Radiosynthesis of a 2-substituted 4,5-dihydro-1H-[2-11C] imidazole: The I2 imidazoline receptor ligand [11C] benazoline

Article abstract of DOI:10.1002/jlcr.746

Benazoline (2-naphthalen-2-yl-4,5-dihydro-1H-imidazole) is a selective high-affinity ligand for the imidazoline 12 receptor. This compound was labelled with carbon-11 (T_1/2 = 20.4 min) at the number two carbon atom of its 2-imidazoline ring. Cyclotron-produced [¹¹C]carbon dioxide reacted with 2-naphthylmagnesium bromide to give 2-[carboxyl- ¹¹C] naphthoic acid in 60% radiochemical yield. The latter was heated with a mixture of ethylenediamine and its dihydrochloride at 300°C to give [¹¹C]benazoline in 16% overall yield, relative to [ ¹¹C]carbon dioxide and with a specific radioactivity of 54GBq/ μmol, decay corrected for end of irradiation. The procedure requires about 45 min from end of cyclotron irradiation. This method should be extendable to other imidazolines. Copyright

Full text of DOI:10.1002/jlcr.746

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2003; 46: 1141–1149.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.746
Research Article
Radiosynthesis of a 2-substituted 4,5-dihydro-
1H-[2-11C] imidazole: the I2 imidazoline receptor
ligand [11C] benazoline
!
D. Roeda*, F. Hinnen and F. Dolle
! ! ! !
Departement de Recherche Medicale, Service Hospitalier Frederic Joliot,
! !
CEA, 4 Place du General Leclerc, Orsay F-91401, France
Summary
Benazoline (2-naphthalen-2-yl-4,5-dihydro-1H-imidazole) is a selective high-
affinity ligand for the imidazoline I2 receptor. This compound was labelled
with carbon-11 (T1=2=20.4 min) at the number two carbon atom of its
2-imidazoline ring. Cyclotron-produced [11C]carbon dioxide reacted with
2-naphthylmagnesium bromide to give 2-[carboxyl-11C]naphthoic acid in 60%
radiochemical yield. The latter was heated with a mixture of ethylenediamine
and its dihydrochloride at 3008C to give [11C]benazoline in 16% overall yield,
relative to [11C]carbon dioxide and with a specific radioactivity of 54 GBq/
mmol, decay corrected for end of irradiation. The procedure requires about
45 min from end of cyclotron irradiation. This method should be extendable to
other imidazolines. Copyright # 2003 John Wiley & Sons, Ltd.
Key Words: [2-11C]imidazolines; [11C]benazoline; imidazoline receptor
Introduction
2-Imidazoline (4,5-dihydro-1H-imidazole), bearing an alkyl or an aryl
group at the 2-position, frequently constitutes a compound of interest in
medicinal chemistry. About 25% of nearly 4000 published compounds
of this structure, are reported on for their pharmacological activity.1
!
!
! !
*Correspondence to: D. Roeda, Departement de Recherche Medicale, Service Hospitalier Frederic
! !
Joliot, CEA, 4 Place du General Leclerc, F-91401 Orsay, France. E-mail: roeda@dsvidf.cea.fr
Copyright # 2003 John Wiley & Sons, Ltd.
Received 9 June 2003
Accepted 2 July 2003

1142
D. ROEDA ET AL.
One of these compounds is 2-naphthalen-2-yl-4,5-dihydro-1H-imida-
zole (benazoline, 1). It has a high affinity for the imidazoline I2 receptor
ðpKi ¼ 9:07Þ with a high selectivity for I2 relative to the a1- and the a2
receptors (2691- and 18621-fold, respectively).2
Our interest in the study of receptor systems in the living human brain
by positron emission tomography (PET) prompted us to envisage
labelling of compound 1 with the short-lived positron-emitting radio-
isotope carbon-11 (half-life = 20.4 min). To date, no suitable PET
ligand has been available for the imidazoline receptor system.
Carbon atom number two of the 2-imidazoline ring, part of the
molecule’s amidine substructure, is the only candidate for labelling with
carbon-11. The other two are practically inaccessible with current
carbon-11 labelling technology, let alone those of the naphthalene
system. To our knowledge, introduction of a carbon-11 atom in this
position has not been published before. Thus, the exploration of the
synthesis of [11C]benazoline may not only provide us with an interesting
new research tool in PET, but also give us a novel labelling pathway
that possibly can be extended to other 2-substituted 2-imidazolines.
Results and discussion
Benazoline (1) was synthesized according to a literature procedure2 in
52% yield from 2-cyanonaphthalene (2) via naphthalene-2-carboximidic
acid methyl ester (3) (Scheme 1). This is an example of the usual method
Scheme 1.
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149

RADIOSYNTHESIS OF A 2-SUBSTITUTED
1143
of synthesis of 2-substituted 2-imidazolines, namely the cyclization
reaction of ethylenediamine with a carboxylic acid or rather one of its
derivatives.3 The reaction sequence in Scheme 1 is rather unsuitable for
our labelling purpose (too many steps) but the approach in general is
convenient to the extent that [11C]carboxylic acids are normally easy to
make in good yields from [11C]carbon dioxide and the appropriate
Grignard reagent. Its disadvantage lies in the fact that the radioactivity
introduction is to be followed by at least one chemical step, but this
drawback looks inevitable for this type of molecule in any case. So we
have developed a radiosynthesis for [11C]benazoline ([11C]-1) according
to the more straightforward route outlined in Scheme 2.
Scheme 2.
2-Naphthylmagnesium bromide (4) was freshly prepared from
2-bromonaphthalene before each [11C]benazoline run, following a
literature procedure that employs a substantial amount of l,2-
dibromoethane as an initiator.4 The final reaction mixture of Grignard
in tetrahydrofuran (THF) was used as such in the radiosynthesis. Before
this, samples of the Grignard preparation were analysed by hydrolysis
to naphthalene and by carbonation to 2-naphthoic acid (5), giving
average yields of 76 and 69%, respectively (HPLC), relative to the
theoretical 0.62 M Grignard concentration.
2-[11C]naphthoic acid ([11C]-5) could be made in about 60%
radiochemical yield from cyclotron-produced no-carrier-added [11C]car-
bon dioxide by passing the latter with an inert vector gas into 400 ml of
the Grignard solution at room temperature. The radioactivity trapping
efficiency was 95%. When the reaction mixture was acidified, 37% of
the radioactivity, most likely non-reacted [11C]carbon dioxide, could be
chased by bubbling through an inert gas. Radio-HPLC analysis showed
that the remainder was only 2-[11C]naphthoic acid. This is in contrast
with an early report on the similar synthesis of 2-[14C]naphthoic acid
(yield 73%), reporting the side product [14C]di-2-naphthylketone.5
2-Naphthylmagnesium bromide (4) is only moderately reactive, seen the
important amount of non-reacted [11C]carbon dioxide. Indeed, at
ꢀ208C we found no reaction at all and increasing the reaction
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149

1144
D. ROEDA ET AL.
temperature to 608C did not improve the yield. When less Grignard
solution was used (100 ml) the [11C]carbon dioxide trapping efficiency
dropped considerably, also at ꢀ208C.
Aryllithium compounds are generally more reactive than arylmagne-
sium compounds and usually give ketones with carbon dioxide.6
Nevertheless, we tried to obtain 2-[11C]naphthoic acid also through
carbonation of 2-naphthyllithium, made from 2-bromonaphthalene and
n-butyllithium, in order to avoid the rather laborious Grignard
preparation before each radiosynthesis. However, this preparation did
not react at all with [11C]carbon dioxide. Possibly the concentration we
used was too low or the 2-naphthyllithium, which was reported to be
particularly sensitive to the Wurtz-coupling, was lost, as was reported
for the reaction of 2-bromonaphthalene and lithium metal that gave
exclusively di-2-naphthyl.7 Although formation of 2-naphthyllithium
from n-butyllithium has been reported in the literature (e.g. ref. 8), we
did not further explore this option, the more so because there would be
a risk of attack of non-reacted n-butyllithium on the naphthalene system
of the formed 2-[11C]naphthoic acid.9
For the second step, the conversion of 2-[11C]naphthoic acid ([11C]-5)
into [11C]benazoline ([11C]-1), we based ourselves on a German patent
from 1938.10 Our method consisted in hydrolysis of the reaction mixture
of [11C]-5, not with HCl as above, but with an aqueous solution of
ethylenediamine and its dihydrochloride. The mixture was then heated
at 3008C during 10 min, with the reaction vessel open to atmosphere,
giving [11C]-1 in 27% radiochemical yield with respect to [11C]-5 and in
an overall radiochemical yield of 16% with respect to [11C]carbon
dioxide. There is no reaction at lower temperatures. The reaction
mixture was taken up in N,N-dimethylformamide (DMF) for HPLC
injection. Water, which perturbs less the HPLC system, can also be
used, but leaves more undissolved solids which can give problems with
the injection. [11C]Benazoline ([11C]-1) was isolated using reversed-phase
semi-preparative HPLC. Apart from very little 2-[11C]naphthoic acid
([11C]-5), two radioactive side products were seen on the chromatogram
in-between [11C]-5 and [11C]-1. The first apparently comes from reaction
of [11C]carbon dioxide with ethylenediamine and is most likely
[11C]ethyleneurea.11 The second one may very well be N-(2-ami-
noethyl)-2-[11C]naphthamide.10 [11C]Benazoline ([11C]-1) was formu-
lated by a C18-Seppak1-cartridge procedure. It was eluted from the
cartridge with a small amount of ethanol followed by physiological
serum or citric acid buffer (pH=4.7), the latter giving a slightly better
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149

RADIOSYNTHESIS OF A 2-SUBSTITUTED
1145
result. The specific radioactivity was about 54 GBq/mmol (decay
corrected for end of irradiation) for a 30 min/30 mA irradiation. The
whole procedure can be carried out in 45 min.
A reagent that can serve as an alternative to ethylenediamine is
ethyleneurea (2-imidazolidone),12 which we confirmed by a non-
radioactive experiment. The yield is higher, but the solidified melt,
after cooling, is difficult to dissolve for HPLC injection and we
discarded this potentially interesting approach for this technical reason.
Two other methods employing ethylenediamine and carboxylic acids
did not work for 2-naphthoic acid in our hands, the first using toluene
as a solvent13 and the second using triphenylphosphine, tetrachlor-
omethane and triethylamine at room temperature.14
The condensation of a carboxylic acid with ethylenediamine, which
we have adopted for the [11C]benazoline synthesis, has a wide scope.10
We assume that the presented procedure can be extended to a general
method to obtain labelled 2-substituted 2-imidazolines.
Experimental
All chemicals were purchased from Aldrich, France except for
2-cyanonaphthalene (Lancaster, UK) and magnesium turnings (Farm-
italia Carlo Erba, Italy) and were used without further purification. An
isotonic citrate buffer (pH 4.7) was prepared from citric acid (0.12 g) and
sodium citrate (0.24 g) in sterile water (10 ml), and then submitted to
sterile filtration. Air- or moisture sensitive reactions were conducted in
oven-dried glassware and under an inert atmosphere. 1H-NMR and 13C-
NMR were recorded on a Bruker AMX (300 MHz) apparatus at room
temperature using TMS as an internal standard. The chemical shifts are
reported in ppm, downfield from TMS. Abbreviations used: s, m, br for
singlet, multiplet and broad, respectively. HPLC was performed on a
reversed-phase column (Xterra MS C18, 7 mm, 7.8 ꢁ 300 mm) using a
mixture of water, acetonitrile and triethylamine (70/30/0.05 v/v/v=sys-
tem A or 60/40/0.1 v/v/v=system B) as the mobile phase (5 ml/min) and
.
employing a UV (254 nm)- and a Geiger–Muller detector.
2-Naphthalen-2-yl-4,5-dihydro-1H-imidazole (benazoline, 1)
A solution of 2-cyanonaphthalene (1 g, 6.54 mmol) and methanol
(0.53 ml, 13.08 mmol) in chloroform (10 ml) was cooled in an ice/water
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149

1146
D. ROEDA ET AL.
bath and gaseous hydrochloric acid was passed through over 45 min.
The yellow reaction mixture was kept for 12 h in a refrigerator at 0–48C.
Diethyl ether (50 ml) was added and the formed precipitate was filtered
off (pale-yellow solid; 2.2 g). This solid, which is highly impure
naphthalene-2-carboximidic acid methyl ester hydrochloride, was added
to a solution of ethylenediamine (6.7 ml, 100 mmol) in absolute ethanol
(300 ml) cooled in an ice/water bath. After the mixture had been stirred
over 1 h at 08C, a mixture of concentrated hydrochloric acid (7 ml) and
ethanol (20 ml) was added carefully, causing a white precipitate. The
mixture was stored in the refrigerator for 24 h and then, after dilution
with ethanol (100 ml), refluxed over 5 h. The precipitate was filtered off
and discarded. The filtrate was evaporated till dryness and the residue
was taken up in chloroform. This organic solution was washed with 3 N
sodium hydroxide and dried on sodium sulphate. Evaporation of the
1
solvent gave a yellow solid (1,667 mg, 52%). H-NMR (d6-DMSO): d:
4.14 (4H, s), 6.80–7.20 (1H, br), 7.56–7.58 (2H, m), 7.93–8.02 (4H, m),
8.35 (1H, s). 13C-NMR (d6-DMSO): d: 49.4 (CH2), 124.3 (CH), 125.9
(CH), 126.2 (CH), 126.4 (CH), 127.0 (CH), 127.1 (CH), 127.9 (CH),
132.0 (Cq), 133.4 (Cq), 163.2 (Cq).
2-Naphthylmagnesium bromide (4)
Magnesium turnings (214 mg, 8.79 mmol) were placed in a reaction
vessel, equipped with magnetic stirring and an addition funnel and
provided with a nitrogen atmosphere, and were covered with dry THF
(3 ml). The addition funnel was charged with 2-bromonaphthalene
(1.668 g, 8.04 mmol) in dry THF (12 ml). 1,2-Dibromoethane (98 ml,
1.14 mmol) was added at once into the reaction vessel, which was then
immersed in a heating bath at 808C. As soon as a refluxing condition
was reached the contents of the addition funnel were added dropwise
over 15 min. Refluxing was continued for 1 h after which all magnesium
had been consumed. The red–brown reaction mixture was used as such
in the radiochemistry procedure (see below). Its quality was checked in
two ways: (1) An aliquot (50 ml) was hydrolyzed with HPLC mobile
phase and analysed for naphthalene (system B, retention time: 6.8 min).
(2) Ten millilitres of carbon dioxide gas were passed through an aliquot
of the Grignard solution (400 ml) in 1 min, followed by hydrolysis with
HPLC mobile phase. The mixture was analysed for 2-naphthoic acid on
HPLC (system A, retention time: 3.8 min).
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149

RADIOSYNTHESIS OF A 2-SUBSTITUTED
1147
2-Naphthyllithium
2-Bromonaphthalene (1 mg, 5 mmol) was dissolved in dry THF (200 ml)
in a septum-capped vial. The solution was cooled in an ice/water bath
and n-butyllithium (7 ml, 2.5 M in hexanes, 15 mmol) was added by
syringe through the septum. The mixture was left at 08C for 5 min and
was then used in its totality for trapping [11C]carbon dioxide at ꢀ108C.
[11C]Carbon dioxide
No-carrier-added [11C]carbon dioxide was produced using the nuclear
reaction 14N(p, a)11C by irradiation of a pressurized nitrogen gas target
(6 bar) with a cyclotron-generated 20 MeV proton beam.15 A current of
30 mA during 30 min gave rise to ꢂ48 GBq (1.3 Ci) of radioactive
carbon dioxide. After irradiation, the radioactive gas was released from
the target holder through a drying column (70 mm ꢁ 4 mm) filled with
phosphorus pentoxide. The [11C]carbon dioxide was frozen out in a
liquid-argon cooled stainless steel coil that was positioned in a shielded
hot cell in which all further radioactivity manipulations were performed
remotely controlled.
[Carboxyl-11C]2-naphthoic acid ([11C]-5)
A solution of freshly prepared 2-naphthylmagnesium bromide (4) in
THF (400 ml, see above) was placed in a small conical reaction vial
equipped with a septum to accommodate the needles connected to the
appropriate in- and outlet Teflon1 tubing. [11C]carbon dioxide was
released from its cooled trap by warming to room temperature and
carried by a nitrogen gas stream (3 ml/min) into the Grignard solution
at room temperature. A trap, containing soda lime, had been placed in
series to collect any [11C]carbon dioxide not trapped in the reaction vial.
When the radioactivity accumulation in the vial had levelled out, the gas
flow was discontinued and the reaction mixture was left to stand for
2 min. At this point the solution was either further used in the synthesis
of [11C]benazoline (see below) or hydrolyzed for HPLC analysis. In
the latter case aqueous HC1 (10 N, 50 ml) in acetonitrile (450 ml) was
added. A nitrogen stream was bubbled through the resulting acidic
mixture until all volatile radioactivity had been chased. A sample of the
mixture was then analysed with HPLC (system A, retention time
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149

1148
D. ROEDA ET AL.
2-[11C]naphthoic acid=3.8 min; note that [11C]carbon dioxide elutes
almost indistinguishably close to 2-[11C]naphthoic acid).
2-Naphthalen-2-yl-4,5-dihydro-1H-[2-11C]-imidazole ([11C]benazoline,
[11C]-1)
A solution of ethylenediamine dihydrochloride (200 mmol, 26.6 mg) and
ethylenediamine (200 mmol, 13.4 ml) in water (100 ml) was added to the
2-[11C]naphthoic acid ([11C]-5) reaction mixture (see above) followed
by another 100 ml of water to rinse all ethylenediamine from the addition
line into the reaction vial. After nitrogen had been set bubbling through
the mixture (20 ml/min), the reaction vial was placed in a heating block
at 3208C (effective maximum temperature inside the vial: 3008C). All
liquid evaporated quickly and the vapours were guided and condensed
via a wide-bore (2 mm id) outlet Teflon1 tube into a receiver. It was
important that this outlet line had a less than 908 bend close to the
reaction vial in order to prevent refluxing of liquid back into the vial.
After 2 min all liquids had evaporated. The nitrogen flow was stopped
and heating was continued for another 8 min. The vial was carefully
cooled down after the nitrogen flow had been restored (to avoid
blocking of the inlet needle). DMF (1 ml) was added and nitrogen
bubbling was continued for 1 min to promote solubilization of the
solids. The DMF solution was then transferred into an HPLC injector
loop and injected into the HPLC (system A). The [11C]benazoline peak
(retention time ꢂ11 min) was collected in a reservoir containing water
(40 ml). The [11C]benazoline was absorbed on a reversed-phase Seppak1
cartridge by passing the above aqueous solution through it, followed by
water (10 ml) to wash out any organic solvents. The [11C]benazoline was
eluted from the cartridge with ethanol (0.5 ml) followed by physiological
saline (4.5 ml) or an isotonic citrate buffer (pH=4.7, 4.5 ml). HPLC
analysis (system A; retention time of benazoline=10.8 min) of this
solution showed a single radioactive product co-eluting with co-injected
authentic synthesized benazoline. The UV trace did not show any
products but benazoline. The specific radioactivity of [11C]benazoline
was determined with the same HPLC system by comparing the
absorbance of [11C]benazoline to that of a known benazoline standard
and correlating the calculated mass with the measured radioactivity.
The radiochemical yield was deduced from the total recovered radio-
activity (waste, soda lime, reaction vial, etc.) that corresponded well
with the expected amount of [11C]carbon dioxide produced.
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149

RADIOSYNTHESIS OF A 2-SUBSTITUTED
1149
Conclusion
We have developed a radiosynthesis for the imidazoline I2 receptor
ligand [11C]benazoline for further exploration with PET. The synthesis
consists in the condensation of a carbon-11-labelled carboxylic acid
with ethylenediamine, introducing the label at the 2-carbon atom
position of a 2-substituted 2-imidazoline ring. We expect that this
method can be extended to many other imidazolines of this type.
Acknowledgements
We wish to thank the cyclotron operators Daniel Gouel, Christophe
#
Lechene and Christophe Peronne for carrying out the irradiations.
References
1. Beilstein database BS0302AB, Beılsteın GmbH and MDL Information
´ ´
Systems GmbH, Germany.
2. Pigini M, Bousquet P, Carotti A, Dontenwil M, Giannella M, Moriconi R,
Piergentili A, Quaglia W, Tayebati SK, Brasili L. Bioorg Med Chem 1997;
5: 833–841.
3. Ferm RJ, Riebsomer JL. Chem Rev 1954; 54: 593–613.
4. Tsuji R, Komatsu K, Inoue Y, Takeuchi K. J Org Chem 1992; 57: 636–641.
5. Heidelberger C, Brewer P, Dauben WG. J Amer Chem Soc 1947; 69:
1389–1391.
6. Gilman H, Van Ess R. J Amer Chem Soc 1933; 55: 1258–1261.
7. Gilman H, Zoellner EA, Selby WM. J Amer Chem Soc 1933; 55:
1252–1257.
8. Beringer FM, Nathan RA. J Org Chem 1969; 34: 685–689.
9. Plunian B, Mortier J, Vaultier M, Toupet, L. J Org Chem 1996; 61:
5206–5207.
10. Hartmann M, Isler HO (CIBA). German Patent 1938; DE687196.
11. Mulvaney JF, Evans RL. Ind Eng Chem 1948; 40: 393–397.
.
12. Kranzlein G, Bastian H, Schnurr W (I.G. Farbenindustrie). German Patent
1937; DE694119.
13. Shi Z, Gu H. Synth Comm 1997; 27: 2701–2707.
.
14. Vorbruggen H, Krolikiewicz K. Tetr Lett 1981; 22: 4471–4474.
(
15. Crouzel C, Langstrom B, Pike VW, Coenen HH. Appl Radiat Isot 1987; 38:
.
601–603.
Copyright # 2003 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 1141–1149