Synthesis of 18F labelled FK960, a candidate anti-dementia drug, and PET studies in conscious monkeys

Article abstract of DOI:10.1002/jlcr.633

The present study demonstrated the synthesis and in vivo study of ¹⁸F-labeled N-(4-acetyl-1-piperazinyl)-p-fluorobenzamide (FK960) which is a novel anti-dementia drug candidate. [¹⁸F]FK960 was prepared by a one-pot, three reaction sequence, using nucleophilic fluorination, with an automated synthetic apparatus using either ethyl-4-trimethylammonium triflate (1a) or ethyl-4-nitrobenzoate (lb) as the precursor for labeling. Though la gave a higher yield, the specific activity was 50-100 fold higher with 1b. The radiochemical yield of [¹⁸F]FK960 was 7-15% (EOB) and the specific activity ranged from 2.0-60.2GBq/μmol depending on the amount of F-18 used. The synthesis time was 2.2-2.9 h. The obtained [¹⁸F]FK960 was injected into 3 conscious monkeys (100-120MBq/kg body weight), and distribution images and pharmacokinetic data for [¹⁸F]FK960 showed similar uptake in different brain regions and 3-fold higher levels of [¹⁸F]FK960 in blood relative to brain. Copyright

Full text of DOI:10.1002/jlcr.633

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2002; 45: 1219–1228.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.633
Research Article
18
Synthesis of F labelled FK960, a candidate
anti-dementia drug, and PET studies in
conscious monkeys
Yoshihiro Murakami1*, Shintaro Nishimura1, Akihiro Noda1,
Norihiro Harada2 and Hideo Tsukada2
1 Advanced Technology Platform Research Laboratories, Fujisawa
Pharmaceutical Co.,Ltd., 5-2-3, Tokodai, Tsukuba, Ibaraki 300-2698, Japan
2 Central Research Laboratory, Hamamatsu Photonics K.K. 5000, Hirakuchi,
Hamakita, Shizuoka 434-8601, Japan
Summary
The present study demonstrated the synthesis and in vivo study of 18F-labeled
N-(4-acetyl-1-piperazinyl)-p-fluorobenzamide (FK960) which is a novel anti-
dementia drug candidate. [18F]FK960 was prepared by a one-pot, three
reaction sequence, using nucleophilic fluorination, with an automated synthetic
apparatus using either ethyl-4-trimethylammonium triflate (1a) or ethyl-4-
nitrobenzoate (1b) as the precursor for labeling. Though 1a gave a higher yield,
the specific activity was 50–100 fold higher with 1b. The radiochemical yield of
[18F]FK960 was 7–15% (EOB) and the specific activity ranged from
2.0–60.2 GBq/mmol depending on the amount of F-18 used. The synthesis
time was 2.2–2.9 h. The obtained [18F]FK960 was injected into 3 conscious
monkeys (100–120 MBq/kg body weight), and distribution images and
pharmacokinetic data for [18F]FK960 showed similar uptake in different brain
regions and 3-fold higher levels of [18F]FK960 in blood relative to
brain. Copyright # 2002 John Wiley & Sons, Ltd.
*Correspondence to: Y. Murakami, The Medical and Pharmacological Research Center Foundation,
Wo32, Inoyama-Cho, Hakui-City, Ishikawa 925-0613, Japan. E-mail: murakami@mprcf.or.jp
Copyright # 2002 John Wiley & Sons, Ltd.
Received 18 February 2002
Revised 15 May 2002
Accepted 31 May 2002

1220
Y. MURAKAMI ET AL.
Introduction
N-(4-acetyl-1-piperazinyl)-p-fluorobenzamide (FK960) is a novel anti-
dementia drug candidate that has displayed improved memory
impairment in a variety of animal models.1,2 FK960 is a somatostatin
releaser3,4 and it is well known that somatostatin releases acetylcholine.5
Therefore, FK960 indirectly activates the release of acetylcholine, and
shows efficacy in animal models of scopolamine-induced amnesia1,2 and
neuronal dysfunction.6
As described above, the mechanism and pharmacological efficacy of
FK960 are becoming clear, however in general, a suitable dose setting is
important in order to succeed in clinical trials. Positron emission
tomography (PET) can provide important information about the
pharmacokinetics of a compound, especially the permeability of the
blood-brain barrier both in animals and humans. In order to perform a
PET study, labelling of the compound is essential. Because FK960
contains a benzoylamide with a F-atom in the para position, which
makes the compound suitable for labelling with 18F, the present study
demonstrates the development of a rapid synthesis of 18F labelled
FK960 using nucleophilic fluorination with a fully automated synthetic
apparatus, and the distribution study of 18F labelled FK960 in
conscious monkey brain using PET.
Results and discussion
Chemistry
We first considered the straightforward synthesis shown in Figure 1. We
attempted a cold synthesis of [18F]FK960 by nucleophilic substitution
reactions of compounds 1 under various conditions. However, the
reaction did not proceed due to lack of reactivity of the aromatic ring.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

SYNTHESIS OF 18F LABELLED FK960
1221
Figure 1. Attempted preparation of [18F]FK960 in one step from compounds 1
Figure 2. Synthetic route for [18F]FK960. *WSC: 1-[3-(Dimethylamino)propyl]-
3-ethylcarbodiimide. **HOBt: N-hydroxybenzotriazole
These results led us to a stepwise route for the synthesis of [18F]FK960
involving nucleophilic substitution on an aromatic ring as the key
fluorination reaction, as shown in Figure 2. An ethyl benzoate
derivative was chosen as the starting material because hydrolysis could
be performed without cleavage of the F-atom. The fluorination
conditions were examined based upon reference to various articles.7–9
For the amidation of 4-fluorobenzoic acid, the reaction mixture was
neutralized with 1N-HCl and the solution containing 4-fluorobenzoic
acid was evaporated. The evaporation was performed to prevent
reaction of methyl sulfoxide (DMSO) with the amidation agent, WSC
(ethyl N,N-dimethylaminopropyl carbodiimide mono hydrochloride),
and to increase the concentration of the reagents. The neutralization
before the amidation was critical, because the pH was shown to be the
most important parameter in studies of cold synthesis. It was confirmed
in our experiments that the amidation reaction proceeds in the pH
3.8 – 9.3 region. After evaporation, the amidation was conducted
without purification in N,N-dimethylformamide (DMF) by addition of
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

1222
Y. MURAKAMI ET AL.
Table 1. Automated synthesis of [18F]FK960
Starting material
1a
1b
Fluorination
Preparation time
Radioactivity (MBq)
Radiochemical purity (%)
Specific activity (GBq/mmol)
1208C, 10 min
2 h 12 min–2 h 53 min
320–940
1508C, 10 min
2 h 15 min–2 h 23 min
115–633
87–97
2.0–3.1
87–95
13.0–60.2
the reagents to the mixture. However, the entire process from starting
material took approximately 2 h owing to the two evaporation steps.
Ethyl 4-trimethylammonium benzoate (2a) was first chosen as a starting
material because of higher reactivity than the corresponding 4-nitro
derivative.10 The fluorination proceeded smoothly and the next two
steps were also efficient. Consequently [18F]FK960 was obtained in
good yield. On the other hand, the material produced in this way had
low specific activity. We considered the reason to be isotopic dilution
from the starting material during the evaporation before neutralization.
In order to avoid the use of the trifluoromethanesulfonate, ethyl 4-
nitrobenzoate (2b) was chosen as the starting material. The fluorination
reaction required 308C higher temperatures than with compound (2a)
and the amounts of side-product resulting from reaction of the product
with dimethyl amine derived from DMF were increased. [18F]FK960
with a specific radioactivity of 13.0–60.2 GBq/mmol was obtained
(Table 1). The synthesis using ethyl 4-nitrobenzoate (2b) produced
specific radioactivities 50–100 fold higher than when the trifluoro-
methanesulfonate (2a) was used.
PET study
The time activity curves shown in Figure 3 display the brain uptake of
no-carrier added [18F]FK960. After injection of [18F]FK960, the
radioactivity in whole blood and plasma decreased gradually, whereas
the radioactivity in the brain increased gradually for 50 min and was
maintained for 2 h. The radioactivity level in plasma was approximately
3 times higher than each ROI in the brain until 2 h after injection. No
difference among the ROIs (the cerebellum, hippocampus, temporal
cortex, occipital cortex, frontal cortex) was observed in the present
study. FK960 is taken up gradually by the brain and is distributed to the
whole brain. FK960 is a somatostatin releaser, however the present
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

SYNTHESIS OF 18F LABELLED FK960
1223
Figure 3. Time activity curves of [18F]FK960 in brain and blood of conscious
monkey
study did not show the specific binding site of FK960 in conscious
monkey brain. Although the target molecule of FK960, and its
distribution in the brain still remain to be elucidated, in the present
study we could measure the distribution and pharmacokinetic data of
FK960 in conscious monkey brain using PET. Furthermore, the present
study suggests the possibility that using the same method, we will be
able to obtain distribution and pharmacokinetic data in humans, which
may provide us important information for clinical dose setting.
Experimental
Materials and reagents
Kryptofix222 was purchased from Aldrich Japan (Tokyo, Japan).
Methyl sulfoxide (DMSO), NaOH 1N, and HCl 1N were purchased
from Nacalai Tesque (Kyoto, Japan). Potassium fluoride, potassium
carbonate, N,N-dimethylformamide (DMF), ammonium acetate, acetic
acid and tetrahydrofuran (THF) were purchased from Wako Pure
Chemical Industries, Ltd (Osaka, Japan). 1-Hydroxybenzotriazole was
purchased from Peptide Institute Inc (Osaka, Japan). WSC was
purchased from Eiweiss (Shizuoka, Japan). DMSO was distilled under
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

1224
Y. MURAKAMI ET AL.
vacuum from calcium hydride and kept over 4A molecular sieves.
Preparative TLC was performed on Merck F-254 silica gel 60 plates
(0.5 mm). Sterile filters were purchased from Millipore Japan (Tokyo,
Japan). Compound (2a) was synthesized according to the literature.10
Analysis
1H-NMR spectra were recorded on an AC200P spectorometer (Bruker
BioSpin K.K. Ibaraki, Japan). HPLC analysis was performed with a
Tosoh CCPM-II pump (Tosoh Co., Ltd, Tokyo, Japan), an Aloka
positron detector RLC-700 (Aloka Co., Ltd, Tokyo, Japan), a UV
detector Tosoh UV-8020.
HPLC analysis of FK960
Column: Jasco finepak SIL C18S (4.6 mm  150 mm, 5 mm) (Jasco Co.,
Ltd, Tokyo, Japan)
Mobile phase: (0.05 N NH4OAc:AcOH=250:1):THF=87.5:12.5.
Flow rate: 1.0 ml/min
Synthesis of FK960
To a solution of potassium fluoride (3 mg, 52 mmol), potassium
carbonate (3 mg, 22 mmol), and kryptofix222 (35 mg, 93 mmol), was
added a solution of the starting material (ethyl 4-trimethylammonium-
benzoate triflate (2a) (18.5 mg, 52 mmol) or ethyl 4-nitrobenzoate
(2b)(9.2 mg, 52 mmol) in DMSO (0.6 ml). The mixture was heated at
1208C for 10 min (starting material 2a) or 1508C for 10 min (starting
material 2b). The reaction mixture was cooled to room temperature. To
this mixture was added 0.8 ml of 1N NaOH and stirred at 808C for
5 min, and then the solvent was removed under reduced pressure. To
this crude product was added 1-acetyl-6-aminopiperazine (5), (10.2 mg,
71 mmol in 0.8 ml 1 M HCl) and stirred at room temperature for 2 min.
To the mixture was added 1-hydroxybenzotriazole (9.6 mg, 71 mmol)
and WSC (13.6 mg, 71 mmol) in DMF (1.0 ml). The mixture was stirred
at room temperature for 10 min, and then the solvent was removed in
vacuo. The purification was performed on preparative TLC (10%
MeOH–CH2Cl2 Rf: 0.3) to give FK960 (2.8 mg) (22.4%).
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

SYNTHESIS OF 18F LABELLED FK960
1225
1H-NMR (DMSO-d6) d ppm 2.02(3H, s), 2.76–2.94(4H, m),
3.45–3.60(4H, m), 7.29(2H, t, J=8.8 Hz), 7.84(2H, dd, J=5.6, 8.8 Hz),
9.59(1H, s)
Synthesis of [18F]FK960
Production of 18FÀ was accomplished via 18O(p, n)18F reaction by
proton bombardment (18 MeV, 20 mA) of a 18O–water target using a
cyclotron-target system (CYPRIS HM-18,Sumitomo Heavy Industries
Co., Ltd). After irradiation, the 18F containing target water was
transferred to a 18F recovery system. The target material was passed
through an AG1-X8 (Bio-Rad) anion exchange resin and recovered as
[18F]KF by elution with aqueous potassium carbonate (0.3 ml, 66 mM)
followed by water (0.3 ml) into a conical glass vial. To this solution
kryptofix2.2.2. (20 mg, 53 mmol) in acetonitrile was added. The
recovered 18FÀ solution was then transferred to the synthesis apparatus
using He gas. The radioisotope detector was used to monitor the arrival
of radioactivity from the glass vial. The aqueous solution was removed
azeotropically with acetonitrile at 1208C. This operation was repeated
three times. To the residue was added a solution of the starting material
(ethyl 4-trimethylammoniumbenzoate triflate (2a) (18.5 mg, 52 mmol) or
ethyl 4-nitrobenzoate (2b)(9.2 mg, 52 mmol) in DMSO (0.6 ml). The
mixture was heated at 1208C for 10 min (starting material 2a) or 1508C
for 10 min (starting material 2b). After the reaction mixture was cooled
to room temperature, 1N-NaOH (0.8 ml) was added. The mixture was
heated 808C for 5 min, and the solvent was evaporated by N2 flush and
reduced pressure. The residue was neutralized in a mixed solution of
1-acetyl-6-aminopiperazine (5), (10.2 mg, 71 mmol in 0.8 ml 1 M HCl)
and water (0.5 ml). The mixture was stirred for 2 min at room
temperature. To this mixed solution was added a mixed solution of
1-hydroxybenzotriazole (9.6 mg, 71 mmol) and 1-(3-dimethylamino-
propyl)-3-ethylcarbodiimide hydrochloride (13.6 mg, 71 mmol) in DMF
(1.2 ml). The mixture was stirred at room temperature for 10 min. After
the reaction, the mixture was evaporated by N2 flush and reduced
pressure in a similar manner. A mixture of water and methanol was
added to the reaction vessel and the residue was dissolved and the
solution was loaded on to an HPLC column. The fraction containing
[18F]FK960 was collected in the evaporator. The eluate was evaporated
by N2 flush and reduced pressure. The residue was dissolved with saline
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

1226
Y. MURAKAMI ET AL.
and passed through a sterile filter into a sterile vial. The retention time
of the product (6.8 min) was identical to that of the authentic sample.
Preparative HPLC was performed with a Rabo-Quortec pump system
(Rabo- Quortec Co., Ltd Tokyo, Japan).
Column: Jasco Megapak SIL C18-10 (7.5 mm  250 mm, 5 mm)
Mobile phase: (0.05N NH4OAc:AcOH=250:1):THF=87.5:12.5.
Flow rate: 4.0 ml/min
Retention time: FK960 (8.5 min)
PET study
Three young-adult male rhesus monkeys (Macaca mulatta) weighing
from 4 to 6 kg were used for the PET measurements. Monkeys were
maintained and handled in accordance with recommendations of the US
National Institutes of Health and also the guidelines of the Central
Research Laboratory, Hamamatsu Photonics. They were trained to sit
on a chair twice a week over a period of more than 3 months. Magnetic
resonance images (MRI) of all monkeys were obtained with a
TOSHIBA MRT-50A/II (0.5 T) under pentobarbital anesthesia. The
stereotaxic coordinates of PET and MRI were adjusted based on the
orbitomeatal (OM) line with a specially designed head holder.11 At least
1 month before the PET study, an acrylic plate, with which the monkey
was fixed to a monkey chair, was attached to the head under
pentobarbital anesthesia as described previously.12
Data were collected on a high-resolution PET scanner (SHR-7700,
Hamamatsu Photonics, Hamamatsu, Japan) with transaxial resolution
of 2.6 mm full width at half maximum (FWHM) and a center-to-center
distance of 3.6 mm.13 The PET camera allowed 31 slices for imaging to
be recorded simultaneously.
After an overnight fast, animals were fixed to the monkey chair with
stereotaxic coordinates aligned parallel to the OM line. A cannula was
implanted into the posterior tibial vein of the monkey for administra-
tion of [18F]FK960, and another cannula was put into the femoral vein
of the other leg to obtain blood samples. PET scans were performed
under dim light.
[18F]FK960 (100–120 MBq/kg body weight) was injected through the
posterior tibial vein cannula, and PET scan was performed for 112 min
with 16 time frames at 1 min intervals, followed by 32 time frames at
3 min.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

SYNTHESIS OF 18F LABELLED FK960
1227
Regions of interest (ROIs), i.e. the cerebellum, hippocampus,
temporal cortex, occipital cortex, frontal cortex, were identified
according to MR images of each monkey brain, and time activity
curves in ROIs were obtained.
Blood samples were obtained 5, 10, 30, 60, 90 and 120 min after
[18F]FK960 injection. Blood samples were centrifuged to separate
plasma, weighed, and the radioactivity was measured.
Conclusion
In the present study, we developed a one-pot reaction for the synthesis
of [18F]FK960, and succeeded to prepare this labelled drug as an
injectable solution, and obtained the distribution and pharmacokinetic
data in conscious monkey brain using a [18F]FK960 PET study. After
injection of [18F]FK960, the radioactivity in the brain increased
gradually for 50 min and then was maintained up to 2 h. The
radioactivity level in plasma was 3 times as high as each ROI in the
brain. Differences between ROI were not observed in this study. The
present study suggests the possibility to measure distribution and
pharmacokinetic data in humans using the same method.
Acknowledgements
We wish to thank Dr. Toshiki Kobayashi of NKK corporation,
Mrs. Satoshi Nakamura, Hajime Miyoshi and Naotaka Sawada of
Dainihonseiki Co Ltd. for helping to make our automated apparatus.
References
1. Yamazaki M, Matsuoka N, Maeda N, Ohkubo Y, Yamaguchi I.
J Pharmacol Exp Ther. 1996; 279: 1157–1173.
2. Matsuoka N, Aigner TG. J Pharmacol Exp Ther 1997; 280: 1201–1209.
3. Matsuoka N, Satoh M. Brain Res 1998; 794: 248–254.
4. Inoue T, Wang F, Moriguchi A, Shirakawa K, Matsuoka N, Goto T.
Brain Res 2001; 892: 111–117.
5. Araujo DM, Lapchak PA, Collier B, Quirion R. J Neurochem 1990; 55:
1546–1555.
6. Tsukada H, Yamazaki S, Noda A. et al. Brain Res 1999; 832: 118–123.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228

1228
Y. MURAKAMI ET AL.
7. Angelini G, Speranza M, Wolf AP, Shiue C.-Y. J Fluorine Chem 1985; 27:
177–192.
8. Attina M, Cacace F, Wolf AP. J Chem Soc Chem Commun 1983; 108–109.
9. Shiue C-Y, Watanabe M, Wolf AP, Fowler J, Salvadori P. J Label Compd
Radiopharm 1984; 21: 533–547.
10. Haka, MS., Kilbourn, MR., Watkins, GL, Toorongian, SA. J Label
Compd Radiopharm 1988; 27: 823–833.
11. Takechi H, Onoe H, Imamura K. Neurosci Lett 1994; 182: 279–282.
12. Onoe H, Inoue O, Suzuki K, et al. Brain Res 1994; 663: 191–198.
13. Watanabe M, Okada H, Shimizu K, et al. IEEE Trans Nucl Sci 1997; 44:
1277–1282.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 1219–1228