Nitroimidazoles and hypoxia imaging: Synthesis of three Technetium-99m complexes bearing a nitroimidazole group: Biological results

Article abstract of DOI:10.1016/S0960-894X(00)00593-X

Several Tc-99m complexes were synthesized, substituted with a nitroimidazole group, in order to visualize hypoxic tissues. The complexes were tested on rats (isolated hearts) and showed no significant uptake under hypoxic conditions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00593-X

Bioorganic & Medicinal Chemistry Letters 11 (2001) 71±74
Nitroimidazoles and Hypoxia Imaging: Synthesis of Three
Technetium-99m Complexes Bearing a Nitroimidazole Group:
Biological Results
, * Amaury du Moulinet d'Hardemare, Sandrine Se
 Á
pe,a
a,
a
FrancË
oise Riche
Laurent Riou, Daniel Fagret and Michel Vidal
b
b
a
aLaboratoire d'Etudes Dynamiques et Structurales de la SeÂlectivit e , Universit e Joseph Fourier, UMR 5616 - BP 53,
8041 Grenoble Cedex 9, France
Laboratoire d'Etudes des Radiopharmaceutiques, Universite Joseph Fourier, Domaine de la Merci, INSERM,
8700 La Tronche, France
3
b
3
Received 5 July 2000; revised 6 October 2000; accepted 23 October 2000
AbstractÐSeveral Tc-99m complexes were synthesized, substituted with a nitroimidazole group, in order to visualize hypoxic tis-
sues. The complexes were tested on rats (isolated hearts) and showed no signi®cant uptake under hypoxic conditions. # 2000
Elsevier Science Ltd. All rights reserved.
Introduction
nitroimidazole group is enzymatically reduced and
trapped in hypoxic cells. Labeled nitroimidazole deri-
vatives are therefore potential radiopharmaceuticals for
imaging hypoxic areas. The ®rst compounds studied
intensively were derived from [1-(3-methoxypropyl-2-
hydroxy)]-2-nitroimidazole (misonidazole) by the intro-
duction of a radioactive halogen, F-18, Br-77 or I-123
A fall in blood pressure can be observed in many dis-
eases. This leads to a reduction in the amount of oxygen
delivered to cells, which become hypoxic while remain-
ing viable. If hypoxia is short-lived, the metabolism can
return to normal. However, if the oxygen supply is
de®cient for more than a few seconds, the damage
caused can be irreversible and lead to cellular death.1
Our aim was to develop a speci®c marker of hypoxic
tissues which could be visualized by external detection.
Imaging of the myocardium and the brain, both high
consumers of dioxygen, are potential applications for
such a marker. At the moment, perfusion imaging can
identify the relative delivery of dioxygen at the cellular
level but does not provide any information on cell
(I-131 for in vitro experiments).5 Images obtained
with ¯uorinated (PET) and brominated (SPECT) com-
pounds present poor contrast between normoxic and
hypoxic areas, while iodinated compounds are too lipo-
philic, making blood clearance excessively long. Given
the advantages of technetium-99m in nuclear medicine
(low cost, good availability, ideal half-life of 6.02 h and
ideal g-energy of 140 keV), research has focused
increasingly on Tc-99m-labeled markers of hypoxia.
Unfortunately, unlike the radioactive halogens, Tc-99m
cannot be directly bonded to the biomolecules, so that a
chelating group has to be linked to the bioactive frag-
ment to incorporate this metallic nuclide. Essentially,
two classes of compounds bearing a nitroimidazole
group have been studied: boronic adducts of technetium
�9
2
vitality. A hypoxia marker would also be very useful in
oncology, since tumors present hypoxic regions which
are more resistant to radiotherapy and therefore require
3
higher irradiation doses. It was shown as long ago as
1944 that the introduction of a nitro group on hetero-
cycles such as nitroimidazoles gives them bacteriostatic
4
10
11
properties. Over the last two decades, numerous papers
dealing with the biological activity of nitroimidazoles
under anaerobic conditions have suggested that the
dioxime and technetium diamine dioximes, both
neutral complexes. The lipophilicity of the ®rst com-
pounds resulted in their being trapped in the hydro-
phobic membrane, but the authors showed that the
nitroimidazole moiety of the complexes was enzymati-
cally reduced in the absence of oxygen, demonstrating
recognition by the enzyme. Dioxime complexes showed
*
4
Corresponding author. Tel.: +33-4-7663-5705; fax: +33-4-7651-
382; e-mail: francoise.riche@ujf-grenoble.fr
0960-894X/01/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00593-X

7
2
F. Rich e et al. / Bioorg. Med. Chem. Lett. 11 (2001) 71±74
relatively good localization in hypoxic cells. Recently,
HL-91, another dioxime with a [TcO]
introduced via mesylate. Deprotection of the amines in
11 and labeling with Tc-99m led to complex 12.
3+
core, showed
1
2
good uptake in tumors. Curiously, this compound
does not include a nitroimidazole group. Apart from
this latter compound, all the derivatives with an anity
for hypoxic tissue include a nitroimidazole group, which
is why we decided to bond this type of structure to
Tc-99m chelating groups.
The intermediate 8 served as a precursor for the pre-
paration of the third macrocyclic ligand 15, as shown in
Scheme 3. The two amide functions were reduced by
1
5
BH ±Me S in THF according to Brown's method with
3
2
a modi®cation: the amines were protected in situ to
facilitate puri®cation. The alcohol function was acti-
vated via mesylate to introduce 2-methyl-4-nitroimida-
zole. After deprotection of the amine groups, 14 could
be cyclized on methylacrylate to give the target molecule
15, then labeling of 15 gave 16.
The present report describes the synthesis of several
ligands, their labeling with technetium-99m and the
biological results. The ligands present a nitroimidazole
group required for hypoxic cell trapping and three che-
lating groups leading to three technetium complexes
with a positive charge, negative charge or no charge, in
order to evaluate the in¯uence of radiopharmaceutical
charge on uptake by hypoxic cells.
The three chelating groups are 1,4,8,11-tetraazacyclotet-
radecan (cyclam), 5,7-dioxo-1,4,8,11-tetraazaundecan
and 2-oxo-1,5,8,12-tetraazacyclotetradecan (oxocyclam)
leading, respectively, to positive, negative or neutral
complexes at physiological pH by complexation of the
+
13,14
[
TcO2] core.
in Figure 1.
The three target complexes are shown
Chemistry and Labeling
First, compound 6 was synthesized as described in
Scheme 1 from cyclam, in which the three amine func-
tions were protected by tert-butyloxycarbonyl groups
before coupling with mesylate of 6-(2-methyl-4-nitro-
imidazolyl)hexan-1-ol 2 to give 4. Next, the amine
groups were deprotected in HCl/EtOH to give 5. The
Scheme 1. Synthesis and complexation of N-[6-(2-methyl-4-nitroimi-
dazolyl)-hexyl]cyclam. Reagents: (a) 2-methyl-5-nitroimidazole,
+
introduction of the TcO2 core into 5, leading to 6, is
described later.
K
2
CO
c) Boc
EtOH (10% v/v), 71%; (f)
3
,
DMF, 78%; (b) CH
3
SO
2
Cl, CH
2
Cl
2
,
pyridine, 98%;
(
2
O, EtOH, 50%; (d) 2, K
2
CO
3
, DMF, 25%; (e) HCl 37%/
99m
�
TcO
4
, Sn-tartrate, H
2
O, pH>11.5.
Synthesis of 12 (Scheme 2) was performed from
5-chloropentan-1-ol, a bifunctional compound allowing
introduction of an imidazole group on one side and a
chelating group on the other side. After protection of
the hydroxyl group, a malonic synthesis led to the
introduction of a three-carbon bridge on the linker in
compound 7. The four amine/amide functions were
easily introduced through the action of an excess of
ethylene diamine (as a solvent) on the diester. After
deprotection of the alcohol and protection of the
amines, the 2-methyl-4-nitroimidazolyl group was
Scheme 2. Synthesis and complexation of 6-[5-(2-methyl-4-nitroimi-
dazolyl)pentyl]-5,7-dioxo-1,4,8,11-tetraazaundecan. Reagents: (a) DHP,
PTSA, THF, 83%; (b) EtONa, CH
N(CH NH , 74%; (d) PTSA, MeOH, 79%; (e) Boc
8%; (f) CH SO Cl, CH Cl , 87%; (g) 2-methyl-5-nitroimidazole,
2
(COOEt)
2
, EtOH, 47%; (c)
H
2
2
)
2
2
2
O, MeOH,
9
3
2
2
2
K
(i)
2
CO
99m
3
,
DMF, 92%; (h) HCl 37%/MeOH (10% v/v), quant;
�
Figure 1. Target compounds.
TcO
4
, Sn-tartrate, H
2
O, pH>11.5.

F. Rich e et al. / Bioorg. Med. Chem. Lett. 11 (2001) 71±74
73
The radiolabeling of the three target ligands was
achieved with sodium [9 Tc]pertechnetate and stan-
nous(II) tartrate as the reducing agent to produce 6, 12
and 16 in good yields (75 to 95%), evaluated by chro-
matography. The charges of the complexes were deter-
mined by electrophoresis on cellulose in bu€ered
medium at physiological pH and were in agreement
with those obtained with chelating groups without
functionalization. The macroscopic (Tc-99) structure of
these non-functionalized complexes has already been
studied in order to con®rm the proposed structure of
to determine whether the perfused hearts remained in
good condition during the experiments. The experi-
mental protocol is described in Figure 2.
9m
9
9m
+
99m
�
After a bolus injection of [ TcO -6] , [ TcO -12]
2
2
or [9 TcO -16] (1.5 MBq/250 mL physiological serum),
9m
2
radioactivity was monitored by a NaI detector for
30 min. Each compound was tested on four isolated
hearts. With this model, retention of a hypoxia tracer is
expected to be much higher in hypoxic conditions than
in normoxic conditions. Typical curves obtained in
normoxia and hypoxia are represented in Figure 3.
1
3,14
these chelates at the microscopic level (Tc-99m).
can thus arm by analogy that the structures of the
We
9
9m
+
functionalized complexes involve the [ TcO2] core
and that deprotonations at pH 7.4 occur for 12 and 16
to give negative and neutral complexes, respectively.
The partition coecients P were measured between a 1-
octanol phase and a bu€er phase (pH 7.4). Radiolabeling
results are given in Table 1.
Although no signi®cant statistical di€erence was found,
the complexes seemed to show slightly higher radio-
activity retention in normoxic conditions compared
with hypoxic conditions. This observation has already
been reported in the case of bisthiosemicarbazone com-
plexes by Dearling et al.,17 who suggest that factors
other than hypoxia govern the uptake and uptake
kinetics of certain complexes.
Biological Results and Discussion
The rapid clearance obtained is not surprising for
[
species which are insuciently lipophilic to cross the
membrane by passive di€usion.
99m
+
99m
�
TcO -6] and [ TcO -12] , since they are charged
2 2
Uptake of target complexes 6, 12 and 16 was performed
in Langendor€ bu€er-perfused rat hearts in normoxic
and hypoxic conditions according to the method of
1
6
Chin et al. Normoxia was obtained with bu€er satu-
rated by 95% O and 5% CO and hypoxia with bu€er
saturated by 95% N and 5% CO . Heart rate and sys-
The result is more surprising for the neutral complex
99m
2
2
[
TcO -16]. The rapid clearance may be correlated
2
2
2
tolic and diastolic pressures were also recorded on-line
with rapid uptake and washout due to the absence of
reduction of the nitro group, which is the initial step of
the cellular trapping process. Although the redox
3
potential of 2-methyl-5-nitroimidazole (� 415 mV ) is
greater than that of the oxido-reductases with the lowest
potentials such as succinate dehydrogenase (� 450 mV),
the di€erence may not be sucient to allow ecient
reduction of this compound. We now envisage the
synthesis of other nitroaromatic compounds such as 2-
nitro or polynitroimidazole derivatives, which present
lower redox potentials.
Figure 2. Diagrammatic representation of the experimental protocol.
Scheme 3. Synthesis and complexation of 10-[5-2-(methyl-4-nitroimi-
dazolyl)pentyl]-2-oxocyclam. Reagents: (a) BMS/THF; (b) MeOH,
HCl; (c) Boc
CH Cl , 95%; (e) 2-methyl-5-nitroimidazole, K
HCl 37%/MeOH (10% v/v), quant; (g) CH
ˆCHCOOMe, MeOH,
O, pH>11.5.
2
O, K
2
CO
3
, MeOH, 75% (a+b+c); (d) CH
3 2
SO Cl,
2
2
2
CO , DMF, 98%; (f)
3
2
9
9m
�
1
5%; (g)
TcO , Sn-tartrate, H
4
2
Table 1. Radiolabeling data for 99mTc-6, 99mTc-12 and 99mTc-16
Compound
Yield (%)
Electrophoresis
pH 7.4, 30 V/cm)
Log P
(
9
9
9
9m
9m
9m
Tc-6
Tc-12
Tc-16
95
75
90
� 3 cm
+ 3 cm
0
� 2.0
� 1.4
� 0.12
Figure 3. Characteristic curves of 99mTc complexes in isolated per-
fused rat heart.

7
4
F. Rich e et al. / Bioorg. Med. Chem. Lett. 11 (2001) 71±74
References
. Berne, R. M.; Levy, M. N. Skeletal physiology. In Physiol-
ogy, 3rd ed.; Berne, R. M., Levy, M. N., Eds; Mosby Year
Eckelman, W. C.; Nunn, A. D. Bioconjugate Chem. 1993, 4,
26.
1. Chan, Y. W.; Romero, V.; Linder, K. E.; Patel, B. C.;
Jayatilak, P. G.; Nunn, A. D.; Rumsey, W. J. Nucl. Med.
3
1
1
1
1
994, 35, 152.
2. Yutani, K.; Kusuoka, H.; Fukuchi, K.; Tatsumi, M.;
Book: St Louis, 1993; pp 292±308.
2
3
1
4
8
5
. Coleman, C. N. J. Natl. Cancer Inst. 1988, 80, 310.
. Nunn, A.; Linder, K.; Strauss, H. W. Eur. J. Nucl. Med.
995, 22, 265.
. Dodd, M. C.; Stillman, N. B. J. Pharm. Exp. Ther. 1944,
2, 11.
. Jerabek, P. A.; Dischino, D. O.; Kilborn, M. R.; Welch, M.
Nishimura, T. J. Nucl. Med. 1999, 40, 854.
3. Riche, F.; Trimcev, I.; Godard, J.; Benabed, A.; Vidal,
M.; Pasqualini, R. Appl. Radiat. Isot. 1993, 44, 907.
4. Riche, F.; Vidal, M.; Bolzati, C.; Uccelli, L.; Marchi, A.;
Duatti, A. Inorg. Chim. Acta 1995, 231, 147.
1
Â
1
Â
1
1
1
5. Brown, H. C.; Choi, Y. M.; Narasimhan, S. J. Org. Chem.
982, 47, 3153.
6. Chin, K. Ng.; Sinusas, A. J.; Zaret, B. L.; Soufer, R. Cir-
J. J. Nucl. Med. 1984, 25, 43.
. Grierson, J. R.; Link, J. M.; Mathis, C. A.; Rasey, J. S.;
Krohn, K. A. J. Nucl. Med. 1989, 30, 343.
6
culation 1995, 92, 1261.
7. Dearling, J. L. J.; Mullen, G. E. D.; Lewis, S. S.; Rae,
7
1
8
. Chapman, J. D.; Franko, A. J.; Sharplin, J. Br. J. Cancer
981, 43, 546.
. Biskupiak, J. E.; Grierson, J. R.; Rasey, J. S.; Martin, G.
1
M. T.; Zweit, J.; Welch, M. J.; Blower, P. J. Technetium,
rhenium and other metals in chemistry and nuclear medicine;
Fifth International Symposium on Technetium in Chemistry
and Nuclear Medicine, Bressanone, Italy, 6±9 September 1998,
p 599.
V.; Krohn, K. A. J. Med. Chem. 1991, 34, 2165.
9
1
1
. Wiebe, L. I.; Jette, D. C.; Chapman, J. D. Nuclear Medizin
984, 23, 63.
0. Linder, K. E.; Chan, Y. W.; Cyr, J. E.; Nowotnik, D. P.;