A one-pot three-component radiochemical reaction for rapid assembly of 125I-labeled molecular probes

Article abstract of DOI:10.1021/ja307926g

Nuclear imaging in conjunction with radioactive tracers enables noninvasive measurements of biochemical events in vivo. However, access to tracers remains limited due to the lack of methods for rapid assembly of radiolabeled molecules with the prerequisite biological activity. Herein, we report a one-pot, three-component, copper(II)-mediated reaction of azides, alkynes, and [ ¹²⁵I]iodide to yield 5-[¹²⁵I]iodo-1,2,3-triazoles. Using a selection of azides and alkynes in a combinatorial approach, we have synthesized a library of structurally diverse ¹²⁵I-labeled triazoles functionalized with bioconjugation groups, fluorescent dyes, and biomolecules. Our preliminary biological evaluation suggests that 5-[¹²⁵I]iodo-1,2, 3-triazoles are resistant to deiodination in vivo, both as small molecular probes and as antibody conjugates. The ability to incorporate radioactive iodide into triazoles directly from the parent azides and alkynes makes the method broadly applicable and offers the potential to rapidly assemble molecular probes from an array of structurally diverse, and readily available, building blocks.

Full text of DOI:10.1021/ja307926g

Article
pubs.acs.org/JACS
A One-Pot Three-Component Radiochemical Reaction for Rapid
125
Assembly of I‑Labeled Molecular Probes
Ran Yan,†,⊥ Kerstin Sander,† Eva Galante,† Vineeth Rajkumar,‡ Adam Badar,§ Mathew Robson,‡
Ethaar El-Emir,‡ Mark F. Lythgoe,§ R. Barbara Pedley,‡ and Erik Årstad*,†
†Institute of Nuclear Medicine and Department of Chemistry, University College London, 235 Euston Road (T-5), London NW1
2BU, U.K.
‡UCL Cancer Institute, University College London, 72 Huntley Street, London WC1E 6BT, U.K.
§Centre for Advanced Biomedical Imaging, Division of Medicine, University College London, 72 Huntley Street, London WC1E
6BT, U.K.
S
* Supporting Information
ABSTRACT: Nuclear imaging in conjunction with radio-
active tracers enables noninvasive measurements of biochem-
ical events in vivo. However, access to tracers remains limited
due to the lack of methods for rapid assembly of radiolabeled
molecules with the prerequisite biological activity. Herein, we
report a one-pot, three-component, copper(II)-mediated
reaction of azides, alkynes, and [125I]iodide to yield 5-
[
125I]iodo-1,2,3-triazoles. Using a selection of azides and
alkynes in a combinatorial approach, we have synthesized a
library of structurally diverse 125I-labeled triazoles function-
alized with bioconjugation groups, fluorescent dyes, and
biomolecules. Our preliminary biological evaluation suggests
that 5-[125I]iodo-1,2,3-triazoles are resistant to deiodination in vivo, both as small molecular probes and as antibody conjugates.
The ability to incorporate radioactive iodide into triazoles directly from the parent azides and alkynes makes the method broadly
applicable and offers the potential to rapidly assemble molecular probes from an array of structurally diverse, and readily available,
building blocks.
increased metabolic stability.1−4 However, the need to
INTRODUCTION
■
independently prepare nonradioactive reference compounds
and precursors for labeling makes structural alterations tedious
and preparation of tracers with complex functional groups
difficult.
Nuclear imaging in conjunction with radioactive tracers enables
noninvasive measurements of biochemical events in vivo, and
has become an invaluable tool for biological research and to
diagnose disease. Yet tracer development remains exceedingly
challenging as few methods are available that allow rapid
assembly of radiolabeled molecules with the prerequisite
biological activity. While the optimal choice of radionuclide
depends on the intended application, as well as the structural
requirements of the biological target, the numerous isotopes of
iodine available support widespread use across biological
research (125I), diagnostic imaging (123,124I), and radiotherapy
We envisaged that the copper-catalyzed reaction of azides,
alkynes, and electrophilic iodine to give 5-iodo-1,2,3-triazoles
could give access to highly functionalized radiolabeled tracers in
a single step (Scheme 1).5 In our initial report, we
demonstrated the concept using a copper(II)-mediated one-
pot reaction to synthesize a 125I-labeled trifunctionalized
reagent for multiscale imaging with optical and nuclear
techniques.6 Here, we disclose a broadly applicable and highly
efficient method that offers the potential to rapidly assemble an
array of structurally diverse 5-[125I]iodo-1,2,3-triazoles directly
from the parent azides and alkynes. Imaging studies with single
photon emission tomography (SPECT) suggest that the
resulting 5-[125I]iodo-1,2,3-triazoles are resistant to deiodina-
tion in vivo.
(
131I).1
Labeling with radioiodine is largely restricted to electrophilic
reactions with activated aromatic groups, and iodo-demetala-
tion of aryls and alkenes, using organotin or silicon
precursors.1−3 Oxidative addition of radioiodine to electron-
rich aromatic moieties offers a convenient method for labeling,
but the resulting tracers are prone to rapid deiodination in vivo,
limiting the practical applications to tyrosine-containing
peptides and proteins.4 The use of organometallic precursors
is often preferred as it enables labeling of nonactivated aromatic
groups, as well as alkenes, and allows design of tracers with
Received: August 9, 2012
Published: November 29, 2012
© 2012 American Chemical Society
703
dx.doi.org/10.1021/ja307926g | J. Am. Chem. Soc. 2013, 135, 703−709

Journal of the American Chemical Society
Article
of triazole [125I]3b in 57 4% RCY (entry 2). Phenylacetylene
(2c) and propargyl alcohol (2d) gave poor RCYs of the
corresponding triazoles [125I]3c and [125I]3d (9 1% and 27
19%, respectively, entries 3 and 4), whereas the propargyl
amide 2e afforded triazole [125I]3e in 60 8% RCY (entry 5).
To investigate the impact of the chemical environment on the
reactivity of azides, we treated the propargyl amide 2e with
phenyl azide (1b), and the aliphatic ester 1c, which provided
the resulting triazoles [125I]3f and [125I]3g in disappointing
RCYs of 3 1% and 17 2%, respectively (entries 6 and 7).
Whereas the alkynes 2b and 2e gave comparable yields when
reacted with benzyl azide (1a), the use of N-propargyl
maleimide (2b) in combination with the aliphatic azide 1c
Scheme 1. Synthesis of 5- [125,127I]Iodo-1,2,3-triazoles
RESULTS AND DISCUSSION
led to a marked increase in RCY (30
2%, entry 8) as
■
Synthesis. To evaluate the impact of the chemical
environment on the labeling efficiency, we investigated the
reactivity of azides and alkynes bearing aliphatic, benzylic,
aromatic, and heterocyclic moieties. Azides 1a−f and alkynes
2a−f (Table 1) were synthesized as previously described or
obtained from commercial sources.6−10 The azide 1g was
obtained in 40% overall yield from 1-benzhydryl-piperazine by
alkylation with 2-bromoethanol, reaction of the corresponding
alcohol with thionyl chloride, and substitution of the resulting
chloride with azide. Reaction of biotin with thionyl chloride and
subsequent treatment of the acyl chloride with N-methyl-
propargylamine provided alkyne 2g in 65% yield. Treatment of
6-chloropurine with sodium hydride, followed by alkylation
with propargyl bromide gave alkyne 2h in 58% yield. A
modified literature method was adopted for the preparation of
the nonradioactive 5-iodo-1,2,3-triazoles 3a−n.5 Treatment of
the respective azides 1a−g and alkynes 2a−h with copper(I)
iodide, triethylamine (TEA), and N-iodosuccinimide (NIS)
provided the target compounds 3a−n (Scheme 1), albeit in
moderate to poor yields. While we aimed to rapidly access the
nonradioactive reference compounds for structural character-
ization, it should be noted that other synthetic methods have
been reported to afford 5-iodo-1,2,3-triazoles in good to
excellent yields.5,11,12
compared with that using 2e.
Our initial results were perplexing, as none of the model
compounds investigated gave comparable RCYs to those of the
azide 1d and alkyne 2f, which were employed for the
explorative studies. However, we noted that for several low-
yielding reactions a precipitate gradually formed, suggesting
that the solubility of the reactants strongly influences the
outcome of the reaction. Heating of the reaction mixtures to 60
°C was sufficient to retain homogeneous solutions and
gratifyingly led to dramatic increases in RCYs. With the
exception of triazoles [125I]3f and [125I]3g, which were
obtained in a moderate 39 1% and 54 16% RCYs (entries
6 and 7), reactions of the azides 1a and 1c with the alkynes 2a−
e for 90 min at 60 °C provided the corresponding triazoles
[
125I]3a−e and [125I]3h in excellent to near quantitative RCYs
(entries 1−5 and 8). While it is conceivable that the increased
reactivity observed with heating is mediated by several factors,
including kinetic energy and concentration of the reaction
mixture within the reaction vial, the moderate effect of heating
on the RCYs for triazoles [125I]3j and [125I]3m (entries 10 and
13) is supportive of the notion that solubility is a deciding
factor for labeling efficiency.
Encouraged by the results, we set out to prepare [125I]-
iodotriazoles, functionalized with selected bioconjugation
groups, fluorescent dyes, and biomolecules, as putative
molecular probes for imaging with SPECT. The target
compounds [125I]3i−m were all obtained in excellent RCYs
(entries 9−13), demonstrating the versatility and robustness of
the reaction. The rhodamine derivative [125I]3i was prepared as
a candidate tracer for imaging of mitochondrial membrane
potentials,13 whereas the biotin-derived triazole [125I]3k was
designed for pretargeting strategies in conjunction with avidin,
or streptavidin, and may find use for imaging (123,124,125I) as well
as radioimmunotherapy (131I).14,15 The dual optical and nuclear
labeling reagents [125I]3j and [125I]3l were prepared to give
rapid access to protein-derived tracers for multiscale imaging
through lysine or thiol conjugation, respectively. The 6-
chloropurine [125I]3m may enable imaging of drug efflux
transporter function16,17 and exemplifies the potential to
prepare tracers suitable for labeling with 18F, as well as
radioiodine. Finally, to investigate if a copper-chelating group,
i.e. piperazine, is tolerated in the radiochemical reaction we
prepared [125I]3n, a derivative of the first-generation antihist-
amine hydroxyzine. Despite attempts to optimize the labeling
conditions, low RCYs were observed (15 19%, entry 14), and
the reproducibility was poor, pointing to a limitation with this
method. In preparative runs using 25−95 MBq of [125I]NaI, the
triazoles [125I]3i−l where obtained in 70−80% RCY after
Radiochemistry. Initially, we used the azide 1d and alkyne
2f (Table 1, entry 10) to screen a number of copper salts,
chelators, oxidizing agents, and solvents, with the aim of
identifying a catalytic system that would allow formation of
[
125I]iodotriazole 3j directly from aqueous non-carrier-added
(n.c.a.) [125I]NaI. While several oxidants were effective in
combination with copper(I) salts, including iron(III) chloride,
the use of copper(II) chloride in combination with TEA
provided a convenient source of both oxidant and catalyst for
the reaction. Under the optimized conditions, the triazole [125I]
3j was formed in 80 3% analytical radiochemical yield (RCY)
by reacting azide 1d, alkyne 2f (1.0 equiv), copper(II) chloride
(1.0 equiv), and TEA (1.5 equiv) with sodium [125I]iodide in a
solution of water and acetonitrile (1:10) for 90 min at room
temperature. Formation of the corresponding non-iodinated
triazole was sluggish, and the azide 1d and alkyne 2f remained
largely intact in the reaction mixture.
To explore the scope of the method, we subjected the
alkynes 2a−e to the labeling conditions described above, using
benzyl azide (1a) as the “click partner”. Unexpectedly, the RCY
for the triazole [125I]3a (20 2%, entry 1) was substantially
lower than that for [125I]3j, despite the near identical chemical
environment in the immediate surroundings of reactive groups
of the azides 1a and 1d, and the alkynes 2a and 2f. N-Propargyl
maleimide (2b) proved more reactive and led to the formation
704
dx.doi.org/10.1021/ja307926g | J. Am. Chem. Soc. 2013, 135, 703−709

Journal of the American Chemical Society
Article
d
Table 1. Radiochemical Yields for 5-[125I]Iodo-1,2,3-triazoles
a
b
c
d
Analytical radiochemical yields standard deviation. n = 8. 0.5 equiv of TEA was used, n = 6. All reactions were carried out using 2−5 MBq
125I]NaI and 1.0 μmol of the azides and alkynes in the presence of copper(II) chloride (1.0 equiv) and TEA (1.5 equiv) at room temperature (yields
in brackets) or with heating to 60 °C for 90 min. All experiments were carried out in triplicates unless otherwise specified.
[
HPLC purification, with a specific activity in the range of 2−3
GBq/μmol.
MBq [125I]NaI. When the water content was increased from 6
to 12 μL, the RCY of [125I]3j decreased sharply from 80 3%
to 31% (entry 2). Doubling the volume of the solvent, while
keeping the ratio of water and acetonitrile to 1:10, had a
similarly detrimental effect (entries 3 and 4). The most
practical solution for preparing higher activity levels of [125I]3j
was therefore to increase the amounts of reagents used from 1.0
μmol to 2.0 μmol, which enabled us to use up to 12 μL/180
MBq of aqueous [125I]NaI. Starting with 100 MBq of [125I]NaI,
As radioactive iodide is supplied in aqueous solutions, the
amount of radioactivity that can be used, if a concentration step
is to be avoided, is governed by the total volume of water that
can be added to the reaction mixture. Using the formation of
[
125I]3j as an example, we investigated how the water content,
the concentration of the reagents, and the temperature affect
the labeling efficiency (Table 2). During our explorative work,
we carried out the reactions on a 1.0 μmol scale in a total
volume of 66 μL of water and acetonitrile (1:10), using 2−5
[
125I]3j was obtained in 74 4% isolated RCY with a specific
activity of 9 GBq/μmol (entry 5). At this scale, 5-[125I]iodo-
705
dx.doi.org/10.1021/ja307926g | J. Am. Chem. Soc. 2013, 135, 703−709

Journal of the American Chemical Society
Article
the copper-catalyzed cycloaddition reaction (Scheme 2, path-
way A).
Table 2. Influence of Water Content and Concentration on
RCYs for Formation of Labeling Reagent [125I]3j
In order to explore this possibility further, we attempted to
react the azide 1a with 1-([125I]iodoethynyl)benzene ([125I]4)
in the presence of copper powder as well as several copper(I)
and copper(II) salts; however, the iodoalkyne [125I]4 rapidly
decomposed under these conditions (30 min at 60 °C). When
c
entry
MeCN:water (μL)
temp. (°C)
RCY (%)
a
1
60: 6
RT
80 3 (n = 8)
d
72 4 (n = 5)
a
2
60:12
120:12
120:12
120:12
RT
RT
60
31
37
a
[
125I]4 was prepared in situ with copper(II) chloride and TEA
3
a
prior to addition of azide 1a, the iodoalkyne [125I]4 remained
intact. Interestingly, all the aforementioned sources of copper
mediated rapid scrambling of iodine when added to a mixture
of [125I]NaI and the nonradioactive iodoalkyne 4. In the
presence of the azide 1a, copper(I) iodide yielded both the
iodotriazole [125I]3c (12% RCY) and [125I]4 (8% RCY).
Comparable results were obtained with copper powder and
copper(I) oxide in the absence of TEA, with formation of
iodoalkyne [125I]4 (50% and 62% RCY) as well as the
iodotriazole [125I]3c (10% and 19% RCY). In contrast,
copper(I) and copper(II) chloride gave [125I]4 as the sole
radiochemical product in 46% and 76% RCY, respectively.
While we cannot exclude the possibility that iodoalkynes are
intermediates in the reaction, the failure of copper(I) and
copper(II) chloride to catalyze the cycloaddition of [125I]4 and
azide 1a makes this unlikely.
Under inert atmosphere, complexes of copper(II) chloride
and TEA can undergo reduction to copper(I), yet in the
presence of air they rapidly revert back to their 2+ oxidation
state.18,19 To investigate the oxidation state of the catalytic
system, hydrogen peroxide (1.0 equiv) was added to the
copper(II) chloride/TEA complex prior to addition of the
alkyne 2c. Subsequent reaction with the azide 1a and [125I]NaI
for 30 min at 60 °C led to formation of 1-([125I]iodoethynyl)-
benzene ([125I]4) in quantitative RCY. The absence of the
triazole [125I]3c from the reaction mixture implies that
copper(I) is required for the cycloaddition reaction to occur.
Yet, addition of the alkyne 2c to a solution of copper(I)
chloride and TEA in acetonitrile results in immediate
precipitation of the corresponding copper(I) acetylide as a
yellow solid. As a homogeneous burgundy solution is obtained
when copper(II) chloride is used, the corresponding TEA
complex is likely to be predominantly in the 2+ oxidation state,
while still providing a source of copper(I).
4
53 13 (n = 3)
87 3 (n = 6)
b
5
RT
d
74 4 (n = 6)
a
A solution of rhodamine 2f (1.0 μmol), CuCl2 (1.0 μmol) and TEA
(1.5 μmol) was added to [125I]NaI in water before addition of azide 1d
b
(1.0 μmol) and left to react for 90 min. The reaction was scaled up to
c
2.0 μmol. RCY is the mean value of n experiments
deviation. When not specified, n = 1. Isolated RCY.
standard
d
1,2,3-triazoles can readily be prepared in sufficient activity levels
for preclinical imaging studies; however, scale-up for clinical
applications may require aqueous solutions of radioiodine to be
concentrated before use.
Mechanistic Studies. When optimizing the labeling of
[
125I]3j we found that the addition sequence of the reagents
was important, with the highest yields obtained when the
alkyne 2f, copper(II) chloride, and TEA were combined prior
to addition of [125I]NaI and the azide 1d. Interestingly, in the
absence of the azide 1d a new radiolabeled compound was
formed in 10−60% RCY that previously had only been
observed as a minor side product (<2% RCY). To investigate
the nature of the side product we used the alkyne 2c as a model
compound. When 2c was treated with copper(II) chloride,
TEA, and [125I]NaI at ambient temperature, no radiochemical
reaction took place. However, heating the reaction mixture to
60 °C for 90 min led to formation of 1-([125I]iodoethynyl)-
benzene ([125I]4) (Scheme 2) in 84 3% (n = 3) RCY. The
Scheme 2. Potential Mechanistic Pathways for Copper(II)-
Mediated Formation of 5- [125I]Iodo-1,2,3-triazoles
The mechanism for formation of 5-iodo-1,2,3-triazoles from
azides and alkynes remains elusive, and none of the pathways
suggested to date can fully explain the reported experimental
data.20 In their pioneering work, Wu et al. proposed formation
of an intermediate copper(I) triazolyl complex, which upon
reaction with electrophiles gives rise to 1,4,5-trisubstituted-
1,2,3-triazoles.5 More recently, Buckley et al. demonstrated that
dinuclear copper(I) ladderane complexes can catalyze the
reaction of azides and iodoalkynes in the presence of TEA to
give 5-iodotriazoles.21 To account for this observation it was
proposed that nucleophilic amines mediate transfer of iodine
from iodoalkynes to a dinuclear copper(I) triazolyl inter-
mediate, and that the liberated terminal alkyne serves to
regenerate the catalytic complex.
Under the conditions described here, it is plausible that
[
125I]NaI is oxidized in the presence of copper(II) chloride and
identity of the radioactive product was confirmed by coelution
with the nonradioactive reference compound. The results bear
a striking resemblance to the RCYs obtained when alkyne 2c
TEA, and that electrophilic 125I subsequently undergoes metal−
halide exchange with a copper(I) triazolyl complex (Scheme 2,
pathway B).22 Indeed, oxidation of [125I]NaI with hydrogen
peroxide (0.36 equiv) followed by addition of copper(I)
chloride, TEA, azide 1a, and the alkyne 2c provided the triazole
was reacted with azide 1a (9
1% and 99
2% at room
temperature and with heating to 60 °C, respectively, Table 1,
entry 3), suggesting that 1-iodoalkynes may be intermediates in
706
dx.doi.org/10.1021/ja307926g | J. Am. Chem. Soc. 2013, 135, 703−709

Journal of the American Chemical Society
125I]3c. While the RCY was lower than that obtained with
Article
[
copper(II) chloride and [125I]NaI (24% vs 45%), precipitation
of the copper(I) acetylide is likely to have impaired the reaction
rate. However, this pathway does not explain why formation of
5-[125I]iodotriazoles is favored over the corresponding 5-
prototriazoles despite the low concentration of n.c.a. [125I]NaI.
While we observed efficient transfer of [125I]iodide to (1-
iodoethynyl)benzene 4 in the presence of copper(II) chloride
and TEA, no further reaction took place in the presence of
azide 1a. This finding contradicts the pathway proposed by
Buckley et al.,21 since reaction of TEA with the iodoalkyne 4 to
give a triethyliodoammonium ion would also provide a source
of the alkyne 2c and hence should result in formation of the
iodotriazole [125I]3c. Instead, under the conditions reported
here, iodide may act as a bridging ligand to promote formation
of a dinuclear copper(I) ladderane complex with enhanced
reactivity (Scheme 2, pathway C).23 Subsequent reaction with
an azide would result in formation of a dinuclear copper(I)−
triazolyl complex with iodide ideally positioned to undergo
sequential oxidation and electrophilic substitution, providing
the resulting 5-iodotriazole. This pathway can explain the high
selectivity for formation of iodotriazoles, and also provides a
rationale for why triiodide ions are more reactive than iodine in
this reaction.24
Figure 1. SPECT/CT image (A) of subcutaneous SW1222 tumor (1
cm3, indicated by arrow) 5 days after administration of 6.8 MBq [125I]
3j/A5B7. The color bar represents counts per second. (B) Confocal
image of a SW1222 xenograft section (100 μm × 350 μm) stained
with 3j/A5B7.
that achieved when labeling A5B7 with a commercially available
dye (Alexa Fluor 546), suggesting that 3j may find use as a low-
cost alternative for fluorescent labeling of antibodies.
Preliminary Biological Evaluation. To explore the
potential applications of 5-[125I]iodo-1,2,3-triazoles for in vivo
imaging, we conjugated the dual optical and nuclear labeling
reagent [125I]3j to A5B7, a mouse monoclonal antibody (mAb)
specific to carcinoembryonic antigen (CEA). CEA is expressed
by most gastrointestinal tumors, and A5B7 has been extensively
used for radioimmunotherapy of colorectal cancer in preclinical
and clinical studies.25−27 Incubation of n.c.a. [125I]3j with A5B7
for 1 h provided the [125I]3j/A5B7 conjugate in 26% RCY after
purification with a size exclusion cartridge (80% protein
recovery). Addition of excess nonradioactive 3j (20 equiv) to
the reaction mixture had a limited effect on the conjugation
efficiency, and provided A5B7 with an average of six to eight
fluorescent groups per antibody (22% RCY).6
As previously reported, in mice bearing human colorectal
xenografts (SW1222), the biodistribution of [125I]3j/A5B7 was
similar to that of the antibody when it was labeled with 125I
alone using the Chloramine-T method. However, conjugation
of [125I]3j to A5B7 reduced the tumor uptake and also led to
increased liver uptake and more rapid blood clearance.6 In
preliminary studies, we did not see any correlation between the
distribution of [125I]3j/A5B7 and the number of fluorescence
groups (1−6) per antibody, suggesting that the biological
properties of A5B7 are affected by the conditions used for
lysine conjugation rather than by the physiochemical properties
of the labeling reagent [125I]3j. In this respect, it is worth noting
that the liver uptake of [125I]3j/A5B7 (6.99 0.16% ID/g 48 h
postinjection) after conjugation with 6−8 fluorescent groups
per antibody still compares favorably to that of other
antibodies, such as [64Cu]rituximab, that have been modified
by lysine conjugation.28
The propensity of iodinated compounds to undergo
deiodination in vivo is limiting the use of radioiodine for
biomedical applications. To assess the metabolic stability of 5-
[
125I]iodo-1,2,3-triazole-derived small-molecule tracers we
compared the distribution of rhodamine [125I]3i and biotin
[
125I]3k to that of [125I]FIAU and [125I]iodide using SPECT/
CT (Figure 2). Images were acquired at 30 min and 2 h
postinjection, using identical threshold levels to allow
comparison of activity distribution across one order of
magnitude. The uptake of radioactivity in the thyroid was
used as a measure of deiodination in vivo. The rhodamine-
derived tracer [125I]3i showed high initial uptake in the spleen,
with widespread distribution of activity in the abdomen at the
later time point (Figure 2, A and B). The biotin derivative [125I]
3k was rapidly excreted to the intestines and showed negligible
uptake in other tissues (Figure 2, C and D). In contrast,
administration of [125I]FIAU, a nucleoside-derived tracer for
imaging of herpes simplex virus type 1 thymidine kinase
(HSV1-tk),29,30 led to rapid accumulation of activity in the
thyroid, stomach and bladder (Figure 2, E and F).
Administration of [125I]iodide gave a distribution pattern
almost indistinguishable from [125I]FIAU (Figure 2, G and H).
At 30 min p.i., the thyroid uptake after administration of
[
125I]iodide was 7.49% of the whole body activity. In
comparison, the corresponding values for the rhodamine
125I]3i, the biotin [125I]3k, and [125I]FIAU were 0.92%,
[
0.51% and 4.22%, respectively. It should be noted that the
metabolic fate of radiolabeled tracers depends on a number of
factors, including physiochemical and biological properties, and
that the rate of deiodination varies considerably even for
compounds bearing iodine on comparable chemical moieties.
In addition, scatter and attenuation limit the quantitative
information that can be derived from SPECT. Nonetheless, the
low uptake of activity in the thyroid, stomach, and bladder
following administration of [125I]3i and [125I]3k suggests that
5-[125I]iodo-1,2,3-triazoles are resistant to deiodination in vivo.
Recently, Valliant and co-workers reported the synthesis and
in vivo distribution of a disubstituted [123I]iodotriazole (I-
TAAG) and its glutamate-urea-lysine conjugate.31 Adminis-
SPECT imaging of [125I]3j/A5B7 demonstrated high
retention in the tumor over the course of the experiment,
with gradual clearance in off-target tissues, resulting in excellent
contrast 5 days postinjection (Figure 1A). In separate
experiments, nonradioactive 3j/A5B7 bearing 6−8 fluorescent
groups per antibody was used to image CEA expression on a
SW1222 xenograft section (Figure 1B). In our hands, the
contrast and fluorescence intensity observed was comparable to
707
dx.doi.org/10.1021/ja307926g | J. Am. Chem. Soc. 2013, 135, 703−709

Journal of the American Chemical Society
Article
Figure 2. SPECT/CT images of BALB/c mice. Color bars represent counts per second with maximum and minimum threshold levels indicated. (A)
Rhodamine [125I]3i at 30 min and (B) 2 h p.i. showing initial uptake in the spleen with widespread distribution in the abdomen at the later time
point. (C) Biotin [125I]3k at 30 min and (D) 2 h postinjection (p.i.) showing rapid and almost exclusive excretion to the intestines. (E) [125I]FIAU at
30 min and (F) 2 h p.i. demonstrating rapid deiodination with uptake in the stomach and thyroid, and excretion to the bladder. (G) For comparison,
the distribution of sodium [125I]iodide is shown at 30 min and (H) 1 h p.i.
tration of I-TAAG in mice led to significant accumulation of
radioactivity in the thyroid within 1 h, indicative of rapid
deiodination in vivo. While further studies are required to
elucidate the biological properties of iodotriazoles, the
increased steric hindrance of trisubstituted triazoles may
account for the discrepancy in metabolic stability between I-
TAGG, and the tracers reported herein.
ASSOCIATED CONTENT
* Supporting Information
Experimental data for the synthetic and radiochemical
procedures, spectroscopic data, radio-HPLC characterization,
and imaging protocols are provided. This material is available
free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
e.arstad@ucl.ac.uk
■
CONCLUSIONS
■
We have developed a one-pot, copper(II)-mediated reaction
that provides 5-[125I]iodo-1,2,3-triazoles directly from aqueous
125I]iodide and the parent alkynes and azides. The method is
Present Address
⊥King’s College London, Division of Imaging Sciences and
Biomedical Engineering Rayne Institute, St. Thomas’ Hospital,
London SE1 7EH, U.K. E-mail: ran.yan@kcl.ac.uk.
[
technically straightforward, robust, versatile, and high yielding,
and provides rapid access to highly functionalized tracers. Using
a combinatorial approach, we have prepared an array of
Notes
The authors declare no competing financial interest.
[
125I]iodotriazoles decorated with bioconjugation groups,
ACKNOWLEDGMENTS
■
fluorescent dyes, and biomolecules. Our preliminary biological
evaluation indicates that 5-[125I]iodo-1,2,3-triazoles are resistant
to deiodination in vivo, both as small molecular probes and as
antibody conjugates. The added flexibility of trisubstituted
triazoles, and their apparent resistance to metabolic deiodina-
tion, has the potential to transform the development of tracers
for biomedical imaging and therapeutic applications.32
This work was undertaken at UCLH/UCL which is funded in
part by the Department of Health’s NIHR Biomedical Research
Centres funding scheme (E.Å., R.Y.). The research leading to
these results has received funding from the European
Community’s Seventh Framework Programme (FP7/2007-
2013) under Grant Agreement Number 201380 (“Euripides”)
(K.S., E.G.). We acknowledge support from Cancer Research
708
dx.doi.org/10.1021/ja307926g | J. Am. Chem. Soc. 2013, 135, 703−709

Journal of the American Chemical Society
Article
(28) Cooper, M. S.; Ma, M. T.; Sunassee, K.; Shaw, K. P.; Williams, J.
D.; Paul, R. L.; Donnelly, P. S.; Blower, P. J. Bioconjugate Chem. 2012,
23, 1029−1039.
(29) Tjuvajev, J. G.; Doubrovin, M.; Akhurst, T.; Cai, S. D.; Balatoni,
J.; Alauddin, M. M.; Finn, R.; Bornmann, W.; Thaler, H.; Conti, P. S.;
Blasberg, R. G. J. Nucl. Med. 2002, 43 (8), 1072−1083.
(30) Kim, E. J.; Hong, S. H.; Choi, T. H.; Lee, E. A.; Kim, K. M.; Lee,
K. C.; An, G. I.; El-Gamal, M. I.; Cheon, G. J.; Choi, C. W.; Lim, S. M.
Appl. Radiat. Isot. 2010, 68 (6), 971−978.
(31) Darwish, A.; Blacker, M.; Janzen, N.; Rathmann, S. M.; Czorny,
S.; Hillier, S. M.; Joyal, J. L.; Babich, J. W.; Valliant, J. F. ACS Med.
Chem. Lett. 2012, 3 (4), 313−316.
(32) Årstad, E.; Yan, R. Processes for producing radiohalogenated
bioconjugates and products thereof. WO/2012/022932 (A1), 2012.
UK (R.B.P., M.R.), European Union Seventh Framework
Programme Grant 201342, and ADAMANT (E.E-E.), and by
King’s College London, UCL Comprehensive Cancer Imaging
Centre CR-UK, and EPSRC, in association with the MRC and
DoH (England) (V.R.).
REFERENCES
■
(1) Adam, M. J.; Wilbur, D. S. Chem. Soc. Rev. 2005, 34, 153−163.
(2) Seevers, R. H.; Counsell, R. E. Chem. Rev. 1982, 82, 575−590.
(3) Årstad, E.; Hoff, P.; Skattebøl, L.; Skretting, A.; Breistøl, K. J.
Med. Chem. 2003, 46, 3021−3032.
(4) Vaidyanathan, G.; Zalutsky, M. R. Nat. Protoc. 2007, 2 (2), 282−
286.
(5) Wu, Y. M.; Deng, J.; Li, Y.; Chen, Q. Y. Synthesis 2005, 18, 1314−
1318.
(6) Yan, R.; El-Emir, E.; Rajkumar, V.; Robson, M.; Jathoul, A. P.;
Pedley, R. B.; Årstad, E. Angew. Chem., Int. Ed. 2011, 50, 6793−6795.
́ ́
(7) Kele, P.; Li, X.; Link, M.; Nagy, K.; Herner, A.; Krisztian, L.; Beni,
S.; Wolfbeis, O. S. Org. Biomol. Chem. 2009, 7, 3486−3490.
(8) Stephen, A.; Hashmi, K.; Molinari, L.; Rominger, F.; Oeser, T.
Eur. J. Org. Chem. 2011, 12, 2256−2264.
(9) Link, M.; Li, X.; Kleim, J.; Wolfbeis, O. S. Eur. J. Org. Chem. 2010,
36, 6922−6927.
(10) Glaser, M.; Årstad, E. Bioconjugate Chem. 2007, 18 (3), 989−
993.
(11) Li, L.; Zhang, G.; Zhu, A.; Zhang, L. J. Org. Chem. 2008, 73,
3630−3633.
(12) Hein, J. E.; Tripp, J. C.; Krasnova, L. B.; Sharpless, K. B.; Fokin,
V. V. Angew. Chem., Int. Ed. 2009, 48, 8018−8021.
(13) Zhou, Y.; Kim, Y. S.; Yan, X.; Jacobson, O.; Chen, X.; Liu, S.
Mol. Pharm. 2011, 8, 1198−1208.
(14) Lesch, H. P.; Kaikkonen, M. U.; Pikkarainen, J. T.; Yla-
Herttuala, S. Expert Opin. Drug Delivery 2010, 7 (5), 551−564.
(15) Mohsin, H.; Jia, F.; Bryan, J. N.; Sivaguru, G.; Cutler, C. S.;
́
Ketring, A. R.; Miller, W. H.; Simon, J.; Frank, R. K.; Theodore, L. J.;
Axworthy, D. B.; Jurisson, S. S.; Lewis, M. R. Bioconjugate Chem. 2011,
22, 2444−2452.
(16) Okamura, T.; Kikuchi, T.; Okada, M.; Toramatsu, C.; Fukushi,
K.; Takei, M.; Irie, T. J. Cereb. Blood Flow Metab. 2009, 29, 504−511.
(17) Okamura, T.; Kikuchi, T.; Toramatsu, C.; Fuku-shi, K.; Irie, T. J.
Med. Chem. 2009, 52, 7284−7288.
(18) Yoke, J. T.; Weiss, J. F.; Tollin, G. Inorg. Chem. 1963, 2 (6),
1210−1216.
(19) Weiss, J. F.; Tollin, G.; Yoke, J. T. Inorg. Chem. 1964, 3 (10),
1344−1348.
(20) Spiteri, C.; Moses, J. E. Angew. Chem., Int. Ed. 2010, 49, 31−33.
(21) Buckley, B. R.; Dann, S. E.; Heaney, H. Chem.Eur. J. 2010, 16,
6278−6284.
(22) This mechanistic pathway was proposed by a reviewer, who also
suggested several of the experiments described.
(23) Hein, J. E.; Fokin, V. V. Chem. Soc. Rev. 2010, 39 (4), 1302−
1315.
(24) Brotherton, W. S.; Clark, R. J.; Zhu, L. J. Org. Chem. 2012, 77,
6443−6455.
(25) El-Emir, E.; Qureshi, U.; Dearling, J. L. J.; Boxer, G. M.;
Clatworthy, I.; Folarin, A. A.; Robson, M. P.; Konerding, M. A.; Nagl,
S.; Pedley, R. B. Cancer Res. 2007, 67, 11896−11905.
(26) Fidarova, E. F.; El-Emir, E.; Boxer, G. M.; Qureshi, U.; Robson,
M. P.; Dearling, J. L. J.; Trott, K. R.; Begent, R. H. J.; Pedley, R. B.
Clin. Cancer Res. 2008, 14, 2639−2646.
(27) Meyer, T.; Gaya, A. M.; Dancey, G.; Stratford, M. R. L.;
Othman, S.; Sharma, S. K.; Wellsted, D.; Taylor, N. J.; Stirling, J. J.;
Poupard, L.; Folkes, L. K.; Chan, P.-S.; Pedley, R. B.; Chester, K. A.;
Owen, K.; Violet, J. A.; Malaroda, A.; Green, A. J.; Buscombe, J.;
Padhani, A. R.; Rustin, G. J.; Begent, R. H. Clin. Cancer Res. 2009, 15,
4484−4492.
709
dx.doi.org/10.1021/ja307926g | J. Am. Chem. Soc. 2013, 135, 703−709