Rapid in situ synthesis of [11C]methyl azide and its application in 11C click-chemistry

Article abstract of DOI:10.1016/j.tetlet.2008.06.020

We synthesized [¹¹C]methyl azide ([¹¹C]MeA) by reacting [¹¹C]methyl iodide ([¹¹C]MeI) in situ with an azide-donor and used it in the synthesis of ¹¹C-labeled 1,2,3-triazoles. A one-pot click approach comprised the infusion of gaseous [¹¹C]MeI into a mixture of NaN₃, ethynylbenzene, and CuI in water at a temperature of 100 °C yielding the ¹¹C-triazole in radiochemical yields (RCY) of 25%. In a two-step labeling protocol, we synthesized the [¹¹C]MeA in acetonitrile in advance to the click step. Using the more soluble Na⁺ / 18 -crown- 6 / N₃^- complex as source of N₃^-, a much higher trapping efficiency of [¹¹C]MeI in this solvent ensured an almost quantitative conversion of [¹¹C]MeI to [¹¹C]MeA within 5-10 min at room temperature. The [¹¹C]MeA was thereafter reacted with ethynylbenzene at 100 °C yielding 1-[¹¹C]methyl-4-phenyl-1H-1,2,3-triazole in preparative RCY of 60%. As a final proof of applicability, we used ¹¹C-click-chemistry for the labeling of N-terminal 4-ethynylbenzene derivatized d-Glu-d-Tyr-[Cys-Tyr-Trp-Lys-Thr-Cys]-Thr, a cyclic water-soluble Tyr³-octreotate derivative.

Full text of DOI:10.1016/j.tetlet.2008.06.020

Tetrahedron Letters 49 (2008) 4824–4827
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Rapid in situ synthesis of [11C]methyl azide and its application
in 11C click-chemistry
Ralf Schirrmacher a, Younes Lakhrissi a, Dean Jolly a, Julian Goodstein a, Philippe Lucas a,
Esther Schirrmacher b,
*
a McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, University Street 3800, Montreal, Canada
b Lady Davis Institute for Medical Research, Jewish General Hospital, Cote Sainte Catherine 3755, McGill University, Montreal, Canada
a r t i c l e i n f o
a b s t r a c t
We synthesized [11C]methyl azide ([11C]MeA) by reacting [11C]methyl iodide ([11C]MeI) in situ with an
azide-donor and used it in the synthesis of 11C-labeled 1,2,3-triazoles. A one-pot click approach com-
prised the infusion of gaseous [11C]MeI into a mixture of NaN3, ethynylbenzene, and CuI in water at a
temperature of 100 °C yielding the 11C-triazole in radiochemical yields (RCY) of 25%. In a two-step label-
ing protocol, we synthesized the [11C]MeA in acetonitrile in advance to the click step. Using the more sol-
uble Naþ=18-crown-6=N3ꢀ complex as source of N3ꢀ, a much higher trapping efficiency of [11C]MeI in this
solvent ensured an almost quantitative conversion of [11C]MeI to [11C]MeA within 5–10 min at room
Article history:
Received 30 April 2008
Revised 2 June 2008
Accepted 4 June 2008
Available online 10 June 2008
Keywords:
Click-chemistry
Carbon-11
temperature. The
[
11C]MeA was thereafter reacted with ethynylbenzene at 100 °C yielding 1-
[
11C]methyl-4-phenyl-1H-1,2,3-triazole in preparative RCY of 60%. As a final proof of applicability, we
Radio-labeling
used 11C-click-chemistry for the labeling of N-terminal 4-ethynylbenzene derivatized
[Cys–Tyr–Trp–Lys–Thr–Cys]–Thr, a cyclic water-soluble Tyr3-octreotate derivative.
D-Glu–D-Tyr–
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
been applied to the synthesis of carbon-11 labeled compounds.
In comparison to carbon-11 with a very short half-life (11C, t1/2
=
It has recently been demonstrated that click-chemistry has
great potential for the preparation of 18F-PET radiopharmaceuti-
cals and might become a valuable tool in labeling chemistry.1–3
The term ‘click-chemistry’ was coined by Sharpless and denotes
chemical reactions characterized by high selectivity providing
high yields of products from easily accessible building blocks.4
The most commonly used click-reaction is the Cu(I)-catalyzed
Huisgen reaction, a 1,3-dipolar cyclo-addition of terminal al-
kynes with azides, yielding 1,4 disubstituted 1,2,3-triazoles un-
der mild conditions. This reaction owns its usefulness to the
relative ease with which both necessary functional moieties,
azide and alkyne, can be introduced into various molecules.5
Both groups are relatively stable to the majority of common
reaction conditions in organic synthesis so that they can be
introduced into target molecules whenever convenient.6 The dis-
covery that Cu(I) catalysis leads to 1,4-regioisomers only while
drastically enhancing reaction rates has led to its application
in 18F-radiochemistry.7 In the case of 18F-click-chemistry either
the azide or the alkyne was 18F-labeled and used for the triazole
formation. To our best knowledge click-chemistry has so far not
20 min), fluorine-18 has a far more convenient physical half-life
of 110 min rendering synthesis times of up to 2 h possible. Such
a long synthesis time is not applicable to the preparation of 11C-
radiopharmaceuticals and it is important to find easy protocols
for labeling. Another important difference between these two
isotopes is the available specific activity (SA, in MBq/lmol)
which defines the degree of dilution by non-radioactive fluo-
rine-19 or carbon-12 of the radioactive isotope. The SA of a
radio-isotope is a function of time, demanding rapid and effi-
cient methods for its introduction into bio-molecules. This is
especially true for the synthesis of 11C-labeled compounds,
where the SA is decreasing rapidly during synthesis. The most
frequently applied labeling precursor for the labeling of small
compounds is [11C]MeI,8–10 which can be routinely prepared in
fully automated synthesis modules. This precursor cannot be
used for the labeling of multifunctional compounds such as
peptides as a result of its low selectivity regarding different
nucleophiles. In this Letter, we report the fast reaction of the
click-reactant [11C]methyl azide ([11C]MeA) with a simple model
compound ethynylbenzene (1) yielding 1-[11C]methyl-4-phenyl-
1H-1,2,3-triazole (2) using two different approaches (one-step
and a two-step approach, Scheme 1) and the transferal of the
latter method to the 11C-labeling of
feasibility.
a peptide as proof of
* Corresponding author. Tel.: +1 514 340 8222 3912; fax: +1 514 340 8716.
E-mail address: esther.schirrmacher@mcgill.ca (E. Schirrmacher).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.06.020

R. Schirrmacher et al. / Tetrahedron Letters 49 (2008) 4824–4827
4825
active peak (HPLC) at 15 min, which we attributed to the [11C]MeA.
It has been described in the literature that various ligands such as
NaN3
N
N
11CH3I
A)
B)
+
L-proline and more complex ligands such as tris((1H-benzo-
N
H2O, Cu(I), 100 oC
11CH3
imidazol-2-yl)methyl)amine can accelerate the cycloaddition
considerably.14 However, in radiochemistry the ultimate aim is
the production of a radiotracer suitable for iv injection into
humans, so we decided not to add any compounds which would
result in a longer, more complex work-up procedure. In order to
overcome the insufficient trapping yield of [11C]MeI in water, to re-
duce the loss of radioactivity at higher temperatures and to get a
2
1
1, CH3CN/
H2O, 100 oC
Cu(I)
Na+/18-crown-6/N3
-
11CH3I
11CH3N3
CH3CN, rt
Scheme 1. (A) One-pot 11C-labeling of alkynylbenzene (1, 10 mg, 0.09 mmol) via in
situ generation of [11C]MeA from [11C]MeI in H2O (1 mL) at 100 °C within 10 min
under CuI (10 mg, 0.052 mmol) catalysis; (B) conversion of [11C]MeI to [11C]MeA in
higher conversion of
[
[
11C]MeI to
[
11C]MeA, we bubbled the
CH3CN (0.5 mL) using Naþ=18-crown-6/N3 (10–15 mg, 30.4–45.6 mmol) at rt for
ꢀ
þ
15
11C]MeI into a solution of Na =18-crown-6/N3
in CH3CN using
ꢀ
5–10 min. An aliquot of [11C]MeA in acetonitrile (0.3–0.5 mL) was subsequently
a sweep flow of nitrogen (10 mL/min). Trapping yields were 85–
90%. According to the HPLC chromatogram (not shown), the con-
version of [11C]MeI into [11C]MeA was 97% after 5–10 min at room
reacted with 1 (10 mg, 0.09 mmol) in water (0.5 mL) at 100 °C for 10 min.
temperature. To remove Naþ=18-crown-6/N3 which might nega-
ꢀ
2. Results and discussion
tively affect the triazole formation and could become a toxic con-
taminant in possible in vivo studies, the solution was passed
through a silica-gel cartridge (Waters) to yield [11C]MeA in a pre-
parative yield of 70% (decay corrected) after 12 min. An aliquot
of [11C]MeA in acetonitrile was taken and transferred to a solution
of 1 in water and heated to 100 °C for 10 min (Scheme 1). The RCYs
of the triazole formation were 85–90% according to the radio-HPLC
chromatogram (not shown) and preparative yields were deter-
mined to be 60% (decay corrected) after solid phase extraction.
To prove feasibility of this chemistry to the labeling of more
complex molecules such as peptides which are used in nuclear
One original requirement for click-chemistry is the use of non-
toxic solvents such as water and we therefore started our investi-
gation in this solvent. The one-pot labeling approach in water is
based on the in situ conversion of routinely available [11C]MeI to
[
11C]MeA which has been proposed by Sreedhar and Surendra.11
By simply bubbling [11C]MeI12 using a sweep flow of N2 (10 mL/
min) into a mixture consisting of NaN3, CuI, and 1 in water at room
temperature and subsequently heating to 100 °C, 2 was formed to
55–60% according to the radio-HPLC chromatogram (Scheme 1,
Fig. 1). At temperatures below 100 °C the efficiency of triazole for-
mation was constantly decreasing. The non-radioactive standard
compound 2 as a reference compound for HPLC was synthesized
as described.11 Although the one-pot reaction yielding 2 proceeded
with acceptable radiochemical yields, (based on HPLC, Fig. 1) the
initial amount of trapped [11C]MeI in water was unreliable and var-
ied between 60% and 75%. This loss in 11C-radioactivity decreased
preparative yields considerably and additional radioactivity was
also lost during the reaction possibly in form of volatile [11C]MeI
and [11C]MeA accumulating in the small gas phase of the reaction
vessel (15–20% loss). The overall RCY of 2 after solid phase C18 car-
tridge purification was 25% (decay corrected) after 12–14 min.13 In
the absence of 1 and CuI, the reaction mixture showed a new radio-
medicine as imaging agents, we synthesized cyclic D-Glu–D-Tyr–
[Cys–Tyr–Trp–Lys–Thr–Cys]–Thr (a water-soluble derivative of
Tyr3-octreotate,16,17TATE) used as a tumor imaging agent. The pep-
tide was further derivatized at its N-terminus with aminooxy-ace-
tic acid (AOAA) (compound 3, Scheme 2) for the chemo selective
coupling of 4-ethynylbenzaldehyde via oxime formation yielding
compound
[
4 as a labeling precursor for the reaction with
11C]MeA. This method of peptide derivatization has already pro-
ven to be useful in the synthesis of peptidic labeling precursors
for 18F-labeling.18–20 The synthesis of the non-radioactive peptide
5 as a standard compound for HPLC (Scheme 2) was achieved by
reacting
3
with 4-(1-methyl-1H-1,2,3-triazol-4-yl)benzalde-
Figure 1. Top: radio-HPLC chromatogram of the one-pot synthesis of 2 [1:10 mg (0.09 mmol); NaN3: 20–40 mg (0.3–0.6 mmol); CuI: 10 mg (0.052 mmol), water: 1 ml;
t = 10 min, T = 100 °C]. Bottom: radio-HPLC chromatogram of the synthesis of 5 [4: 0.3 mg (0.22 mol); CuI: 2.5 mg (0.013 mmol); Na-ascorbate: 25 mg (0.13 mmol); water
(350 L)/DMF (100 L); t = 10 min; T = 70 °C].
L)/[11C]MeA in CH3CN (100
l
l
l
l

4826
R. Schirrmacher et al. / Tetrahedron Letters 49 (2008) 4824–4827
approach commercially attractive. For final purification, the crude
O
R =
-Glu-
O
reaction mixture was diluted with water (0.5 mL) and passed
through a C18 cartridge to trap the labeled peptide. The unreacted
D
D
-Tyr-[Cys-Tyr-Trp-Lys-Thr-Cys]-Thr
H2N
A
R
[
11C]MeA showed almost no retention (<5%). After elution with
3
O
ethanol (1 mL), the overall preparative, decay corrected yield after
a synthesis time of 30 min was determined to be 32–35% (11.3–
12.4% non-decay corrected after 1.5 half-lifes). The radiochemical
purity was 95% according to HPLC and specific activity was
O
R
H
O
N
CH3CN, H2O (TFA, pH4)
O
O
>25 GBq/lmol (UV-calibration curve).
N
R
[
11C]MeN3
3. Conclusions
o
70 C,
N
N
10 min
In summary, we described the high yield in situ synthesis of
H311C
N
4
5
[
11C]MeA and its applicability in the formation of 11C-labeled tria-
B
42-55% RCY
zoles. We successfully transferred the labeling of the simple model
compound 2 by a two-step labeling approach to the preparation of
Scheme 2. (A) Synthesis of 4 by reacting 3 with ethynylbenzaldehyd in aceto-
nitrile/water (TFA, pH 4); (B) radio-synthesis of 5 [(4 (0.3 mg, 0.22
mol), [11C]MeA
in CH3CN (50–100 L), H2O (350 L), DMF (100 L), DIEA (25 L), CuI (2.5 mg,
0.013 mmol), Na-ascorbate (25 mg, 0.13 mmol), 70 °C, 10 min].
a
11C-labeled TATE derivative potentially useful in nuclear medi-
l
l
cine. The reaction of [11C]MeA with our alkyne-derivatized TATE
demonstrates that reaction conditions for the click step are differ-
ent from those which were already described in 18F-click-chemis-
try.1,2 Triazole formation in this particular case with [11C]MeA
required higher temperatures pointing at a lower reactivity of
l
l
l
hyde11,21 as described for 4. Structural integrity for all non-radioac-
tively synthesized peptides was proven by mass spectroscopy and
purity was determined by HPLC.22 It is noteworthy that unlike for
the synthesis of 18F-labeled peptides,23 a simple labeling procedure
with 11C is not available. Henriksen et al. reported the use of
4-[11C]methoxy-benzaldehyde which was reacted with a carbohy-
drate analogue of TATE in a reaction time of 1 h yielding the
labeled peptide in an overall decay corrected yield of 21% (2.6%
non-decay corrected after 3 half-lifes).24 Two HPLC-purification
steps were needed. It could be demonstrated by the investigators
that their compound, despite the short half life of 11C, shows suit-
able in vivo kinetics for PET imaging of somatostatin-receptor
expressing tumors.
[
11C]MeA. Regardless of this, we demonstrated a simple and fast
conversion of [11C]MeI to [11C]MeA and showed that click-chemis-
try can be applied to the synthesis of short-lived 11C-compounds
avoiding cumbersome workup procedures such as multiple HPLC
purifications.
Acknowledgement
The authors would like to thank the cyclotron personnel at the
Montreal Neurological Institute.
References and notes
In a one-pot labeling approach, peptide 4, NaN3 and Cu(I)I were
mixed in water and [11C]MeI was bubbled through the reaction
mixture as described for the synthesis of 2.25 The RCY for the
formation of 5 was 5% (determined by HPLC, cf. Table 1, entry 1).
Neither lower nor higher reaction temperatures had a positive
effect on the RCYs. Using [11C]MeA in acetonitrile, we applied the
optimized conditions described by Marik and Sutcliffe1 as well as
Glaser and Årstad2 for the formation of 18F-labeled triazoles, but
neither of those procedures yielded 5 in RCY of more than 10%
(Table 1, entries 2 and 3). During the optimization of the two-step
approach, we found that it is crucial to maintain a high water con-
tent while avoiding too much DMF (although in complete absence
of DMF, the yields decreased considerably). Using a minimum
1. Marik, J.; Sutcliffe, J. L. Tetrahedron Lett. 2006, 47, 6681–6684.
2. Glaser, M.; Årstad, E. Bioconjugate Chem. 2007, 18, 989–993.
3. Li, Z.-B.; Wu, Z.; Chen, K.; Chin, F. T.; Chen, X. Bioconjugate Chem. 2007, 18,
1987–1994.
4. Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2001, 40, 2004–
2021.
5. Bock, V.; Hiemstra, H.; von Maarseveen, J. H. Eur. J. Org. Chem. 2006, 51–68.
6. Gothelf, K. V.; Jorgensen, K. A. Chem. Rev. 1998, 98, 863–909.
7. (a) Torne, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057–3064;
(b) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem., Int.
Ed. 2002, 41, 2596–2599.
8. Långström, B.; Lundquist, H. Int. J. Appl. Radiat. Isot. 1976, 27, 357–363.
9. Larsen, P.; Ulin, J.; Dahlström, K.; Jensen, M. Int. J. Appl. Radiat. Isot. 1997, 48,
153–157.
10. For a review of 11C-chemistry: Antoni, G.; Kihlberg, T.; Långström, B. Aspects on
the synthesis of 11C-labelled compounds. In Welch, M., Redvanly, C. S., Eds.;
Handbook of Radiopharmaceuticals, Chemistry and Applications; Wiley:
Chichester, 2003; pp 141–193.
amount of 4 (0.3 mg, 0.22 lmol) 42–55% RCY of 5 were obtained
under optimized conditions after 10 min at 70 °C (Fig. 1, Table 1,
11. Sreedhar, B.; Surendra, P. Synth. Commun. 2007, 37, 805–812; Compound 2 as a
standard compound for HPLC and reactant for the synthesis of non-radioactive
peptide 4 was synthesized according to this reference.
entry 4).26 It is noteworthy that if higher amounts of 4 (0.6–
1 mg, 0.44–0.73 lmol) were used, the RCY of 5 increased consider-
12.
[
11C]MeI was synthesized using a commercially available synthesis unit MEI
ably up to 85% under the same reaction conditions. As most pep-
tides used in nuclear medicine are relatively costly we aimed at
keeping the precursor amount as low as possible to make this
plus by Bioscan. [11C]CO2 (produced by the 14N(p, 11C nuclear reaction using a
a)
cyclotron) is converted into [11C]methanolate with LiAlH4 in anhydrous THF
and further treated with HI to yield [11C]MeI.
Table 1
Synthesis of peptide 5 using different conditions
Entry
Catalytic system
Solvent/[11C]MeA introduction/temperature (°C)
Time (min)
Yieldc (%)
1
2
3
4
CuI
Water, NaN3, [11C]MeI was bubbled into solution, 100 °C
Water/acetonitrile, [11C]MeA in acetonitrile, rta
Water/DMF/[11C]MeA in acetonitrile/DIEA, rtb
Water/DMF/[11C]MeA in acetonitrile/DIEA, 70 °C
10
10
10
10
5
<5
5–10
42–55
CuII/ascorbate
CuI/ascorbate
CuI/ascorbate
a
Conditions from Glaser and Årstad.2
Conditions from Marik and Sutciffe.1
Yields are based on HPLC.
b
c

R. Schirrmacher et al. / Tetrahedron Letters 49 (2008) 4824–4827
4827
13. The reaction mixture was diluted with water (3 mL) and passed over a small
C18 Sep-Pak cartridge (Waters), washed with water (2 mL) and eluted with
CH3CN (1 mL) to give compound 2 in a purity of 98.8% (as determined by
HPLC). HPLC condition: HPLC column: Whatman Partisil 10 ODS2
9.5 ꢁ 250 mm; eluent: CH3CN/0.01 M H3PO4 40/60; flow: 2 mL/min.
14. (a) Feldman, A. K.; Colasson, B.; Fokin, V. V. Org. Lett. 2004, 6, 3897–3899; (b)
Liu, F.; Ma, D. J. Org. Chem. 2007, 72, 4844–4850; (c) Rodionov, V. O.; Presolski,
S. I.; Diaz Diaz, D.; Fokin, V. V.; Finn, M. G. J. Am. Chem. Soc. 2007, 129, 12705–
12712.
21. 1H NMR (600 MHz, CDCl3) d 10 (s, 1H), 7.9–7.99 (m, 4H), 7.85 (s, 1H), 4.16 (s,
1H); 13C (600 MHz, CDCl3) d 37.07, 122.05, 126.18, 130.56, 133.27, 135.95,
136.54, 191.87; MS (CI) m/z 188.13 (M+1).
22. 3, 4 (peptidic part + AOAA) and non-radioactive 5 (peptidic part + AOAA) were
synthesized by Fmoc solid-phase peptide synthesis. Purification of all peptides
was accomplished using the following conditions: Lichrosphere RPselectB C18
100 ꢁ 4.6 (Merck); gradient eluent: 100% H2O + 0.1% TFA after 30 min 100%
CH3CN + 0.1% TFA; flow: 2 mL/min. Labeling precursor 4 and non-radioactive
peptide 5 were synthesized coupling either ethynylbenzaldehyde or non-
15. Naþ=18-crown-6/N3 was prepared by freeze-drying an equimolar solution
radioactive 2 to the N-terminal aminooxy moiety. MALDI-TOF molecular
ꢀ
made from NaN3 and 18-crown-6 in water.
weight determination was carried out using a BRUKER DALTONICS/microflexLT
apparatus. 3 m/z = 1266.3 [M]+ (calcd 1266.5); 4 m/z = 1378.3 [M]+ (calcd
1378.5); compound, 5 m/z = 1435.6 [M]+ (calcd 1435.5).
16. Wängler, B.; Beck, C.; Wagner-Utermann, U.; Schirrmacher, E.; Bauer, C.; Rosch,
F.; Schirrmacher, R.; Eisenhut, M. Tetrahedron Lett. 2006, 47, 5985–5988.
17. Graham, K. A. N.; Wang, Q.; Eisenhut, M.; Haberkorn, U.; Mier, W. Tetrahedron
Lett. 2002, 43, 5021–5024.
23. for review Schirrmacher, R.; Wängler, C.; Schirrmacher, E. Mini-Rev. Org. Chem.
2007, 4, 317–329.
18. Thumshirn, G.; Hersel, U.; Goodmann, H.; Kessler, H. Chem. Eur. J. 2003, 9,
2717.
24. Henriksen, G.; Schottelius, M.; Poethko, T.; Hauser, A.; Wolf, I.; Schwaiger, M.;
Wester, H.-J. Eur. J. Nucl. Med. 2004, 31, 1653–1657.
19. Poethko, T.; Schottelius, M.; Thumshirn, G.; Hersel, U.; Herz, M.; Henriksen,
G.; Kessler, H.; Schwaiger, M.; Wester, H. J. J. Nucl. Med. 2004, 45, 892–902.
20. Schirrmacher, R.; Bradtmöller, G.; Schirrmacher, E.; Thews, O.; Tillmanns, J.;
Siessmeier, T.; Buchholz, H. G.; Bartenstein, P.; Wängler, B.; Niemeyer, C. M.;
Jurkschat, K. Angew. Chem., Int. Ed. 2006, 45, 6047–6051.
25. Peptide 4 (0.3 mg, 0.22 lmol), NaN3 (40 mg, 0.61 mmol), Cu(I)I (10 mg,
0.052 mmol).
26. If 0.3 mg (0.22
l
mol) of 4 were used, the optimized parameters were found to
L)/DMF (100 L), CuI
L)/[11C]MeA in CH3CN (100
(2.5 mg, 0.013 mmol), Na-ascorbate (25 mg, 0.13 mmol).
be as follows: water (350
l
l
l