to enhance the reactivity of the sulfonyl chloride
reagents, and to suppress any further reactions of the desired
nonaflate 8 with liberated halide ion, AgOTf was added to the
sulfonylation reaction. Surprisingly, the only isolated product was
perfluorobutanoate 9 (Scheme 3).
The low reactivity of the perfluorosulfonyl halide reagents
towards the alcohol hindered secondary alcohol nucleophile led
us to reconsider the perfluorosulfonate formation conditions. It
was believed that the sodium salt of a suitably protected mannose
derivative would react more effectively with a perfluoroalkylsul-
fonyl halide to give the desired sulfonyl ester.22,24 This would
require a new protecting group strategy in order to accomplish
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 564–575 | 565
©

View Article Online
our overall aims. The protection chosen had to be stable to
the conditions used to prepare a sugar sulfonate linkage, which
would require strong base to deprotonate the secondary alcohol
at the 2-position of the sugar. It would also be desirable if the
revised protected mannose perfluoroalkylsulfonate could undergo
a more efficient fluoridation than that of the tetraacetate 1, which
is a poor substrate. Finally, it was imperative that all of the
protecting groups could be removed rapidly in one simple and
efficient step that would cause negligible decomposition of FDG.
Fortunately, literature precedent suggested that the use of a 4,6-
benzylidene acetal in combination with acid-labile protection at
the 1- and 3-positions might satisfy all of the above demands.12c,25
The ethoxymethyl group was selected for protection of the C3
alcohol as it can be removed under similar conditions used to de-
block the benzylidene and anomeric methoxy protecting groups.
Our synthetic approach began with the commercially available
D-glucose derivative, methyl b-D-glucopyranoside (10), which
was converted in near quantitative yield to its crystalline 4,6-
benzylidene acetal 11.26 Reaction of the diol 11 with EOM-Cl
gave a mixture of products, from which the desired C2-carbinol
13 was isolated in satisfactory yield (Scheme 4). Inversion of
stereochemistry at the 2-position was achieved by means of a
two-step sequence involving oxidation of 13 and subsequent
stereoselective reduction of the resulting C2-carbonyl function
using sodium borohydride to give the protected D-mannose
derivative 15 in good yield.25 The use of the b-anomer was required
in order to direct the approach of the hydride reagent to the a-
face of the ketone (isolated d.r. ~9 : 1), and to ultimately facilitate
nucleophilic attack of fluoride ion at C2, which would also occur
from the a-face of the sugar.
of alcohol 15 with NaH, reacted smoothly with nonafluorobu-
tanesulfonyl fluoride in THF to give the desired nonaflate 16
(Scheme 5). Furthermore, the subsequent reaction of 16 with
Kryptofix [2.2.2] and KF in CH3CN afforded protected FDG
18 in a substantialy improved 53% yield along with a smaller
quantity of an eliminated byproduct 17. It is significant that similar
preparative scale fluoridolysis carried out on the triflate 5 gave
an amount of protected [19F]FDG that was barely observable
1
in the H NMR spectrum of the crude reaction mixture,20 thus
indicating that 1 was not only less reactive than 16 towards
nucleophilic displacement but also significantly more prone to
give rise to byproducts. Heating 18 at reflux in 6 N HCl for
10 min resulted in removal of the protecting groups to afford
unlabeled FDG [19F]-3. The mixture of isomeric fluorosugars
was peracetylated to give the a- and b-anomers of 2-deoxy-2-
fluoro-b-D-glucopyranoside tetraacetate, which were separated
and fully characterised as single compounds.27 The structures
were confirmed by comparison against authentic samples prepared
from a commercial [19F]FDG standard, under identical conditions.
A more detailed optimization of the conditions for deprotection of
18 was postponed until the radiochemistry studies. These studies
clearly showed that the acetal protected substrate 16 was a superior
precursor to FDG in terms of the chemical synthesis.
Scheme 5 Synthesis of [19F]FDG from a model for the solid-supported
perfluoroalkylsulfonate. Reagents and conditions: (i) C4F9SO2F, NaH,
THF, rt; (ii) 4, KF, MeCN, reflux, 10 min; (iii) 6 M HCl, reflux, 10 min;
(iv) Ac2O, pyridine.
Now that the viability of the overall approach had been
demonstrated using the nonaflate 16 as a model for a perfluo-
roalkylsulfonate linker, we were able to turn our attention to the
synthesis of a sugar-linker construct. Recent publications have
described a perfluoroalkylsulfonyl fluoride linker, which when
attached to a resin allows coupling of phenols and subsequent
release of a range of substituted arenes, biaryls and biaryl amides
into solution (Scheme 6).28
Scheme 4 Synthesis of the FDG precursor 15 with base-stable protecting
groups. Reagents and conditions: (i) benzaldehyde dimethylacetal, CSA,
MeCN, Et3N, rt; (ii) chloromethylethyl ether, NaH, THF, rt; (iii) DMSO,
Ac2O, rt; (iv) NaBH4, THF, CH3OH, rt.
Following the protecting group switch, we were delighted to
discover that the sodium alcoholate, obtained from treatment
566 | Org. Biomol. Chem., 2009, 7, 564–575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
oxapentanesulfonyl fluoride. Initial formation of the sulfonate
ester gave poor repeatability and yields using NaH as the base, but
the results were dramatically improved by using NaHMDS to pre-
form the alkoxide. With the iodide 25 in hand, radical-mediated
coupling reactions were carried out with a series of enoic acids
of varying chain lengths.29 The lower yield observed using the
l6-heptadecenoic acid was due to the poor solubility of this olefin
in the reaction solvents. Reaction of acrylic acid with iodide 25
provided the reduced product 29 directly, whereas transfer of the
iodine atom was observed in reactions of the longer alkenoic acids.
The resulting iodoacids 26–28 were easily de-iodinated by refluxing
in Et2O–AcOH in the presence of zinc powder.30 Gratifyingly, the
protecting groups on the sugar and the perfluoroalkylsulfonate
ester survived these acidic conditions by careful control of reaction
time and temperature.
Scheme
6
Previously reported syntheses of perfluoroalkylsulfonate
linkers.28 Reagents and conditions: (i) (a) (CO)2Cl2, CH2Cl2, DMF;
(b) amine resin, CH2Cl2, i-Pr2NEt; (ii) ArOH, K2CO3, DMF.
However, this linker proved unsatisfactory for our application
due to a number of significant shortcomings. Firstly, the yields
obtained for the coupling of mannose derivative to the linker
in solution or after attachment to the resin were poor, and we
observed elimination of [19F]fluoride ion from the linker under
basic conditions. In addition, when the coupling of the sugar is
carried out on a resin-bound sulfonyl fluoride, any resin-bound
sulfonyl [19F]fluoride 21 unreacted in the coupling could exchange
with [18F]fluoride ion in the subsequent labeling studies, thereby
producing unlabeled FDG and lowering the radiochemical yield
and purity of the product. Potential for loss of [19F]fluoride ion
from the linker was a major concern because limiting [18F]fluoride
ion (<10 mmol) is used in the synthesis of [18F]FDG. It was
therefore decided that a different linker was required, which
would be less prone to elimination and that would be suitable
for the formation of a sugar–linker conjugate in solution ready for
coupling with the resin.
The strategy that ultimately proved to be successful involved
preparation of the linker–sugar conjugate in solution and sub-
sequent attachment to an amino resin (Scheme 7). Importantly,
this approach should minimise the potential for release of
[19F]fluoride ion from the linker, either due to E1cb elimination
of H19F or through exchange with residual unreacted SO2F
groups present within the resin. A variety of different linker–
sugar conjugates were prepared (29–32) by firstly coupling the
alkoxide of the mannose derivative 15 to 5-iodooctafluoro-3-
A range of different coupling agents and conditions were
investigated for the attachment of acids 29–32 to aminomethyl
polystyrene (–NH2 = 1.5 mmol g-1), and it was found that the
resins 33–36 could be prepared in high yields (estimated from
masses of dry resins) using diphenylphosphoryl chloride as the
source of activation. Successful formation of the desired supported
sulfonate esters was confirmed by on-bead IR, elemental analysis,
1
19F NMR, MAS H and 13C NMR. With a view to reducing the
ultimate cost of the resin for commercial application, resins 35
with lower loading levels were also synthesized by using limiting
amounts of the acid 31. Any unreacted aminomethyl groups were
capped by treatment with Ac2O and pyridine. The synthesis of
one resin 35 (0.1 mmol g-1 by sulfur analysis) was conducted on a
multi-kilogram scale.
Fluoridation experiments using [19F]fluoride ion were carried
out in order to demonstrate that protected [19F]FDG could be
released from the resin (Scheme 8). Replicating the radiochemical
fluoridation conditions on a preparative scale is difficult to achieve
due to the large ratio of FDG precursor to fluoride ion that is typi-
cally used in the radiosynthesis, so one would expect a significantly
different product distribution. Nonetheless, these experiments
would provide insight into the identity of any potential byproducts
Scheme 7 Preparation of the solid-supported FDG precursors 33–36. Reagents and conditions: (i) ICF2CF2OCF2CF2SO2F, NaHMDS (1.0 M in THF),
THF, rt; (ii) CH2=CH(CH2)nCOOH where n = 2, 8 or 14; NaHCO3, Na2S2O4, MeCN, H2O, rt; (iii) Zn, AcOH, Et2O, reflux; (iv) CH2=CHCOOH;
NaHCO3, Na2S2O4, MeCN, H2O, rt; (v) amino-methylated polystyrene, (PhO)2POCl, i-Pr2NEt, CH2Cl2, rt.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 564–575 | 567
©

View Article Online
resins should be modified to avoid the use of acetate capping. It
is noteworthy that b-D-glucose pentaacetate (6) is a major side
product from the fluoridation of the mannose triflate 1, but in this
case, the acetate comes from degradation of the starting material.
Fluoridolysis using limiting fluoride proceeded similarly to
give 17 and 18 as the major products, and the recovered resin
was shown to release further protected [19F]FDG for up to four
cycles. These conditions more closely resembled the proposed
radiochemistry and clearly demonstrated that unreacted sulfonate
ester remained attached to the solid-phase. Collectively, the above
experiments showed that we had successfully developed a resin
that was capable of releasing a protected FDG derivative 18
when exposed to [19F]fluoride ion, and that the crude product
was readily separated from any unreacted starting material and
the reagents. Furthermore, the crude protected FDG prepared
in this way was of vastly superior chemical purity compared to
the corresponding crude tetraacetate [19F]-2 obtained using the
conventional fluoridation procedure (see Scheme 2).20 Finally, any
concerns relating to the efficiency of nucleophilic cleavage due to
poor swelling of the polystyrene resin in MeCN were allayed.
Scheme 8 [19F]Fluoridation studies using resin 35.
Radiochemical results
that could be generated during the radiochemistry. Fluoridolysis
of the supported D-mannose derivative 35 (0.1 mmol g-1) was
explored by using excess [19F]fluoride ion (10 equiv) in the first
instance, in order to cleave all of the material from the solid-
phase. The two expected major products 17 and 18 and two
lesser ones 37 and 38 were released from the resin in ratios of
31 : 80 : 1 : 13 (17 : 18 : 37 : 38; estimated from 1H NMR and HPLC
analysis of the crude product). The NMR (1H and 19F) spectra also
showed the presence of a large quantity of Kryptofix [2.2.2] (4) and
complexed K19F in the crude product. During the radiosynthesis of
[18F]FDG, these contaminants are removed by sequential elution
through disposable cartridges (e.g. C18, alumina or ion exchange).
For our purposes, passage through a short plug of silica gel
served to remove the excess reagents, leaving only three significant
components (17, 18 and 38) observable by 1H NMR and only one
fluorinated product 18 (19F NMR: d = -37.30 ppm, ref. C6F6).
Complete cleavage of the sugar from the resin was confirmed by
the loss of the 19F NMR signal at -114 ppm (referenced to CFCl3)
due to the diastereotopic CF2SO3R fluorines, and the presence of
an additional signal at -118 ppm due to the new CF2SO3K group.
Formation of eliminated byproduct 17 was expected from the
fluoridation studies carried out on nonaflate 16, and the minor
byproduct 37 arises from further degradation of 17. However,
the observation of a new substitution product 38 resulting from
acetate displacement of the sulfonate linker was not expected from
earlier solution studies. The formation of the glucose derivative
38 could either be due to acetate present due to the capping
step or from degradation of the acetonitrile solvent. In order to
establish the origin of the acetate group, the reaction mixture from
the fluoridation of nonaflate 16 was carefully reevaluated with
the confirmation that acetate 38 was not formed in the solution
reaction. This result strongly suggested that the acetate was coming
from the capping step performed during the synthesis of the
lower loading resin. The protected glucose derivative 38 is not
considered to be harmful due to its conversion to D-glucose upon
deprotection. However, these findings may be significant for the
fluoridation of other substrates and the synthesis of lower loading
The excellent results obtained from the [19F]fluoridolysis using
the resin-bound D-mannose derivative 35 set the stage for the
radiochemistry studies. Low activity labeling experiments were
carried out manually using K18F (185–350 MBq) and Kryptofix
[2.2.2] (4) in CH3CN, and radiochemical yields were established
by reversed-phase HPLC analysis of the protected product
[18F]-18 using UV and g-detection (Scheme 9). Conventional
solution-phase 18F labeling typically results in 80–90% incorpo-
◦
ration after 2 minutes at 86 C in 1–2 mL reaction volumes.9
In the present study involving heterogeneous reaction mixtures
containing the solid-supported precursors 33–36, advantage was
gained from reducing the reaction volume to 0.2 mL to increase
the concentration of the [18F]fluoride ion solution. Using resin 35,
labeling times of 10–15 minutes were initially used to ensure maxi-
mum incorporation of 18F into [18F]-18. However, it transpired that
even after 3–4 minutes, levels of [18F] incorporation between 70–
91% were achieved and no further improvement was obtained from
longer reaction times (10–15 minutes). Incorporation gradually
Scheme 9 Synthesis of [18F]-FDG from resin-bound precursors.
568 | Org. Biomol. Chem., 2009, 7, 564–575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
◦
being collected. Deprotection of [18F]-18 using 6 M HCl, 125 C
Table 1 Determination of optimum conditions for global deprotection
of [18F]-18
for 5 minutes provided a radioactive ion-exchange chromatogram
with one main peak ([18F]FDG) and one small peak (<3%)
corresponding to partially deprotected [18F]FDG (Fig. 1). The
major radioactive peak (black trace) was shown to co-elute with an
authentic sample of [19F]FDG (not shown) indicating that almost
complete deprotection had occurred. The chromatogram from the
electrochemical detector (grey trace) shows that the protected
glucose formed as a side product in the initial fluoridation
reaction to prepare [18F]-18 has been deprotected and some
other decomposition products formed. The formation of these
by-products are a consequence of residual water present in the
[18F]fluoride solution and elimination due to the use of excess
base in the labeling process. Comparison of the chromatograms
generated using electrochemical and g-detectors indicates that
[18F]FDG was produced with high specific activity. The neutralized
deprotection reaction mixture was analysed further for non-polar
impurities by reversed phase HPLC, identifying benzaldehyde as
the major non-polar impurity. Benzaldehyde was readily removed
Entry
HCl (M)
Temp (◦C)
Time (min)
% Cleavagea
1
2
3
4
5
6
6
6
6
6
5
4
110
110
125
125
125
125
10
5
10
5
5
5
90
52
97
97
81
52
a Determined by ion exchange HPLC measuring radioactivity.
◦
improved with increasing temperature up to 86 C; beyond this,
there was no significant effect upon the radiochemical yield. High
activity labeling studies (5.18–6.16 Ci) were also conducted, and
excellent incorporation yields in the range of 68–77% were realised.
Five sequential labeling experiments re-using the same sample
of resin 35 all led to the formation of protected [18F]FDG with
consistently high radiochemical yields. These experiments clearly
demonstrate that the majority of the protected mannose derivative
remains attached to the resin through the fluoroalkylsulfonyl linker
during the [18F]-labeling experiment, and that one batch of the resin
could be used to generate multiple doses of the radiotracer.
To identify the optimum conditions required for removal of the
protecting groups, a sample of the protected [18F]fluoro-sugar [18F]-
18 was subjected to a range of strongly acidic conditions (Table
1). Dionex anion exchange HPLC with g-detection was used to
quantify [18F]FDG produced. Heating the protected product in
6 N HCl at 110 ◦C for ten minutes achieved a high percentage
deprotection. However, reducing the heating time to five minutes
resulted in a significant drop in the yield of [18F]FDG.
R
by passage of the reaction mixture through a C18 Sep-Pakꢀ
cartridge (benzaldehyde retained).
As a qualitative comparison of the chemical purity of the
protected FDG [18F]-18 from the resin precursor 35 with the
protected FDG product [18F]-2 obtained from the conventional
solution synthesis method, HPLC analyses of the crude products
were undertaken. As expected, both radioactivity traces indicated
that the protected fluorinated sugars and [18F]fluoride were the
major [18F]-containing compounds present. By contrast, the
corresponding UV activity traces showed that the protected [18F]-
sugar obtained from the resin precursor 35 contained significantly
reduced levels of chemical impurities.
We also investigated the effect of altering the linker chain
length on the radiochemical yield for resins 33–36 (Table 2).
Radiochemical yield improved with increasing alkyl chain length
up to four methylene groups, after which it leveled off and then
began to fall. Economically and synthetically, it was more practical
to use the ten-carbon spacer, and given the high radiochemical
yield achieved, immobilized sulfonate ester 35 (y = 9) contained
the linker length of choice.
In radiosynthesis using [18F]fluoride ion, it is desirable to
keep the reaction time as short as possible and the preferred
conditions for the generation of [18F]FDG required 18F labeling
◦
at 86 C for 4 minutes in CH3CN followed by deprotection with
6 M HCl for five minutes at 125 ◦C. This produced an average
radiochemical yield of 73% (decay corrected) and activity losses
on the resin ranged between 3–8%, with 91–97% of the activity
Fig. 1 Dionex anion exchange resin HPLC of the reaction mixture from entry 4, Table 1. Upper line; radioactivity trace. Lower line; electrochemical
detector.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 564–575 | 569
©

View Article Online
64.39, 57.67, 15.02; MS (ES+) m/z 363.1 ([M + Na]+); Anal. Calcd
for C17H24O7: C, 59.99; H, 7.11. Found: C, 59.83; H, 7.14.
Table 2 Showing the effect of increasing linker chain length on the
radiochemical yield of [18F]-18 and remaining [18F]fluoride ion. (C1 resin
was synthesised by coupling methyl-4,6-O-benzylidene-3-O-ethoxymethyl-
b-D-mannopyranoside (15) to resin 21)
Data for methyl-4,6-O-benzylidene-2-O-ethoxymethyl-b-D-
glucopyranoside (12)
Radiochemical
yield (%)a
[18F]fluoride
Entry
Linker length (y)
remaining (%)a
Mp 103–105 ◦C; [a]D -20.8◦ (c = 0.606, CHCl3); nmax (neat, cm-1)
1
1
2
3
4
5
0
1
3
9
15
45–50
75
92
85–90
77
50–55
25
8
10–15
23
3437, 2975, 2879, 1453, 1390, 1089, 1046, 1025; H NMR (300
MHz, CDCl3) d 7.51–7.30 (5H, m), 5.56 (1H, s), 4.89 (1H, d, J =
6.6 Hz), 4.79 (1H, d, J = 7.4 Hz), 4.38 (1H, d, J = 7.4 Hz), 4.36
(1H, dd, J = 5.2, 10.3 Hz), 4.02 (1H, s), 3.85–3.75 (3H, m), 3.62–
3.55 (2H, m), 3.56 (3H, s), 3.46 (1H, ddd, J = 4.4, 9.5, 10.3 Hz),
3.34 (1H, t, J = 8.1 Hz), 1.21 (3H, t, J = 6.6 Hz); 13C NMR (400
MHz, CDCl3) d 137.08, 129.14, 128.26, 126.30, 103.47, 101.88,
96.67, 83.34, 80.59, 72.71, 68.70, 66.08, 64.33, 57.41, 15.01; MS
(ES+) m/z 363.3 ([M + Na]+).
a Determined by HPLC with g-detection (decay corrected).
Conclusions
We have described the synthesis of solid supports that liberate
protected [18F]FDG on treatment with [18F]fluoride ion in high
radiochemical yield, and we have shown that most of the unreacted
precursor remains on the resin thereby avoiding the presence
of excess triflate in subsequent deprotection steps. The product
[18F]-18 can be easily and rapidly deprotected and purified to give
the widely used radiotracer [18F]FDG with high radiochemical
purity. The general approach lends itself to automation of the
radiosynthesis because the solid-supported precursor could be
packaged in a suitable cartridge format. This new platform techno-
logy would be suited to the production of [18F] radiopharmaceuti-
cals where the presence of excess starting materials would have
a deleterious effect on the efficacy of the product or the patient,
particularly when purification of the radiotracer is impractical.
Data for methyl-4,6-O-benzylidene-2,3-O-di(ethoxymethyl)-b-D-
glucopyranoside (14)
◦
Mp 73–75 C; [a]D -62.2◦ (c = 0.475, CHCl3); nmax (neat, cm-1)
1
2975, 2880, 1455, 1390, 1095, 1046, 1028; H NMR (300 MHz,
CDCl3) d 7.50–7.30 (5H, m), 5.52 (1H, s), 4.90 (1H, d, J =
6.6 Hz), 4.89 (2H, s), 4.83 (1H, d, J = 6.6 Hz), 4.36 (1H, d,
J = 8.1 Hz), 4.35 (1H, dd, J = 4.4, 10.3 Hz), 3.87 (1H, t, J =
9.2 Hz), 3.77 (1H, t, J = 10.3 Hz), 3.71–3.52 (6H, m), 3.54 (3H,
s), 3.43 (1H, ddd, J = 5.1, 9.6, 10.3 Hz), 1.22 (3H, t, J = 7.0 Hz),
1.05 (3H, t, J = 6.6 Hz); 13C NMR (300 MHz, CDCl3) d 137.35,
129.18, 128.35, 126.29, 104.73, 101.65, 95.97, 80.89, 78.13, 76.89,
68.95, 66.23, 63.98, 63.75, 57.39, 15.13, 14.91; MS (ES+) m/z 421.4
([M + Na]+).
Methyl-4,6-O-benzylidene-3-O-ethoxymethyl-b-D-
mannopyranoside (15)
Experimental section
Methyl-4,6-O-benzylidene-3-O-ethoxymethyl-b-D-glucopyranoside
(13)
A
solution methyl-4,6-O-benzylidene-3-O-ethoxymethyl-b-D-
glucopyranoside (13, 14.0 g, 41.2 mmol) in DMSO (168 mL) and
Ac2O (84.5 mL, 0.943 mol) was stirred at room temperature for
24 h. The reaction was diluted with EtOAc (1 L) and washed with
a saturated aqueous solution of K2CO3 (600 mL). The organic
layer was separated, dried (Na2SO4) and concentrated to dryness
in vacuo giving the crude ketone as a white solid (20 g). The crude
ketone ◦(20 g) was re-dissolved in MeOH (200 mL) and cooled
to -20 C before NaBH4 (1.69 g, 44.4 mmol) was added slowly,
with stirring. The reaction mixture was allowed to warm to 25 ◦C
then stirred at this temperature for a further 48 h. The reaction was
concentrated in vacuo to give a gum which was partitioned between
EtOAc (250 mL) and saturated aqueous K2CO3. The organic layer
was separated re-extracting the aqueous with EtOAc (2 ¥ 100 mL).
The combined organic phase was dried (Na2SO4) and concentrated
in vacuo. Purification of the resulting off-white solid by silica gel
column chromatography (hexane–EtOAc, 1 : 3) afforded the title
compound as a white solid (15, 10.62 g, 31.2 mmol, 76%) in
addition to some mixed fractions containing some of compound
15 (2.74 g) and pure 13 (1.54 g, 4.0 mmol, 10%). Data for 15: mp
106–108 ◦C; [a]D -38.7◦ (c = 0.375, CHCl3); nmax (neat, cm-1) 3473,
2971, 2891, 1738, 1375, 1092, 1030; 1H NMR (300 MHz, CDCl3)
d 7.50–7.30 (5H, m), 5.56 (1H, s), 4.91 (1H, d, J = 6.9 Hz), 4.83
(1H, d, J = 6.9 Hz), 4.50 (1H, s), 4.34 (1H, dd, J = 5.2, 10.3 Hz),
4.17 (1H, dd, J = 1.5, 2.2 Hz), 4.06 (1H, t, J = 9.6 Hz), 3.91 (1H,
To methyl-4,6-O-benzylidene-b-D-glucopyranoside (11, 847 mg,
3.0 mmol) in anhydrous THF (40 mL) was added NaH (60%
in mineral oil, 192 mg, 4.8 mmol) and the mixture stirred,
under argon, for 30 min. Chloromethylethyl ether (425 mg,
0.42 mL, 4.5 mmol) was added dropwise, and the reaction
stirred at room temperature for 20 h. The reaction was quenched
with methanol (0.5 mL) and concentrated in vacuo. The crude
product was dissolved in CH2Cl2, washed with water and the
aqueous phase extracted with CH2Cl2. The combined organic
phase was washed with water, brine, dried (anhydrous Na2SO4)
and concentrated in vacuo. Purification by silica gel column
chromatography (hexane–EtOAc, 2 : 1) afforded three isolated
products: compound 12 (white solid, 260 mg, 0.76 mmol, 25%),
compound 13 (white solid, 500 mg, 49%) and compound 14 (white
◦
solid, 82 mg, 0.21 mmol, 7%). Data for 13: mp 99–102 C; [a]D
-11.3◦ (c = 0.488, CHCl3); nmax (neat, cm-1) 3450, 2876, 1727, 1380,
1100, 1069, 1026; 1H NMR (300 MHz, CDCl3) d 7.50–7.30 (5H,
m), 5.53 (1H, s), 4.85 (1H, d, J = 7.4 Hz), 4.78 (1H, d, J = 7.4 Hz),
4.38 (1H, d, J = 7.4 Hz), 4.37 (1H, dd, J = 5.2, 10.3 Hz), 4.00
(1H, d, J =1.5 Hz), 3.84–3.74 (2H, m), 3.66 (1H, t, J = 8.1 Hz),
3.62–3.56 (2H, m), 3.60 (3H, s), 3.58–3.40 (2H, m), 1.21 (3H, t, J =
6.6 Hz); 13C NMR (300 MHz, CDCl3) d 137.25, 129.28, 128.40,
126.33, 104.53, 101.77, 96.68, 82.17, 79.71, 74.09, 68.86, 66.55,
570 | Org. Biomol. Chem., 2009, 7, 564–575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
dd, J = 2.9, 9.5 Hz), 3.88 (1H, t, J = 10.3 Hz), 3.73–3.60 (2H,
m), 3.58 (3H, s), 3.40 (1H, ddd, J = 5.2, 9.6, 10.3 Hz), 2.59 (1H,
d, J = 1.5 Hz), 1.17 (3H, t, J = 7.0 Hz); 13C NMR (300 MHz,
CDCl3) d 137.48, 129.15, 128.36, 126.25, 101.87, 101.50, 94.91,
77.65, 74.89, 70.56, 68.74, 67.13, 63.79, 57.50, 15.16; MS (ES+)
m/z 703.2 ([2M + Na]+); HRMS (ES) Calcd for C17H24O7Na:
363.1414. Found 363.1416; Anal. Calcd for C17H24O7: C, 59.99; H,
7.11. Found: C, 59.90; H, 7.12.
(d, J = 18.8 Hz), 68.69, 66.38, 63.52, 57.61, 14.90; 19F NMR
(282 MHz, CDCl3, ref. C6F6) d -37.30 (ddd, J = 4.6, 16.2,
53.3 Hz); MS (ES+) m/z 365.1 ([M + Na]+); HRMS (ES) Calcd
for C17H23O6FNa: 365.1371. Found 365.1369.
Data for methyl-4,6-O-benzylidene-3-O-ethoxymethyl-2-deoxy-b-
D-erythro-hex-2-enopyranoside (17)
Mp 77–79 C; [a]D +11.9◦ (c = 0.118, CHCl3); nmax (film, cm-1)
◦
1
2927, 2865, 1692, 1590, 1248, 1138, 1094; H NMR (300 MHz,
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(1,1,2,2,3,3,4,4,4-
nonafluoro-butane-sulfonate)-b-D-mannopyranoside (16)
CDCl3) d 7.50–7.34 (5H, m), 5.61 (1H, s), 5.39 (1H, dd, J = 1.5,
2.2 Hz), 5.10 (2H, s), 4.99 (1H, t, J = 1.5 Hz), 4.40 (1H, dt, J =
2.2, 8.1 Hz), 4.32 (1H, dd, J = 4.4, 9.5 Hz), 3.90 (1H, t, J =
9.5 Hz), 3.80 (1H, ddd, J = 4.4, 8.1, 10.3 Hz), 3.68 (2H, q, J =
7.4 Hz), 3.47 (3H, s), 1.22 (3H, t, J = 7.4 Hz); 13C NMR (300
MHz, CDCl3) d 153.03, 137.21, 129.15, 128.37, 126.42, 102.37,
100.06, 99.73, 92.86, 74.76, 69.19, 68.94, 64.73, 54.68, 15.11; MS
(ES+) m/z 255.0 ([M - C6H5CH + Na]+), 233.1 ([M - C6H5CH]+).
To a solution of 15 (204 mg, 0.6 mmol) in anhydrous THF
(8 mL) was added NaH (60% in mineral oil, 43 mg, 1.1 mmol)
and the reaction stirred, under argon, at room temperature for
30 min. Perfluoro-n-butylsulfonyl fluoride (290 mg, 0.1 mmol),
was added, dropwise, and the reaction stirred for a further 1.5 h.
The reaction was quenched with methanol and concentrated
in vacuo. The crude product was dissolved in CH2Cl2, washed
with water and the aqueous phase extracted with CH2Cl2. The
combined organic phase was washed with water, brine, dried
(Na2SO4) and concentrated in vacuo. Purification by silica gel
column chromatography (hexane–EtOAc, 2 : 1) afforded the title
compound as a white solid (16, 233 mg, 62%). Mp 93–95 ◦C; [a]D
-56.7◦ (c = 0.425, CHCl3); nmax (film, cm-1) 2944, 1714, 1410, 1353,
Preparation of FDG tetraacetates [19F]2 and 19
A solution of 18 (59 mg, 0.17 mmol) in 6 N HCl (2 mL) was refluxed
at 140 ◦C (bath) for 10 min. The reaction was concentrated in
vacuo and Ac2O (0.5 mL) and pyridine (1.0 mL) were added. The
reaction was stirred at room temperature overnight, concentrated
and purified by silica gel column chromatography (hexane–EtOAc
2 : 1) to afford compounds [19F]2 (22 mg, 36%) and 19 (21 mg,
35%).
1
1197, 1143, 1078, 934; H NMR (300 MHz, CDCl3) d 7.49–7.33
(5H, m), 5.59 (1H, s), 5.15 (1H, d, J = 3.0 Hz), 4.87 (1H, d, J =
6.9 Hz), 4.78 (1H, d, J = 7.4 Hz), 4.62 (1H, s), 4.36 (1H, dd, J =
5.0, 10.9 Hz), 4.15 (1H, dd, J = 3.0, 9.9 Hz), 3.91 (1H, t, J =
9.4 Hz), 3.90 (1H, t, J = 10.4 Hz), 3.74–3.59 (2H, m), 3.58 (3H, s),
3.46 (1H, ddd, J = 4.5, 9.5, 9.9 Hz), 1.15 (3H, t, J = 6.9 Hz); 13
C
Data for 2-fluoro-2-deoxy-b-D-glucopyranoside tetraacetate [19F]2
[a]D +43◦ (c = 0.41, CH2Cl2), lit.31 [a]D +50◦; nmax (film, cm-1)
NMR (400 MHz, CDCl3) d 137.10, 129.29, 128.37, 126.19, 101.93,
99.15, 94.09, 83.83, 77.36, 71.01, 68.44, 67.59, 63.98, 57.49, 14.99;
19F NMR (282 MHz, CDCl3, ref. C6F6) d 81.27, 52.03, 40.89,
35.83.
1
2957, 1748, 1434, 1369, 1212, 1072, 1035; H NMR (300 MHz,
CDCl3) d 5.79 (1H, dd, J = 3.2, 8.1 Hz), 5.38 (1H, td, J = 9.6,
15.5 Hz), 5.08 (1H, t, J = 9.6 Hz), 4.45 (1H, ddd, J = 8.1, 9.6,
51 Hz), 4.30 (1H, dd, J = 4.4, 12.5 Hz), 4.10 (1H, dd, J = 2.0,
12.5 Hz), 3.87 (1H, ddd, J = 2.0, 4.4, 9.6 Hz), 2.19 (3H, s), 2.10
(3H, s), 2.09 (3H, s), 2.05 (3H, s); 13C NMR (300 MHz, CDCl3) d
170.72, 170.04, 169.70, 168.99, 91.65 (d, J = 24 Hz), 88.60 (d, J =
192 Hz), 73.17, 73.12 (d, J = 19 Hz), 68.03 (d, J = 8 Hz), 61.75,
20.98, 20.86, 20.80, 20.71; 19F NMR (282 MHz, CDCl3, ref. C6F6)
d -39.05 (m). Spectroscopic data consistent with those reported
previously in the literature.20a
Solution-phase fluoridation: preparation of 17 and 18
A solution of 16 (124 mg, 0.20 mmol), 1,10-diaza-4,7,13,16,21,24-
hexaoxabicyclo[8,8,8] hexacosan (90 mg, 0.24 mmol) and KF
(14 mg, 0.24 mmol) in MeCN (4 mL) was vigorously refluxed,
under argon, for 10 min. The reaction was cooled and concen-
trated in vacuo. The crude product was dissolved in CH2Cl2,
washed with water and the aqueous phase re-extracted with
CH2Cl2. The combined organic phase was washed with water,
brine, dried (Na2SO4) and concentrated in vacuo. Purification by
silica gel column chromatography (hexane–Et2O, 2 : 1) afforded
compound 18 (36 mg, 53%) and compound 17 (11 mg, 17%).
Data for methyl-4,6-O-benzylidene-3-O-ethoxymethyl-2-deoxy-2-
fluoro-b-D-glucopyranoside (18): mp 44–46 ◦C; [a]D -22.9◦ (c =
0.431, CHCl3); nmax (film, cm-1) 2972, 2885, 1454, 1383, 1095, 1030;
1H NMR (300 MHz, CDCl3) d 7.48–7.32 (5H, m), 5.53 (1H, s),
4.91 (1H, d, J = 6.6 Hz), 4.83 (1H, d, J = 6.6 Hz), 4.52 (1H,
dd, J = 3.7, 7.7 Hz), 4.37 (1H, dd, J = 5.2, 10.3 Hz), 4.26 (1H,
td, J = 7.7, 49.3 Hz), 4.09 (1H, td, J = 9.2, 15.4 Hz), 3.78 (1H,
t, J = 10.3 Hz), 3.69–3.57 (3H, m), 3.60 (3H, s), 3.51–3.45 (1H,
m), 1.12 (3H, t, J = 7.4 Hz); 13C NMR (300 MHz, CDCl3) d
137.12, 129.29, 128.39, 126.25, 102.30 (d, J = 23.7 Hz), 101.68,
95.39, 92.55 (d, J = 186.8 Hz), 80.13 (d, J = 9.2 Hz), 75.15
Data for 2-deoxy-2-fluoro-a-D-glucopyranoside tetraacetate (19)
[a]D +119◦ (c = 0.53, CH2Cl2), lit.32 [a]D +147◦ (c = 1.0, CHCl3);
1
nmax (film, cm-1) 2963, 1746, 1434, 1370, 1213, 1078, 1036; H
NMR (300 MHz, CDCl3) d 6.44 (1H, d, J = 4.4 Hz), 5.50 (1H,
td, J = 9.6, 12.5 Hz), 5.08 (1H, t, J = 9.6 Hz), 4.66 (1H, ddd,
J = 4.4, 9.6, 49 Hz), 4.30 (1H, dd, J = 4.4, 12.5 Hz), 4.13–4.00
(2H, m), 2.21 (3H, s), 2.10 (3H, s), 2.09 (3H, s), 2.05 (3H, s);
13C NMR (300 MHz, CDCl3) d 170.59, 170.17, 169.50, 168.70,
88.72 (d, J = 22 Hz), 86.60 (d, J = 195 Hz), 70.97 (d, J =
19 Hz), 69.96, 67.83 (d, J = 8 Hz), 61.76, 20.88, 20.70, 20.56; 19
F
NMR (282 MHz, CDCl3, ref. C6F6) d -40.64 (dd, J = 10, 48 Hz).
Spectroscopic data consistent with those reported previously in the
literature.32
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 564–575 | 571
©

View Article Online
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(5-iodooctafluoro-3-
oxapentane sulfonate)-b-D-mannopyranoside (25)
NMR (282 MHz, CDCl3, ref. C6F6) d 79.74, 72.97, 47.56, 43.89;
MS (ES-) m/z 844.7 ([M]-).
To methyl-4,6-O-benzylidene-3-O-ethoxy methyl-b-D-mannopy-
ranoside (15, 315 mg, 0.93 mmol) in anhydrous THF (10 mL)
was added NaHMDS (1.0 M solution in THF, 1.1 mL, 1.1 mmol)
and the reaction stirred, under argon, at room temperature for
20 min. 5-Iodooctafluoro-3-oxapentanesulfonyl fluoride (437 mg,
1.11 mmol), was added, dropwise, and the reaction stirred for a
further 45 min. The reaction was concentrated in vacuo, dissolved
in Et2O, washed with saturated K2CO3 solution and water and
the aqueous phase re-extracted with Et2O. The combined organic
phase was washed with brine, dried (anhydrous MgSO4) and
concentrated in vacuo. Purification by silica gel column chromatog-
raphy (hexane–EtOAc, 3 : 1) afforded the desired product 25 as
a white solid (540 mg, 77%). [a]D -4.0◦ (c = 0.028, CHCl3); nmax
(film, cm-1) 2971, 2890, 1738, 1411, 1380, 1208, 1146, 1092, 1025,
915; 1H NMR (300 MHz, CDCl3) d 7.50–7.30 (5H, m), 5.57 (1H,
s), 5.13 (1H, d, J = 2.9 Hz), 4.86 (1H, d, J = 7.4 Hz), 4.78 (1H,
d, J = 7.4 Hz), 4.57 (1H, s), 4.34 (1H, dd, J = 5.1, 10.3 Hz),
4.12 (1H, dd, J = 2.9, 9.6 Hz), 3.94–3.82 (2H, m), 3.74–3.55 (2H,
m), 3.56 (3H, s), 3.44 (1H, ddd, J = 4.4, 9.5, 10.3 Hz), 1.13 (3H,
t, J = 6.6 Hz); 13C NMR (75 MHz, CDCl3) d 137.13, 129.28,
128.37, 126.19, 101.92, 99.16, 94.07, 83.61, 77.32, 71.04, 68.44,
67.58, 63.96, 57.50, 15.01; 19F NMR (282 MHz, CDCl3, ref. C6F6)
d 96.98, 79.97, 76.42, 48.24; MS (ES+) m/z 769.1 ([M + Na]+);
Anal. Calcd for C21H23O10F8IS: C, 33.79; H, 3.11. Found: C, 33.86;
H, 3.13.
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
12,12,13,13,15,15,16,16-octa fluoro-10-iodo-hexadecanoic
acid-16-sulfonate)-b-D-mannopyranoside (27)
Using the general procedure A with the following amounts: iodide
25 (300 mg, 0.40 mmol) and undecylenic acid (89 mg, 0.48 mmol)
in CH3CN (2 mL), H2O (1 mL) with NaHCO3 (41 mg, 0.48 mmol)
and Na2S2O4 (85%, 97 mg, 0.48 mmol). Purification by silica gel
column chromatography twice, first eluting with CH2Cl2–CH3OH
(98 : 2), secondly eluting with CH2Cl2–CH3OH (99 : 1) afforded
the desired product 27 as a colourless oil (200 mg, 54%). nmax
(film, cm-1) 2931, 2858, 1709, 1410, 1192, 1147, 1093, 1025, 919;
1H NMR (300 MHz, CDCl3) d 7.50–7.31 (5H, m), 5.58 (1H, s),
5.15 (1H, d, J = 2.9 Hz), 4.85 (1H, d, J = 7.4 Hz), 4.79 (1H, d,
J = 7.4 Hz), 4.61 (1H, s), 4.40–4.25 (2H, m), 4.17 (1H, dd, J = 2.9,
10.3 Hz), 3.95–3.85 (2H, m), 3.75–3.60 (2H, m), 3.58 (3H, s), 3.51–
3.42 (1H, m), 3.00–2.65 (2H, m), 2.35 (2H, t, J = 7.4 Hz), 1.90–1.30
(14H, m), 1.15 (3H, t, J = 6.6 Hz); 13C NMR (75 MHz, CDCl3)
d 178.22, 136.01, 128.10, 128.18, 126.46, 102.20, 99.47, 94.26,
83.84, 77.58, 71.39, 68.71, 67.90, 64.25, 57.73, 41.65, 40.69, 34.31,
29.86, 29.53, 29.49, 29.37, 28.82, 25.04, 21.48, 15.23; 19F NMR
(282 MHz, CDCl3, ref. C6F6) d 79.78, 74.37, 47.92, 43.70; MS
(ES-) m/z 929.3 ([M]-); HRMS (ES+) Calcd for C32H43O12F8ISNa:
953.1284. Found 953.1266.
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
18,18,19,19,21,21,22,22-octafluoro-16-iodo-docosanoic
acid-22-sulfonate)-b-D-mannopyranoside (28)
General procedure A: the alkylation of iodide 25 to provide acids
26–29
To the iodide 25 and the carboxylic acid (1.2 equiv) in MeCN
was added H2O followed by NaHCO3 (1.2 equiv) and Na2S2O4
(85%, 1.2 equiv) and the reaction stirred at room temperature for
30 min. The reaction was concentrated in vacuo, dissolved in Et2O
(100 mL), washed with water (100 mL) and the aqueous phase
re-extracted with Et2O (50 mL). The combined organic phase was
washed with brine (100 mL), dried (MgSO4), concentrated in vacuo
and purified as described.
Using the general procedure A with the following amounts:
iodide 25 (1.0 g, 1.34 mmol) and 16-heptadecenoic acid (378 mg,
1.41 mmol) in CH3CN (60 mL), H2O (30 mL) with NaHCO3
(135 mg, 1.61 mmol) and Na2S2O4 (85%, 322 mg, 1.61 mmol). The
reaction remained cloudy and was stirred at room temperature
for 1 h. Purification by silica gel column chromatography, eluting
with EtOAc–hexane (1 : 3) afforded the desired product 28 as a
colourless oil (230 mg, 17%). nmax (film, cm-1) 2927, 2855, 1725,
1412, 1194, 1149, 1096, 1026, 921; 1H NMR (300 MHz, CDCl3) d
7.50–7.31 (5H, m), 5.58 (1H, s), 5.15 (1H, d, J = 2.9 Hz), 4.86 (1H,
d, J = 7.4 Hz), 4.80 (1H, d, J = 7.4 Hz), 4.63 (1H, s), 4.40–4.28
(2H, m), 4.17 (1H, dd, J = 2.9, 10.3 Hz), 3.97–3.87 (2H, m), 3.75–
3.62 (2H, m), 3.60 (3H, s), 3.52–3.42 (1H, m), 3.00–2.70 (2H, m),
2.37 (2H, t, J = 7.4 Hz), 1.85–1.40 (4H, m), 1.30–1.10 (22H, m),
1.15 (3H, t, J = 7.4 Hz); 13C NMR (100 MHz, CDCl3) d 179.04,
137.56, 129.67, 128.76, 126.62, 102.36, 99.64, 94.47, 84.00, 77.59,
71.55, 68.87, 68.06, 64.41, 57.91, 42.02, 41.82, 41.62, 40.91, 34.48,
30.17, 30.14, 30.10, 29.99, 29.93, 29.81, 29.65, 29.10, 25.32, 21.79,
15.40; 19F NMR (282 MHz, CDCl3, ref. C6F6) d 80.20, 73.14,
47.87, 43.76; MS (ES-) m/z 1013.1 ([M]-), 1127.5 ([M + TFA]-).
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
6,6,7,7,9,9,10,10-octafluoro-4-iodo-decanoic
acid-10-sulfonate)-b-D-mannopyranoside (26)
Using the general procedure A with the following amounts: iodide
25 (400 mg, 0.54 mmol) and 4-pentenoic acid (56 mg, 0.56 mmol)
in CH3CN (4 mL), H2O (2 mL) with NaHCO3 (59 mg, 0.70 mol)
and Na2S2O4 (85%, 140 mg, 0.70 mmol). Purification by silica
gel column chromatography eluting with EtOAc–hexane (1 : 2)
afforded the desired product 26 as a colourless oil (305 mg, 67%).
nmax (film, cm-1) 2975, 2878, 1714, 1410, 1192, 1148, 1093, 1025,
920; 1H NMR (300 MHz, CDCl3) d 7.50–7.25 (5H, m), 5.52 (1H,
s), 5.17 (1H, d, J = 2.9 Hz), 4.86 (1H, d, J = 7.4 Hz), 4.80 (1H,
d, J = 7.4 Hz), 4.60 (1H, s), 4.34–4.24 (2H, m), 4.09 (1H, dd,
J = 2.9 Hz), 3.90–3.77 (2H, m), 3.70–3.50 (2H, m), 3.50 (3H, s),
3.45–3.32 (1H, m), 2.98–2.28 (4H, m), 2.18–1.90 (2H, m), 1.05
(3H, t, J = 7.4 Hz); 13C NMR (75 MHz, CDCl3) d 177.32, 137.12,
129.28, 128.36, 126.20, 101.93, 99.19, 94.02, 83.66, 77.28, 71.10,
68.43, 67.61, 64.01, 57.54, 41.37, 35.14, 34.21, 19.09, 14.98; 19F
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
4,4,5,5,7,7,8,8-octafluoro-octanoic acid-8-sulfonate)-b-D-
mannopyranoside (29)
Using the general procedure A with the following amounts: iodide
25 (1.0 g, 1.34 mmol) and acrylic acid (101 mg, 1.41 mmol) in
CH3CN (8 mL), H2O (4 mL) with NaHCO3 (135 mg, 1.61 mmol)
572 | Org. Biomol. Chem., 2009, 7, 564–575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
and Na2S2O4 (85%, 322 mg, 1.61 mmol). Purification by silica
gel column chromatography, eluting with EtOAc–hexane (1 : 4)
to EtOAc–hexane (1 : 0) afforded the desired product 29 as a
colourless oil (419 mg, 38%). nmax (film, cm-1) 2972, 1722, 1412,
1193, 1148, 1096, 1026, 921; 1H NMR (300 MHz, CDCl3) d 7.52–
7.32 (5H, m), 5.58 (1H, s), 5.15 (1H, d, J = 2.9 Hz), 4.85 (1H, d,
J = 7.4 Hz), 4.80 (1H, d, J = 7.4 Hz), 4.61 (1H, s), 4.30 (1H, q,
J = 5.2 Hz), 4.16 (1H, dd, J = 2.9, 10.3 Hz), 3.90–3.78 (2H, m),
3.75–3.62 (2H, m), 3.58 (3H, s), 3.51–3.42 (1H, m), 2.68 (2H, t,
4.35–4.25 (1H, q, J = 5.8 Hz), 4.17 (1H, dd, J = 2.9, 10.3 Hz),
3.95–3.86 (2H, m), 3.75–3.60 (2H, m), 3.58 (3H, s), 3.51–3.42 (1H,
m), 2.35 (2H, t, J = 7.4 Hz), 2.15–1.90 (2H, m), 1.65–1.25 (14H,
m), 1.10 (3H, t, J = 7.4 Hz); 13C NMR (75 MHz, CDCl3) d 179.34,
129.39, 128.48, 126.34, 102.08, 99.38, 94.14, 83.57, 77.46, 71.31,
68.59, 67.78, 64.11, 57.58, 34.22, 30.89, 30.67, 30.45, 29.57, 29.45,
29.36, 29.30, 24.95, 20.56, 15.09; 19F NMR (282 MHz, CDCl3, ref.
C6F6) d 79.8, 74.2, 47.9, 43.5; MS (ES-) m/z 803.3 ([M]-); HRMS
(ES+) Calcd for C32H44O12F8SNa: 827.2318. Found 827.2299.
J = 7.4 Hz), 2.55–2.36 (2H, m), 1.65–1.15 (3H, t, J = 7.4 Hz); 13
C
NMR (75 MHz, CDCl3) d 176.63, 137.11, 129.31, 128.38, 126.20,
101.92, 99.19, 93.93, 83.61, 77.23, 71.09, 68.41, 67.58, 64.01, 57.54,
25.97, 25.69, 14.94; 19F NMR (282MHz, CDCl3, ref. C6F6) d 79.82,
73.56, 47.73, 43.27; MS (ES-) m/z 804.8 ([M + TFA]-); HRMS
(ES+) Calcd for C24H28O12F8IS Na: 715.1066. Found 715.1068.
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
18,18,19,19,21,21,22,22-octafluoro-docosanoic
acid-22-sulfonate)-b-D-mannopyranoside (32)
Using the general procedure B with the following amounts: acid 28
(200 mg, 0.20 mmol), zinc (80 mg, 1.23 mmol) in Et2O (4 mL) and
AcOH (2 mL). Purification by silica gel column chromatography,
eluting with EtOAc–hexane (1 : 3) afforded the desired product 32
as a colourless oil (132 mg, 75%). nmax (film, cm-1) 2926, 2854, 1711,
1412, 1191, 1149, 1096, 1027, 920; 1H NMR (300 MHz, CDCl3) d
7.50–7.35 (5H, m), 5.60 (1H, s), 5.15 (1H, d, J = 2.9 Hz), 4.86 (1H,
d, J = 7.4 Hz), 4.79 (1H, d, J = 7.4 Hz), 4.61 (1H, s), 4.36 (1H, q,
J = 5.1 Hz), 4.20–4.12 (1H, m), 3.95–3.85 (2H, m), 3.75–3.60 (2H,
m), 3.58 (3H, s), 3.51–3.42 (1H, m), 2.37 (2H, t, J = 7.4 Hz), 1.70–
1.55 (4H, m), 1.40–1.20 (26H, m), 1.15 (3H, t, J = 7.4 Hz); 13C
NMR (75 MHz, CDCl3) d 179.40, 137.07, 129.51, 128.60, 126.45,
102.19, 99.49, 94.28, 83.70, 77.58, 71.43, 68.71, 67.89, 64.29, 57.78,
34.46, 31.02, 30.80, 30.58, 30.04, 29.99, 29.84, 29.80, 29.65, 29.49,
25.17, 21.43, 20.70, 15.21; 19F NMR (282 MHz, CDCl3, ref. C6F6)
d 80.01, 74.08, 47.96, 43.34; MS (ES-) m/z 887.1 ([M]-); HRMS
(ES+) Calcd for C38H56O12F8SNa: 911.3257. Found 911.3275.
General procedure B: the reduction of acids 26–28 to provide acids
30–32
To acid (26–28) in Et2O was added zinc (99.998%, 100 mesh,
6 equiv) and AcOH and the reaction heated at reflux, under argon,
for 3 h (bath temp = 80 ◦C). The reaction was allowed to cool to
room temperature and filtered through celite, washing with Et2O
(3 ¥ 30 mL). The filtrate and combined washings were concentrated
to dryness in vacuo and purified as described.
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
6,6,7,7,9,9,10,10-octafluoro-decanoic acid-10-sulfonate)-
b-D-mannopyranoside (30)
Using the general procedure B with the following amounts: acid 26
(300 mg, 0.36 mmol), zinc (93 mg, 1.42 mmol) in Et2O (2 mL) and
AcOH (1 mL). Purification by silica gel column chromatography,
eluting with EtOAc–hexane (1 : 3) afforded the desired product
30 as a colourless oil (70 mg, 27%). nmax (film, cm-1) 2932, 1723,
General procedure C: coupling of acids 29–32 to provide resins
33–36
1
1410, 1192, 1149, 1115, 1027, 922; H NMR (300 MHz, CDCl3)
To amino-methylated polystyrene (NovaBiochem, 50–100 mesh,
loading: 1.5 mmol g-1) and acid (29–32, 1.3 equiv) in anhydrous
CH2Cl2 was added N,N -diisopropylethylamine (2.6 equiv) fol-
lowed by diphenylphosphoryl chloride (1.3 equiv). The reaction
was stirred gently, under argon, at room temperature for 18 h. The
resin was removed by filtration, washed with CH2Cl2 (3 ¥ 10 mL),
CH3OH (3 ¥ 10 mL), Et2O (3 ¥ 10 mL) and dried in vacuo, at 40 ◦C
for 48 h.
Lower loading resins 35 were synthesised using reduced
amounts of the acid 31. Residual aminomethyl groups were capped
using excess Ac2O and pyridine. A Kaiser test was carried out for
qualitative detection of amino groups. If the test was positive, the
capping process was repeated.
d 7.50–7.31 (5H, m), 5.58 (1H, s), 5.15 (1H, d, J = 2.9 Hz), 4.87
(1H, d, J = 7.4 Hz), 4.79 (1H, d, J = 7.4 Hz), 4.63 (1H, s), 4.36
(1H, q, J = 5.2 Hz), 4.17 (1H, dd, J = 2.9, 10.3 Hz), 3.95–3.86
(2H, m), 3.75–3.60 (2H, m), 3.58 (3H, s), 3.52–3.43 (1H, m), 2.42
(2H, t, J = 7.4 Hz), 2.20–2.00 (2H, m), 1.80–1.60 (4H, m), 1.15
(3H, t, J = 7.4 Hz); 13C NMR (75 MHz, CDCl3) d 178.65, 137.11,
129.29, 128.37, 126.20, 101.92, 99.19, 93.99, 83.50, 77.27, 71.08,
68.42, 67.60, 64.00, 57.54, 33.55, 30.29, 24.16, 20.07, 14.97; 19F
NMR (282 MHz, CDCl3, ref. C6F6) d 79.72, 73.61, 47.91, 43.70;
MS (ES-) m/z 719.0 ([M]-), 832.9 ([M + TFA]-); HRMS (ES+)
Calcd for C26H32O12F8SNa: 743.1379. Found 743.1365.
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
12,12,13,13,15,15,16,16-octafluoro-hexadecanoic
acid-16-sulfonate)-b-D-mannopyranoside (31)
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
4,4,5,5,7,7,8,8-octafluoro-octanoic acid-8-sulfonate)-
b-D-mannopyranoside polystyryl amide (33)
Using the general procedure B with the following amounts: acid 27
(190 mg, 0.20 mmol), zinc (41 mg, 0.61 mmol) in Et2O (2 mL) and
AcOH (1 mL). Purification by silica gel column chromatography,
eluting with EtOAc–hexane (1 : 3) afforded the desired product
31 as a colourless oil (97 mg, 59%). nmax (film, cm-1) 2929, 2858,
2858, 1710, 1411, 1147, 1093, 1025, 994, 918; 1H NMR (300 MHz,
CDCl3) d 7.50–7.31 (5H, m), 5.58 (1H, s), 5.15 (1H, d, J = 2.9 Hz),
4.85 (1H, d, J = 7.4 Hz), 4.79 (1H, d, J = 7.4 Hz), 4.63 (1H, s),
Using general procedure C with the following amounts: amino-
methylated polystyrene (145 mg, 0.218 mmol), acid 29 (200 mg,
0.289 mmol) in CH2Cl2 (2 mL), N,N -diisopropylethylamine
(75 mg, 100 mL, 0.579 mmol) and diphenylphosphoryl chloride
(68 mg, 55 mL, 0.289 mmol). This gave the title resin 33 as a
pale yellow solid (283 mg, 94%). nmax (on-bead, cm-1) 2970, 1739,
1418, 1366, 1216, 1093; 19F NMR (282 MHz, CDCl3, ref. CFCl3)
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 564–575 | 573
©

View Article Online
d -81.84, -87.79, -113.67, -117.96; Loading Calcd: 0.75 mmol
eluting with hexane–i-PrOH (9 : 1 → 7 : 3) to afford compound
[19F]-18 (13 mg, 0.04 mmol, 16% based on KF). Compounds 17,
37 and 38 were also present in the crude isolated reaction mixture.
g-1. Found (F elemental analysis) 0.80 mmol g-1.
1
For H and 19F NMR spectra of the crude reaction mixture, see
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
6,6,7,7,9,9,10,10-octafluoro-decanoic acid-10-sulfonate)-
b-D-mannopyranoside polystyryl amide (34)
ESI.† Data for compound 38: nmax (neat, cm-1) 2975, 2884, 1751;
1H NMR (300 MHz, CDCl3) d 7.50–7.33 (5H, m), 5.54 (1H, s),
5.00 (1H, dd, J = 8.0, 9.3 Hz), 4.85 (1H, d, J = 7.0 Hz), 4.67
(1H, d, J = 7.0), 4.43 (1H, d, J = 8.0 Hz), 4.37 (1H, dd, J = 5.0,
9.3 Hz), 3.97 (1H, t, J = 9.3 Hz), 3.80 (1H, t, J = 10.3 Hz), 3.69
(1H, t, J = 9.3 Hz), 3.62–3.45 (3H, m), 3.51 (3H, s), 2.12 (3H, s),
1.12 (3H, t, J = 7.1 Hz); 13C NMR (300 MHz, CDCl3) d 169.43,
137.05, 129.05, 128.20, 126.07, 102.46, 101.57, 94.34, 80.96, 75.48,
72.78, 68.68, 66.25, 63.39, 57.00, 20.90, 14.80; MS (ES+) m/z 405
([M + Na]+).
Using general procedure C with the following amounts: amino-
methylated polystyrene (45 mg, 0.067 mmol), acid 30 (58 mg,
0.289 mmol) in CH2Cl2 (1 mL), N,N -diisopropylethylamine
(21 mg, 28 mL, 0.162 mmol) and diphenylphosphoryl chloride
(19 mg, 16 mL, 0.081 mmol). This gave the title resin 34 as a pale
yellow solid (87 mg, 99%). nmax (on-bead, cm-1) 2931, 1662, 1493,
1452, 1410, 1275, 1146, 1094, 1025, 919; 19F NMR (282 MHz,
CDCl3, ref. CFCl3) d -82.01, -88.13, -113.86, -118.23; Loading
Calcd: 0.73 mmol g-1. Found (F elemental analysis) 0.55 mmol g-1.
Procedure for the manual radiosynthesis of
[18F]-2-fluoro-2-deoxy-D-glucose ([18F]-3) from a resin-bound
b-D-mannose derivative 35
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
12,12,13,13,15,15,16,16-octafluoro-hexadecanoic
Labeling. A carbon glass reaction vessel was placed in a brass
heater and the pot lid (with three lines attached to allow evapora-
tion, nitrogen flow, and addition of reagents) tightened down and
the whole system leak tested. Kryptofix [2.2.2] (22 mg) in CH3CN
(300 mL) and K2CO3 (8 mg) in water (300 mL) were transferred
using a plastic syringe (1 mL) into the carbon glass reaction vessel.
The [18F]fluoride ion (185–370 MBq) in water (0.5–2 mL) was
added and heated to 125 ◦C. After 15 min, three aliquots of
CH3CN (0.5 mL) were added at 1 min intervals. [18F]Fluoride ion
was dried up to 40 min in total. The heater was cooled to room tem-
perature, the pot lid removed and CH3CN (0.2 mL) was added. The
pot lid was replaced and t◦he lines were capped off with stoppers.
The heater was set at 100 C for 10 min and the [18F]fluoride ion
redissolved. After cooling to room temperature, the [18F]fluoride
ion solution (0.2 mL) was transferred to a second carbon glass re-
action vessel containing the resin (20–25 mg) using a plastic syringe
(1 mL). This carbon glass vessel was transferred to an ion chamber
and the labeling activity measured. The carbon glass vessel was
replaced in the brass heater and the capped pot lid tightened. The
reaction was heated to 86 ◦C for 4 min before cooling with com-
pressed air. The pot lid was removed, CH3CN (1.0 mL) was added
and the activity in the reaction vessel was measured. The resin
was mixed and drawn up into a plastic syringe and then passed
through a sintered syringe and into a collection vial. The reaction
vessel was washed with a further volume of CH3CN (0.5 mL) and
passed through the sintered syringe. The activity in the collection
vial, on the resin and of the sintered syringe was measured. The
reaction mixture was diluted with water (1 mL), passed through a
C18 Sep-Pak (Waters) that had been pre-conditioned with EtOH
(5 mL) and water (10 mL). This solution was then passed through
a silica Sep-Pak (Waters) that had been pre-conditioned with Et2O
(5 mL). The activities of all solutions were measured at each stage.
acid-16-sulfonate)-b-D-mannopyranoside polystyryl amide (35)
Using general procedure C with the following amounts: amino-
methylated polystyrene (50 mg, 0.07 mmol), acid 31 (80 mg,
0.10 mmol) in CH2Cl2 (1 mL), N,N -diisopropylethylamine (35 mL,
0.20 mmol) and diphenylphosphoryl chloride (24 mg, 0.10 mmol).
This gave the title resin 35 as a pale yellow solid (103 mg, 90%).
nmax (on-bead, cm-1) 2925, 1662, 1493, 1453, 1411, 1146, 1095,
1025; 19F NMR (282 MHz, CDCl3, ref. CFCl3) d -82.1, -88.1,
-113.9, -118.3; Theoretical loading calcd: 0.69 mmol g-1. Found
(F elemental analysis) 0.59 mmol g-1.
A lower loading resin 35, capped with Ac2O, gave the following
data: 19F NMR (282 MHz, CDCl3, ref. CFCl3) d -82.0, -88.1,
-113.8, -118.2; Loading (S elemental analysis) 0.1 mmol g-1.
Methyl-4,6-O-benzylidine-3-ethoxymethyl-2-(3-oxa-
18,18,19,19,21,21,22,22-octafluoro-docosanoic
acid-22-sulfonate)-b-D-mannopyranoside polystyryl amide (36)
Using general procedure C with the following amounts: amino-
methylated polystyrene (62 mg, 0.093 mmol), acid 32 (108 mg,
0.121 mmol) in CH2Cl2 (3 mL), N,N -diisopropylethylamine
(42 mL, 0.243 mmol) and diphenylphosphoryl chloride (29 mg,
0.121 mmol). This gave the title resin as a pale yellow solid
(36, 136 mg, 86%). nmax (on-bead, cm-1) 2925, 1662, 1493, 1453,
1411, 1146, 1095, 1025; 19F NMR (282 MHz, CDCl3, ref. CFCl3)
d -82.06, -88.14, -113.93, -118.34; Theoretical loading calcd:
0.65 mmol g-1. Found (F elemental analysis) 0.52 mmol g-1.
Solid-phase fluoridation using K19F: preparation of [19F]-18
To a solution of KF (14 mg, 0.25 mmol) and 1,10-diaza-4,7,13,
16,21,24-hexaoxabicyclo[8,8,8] hexacosan (94 mg, 0.25 mmol) in
MeCN (15 mL) in a standard glass reaction vessel, was added resin
35 (5.0 g, 0.1 mmol g-1, 0.5 mmol) and the suspension was refluxed
at 86 ◦C, under nitrogen, for 1 h. The cooled reaction mixture was
filtered and the resin washed with CH2Cl2 (30 mL). The combined
filtrate and washing were concentrated in vacuo. The crude product
was passed through a short plug of silica gel eluting with EtOAc–
hexane (4 : 6) to give a mixture of products (31 mg) which were
subsequently purified by preparative HPLC on a silica gel column
Deprotection. The solution of [18F]-18 after treatment through
the Sep-Pak cartridges was transferred into a third carbon glass
reaction vessel and the activity measured. The reaction vessel was
placed in the brass heater and the pot lid was tightened. Lines
from the pot were connected to the nitrogen supply and waste vial
and then the solution heated at 100 ◦C for 10 min to evaporate the
solvent. The vessel was cooled with compressed air before removal
of the pot lid and addition of HCl (6 M, 0.5 mL). A second pot lid
574 | Org. Biomol. Chem., 2009, 7, 564–575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
was secured with no lines to enable heating under pressure. The
heater was set at 120 ◦C and left for 5 min. After cooling to rt, the
pot lid was removed, water (1 mL) was added and the activity in
the reaction vessel measured. The reaction mixture was transferred
to a collection vial and once again activity in the new vial and
empty reaction vessel was measured. Reaction mixture (100 mL)
was added to the Dionex eluent [NaOH (50–52%)–water (600 mL)]
(4–5 mL) to give a solution with a pH of 7 2 ready for ion exchange
and reverse phase HPLC analysis. For further purification, the
neutralised solution was flushed through a conditioned C18 Sep-
Pak and the eluate analysed by reverse phase HPLC.
and B. Johannsen, Appl. Radiat. Isot., 1996, 47, 61; (c) C. Lemaire,
P. Damhaut, B. Lauricella, C. Mosdzianowski, J.-L. Morelle, M.
Monclus, J. Van Naeman, E. Mulleneers, J. Aerts, A. Plenevaux, C.
Brihayne and A. Luxen, J. Label. Compd. Radiopham., 2002, 45, 435;
(d) F. Fu¨chtner, J. Steinbach, P. Ma¨ding and B. Johannsen, Appl.
Radiat. Isot., 1996, 47, 61; (e) C. Lemaire, P. Damhaut, B. Lauricella,
C. Mosdzianowski, J.-L. Morelle, M. Monclus, J. Van Naeman, E.
Mulleneers, J. Aerts, A. Plenevaux, C. Brihayne and A. Luxen, J. Label.
Compd. Radiopham., 2002, 45, 435.
14 Y. Kuge, K. Nishijima, K. Nagatsu, K. Seki, K. Ohkura, A. Tanaka,
M. Sasaki, E. Tsukamoto and N. Tamaki, Nucl. Med. Biol., 2002, 29,
275.
15 (a) For a preliminary communication of this work: L. J. Brown, D. R.
Bouvet, S. Champion, A. M. Gibson, Y. Hu, A. Jackson, I. Khan, N.
Ma, N. Millot, H. Wadsworth and R. C. D. Brown, Angew. Chem.-Int.
Edit., 2007, 46, 941.
Labeling the resin 35 using the manual protocol described
above, incorporation yields were consistently in the range 73–
91% (reverse phase HPLC, average 80–85%), and deprotection
proceeded in 91–97% (ion exchange HPLC).
16 The solid-phase synthesis of [11C]-labeled N-methylated sulfonamides
and amines using has been reported using [11C]-methyl iodide and
safety-catch linkers: D. Maclean, J. Zhu, M. Y. Chen, R. Hale, N.
Satymurthy and J. R. Barrio, J. Am. Chem. Soc., 2003, 125, 10168.
17 (a) J. Jones, Amino acid and peptide synthesis 2nd ed.; Oxford University
Press, 2001; (b) W. Chan, P. White, Fmoc solid-phase peptide synthesis: a
practical approach, Oxford University Press, 1999; (c) M. H. Caruthers,
Science, 1985, 230, 281; (d) M. H. Caruthers, Acc. Chem. Res., 1991, 24,
278; (e) M. J. Davies, A. Shah and I. J. Bruce, Chem. Soc. Rev., 2000, 29,
97; (f) P. Seneci, Solid-phase synthesis and combinatorial technologies,
John Wiley & Sons Inc., 2000.
Acknowledgements
We thank the Royal Society for the award of a University Research
Fellowship (RCDB). We acknowledge the use of the EPSRC’s
Chemical Database Service at Daresbury.33
18 (a) B. H. Mock, M. T. Vavrek and G. K. Mulholland, Nucl. Med. Biol.,
1996, 23, 497; (b) J. S. Fowler, R. R. MacGregor, A. P. Wolf, A. A.
Farrell, K. I. Karlstrom and T. J. Ruth, J. Nucl. Med., 1981, 22, 376;
(c) K. Hamacher, G. Blessing and B. Nebeling, Appl. Radiat. Isot., 1990,
41, 49.
19 (a) For some examples of sulfonate ester linkers: fluoroalkylsulfonate
linkers: Y. J. Pan and C. P. Holmes, Org. Lett., 2001, 3, 2769; (b) Y.
Pan, B. Ruhland and C. P. Holmes, Angew. Chem. Int. Ed., 2001, 40,
4488; (c) Fluoroarylsulfonate linkers:A. N. Cammidge and Z. Ngaini,
Chem. Commun., 2004, 1914; (d) J. D. Revell and A. Ganesan, Chem.
Commun., 2004, 1916; (e) Aryl- and alkylsulfonate linkers: J. A. Hunt
and W. R. Roush, J. Am. Chem. Soc., 1996, 118, 9998; (f) M. S.
Shchepinov and D. A. Stetsenko, Bioorg. Med. Chem. Lett., 1997, 7,
1181; (g) J. K. Rueter, S. O. Nortey, E. W. Baxter, G. C. Leo and A. B.
Reitz, Tetraherdon Lett., 1998, 39, 975; (h) T. Takahashi, S. Ebata and
T. Doi, Tetrahedron Lett., 1998, 39, 1369; (i) I. Hijikuro, T. Doi and T.
Takahashi, J. Am. Chem. Soc., 2001, 123, 3716.
20 (a) For NMR spectra taken of the crude reaction mixture see supporting
information. Compounds [19F]-2, 6 and 7 gave spectroscopic data
consistent with the literature. Compound [19F]-2: P. Kovac, H. J. C. Yeh
and C. P. J. Glaudemans, Carbohydr. Res., 1987, 169, 23; (b) Compound
6: D. Y. Gagnaire, F. R. Taravel and M. R. Vignon, Carbohydr. Res.,
1976, 51, 157; (c) H. Shimizu, J. M. Brown, S. W. Homans and R. A.
Field, Tetrahedron, 1998, 54, 9489; (d) Compound 7: M. Hayashi, K.
Yamada and O. Arikita, Tetrahedron, 1999, 55, 8331.
Notes and references
1 (a) H. R. Herschman, Science, 2003, 302, 605; (b) A. Del Guerra, N.
Belcari, A. Motta, G. Di Domenico, N. Sabba and G. Zavattini, Nucl.
Instrum. Meth. A, 2003, 513, 13; (c) J. Czernin and M. E. Phelps, Annu.
Rev. Med., 2002, 53, 89; (d) R. Nutt, Mol. Imag. Biol., 2002, 4, 11; (e) R.
G. Blasberg and J. G. Tjuvajev, J. Clin. Invest., 2003, 111, 1620.
2 (a) H. Anderson and P. Price, Eur. J. Cancer, 2000, 36, 2028; (b) H.
A. Nabi and J. M. Zubeldia, J. Nucl. Med. Tech., 2002, 30, 3; (c) S. S.
Gambhir, Nature Revs., 2002, 2, 683.
3 (a) R. O. Robinson, C. D. Ferrie, M. Capra and M. N. Maisey, Arch.
Dis. Child., 1999, 81, 263; (b) R. S. Fisher and J. J. Frost, J. Nucl. Med.,
1991, 32, 651; (c) M. Iacoboni, J. C. Baron, R. S. J. Frackowiak, J. C.
Mazziotta and G. L. Lenzi, Clin. Neurophys., 1999, 110, 2.
4 M. N. Maisey, Nucl. Med. Comm., 2000, 21, 234.
5 J. Wang and L. Maurer, Curr. Top. Med. Chem., 2005, 5,
1053.
6 (a) For an excellent review of chemistry with [18F]fluoride ion: L. S. Cai,
S. Y. Lu and V. W. Pike, Eur. J. Org. Chem., 2008, 2853.
7 Methods for the synthesis of [18F]FDG and its application in PET have
been reviewed: B. Beuthien-Baumann, K. Hamacher, F. Oberdorfer
and J. Steinbach, Carbohydr. Res., 2000, 327, 107.
8 (a) For the synthesis of [18F]-FDG by electrophilic fluorination: T. Ido,
C. Wan, V. Casella, J. S. Fowler, A. P. Wolf, M. Reivich and D. E. Kuhl,
J. Label. Compd. Radiopharm., 1978, 14, 175; (b) J. S. Fowler, R. R.
MacGregor, A. P. Wolf, A. A. Farrell, K. I. Karlstrom and T. J. Ruth,
J. Nucl. Med., 1981, 22, 376; (c) For the synthesis of [19F]-FDG by
electrophilic fluorination: J. Adamson, A. B. Foster, L. D. Hall and R.
N. Johnson, Carbohyd. Res., 1970, 15, 351; (d) T. Ido, C. N. Wan, J. S.
Fowler and A. P. Wolf, J. Org. Chem., 1977, 42, 2341.
21 (a) K. Hamacher, Carbohydr. Res., 1984, 128, 291; (b) V. Pavliak and
K. Pavol, Carbohydr. Res., 1991, 210, 333.
22 Synthesis of mannose tetraacetate: V. Pozsgay, C. P. J. Glaudemans, J.
B. Robbins and R. Schneerson, Tetrahedron, 1992, 48, 10249.
23 P. J. Stang, M. Hanack and L. R. Subramanian, Synthesis, 1982,
85.
9 For a detailed discussion of methods for the synthesis of [18F]-FDG: T.
J. Tewson, Nucl. Med. Biol., 1989, 16, 533.
24 (a) R. K. Sehgal, B. Almassian, D. P. Rosenbaum, R. Zadrozny and S.
K. Sengupta, J. Med. Chem., 1988, 31, 790; (b) J. Murata, M. Tamura
and A. Sekiya, J. Fluor. Chem., 2002, 113, 97.
10 For a synthesis of [18F]-FDG using continuous-flow microreactors: C.
C. Lee, G. D. Sui, A. Elizarov, C. Y. J. Shu, Y. S. Shin, A. N. Dooley,
J. Huang, A. Daridon, P. Wyatt, D. Stout, H. C. Kolb, O. N. Witte, N.
Satyamurthy, J. R. Heath, M. E. Phelps, S. R. Quake and H. R. Tseng,
Science, 2005, 310, 1793.
25 B. Doboszewski, G. W. Hay and W. A. Szarek, Can. J. Chem., 1986, 65,
412.
26 D. R. Bundle, J. Chem. Soc.-Perkin Trans. 1, 1979, 2751.
27 K. Dax, B. I. Gla¨nzer, G. Schulz and H. Vyplel, Carbohydr. Res., 1987,
162, 13.
28 See ref. 19a,b and: C. P. Holmes, Y. J. Pan, 2002, WO 02/055026 A2.
29 G. B. Rong and R. Keese, Tetrahedron Lett., 1990, 31, 5615.
30 N. O. Brace, J. Org. Chem., 1975, 41, 766.
31 J. Adamson, A. B. Foster, L. D. Hall, R. N. Johnson and R. H. Hesse,
Carbohydr. Res., 1970, 15, 351.
32 K. Dax, B. I. Glanzer, G. Schulz and H. Vyplel, Carbohydr. Res., 1987,
162, 13.
11 K. Hamacher, H. H. Coenen and G. Sto¨cklin, J. Nucl. Med., 1986, 27,
235.
12 (a) For early approaches to the synthesis of labeled and unlabeled
FDG by nucleophilic fluorination. [19F]-FDG: J. Paca´k, Z. Toc´ıc and
M. Cerny, Chem. Commun., 1969, 77; (b) [18F]-FDG:P. A. Beeley, W.
A. Szarek, G. W. Hay and M. M. Perlmutter, Can. J. Chem.-Rev. Can.
Chim., 1984, 62, 2709; (c) S. Levy, E. Livni, D. Elmaleh and W. Curatolo,
J. Chem. Soc.-Chem. Commun., 1982, 972.
33 D. A. Fletcher, R. F. McMeeking and D. Parkin, J. Chem. Inf. Comput.
Sci., 1996, 36, 746.
13 (a) Acid deprotection: G. K. Mulholland, Nucl. Med. Biol., 1995,
22, 19; (b) base deprotection: F. Fu¨chtner, J. Steinbach, P. Ma¨ding
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 564–575 | 575
©