Synthesis and brain regional distribution of [11C]NPS 1506 in mice and rat: An N-methyl-D-aspartate (NMDA) receptor antagonist

Article abstract of DOI:10.1248/bpb.26.1570

NPS 1506 [3-fluoro-γ-(3-fluorophenyl)-N-methylbenzenepropamine] is representative of a non-psychotomimetic class of N-methyl-D-aspartate (NMDA) receptor antagonists. [¹¹C]NPS 1506 was prepared at high radiochemical purity (γ98%) with a specific activity of around 50 GBq/μmmol at the end of synthesis by methylation of the desmethyl precursor with [¹¹C]methyl iodide in the presence of NaH. Biodistribution of [¹¹C]NPS 1506 in mice and rat demonstrated that uptake into the brain was rapid and occurred at high levels. [¹¹C]NPS 1506 showed no appreciable specific binding in rodent brains under in vivo conditions, possibly because of both a large non-specific bound fraction and low in vitro binding affinity for NMDA receptors.

Full text of DOI:10.1248/bpb.26.1570

1570
Biol. Pharm. Bull. 26(11) 1570—1573 (2003)
Vol. 26, No. 11
Synthesis and Brain Regional Distribution of [11C]NPS 1506 in Mice and
Rat: an N-Methyl-D-aspartate (NMDA) Receptor Antagonist
Takeshi FUCHIGAMI,a Terushi HARADAHIRA,b,d Takuya ARAI,b,d Takashi OKAUCHI,c,d Jun MAEDA,c,d
Kazutoshi SUZUKI,b Fumihiko YAMAMOTO,a,d Tetsuya SUHARA,c,d Shigeki SASAKI,a,d and
,a,d
*
Minoru MAEDA
a Graduate School of Pharmaceutical Sciences, Kyushu University; 3–1–1 Maidashi, Higashi-ku, Fukuoka 812–8582,
Japan: b Department of Medical Imaging, National Institute of Radiological Sciences; c Brain Imaging Project, National
Institute of Radiological Sciences; 4–9–1 Anagawa, Inage-ku, Chiba 263–8555, Japan: and d CREST, Japan Science and
Technology Corporation; 4–1–8 Honmachi, Kawaguchi, Saitama 332–0012, Japan.
Received June 16, 2003; accepted August 20, 2003
NPS 1506 [3-fluoro-g-(3-fluorophenyl)-N-methylbenzenepropamine] is representative of a non-psychoto-
mimetic class of N-methyl-D-aspartate (NMDA) receptor antagonists. [11C]NPS 1506 was prepared at high radio-
chemical purity (Ͼ98%) with a specific activity of around 50 GBq/mmol at the end of synthesis by methylation of
the desmethyl precursor with [11C]methyl iodide in the presence of NaH. Biodistribution of [11C]NPS 1506 in
mice and rat demonstrated that uptake into the brain was rapid and occurred at high levels. [11C]NPS 1506
showed no appreciable specific binding in rodent brains under in vivo conditions, possibly because of both a large
non-specific bound fraction and low in vitro binding affinity for NMDA receptors.
Key words NPS 1506; carbon-11; brain distribution; N-methyl-D-aspartate (NMDA) receptor
The glutamatergic N-methyl-D-aspartate (NMDA) recep- MATERIALS AND METHODS
tors have been considered prime therapeutic targets for the
development of useful neuroprotective strategies.1) At pre-
Nuclear magnetic resonance (1H-NMR) spectra were
sent, many NMDA receptor antagonists, as exemplified by recorded on a JEOL GX-270 spectrometer 270 MHz with
high-affinity channel blocker MK-801 and PCP, have serious tetramethylsilane as an internal standard. All chemical shifts
unwanted effects, such as the induction of a psychotomimetic (d) are reported in parts per million (ppm) downfield from
effect or cardiovascular dyregulation.2)
the standard. Mass spectra were obtained on a JMS DX-610
Recently a novel non-psychotomimetic class of NMDA re- spectrometer. Infrared (IR) spectra were recorded with a
ceptor open-channel blockers with a diphenylpropylamine JASCO IR Report-100 spectrometer. Column chromatogra-
skeleton has been described. NPS 1506 [3-fluoro-g-(3-fluo- phy was done on Merck Kieselgel 60 (70—230 mesh), and
rophenyl)-N-methylbenzenepropamine] is representative of analytical thin layer chromatography (TLC) was carried out
this class and currently used in human clinical trials for the on Silica gel 60 F254 plates (Merck). In the synthetic proce-
treatment of acute ischemic stroke.3,4) Pharmacological stud- dures, organic extracts were routinely dried over anhydrous
ies on NPS 1506 have shown5): (1) that this molecule is a Na2SO4 and evaporated with a rotatory evaporator under re-
moderate-affinity uncompetitive NMDA receptor antagonist duced pressure. Radioactivity was quantified with a IGC-3R
(IC50ϭ664 nM at displacing radioligand from [3H]MK-801 Curiemeter (Aloka). High-pressure liquid chromatography
labeled binding sites); and (2) that NPS 1506 is neuroprotec- (HPLC) was done using a Waters HPLC system for nonra-
tive in a variety of animal models of stroke and head injury; dioactive runs or a JASCO HPLC system for radioactive
(3) that, importantly, at doses greater than those required for runs. Effluent radioactivity from the HPLC was determined
neuroprotection, NPS 1506 did not elicit in rodents the char- using a NaI (Tl) scintillation detector system. Carbon-11
acteristic side effect profile and had no PCP-like psy- was generated by the 14N(p, a)11C nuclear reaction using a
chotomimetic effects at doses between 1—5 mg/kg i.p. in CYPRIS HM-18 cyclotron (Sumitomo Heavy Industries,
rats.
Ltd.). Preparation of [11C]CH3I and subsequent 11C-methyla-
Such a pharmacological profile, different from the previ- tion were carried out automatically by using a synthetic ap-
ous NMDA antagonists, led us to label NPS 1506 with car- paratus for 11C-labeled compounds developed by Suzuki et
bon-11 and study the distribution and pharmacokinetics of al.7) 3-Nitropropionic acid was purchased from Sigma Co.
the tracer in animals. Positron emission tomography (PET) is and all reagents were used as received unless otherwise
an imaging modality for in vivo assessment of drug distribu- stated. The animal experiments were carried out according to
tion and interaction with biochemical target systems by the the recommendations of the committee for the care and use
use of tracers labeled with positron emitters such as carbon- of laboratory animals, National Institute of Radiological Sci-
11 (T1/2ϭ20.4 min).6) In this paper, we describe the synthesis ences.
of 11C-labeled NPS 1506 and preliminary results on regional
brain distribution studies by the tissue dissection method in (NPS 1506) To a stirred solution of the hydrochloride salt
3-Fluoro- -(3-fluorophenyl)-N-methylbenzenepropamine
g
mice and by PET in a rat, with special attention to the ability of 3-fluoro-g-(3-fluorophenyl)benzenepropamine (1)8) (100
of specific binding in vivo.
mg, 0.35 mmol) in a mixture of water and methanol (4 : 1,
2.5 ml) was added cyclopentadiene (130 ml, 1.58 mmol) and
an aqueous solution of 37% formaldehyde (64 ml, 0.78
To whom correspondence should be addressed. e-mail: maeda@phar.kyushu-u.ac.jp
© 2003 Pharmaceutical Society of Japan

November 2003
1571
mmol). After 66 h, the reaction mixture was neutralized with umn; nacalai tesque COSMOSIL 5C-18-AR, 4.6ϫ250 mm,
a 5% sodium bicarbonate solution and extracted with methyl- UV at 254 nm, mobile phase; CH3CN : 0.2% TFA in H2Oϭ
ene chloride (20 mlϫ3). The combined extracts were washed 35 : 65, flow rate; 1.5 ml/min). Confirmation of the identity of
with a saturated NaCl aqueous solution (30 ml), dried [11C]NPS 1506 was achieved with authentic NPS 1506. The
and evaporated to dryness. The crude product was chro- specific activity at end of synthesis (EOS) was calculated by
matographed on a silica gel with CHCl3 : MeOH : NH4OHϭ UV spectroscopy (254 nm) to be an average of 50 GBq/mmol.
49 : 1 : 0.1 to provide 2-azanorbornene adduct (2) (4.4 mg,
Brain Distribution in Mice [11C]NPS 1506 (in 0.2 ml
1
91%) as a colorless oil. H-NMR (CDCl3) d: 1.43 (2H, t, saline, ca. 7.3 MBq) was injected intravenously via a tail vein
Jϭ6.4 Hz), 2.37—2.06 (7H, m), 3.19 (1H, q, Jϭ3.9 Hz), into ddY mice (40—42 g). These mice (nϭ3 per time point)
3.96 (1H, t, Jϭ7.5 Hz), 5.93 (1H, dd, Jϭ5.6, 2.0 Hz), 6.31 were killed at 1, 20, and 60 min by decapitation, the whole
(1H, dd, Jϭ5.6, 3.0 Hz), 7.02—6.84 (6H, m), 7.24 (2H, dt, brains were rapidly removed and dissected into the cerebral
Jϭ14.3, 7.5 Hz). IR (CHCl3) cmϪ1: 2864, 1610, 1589. FAB- cortex, striatum, hippocampus, and cerebellum. The samples
MS (m/z): 326 (MϩH)ϩ.
were weighed and their radioactivities were measured with a
To a solution of the 2-azanorbornene (2) (35 mg, 0.108 Packard g-counter and corrected for decay. A sample of
mmol) in CHCl3 (1 ml) was added trifluoroacetic acid (1 ml) blood was also taken. The results were expressed as the
and triethylsilane (165 ml, 1.03 mmol). The resulting mixture percent of administered dose per gram of tissue (% dose/g).
was stirred at room temperature for 72 h. The solvent was re- In separate experiments, the effect of the carrier dose on
moved under reduced pressure. The crude product was dis- the brain uptake was investigated. Unlabeled NPS 1506
solved in chloroform (0.5 ml), treated with 10% hydrochloric (3 mg/kg) was co-injected with [11C]NPS 1506 into the mice,
acid (5.0 ml), and washed with a mixture of hexane and ether which were sacrificed at 20 min after injection. The brain was
(1 : 1). The aqueous layer was neutralized with a 5% sodium removed, dissected and regional radioactivity was determined
bicarbonate solution and then the product was extracted with as described above. The control mice were treated with a
methylene chloride. The combined extracts were washed saline solution under identical conditions.
with a saturated NaCl aqueous solution (20 ml), dried and
PET of [11C]NPS 1506 in a Rat One conscious rat
evaporated to dryness. The residue was purified by HPLC (male Sprague-Dawley, 10 weeks old) was used for control
(column; nacalai tesque COSMOSIL 5C-18-MS, 10ϫ250 scan and subsequent drug-pretreatment scan. The rat was
mm, mobile phase; CH3CN: 0.2% TFA in H2Oϭ35 : 65, flow fixed with a specifically designed head-hold device,9) and
rate; 4.0 ml/min) to give NPS 1506 (23 mg, 81%) as a color- PET scans were performed with a high-resolution SHR-7700
1
less oil. H-NMR (CDCl3) d: 2.23 (2H, q, Jϭ7.5 Hz), 2.41 PET camera (Hamamatsu Photonics Shizuoka, Japan) which
(3H, s), 2.56 (2H, t, Jϭ7.3 Hz), 4.04 (1H, t, Jϭ7.8 Hz), provides with 31 transaxial slices 3.6 mm (center-to-center)
7.03—6.85 (6H, m), 7.24 (2H, dt, Jϭ14.2, 6.2 Hz). IR (KBr) apart and a 14.8 cm field of view. After transmission scan for
cmϪ1: 3022, 1676. FAB-MS (m/z): 262 (MϩH)ϩ.
attenuation correction, dynamic emission scans were per-
3-Fluoro-g-(3-fluorophenyl)-N-[11C]methylbenzene- formed for 60 min (30 sϫ6 scans, 1 minϫ7 scans, 2 minϫ10
propamine ([11C]NPS 1506) The hydrochloride salt of scans, 5 minϫ6 scans) following the [11C]NPS 1506 injection
(1) was further purified by HPLC (column; nacalai tesque (30 MBq and 54 MBq for control and pretreatment experi-
COSMOSIL 5C-18-MS, 10ϫ250 mm, mobile phase; CH3CN: ment, respectively) via the tail vein. The scan images were
0.2% TFA in H2Oϭ35 : 65, flow rate; 4.0 ml/min) and the tri- reconstructed with a Hanning filter of 4 mm, and circular
fluoroacetic acid salt obtained (10 mg, 0.027 mmol) was neu- regions of interest (ROIs) with a 5 mm diameter were placed
tralized with a 5% sodium bicarbonate solution (5 ml) and over the striatum and hippocampus using image analysis
extracted with ether (5 mlϫ4). The combined extracts were software (in-house software PET Analyzer). The ROI values
washed with a saturated NaCl aqueous solution (20 ml), were expressed in kBq/ml, normalized to the injected
dried and evaporated to dryness to give the free base of (1) radioactivity of 37 MBq, and plotted against time (Fig. 2).
(7 mg, 100%). Preparation of [11C]CH3I and subsequent 11C- In the pretreatment experiment, 3-nitropropionic acid (20
methylation of (1) to [11C]NPS1506 were achieved automati- mg/kg, 11.6 mg/0.8 ml saline) was injected intraperitoneally
cally using specially designed equipment.7) The [11C]CH3I 3 h prior to the [11C]NPS 1506 injection.
prepared was trapped in anhydrous DMF (250 ml) containing
the free base of (1) (1.0 mg) and NaH (4.5 ml, 1 M in DMF) at RESULTS AND DISCUSSION
Ϫ15—Ϫ20 °C. The reaction vessel was heated at 30 °C and
kept there for 3 min. After addition of an HPLC mobile
NPS 1506 presents in its chemical structure one N-methyl
phase (500 ml), the radioactive mixture was transferred onto group, which was considered to the most accessible site for
an HPLC (column; nacalai tesque COSMOSIL 5C-18-MS, labeling with carbon-11 using [11C]methyl iodide. The pre-
10ϫ250 mm; mobile phase; CH3CN : 0.2% TFA in H2Oϭ cursor for 11C-labeling, 3-fluoro-g-(3-fluorophenyl)benzene-
35 : 65, flow rate; 4.0 ml/min). A radioactive fraction having propamine (1) was prepared according to the published
a retention time of 10 min was collected in a flask. After method.8) NPS 1506 as the reference compound was pre-
evaporation of the solvents from the flask under reduced pared by the two-step N-methylation via a retro aza Diels–
pressure, the pure [11C]NPS 1506 was reconstructed in sterile Alder reaction (Chart 1).10) The immonium ion generated in
0.9% saline. By this procedure, radiochemical pure (Ͼ98%) situ from the hydrochloride salt of (1) and aqueous formalde-
[11C]NPS 1506 (300—370 MBq) was obtained in a synthesis hyde in water underwent a facile cyclocondenzation with cy-
time of 20 min from end of bombardment (EOB) after a 15— clopentadiene to give the 2-azanorbornene adduct (2) in 91%
20 min proton bombardment at a beam current of 15 mA. Ra- yield. Subsequent treatment of (2) in CHCl3/CF3CO2H with
diochemical purity was assayed by analytical HPLC (col- triethylsilane gave the desired NPS 1506 in 81% yield. Com-

1572
Vol. 26, No. 11
Chart 1. Synthesis of Target Compound NPS 1506
Fig. 1. Uptake of Radioactivity in Mouse Regional Brain and Blood at
20 min after Intravenous Injection of [11C]NPS 1506 Together with NPS
1506 (3 mg/kg) or without NPS 1506 (Control)
Chart 2. Radiosynthesis of [11C]NPS 1506
The results are means (nϭ3)ϮS.D. expressed as % dose/g. No significant differences
were detected. CTX: cerebral cortex, STR: striatum, HIP: hippocampus, CB: cerebel-
lum.
Table 1. Biodistribution of [11C]NPS 1506 in Mouse Brain and Blood
% dose/g (mean)ϮS.D.a)
Tissue
1 min
20 min
60 min
Blood
Cerebral cortex
Striatum
Hippocampus
Cerebellum
0.87Ϯ0.07
1.98Ϯ0.32
1.76Ϯ0.22
1.69Ϯ0.25
1.98Ϯ0.24
0.42Ϯ0.05
3.84Ϯ0.35
3.34Ϯ0.34
3.31Ϯ0.44
3.41Ϯ0.24
0.42Ϯ0.06
4.16Ϯ0.53
4.42Ϯ0.39
4.00Ϯ0.67
3.99Ϯ0.41
a) Average of three mice.
pound (1), (2) and NPS 1506 have analytical data such as 1H-
NMR and MS in agreement with the indicated structures.
[11C]NPS 1506 was prepared by methylation of (1) with
[11C]methyl iodide in DMF in the presence of NaH at 30 °C
for 3 min (Chart 2). The crude 11C-reaction mixture was sub-
jected to HPLC purification to give [11C]NPS 1506 as an
aqueous solution with Ͼ98% radiochemical purity. The pre-
cursor (1) was not detected by analytical HPLC system in the
final solution of [11C]NPS 1506. The total synthesis time was
20 min from EOB, and the specific activity was around
50 GBq/mmol at EOS. Although the reaction conditions have
not been optimized, an average 330 MBq of [11C]NPS 1506
after a 15—20 min proton bombardment at a beam current of
15 mA was obtained at EOS.
The brain biodistribution of [11C]NPS 1506 after intra-
venous injection was investigated using mice (Table 1).
Blood radioactivity was very low at 1 min (ca. 1% dose/g)
and fell to Ͻ0.5% dose/g at 20 min after injection, indicating
a rapid distribution of [11C]NPS 1506 into the tissues.
Radioactivity uptake in the brain was high and increased
slightly with time, indicative of good blood–brain penetra-
tion. The regional brain distribution was fairly uniform and
did not clearly reflect the expected distribution which is
based on the known localization of NMDA receptors in the
rodent brain.11,12) We evaluated the effect of co-injection
of unlabeled NPS 1506 (3 mg/kg per mouse) along with
[11C]NPS 1506 20 min after the tracer injection. As can be
seen from Fig. 1, treatment with unlabeled NPS 1506 did not
cause any change of the uptake in the mouse brain compared
to the control group, suggesting that the brain uptake ob-
served was unlikely to be mediated by a saturable, specific
mechanism.
Fig. 2. Time Activity–Curves (TAC) in the Hippocampus and Striatum of
Rat Obtained by PET after Intravenous Injection of [11C]NPS 1506; Control
(᭹) and Pretreatment with 20 mg/kg of 3-Nitropropionic Acid at 3 h before
the Tracer Injection (᭝)
Each point was normalized to the injected dose of 37 MBq.
In vivo distribution was further assessed in the rat brain.
Figure 2 shows the time–activity curves (TACs) in the hip-
pocampus and striatum containing high densities of NMDA
receptor for 60 min PET scanning after intravenous injection
of [11C]NPS 1506. Regional radioactivity in the hippocam-
pus and striatum gradually increased with time, having a
plateau from 30 min onwards, again reflecting good penetra-
tion of the tracer across the BBB. Recent studies have shown
that the channel open probability of the NMDA receptors is
not very high at the normal brain activity.13) It is known that
3-nitropropionic acid (3-NP) causes NMDA receptor activa-
tion in the striatum and cortex when administrated acutely

November 2003
1573
to rats, as demonstrated by an increase in [3H]MK-801 bind-
ing.14) We examined the effect of prior treatment with 3-NP
on the brain uptake of [11C]NPS 1506. As can be seen from
Fig. 2, pretreatment with 3-NP (20 mg/kg) given intraperi-
toneally 3 h prior to tracer injection did not produce any
change in the regional uptake in the hippocampus nor the
striatum.
ported in part by CREST of Japan Science and Technology
Corporation.
REFERENCES AND NOTES
1) Brauner-Osborne H., Egebjerg J., Nielsen E. O., Madsen U., Krogs-
gaard-Larsen P., J. Med. Chem., 43, 2609—2645 (2000).
2) Kemp J. A., Mckernan R. M., Nature Neurosci. Suppl., 5, 1039—1042
(2002).
In summary, the labeling of NPS 1506 with carbon-11 in
the N-methyl position was accomplished with high specific
activity and high radiochemical purity. [11C]NPS 1506 ad-
ministered in tracer doses readily crosses the BBB consistent
with its partition coefficient (log Pϭ3.81).15) It has been re-
ported that a number of NMDA receptor antagonists acting
at the glycine site cross the BBB very poorly, despite their
lack of dose-limiting side effects.16) The good access of NPS
1506 to brain tissue gives it an advantage as an effective ther-
apeutic agent over other antagonists. We could not prove that
the accumulation of [11C]NPS 1506 in rat and mouse brains
was due to the specific binding to NMDA receptor binding
sites. This could possibly result from both low in vitro affin-
ity for NMDA receptors and large non-specific bound frac-
tions, which might have masked the interaction with specific
binding sites. In the present work, no metabolite studies were
undertaken, and therefore whether the unspecific pattern of
brain distribution of [11C]NPS 1506 was due to radiolabeled
metabolites that might have been formed in brain or pene-
trate the BBB remains uncertain. The observed brain reten-
tion of radioactivity may be related mainly to the affinity of
[11C]NPS 1506 itself for high-capacity, nonspecific binding
sites in brain tissues, similar to the accumulation mechanism
proposed for other lipophilic amines such as N-isopropyl-4-
3) Moe S. T., Shimizu S. M., Smith D. L., Van Wagenen B. C., DelMar E.
G., Balandrin M. F., Chien Y. (E)., Raszkiewicz J. L., Artman L. D.,
Mueller A. L., Lobkovsky E., Clardy J., Bioorg. Med. Chem. Lett., 9,
1915—1920 (1999).
4) Moe S. T., Smith D. L., DelMar E. G., Shimizu S. M., Van Wagenen
B. C., Balandrin M. F., Chien Y. (E)., Raszkiewicz J. L., Artman L. D.,
White H. S., Mueller A. L., Bioorg. Med. Chem. Lett., 10, 2411—2415
(2000).
5) Mueller A. L., Artman L. D., Balandrin M. F., Brady E., Chien Y. (E).,
DelMar E. G., George K., Kierstead A., Marriott T. B., Moe S. T.,
Newman M. K., Raszkiewicz, J. L., Sanguinetti E. L., Van Wagenen B.
C., Wells D., Ann. N.Y. Acad. Sci., 890, 450—461 (1999).
6) Gibson R. E., Burns H. D., Hamill T. G., Eng W., Francis B. E., Ryan
C., Curr. Pharm. Design, 6, 973—989 (2000).
7) Suzuki K., Inoue O., Hashimoto K., Yamasaki T., Kuchiki M., Tamate
K., Appl. Radiat. Isot., 36, 971—976 (1985).
8) Moe S. T., Smith D. L., By K., Egan J. A., Filer C. N., J. Labelled
Compds. Radiopharm., 41, 535—543 (1998).
9) Okauchi T., Suhara T., Maeda J., Kawabe K., Obayashi S., Suzuki K.,
Synapse, 41, 87—95 (2001).
10) Griecok P. A., Bahsas A., J. Org. Chem., 52, 5746—5749 (1987).
11) Wong E. H. F., Knight A. R., Woodruff G. N., J. Neurochem., 50,
274—281 (1988).
12) Ferrarese C., Guidotti A., Costa E., Miletich R. S., Rice K. C., de
Costa B. R., Fulham M. J., Di Chiro G., Neuropharmacol., 30, 899—
905 (1991).
13) Chen N., Luo T., Raymond L. A., J. Neurosci., 19, 6844—6854 (1999)
and references cited therein.
[
123I]-iodoamphetamine.17) Further experiments are necessary
14) Wüllner U., Young A. B., Penney J. B., Beal M. F., J. Neurochem., 63,
1772—1781 (1994).
to characterize the in vivo behavior of NPS 1506.
15) Calculated value using Advanced Chemistry Development (ACD)
Software Solaris V4.67, which is shown in STN databases.
16) Danysz W., Parsons C. G., Pharmacol. Rev., 50, 597—664 (1998).
17) Winchell H. S., Horst W. D., Braun L., Oldendorf W. H., Hattner R.,
Parker H., J. Nucl. Med., 21, 947—952 (1980).
Acknowledgments We thank the cyclotron crew of the
National Institute of Radiological Sciences for their help in
the 11C production of this works. This work has been sup-