Synthesis of N-(piperidin-1-yl)-5-(4-methoxyphenyl)-1-(2-chlorophenyl)-4- [18F]fluoro-1H-pyrazole-3-carboxamide by nucleophilic [18F] fluorination: A PET radiotracer for studying CB1 cannabinoid receptors

Article abstract of DOI:10.1002/jlcr.647

The feasibility of nucleophilic displacement of bromide in the 4-bromopyrazole ring with [¹⁸F]fluoride has been demonstrated by the synthesis of two radiolabeled compounds: N-(piperidin-1-yl)-5-(4-methoxyphenyl)-1- (2-chlorophenyl)-4-[¹⁸F]fluoro-1H -pyrazole-3-carboxamide, ([¹⁸F] NIDA-42033) 1b and 1-(2-chlorophenyl)-4-[¹⁸F]fluoro-5-(4-methoxyphenyl)- 1H-pyrazole-3-carboxylic acid, ethyl ester 4. The radiochemical yields were in the range of 1-6%. [¹⁸]NIDA-42033, a potential radiotracer for the study of CB₁ cannabinoid receptors in the animal brain by positron emission tomography, has been synthesized in sufficient quantities with specific radioactivity greater than 2500 mCi/μmol and radiochemical purity >95%. Copyright

Full text of DOI:10.1002/jlcr.647

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2003; 46: 93–98.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.647
Research Article
Synthesis of N-(piperidin-1-yl)-5-
(4-methoxyphenyl)-1-(2-chlorophenyl)-
4-½18Fꢀfluoro-1H-pyrazole-3-carboxamide
by nucleophilic ½18Fꢀ fluorination: a PET
radiotracer for studying CB1 cannabinoid
receptors
Reeti Katoch-Rouse and Andrew G. Horti*
Neuroimaging Research Branch, National Institute on Drug Abuse,
Intramural Research Program, National Institutes of Health, 5500 Nathan
Shock Drive, Baltimore, MD 21224, USA
Summary
The feasibility of nucleophilic displacement of bromide in the 4-bromopyrazole
ring with ½18Fꢀfluoride has been demonstrated by the synthesis of two
radiolabeled compounds: N-(piperidin-1-yl)-5-(4-methoxyphenyl)-1-(2-chloro-
phenyl)-4-½18Fꢀfluoro-1H-pyrazole-3-carboxamide, (½18Fꢀ NIDA-42033) 1b and
1-(2-chlorophenyl)-4-½18Fꢀfluoro-5-(4-methoxyphenyl)-1H-pyrazole-3-carboxylic
acid, ethyl ester 4. The radiochemical yields were in the range of 1–6%.
½18FꢀNIDA-42033, a potential radiotracer for the study of CB1 cannabinoid
receptors in the animal brain by positron emission tomography, has been
synthesized in sufficient quantities with specific radioactivity greater than
2500 mCi=mmol and radiochemical purity > 95%. Copyright # 2002 John
Wiley & Sons, Ltd.
Key Words: positron emission tomography; cannabinoid receptors;
radiosynthesis; 18F; nucleophilic radiofluorination
*Correspondence to: A.G. Horti, NIDA-IRP/NIH, Neuroimaging Research Branch, 5500 Nathan
Shock Drive, Baltimore, MD 21224, USA. E-mail: ahorti@intra.nida.nih.gov
Copyright # 2002 John Wiley & Sons, Ltd.
Received 19 June 2002
Revised 30 June 2002
Accepted 30 July 2002

94
R. KATOCH-ROUSE AND A. G. HORTI
Figure 1.
Introduction
Non-invasive in vivo imaging of the central CB1 receptors by positron
emission tomography (PET) would enable the investigation of the
pharmacological and physiological role played by the cannabinoids in
normal and disease states. Recently, several PET and SPECT radio-
tracers based on the selective and potent CB1 antagonist SR141716 have
been reported.124 Unfortunately, these radiotracers display very high
non-specific binding due to their high lipophilicity.
We have recently synthesized NIDA-42033, N-(piperidin-1-yl)-5-
(4-methoxyphenyl)-1-(2-chlorophenyl)-4-fluoro-1H-pyrazole-3-carbox-
amide, an SR1417165 analogue with lower lipophilicity ðclog D ¼ 3:43Þ
and high affinity at CB1 receptors ðKi ¼ 18 nMÞ in in vitro studies
(Figure 1).6 Herein, we report the radiosynthesis of 18F labeled
NIDA-42033 by no-carrier-added nucleophilic ½18Fꢀfluorination of
NIDA-42055, N-(piperidin-1-yl)-5-(4-methoxyphenyl)-1-(2-chlorophenyl)-
4-bromo-1H-pyrazole-3-carboxamide. The presented radiosynthesis uti-
lizes a novel nucleophilic radiofluorination reaction with 4-bromopyrazole
precursors7 yielding 4-½18Fꢀfluoropyrazole derivatives.
Results and discussion
In the past, synthesis of non-labeled fluorinated pyrazoles has been
performed mainly via three routes: (i) condensation of fluori-
nated diketo substrates with substituted hydrazines,8 (ii) direct ring
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 93–98

[
18F] CANNABINOID RECEPTOR LIGAND
95
Scheme 1.
fluorination using gaseous fluorine,9 and (iii) diazotization and
photochemical irradiation of the diazonium salts in tetrafluoroboric
acid.10 Any examples of nucleophilic fluorination of 4-substituted
pyrazole derivatives were not found even after an exhaustive literature
search. Because the pyrazole moiety can be found in many CNS drugs, a
practical method of nucleophilic radiofluorination of pyrazole ring in
position C-4 can be potentially beneficial for the development of novel
PET radiotracers including novel cannabinoid ligands. As a result, we
proposed to synthesize a 4-½18Fꢀfluoropyrazole derivative by nucleo-
philic radiofluorination of the corresponding 4-bromopyrazole. The
previously synthesized cannabinoid ligands NIDA-42055 and NIDA-
42033 by our group6 appeared to be interesting objects for such a study.
No-carrier-added radiofluorination of NIDA-42055 in DMSO
solution in the presence of Kryptofix1 222 at 180–1908C for 15 min
gave the corresponding ½18FꢀNIDA-42033 (Scheme 1). The average
radiochemical yield (not decay corrected) was 1% based on five
syntheses and the average specific radioactivity was greater than 2500
mCi=mmol calculated at the end of the synthesis. There were two
radioactive hydrophilic by-products in the reaction mixture as
determined by the preparative HPLC. The first peak was ½18Fꢀ fluoride
and the second peak was unknown. The purification of the product was
performed with reverse phase semi-preparative HPLC. The total time of
the synthesis including purification was 45–60 min. After the evapora-
tion of HPLC solvent, the final product was formulated as a saline
solution ð5 mlÞ with the addition of 10 ml Tween 80 (Aldrich). Tween 80
was added to improve the solubility of the radiotracer in saline.
Microwave heating at 2508C for 2 min gave a similar radiochemical
yield. The final product was shown to be radiochemically pure (>95%
by radioHPLC) and coeluted with the standard.6; 7 The relatively low
yield of ½18FꢀNIDA-42033 may be due to the instability of the hydrazide
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 93–98

96
R. KATOCH-ROUSE AND A. G. HORTI
Scheme 2.
moiety under the drastic radiofluorination conditions. Under similar
conditions, 1-(2-chlorophenyl)-4-½18Fꢀfluoro-5-(4-methoxyphenyl)-1H-
pyrazole-3-carboxylic acid ethyl ester, 4 was obtained by radiofluorina-
tion of the corresponding bromo analogue 3 with a radiochemical yield
of 6% (Scheme 2) (3 runs, experiment not presented). The higher yield
in case of 4 as opposed to 1b suggests that the nucleophilic radio-
fluorination reaction may prove more or less efficient based on the
stability of other substituents attached to the pyrazole backbone and/or
activation of 4-position for the nucleophilic substitution.
Conclusion
In summary, a simple one-step radiosynthesis of a high specific activity
radioligand for studying the cerebral cannabinoid receptors by PET
has been developed. The synthesis yields sufficient quantities of the
radioligand for future studies of CB1 receptors by PET. A novel
nucleophilic radiofluorination at C-4 position of the pyrazole ring has
been reported. This nucleophilic radiofluorination procedure may prove
useful for the preparation of various 4-½18Fꢀfluoropyrazole based
radiotracers. Potentially, the yield of 4-½18Fꢀfluoropyrazole derivatives
may be improved by employing precursors with a better leaving group,
such as, 4-trimethylammoniumpyrazole derivatives.
Experimental
Materials and Methods
All reagents were ACS or HPLC grade and were purchased from
Aldrich. HPLC analyses and purification were performed with two
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 93–98

[
18F] CANNABINOID RECEPTOR LIGAND
97
Waters 600/610 HPLC pumps, an in-line UV-detector (Waters,
254 nm), and a flow-count radioactivity detector (Bioscan 3200). HPLC
chromatograms were recorded on a Rainin Dynamax dual channel
control/ interface module connected to a Macintosh computer with
Dynamax v.1.4.2 software. A dose calibrator (Capintec CRC-35R) was
used for all radioactivity measurements. ½18FꢀFluoride was prepared
using an RDS111 cyclotron (CTI, Knoxville, TN). The radiofluorina-
tion was performed either with an automated radiochemistry module
(CPCU, CTI, Knoxville, TN) or, manually, with a microwave oven
(CEM, Matthews, NC). All precursors and standards were prepared in
our lab.6; 7
Radiochemistry
N-(piperidin-1-yl)-5-(4-methoxyphenyl)-1-(2-chlorophenyl)-4-½18Fꢀ fluoro-
1H-pyrazole-3-carboxamide (1b). N-(piperidin-1-yl)-5-(4-methoxyphenyl)-
1-(2-chlorophenyl)-4-bromo-1H-pyrazole-3-carboxamide (NIDA-42055)
(3 mg, 6 mmol) was dissolved in anhydrous DMSO ð0:5 mlÞ and trans-
ferred to a reaction vessel containing K½18FꢀF=Kryptofix1222=K2CO3
complex prepared by the Hamacher method11 with 24 mg Kryptofix1222
and 4:5 mg K2CO3. The reaction mixture was heated at 1828C for 15 min,
cooled down, diluted with 0:5 ml water and injected onto the Waters
PrepPak NOVA C18 cartridge ð25 ꢁ 100 mm2Þ. The reaction mixture was
eluted with mixture of acetonitrile : 0:1 M ammonium formate buffer
(53:47) at a flow rate of 8 ml=min. The radioactive peak with a retention
time of 14:2 min ðk0 ¼ 3:05Þ corresponding to the NIDA-42033 standard
was collected. The retention time and the capacity factor of the precursor,
NIDA-42055, under the separation conditions were 17:25 min and 3.93,
respectively. There were three more substantial unidentified non-radio-
active peaks in the reaction mixture with retention times of 4.7, 5.0 and
8:8 min.
An aliquot of the final solution of known volume and radioactivity
was applied to an analytical HPLC column (Waters Symmetry C18
column, 4:6 ꢁ 250 mmÞ. A mobile phase of acetonitrile: 0:1 M
ammonium formate buffer (50:50) at a flow rate of 3 ml=min was used
to elute the radioligand, which had a retention time of 4:8 min. The area
of the UV absorbance peak measured at 254 nm corresponding to
carrier product was measured and compared to the cold standard for
specific radioactivity calculation. The radiochemical product coeluted
with a sample of authentic NIDA-42033.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 93–98

98
R. KATOCH-ROUSE AND A. G. HORTI
Acknowledgements
The authors would like to thank Dr Alane Kimes for administrative
support and Ms Alvina Walters for secretarial assistance.
References
1. Mathews WB, Ravert HT, Musachio JL, et al. J Label Compd Radiopharm
1999; 42: 589–596.
2. Mathews WB, Scheffel U, Finley P, et al. Nucl Med Biol 2000; 27: 757–762.
3. Lan R, Gatley SJ, Makriyannis A. J Label Compd Radiopharm 1996; 38:
875–881.
4. Gatley SJ, Lan R, Volkow ND, et al. J Neurochem 1998; 70: 417–423.
!
5. Rinaldi-Carmona M, Barth F, Heaulme M, et al. FEBS Lett 1994; 350:
240–244.
6. Katoch-Rouse R, Pavlova OA, Kimes AS, Mukhin AG, Horti AG.
Abstracts of Papers, 223rd ACS National Meeting, Orlando, FL, United
States, 2002 April 7–11.
7. The 4-bromoesters and 4-fluoroesters were prepared from the correspond-
ing 4-unsubstituted esters via a bromination reaction with Br2=CH2Cl2 at
08C and via an electrophilic fluorination reaction in CH3CN at 708C,
correspondingly.
8. Penning TD, Talley JJ, Bertenshaw SR, et al. J Med Chem 1997; 40:
1347–1365.
9. Storer R, Ashton CJ, Baxter AD, et al. Nucleosides Nucleotides 1999; 18:
203–216.
10. Fabra F, Vilarrasa J, Coll J. J Heterocyclic Chem 1978; 15: 1447–1449.
11. Hamacher K, Coenen HH, Stocklin G. J Nucl Med 1986; 27: 235–238.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2003; 46: 93–98