[18F]Ethenesulfonyl Fluoride as a Practical Radiofluoride Relay Reagent

Article abstract of DOI:10.1002/chem.201900930

Fluorine-18 is the most utilized radioisotope in positron emission tomography (PET), but the wide application of fluorine-18 radiopharmaceuticals is hindered by its challenging labelling conditions. As such, many potentially important radiotracers remain

Full text of DOI:10.1002/chem.201900930

A Journal of
Accepted Article
Title: [¹⁸F]Ethenesulfonyl fluoride as a novel radiofluoride relay reagent
Authors: Bo Zhang, Benjamin H Fraser, Mitchell A Klenner, Zhen
Chen, Steven H Liang, Massimiliano Massi, Andrea J
Robinson, and Giancarlo Pascali
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201900930
Link to VoR: http://dx.doi.org/10.1002/chem.201900930
Supported by

Chemistry - A European Journal
10.1002/chem.201900930
COMMUNICATION
[18
F]Ethenesulfonyl fluoride as a novel radiofluoride relay reagent
Bo Zhang[a],[b], Benjamin H. Fraser *, Mitchell A. Klenner[a],[c], Zhen Chen , Steven H. Liang ,
[a]
[d]
[d]
Massimiliano Massi , Andrea J. Robinson and Giancarlo Pascali[a],[e]*
[
c]
[b]
1
8
Abstract: Fluorine-18 is the most utilized radioisotope in Positron
Emission Tomography (PET), but the wide application of fluorine-18
radiopharmaceuticals is hindered by its challenging labelling
conditions. As such, many potentially important radiotracers remain
[ F]fluoride (i.e. instead of azeotropic drying), it is unclear
whether such reagents could be shipped conveniently for off-site
use on a wider range of precursors.
1
8
Thereafter, [ F]ethenesulfonyl fluoride (ESF) started to attract
our attention. [ F]ESF containing structures were recently
18
18
underutilized. Herein, we describe the use of [ F]ethenesulfonyl
fluoride (ESF) as a novel radiofluoride relay reagent that allows
radiofluorination reactions to be performed in minimally equipped
18
reported to release [ F]fluoride and therefore were not useful as
PET tracers. However, ESF is liquid at room temperature, and has
facile synthesis, simple purification and the feasibility to be
trapped in a cartridge.[12] Therefore, [ F]ESF might qualify as an
optimal radiofluoride relay reagent with the capability of remote
18
satellite nuclear medicine centres. [ F]ESF has a simple and reliable
production route and can be stored on inert cartridges. The cartridges
18
18
can then be shipped remotely and the trapped [ F]ESF can be
liberated by simple solvent elution. We have tested 18 radiolabelling
precursors, inclusive of model and clinically used structures, and most
precursors have demonstrated comparable radiofluorination
18
shipping. In this paper we test the application of [ F]ESF to the
radiofluorination of a variety of precursors, and compare the
radiochemical yields (RCYs) to those using conventionally dried
18
18
efficiencies to those obtained using a conventionally dried [ F]fluoride
source.
[ F]fluoride method.
1
8
The synthesis of [ F]ESF was performed in a microfluidic system
using 2,4,6-trichlorophenylethenesulfonate and non-dried
tetraethylammonium bicarbonate (TEAB)/[ F]fluoride complex,
and the produced [ F]ESF was distilled onto a Silica-plus
cartridge (Figure 1 & ESI). We usually produced 0.5–2.5 GBq
of [ F]ESF with an average RCY of 57±11% (n=24) and
radiochemical purity (RP) >95%. The cartridge was then
transported to the site of usage and eluted with chosen solvent.
18
Development of novel and efficient methods for incorporating
fluorine-18 radioisotope into pharmaceutically relevant structures
is of paramount importance for guaranteeing a wide access to
PET tracers.[1-2] The traditional approach of nucleophilic
1
8
[12]
18
18
substitution uses dried [ F]fluoride and can only be performed in
few centres with the appropriate level of equipment and expertise
18
Stability was assessed by measuring the RP of the [ F]ESF
(
e.g. cyclotron, synthesizers and bulky hot cells).[3] On the other
eluted from a dedicated cartridge after 4 h and the value remained
unchanged. Alternatively, several Silica-light cartridges could be
used to distribute the radioactivity over these supports, that could
then be eluted with different solvents, providing further
experimental flexibility. We believe the trapping of [ F]ESF onto
the Si matrix is mediated by non-covalent interactions involving
9
9m
hand, in most nuclear medicine departments where
Tc is used,
“
shake & bake” labelling reactions are performed using minimal
equipment,[ thus providing access to a wide array of 99mTc
4]
radiotracers.[5] Similar protocols can also be applied to other
radiometal labellings (e.g. Ga, Cu), but such simplicity has not
yet been achieved for C- F labellings.
18
68
64
1
8
[6-8]
1
8
the double bound, therefore making [ F]ESF the smallest sulfonyl
fluoride species with useful trapping features.
To achieve such an important target, we focused on fluoride relay
reagents,[9] where reactive fluoride species can be released
following simple chemical or physical interaction. This interesting
concept was reported recently by Pees, et al.[10] using [ F]triflyl
fluoride, and previously by DeGrado’s group[11] using [ F]acetyl
fluoride. Both reagents were produced as gases and used
immediately in their respective radiofluorination reactions.
DeGrado’s group also reported the feasibility of trapping
[13]
The scale of
activity we used was intended to mimic levels that could be
employed by a clinical radiopharmacy to obtain few doses of a
particular 18F radiopharmaceutical, or could be easily integrated
in existing minimalistic systems already available to produce
18
18
99mTc, 68
Ga or other radiometal-based radiopharmaceuticals.[14-16]
As a limit to this approach, the use of low activity did not allow to
detect any UV signal for molar activity calculations; however, an
estimation of molar activity obtained using no carrier added
[18
F]acetyl fluoride in a cartridge but limited information was given
and the reaction scope was relatively narrow. While these
approaches may potentially represent simpler ways to activate
[18
F]ESF can be found in previous literature.[12]
In this work, our scope was to assess the feasibility of using
[18
F]ESF on radiofluorinating a wide variety of precursors.
Therefore, we did not optimize the RCY for each precursor, but
instead looked for the impact of experimental conditions on RCYs.
It is worth noticing that dedicated optimization studies would be
needed for each precursor to obtain the best RCY.
[
a]
Mr. B. Zhang, Dr. B.H. Fraser*, Mr. M.A. Klenner, Dr. G. Pascali*
Australian Nuclear Science and Technology Organisation (ANSTO)
New Illawarra Rd, Lucas Heights, NSW, Australia
*
E-mail: bfr@ansto.gov.au, gianp@ansto.gov.au
[b]
[c]
[d]
[e]
Mr. B. Zhang, Prof. A.J. Robinson
Faculty of Science, Monash University
Twelve precursors (Figure 1) were initially chosen to test the
radiofluorination reactions with [ F]ESF, of which 1-5 are model
aromatic compounds, while 6-12 represent a set of commonly
18
Wellington Rd, Clayton, Victoria, Australia
Mr. M.A. Klenner, Prof. M. Massi
School of Molecular and Life Sciences, Curtin University
Kent St, Bentley, Western Australia, Australia
Mr. C. Zhen, Prof. S.H. Liang
used radiotracer precursors for PET imaging. In a typical test,
1
8
[ F]ESF was eluted from the cartridge to obtain a stock solution
in CH CN with a radioactivity concentration of 100–250 MBq/mL.
3
Massachusetts General Hospital and Harvard Medical School
An aliquot (200 µL, 20–50 Mbq) of this radiolabelling solution was
then added to 800 µL of a premade solution or suspension of the
55 Fruit St, Boston, MA, USA
Dr. G. Pascali
Brain and Mind Centre, The University of Sydney
Mallett St, Camperdown, NSW, Australia
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900930
COMMUNICATION
Figure 1 Tested precursors and best RCYs obtained, fluorination sites in red. General procedure: 200 µL [18F]ESF in CH
3
CN was added to 800 µL of precursor
3
and TEAB in solvent and heated for 15 min. Conditions: 3, 5-9, and 11 (2mg/mL in CH CN, T=100°C), 10 (2 mg/mL in DMSO, T=100°C), 1, 4 and 12 (1 mg/mL in
DMSO, T=130°C).
desired precursor inside a single-use 4 mL glass vial. A suitable
amount of TEAB was also added to release and activate the
F]fluoride. Alternatively, it is also possible to elute each
was used to calculate the statistical significance between different
sets of reactions. Precursor 2 was subjected to all the reaction
sets, but no product was observed in any conditions.
[18
cartridge directly with chosen solvent into a premade vial
containing the dry precursors and TEAB. The reaction mixture
was then magnetically stirred and heated using standard
laboratory equipment. Each experiment was typically repeated
three times and the RCY was calculated by analysing the reaction
mixture with HPLC and correcting for the efficiency of radioactivity
recovery of the liquid from the reaction vial (RCY = RCYHPLC
Impact of TEAB was tested by varying its concentration from 0.01
to 0.50 mg/mL using a fixed amount of 11 (1 mg/mL). We found
that TEAB was essential for the reaction, both acting as a base
1
8
and a phase transfer catalyst to convert [ F[ESF into reactive
18
species [ F]TEAF. In the absence of TEAB, Michael addition
reaction could happen on free amines, as for substrate 12 at high
temperatures; however, we verified that post-addition of TEAB to
the same mixture successfully results in radiofluorination. The use
×
%Recovery, see ESI). This correction, together with the use of a
[
17]
monolithic HPLC column, allowed us to account for potential
activity lost in the apparatus, therefore better estimating real-
world production yields. As for the RCYs of 1 and 12, we
combined the RCYs for fluorination and for the radioproduct
resulting from, respectively, dissociation of Re (unpublished work)
and Boc deprotection.[18-19] A two-tail unpaired two-sample t-test
3
of different bases (i.e. trimethylamine and KHCO ) did not result
in any radiofluorination. Good RCYs of 66% were already
obtained using 0.05 mg/mL of TEAB; however, in order to simplify
the operative parameters and to account for potential differences
in precursor reactivity, we chose 0.5 mg/mL for the whole set of
precursors. Such amount could be, in some cases, too basic for
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900930
COMMUNICATION
some precursors and would need to be tuned appropriately in
dedicated optimization studies.
Reaction time was next evaluated on a subset of precursors and
RCYHPLC were recorded at different time points (5, 10, 15 and 30
min, see ESI), sampling the reaction mixture each time after
quenching via ice cooling. For most of the precursors, longer
consistent quality of this labelling reagent compared to the
1
8
[ F]TEAF coming from a drying process that, even if automated,
might introduce varying grades of moisture or other contaminants
on each day it was performed, or due to the dissolution of the
dried reagent back into solution.
18
We then moved our interest towards the use of [ F]ESF for
radiolabelling emerging precursors with novel leaving groups (i.e.
reaction times led to improved RCYHPLC; however, such
improvement would not always generate more final product due
to radioactivity decay. In addition, ice cooling was essential to
minimize the chances of producing hazardous radioactive
volatiles; however, this operation did not warrant reaching the
target temperature for all the duration of the time point tested.[20]
We deemed that a reaction time of 15 min would be utilizable for
obtaining acceptable radioactive incorporation with most of the
precursors.
The effect of reaction temperature was tested on all the
precursors under 3 conditions: 70, 100 and 130 °C (see ESI).
Most precursors had very low RCYs (<10%) at 70 °C, while the
18
18
precursors for [ F]FDG and [ F]fallypride (6 and 7) had RCYs
30%. RCYs were always statistically improved from 70 to 100 °C,
>
but when reaching 130 °C, 5-9 and 11 provided lower RCYs. This
is probably due to the decomposition of precursors at such higher
temperature; in addition, we employed DMSO for all the reactions
at 130 °C, even for the precursors which we previously
investigated in CH CN (3-9 and 11), and such change could have
3
also contributed to different RCYs.
Figure 2 Comparative fluorination RCYs between [18F]ESF and traditionally
dried [18F]TEAF at precursor concentration of 1 mg/mL, reaction time of 15 min
and T=100 °C (n = 3), Whereas indicated, statistical significance was calculated
by a two-tail unpaired two sample t-test (* p<0.05, ** p<0.005, ***p<0.0005).
Lastly, we studied the effect of precursor amount using
concentrations of 0.5, 1 or 2 mg/mL, while the TEAB amount was
kept at half of concentrations of the precursors (see ESI). The
variation of RCYs due to this parameter was highly dependent on
the precursors tested, with 6, 7, 9, 10 and 12 being the only ones
showing statistically significant increases in RCYs when rising
both from 0.5 to 1 mg/mL and from 1 to 2 mg/mL. The highest
RCYs obtained and relative reaction conditions for all precursors
after testing are summarised in Figure 1.
boronic acids and iodonium ylides, Figure 3).[21-23] For boronic
acid reactions, we started using 800 µL of 1 mg/mL of 13 in
dimethylacetamide (DMA) added with 5 eq of Cu(OTf)
pyridine and 0.5 mg TEAB, and reacted with 200 µL of [ F]ESF
2
, 125 eq of
18
3
in CH CN at 110 °C for 20 min. Under these conditions, we
obtained RCYs significantly lower than the ones reported in
literature.[21] Given the potential impact of basic conditions on this
type of reaction, we investigated decreasing the TEAB amount
and realized that radiofluorination occurred even in absence of
TEAB, albeit with the same low RCYs (see ESI). The effect of
solvent was next tested, and we found that higher RCYs were
Bigger scale reactions employing ~100 MBq were also conducted
on 7-9, 11 and 12. For 7 and 8, the reaction mixture was passed
through an alumina column and RCYs of 32% and 57% were
obtained. For 9, 11, and 12, radiofluorination did not provide the
final product as additional steps were required to synthesize the
target molecule. For example, 11 required two extra steps of a
hydrolysis and a conjugation reaction to give the final product,
obtained in absence of CH
3
CN (i.e. in 100% DMA or DMF). In a
last attempt to further improve RCYs, the effects of Cu(OTf)
2
,
pyridine and precursor concentration were investigated. We found
a slightly increased RCY when 300 eq of pyridine were used.
1
8
[
F]SFB, in 52% RCY (see ESI). We have also successfully
However, increasing the amount of Cu(OTf)
2
or decreasing
tested acid hydrolysis for 9 and high-temperature deprotection for
amount of precursor did not provide any improvement. We
therefore proceeded with the testing on all the precursors by
1
8
1
2, which further indicated that [ F]ESF conditions do not
interfere with subsequent additional reactions.
18
reacting DMF eluted [ F]ESF with 1 mg/mL precursor solutions
Comparative radiolabelling reactions using traditionally dried
2
in DMF, with 5 eq of Cu(OTf) , 125 eq of pyridine and without
TEAB, heated at 110°C for 20 min. The RCYs obtained were
[18
homogenous
F]tetraethylammonium fluoride (TEAF) were conducted using a
set of conditions (same solvent,
reported in Figure 3 and were compared to equivalent reactions
tetraethylammonium cation, precursor at 1 mg/mL, reaction time
18
performed using traditionally dried [ F]TEAF. For all the reported
of 15 min, T=100 °C, see ESI). As shown in Figure 2, the RCYs
18
precursors, the RCYs obtained using [ F]ESF were lower than
1
8
using [ F]ESF method were comparable to the use of traditionally
18
the ones obtained using [ F]TEAF. However, we were surprised
1
8
dried [ F]fluoride complex. In particular, the statistical test
indicated that 6 precursors showed insignificant difference in
that this latter yield was drastically higher (18% in average) than
equivalent literature values. This fact might be due to the use of
1
8
RCYs, while for 4 precursors (4, 5, 10 and 12), the [ F]ESF
method provided higher RCYs; in the case of 3, the traditional
[18
F]TEAF instead of K2.2.2-based complexes reported previously,
to different automated drying sequences or to other currently
uncontrolled parameters. This could suggest that a deeper
1
8
dried [ F]TEAF provided a higher RCY. It was also observed that
the standard deviations for these radiofluorinations were
18
generally lower when using [ F]ESF, which could be linked to the
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900930
COMMUNICATION
understanding of the crucial conditions for such Cu-mediated
radiofluorinations is still needed.[24]
Financial supports from Monash University (MGS and MIPRS)
and The Australian Institute of Nuclear Science and Engineering
(
AINSE PGRA 12074) are gratefully acknowledged. We also
1
8
thank the ANSTO cyclotron team for providing [ F]fluoride, and
Dr. Tien Pham for his useful comments on manuscript editing.
Keywords: fluorine-18 • sulfonyl fluoride • radiofluorination •
PET • ESF
[
[
[
1]
2]
3]
D. van der Born, A. Pees, A. J. Poot, R. V. A. Orru, A. D. Windhorst, D.
J. Vugts, Chem. Soc. Rev. 2017, 46, 4709-4773.
G. Pascali, L. Matesic, B. Zhang, A. T. King, A. J. Robinson, A. T. Ung,
B. H. Fraser, EJNMMI radiopharm chem 2017, 2, 9.
M. Lazari, K. M. Quinn, S. B. Claggett, J. Collins, G. J. Shah, H. E.
Herman, B. Maraglia, M. E. Phelps, M. D. Moore, R. M. van Dam,
EJNMMI Res 2013, 3, 52.
[
[
4]
5]
T. B. Saleh, in Basic Sciences of Nuclear Medicine (Ed.: M. M. Khalil),
Springer Berlin Heidelberg, Berlin, 2011, pp. 41-53.
P. Y. Keng, M. Esterby, R. M. Van Dam, in Positron Emission
Tomography - Current Clinical and Research Aspects (Ed.: C.-H. Hsieh),
IntechOpen, 2012.
[
[
6]
7]
D. Satpati, A. Shinto, K. K. Kamaleshwaran, S. Sane, S. Banerjee, Mol
Imaging Biol 2016, 18, 420-427.
V. Bernard-Gauthier, M. L. Lepage, B. Waengler, J. J. Bailey, S. H. Liang,
D. M. Perrin, N. Vasdev, R. Schirrmacher, J Nucl Med 2018, 59, 568-572.
R. Schirrmacher, A. Kostikov, C. Wängler, K. Jurkschat, V. Bernard‐
Gauthier, E. Schirrmacher, B. Wängler, in Radiochemical Syntheses
Figure 3 RCYs of novel types of precursors using [18F]ESF and dried [18F]TEAF
under comparable conditions, fluorination sites are shown in red.
[8]
(Ed.: P. J. Scott), John Wiley & Sons, Inc., 2015.
Radiofluorination reactions employing iodonium ylides were
[9]
H. Sun, S. G. DiMagno, ChemComm 2007, 528-529.
1
8
performed at 120 °C for 15 min using [ F]ESF in CH
3
CN and 17
[10] A. Pees, C. Sewing, M. J. W. D. Vosjan, V. Tadino, J. D. M. Herscheid,
and 18 in toluene and DMF respectively (see ESI). For these
A. D. Windhorst, D. J. Vugts, ChemComm 2018.
[
[
11] H. Jiang, S. G. DiMagno, T. R. DeGrado, J Fluor Chem 2015, 180, 181-
185.
substrates, RCY of 17 (64±2%) was higher than the literature
value (46%)[
23]
while RCY of 18 (5±1%) was lower than the
12] B. Zhang, G. Pascali, N. Wyatt, L. Matesic, M. A. Klenner, T. R. Sia, A.
J. Guastella, M. Massi, A. J. Robinson, B. H. Fraser, J Labelled Comp
Radiopharm 2018, 61, 847-856.
literature value (25%).[25] This fact was not surprising, as iodonium
reactivity can vary on each batch of precursor. In our setting,
18
RCYs using [ F]ESF were similar or better than using dried
[
13] E. Papirer, Adsorption on Silica Surfaces, Taylor & Francis, 2000.
1
8
[
F]TEAF.
[14] R. Vis, J. Lavalaye, E. M. van de Garde, EJNMMI Res 2015, 5, 27.
[15] M. D. Decker, J. H. Turner, Cancer Biother Radiopharm 2012, 27, 72-76.
[16] M. Asti, G. Atti, M. Iori, D. Farioli, A. Filice, A. Versari, Nucl Med Commun
2012, 33, 1144-1152.
1
8
In conclusion, we have demonstrated that [ F]ESF is a simple to
use radiofluoride relay reagent, providing RCYs comparable to
1
8
dried [ F]TEAF in most cases, but greatly reducing the reaction
equipment needed, in the simplest case to a heating/stirring
aluminium block and single-use vials. Currently, fluorine-18 is
shipped either in an aqueous fluoride solution or inside an anion-
[
17] D. Ory, J. Van den Brande, T. de Groot, K. Serdons, M. Bex, L. Declercq,
F. Cleeren, M. Ooms, K. Van Laere, A. Verbruggen, G. Bormans, J.
Pharm. Biomed. Anal. 2015, 111, 209-214.
[
18] T. M. Shoup, D. L. Yokell, P. A. Rice, R. N. Jackson, E. Livni, K. A.
Johnson, T. J. Brady, N. Vasdev, J Label Compd Radiopharm 2013, 56,
736-740.
1
8
exchange cartridge. Unlike [ F]ESF cartridges, these forms
would require additional drying steps, and cannot be used for
radiofluorination in minimally equipped hospital or academia
settings. As of now, in our simplified labelling approach,
chromatographic purification and formulation is still needed to
obtain the desired product. However, we believe that the ease
[19] M. A. Klenner, G. Pascali, B. Zhang, T. R. Sia, L. K. Spare, A. M. Krause-
Heuer, J. R. Aldrich-Wright, I. Greguric, A. J. Guastella, M. Massi, B. H.
Fraser, Chemistry 2017, 23, 6499-6503.
[
[
[
20] M. Lazari, J. Irribarren, S. Zhang, R. M. van Dam, Appl. Radiat. Isot. 2016,
08, 82-91.
1
1
8
introduced by [ F]ESF fluorination process would justify the
investigation into novel kit- or cartridge-based solutions. If such
scenario will be realized, the miniaturization and compactification
of quality control tests would provide the final tile to achieve an
unprecedented flexibility of access to 18F radiopharmaceuticals,
thus providing the most personalized diagnoses possible.
21] A. V. Mossine, A. F. Brooks, K. J. Makaravage, J. M. Miller, N. Ichiishi,
M. S. Sanford, P. J. Scott, Org. Lett. 2015, 17, 5780-5783.
22] J. Cardinale, J. Ermert, S. Humpert, H. H. Coenen, RSC Advances 2014,
4, 17293-17299.
[23] B. H. Rotstein, L. Wang, R. Y. Liu, J. Patteson, E. E. Kwan, N. Vasdev,
S. H. Liang, Chem Sci 2016, 7, 4407-4417.
[
[
24] X. Zhang, F. Basuli, R. E. Swenson, J Labelled Comp Radiopharm 2019.
25] B. H. Rotstein, N. A. Stephenson, N. Vasdev, S. H. Liang, Nat Commun
2014, 5, 4365.
Acknowledgements
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900930
COMMUNICATION
Entry for the Table of Contents
COMMUNICATION
[18
F]ESF can be trapped into
B. Zhang, B.H. Fraser*, M.A.
inert cartridges and released
to provide a ready-to-use
labelling solution, reducing the
radiochemistry equipment
needs.
Klenner, Z. Chen, S.H. Liang, M.
Massi, A.J. Robinson, G. Pascali*
Page No. – Page No.
[18
F]Ethenesulfonyl fluoride as
a novel radiofluoride relay
reagent
This article is protected by copyright. All rights reserved.