726
Vol. 33, No. 4
reaction, the mixture was filtered and the filtrate was washed
with brine (20 ml) and saturation NaHCO3 (20 ml). The
phases were separated, the organic phase dried over anhy-
drous MgSO4, filtered, and the solvent removed under vac-
uum. The residue was further purified by alumina column
chromatography to afford the product.
Reagents and reaction conditions: a: MgSO4, dry THF, rt, overnight; b: TiCl4, dry
THF, rt, 3 d; c: MgSO4, benzene, 80 °C, overnight, d: NaBH4, MeOH, 0 °C,
20 min.
1
Compound 2: Yꢀ87%, H-NMR (CDCl3) d: 7.75—7.72
(m, 4H, CH), 7.50—7.51 (t, 6H, Jꢀ2.5 Hz, CH), 4.11—4.12
(d, 2H, Jꢀ5 Hz, CH2), 2.22—2.21 (t, 1H, Jꢀ2.5 Hz, CH).
ESI-MS m/z: 220.2 (Mꢁ1)ꢁ, IR (KBr, cmꢂ1): nCꢀNꢀ1658.
Chart 1. Synthesis of Benzylidene-prop-2-ynyl-amines 1—23
Procedure c7) To a solution of aldehyde (2 mmol) in RESULTS AND DISCUSSION
benzene (8 ml) was added propargylamine (0.205 ml) and
MgSO4 (0.2 g). The reaction mixture was refluxed at 80 °C
Firstly, compounds 1—21 were synthesized from alde-
for 24 h. The progress of the reaction was monitored by TLC. hydes and propargylamine, and the experiments were sum-
After completion of the reaction, the reaction mixture was marized in Chart 1. Condensation of aldehydes (or ketone)
filtered and collected filtered. The solvent was distilled, water with propargylamine in the presence of dehydrating agents
(20 ml) was added, and the mixture was extracted with MgSO4 afforded benzylidene-prop-2-ynyl-amines in good
dichloromethane. The organic layer was separated, washed yields (70—99%). Then compound 3 and 21 were reduced to
with brine (20 ml) and saturation NaHCO3 (20 ml). The the corresponding secondary amines 22 and 23 with NaBH4.
phases were separated, the organic phase dried over anhy- All compounds were determined by MS and 1H-NMR.
drous MgSO4, filtered, and the solvent removed under vac-
uum. The residue was further purified by alumina column A/B, IC50 was determined with SPSS software using our re-
chromatography to afford the product.
ported fluorimetric method.11) All data including selectivity
To verify the ability of these compounds to inhibit MAO-
1
Compound 20: Yꢀ90%, H-NMR (CDCl3) d: 8.39—8.38 index (SI) toward MAO-A and B are presented in Table 1.
(t, 1H, Jꢀ2.5 Hz, CHꢀN), 6.93 (s, 2H, CH), 4.43—4.42 (t, Clorgyline and pargyline are commercial MAO inhibitors
2H, Jꢀ2.5 Hz, CH), 3.82 (s, 9H, CH3O), 1.16—1.15 (m, 1H, and their IC50 were included as references (24, 25 in
Jꢀ2.5 Hz, CH), ESI-MS m/z: 234.3 (Mꢁ1)ꢁ, IR (KBr, Table 1).12,13) As expected, nearly all compounds show good
cmꢂ1): nCꢀNꢀ1640.
inhibitory activity toward MAO-A/B. As can be seen from
Precedure d8) Benzylidene-prop-2-ynyl-amines (0.1 Table 1, benzhydrylidene-prop-2-ynyl-amine (2, IC50ꢀ
mmol) were dissolved in MeOH and cooled to 0 °C. The so- 32 nM) and (3,4-dimethoxy-benzylidene)-prop-2-ynyl-amine
lution was added NaBH4 (0.15 mmol). After 20 min, the sol- (10, IC50ꢀ14 nM) are the best inhibitors toward monoamine
vent was distilled. Water (20 ml) was added, and the mixture oxidase A and B respectively, while compound 1 (SIꢀ58.96)
was extracted with dichloromethane. The organic layer was is the most selective inhibitor toward MAO-A and compound
separated, washed with brine (20 ml) and saturation NaHCO3 2 (SIꢀ0.34) is the most selective inhibitor toward MAO-B.
(20 ml). The phases were separated, the organic phase dried
Structure–activity relationships (SARs) were examined
over anhydrous MgSO4, filtered, and the solvent removed based on IC50 and SI values (Table 1). Firstly, compound 16
under vacuum. The residue was further purified by alumina (nꢀ2) shows much better inhibitory activity than the others
column chromatography to afford the product.
(nꢀ0), and this phenomenon indicates that the distance be-
1
Compound 23: Yꢀ73%, H-NMR (CDCl3) d: 7.33—7.31 tween imine group and aromatic ring affects the anti-MAO
(d, 2H, Jꢀ10 Hz, CH), 7.30—7.28 (t, 2H, Jꢀ5 Hz, CH), activity. Additionally, most compounds with electron donat-
7.24—7.26 (d, 1H, Jꢀ10 Hz, CH), 3.88 (s, 2H, CH2), 3.40— ing group such as 4, 7—10, and 18, display better inhibitory
3.39 (d, 2H, Jꢀ5 Hz, CH2), 2.25—2.24 (t, 1H, Jꢀ2.5 Hz, activity than comounds 3, 5, 15, 17, 19 bearing electron
NH), 2.04 (s, 1H, CH). ESI-MS m/z: 146.3 (Mꢁ1)ꢁ, IR withdrawing groups. We also noticed that compounds 2 and
(KBr, cmꢂ1): nN–Hꢀ1495.
12 showed higher inhibitory activity toward MAO-B than all
Biochemistry Briefly, newly synthesized amines (ranged the other compounds, which implied imines with bulky
from 1—1000 mM), mitochondria (6 mg/ml) and probe (0.04 group were favorable inhibitors against MAO-B. In addition,
mM) were added to borate buffer (0.05 M, pH 8.4). The assays in order to study the function of imines, compounds 3 and 21
were carried out at 37 °C for 3 h, and the fluorescent signals were reduced to the corresponding amines 22 and 23 by
were recorded at lex 360 nm and lem 460 nm on a Spectrum sodium borohydride. Although no general rules could be
M2 spectrofluorometer.
found whether double bond exerts positive effect on the in-
Molecular Modeling Molecular modeling and graphics hibitory activity toward MAO-A/B, comparison of com-
manipulations were performed by using the Discover studio pounds 3 with 22, 21 with 23 indicates that reduced species
2.1 software. Protein Data Bank crystallographic models of increase the inhibitor selectivity toward MAO-B: the values
human MAO-A (2BXR)9) and MAO-B (1GOS)10) were con- of SI decrease from 40.6 and 39.83 to 2.63 and 27.13 respec-
sidered for the docking studies. Both structures were ob- tively.
tained as adducts with two similar compounds: clorgyline for
To better understand MAO-inhibitor interactions and the
2BXR and pargyline for 1GOS, covalently linked to the FAD selectivity toward different MAO isoforms, docking studies
N5 nitrogen. Starting geometries of 1 and 2 complexes with were performed in the following discussions. Crystal struc-
MAO-A and MAO-B were generated by the Flexible docking tures of MAO-A and B were retrieved from the Protein Data
method.
Bank (PDB) as targets, and selective inhibitors 1 and 2 were
considered for docking studies by using the Discover studio