J. Kim, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Table 3
5-HT2C was mostly abolished while binding to SERT was less affected
(11a-11f). In stark contrast, a longer linker by one carbon (11g-11l)
even further enhanced association with 5-HT2A, 5-HT2C, and SERT: 11j
showed IC50s of 104 nM (5-HT2A), 38 nM (5-HT2c), and 488 nM (SERT),
respectively. Finally, acetamides (12a-12c) were explored instead of
ethyl acetates and comparable binding affinities to corresponding
compounds (11h, 11j, and 11l) were observed (Table 3).
Biological evaluation of acetamide compounds.
Among several promising compounds, 11j and 11l were subjected
to hERG binding assay to assess the risk of fatal arrhythmias that some
was observed that both 11j and 11l did not efficiently inhibit binding of
astemizole. Furthermore, CYP450 inhibitory activities were briefly
tested against CYP3A4 and CYP2C19 and it was observed that 11l
barely inhibited both CYP3A4 and CYP2C19. Although 11j moderately
inhibited CYP2C19 with IC50 of 2.2 μM, but its inhibition appears
weaker than fluoxetine's.
Compounds R1
Inhibition at
10 μM for 5-
HT2A (%)
IC50 for Inhibition at
IC50 for IC50 for
5-HT2A
(nM)
10 μM for 5-
HT2c (%)
5-HT2c
(nM)
SERT
(nM)
12a
12b
12c
H
100
256
144
158
94
69
72
69
429
160
5-F 98
6-F 88
2014
1422
a
Not determined.
In vivo efficacy of 11j and 11l was assessed by the forced swimming
test measuring the total duration time showing the physical immobility
after mice were placed in water chamber (Fig. 3).14,17 Fluoxetine was
50 mg/kg of fluoxetine and 5, 10, 25, and 50 mg/kg of 11j or 11l were
administered to mice intraperitoneally 30 min before the experiments,
and total duration of immobility was recorded during the last 5 min of
the 6-minute testing period. In our setting, fluoxetine exhibited anti-
depressant effects at 50 mg/kg and 11j also showed comparable in vivo
efficacy to fluoxetine at the same dose. Furthermore, 11l presented
significant anti-depressant effects inducing almost 40% reduction of the
duration time of immobility starting from 25 mg/kg. Finally, we in-
vestigated potential synergistic effects between fluoxetine and 11j
(Fig. 4). In another forced swimming test, 50 mg/kg of fluoxetine was
dosed orally 60 min before the test, which resulted in reduction of the
immobility duration by approximately 20%. Oral treatment of 25 mg/
kg of 11j alone did not exhibit any noticeable effect. In contrast, when
we pre-treated 25 mg/kg of 11j orally 60 min before treatment of
50 mg/kg of fluoxetine, we were able to observe substantial additive
anti-depressant effects as the immobility was decreased by approxi-
mately 60%.
bromopropanoate (7a) with variously substituted 4-phenylpiperazines
and subsequent hydrolysis of ethyl esters to acids. The final amide
formation between 6a and 6c and 9a-9p employing EDC, HOBT, and N-
methylmorpholine furnished desired phthalazinone compounds 10a-
10t. We also seek to explore effects of the linker length of acids on
binding affinities with especially (3,4- or 2,3-dimethylphenyl)piper-
azines (8m or 8n) (Scheme 2). For this purpose, either ethyl 2-bro-
or 8n producing 9q-9t that were further reacted with 6a-6c to generate
desired 11a-11l in good yield. Furthermore, in order to potentially
enhance metabolic stability, amides (12a-12c) were synthesized from
11h, 11j, and 11l by amination of the ethyl ester with ammonium
(10a-10t, 11a-11l, and 12a-12c) against 5-HT2A, 5-HT2C, and SERT by
competition with [3H]Ketanserin for 5-HT2A, [3H]Mesulergine for 5-
HT2C, and [3H]Imipramine for SERT in cells expressing each receptor or
transporter, respectively (Tables 1–3). First, we examined effects of
fluoromethyl-, and nitro substituents did not show efficient binding to
any of these (10a-10g). Even methoxy- and methyl groups at different
locations (10h-10l) did not help to achieve potent binding, either. As
shown in the study of the quinazolinone compounds,15 simultaneous
installation of two methyl groups on the phenyl ring finally realized
great binding affinities against 5-HT2A, 5-HT2C, and SERT: especially
3,4-dimethyl- and 2,3-dimethylphenyl groups (10m and 10n) helped to
accomplish efficient binding to 5-HT2A, 5-HT2C, and SERT with mostly
stronger than 800 nM of IC50s, respectively. Excellent binding was
consistently observed after introduction of fluoro substituents at 5- or 6-
positions of the phthalazinone ring as exemplified in compounds 10t
that bound to 5-HT2A, 5-HT2C, and SERT with IC50s of 435 nM, 35 nM,
and 141 nM, respectively.
In summary, we have developed phthalazinone-based compounds
that potently bind to 5-HT2A, 5-HT2C, and the serotonin transporter
showing promising anti-depressant activities. The representative com-
pounds 11j and 11l displayed strong binding affinities against these
targets, and showed favorable toxicity profiles as determined by hERG
binding and CYP inhibition assays. Furthermore, these compounds
promoted comparable or even superior anti-depressant effects as com-
pared to fluoxetine in forced swimming test. Finally, potential sy-
nergistic in vivo anti-depressant effects of 11j with fluoxetine implicates
further optimization of current compounds could lead to production of
promising candidates for the treatment of major depressive disorder.
Declaration of Competing Interest
Then, we investigated effects of the linker length between the amide
group and piperazine moiety on the binding with (3,4- or 2,3-di-
methylphenyl)piperazines that conferred great binding affinities
(Table 2). When the linker was shortened, binding to both 5-HT2A and
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Table 4
hERG binding and CYP inhibitory activities of 11j and 11l.
Compounds
IC50 for inhibition hERG binding of [3H]Astemizole (μM)
IC50 for CYP3A4 inhibition (μM)
IC50 for CYP2C19 inhibition (μM)
11j
> 10
> 10
> 10
> 10
> 10
> 10
2.2
11l
> 10
0.065
Fluoxetine
4
Kim, Jihye
