[18F]FPyZIDE: A versatile prosthetic reagent for the fluorine-18 radiolabeling of biologics via copper-catalyzed or strain-promoted alkyne-azide cycloadditions

Article abstract of DOI:10.1002/jlcr.3701

Methods for the radiolabeling of biologics with fluorine-18 have been of interest for several decades. A common approach consists in the preparation of a prosthetic reagent, a small molecule bearing a fluorine-18 that is conjugated with the macromolecule to an appropriate function. Click chemistry, and more particularly cycloadditions, is an interesting approach to radiolabel molecules thanks to mild reaction conditions, high yields, low by-products formation, and strong orthogonality. Moreover, the chemical functions involved in the cycloaddition reaction are stable in the drastic radiofluorination conditions, thus allowing a simple radiosynthetic route to prepare the prosthetic reagent. We report herein the radiosynthesis of ¹⁸F-FPyZIDE, a pyridine-based azide-bearing prosthetic reagent. We exemplified its conjugation via copper-catalyzed cycloaddition (CuAAC) and strain-promoted cycloaddition (SPAAC) with several terminal alkyne or strained alkyne model compounds.

Full text of DOI:10.1002/jlcr.3701

1
8
[
F]FPyZIDE : a versatile prosthetic reagent for the fluorine-18 radiolabeling of
biologics via copper-catalyzed or strain-promoted alkyne-azide cycloadditions
Mélanie Roche, Simon Specklin, Mylène Richard, Françoise Hinnen, Kevin Génermont,
Bertrand Kuhnast*
IMIV, Service Hospitalier Frédéric Joliot, CEA, Inserm, Université Paris Sud, CNRS,
Université Paris-Saclay, Orsay, France
*
Corresponding author: Bertrand Kuhnast, CEA, 4 place du général Leclerc, 91401 Orsay
Cedex, France, bertrand.kuhnast@cea.fr; + 33 1 69 86 77 36
ORCID iD Bertrand Kuhnast : 0000-0002-5035-4072
Abstract
Methods for the radiolabeling of biologics with fluorine-18 have been of interest for several
decades. A common approach consists in the preparation of a prosthetic reagent, a small
molecule bearing a fluorine-18 that is conjugated with the macromolecule to an appropriate
function. Click chemistry, and more particularly cycloadditions, are an interesting approach
to radiolabel molecules thanks to mild reaction conditions, high yields, low by-products
formation and strong orthogonality. Moreover, the chemical functions involved in the
cycloaddition reaction are stable in the drastic radiofluorination conditions, thus allowing a
simple radiosynthetic route to prepare the prosthetic reagent. We report herein the
1
8
radiosynthesis of F-FPyZIDE, a pyridine-based azide-bearing prosthetic reagent. We
exemplified its conjugation via copper-catalyzed cycloaddition (CuAAC) and strain-
promoted cycloaddition (SPAAC) with several terminal alkyne or strained alkyne model
compounds.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/jlcr.3701
This article is protected by copyright. All rights reserved.

Introduction
Biologics-based radiopharmaceuticals have been shown for several decades to be valuable
tools for diagnostic imaging and vectorized radiotherapy. Biologics may be defined as
components that are produced by living organisms or resulting from chemical manufacturing
or biotechnology while mimicking biological compounds. They comprise peptides, proteins,
polysaccharides, nucleic acids but also a wide range of cellular and tissue components. In the
field of Positron Emission Tomography (PET) imaging, peptides, polysaccharides or nucleic
1
-5
acids are key-actors in the toolbox of radiotracers and radiopharmaceuticals . Their
radiolabeling with a positron emitter is usually achieved either via the conjugation of a
6
-8
fluorine-18-labeled prosthetic reagent or via their functionalization with a bifunctional
chelating agent prior to the complexation of a radiometal (gallium-68, copper-64 or
9
-12
zirconium-89).
Other options taking advantage of the silicon-fluorine, boron-fluorine or
1
3-19
aluminum-fluorine bond have been described and are increasingly exploited.
Nonetheless, the prosthetic approach remains a strategy of choice due to its robustness and
reliability as witnessed by the numerous reports published, but there is still room available for
innovation in this fruitful field. Besides the acylation, S-alkylation or oxime formation
7
, 8, 20
reactions that have been largely exploited,
the emergence of the concept of click
2
1-23
chemistry about 15 years ago revived the design of prosthetic reagents.
This concept
encompasses a wide range of chemical reactions characterized by high yields, high reaction
rates, moderate temperatures and a large tolerance of solvents. The most emblematic reaction
2
4
within this family is the copper-catalyzed alkyne-azide cycloaddition (CuAAC), a copper-
catalyzed variant of the Huisgen’s cycloaddition. This reaction is particularly well adapted to
the prosthetic radiofluorination of biologics in many respects: i) the mild reaction conditions
do not threaten the chemical integrity of the biologics; ii) the high conjugation rates and
yields are compatible with the half-life of fluorine-18 (t1/2 = 109.8 minutes); iii) the strong
orthogonality of the chemical moieties involved in the cycloaddition prevents side-reactions
with amines, alcohols, carboxylic acids and other functions borne by the biologics; and iv)
the chemical functions involved in the CuAAC (azide or alkyne) are stable toward the drastic
reaction conditions for the radiofluoration
In spite of these numerous assets, the need of copper (generally as CuSO /sodium ascorbate)
4
to promote the cycloaddition reaction, via in situ generation of Cu(I), is perceived as a
drawback of the CuAAC. It indeed prevents direct in vivo applications of the reaction and it
2
5
is recognized that reactive oxygen species (ROS) may be produced. These species could
induce molecular damages during the chemical functionalization of biologics like DNA or
RNA and also in living cell. It has been reported that nucleobases can undergo oxidative
2
6
modification in presence of copper leading to DNA or RNA scission. Recently, a
methodological study tended to measure the degree of copper-mediated oxidation on a library
2
7
of 1200 tetrapeptides and a model protein. The use of catalytic amounts of copper(I)
2
8
29
stabilized with appropriate ligands, nanoparticular catalytic systems, or even pseudo-
3
0, 31
ligandless reaction by a tight control of the copper amount and presence of acetonitrile
may limit such drawbacks of the CuAAC reaction. As a clever alternative to CuAAC, the
strain-promoted alkyne-azide cycloaddition (SPAAC) reaction,
3
2, 33
34
is also widely used. Its
reactivity relies on lower energy of activation for the reaction of strained cyclooctynes with
azides. SPAAC reaction shows equivalent advantages to CuAAC in terms of orthogonality,
This article is protected by copyright. All rights reserved.

yields and stability of strained alkyne in radiofluorination conditions, even if its reaction rate
3
5
is slower.
The compatibility of terminal alkynes, azides or even cyclooctynes with the standard
radiofluorination conditions allows the preparation of the radiofluorinated prosthetic reagent
in only one radiochemical step, contrary to the standard acylating or alkylating prosthetic
reagents often presenting sensitive activated esters or maleimides. Their preparation thus
requires multi-step processes that are limiting regarding the half-life of fluorine-18 and the
automation access. Consequently, many alkyne- or azide-containing tags have flourished in
1
8
18
18
the literature displaying [ F]fluoroalkyl, [ F]fluoraryl, [ F]fluoroheteroaryl or
18
7, 22, 23, 36-42
[
F]fluorocarbohydrate motives.
18
On the basis of our experience in the preparation of [ F]fluoropyridine-based prosthetic
4
3-45
42, 46, 47
reagents
and analogously to already developed clickable ones,
we report the
1
8
preparation of a novel azide-containing prosthetic reagent ([ F]FPyZIDE). Its versatility in
cycloaddition reactions has been illustrated using a propargylated cysteine derivative and two
cyclooctyne-models. For both CuAAC and SPAAC, reaction conditions (time, temperature,
concentrations, catalytic system for CuAAC) have been modulated to evaluate the
performances of [ F]FPyZIDE regarding these two cycloadditions and to establish the best
reaction compromises regarding fluorine-18 constraints.
1
8
Experimental
1
1
. General
.1 Chemicals
Chemicals were purchased from Sigma-Aldrich (France) and were used without further
purification.
1
1
.2 Chromatographies.
.2.1 High Pressure Liquid Chromatography.
High Pressure Liquid Chromatography (HPLC) was performed on the following systems:
[HPLC A]: Semi-preparative equipment integrated to the TRACERLab FX-FN/FX N Pro
(GE Medical Systems): S1122 Solvent Delivery System (Sykam); U.V. Detector K-2501
(Knauer); radioactivity  detector; column Zorbax SB C18, particle size 5 µm, 250x9.4 mm
(Agilent);
Mobile Phase 1 (MP-1): H O/MeOH/THF/H PO : 650 mL / 175 mL / 175 mL / 0.1 mL
2
3
4
(v/v/v/v); flow rate: 5 mL/min; detection  = 254 nm; temperature: rt.
Mobile Phase 2 (MP-2): H O/MeCN/TFA: 700 mL / 300 mL / 0.1 mL (v/v/v); flow rate: 5
2
mL/min; detection  = 254 nm; temperature: rt.
[
HPLC B]: Waters Alliance 2690 equipped with a UV spectrophotometer (Photodiode Array
Detector, Waters 996 (Waters)) and a Berthold LB509 radioactivity detector; column:
®
analytical Symmetry-M C-18, 50 x 4.6 mm, 5 µm (Waters); solvent A : H O containing
2
®
Low-UV PIC B7 reagent (20 mL for 1000 mL), solvent B: H O / CH CN: 30:70 (v/v)
2
3
This article is protected by copyright. All rights reserved.

®
containing Low-UV PIC B7 reagent (20 mL for 1000 mL), isocratic elution; flow rate: 2.0
mL/min; U.V. detection at λ = 254 nm; temperature: rt.
Mobile Phase 3 (MP-3): A/B 60/40 (v/v).
Mobile Phase 4 (MP-4): A/B 70/30 (v/v).
Mobile Phase 5 (MP-5): A/B 45/55 (v/v).
1
.2.2 Thin Layer Chromatography.
Thin Layer Chromatography (TLC) was performed on pre-coated plates of silica gel 60F254
Merck) and eluted with appropriate mixtures of organic solvents. Non-radioactive
compounds were localized at 254 nm using a UV-lamp and/or by dipping the TLC-plates in a
% ethanolic ninhydrin solution or a basic potassium permanganate aqueous solution and
heating on a hot plate.
(
1
1
.2.3 Column Chromatography.
Column chromatographies were conducted on silica gel 63-200 µm (Merck) at 0.3 bars
compressed air.
1
.3 Spectroscopies.
1
13
H and C NMR spectra were recorded on a Avance 400 MHz apparatus (Bruker), and
chemical shifts were referenced to the hydrogenated residue of the deuterated solvent
1
(
δ[CHCl ] = 7.24 ppm) for H NMR and to the deuterated solvent (δ[CDCl ] = 77.2 ppm) for
3
3
1
3
C NMR experiments. The chemical shifts are reported in ppm, downfield from TMS (s, d, t,
q and m for singlet, doublet, triplet, quadruplet and multiplet respectively). The mass
spectrometry analyses were performed by electrospray with positive (ESI+) or negative
(ESI−) ionization mode using a LTQ Velos Pro mass spectrometer (Thermo Scientific) and
XCalibur 2.2 software (Thermo Scientific).
1
.4 Radioisotope availability and handling.
1
8
18
18
Fluorine-18, as no-carrier-added aqueous [ F]fluoride ion was produced via the O(p,n) F
18
nuclear reaction by irradiation of a 2 mL [ O]water target (> 97%-enriched, CortecNet) on
an Cyclone-18/9 cyclotron (18 MeV proton beam, IBA) and was transferred to the
appropriate hot cell. Target hardware: commercial, 2-mL, two-port, stainless steel target
holder equipped with a domed-end niobium cylinder insert. Target to hot cell liquid-transfer
system : 60 m PTFE line (0.8 mm internal diameter ; 1/16 inch external diameter), 2.0 bar
1
8
helium drive pressure, transfer time 3-6 minutes. Typical production of [ F]fluoride ion at
the end of bombardment for a 20 µA, 30 minutes irradiation: 27-30 GBq. Radiosyntheses
using fluorine-18, including the HPLC purifications, were performed in a 5 cm lead shielded
cell using a TRACERLab FX-FN or FX N Pro (GE Medical Systems).
2
2
. Chemistry
-(2-(2-Azidoethoxy)ethoxy)ethan-1-ol
(2).
To
a
solution
of
2-(2-(2-
chloroethoxy)ethoxy)ethan-1-ol (1) (4.31 mL, 29.6 mmol) in water (25 mL) was added
sodium azide (3.85 g, 59.3 mmol) and the resulting mixture was stirred 24 h at 75 °C. After
cooling to room temperature, water was removed under vacuum and co-evaporated with
This article is protected by copyright. All rights reserved.

toluene (3x25 mL). Et O (50 mL) was added to the residue and the resulting suspension was
2
filtrated. The filtrate was concentrated under vacuum to afford 2 (5.22 g, 91%) as a colourless
1
oil. Rf (dichloromethane/MeOH 9:1): 0.63. H NMR (CDCl , 400 MHz): δ 3.74-3.64 (m,
8
3
3
3
13
H); 3.60 (t, J=4.5 Hz, 2H); 3.38 (t, J=5.0 Hz, 2H); 2.51 (bs, 1H). C NMR (CDCl , 100
3
MHz): δ 72.6 (CH ); 70.7 (CH ); 70.5 (CH ); 70.1 (CH ); 61.8 (CH ); 50.7 (CH ). ESI(+)-
2
2
2
2
2
2
+
+
MS m/z 198 [M+Na] (Calculated for C H N O Na : 198.1).
6
13
3
3
2
5
-(2-(2-azidoethoxy)ethoxy)ethyl 4-methylbenzenesulfonate (3). To a solution of 2 (1.00 g,
.71 mmol) in anhydrous dichloromethane (40 mL) were added p-toluenesulfonyl chloride
(1.19 g, 6.28 mmol), triethylamine (0.87 mL, 6.28 mmol) and dimethylamino pyridine (0.07
g, 0.57 mmol). The mixture was stirred overnight at room temperature. A solution of
saturated NH Cl (40 mL) was added and the mixture was extracted with dichloromethane
4
(
2x50 mL). The organic layers were combined, dried over Na SO and concentrated under
2 4
1
vacuum to afford 3 (1.37 g, 73%) as a pale yellow oil. Rf (heptane/EtOAc 1:1): 0.27. H
3
3
NMR (CDCl , 400 MHz): δ 7.78 (d, J=8.4 Hz, 2H); 7.33 (d, J=8.4 Hz, 2H); 4.14 (t,
3
3
3
3
J=4.8Hz, 2H); 3.68 (t, J=4.8 Hz, 2H); 3.62 (t, J=5.1 Hz, 2H); 3.59-3.57 (m, 4H); 3.34 (t,
3
13
J=5.1 Hz, 2H); 2.43 (s, 3H). C NMR (CDCl , 100 MHz): δ 144.9 (C); 133.0 (C); 129.9
3
(
2xCH); 128.0 (2xCH); 70.8 (CH ); 70.7 (CH ); 70.1 (CH ); 69.3 (CH ); 68.8 (CH ); 50.7
2
2
2
2
2
+
+
(CH ); 21.7 (CH ). ESI(+)-MS m/z 352 [M+Na] (Calculated for C H N O SNa : 352.1).
2
3
13 19
3
5
3
-(2-(2-(2-Azidoethoxy)ethoxy)ethoxy)-2-fluoropyridine (5). To a solution of 2-fluoro-3-
hydroxypyridine (4) (361 mg, 3.19 mmol) in anhydrous DMF (15 mL) was added potassium
carbonate (588 mg, 4.25 mmol). The mixture was stirred 20 min at room temperature under
argon and a solution of 3 (700 mg, 2.2 mmol) in DMF (5 mL) was added to the mixture. The
solution was stirred overnight at 90 °C. After cooling to room temperature, water was added
(50 mL) and the resulting solution was extracted with EtOAc (2x20 mL). The organic layers
were combined and successively washed with 0.1 M sodium hydroxide (2x30 mL), water (20
mL), 1 M HCl (20 mL), brine (20 mL), dried over Na SO and concentrated under vacuum.
2
4
The residue was purified by column chromatography (silica gel, heptane/EtOAc 7:3) to
1
afford 5 (646 mg, 75%) as a yellow oil. Rf (heptane/EtOAc 1:1): 0.42. H NMR (CDCl , 400
3
3
4
4
4
MHz): δ 7.75 (ddd, J=4.8 Hz, J=1.6 Hz, J =1.6 Hz, 1H); 7.33 (ddd, J =10.1 Hz,
J=7.6 Hz, J=1.6Hz, 1H); 7.10 (dd, J=7.6 Hz, J= 4.8 Hz, 1H); 4.21 (t, J=4.8Hz, 2H); 3.91-
.89 (m, 2H); 3.76-3.73 (m, 2H); 3.69-3-67 (m, 4H); 3.39 (t, J=4.8Hz, 2H). C NMR
CDCl , 100 MHz): δ 154.0 (C, d, J =237.0Hz); 142.3 (C, d, J =25.3Hz); 137.7 (CH, d,
J =13.3Hz); 123.4 (CH, d, J =4.3Hz); 121.8 (CH, d, J =4.3Hz); 71.1 (CH ); 70.8
CH ); 70.2 (CH ); 69.6 (CH ); 69.1 (CH ); 50.8 (CH ). ESI(+)-MS m/z 271 [M+H]
2 2 2 2 2
H-F
3
H-F
3
4
3
3
3
13
3
1
2
(
3 C-F C-F
3
3
4
C-F
C-F
C-F
2
+
(
+
(Calculated for C H FN O : 271.1).
1
1
16
4
3
3
-(2-(2-(2-Azidoethoxy)ethoxy)ethoxy)-2-nitropyridine (7). To a solution of 2-nitro-3-
hydroxypyridine (6) (192 mg, 1.37 mmol) in anhydrous DMF (15 mL) was added potassium
carbonate (252 mg, 1.83 mmol). The mixture was stirred 20 min at room temperature under
argon and a solution of 3 (300 mg, 0.91 mmol) in DMF (5 mL) was added. The solution was
stirred 48 h at 90 °C. After cooling to room temperature, water was added (50 mL) and the
resulting solution was extracted with EtOAc (2x20 mL). The organic layers were combined
This article is protected by copyright. All rights reserved.

and successively washed with sodium hydroxide 0.1 M (2x30 mL), water (20 mL), HCl 1 M
20 mL), brine (20 mL), dried over Na SO and concentrated under vacuum. The residue was
(
2
4
purified by column chromatography (silica gel, heptane/EtOAc 8:2 to 1:1) to afford 7 (248
1
mg, 92%) as a white powder. Mp: 50-53 °C. Rf (heptane/EtOAc 1:1): 0.14. H NMR
3
4
3
4
(
1
3
CDCl , 400 MHz): δ 8.06 (dd, J=4.4 Hz , J=1.0 Hz, 1H); 7.58 (dd, J=8.4 Hz , J=1.0 Hz,
3
3
4
3
3
H); 7.50 (dd, J=8.4 Hz , J=4.4 Hz, 1H); 4.28 (t, J=4.4 Hz, 2H); 3.88 (t, J=4.8 Hz, 2H);
.71-3.68 (m, 2H); 3.65-3.62 (m, 4H); 3.36 (t, J=5.2 Hz, 2H). C NMR (CDCl , 100 MHz):
3
13
3
δ 149.2 (C); 147.4 (C); 139.5 (CH); 128.6 (CH); 124.4 (CH); 71.1 (CH ); 70.7 (CH ); 70.1
2
2
+
(
CH ); 69.9 (CH ); 69.4 (CH ); 50.7 (CH ). ESI(+)-MS m/z 320 [M+Na] (Calculated for
2 2 2 2
+
C H N O Na : 320.1).
11
15
5
5
2
-Dimethylamino-3-hydroxypyridine (9). To a solution of 2-amino-3-hydroxypyridine (8)
(2.00 g, 18.1 mmol) in MeCN (30 mL) were successively added a 36% aqueous
formaldehyde solution (13.9 mL, 181 mmol), acetic acid (1.04 mL, 18.1 mmol) and
NaBH CN (3.42 g, 54.5 mmol) at 0 °C. The resulting mixture was stirred 48 h at room
3
temperature. A saturated solution of NaHCO (50 mL) was then added and the mixture was
3
extracted with EtOAc (3x40 mL). The organic layers were combined, dried over Na SO and
2
4
concentrated under vacuum. The residue was purified by column chromatography (silica gel,
heptane/EtOAc 9:1 to 7:3) to afford 9 (1.80 g, 72%) as a white solid. Rf (heptane/EtOAc
1
3
4
1
1
2
1
:1): 0.49. Mp: 51.0-53.2 °C. H NMR (CDCl ; 400 MHz): δ 7.89 (dd, J=4.8 Hz, J=1.6 Hz,
3
3
4
3
3
H); 7.16 (dd, J=8.0 Hz, J=1.6 Hz, 1H); 6.94 (dd, J=8.0 Hz, J=4.8 Hz, 1H); 5.37 (bs, 1H);
1
3
.78 (s, 6H). C NMR (CDCl , 100 MHz): δ 153.1 (C); 145.8 (C); 139.2 (CH); 121.9 (CH);
3
+
+
20.6 (CH); 42.4 (2xCH ). ESI(+)-MS m/z 139 [M + H] (Calculated for C H N O : 139.1).
3
7
11
2
3
-(2-(2-(2-Azidoethoxy)ethoxy)ethoxy)-N,N-dimethylpyridin-2-amine (10). To a solution
of 9 (1.26 g, 18.2 mmol) in anhydrous DMF (30 mL) was added K CO (1.68 g, 24.3 mmol).
2
3
The mixture was stirred 20 min at room temperature under argon and a solution of 3 (2.00 g,
.08 mmol) was added. The mixture was stirred overnight at 90 °C. After cooling to room
temperature, water was added (50 mL) and the resulting solution was extracted with EtOAc
2x30 mL). The organic layers were combined and successively washed with 0.1 M sodium
hydroxide (2x30 mL), water (20 mL), 1 M HCl (20 mL), brine (20 mL), dried over Na SO
6
(
2
4
and concentrated under vacuum. The residue was purified by column chromatography (silica
gel, toluene/EtOAc 1:1) to afford 10 (1.02 g, 57%) as a yellow oil.
1
3
4
Rf (heptane/EtOAc 1:1): 0.33. H NMR (CDCl , 400 MHz): δ 7.84 (dd, J=4.8 Hz, J=1.2
3
3
4
3
3
Hz, 1H); 7.00 (dd, J=7.6 Hz, J=1.2 Hz, 1H); 6.71 (dd, J=7.6 Hz, J=4.8 Hz, 1H); 4.12 (t,
J=4.8 Hz, 2H); 3.89 (t, J=4.8 Hz, 2H); 3.76-3.71 (m, 2H); 3.69-3.65 (m, 4H); 3.37 (t, J=4.8
Hz, 2H); 2.99 (s, 6H). C NMR (CDCl , 100 MHz): δ 153.4 (C); 145.7 (C); 139.3 (CH);
3
3
3
13
3
1
19.1 (CH); 115.7 (CH); 71.0 (CH ); 70.9 (CH ); 70.3 (CH ); 69.9 (CH ); 67.8 (CH ); 50.8
2 2 2 2 2
+
+
(CH ); 41.0 (2xCH ). ESI(+)-MS m/z 296 [M+H] (Calculated for C H N O : 296.2).
2 3 13 22 5 3
3
-(2-(2-(2-Azidoethoxy)ethoxy)ethoxy)-N,N,N-trimethylpyridin-2-aminium (11). To a
solution of 10 (60 mg, 0.20 mmol) in toluene (3 mL) was added methyl
trifluoromethanesulfonate (29 µL, 0.26 mmol) at 0 °C. The mixture was stirred 2 h at room
temperature and concentrated under vacuum to afford 11 (100 mg, 99%) as a brown oil
This article is protected by copyright. All rights reserved.

1
without further purification. Rf (dichloromethane/MeOH 95:5): 0.19. H NMR (DMSO-d ,
6
3
4
3
4
4
(
00 MHz): δ 8.18 (dd, J=4.4 Hz, J=1.2 Hz, 1H); 7.97 (dd, J=8.4 Hz, J=1.2 Hz, 1H); 7.74
dd, J=8.4 Hz, J=4.4 Hz, 1H); 4.44 (t, J=4 Hz, 2H); 3.88 (t, J=4 Hz, 2H); 3.65-3.60 (m,
1H); 3.590-3.56 (m, 4H); 3.38 (t, J=4.8 Hz, 2H). C NMR (DMSO-d , 100 MHz): δ 147.0
6
3
3
3
3
3
13
1
(
6
C); 142.9 (C); 138.5 (CH); 128.3 (CH); 125.4 (CH); 69.7 (CH ); 69.6 (CH ); 69.2 (CH );
2 2 2
+
8.9 (CH ); 68.1 (CH ); 53.3 (3xCH ); 49.9 (CH ). ESI(+)-MS m/z 310 [M+H] (Calculated
2 2 3 2
+
for C H N O : 311.2).
1
4
25
5
3
But-3-yn-1-yl tosylate (13). To a solution of 3-butyn-1-ol (12) (1.08 mL, 14.3 mmol) in
dichloromethane (40 mL) were added p-toluenesulfonyl chloride (3.27 g, 17.1 mmol) and
triethylamine (5.78 mL, 42.8 mmol). The mixture was stirred 48 h at room temperature under
argon. Water was then added (20 mL) and the resulting solution was extracted with
dichloromethane (3x15 mL). The organic layers were combined and successively washed
with 1 M HCl (3x20 mL) and water (2x20 mL), dried over Na SO and concentrated under
2
4
vacuum. The crude residue was purified by column chromatography (silica gel,
heptane/EtOAc 9:1) to afford 13 (2.54 g, 79%) as a light yellow oil. Rf (heptane/EtOAc 7:3):
1
3
3
0
.51. H NMR (CDCl , 400 MHz): δ 7.81 (d, J=8.0 Hz, 2H); 7.35 (d, J=8.0 Hz, 2H); 4.11
3
3
3
4
4
(
t, J=6.8 Hz, 2H); 2.56 (dt, J=6.8 Hz, J=2.4 Hz, 2H); 2.45 (s, 3H); 1.97 (t, J=2.4 Hz, 1H).
C NMR (CDCl , 100 MHz): δ 145.1 (C); 132.9 (C); 130.0 (2xCH); 128.0 (2xCH); 78.5
C); 70.9 (CH); 67.6 (CH ); 21.7 (CH ); 19.5 (CH ). ESI(+)-MS m/z 225 [M+H] (Calculated
1
3
3
+
(
2
3
2
+
for C H O S : 255.1).
1
1
13
3
Methyl N-acetyl-S-(but-3-yn-1-yl)cysteinate (14). To a solution of N-acetyl-L-cysteine
methyl ester (500 mg, 2.82 mmol) in anhydrous DMF (30 mL) were added sodium hydride
(81 mg, 3.4 mmol) and 13 (759 mg, 3.39 mmol). The mixture was stirred 48 h at room
temperature under argon. Water (50 mL) was then added and the resulting solution was
extracted with EtOAc (3x30 mL). The organic layers were combined and successively
washed with sodium hydroxide 0.1 M (20 mL), water (20 mL), brine (20 mL), dried over
Na SO4 and concentrated under vacuum. The residue was purified by column
2
chromatography (silica gel, dichloromethane/acetone 100:0 to 95:5) to afford 14 (247 mg,
1
3
8%) as a yellow oil. Rf (dichloromethane/acetone 95:5): 0.23. H NMR (CDCl , 400 MHz):
3
3
3
3
δ 6.44 (d, J=7.4 Hz, 1H); 4.81 (dt, J=7.4 Hz, J=5.0 Hz, 1H); 3.75 (s, 3H); 3.02 (split
quartet AB, J=13.5 Hz, J=5.0 Hz, 2H); 2.66 (t, J=7.2 Hz, 2H); 2.43 (td, J=7.2 Hz, J=2.8
Hz, 2H); 2.02 (s, 3H); 2.02 (t, J=2.8 Hz, 1H). C NMR (CDCl , 100 MHz): δ 171.3 (C);
2
3
3
3
4
4
13
3
1
1
2
70.0 (C); 82.2 (C); 69.9 (CH); 52.8 (CH); 52.0 (CH ); 34.3 (CH ); 31.5 (CH ); 23.2 (CH );
3 2 2 3
+
+
+
9.9 (CH ). ESI(+)-MS m/z 230 [M+H] , 252 [M+Na] (Calculated for C H NO SNa :
2
10 15
3
52.1).
Methyl
,2,3-triazol-4-yl)ethyl)-L-cysteinate (15). To a solution of 5 (59 mg, 0.22 mmol) in a
mixture of H O/tBuOH 1/1 (5 mL) were added 14 (50 mg, 0.22 mmol), CuSO (3.5 mg,
N-acetyl-S-(2-(1-(2-(2-(2-((2-fluoropyridin-3-yl)oxy)ethoxy)ethoxy)ethyl)-1H-
1
2
4
0
.022 mmol) and sodium ascorbate (86 mg, 0.44 mmol. The mixture was stirred 6 h at room
temperature. The mixture was concentrated and the residue dissolved in dichloromethane (5
mL). The solution was washed with a saturated solution of NH Cl (10 mL) and the aqueous
4
This article is protected by copyright. All rights reserved.

layer was extracted with dichloromethane (2x5 mL). The organic layers were combined,
dried over Na SO and concentrated under vacuum. The residue was purified by column
2
4
chromatography (silica gel, dichloromethane/MeOH 100/0 to 98:2) to afford 15 (62 mg,
1
5
(
7%) as a pale yellow oil. Rf (toluene/MeOH 8:2): 0.39. H NMR (CDCl , 400 MHz): δ 7.74
3
3
4
3
4
dt, J=4.8 Hz, J=1.6 Hz, 1H); 7.53 (s, 1H); 7.31 (ddd, J =9.6 Hz, J=8.0 Hz, J=1.6 Hz,
H); 7.10 (dd, J=8.0 Hz, J=4.8 Hz, 1H); 6.68 (d, J=7.8 Hz, 1H); 4.81 (dt, J=7.8 Hz,
J=5.0 Hz, 1H); 4.50 (t, J=5.0 Hz, 2H); 4.18 (t, J=4.8 Hz, 2H); 3.87-3.82 (m, 4H); 3.73 (s,
H); 3.69-3.66 (m, 2H); 3.62-3.60 (m, 2H); 2.99-2.97 (m, 2H); 2.95-2.91 (m, 2H); 2.86-2.80
m, 2H); 2.05 (s, 3H). C NMR (CDCl , 100 MHz): δ 171.4 (C=O); 170.2 (C=O); 153.9 (C,
H-F
3
3
3
3
1
3
3
3
3
(
1
3
3
1
2
3
d, J =237.0 Hz); 142.2 (C, d, J =25.9 Hz); 137.8 (CH, d, J =13.0 Hz); 123.4 (CH, d,
J =4.6 Hz); 121.9 (CH, d, J =3.8 Hz); 71.0 (CH ); 70.6 (CH ); 69.6 (CH ); 69.5 (CH );
C-F
C-F
C-F
3
4
C-F
C-F
2
2
2
2
6
9.2 (CH ); 52.8 (CH ); 52.2 (CH); 50.4 (CH ); 34.2 (CH ); 32.4 (CH ); 26.0 (CH ); 23.2
2 3 2 2 2 2
+
+
+
(CH ). ESI(+)-MS m/z 500 [M+H] , 522 [M+Na] (Calculated for C H FN O SNa :
3
21 30
5
6
5
22.2).
1
3
The C NMR signals of triazole moiety are not detected.
((5aR,6S,6aS)-1-(2-(2-(2-((2-fluoropyridin-3-yl)oxy)ethoxy)ethoxy)ethyl)-
1
,4,5,5a,6,6a,7,8-octahydrocyclopropa[5,6]cycloocta[1,2-d][1,2,3]triazol-6-yl)methanol
(17). To a solution of 5 (11.0 mg, 0.041 mmol) in anhydrous dichloromethane (0.2 mL) was
added (1R,8S,9s)-bicyclo[6.1.0]non-4-yn-9-ylmethanol (16) (5.9 mg, 0.039 mmol). The
mixture was stirred 4 h at room temperature and was then concentrated under vacuum. The
residue was purified by column chromatography (silica gel, heptane/acetone 10:0 to 3:7) to
1
afford 17 (16 mg, 97%) as a colorless oil. Rf (heptane/acetone 1:1): 0.23. H NMR (CDCl ,
3
3
4
4
00 MHz): δ 7.74 (ddd, J=4.8 Hz, J=1.6 Hz, J =1.6 Hz, 1H); 7.31 (ddd, J =10.0 Hz,
H-F H-F
J=7.6 Hz, J=1.6 Hz, 1H); 7.11 (ddd, J=7.6 Hz, J= 4.8 Hz, J =0.4 Hz, 1H); 4.40
td, J=5.0 Hz, J =2.0 Hz, 2H); 4.17 (t, J=5.0 Hz, 2H); 3.84 (t, J=5.4 Hz, 2H); 3.81 (t, J=4.6
Hz, 2H); 3.74 (dd, J=11.3 Hz, J=7.4 Hz, 1H), 3.68-3.60 (m, 3H); 3.59-3.52 (m, 2H); 3.10
ddd, J=15.6 Hz, J=8.0 Hz, J=3.6 Hz, 1H); 2.99 (ddd, J=16.0 Hz, J=7.2 Hz, J=3.2 Hz,
H); 2.84 (ddd, J=15.6 Hz, J=9.2 Hz, J=3.6 Hz, 1H); 2.69 (ddd, J=16.0 Hz, J=10.4 Hz,
J=3.2 Hz, 1H); 2.28-2.16 (m, 2H); 1.59-1.47 (m, 2H); 1.18 (q , J=8.4 Hz, 1H); 1.08-0.95
m, 2H). C NMR (CDCl , 100 MHz): δ 154.0 (C, d, J =237 Hz); 144.7 (C); 142.3 (C, d,
3
4
3
3
H-F
3
(
H-F
2
3
2
3
4
2
3
4
(
2
3
4
2
3
1
4
t 3
1
3
1
(
3 C-F
2
3
3
J =25.0 Hz); 137.8 (CH, d, J =13.0 Hz); 134.5 (C); 123.4 (CH, d J =3.8 Hz); 121.9
C-F
C-F
C-F
4
(
CH, J =3.8 Hz); 71.0 (CH ); 70.7 (CH ); 70.4 (CH ); 69.5 (CH ); 69.1 (CH ); 59.9 (CH );
C-F 2 2 2 2 2 2
4
7.9 (CH ); 26.1 (CH ); 23.2 (CH ); 23.0 (CH ); 22.4 (CH ); 21.4 (CH); 20.1 (CH); 19.7
2 2 2 2 2
+
+
(CH). ESI(+)-MS m/z 420 [M+H] (Calculated for C H FN O : 420.2).
21 29 4 4
N-Boc-dibenzocyclooctyne-amine (19). To a solution of dibenzocyclooctyne-amine (18)
20.0 mg, 0.072 mmol) in anhydrous dichloromethane (1 mL) was added di-tert-butyl
(
dicarbonate (32 mg, 0.14 mmol). The mixture was stirred 2 h at room temperature and then
concentrated under vacuum. The residue was purified by column chromatography (silica gel,
heptane/EtOAc 9:1 to 7:3) to afford 19 (22 mg, 84%) as a pale yellow oil. Rf (heptane/EtOAc
1
3
1
:1): 0.42. H NMR (CDCl , 400 MHz): δ 7.68 (d, J=7.2 Hz, 1H); 7.42-7.27 (m, 7H); 5.14
3
2
2
(
d, J=14.0 Hz, 1H); 5.03 (m, 1H); 3.69 (d, J=14.0 Hz, 1H); 3.26-3.09 (m, 2H); 2.49 (ddd,
J=16.4 Hz, J=7.6 Hz, J=4.0 Hz, 1H); 1.93 (ddd, J=16.4 Hz, J=6.0 Hz, J=4.0 Hz, 1H);
2
3
3
2
3
3
This article is protected by copyright. All rights reserved.

1
3
1
1
1
2
3
.39 (s, 9H). C NMR (CDCl , 100 MHz): δ 172.4 (C); 155.9 (C); 151.3 (C); 148.1 (C);
3
32.2 (CH); 129.2 (CH); 128.6 (CH); 128.4 (2 CH); 127.9 (CH); 127.3 (CH); 125.8 (CH);
23.1 (C); 122.7 (C); 115.1 (C); 107.8 (C); 79.1 (C); 55.6 (CH ); 36.6 (CH ); 35.1 (CH );
8.5 (3xCH ). ESI(+)-MS m/z 377 [M+H] , 399 [M+Na] (Calculated for C H N O Na :
2
2
2
+
+
+
3
23 24
2
3
99.2).
tert-Butyl (3-(3-(2-(2-(2-((2-fluoropyridin-3-yl)oxy)ethoxy)ethoxy)ethyl)-3,9-dihydro-8H-
dibenzo[b,f][1,2,3]triazolo[4,5-d]azocin-8-yl)-3-oxopropyl)carbamate (20). To a solution
of 19 (9.4 mg, 0.025 mmol) in anhydrous dichloromethane (1 mL) was added 5 (6.8 mg,
0
.025 mmol). The mixture was stirred 5 h at room temperature and then concentrated under
vacuum. The residue was purified by column chromatography (silica gel,
dichloromethane/MeOH 100:0 to 98:2) to afford 20 (13.6 mg, 84%) as a pale yellow oil. Rf
1
(
6
1
dichloromethane/MeOH 95:5): 0.5. H NMR (CDCl , 400 MHz): δ 7.77-6.98 (m, 11H);
3
2
3
3
.06 (d, J=16.4 Hz, 1H); 5.09 (t, J=5.8 Hz, 0.4H); 5.02 (t, J=5.8 Hz, 0.6H); 4.67-3.54 (m,
3H); 3.18-3.01 (m, 2H); 1.99 (m, 1H); 1.71 (m, 1H); 1.40 (s, 9H). ESI(+)-MS m/z 647
+
+
[
M+H] (Calculated for C H FN O : 647.3).
34 40 6 6
1
13
The H NMR signals of both regioisomers are not all distinguishable. The C NMR signals of
both regioisomers are not all distinguishable and could not be correctly assigned.
3
3
. Radiochemistry
-(2-(2-(2-Azidoethoxy)ethoxy)ethoxy)-2-[ F]fluoropyridine ([ F]-5). The aqueous
1
8
18
18
solution containing [ F]fluoride anions was automatically transferred to the TRACERLab
FX-FN or FX N Pro after the end of irradiation. The irradiated water is then sucked through a
QMA cartridge (Waters) to fix the [ F]fluoride anions and remove the enriched water which
was separately collected. The [ F]fluoride anions were eluted from the resin and transferred
to the reactor with a K CO /K solution (water and acetonitrile 30/70 (1 mL) containing 1.5
18
1
8
2
3
222

18
mg of K CO and 12 to 15 mg of Kryptofix 222). Finally, the K[ F]F-K complex was
2
3
222
prepared by evaporation of the solution in two heating steps : (i) first at 60 °C for 7 minutes
at a pressure ranging between 30 and 35 kPa and then (ii) at 120 °C for 5 minutes under
vacuum. Radiofluorination was carried out by addition of the precursors 7 or 11 (4 to 5 mg)
18
in solution in dimethyl sulfoxide (0.7 mL) to the dried K[ F]F-K
complex and the
22
2
resulting mixture was heated to 160 °C for 5 minutes. After cooling to 50 °C, the crude was
diluted twice with HPLC-solvent (2x2 mL) and transferred through an Alumina N cartridge
(Waters) and finally collected before HPLC injection ([HPLC A]). HPLC elution was
followed by both UV and radioactivity detection. When using precursor 7 [HPLC A / MP-1],
1
8
the retention times were 10-11 minutes for [ F]-5 and 16-17 minutes for 7. When using
1
8
precursor 11 [HPLC A / MP-2], the retention time was 12-13 minutes for [ F]-5 (not
1
8
determined for precursor 11). The fraction containing [ F]-5 was collected and diluted with
water (20 mL). The final formulation of the radiotracer was performed automatically using a

Sep-Pak Plus C18-based system (Waters). The purified tracer was finally recovered after
elution of the C18 cartridge with ethanol (2 mL). Chemical and radiochemical purities were
This article is protected by copyright. All rights reserved.

assessed on an aliquot by analytical HPLC, with an authentic sample of 5, as standard
[HPLC B / MP-3]: t = 2.6 min). Molar activity of the radiotracer was calculated from three
(
R
consecutive HPLC ([HPLC B / MP-3]) analyses (average) and determined as follows: the
area of the UV absorbance peak corresponding to the radiolabeled product was measured
(integrated) on the HPLC-chromatogram and compared with a standard curve relating mass
to UV absorbance.
3
.2 Kinetic studies
General procedure.
18
The kinetic studies were carried out as follows: an aliquot of [ F]-5 (150 to 300 MBq, in a
mixture of H O/EtOH 50/50) corresponding to 6 to 12 nmoles (calculated on the basis of the
2
molar activity for each production batch, see details Table 1) was mixed with the appropriate
precursor (14, 16 or 19) in acetonitrile, a catalyst in the case of CuAAC and further diluted
with water and tert-butanol to reach a final 1 mL volume. Reactions were all performed at
room temperature and 40 °C. Aliquots were taken at 5, 15 and 30 minutes and analyzed by
analytical HPLC [HPLC B / MP-4 or MP-5]. Conversions were evaluated by integration and
1
8
18
18
18
comparison of the areas of the peaks of compounds [ F]-5 to [ F]-15, [ F]-17 or [ F]-20,
derived from radioactive chromatograms.
18
Radiosynthesis of [ F]-15 via CuAAC.
18
Kinetics of the formation of [ F]-15 was studied firstly in two different catalytic conditions:
18

CuI/sodium ascorbate/DIEA: [ F]-5 (200 µL) mixed with 14 in MeCN (1 mg, 4
µmol, 100 µL) and aq. solutions of CuI (8.2 mg, 43 µmol, 10 eq, 100 µL)/sodium ascorbate
(
17 mg, 87 µmol, 20 eq,100 µL)/DIEA (4 µL) diluted with H O (300 µL) and tBuOH
2
(
200 µL).
1
8

CuSO /sodium ascorbate: [ F]-5 (200 µL) was mixed with 14 in MeCN (1 mg, 4
4
µmol, 100 µL) and aq. solutions of CuSO (3.4 mg, 21 µmol, 5 eq, 100µL)/sodium ascorbate
4
(
17 mg, 87 µmol, 20 eq, 100 µL) diluted with H O (300 µL) and tBuOH (200 µL). The
2
reaction was repeated with decreasing amounts of 14 (400 nmol and 40 nmol).
18
18
Retention times were 5.2 min for [ F]-5 and 2.8 min for [ F]-15 [HPLC B / MP-4].
1
8
18
Radiosynthesis of [ F]-17 and [ F]-20 via SPAAC.
18

SPAAC with BCN: [ F]-5 (200 µL) was mixed with 16 in MeCN/DMSO 2/1 (0.5 mg,
3
.3 µmol, 100 µL) diluted with H O (400 µL) and tBuOH (300 µL).
2
18
18
Retention times were 4.8 min for [ F]-5 and 2.4 min for [ F]-17 [HPLC B / MP-4].
1
8

SPAAC with DBCO derivative: [ F]-5 (200 µL) was mixed with 19 in MeCN/DMSO
4
/1 (0.5 mg, 1.3 µmol, 100 µL) H O (400 µL) and tBuOH (300 µL). The reaction was
2
repeated with decreasing amounts of 19 (130 nmol and 13 nmol).
18
18
Retention times were 1.3 min for [ F]-5 and 4.8 min for [ F]-20 ([HPLC B / MP-5]).
This article is protected by copyright. All rights reserved.

Results and discussion
Synthesis of FPyZIDE and precursors for radiolabeling.
Non-radioactive reference FPyZIDE (5) and the radiolabeling precursors 7 and 11 were
prepared in one, two or three steps starting from the appropriate 3-pyridinols 4, 6 and 8 and
the key intermediate 3 (Scheme 1). The azide derivative 3 was prepared in two steps by
substitution of the chlorodiethoxyethanol 1 with sodium azide followed by the tosylation of
the alcohol 2. Compound 3 was obtained in 66% yield over two steps. Such PEGylated
heterobifunctional chains, often used to prepare azide- or alkyne-containing prosthetic
4
6
reagents, are commonly synthesized in only one step starting from a di-alcohol or di-
4
2
tosylated PEG-chain but in lower yields due to the risk of homo-disubstitution. FPyZIDE
5) and the nitro-precursor 7 were prepared in only one step by alkylation of the 2-fluoro-3-
(
hydroxypyridine (4) and 2-nitro-3-hydroxypyridine (6) with 3. Compounds 5 and 7 were
obtained in 75% and 92% yields, respectively. The trimethylammonium precursor 11 was
synthesized in three steps. The 2-amino-3-hydroxypyridine (8) was first di-methylated by
reductive amination to afford the dimethylaminopyridine 9. Compound 9 was then alkylated
in the same conditions as for compounds 5 or 7, affording compound 10 that was methylated
using methyl triflate. Compound 11 was obtained in 41% yield over three steps.
Scheme 1: Synthesis of the non-radioactive reference FPyZIDE and its radiolabeling
precursors
Synthesis of the non-radioactive triazole references resulting from the click reactions.
As model for the CuAAC reaction, a propargylated derivative of cysteine (14) was chosen
(Scheme 2). The alcohol 12 was first tosylated in standard conditions to obtain compound 13
in 79% yield. N-Acetyl-L-cysteine methyl ester was then alkylated using 13 to afford the
propargylated derivative 14 in 38% yield. The non-radioactive reference 15 was finally
This article is protected by copyright. All rights reserved.

prepared by reaction of 14 with FPyZIDE (5) in standard copper-catalyzed conditions using
copper sulfate and sodium ascorbate. Compound 15 was obtained in 17% yield over three
steps.
As model for the SPAAC reaction, commercially available BCN (16) and DBCO (18) were
selected. These two cyclooctynes are frequently used for the tagging or engineering of
biologics using SPAAC reactions and therefore numerous fluorescent, biotinylated or
peptide-based tags derived from BCN and DBCO are described in the literature and
marketed. Considering the reaction rate of cyclooctynes in SPAAC reaction, BCN and DBCO
4
8
49
are among those displaying the highest rates. Sletten and Bertozzi and Dommerholt et al.
summarized the second order rate constants of different cyclooctynes in a SPAAC reaction
-1 -1
-
with benzyl azide or aliphatic azides. They reported rate constants of 0.14 M s and 0.31 M
1
-1
s for BCN and DBCO, respectively. This is higher than the rate constants described for
-1 -1
fluorinated cyclooctynes like DIFO (k = 0.0076 M s ) or dibenzyl cyclooctynes like DIBO
-1 -1
-1 -1
(k = 0.057 M s ) or TMDIBO (k = 0.094 M s ). Only the dibenzyl cyclooctyne BARAC
-
1 -1
displayed a higher rate constant (k = 0.96 M s ) than BCN or DBCO but with lower
commercial availability and with stability issues due to rearrangements depending on the
functionalization of the cyclooctyne ring.
5
0
Non-radioactive BCN-based reference 17 was prepared by reaction of FPyZIDE with 16 and
was obtained in 99% yield. Non-radioactive DBCO-based reference 20 was synthesized in
two steps starting from DBCO 18. For practical reasons related to the polarity of the amine
function borne by 18, we have chosen to protect it with a Boc moiety leading to compound 19
that was then reacted with FPyZIDE (5) and afforded non-radioactive reference 20 in 70%
yield over two steps.
Scheme 2: Synthesis of the non-radioactive model compounds 15, 17 and 20 resulting from
CuAAC and SPAAC reactions
This article is protected by copyright. All rights reserved.

Radiochemistry.
18
18
[
F]FPyZIDE ([ F]-5) was prepared by nucleophilic heteroaromatic substitution (Scheme
). Pyridine ring presents a higher reactivity toward aromatic nucleophilic substitution
3
compared to benzene. Pyridine-based radiofluorinated prosthetic reagents are now often
4
2-44, 46, 47, 51-53
encountered.
Two precursors displaying either a nitro (7) or a
18
trimethylammonium (11) moiety as efficient leaving group have been tested. [ F]FPyZIDE
was synthesized following a standard, fully automated, radiolabeling process comprising: (i)
1
8
the reaction of the precursor (7 or 11) with the activated complex of fluorine-18 K[ F]F-K
222
in dimethylsulfoxide at 160 °C for five minutes, (ii) the pre-purification of the reaction
mixture on an alumina cartridge, (iii) the purification using reverse-phase HPLC with an
1
8
adapted mobile phase to efficiently separate the exceeding precursor from [ F]-5, and (iv)
the C18-based formulation of the final product in ethanol. For the HPLC purification
different mobile phases were used depending on the precursor. When the
trimethylammonium precursor 11 was used, a standard mixture of water and acetonitrile with
1
8
trifluoroacetic acid was efficient to separate in short purification times [ F]-5 (t = 12-13
R
min) from 11. When the nitro precursor 7 was used, the similar polarity of nitro- and fluoro-
compounds necessitated adapted mobile phase to allow for an efficient separation by reverse-
phase HPLC. Using a three-component mixture of water/methanol/tetrahydrofuran acidified
18
with phosphoric acid enabled the isolation of [ F]-5 (t = 16-17 min) from precursor 7 with a
R
satisfactory purity.
1
8
Starting from precursors 7 and 11, [ F]FPyZIDE was respectively obtained in less that one
hour in 39% ± 10 (n = 6) and 25% ± 8 (n = 12) radiochemical yields, based on starting [ F]F
radioactivity. The average decay-corrected molar activity was 145 ± 27 GBq/µmol (n = 11).
1
8
-
1
8
These yields are comparable with those obtained for the preparation of other [ F]pyridine-
1
8
based prosthetic reagents. [ F]FPyKYNE was prepared in 30-35% decay-corrected yield in
4
5
18
18
6
[
0 to 70 minutes , and [ F]FPy5yne, an alkyne bearing analog of [ F]FPyZIDE, or PEG-
1
8
F]FPyKYNE in 35 to 45% decay-corrected yields but in longer synthesis times ranging
4
6, 47
18
from 80 to 100 minutes.
high 70% non-decay-corrected yield.
Only the analog F-BMT-187144 was prepared in a surprisingly
4
2
The chemical and radiochemical purities were greater than 95% and 99%, respectively, as
assessed by the quality control. The average decay corrected molar activity was 145
GBq/µmol, whatever the precursor used. Even if trimethylammonium precursors are usually
considered to be more reactive than the nitro ones, we observed a more efficient radiolabeling
using precursor 7 than 11. The HPLC chromatograms (see ESI – Figure S1) of isolation of
1
8
[
F]-5 showed several radioactive by-products when precursor 11 was used, compared to
18
precursor 7. Such a trend was already observed in the preparation of [ F]FPy5yne, where the
nitro- and trimethylammonium precursors led to isolated [ F]FPy5yne in 50% and 42%
decay-corrected yields, respectively.
18
4
7
Moreover, when precursor 11 was used, a
demethylation reaction of the trimethylammonium in competition with the nucleophilic
5
4
substitution was observed, contributing to the lower final yields.
This article is protected by copyright. All rights reserved.

18
Scheme 3: Preparation of [ F]FPyZIDE
Cycloadditions (Scheme 4) were conducted in different reaction conditions summarized in
Table 1. A typical series of chromatograms for the follow up of the cycloaddition can be seen
in Supplementary Information (ESI – Figure S2).
For the CuAAC reaction, two standard catalysis conditions were used, either copper iodide as
direct source of Cu(I) with sodium ascorbate to avoid copper oxidation and a base, or copper
sulfate with sodium ascorbate to generate in situ Cu(I).
1
8
18
18
Scheme 4: Cycloaddition reactions to prepare [ F]-15, [ F]-17 and [ F]-20
The CuAAC reaction catalyzed with Cu(I) at room temperature (Table 1, conditions #1)
showed a slow conversion rate and did not exceed 75% after 30 minutes (Figure 1, #1). At 40
1
8
°
[
C with the same conditions (Table 1, conditions #2), a higher 70% conversion of [ F]-5 into
18
F]-15 was observed after 5 minutes of reaction (Figure 1, #2) Surprisingly, the conversion
yield decreased at longer times (15 and 30 minutes) and a more polar radioactive by-product
1
8
18
(
t < 1 min) than [ F]-15 (t = 2.8 min) was detected on analytical HPLC whereas [ F]-5 (t
R
R
R
=
5.2 min) could not be detected anymore (Figure 1, #2). The aqueous basic reaction medium
could induce a hydrolysis of the methyl ester protecting the carboxylic acid of the cysteine
residue leading to a polar radioactive compound.
The CuAAC reaction catalyzed with Cu(II)/sodium ascorbate at room temperature (Table 1,
conditions #3) showed a similar conversion yield to those observed with Cu(I) (Figure 1,
This article is protected by copyright. All rights reserved.

#
3). At 40 °C (Table 1, conditions #4), the reaction rate increased drastically and the
1
8
18
conversion of [ F]-5 into [ F]-15 was almost completed after 15 minutes (Figure 1, #4).
When the amount of alkyne 14 decreased while maintaining the same amount of copper
sulfate and sodium ascorbate (Table 1, conditions #5 and #6), a very fast reaction was
1
8
observed leading to a total conversion of [ F]-5 after 5 minutes (Figure 1, #5 and #6). For
18
longer reaction times (30 minutes), a degradation of [ F]-15 was observed but in limited
amounts (less than 10%). A polar radioactive by-product was detected on HPLC
chromatograms. This could be due to the oxidative character of the catalytic system leading
18
27
to a partial degradation of [ F]-15 .
Table 1. Summary of the reaction conditions for CuAAC and SPAAC reactions
1
8
Eq.
alkyne
#
Alkyne
[ F]-5
Temp.
Catalysis
Entry
#
CuI
(µM)
Na Asc
(µM)
CuSO4
(µM)
Compd
µM
µM
DIEA
1
2
3
4
5
6
7
8
9
14
14
14
14
14
14
16
16
19
19
19
19
19
19
4.3
4.3
4.3
0.009
0.009
0.012
0.012
0.012
0.012
0.011
0.011
0.008
0.010
0.009
0.006
0.009
0.006
477
477
358
358
35.8
3.58
300
300
162
130
14
rt
40 °C
rt
40 °C
40 °C
40 °C
rt
40 °C
rt
40 °C
rt
40 °C
rt
43
43
--
--
--
23
23
87
87
87
87
87
87
--
--
22
22
22
22
4.3
0.43
0.043
3.3
3.3
1.3
--
none
none
none
none
none
none
none
none
1
1
1
1
1
0
1
2
3
4
1.3
0.13
0.13
0.013
0.013
21
1.4
2.1
40 °C
This article is protected by copyright. All rights reserved.

1
8
Figure 1: [ F]-15 formation kinetics in different copper-catalysis conditions via CuAAC
The SPAAC reaction was tested with two different cyclooctynes, BCN (16) and a DBCO
derivative 19. At room temperature or at 40 °C (Table 1, conditions #7 and #8), BCN (16)
1
8
reacted slowly with [ F]-5 and the conversion yields did not exceed 14% and 27% at room
temperature and 40 °C, respectively (Figure 2, #7 and #8). Longer reaction times (60
minutes) only slightly increased the conversion (data not shown). The DBCO derivative 19
18
reacted more rapidly with [ F]-5 (Table 1, conditions #9 and #10). The conversion was
almost complete after 15 minutes (Figure 2, #9, #10). No major difference of reaction rate
could be observed at room temperature and 40 °C (Figure 2, #9, #10). The difference of
reactivity between 16 and 19 in SPAAC reaction is in accordance with the rates reported in
4
9
the literature . At lower concentrations of 19, the conversion proceeded more slowly (Figure
, #11 to #14). Similarly, reaction temperature had no influence on conversion rates. In quasi-
2
1
8
equimolar conditions between 19 and [ F]-5 corresponding to nanomolar concentrations, the
conversion does not exceed 5 to 7% (Figure 2 #13 and #14).
1
8
18
Figure 2: [ F]-17 and [ F]-20 formation kinetics via SPAAC
This article is protected by copyright. All rights reserved.

Among the different reaction conditions tested, it appeared that the fastest conversions were
obtained (i) in CuAAC at 40 °C with an excess of Cu(II) and sodium ascorbate and (ii) in
SPAAC using an excess of DBCO derivative 19 as strained alkyne. Moreover, in CuAAC
18
reaction, the molar ratio between terminal alkyne (14) and azide ([ F]-5) seemed to influence
the reaction rate. In our experimental conditions, the decrease of this ratio, tending to an
equimolar proportion of both reaction partners, positively influenced the conversion rate and
allowed a reaction completion in only 5 minutes. This can be advantageous when considering
the radiolabeling of expensive biologics. For the SPAAC reaction, the temperature had a
limited benefit on the conversion yields, whatever the cyclooctyne used. Moreover, contrarily
to the CuAAC, the dilution of strained alkyne 19 negatively influenced the conversion rate.
Our series of experiments showed that a faster reaction could be achieved with SPAAC
(Figure 2, #9) than CuAAC (Figure 1, #3) at room temperature. At 40 °C, conversion rates
were comparable (Figure 1, #4 – Figure 2, #9) unless working with a huge excess of copper
sulfate and sodium ascorbate. In this latter case (Table 1, conditions #5 and #6), at 40 °C, the
rate of CuAAC reaction was higher than SPAAC reaction. At high dilution of the reagents
1
8
18
(
Figure 1, #5 and #6 / Figure 2 #11 to#14), the conversion of [ F]-5 into [ F]-15 via CuAAC
18
reaction is significantly faster than the formation of [ F]-20 via SPAAC reaction. These
results were in accordance with classically reported reaction rates for CuAAC and SPAAC.
35
Conclusion.
18
18
18
[
F]FPyZIDE ([ F]-5), a new [ F]fluoropyridine-based azide-containing prosthetic reagent,
was prepared in one radiochemical step in good yields (39% decay-corrected) in less than one
hour starting from the corresponding nitro-precursor 7. We showcased with model
compounds that this prosthetic reagent could be used in a versatile manner either in CuAAC
or SPAAC reaction in high yields and short reaction times that are fully compatible with
fluorine-18 half-life. Depending on the chemical structure of the biologics that will be
1
8
radiolabeled with fluorine-18 using [ F]FPyZIDE, CuAAC at 40 °C will be preferred for
compounds that are not sensitive to copper and temperature (short peptides, sugars) whereas
SPAAC will be preferred for more sensitive biologics (nucleic acids, proteins).
Acknowledgements.
The authors wish to thank Daniel Gouel and Christophe Lechene for cyclotron operation.
Mélanie Roche (PhD student) was funded by the intramural CEA-IRTELIS doctoral program.
Mylène Richard (PostDoc student) was funded by the intramural DRF-Impulsion IRIP
program.
This article is protected by copyright. All rights reserved.

References
Jackson IM, Scott PJH, Thompson S. Clinical applications of radiolabeled peptides
for PET. Seminars Nucl Med 2017;47(5):493-523.
Kuhnast B, El Hadri A, Boisgard R, et al. Synthesis, radiolabeling with fluorine-18
1
.
2
.
and preliminary in vivo evaluation of a heparan sulphate mimetic as potent angiogenesis and
heparanase inhibitor for cancer applications. Org Biomol Chem. 2016;14(6):1915-20.
3
2
3
.
Li Y, Schaffer P, Perrin DM. Dual isotope labeling: Conjugation of P-
1
8
oligonucleotides with F-aryltrifluoroborate via copper(I) catalyzed cycloaddition. Bioorg
Med Chem Lett. 2013;23(23):6313-6.
4
.
Takemiya K, Ning X, Seo W, et al. Novel PET and Near Infrared Imaging probes for
the specific detection of bacterial infections associated with cardiac devices. JCCC:
Cardiovascular Imaging. 2018;In press.
5
.
Chatalic KLS, Kwekkeboom DJ, de Jong M. Radiopeptides for imaging and therapy:
A radiant future. J Nucl Med. 2015;56(12):1809-12.
1
8
6
.
Arimitsu K, Kimura H, Arai Y, et al. F-Containing Positron Emission Tomography
probe conjugation methodology for biologics as specific binders for tumors. Curr Topics Med
Chem. 2016;16(24):2703-24.
7
.
Kuhnast B, Dollé F. The challenge of labeling macromolecules with fluorine-18 :
Three decades of research. Curr Radiopharm. 2010;3:174-201.
8
.
Schirrmacher R, Wangler C, Schirrmacher E. Recent developments and trends in
1
8
Fradiochemistry: Syntheses and applications. Mini Rev Org Chem. 2007;4(4):317-29.
.
4)NH : a heterodimeric targeting vector for positron emission tomography imaging of
6
4
9
1
Jackson AB, Nanda PK, Rold TL, et al. Cu-NO2A-RGD-Glu-6-Ahx-BBN(7-
2
prostate cancer. Nucl Med Biol. 2012;39(3):377-87.
68
1
2
1
0.
014;4(1):47-80.
1. Yu HM, Chen JH, Lin KL, et al. Synthesis of Ga-labeled NOTA-RGD-GE11
Velikyan I. Prospective of Ga-radiopharmaceutical development. Theranostics.
6
8
heterodimeric peptide for dual integrin and epidermal growth factor receptor-targeted tumor
imaging. J Label Compds Radiopharm. 2015;58(7):299-303.
1
2.
Cook BE, Adumeau P, Membreno R, et al. Pretargeted PET imaging using a site-
specifically labeled immunoconjugate. Bioconjugate Chem. 2016;27(8):1789-95.
1
8
1
3.
Amigues E, Schulz J, Szlosek-Pinaud M, et al. F-Si-RiboRGD: From design and
synthesis to the imaging of v3 integrins in melanoma tumors. ChemPlusChem.
2
1
012;77(5):345-9.
4. Lindner S, Michler C, Leidner S, et al. Synthesis and in vitro and in vivo evaluation of
SiFA-tagged Bombesin and RGD peptides as tumor imaging probes for positron emission
tomography. Bioconj Chem. 2014;25(4):738-49.
1
8
1
5.
Liu ZB, Amouroux G, Zhang ZX, et al. F-Trifluoroborate derivatives of Des-
Arg(10) Kallidin for imaging bradykinin B1 receptor expression with positron emission
tomography. Mol Pharm. 2015;12(3):974-82.
1
6.
Malik N, Baur B, Winter G, et al. Radiofluorination of PSMA-HBED via
18
2+
(Al[ F]F) chelation and biological evaluations in vitro. Mol Imaging Biol. 2015;17(6):777-
8
1
5.
7.
1
8
McBride WJ, Sharkey RM, Karacay H, et al. A novel method of F-radiolabeling for
PET. Bioconj Chem. 2009;50(6):991-8.
8. Schirrmacher R, Bradtmöller G, Schirrmacher E, et al. F-labeling of peptides by
means of an organosilicon-based fluoride acceptor. Angew Chem Int Ed. 2006;45(36):6047-
0.
1
8
1
5
This article is protected by copyright. All rights reserved.

1
9.
Ting R, Adam MJ, Ruth TJ, et al. Arylfluoroborates and alkylfluorosilicates as
1
8
potential PET imaging agents: High-yielding aqueous biomolecular F-labeling. J Am Chem
Soc. 2005;127(38):13094-5.
2
18
0.
Olberg DE, Hjelstuen OK. Labeling strategies of peptides with F for Positron
Emission Tomography. Curr Topics Med Chem. 2010;10(16):1669-79.
1
8
2
1.
Marik J, Sutcliffe JL. Click for PET: rapid preparation of [ F]fluoropeptides using
Cu-I catalyzed 1,3-dipolar cycloaddition. Tet Lett. 2006;47(37):6681-4.
2
1
2
2.
Meyer JP, Adumeau P, Lewis JS, et al. Click chemistry and radiochemistry: The first
0 Years. Bioconjugate Chem. 2016;27(12):2791-808.
3.
1
8
Kettenbach K, Schieferstein H, Ross TL. F-Labeling using click cycloadditions.
Biomed Res Intl. 2014.
4. Kolb HC, Finn MG, Sharpless KB. Click chemistry: Diverse chemical function from a
few good reactions. Ang Chem Intl Ed. 2001;40(11):2004-9.
5. Houghton EA, Nicholas KM. In vitro reactive oxygen species production by histatins
and copper(I,II). J Biol Inorg Chem. 2009;14(2):243-51.
6. Burrows CJ, Muller JG. Oxidative nucleobase modifications leading to strand
scission. Chem Rev. 1998;98(3):1109-52.
7. Li S, Cai H, He J, et al. Extent of the oxidative side reactions to peptides and proteins
during the CuAAC reaction. Bioconjugate Chem. 2016;27(10):2315-22.
8. Li L, Zhang Z. Development and applications of the Copper-Catalyzed Azide-Alkyne
Cycloaddition (CuAAC) as a bioorthogonal reaction. Molecules. 2016;21(10).
9. Clavadetscher J, Hoffmann S, Lilienkampf A, et al. Copper catalysis in living systems
and in situ drug synthesis. Angew Chem Int Ed. 2016;55(50):15662-6.
2
2
2
2
2
2
3
0.
Mack S, Fouz MF, Dey SK, et al. Pseudo-ligandless click chemistry for
oligonucleotide conjugation. Curr Prot Chem Biol. 2016;8(2):83-95.
1. Paredes E, Das SR. Optimization of acetonitrile co-solvent and copper stoichiometry
for pseudo-ligandless click chemistry with nucleic acids. Bioorg Med Chem Lett.
3
2
3
012;22(16):5313-6.
2. Agard NJ, Prescher JA, Bertozzi CR. A strain-promoted [3+2] azide-alkyne
cycloaddition for covalent modification of blomolecules in living systems. J Am Chem Soc.
2
3
004;126(46):15046-7.
3. Jewett JC, Bertozzi CR. Cu-free click cycloaddition reactions in chemical biology.
Chem Soc Rev. 2010;39(4):1272-9.
4. McKay Craig S, Finn MG. Click chemistry in complex mixtures: bioorthogonal
bioconjugation. Chem Biol. 2014;21(9):1075-101.
3
3
2
3
5.
014;9(1):16-20.
6. Pretze M, Pietzsch D, Mamat C. Recent trends in bioorthogonal click-radiolabeling
Lang K, Chin JW. Bioorthogonal reactions for labeling proteins. ACS Chem Biol.
reactions using fluorine-18. Molecules. 2013;18(7):8618-65.
1
8
3
7.
Denk C, Wilkovitsch M, Skrinjar P, et al. [ F]Fluoroalkyl azides for rapid
radiolabeling and (re)investigation of their potential towards in vivo click chemistry. Org
Biomol Chem 2017;15:5976-82.
1
8
3
8.
emission tomography. Bioconjugate Chem. 2007;18(3):989-93.
9. Hashizume K, Hashimoto N, Miyake Y. Synthesis of positron labeled photoactive
Glaser M, Arstad E. "Click labeling" with 2-[ F]fluoroethylazide for positron
3
1
8
compounds - F-labeled aryl azides for positron labeling of biochemical molecules. J Org
Chem. 1995;60(21):6680-1.
This article is protected by copyright. All rights reserved.

4
0.
Hausner SH, Carpenter RD, Bauer N, et al. Evaluation of an integrin a b -specific
v 6
1
8
peptide labeled with F-fluorine by copper-free, strain-promoted click chemistry. Nucl Med
Biol. 2013;40(2):233-9.
4
1
8
1.
Collet C, Maskali F, Clement A, et al. Development of 6- Ffluoro-carbohydrate-
based prosthetic groups and their conjugation to peptides via click chemistry. J Label
Compds Radiopharm. 2016;59(2):54-62.
4
2.
Donnelly DJ, Smith RA, Morin P, et al. Synthesis and biologic evaluation of a Novel
1
8
F-labeled adnectin as a PET radioligand for imaging PD-L1 expression. J Nucl Med.
2
4
018;59(3):529-35.
1
8
3.
de Bruin B, Kuhnast B, Hinnen F, et al. 1-[3-(2-[ F]fluoropyridin-3-
yloxy)propyl]pyrrole-2,5-dione: Design, synthesis, and radiosynthesis of a new
18
[
F]fluoropyridine-based maleimide reagent for the labeling of peptides and proteins.
Bioconjugate Chem. 2005;16(2):406-20.
4
[
4.
Kuhnast B, de Bruin B, Hinnen F, et al. Design and synthesis of a new
18
F]fluoropyridine-based haloacetamide reagent for the labeling of oligonucleotides: 2-
1
8
Bromo-N-[3-(2-[ F]fluoropyridin-3-yloxy)propyl]acetamide.
Bioconjugate
Chem.
2
4
004;15(3):617-27.
5. Kuhnast B, Hinnen F, Tavitian B, et al. [ F]FPyKYNE, a fluoropyridine-based
1
8
alkyne reagent designed for the fluorine-18 labelling of macromolecules using click
chemistry. J Label Compds Radiopharm. 2008;51(9-10):336-42.
4
6.
Inkster J, Lin KS, Ait-Mohand S, et al. 2-Fluoropyridine prosthetic compounds for the
1
8
F-labeling of bombesin analogues. Bioorg Med Chem Lett. 2013;23(13):3920-6.
4
[
7.
Inkster JA, Guerin B, Ruth TJ, et al. Radiosynthesis and bioconjugation of
1
8
F]FPy5yne, a prosthetic group for the F-18 labeling of bioactive peptides. J Label Compds
Radiopharm. 2008;51(13-14):444-52.
8. Sletten EM, Bertozzi CR. From mechanism to mouse: A tale of two bioorthogonal
reactions. Acc Chem Res. 2011;44(9):666-76.
9. Dommerholt J, Rutjes FPJT, van Delft FL. Strain-promoted 1,3-dipolar cycloaddition
of cycloalkynes and organic azides. Top Curr Med. 2016;374(2):16.
0. Chigrinova M, McKay CS, Beaulieu LPB, et al. Rearrangements and addition
4
4
5
reactions of biarylazacyclooctynones and the implications to copper-free click chemistry. Org
Biomol Chem. 2013;11(21):3436-41.
1
8
5
1.
Berndt M, Pietzsch J, Wuest F. Labeling of low-density lipoproteins using the F-
18
labeled thiol-reactive N-[6-(4-[ F]fluorobenzylidene)aminooxyhexyl]maleimide. Nucl Med
Biol. 2007;34(1):5-15.
1
8
5
[
2.
Cai WB, Zhang XZ, Wu Y, et al. A thiol-reactive F-labeling agent, N-[2-(4-
F]fluorobenzamido)ethyl]maleimide, and synthesis of RGD peptide-based tracer for PET
1
8
imaging of a b integrin expression. J Nucl Med. 2006;47(7):1172-80.
v 3
18
5
3.
Cavani M, Bier D, Holschbach M, et al. Efficient synthesis of [ F]FPyME: A new
approach for the preparation of maleimide-containing prosthetic groups for the conjugation
with thiols. J Label Compds Radiopharm. 2017;60(1):87-92.
5
4.
Sun H, DiMango SG. Competitive demethylation and substitution in N,N,N-
trimethylanilinium fluorides. J Fluor Chem. 2007;128:806-12.
This article is protected by copyright. All rights reserved.

Graphical abstract.
18
[
F]FPyZIDE : a versatile prosthetic reagent for the fluorine-18 radiolabeling of biologics
via copper-catalyzed or strain-promoted alkyne-azide cycloadditions
This article is protected by copyright. All rights reserved.