Simplified heterocyclic analogues of fluoxetine inhibit inducible nitric oxide production in lipopolysaccharide-induced BV2 cells

Article abstract of DOI:10.1248/bpb.34.538

A series of fluoxetine, where the N-methylamino group was replaced and then simplified, were synthesized and their inhibitory effect was tested for nitric oxide (NO) production and inducible NO synthase (iNOS) expression in lipopolysaccharide (LPS)-induced BV2 cells. Although the synthesized compounds generally revealed weaker activity or greater cytotoxicity than fluoxetine, compound 10a, in which the N-methylamino group in fluoxetine was replaced by morpholine, and the trifluoromethylphenyl ring was substituted with simple oxo group, suppressed NO production dose-dependently at 10, 20 and 40 μM concentrations with less cytotoxicity than fluoxetine, and inhibited iNOS mRNA and protein expression at the same concentrations in LPS-induced BV2 cells. The results suggested that the trifluoromethylphenyl ring moiety in fluoxetine is not necessary for the suppression of NO production and that 10a has the potential as a potent inhibitor of NO production.

Full text of DOI:10.1248/bpb.34.538

538
Regular Article
Biol. Pharm. Bull. 34(4) 538—544 (2011)
Vol. 34, No. 4
Simplified Heterocyclic Analogues of Fluoxetine Inhibit Inducible Nitric
Oxide Production in Lipopolysaccharide-Induced BV2 Cells
Ju-Young PARK,a Seung-Woo KIM,b Ja-Kyeong LEE,b Weon Bin IM,a Byung Kwan JIN,c and
,a
*
Sung-Hwa YOON
a Department of Molecular Science and Technology, Ajou University; San 5, Woncheon, Yeongtong, Suwon 443–749,
Republic of Korea: b Department of Anatomy, Inha University; 3rd St. 7–241, Jung, Shinheung, Inchon 400–712, Republic
of Korea: and c Department of Biochemistry & Molecular Biology, Neurodegeneration Control Research Center, School of
Medicine Kyung Hee University; 1 Hoegi, Dongdaemoon, Seoul 130–701, Republic of Korea.
Received August 12, 2010; accepted January 13, 2011; published online January 20, 2011
A series of fluoxetine, where the N-methylamino group was replaced and then simplified, were synthesized
and their inhibitory effect was tested for nitric oxide (NO) production and inducible NO synthase (iNOS) expres-
sion in lipopolysaccharide (LPS)-induced BV2 cells. Although the synthesized compounds generally revealed
weaker activity or greater cytotoxicity than fluoxetine, compound 10a, in which the N-methylamino group in
fluoxetine was replaced by morpholine, and the trifluoromethylphenyl ring was substituted with simple oxo
mM concentrations with less cytotoxicity
group, suppressed NO production dose-dependently at 10, 20 and 40
than fluoxetine, and inhibited iNOS mRNA and protein expression at the same concentrations in LPS-induced
BV2 cells. The results suggested that the trifluoromethylphenyl ring moiety in fluoxetine is not necessary for the
suppression of NO production and that 10a has the potential as a potent inhibitor of NO production.
Key words fluoxetine analogue; microglia; BV2 cell; nitric oxide; inducible nitric oxide synthase
Microglial cells are a type of glial cell that acts as the first matory and neurodegenerative diseases.15—17)
and main form of active immune defense in the central ner-
Fluoxetine, the selective serotonin reuptake inhibitor that
vous system, and become activated by stimuli such as infec- is widely used in clinical practice for the treatment of depres-
tion and injury.1) The activation of microglia has been ob- sion, obsessive-compulsive disorder, bulimia and panic dis-
served in neurodegenerative disease such as Alzheimer’s dis- order,18,19) has been shown to exert antioxidant potential via
ease (AD), Parkinson’s disease, amyotrophic lateral sclerosis, reversal of oxidative damage by enhancing the in vivo antiox-
multiple sclerosis (MS) and human immunodeficiency virus- idant defenses and improving the cellular antioxidant status
associated dementia as the hallmark of brain inflamma- following a stress-induced decline.20) Fluoxetine has been
tion.2,3) Once activated, microglia release several inflamma- found to have beneficial effects in animal models of stroke,
tory mediators such as reactive oxygen intermediates, nitric MS, and epilepsy, even though fluoxetine has no positive ef-
oxide (NO), interleukin 1b (IL-1b), interferon g (IFN-g) and fect in clinical studies with Parkinson’s disease, AD and
tumor necrosis factor alpha (TNF-a), as well as neurotoxic Huntington’s disease.21) Recently, we reported that fluoxetine
substances.4,5) These mediators are thought to be accountable exerts an anti-inflammatory effect by inhibiting nuclear fac-
for brain diseases including ischemia, AD and neuronal tor-kB (NF-kB) activity in the postischemic brain and NO in
death.6) Therefore, regulating microglial activation or sup- LPS-treated primary microglia.22)
pressing these proinflammatory mediators could be a poten-
tial therapeutic approach to reducing the progression of these duction, a series of fluoxetine derivatives and analogues,
neurodegenerative disease.7)
where the N-methylamine group in fluoxetine was replaced
In our ongoing efforts to find a potent inhibitor of NO pro-
Although intrinsic molecules of microglia activation have by various heterocyclic amines, and then the trifluoro-
not yet been identified, experimental inducers of microglia, methylphenyl ring was simplified, were investigated in this
including lipopolysaccharide (LPS), amyloid-b and IFN-g, study for their inhibitory activities on NO production and
are known.8,9) LPS, an outer-membrane component of gram- iNOS expression on BV2 cells stimulated with LPS.
negative bacteria, is induced in inflammatory responses such
as activated macrophages and microglia, produces TNF-a, MATERIALS AND METHODS
IL-1b and NO, and is involved in immune responses and host
defense.10) Among these chemicals, NO produced by in-
Instruments Melting points were determined on a
ducible NO synthase (iNOS) can act as a cytotoxic agent to Fisher–Johns melting point apparatus and were left uncor-
1
the cells by reacting with superoxide anions to produce rected. H- and 13C-NMR spectra were recorded on a Varian
highly reactive peroxynitrite anions (ONOOꢀ) in pathophysi- Gemini 400 spectrometer at 400 MHz and 100 MHz, respec-
ologic state,11,12) leading to oxidative damage in host tissue, tively. The chemical shifts given are relative to tetramethylsi-
modification of DNA bases, and disruption of enzyme func- lane. Infrared spectra were recorded on a Nicolet 6700
tion and structural proteins.13,14) Therefore, the high level of Fourier transform (FT)-IR spectrometer. Electrospray ioniza-
NO produced by iNOS has been considered an indicator of tion (ESI)-MS spectra were obtained by Shimadzu LCMS-
cytotoxicity in inflammation, NO production is a critical step 2010EV (Japan). Column chromatography was carried out
in numerous neurodegenerative disease, and reduced NO using Merck silica gel 60 (230—400 mesh) (U.S.A.). All re-
production is a promising therapeutic strategy for the reduc- actions were performed under a nitrogen atmosphere.
tion of neuronal cell injury or death in various neuroinflam-
3-Chloro-1-phenylpropan-1-ol (3) To a solution of 3-
© 2011 Pharmaceutical Society of Japan
∗ To whom correspondence should be addressed. e-mail: shyoon@ajou.ac.kr

April 2011
539
chloro-1-propiophenone 2 (25.0 g, 148 mmol) in methanol 33.50, 54.30, 57.33, 74.99, 125.18, 126.44, 127.78, 144.66.
(300 ml) was added sodium borohydride (16.8 g, 445 mmol). 3-Morpholino-1-phenylpropan-1-ol (5d) Morpholine
The mixture was stirred for 24 h at room temperature. After (1.67 ml, 19.1 mmol) and 3-iodo-1-phenylpropan-1-ol 4
the reaction was complete, the solution was quenched with (1.00 g, 3.82 mmol) were subjected to the same reaction de-
sodium dihydrogen phosphate (66.7 g, 556 mmol). The sol- scribed for the synthesis of the compound 5a to give the title
vent was removed by rotary evaporator and the residue was compound as a pale yellow solid (450 mg, 53%). mp: 60 °C.
1
extracted with hexane and the organic layer was washed with IR (KBr) cmꢀ1: 3162. H-NMR (CDCl3) d: 1.85 (2H, m),
water, dried over Na2SO4 and concentrated in vacuo to give 2.47 (2H, s), 2.60 (4H, m), 3.72 (4H, t, Jꢁ4.8 Hz), 4.90 (1H,
the title compound as a colorless oil (24 g, 95%). IR (KBr) t, Jꢁ5.6 Hz), 7.22 (1H, m), 7.32 (4H, m). 13C-NMR (CDCl3)
1
cmꢀ1: 3235. H-NMR (CDCl3) d: 2.00 (1H, m), 2.10 (1H, d: 33.41, 53.52, 57.33, 66.77, 75.19, 125.22, 126.71, 127.96,
m), 2.80 (1H, s), 3.40 (1H, m), 3.60 (1H, m), 7.30 (5H, m). 144.41. MS m/z: 222.00 ([MꢂH]ꢂ).
13C-NMR (CDCl3) d: 41.28, 41.64, 71.04, 125.55, 127.58,
128.34, 143.32.
3-(4-Methylpiperazin-1-yl)-1-phenylpropan-1-ol
1-Methylpiperazine (2.12 ml, 19.1 mmol) and 3-iodo-1-
(5e)
3-Iodo-1-phenylpropan-1-ol (4) To a solution of 3- phenylpropan-1-ol 4 (1.00 g, 3.82 mmol) were subjected to
chloro-1-phenyl-1-propanol 3 (10.0 g, 58.6 mmol) in acetone the same reaction described for the synthesis of the com-
(250 ml) was added sodium iodide (14.9 g, 99.6 mmol). The pound 5a to give the title compound as a white solid
mixture was heated to reflux temperature for 48 h. After the (450 mg, 50%). mp: 71 °C. IR (KBr) cmꢀ1: 3114, 2948,
1
solution was cooled to room temperature, the solvent was 2806. H-NMR (CDCl3) d: 1.71 (2H, m), 2.11 (3H, s), 2.33
evaporated in vacuo. The residue was extracted with diethyl (6H, m), 2.48 (4H, m), 4.70 (1H, t, Jꢁ5.6 Hz), 7.08 (1H, m),
ether, the organic layer was washed with brine, dried over 7.20 (4H, m). 13C-NMR (CDCl3) d: 33.79, 45.27, 52.30,
Na2SO4 and evaporated in vacuo to provide the title com- 54.35, 55.89, 73.91, 124.88, 126.18, 127.43, 144.40.
pound as a yellow solid (15 g, 97%). mp: 33 °C. IR (KBr)
tert-Butyl 4-(3-Hydroxy-3-phenylpropyl)piperazine-1-
cmꢀ1: 3251. 1H-NMR (CDCl3) d: 2.10 (2H, m), 2.54 (1H, s), carboxylate (5f) Piperazine (5.92 g, 68.7 mmol) and 3-
3.11 (1H, m), 3.23 (1H, m), 4.71 (1H, m), 7.29 (5H, m). 13C- iodo-1-phenylpropan-1-one 4 (3.00 g, 11.4 mmol) were sub-
NMR (CDCl3) d: 2.97, 42.15, 73.93, 125.61, 127.68, 128.40, jected to the same reaction described for the synthesis of the
143.09.
compound 5a. To a solution of crude product in THF (10 ml)
1-Phenyl-3-(pyrrolidin-1-yl)propan-1-ol (5a) To a so- was added sodium bicarbonate (3.81 g, 45.4 mmol) and di-
lution of pyrrolidine (1.59 ml, 19.1 mmol) in tetrahydrofuran tert-butyl dicarbonate (2.38 g, 10.9 mmol) in H2O/THF (1/1,
(THF) (7.5 ml) was added 3-iodo-1-phenylpropan-1-ol 4 60 ml). The solution was stirred for 1 h at room temperature.
(1.00 g, 3.82 mmol) in THF (2 ml) dropwise. The resulting The mixture was extracted with ethyl acetate, washed with
solution was heated to reflux temperature for 4 h. After being brine, dried over Na2SO4, and concentrated in vacuo. The
cooled to room temperature, the mixture was extracted with crude product was purified by silica gel column chromatog-
dichloromethane and the organic layer was washed with raphy (EA/HX, 5/1) to give the title compound as a colorless
brine, dried over Na2SO4, and concentrated in vacuo. The oil (1 g, 27%). IR (KBr) cmꢀ1: 3120, 1686. 1H-NMR
crude product was purified by silica gel column chromatog- (CDCl3) d: 1.46 (9H, s), 1.86 (2H, m), 2.43 (2H, m), 2.58
raphy (dichloromethane/methanol, 15/1) to provide the title (4H, m), 2.66 (1H, m), 3.47 (4H, t, Jꢁ4.8 Hz), 4.92 (1H, t,
compound as a pale yellow solid (450 mg, 57%). mp: 72 °C. Jꢁ6.0 Hz), 7.24 (1H, m), 7.33 (4H, m). 13C-NMR (CDCl3)
1
IR (KBr) cmꢀ1: 3111. H-NMR (CDCl3) d: 1.79 (4H, m), d: 28.47, 33.74, 53.06, 57.12, 75.41, 79.79, 125.33, 126.83,
1.87 (2H, m), 2.53 (2H, m), 2.62 (3H, m), 2.89 (1H, m), 4.95 128.09, 144.46, 154.37.
(1H, m), 7.21 (1H, m), 7.34 (4H, m). 13C-NMR (CDCl3) d:
1-(3-Phenyl-3-(4-(trifluoromethyl)phenoxy)propyl)-
23.48, 35.74, 54.03, 54.78, 75.73, 125.36, 126.60, 127.93, pyrrolidine (6a) To solution of NaH (250 mg,
144.94. 10.0 mmol) in N,N-dimethylethanamide (DMA) (10 ml) was
a
3-(1H-Imidazol-1-yl)-1-phenylpropan-1-ol (5b) Imida- added 1-phenyl-3-(pyrrolidin-1-yl)propan-1-ol 5a (410 mg,
zole (1.30 g, 19.1 mmol) and 3-iodo-1-phenylpropan-1-ol 4 2.00 mmol). The mixture was stirred at 70 °C for 30 min. p-
(1.00 g, 3.82 mmol) were subjected to the same reaction de- Chlorobenzotrifluoride (1.34 ml, 10.0 mmol) was added and
scribed for the synthesis of the compound 5a to give the title the mixture was heated to 100 °C for 2 h. After being cooled
compound as a pale yellow solid (350 mg, 45%). mp: 102 °C. to room temperature, the mixture was quenched with 1 N
1
IR (KBr) cmꢀ1: 3169. H-NMR (CDCl3) d: 1.98 (2H, m), NaOH solution (10 ml) and extracted with toluene. The or-
3.83 (1H, m), 4.02 (1H, m), 4.37 (1H, m), 6.73 (2H, d, ganic layer was washed with brine, dried over Na2SO4 and
Jꢁ7.2 Hz), 7.17 (6H, m). 13C-NMR (CDCl3) d: 39.96, 43.65, concentrated in vacuo. The crude product was purified by
69.54, 118.61, 125.45, 127.12, 128.13, 128.39, 136.83, silica gel column chromatography (EA/HX, 1/1) to give the
144.40.
title compound as a pale yellow oil (200 mg, 29%). 1H-NMR
1-Phenyl-3-(piperidin-1-yl)propan-1-ol (5c) Piperidine (CDCl3) d: 1.43 (2H, m), 1.60 (4H, m), 2.05 (1H, m), 2.22
(1.89 ml, 19.2 mmol) and 3-iodo-1-phenylpropan-1-ol 4 (1H, m), 2.43 (4H, s), 5.25 (1H, q, Jꢁ4.8 Hz), 6.88 (2H, d,
(1.00 g, 3.82 mmol) were subjected to the same reaction de- Jꢁ8.8 Hz), 7.22 (1H, m), 7.30 (4H, m), 7.38 (2H, d,
scribed for the synthesis of the compound 5a to give the title Jꢁ8.8 Hz). 13C-NMR (CDCl3) d: 23.52, 34.43, 52.43, 35.66,
compound as a white solid (480 mg, 57%). mp: 59 °C. IR 54.40, 115.60, 122.64 (JC—Fꢁ31.9 Hz), 125.55, 126.65,
1
(KBr) cmꢀ1: 3108. H-NMR (CDCl3) d: 1.37 (2H, s), 1.55 126.68, 128.27, 128.87, 138.66, 159.36. MS m/z: 350.15
(4H, m), 1.75 (2H, m), 2.34 (2H, s), 2.45 (2H, m), 2.55 (2H, ([MꢂH]ꢂ).
m), 4.80 (1H, t, Jꢁ5.2 Hz), 6.84 (1H, s), 7.11 (1H, t,
1-(3-Phenyl-3-(4-(trifluoromethyl)phenoxy)propyl)-1H-
Jꢁ6.4 Hz), 7.23 (4H, m). 13C-NMR (CDCl3) d: 23.95, 25.70, imidazole (6b) Sodium hydride (178 mg, 7.42 mmol), 3-

540
Vol. 34, No. 4
(1H-imidazol-1-yl)-1-phenylpropan-1-ol 5b (300 mg, 1.48 rated in vacuo, the residue was dissolved in ethyl acetate and
mmol) and p-chlorobenzotrifluoride (1.00 ml, 7.42 mmol) then basified with 1 N NaOH solution. The organic layer was
were subjected to the same reaction described for the synthe- washed with brine, dried over Na2SO4 and concentrated in
sis of the compound 6a to give the title compound as a pale vacuo to give the title compound as a yellow oil (500 mg,
1
1
yellow oil (140 mg, 27%). H-NMR (CDCl3) d: 2.28 (1H, 44%). H-NMR (CDCl3) d: 1.84 (1H, s), 2.00 (1H, m), 2.20
m), 2.45 (1H, m), 4.10 (1H, m), 4.24 (1H, m), 5.00 (1H, q, (1H, m), 2.45 (6H, m), 2.90 (4H, m), 5.27 (1H, q, Jꢁ5.2 Hz),
Jꢁ3.6 Hz), 6.85 (2H, t, Jꢁ8.8 Hz), 7.06 (1H, s), 7.26 (2H, 6.89 (2H, d, Jꢁ8.8 Hz), 7.23 (1H, m), 7.32 (4H, m), 7.41
m), 7.32 (2H, m), 7.43 (2H, m), 7.50 (1H, d, Jꢁ8.4 Hz), 7.75 (2H, d, Jꢁ8.4 Hz). 13C-NMR (CDCl3) d: 35.75, 46.05, 54.46,
(1H, d, Jꢁ8.8 Hz), 7.90 (1H, s). 13C-NMR (CDCl3) d: 39.82, 54.98, 78.48, 115.49, 122.18 (JC—Fꢁ31.8 Hz), 122.48,
43.18, 76.37, 115.36, 119.20, 122.56 (JC—Fꢁ23.7 Hz), 126.42, 126.46, 127.53, 128.45, 140.77, 160.30. MS m/z:
125.33, 126.59, 126.62, 128.00, 128.20, 128.72, 137.30, 364.90 ([MꢂH]ꢂ).
139.28, 159.49. MS m/z: 347.15 ([MꢂH]ꢂ).
3-Iodo-1-phenylpropan-1-one (9a) To a solution of 3-
1-(3-Phenyl-3-(4-(trifluoromethyl)phenoxy)propyl)- chloro-1-propiophenone 2 (5.00 g, 29.7 mmol) in acetone
piperidine (6c) Sodium hydride (246 mg, 10.3 mmol), 1- (100 ml) was added sodium iodide (7.56 g, 50.4 mmol). The
phenyl-3-(piperidin-1-yl)propan-1-ol 5c (450 mg, 2.05 mmol) mixture was heated to reflux temperature for 4 h. After the
and p-chlorobenzotrifluoride (1.38 ml, 10.3 mmol) were solution was cooled to room temperature, the solvent was
subjected to the same reaction described for the synthesis of evaporated in vacuo. The residue was extracted with diethyl
the compound 6a to give the title compound as a pale red oil ether, the organic layer was washed with brine, dried over
1
(150 mg, 20%). H-NMR (CDCl3) d: 1.25 (2H, s), 1.41 (4H, Na2SO4 and evaporated in vacuo to provide the title com-
m), 1.85 (1H, m), 2.07 (1H, m), 2.20 (4H, s), 2.28 (2H, m), pound as a yellow solid (4.14 g, 54%). mp: 56 °C. IR (KBr)
1
5.13 (1H, t, Jꢁ8.0 Hz), 6.74 (2H, d, Jꢁ8.4 Hz), 7.03 (1H, cmꢀ1: 1675. H-NMR (CDCl3) d: 3.32 (2H, t, Jꢁ6.4 Hz),
m), 7.11 (2H, t, Jꢁ8.0 Hz), 7.20 (4H, m). 13C-NMR (CDCl3) 3.48 (2H, t, Jꢁ7.6 Hz), 7.34 (2H, m), 7.44 (1H, m), 7.79
d: 24.29, 25.88, 35.86, 54.35, 54.92, 78.59, 115.44, 122.06 (2H, m). 13C-NMR (CDCl3) d: 3.58, 42.31, 127.62, 128.36,
(JC—Fꢁ32.6 Hz), 125.51, 126.21, 126.28, 127.33, 128.27, 133.13, 135.63, 196.90.
140.83, 160.36. MS m/z: 363.95 ([MꢂH]ꢂ), 379.95
2-Bromo-1-phenylethanone (9b) To a solution of tri-
phenylphosphine (2.89 g, 11.1 mmol) and carbon tetrabro-
([MꢂNH4]ꢂ).
4-(3-Phenyl-3-(4-(trifluoromethyl)phenoxy)propyl)mor- mide (3.69 g, 11.1 mmol) in acetonitrile (10 ml) and dichlo-
pholine (6d) Sodium hydride (217 mg, 9.04 mmol), 3-mor- romethane (15 ml) at 0 °C was added 2-hydroxyaceto-
pholino-1-phenylpropan-1-ol 5d (400 mg, 1.81 mmol) and p- phenone 7 (500 mg, 3.67 mmol) in acetonitrile (5 ml). After
chlorobenzotrifluoride (1.22 ml, 9.04 mmol) were subjected being stirred for 15 min, the mixture was filtered off and the
to the same reaction described for the synthesis of the com- residue was concentrated in vacuo. The crude product was
pound 6a to give the title compound as a pale yellow oil purified by silica column chromatography (EA/HX, 1/3) to
(150 mg, 23%). 1H-NMR (CDCl3) d: 1.90 (1H, m), 2.13 (1H, give the title compound as a yellow oil (200 mg, 73%). IR
1
m), 2.36 (6H, m), 3.59 (4H, t, Jꢁ4.4 Hz), 5.18 (1H, q, (KBr) cmꢀ1: 1680. H-NMR (CDCl3) d: 4.37 (2H, s), 7.39
Jꢁ5.2 Hz), 6.78 (2H, d, Jꢁ8.4 Hz), 7.12 (1H, m), 7.20 (4H, (2H, t, Jꢁ7.6 Hz), 7.51 (1H, t, Jꢁ7.6 Hz), 7.87 (2H, m). 13C-
m), 7.34 (2H, d, Jꢁ8.8 Hz). 13C-NMR (CDCl3) d: 35.66, NMR (CDCl3) d: 31.06, 128.68, 128.72, 133.76, 190.93.
53.70, 54.84, 66.94, 78.44, 115.58, 122.38 (JC—Fꢁ32.6 Hz),
4-Iodo-1-phenylbutan-1-one (9c) To a solution of 4-
125.67, 126.52, 126.56, 127.68, 128.57, 140.76, 160.36. MS chloro-1-phenylbutan-1-one 8 (1.00 g, 5.47 mmol) in acetone
m/z: 365.95 ([MꢂH]ꢂ).
(30 ml) was added sodium iodide (1.22 g, 8.21 mmol). The
1-Methyl-4-(3-phenyl-3-(4-(trifluoromethyl)phenoxy)- mixture was heated to reflux temperature for 18 h. After the
propyl)piperazine (6e) Sodium hydride (205 mg, 8.53 solution was cooled to room temperature, the solvent was
mmol), 3-(4-methylpiperazin-1-yl)-1-phenylpropan-1-ol 5e evaporated in vacuo. The residue was extracted with diethyl
(400 mg, 1.71 mmol) and p-chlorobenzotrifluoride (1.15 ml, ether, the organic layer was washed with brine, dried over
8.53 mmol) were subjected to the same reaction described for Na2SO4 and evaporated in vacuo to provide the title com-
the synthesis of the compound 6a to give the title compound pound as a yellow solid (800 mg, 80%). IR (KBr) cmꢀ1:
1
1
as a pale yellow oil (100 mg, 15%). H-NMR (CDCl3) d: 1693. H-NMR (CDCl3) d: 2.19 (2H, q, Jꢁ6.4 Hz), 3.06
1.85 (1H, m), 2.08 (1H, m), 2.16 (3H, s), 2.32 (10H, m), 5.15 (2H, t, Jꢁ6.8 Hz), 3.25 (2H, t, Jꢁ6.8 Hz), 7.38 (2H, m), 7.48
(1H, m), 6.77 (2H, d, Jꢁ8.8 Hz), 7.09 (1H, m), 7.18 (4H, m), (1H, m), 7.88 (2H, m). 13C-NMR (CDCl3) d: 6.97, 27.61,
7.27 (2H, d, Jꢁ8.4 Hz). 13C-NMR (CDCl3) d: 35.88, 45.85, 38.99, 127.87, 128.51, 133.08, 136.53, 198.29.
52.98, 54.17, 55.00, 78.40, 115.44, 125.42, 122.13 (JC—
Fꢁ32.6 Hz), 126.33, 126.36, 127.45, 128.36, 140.70, 160.27. line (8.40 ml, 96.0 mmol) and 3-iodo-1-phenylpropan-1-one
MS m/z: 379.20 ([MꢂH]ꢂ).
9a (500 mg, 1.92 mmol) were subjected to the same reaction
3-Morpholino-1-phenylpropan-1-one (10a) Morpho-
1-(3-Phenyl-3-(4-(trifluoromethyl)phenoxy)propyl)- described for the synthesis of the compound 5a to give the
piperazine (6f) Sodium hydride (374 mg, 15.6 mmol), tert- title compound as a yellow oil (350 mg, 83%). The title com-
butyl 4-(3-hydroxy-3-phenylpropyl)piperazine-1-carboxylate pound was previously synthesized from dimethyaminopro-
5f (1.00 g, 3.12 mmol) and p-chlorobenzotrifluoride (2.10 ml, piophenone as a hydrochloride salt (mp23)ꢁ175—177 °C). IR
1
15.6 mmol) were subjected to the same reaction described for (KBr) cmꢀ1: 1685. H-NMR (CDCl3) d: 2.43 (4H, t, Jꢁ
the synthesis of the compound 6a to give crude yellow oil. 4.4 Hz), 2.75 (2H, t, Jꢁ7.2 Hz), 3.10 (2H, t, Jꢁ7.6 Hz), 3.63
The crude oil was dissolved in 1,4-dioxane (5 ml) and 4 N (4H, t, Jꢁ4.8 Hz), 7.37 (2H, t, Jꢁ8.0 Hz), 7.47 (1H, t,
HCl in 1,4-dioxane (30 ml) added. This solution was stirred Jꢁ7.2 Hz), 7.86 (2H, m). 13C-NMR (CDCl3) d: 35.98, 53.51,
for 3 h at room temperature. After the solution was evapo- 53.66, 66.84, 127.81, 128.44, 132.95, 136.63, 198.57. MS

April 2011
541
m/z: 220.00 ([MꢂH]ꢂ).
2-Morpholino-1-phenylethanone (10b) Morpholine
mycin (Gibco, U.S.A.).24)
Determination of NO Release BV2 cells (2ꢃ105) were
(439 ml, 5.00 mmol) and 2-bromo-1-phenylethanone 9b (200 treated with LPS (100 ng/ml) for 24 h. To measure the
mg, 1.00 mmol) were subjected to the same reaction de- amount of NO produced by the BV2 cells, 100 ml of condi-
scribed for the synthesis of the compound 5a to give the title tioned medium was mixed with an equal volume of Griess
compound as a yellow oil (500 mg, 25%). IR (KBr) cmꢀ1: reagent (0.5% sulfanilamide and 0.05% N-1-naphthylethyl-
1680. 1H-NMR (CDCl3) d: 2.62 (4H, m), 3.78 (4H, m), 3.83 enediamine), and incubated for 10 min at room temperature.
(2H, s), 7.45 (2H, m), 7.56 (1H, m), 7.97 (2H, m).13C-NMR The absorbance of the mixture at 550 nm was measured
(CDCl3) d: 53.88, 64.64, 66.78, 127.90, 128.44, 133.19. MS using a microplate reader.24)
m/z: 206.00 ([MꢂH]ꢂ).
Cell Viability Cell respiration, an indicator of cell via-
4-Morpholino-1-phenylbutan-1-one (10c) Morpholine bility, was determined by the mitochondrial-dependent re-
(1.12 ml, 12.8 mmol) and 4-iodo-1-phenylbutan-1-one 9c duction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-
(0.70 g, 2.55 mmol) were subjected to the same reaction de- zoium (MTT; Sigma-Aldrich Corp., U.S.A.) to formazan.
scribed for the synthesis of the compound 5a to give the title MTT solution (50 ml, 0.5 mg/ml) was added to BV2 cell cul-
compound as a yellow oil (450 mg, 40%). IR (KBr) cmꢀ1: tures. The conversion of MTT to formazan by metabolically
1
1686. H-NMR (CDCl3) d: 1.94 (2H, q, Jꢁ6.8 Hz), 2.40 visible cells was measured with a microplate reader at
(6H, m), 3.00 (2H, t, Jꢁ7.2 Hz), 3.63 (4H, t, Jꢁ4.4 Hz), 7.43 595 nm.15)
(2H, m), 7.53 (1H, m), 7.94 (2H, m). 13C-NMR (CDCl3) d:
Reverse Transcription-Polymerase Chain Reaction
21.21, 36.02, 53.46, 57.99, 66.73, 127.73, 128.22, 132.56, Total RNA was prepared with TriZol reagent (Gibco BRL,
136.83, 199.43. MS m/z: 234.00 ([MꢂH]ꢂ).
U.S.A.), and 1-mg RNA samples were used for cDNA syn-
3-Morpholino-1-phenylpropan-1-one oxime (11a) To a thesis using RT-PCR kit (Roche, Germany). The primer se-
solution of 3-morpholino-1-phenylpropan-1-one 10a (350 quences of rat inducible NO synthase (iNOS) was described
mg, 1.60 mmol) in ethanol (10 ml) was added hydroxylamine previously.25)
hydrochloride (155 mg, 2.23 mmol) and pyridine (180 ml,
Immunoblotting BV2 cells were washed twice with
2.23 mmol). The solution was heated to reflux temperature cold phosphate buffered saline (PBS) and lysed in RIPA
for 1 h. After being cooled to room temperature, the mixture buffer (50 mM Tris–HCl, pH 7.4, 1% NP-40, 0.25% sodium-
was extracted with ethyl acetate. The organic layer was deoxycholate, 150 mM NaCl, 1 mM Na3VO4, and 1 complete
washed with brine, dried over Na2SO4 and concentrated in Mini protease inhibitor cocktail tablet (Roche Diagnostics,
vacuo. The residue was purified by silica gel column chro- Switzerland)). The lysate was centrifuged for 15 min at
matography (dichloromethane/methanol, 20/1) to give the 12000 rpm at 4 °C and supernatants were loaded onto 6%
title compound as a white powder (30 mg, 10%). mp: 143 °C. sodium dodecyl sulfate-polyacrylamide gel electrophoresis
1
IR (KBr) cmꢀ1: 2972, 2845. H-NMR d: 2.57 (4H, s), 2.63 (SDS-PAGE). Primary antibodies were diluted as follows:
(2H, t, Jꢁ8.8 Hz), 3.07 (2H, t, Jꢁ6.8 Hz), 3.75 (4H, m), 7.30 1 : 1000 for anti-iNOS (BD Bioscience, U.S.A.). The antigen-
(3H, m), 7.61 (2H, t, Jꢁ4.4 Hz). 13C-NMR d: 23.94, 53.40, antibody complex was detected by BM Chemiluminescence
54.88, 66.72, 126.02, 128.40, 128.97, 135.60, 157.04. MS Blotting Substrate (Roche, Germany) using donkey anti-
m/z: 235.00 ([MꢂH]ꢂ).
rabbit horseradish peroxidase-conjugated secondary antibody
3-Morpholino-1-phenylpropan-1-one O-ethyl oxime (Santa Cruz Biotechnology, U.S.A.).25)
(11b) 3-Morpholino-1-phenylpropan-1-one 10a (500 mg,
2.28 mmol), O-ethylhydroxylamine hydrochloride (267 mg, RESULTS AND DISCUSSION
2.74 mmol) and pyridine (221 ml, 2.74 mmol) were subjected
to the same reaction described for the synthesis of the com-
Chemistry of Fluoxetine Analogues The syntheses of
pound 11a to give the title compound as a colorless oil the heterocyclic analogues of fluoxetine started from com-
1
(50 mg, 8%). H-NMR (CDCl3) d: 1.31 (3H, t, Jꢁ7.2 Hz), mercially available 3-chloropropiophenone 2, as depicted in
2.50 (6H, m), 2.96 (2H, m), 3.70 (4H, t, Jꢁ4.8 Hz), 4.22 Chart 1. Since racemic fluoxetine is used therapeutically, all
(2H, q, Jꢁ7.2 Hz), 7.34 (3H, m), 7.62 (2H, m). 13C-NMR compounds were synthesized as a racemic mixture. Chloro-
(CDCl3) d: 21.21, 36.02, 53.46, 57.99, 66.73, 127.73, propanone 2 was first reduced to chloropropanol 3 using 3 eq
128.22, 132.56, 136.83, 199.43. MS m/z: 263.00 ([MꢂH]ꢂ).
of NaBH4 in methanol at room temperature for 24 h in excel-
3-Morpholino-1-phenylpropan-1-one O-benzyl oxime lent yield (95%). The chloropropanol 3 was converted to
(11c) 3-Morpholino-1-phenylpropan-1-one 10a (500 mg, iodopropanol 4 using NaI with acetone via a Finkelstein re-
2.28 mmol), O-benzylhydroxylamine hydrochloride (437 mg, action in 97% yield and the completion of the reaction was
2.74 mmol) and pyridine (221 ml, 2.74 mmol) were subjected judged by the 1H-NMR spectrum.26) Reactions of iodo-
to the same reaction described for the synthesis of 11a to propanol 4 with various heterocyclic amines produced the
give the title compound as a colorless oil (100 mg, 14%). 1H- crude products of corresponding amine-propanols 5a—f con-
NMR (CDCl3) d: 2.42 (4H, m), 2.52 (2H, m), 2.97 (2H, m), taining excess amines, which were purified by chromatogra-
3.63 (4H, t, Jꢁ4.4 Hz), 5.21 (2H, s), 7.27 (1H, m), 7.32 (5H, phy on silica gel with dichloromethane/methanol (v/v, 15/1)
m), 7.38 (2H, m) 7.60 (2H, m). 13C-NMR (CDCl3) d: 24.66, to give pure amine-propanols (5a—f). Finally, the hetero-
53.35, 54.93, 66.82, 126.08, 127.53, 127.91, 128.08, 128.20, cyclic analogues 6a—f were prepared by nucleophilic aro-
128.86, 125.41, 137.69, 156.77. MS m/z: 324.95 ([MꢂH]ꢂ).
matic substitution of p-chlorobenzotrifluoride with the corre-
BV2 Cell Culture BV2 cells were grown in Dulbecco’s sponding alkoxide anions of compounds (5a—f), which were
modified Eagle’s medium (DMEM; Gibco, U.S.A.) contain- generated by reacting with sodium hydride in DMA.
ing 5% FBS (Hyclone, U.S.A.) and 1% penicillin–strepto-
The simplified analogues of fluoxetine, where the trifluo-

542
Vol. 34, No. 4
Reagents and conditions: (a) NaBH4, MeOH, r.t., 24 h; (b) NaI, acetone, reflux, 48 h; (c) R1H, THF, reflux, 4 h; and (d) NaH, DMA, p-chlorobenzotrifluoride, 70 to
100 °C, 2 h.
Chart 1
Reagents and conditions: (a) NaI, acetone, reflux, 4 h; (b) CBr4, PPh3, ACN/CH2Cl2 (1/1), 0 °C, 15 min; (c) NaI, acetone, reflux, 18 h; (d) morpholine, THF, reflux, 4 h; (e)
R1 hydroxylamine hydrochloride, EtOH, pyridine, reflux, 1 h.
Chart 2
romethyl phenyl ring was removed, were synthesized as de-
picted in Chart 2. Chloropropanone 2 was converted to
iodopropanone 9a using NaI in acetone, which was then re-
acted with excess morpholine at reflux temperature for 4 h to
give the crude residue containing product as well as excess
morpholine. The desired morpholino-propanone 10a was ob-
tained by chromatography on silica gel with dichloromethane/
methanol (v/v, 15/1). The various oxime analogues (11a—c)
were prepared by condensation reaction of 10a with corre-
sponding hydroxylamine hydrochlorides.
Table 1. Effects of Fluoxetine Heterocyclic Analogues on LPS-Induced
NO Production
Entry
Substituent (R1)
NHCH3
Inhibition of NO release (%) at 10 mM
1
20.8ꢄ0.9
To investigate the influence of the alkyl chain length in
10a on the activity, two different chain-length analogues
(10b, c) were synthesized as depicted in Chart 2, respectively.
For the synthesis of 10b, the alcohol 7 was first converted
into the bromide 9b using triphenylphosphine (PPh3) and
tetrabromide (CBr4) via an Appel reaction, and then reacted
with excess morpholine to produce the crude product 10b,
which was purified by silica gel column chromatography to
give pure 10b. Similarly, the crude product 10c was prepared
6a
6b
6c
6d
6e
6f
0
0
0
6.00ꢄ0.9
Cytotoxic (64.3ꢄ6.7)
Cytotoxic (79.6ꢄ2.3)
from the conversion of 4-chlorobutanone 8 to its correspond- 79.6ꢄ2.3%, respectively) comparable to that of fluoxetine
ing iodide 9c using NaI in acetone followed by the reaction (20.8ꢄ0.9%), while the other heterocyclic analogues exhib-
of 9c with excess morpholine. Pure product 10c was obtained ited no activity at a concentration of 10 mM in NO produc-
by column chromatography on silica gel.
Biological Effect of Fluoxetine Analogues Since LPS
tion.
To determine whether the inhibitory effects of fluoxetine
considerably induced NO production in BV2 cells and the and its analogues on NO production were related to their
level of NO production was detected by nitrite, in this study, toxicity, the cytotoxicity of fluoxetine analogues was evaluat-
fluoxetine and its analogues (6a—f) were first evaluated for ed by MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetra-
inhibitory effects on the production of NO in LPS-induced zolium bromide) assay.27) As shown in Fig. 1, the relative cell
BV2 cells. After BV2 cells were co-treated with fluoxetine or viability of the BV2 cells treated with fluoxetine at the con-
its analogues at a concentration of 10 mM and LPS (100 ng/ centrations of 10 mM, 20 mM and 40 mM was 52.3ꢄ3.4%,
ml) for 24 h, the nitrite concentration was determined in the 73.9ꢄ2.2% and 20.1ꢄ0.6%, respectively, which indicates
supernatant by the Griess assay, and their results are pre- that fluoxetine did not suppress the cell viability of the BV2
sented in Table 1. Among the fluoxetine analogues possess- cell at the concentrations of 10 mM or 20 mM, but did exhibit
ing various heterocyclic amine rings (6a—f), 6d, 6e, and 6f cytotoxicity at the concentration of 40 mM. This result was
showed inhibitory activities (6.00ꢄ0.9%, 64.3ꢄ6.7% and consistent with a previous report.22) The relative cell viability

April 2011
543
Table 2. Effects of Simplified Fluoxetine Analogues on LPS-Induced NO
Production
Entry Substituent (R1)
n
Inhibition of NO release (%) at 10 mM
1
Fluoxetine
20.8ꢄ0.9
42.0ꢄ1.1 (20 mM)
5d
10a
–OH
ꢁO
1
1
0
22.1ꢂ4.8
35.2ꢄ3.4 (20 mM)
83.7ꢄ1.4 (40 mM)
10b
10c
11a
11b
11c
ꢁO
ꢁO
ꢁN–OH
ꢁN–OEt
ꢁN–OBn
0
2
1
1
1
0
0
0
0
0
Fig. 1. Effect of Fluoxetine and Its Analogues on the Cell Viability of
BV2 Cells
Cells (2ꢃ105 cells) were incubated with LPS (1 mg/ml) and each sample (fluoxetine
and its analogues) for 24 h. Cell viability was measured by MTT reduction assays and
was expressed relative to the culture treated only with LPS (control; 48.8%). Each
value indicated the meanꢄS.E.M. and was representative of results obtained from three
independent experiments.
of the BV2 cells treated with fluoxetine analogues (6e, f) was
also decreased to 32.1ꢄ0.7% and 20.2ꢄ0.5%, respectively
(Fig. 1), which clearly indicated that the inhibitory effect on
NO production of the fluoxetine analogues (6e, f) was due to
their cytotoxicity rather than their intrinsic activities. In con-
trast, 6d containing the morpholine moiety did not show any
cytotoxicity even at the concentration of 40 mM (data not
shown), despite of its weak inhibitory activity in NO produc-
tion. Due to the absence of any cytotoxic effect, 6d was se-
lected as a lead compound for further modification study.
The structure of 6d was simplified by removing the trifluo-
romethyl phenyl ring. Various functional moieties such as
hydroxyl, ketone and oxime groups were introduced for re-
placing the ether linkage in fluoxetine while the morpholine
moiety was fixed. The simplified compounds were tested for
their inhibitory effects in NO production on BV2 cells. As
shown in Table 2, none of the tested compounds (5d, 11a—
c) exhibited any inhibitory effect in NO production on BV2
Fig. 2. Effects of Fluoxetine and 10a on LPS-Induced NO Production in
BV2 Cells
BV2 cells were stimulated with LPS (100 ng/ml), co-treated with LPS (100 ng/ml)
and fluoxetine at a concentrations of 10 and 20 mM or co-treated with LPS (100 ng/ml)
and 10a at a concentrations of 10, 20 and 40 mM for 24 h. The amounts of NO were de-
termined using the Griess reagent. All the data of each group are expressed as the
meanꢄS.E.M. and are representative of results obtained from three independent experi-
ments.
cells, except 10a containing a ketone group. 10a revealed al- tween the ketone group and the morpholine moiety in 10a on
most the same inhibitory activity as that of fluoxetine at the the activity, two analogues (10b, c) were also evaluated for
concentration of 10 mM. When we tested the inhibitory activi- the NO production activities in BV2 cells. Interestingly, these
ties at different concentrations, 10a reduced the LPS-induced two analogues did not exhibit any effect at a concentration of
nitrite production in a dose-dependent manner. Accordingly, 10 mM, which implied that the chain length between the oxo
the percentage of inhibited NO production in 10a was group and the heteroamine moiety plays an important role in
22.1ꢄ4.8, 35.2ꢄ3.4 and 83.7ꢄ1.4% at the concentrations of maintaining the activity.
10, 20 and 40 mM, respectively (Table 2).
NO is biosynthesized by NOS, including, for example, en-
To confirm that the suppressed effect of 10a on NO pro- dothelial NOS (eNOS), neuronal NOS (nNOS) and iNOS.
duction was not caused by cytotoxicity, the cell viability was The iNOS-derived NO is generated from various cells such
measured. LPS dramatically increased the concentration of as macrophages, monocytes and microglial cells and the
nitrite from the cells (47.1ꢄ2.3 mM) whereas a small amount overproduction of NO is caused by inflammation. Therefore,
of NO was produced (5.47ꢄ1.22 mM) in unstimulated BV2 the reduced NO production is due to the inhibited iNOS ex-
cells. The relative cell viability of the BV2 cells treated with pression. To examine whether the NO production was caused
10a at the concentrations of 10 mM, 20 mM and 40 mM was by a decrease in the mRNA level of iNOS, LPS-stimulated
60.2ꢄ1.2%, 61.1ꢄ1.9% and 68.8ꢄ1.6%, respectively (Fig. BV2 cells were treated with 10a and then the amount of
1), which indicated that 10a did not exhibit any cytotoxicity iNOS was measured by reverse transcription polymerase
even at the concentration of 40 mM, while fluoxetine at the chain reaction (RT-PCR). The expression of iNOS mRNA in
same concentration did. 10a also exhibited NO inhibitory ac- LPS-treated BV2 cells was significantly increased in compar-
tivity in a dose-dependent manner (Fig. 2).
ison with the non-treated control cells. When the LPS-treated
To investigate the influence of the alkyl chain length be- BV2 cells were treated with 10a at different concentrations

544
Vol. 34, No. 4
REFERENCES
1) Meng X. L., Yang J. Y., Chen G. L., Zhang L. J., Wang L. H., Li J.,
Wang J. M., Wu C. F., Int. Immunopharmacol., 8, 1074—1082 (2008).
2) Blaylock R. L., Journal of American Physicians and Surgeons, 9, 46—
51 (2004).
3) Takeuchi H., Mizuno T., Zhang G., Wang J., Kawanokuchi J., Kuno R.,
Suzumura A., J. Biol. Chem., 280, 10444—10454 (2005).
4) Gehrmann J., Matsumoto Y., Kreutzberg G. W., Brain Res. Rev., 20,
269—287 (1995).
5) Banati R. B., Gehrmann J., Schubert P., Kreutzberg G. W., Glia, 7,
111—118 (1993).
6) Moon D., Choi Y. H., Kim N., Park Y., Kim G., Int. Immunopharma-
col., 7, 506—514 (2007).
Fig. 3. Effects of Fluoxetine and 10a on iNOS Production in LPS-Induced
7) Jin C., Moon D., Lee K., Kim M., Lee J., Choi Y. H., Park Y., Kim G.,
Pharmacol. Res., 54, 461—467 (2006).
8) Jantzen P. T., Connor K. E., Dicarlo G., Wenk G. L., Wallace J. L., Ro-
jiani A. M., Coppola D., Morgan D., Gordon M. N., J. Neurosci., 22,
2246—2254 (2002).
BV2 Cells
BV2 cells were treated with LPS (100 ng/ml or 200 ng/ml) and in the absence or
presence of fluoxetine and 10a. The expression levels of iNOS mRNA in fluoxetine (A)
and 10a (B) were determined by RT-PCR. The production of iNOS protein level in
fluoxetine and 10a (C) was determined by immunoblotting with antibody iNOS.
9) Takeuchi H., Wang J., Kawanokuchi J., Mitsuma N., Mizuno T., Suzu-
mura A., Neurobiol. Dis., 22, 33—39 (2006).
10) Chen R., Chen T., Chen T., Lin L., Chang C., Chang H., Wu C., Ann.
N.Y. Acad. Sci., 1042, 262—271 (2005).
(10, 20, 40 mM), the cells were decreased to the density of
iNOS mRNA in a dose-dependent manner (Figs. 3A, B). To
confirm the effects of 10a on iNOS protein expression, the
11) Aktan F., Life Sci., 75, 639—653 (2004).
amount of iNOS protein expression was measured by the im- 12) Rafi M. M., Yadav P. N., Reyes M., J. Food Sci., 72, S69—S74 (2007).
13) Cui Y., Zhang L., Zhang T., Luo D., Jia Y., Guo Z., Zhang Q., Wang
X., Wang X., Clin. Exp. Pharmacol. Physiol., 37, 422—428 (2010).
14) Jung K. K., Lee H. S., Cho J. Y., Shin W. C., Rhee M. H., Kim T. G.,
munoblotting method. As shown in Fig. 3C, no iNOS protein
was detected in unstimulated BV2 cells, whereas iNOS pro-
tein was induced in LPS-stimulated BV2 cells. When the
Kang J. H., Kim S. H., Hong S., Kang S. Y., Life Sci., 79, 2022—2031
LPS-treated BV2 cells were treated with 10a at different con-
centrations (10, 20, 40 mM), the densities of iNOS protein in
the cells were decreased in a dose-dependent manner. These
results indicated that the inhibitory action of 10a on NO pro-
duction was related to a modulation of iNOS induction.
10a was shown to have weak activities against the EMT6
tumor in vitro and against ADP-induced aggregation of
human blood platelets in vitro,28,29) while there was no effect
in inhibition of edema in the test for anti-inflammatory activ-
ity in CF1 mice at 8 mg/kg as a series of b-alkylamino-
ketones.30) However, there is no previous report regarding
the suppression of NO production by 10a in LPS-induced
BV2 cells.
In conclusion, by modifying the structure of fluoxetine, we
discovered an analogue of fluoxetine 10a, which inhibited not
only NO production but also iNOS expression in LPS-
induced BV2 cells without much cytotoxicity. The trifluo-
romethyl phenyl moiety was not essential for the suppression
in NO production. The simplified structure of 10a can be
used for the further development of potent inhibitors of NO
production.
(2006).
15) Park Y., Koketsu M., Kim J. M., Yeo J., Ishihara H., Lee K., Kim S. Y.,
Kim C., Biol. Pharm. Bull., 26, 1657—1660 (2003).
16) Hou R. C., Chen H., Tzen J. T. C., Jeng K. G., Neuro Report, 14,
1815—1819 (2003).
17) Choi S., Park Y., Choi D., Chang H., J. Microbiol. Biotechnol., 18,
1990—1996 (2008).
18) Abdel-Salam O. M. E., Baiuomy A. R., Arbid M. S., Pharmacol. Res.,
49, 119—131 (2004).
19) Wilde M. I., Benfield P., Pharmacoeconomics, 13, 543—561 (1998).
20) Zafir A., Banu N., Eur. J. Pharmacol., 572, 23—31 (2007).
21) Mostert J. P., Koch M. W., Heerings M., Heersema D. J., Keyser J. D.,
CNS Neurosci. Ther., 14, 153—164 (2008).
22) Lim C., Kim S., Park J., Kim C., Yoon S., Lee J., J. Neurosci. Res., 87,
1037—1045 (2009).
23) Synder H. R., Brewster J. M., J. Am. Chem. Soc., 70, 4230—4232
(1948).
24) Yu Y., Kim J., Lee K., Kim S. Y., Han P., Lee J., Stroke, 36, 2238—
2243 (2005).
25) Kim S., Jeong J., Kim H. J., Seo J., Han P., Yoon S., Lee J., Neurotox.
Res., 17, 39—49 (2010).
26) Robertson D. W., Krushinski J. H., Fuller R. W., Leander J. D., J. Med.
Chem., 31, 1412—1417 (1988).
27) Fotakis G., Timbrell J. A., Toxicol. Lett., 10, 171—177 (2006).
28) Dimmock J. R., Erciyas E., Raghavan S. K., Kirkpatrick D. L., Phar-
mazie, 45, 755—757 (1990).
29) Cazin M., Lesieur I., Boniface B., Boniface M., Lesieur D., Cazin J.
C., Acta Therapeutica, 6, 205—211 (1980).
30) Huang Y., Hall I. H., Res. Commun. Pharmacol. Toxicol., 1, 17—38
(1996).
Acknowledgments This study was financially supported
by SH pharmaceutical company and the Post Brain Korea 21
program.