4992
C. J. Goodacre et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4989–4993
Table 2. The 5-HT2A/2B/2C receptor binding affinities,a selectivities over 5-HT2A and 5-HT2B, and in vivo activitye of [4-methoxy-3-(2-piperidin-1-yl-
ethoxy)-phenyl]-3-aryl-1,3-dihydro-imidazol-2-ones 19 and 20 and 3-(3-fluoro-phenyl)-1-[4-methoxy-3-(2-piperidin-1-yl-ethoxy)-phenyl]-imidazoli-
dine-2,4-dione 21
O
O
R1
N
N
N
O
X
Y
R1
pKi 5-HT2C
pKi (selectivity)
% reversal at 2 he
b
Compound
X
Y
c
d
5-HT2A
5-HT2B
19
20
21
3F
3,5-diF
3F
CH@CH
CH@CH
CO–CH2
8.2
8.1
7.3
<6.0 (>160)
<5.0 (>1250)
<5.0 (>200)
5.7 (320)
<5.1 (>1000)
<5.2(130)
IA
—
—
a–eSee corresponding footnotes in Table 1.
dihydro-imidazol-2-ones (19, 20). Hydrogenation of 20
allowed the preparation of 22. Compound 21 (Scheme
3) was prepared by protection of the phenol, nitro
reduction and subsequent aniline alkylation with ethyl
bromoacetate. Treatment with 3-fluorophenyl isocya-
nate and heating yielded the imidazol-2,4-dione core.
Removal of the benzyl protecting group and alkylation
with 1-(2-chloroethyl)piperidine afforded the final
compound.
tween 5-HT2C affinity and in vivo activity in the hypol-
ocomotion model was observed. However,
pharmacokinetic profiling in the rat revealed that
weakly active compounds 10 and 11 showed high in vi-
vo clearance (>120 mL/min/kg). In contrast, much low-
er clearance (45 mL/min/kg) was observed for
compound 8. Confirmation that a basic amine substitu-
ent was essential for 5-HT2C affinity in this series was
seen with compound 13. Compounds 17 and 18 incor-
porated additional aspects from the indolinyl ureas ser-
ies whilst maintaining the 3,4-diCl aromatic
substitution pattern. Changing the central aryl core
to a pyridine gave disappointing 5-HT2C binding affin-
ities so were not progressed to the hypolocomotion
model.
To investigate whether the SAR from the original urea
and cinnamide series could be transferred to the new
cyclic urea series, compounds 5 and 6 were prepared.
Although 5 maintained good 5-HT2C affinity with very
similar selectivity over 5-HT2B compared to 2 and
showed good solubility (0.82 mg/mL in 0.1 N HCl for
the oxalate salt), it was inactive in the rat hypolocomo-
tion model. In this centrally mediated PD model of 5-
HT2C function, the ability of compounds to block the
hypolocomotion in rats produced by a standard dose
of the moderatively selective 5-HT2C agonist meta
chlorophenylpiperazine (mCPP) was measured.2 In
contrast, the 2-Cl analogue 6 showed a significant
reduction in 5-HT2C affinity and proved 25-fold less
potent than the unsubstituted analogue 9. This prompt-
ed a further investigation of alternative aromatic sub-
stitution. Incorporation of an additional chlorine
substituent at the 3 position to give 7 gave some
improvement in 5-HT2C affinity. However, transposing
the unfavourable 2-Cl substituent to the 4 position
gave a highly potent and selective compound 8, which
showed significant activity in the rat hypolocomotion
model. In addition, the solubility of compound 8
(190 lg/mL in 0.1 N HCl for the hydrochloride salt)
was improved over the original urea 3. With the aim
of improving this further, the less lipophilic analogue
10 was targeted. Compound 10 also showed an excel-
lent binding profile, improved solubility (480 lg/mL
in 0.1 N HCl for the hydrochloride salt) although
much reduced activity in the PD model. Our attention
then moved to optimization of the basic amino group.
In comparison with the morpholine and dimethyl ami-
no compounds, the piperidine analogues showed the
best combination of 5-HT2C affinity with good selectiv-
ity over 5-HT2A and 5-HT2B. The morpholine ana-
logues consistently showed reduced 5-HT2B selectivity
particularly in compound 16. No direct correlation be-
The effects of introducing unsaturation (19, 20) or an
additional carbonyl group (21) were also studied. Both
modifications resulted in lower 5-HT2C affinity though
selectivity over 5-HT2A and 5-HT2B was still achievable
(Table 2). Interestingly, compound 22 (Table 1)
obtained from the double bond reduction of 20 and
containing a 3,5 disubstituted phenyl combined high 5-
HT2C affinity, exceptional selectivity and activity in the
hypolocomotion model. Further modifications will be
reported in future publications.
In summary, a series of cyclic urea analogues has been
prepared with excellent 5-HT2C affinity and selectivity
over 5-HT2A and 5-HT2B. This series has been shown
to be selective against a range of other 5-HT and dopa-
minergic receptor subtypes and antagonist activity at
5-HT2C has been confirmed in a functional assay. Im-
proved solubility over the indolinyl urea 3 has also been
achieved along with improved metabolic stability over
the indolinyl cinnamide 4. Compounds 8, 12 and 22
have also shown significant oral activity in the rat hypol-
ocomotion model.
References and notes
1. (a) Humphrey, P. P. A.; Hartig, P.; Hoyer, D. Trends
Pharmacol. Sci. 1993, 14, 233; (b) Baxter, G.; Kennett, G.
A.; Blaney, F.; Blackburn, T. P. Trends Pharmacol. Sci.
1995, 16, 105.