Diaryl substituted pyrrolidinones and pyrrolones as 5-HT2C inhibitors: Synthesis and biological evaluation

Article abstract of DOI:10.1016/j.bmcl.2006.05.034

Within the continuous quest for the discovery of novel compounds able to treat anxiety and depression, the generation of a pharmacophore model for 5-HT_2C receptor antagonists and the discovery of a new class of potent and selective 5-HT_2C<

Full text of DOI:10.1016/j.bmcl.2006.05.034

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3906–3912
Diaryl substituted pyrrolidinones and pyrrolones as 5-HT2C
inhibitors: Synthesis and biological evaluation
Fabrizio Micheli,a,* Alessandra Pasquarello,a Giovanna Tedesco,a Dieter Hamprecht,a
Giorgio Bonanomi,a Anna Checchia,a Albert Jaxa-Chamiec,b Federica Damiani,a
Silvia Davalli,a Daniele Donati,a Chiara Gallotti,a Marcella Petrone,a Marilisa Rinaldi,a
Graham Riley,b Silvia Terrenia and Martyn Woodb
aGlaxoSmithKline Psychiatry Centre of Excellence for Drug Discovery, Via Fleming, 4, 37135 Verona, Italy
bNFSP, Harlow, Essex, UK
Received 21 April 2006; revised 10 May 2006; accepted 11 May 2006
Available online 30 May 2006
Abstract—Within the continuous quest for the discovery of novel compounds able to treat anxiety and depression, the generation of
a pharmacophore model for 5-HT2C receptor antagonists and the discovery of a new class of potent and selective 5-HT2C molecules
are reported.
Ó 2006 Elsevier Ltd. All rights reserved.
During the last decade, one approach of GSK research
in CNS drug discovery was aimed at the discovery of
novel chemical entities (NCE) able to modulate the
serotoninergic neurotransmission. The 5-HT2 receptor
family belongs to the seven-membrane-spanning G-pro-
tein-coupled receptor superfamily and represents one of
the 7 classes of 5-HT receptors.1 The 5-HT2 receptor
family consists of 5-HT2A, 5-HT2B and 5-HT2C recep-
tors that share common primary structure, secondary-
messenger system and pharmacological profile.1,2
to identify additional NCEs and exploiting the data gen-
erated over the years in GSK, a pharmacophore model
was developed in house and used both to give priority
to the synthesis of different chemical classes and as a
filter to screen in silico the compound collection.
This manuscript reports a new and potent class of
5-HT2C antagonists, while the in silico screening
activities will be described in a separate manuscript.
Pharmacophore modelling was extensively applied as
part of an integrated medicinal and computational
chemistry strategy aimed at the identification of novel
5-HT2C antagonists.
The functional role of the different 5-HT2 receptor sub-
types however has not been fully elucidated yet due to
the limited availability of selective ligands. Several lines
of evidence suggest that selective 5-HT2C antagonists
may be useful in the treatment of anxiety and depres-
sion.3–5 More recently, published data have suggested
that 5-HT2C antagonists may offer interesting therapeu-
tic opportunities for the treatment of schizophrenia,6
migraine7 and Parkinson’s disease.8
At this purpose, a small set of representative structures
was selected from the wide in house collection to span
the available chemical diversity. All compounds includ-
ed in the study are endowed with a high affinity in a
5-HT2C binding assay (pKi > 8.5),4 and are at least
100-fold selective towards 5-HT2A/2B. In addition, they
were assumed to bind at the 5-HT2C receptors with a
common binding mode according to the available
SAR data for the selected molecules. Representative
examples of compounds included in the study are shown
below in Table 1.
GSK showed a long-standing interest in the 5-HT2C
receptor and a number of potent different selective tem-
plates were disclosed and investigated.9–12 With the aim
Keywords: Serotonin; 5-HT; 5-HT2C; 7-TM; Pharmacophore.
*
All pharmacophore modelling work was performed with
program Catalyst 4.6.13 Ligand conformational searches
Corresponding author. Tel.: +39 045 8218515; fax: +39 045
8218196; e-mail: fm20244@gsk.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.034

F. Micheli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3906–3912
Table 1. Representative compounds included into the pharmacophore modelling exercise
3907
19,10
211
312
O
Cl
Cl
CH3
Cl
Cl
O
O
O
O
N
N
O
N
CH3
O
N
N
H3C
N
N
O
N
N
O
H2N
CH3
were carried out using the BEST routine (max 200 con-
formations, energy window of 10 Kcal), while common-
feature alignments were performed using the HipHop
algorithm. Standard catalyst pharmacophore features
(H-bond acceptor, positive ionisable, aromatic rings,
hydrophobic groups) were selected according to the
extensive SAR data available in house for these com-
pound series.
discarded as it is irrelevant in determining the high
5-HT2C affinity.
Considering the wealth of data available and the chem-
ical diversity of the 5-HT2C antagonists available, new
chemical templates were rationally designed and evalu-
ated for chemical feasibility. Subsequently, to give the
appropriate priority to the synthesis of the possible
NCEs, the above-described pharmacophore model was
used as a sieve and the best fitting compounds were
given a higher priority in the synthetic activities.
In Figure 1, the best pharmacophore model obtained is
shown superimposed to some representative compounds
(3, 4, and 5) included in the study.
The appropriately decorated lactamic scaffolds (6, 7, 8, 9
and Fig. 2) resulted among the top scorers in terms of
fitting to the pharmacophore model as exemplified by
the fitting of compound 7a in Figure 3.
As can be seen from the above figure, the pharmaco-
phore model consists of a positive ionisable group (red
sphere), which is mapped by the basic nitrogen on the
piperidine side chain of compound 3, a H-bond acceptor
(green spheres), mapped by the carbonyl oxygen on the
cyclic urea moiety of compound 3, an aromatic ring (yel-
low sphere) and 3 hydrophobic groups (cyan spheres).
The pharmacophore was validated by means of SARs
derived from data available in house, and with the syn-
thesis and testing of a small set of ad hoc compounds
(data not shown). As a result of that, it was demonstrat-
ed the key role played in particular by both the positive
ionisable group and the H-bond acceptor functionality
to achieve high 5-HT2C affinity. A second H-bond
acceptor functionality, mapped by the ether oxygen on
the piperidine side chain of compound 3, was instead
Accordingly, a synthetic strategy was devised for the
preparation of both series as outlined in Scheme 1.14
The alkylation of appropriately substituted phenyl ace-
tic acid was performed by generating the bis anion with
2 mol of lithium HMDSA and with the subsequent
quench with allyl iodide. In the next step, the corre-
sponding acid chlorides were prepared with oxalyl chlo-
ride and reacted with suitably decorated anilines.
Treatment with NaIO4 and OsO4 afforded immediately
the ring closure to give the 5-hydroxy-1,3-diphenyl-2-
pyrrolidinone 16, a key intermediate to obtain either
2
2
R
R
N
N
1
1
R
R
O
O
6
7
2
2
R
R
N
N
1
1
R
O
O
R
Figure 1. Pharmacophore model for the 5-HT2C ligands. Colour
coding of pharmacophoric features: green, H-bond acceptor; red,
positive ionisable; cyan, hydrophobic; yellow, aromatic ring.
8
9
Figure 2. The newly designed 5-HT2C antagonists.

3908
F. Micheli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3906–3912
lactams series. The unsaturated lactams 7 were obtained
dissolving compounds 16 in neat TFA, while the satu-
rated lactams 6 were isolated when the reaction was per-
formed in presence of Et3SiH. Biological data for the
unsaturated lactams 7 are reported in Table 2 and con-
firm the working hypothesis.
The key features of the selective 5-HT2C pharmacophore
are clearly conserved and this confers high selectivity to
all the compounds reported below. Considering com-
pound 7a as an example of the lactam class, no affinity
Figure 3. Compound 7a superimposed to the 5-HT2C pharmacophore.
4
NH2
4
4
OH
H
4
R
R
R
R
(i)
(ii)
(iii)
N
1
2
O
COOH
3
R, R
COCl
3
O
+
2
R
3
R
3
R
R
R
1
R
10
11
12
14
13
1
2
(iv)
OH
R, R
4
R
N
(vi)
O
3
R
OH
N
1
2
OH
H
4
4
7
R, R
R
R
N
1
2
1
2
R, R
4
R, R
O
3
O
3
R
(v)
N
(vii)
R
R
3
O
R
6
16
15
Scheme 1. Reagents and conditions: (i) LiHMDS 1 M in THF, allyl iodide, THF, ꢀ78 to 0 °C, 78%; (ii) (COCl)2, CH2Cl2, DMF (cat.), 0 °C (15 min)
to rt, 18 h; (iii) TEA, CH2Cl2, 0 °C to rt, 5 h, 84–57%; (iv) OsO4 (4 wt % soln in H2O), 4-methyl-morpholine N-oxide, acetone/H2O 8:1, rt, 5 h; (v)
NaIO4, THF/H2O, rt, 2 h, 62% two steps; (vi) TFA, rt, 8 h, 56%; (vii) Et3SiH, TFA, CH2Cl2, 63%.
Table 2. Affinity results associated with class 7
4
O
2
R
R
1
R
N
3
R
O
N
Compound
R1
R2
R3
R4
pKi
5-HT2C
5-HT2A
5-HT2B
7a
7b
7c
7d
7e
7f
H
OCH3
Cl
Cl
Cl
Cl
Cl
Cl
Cl
F
Cl
Cl
9.1
7.2
<5.9
7.3
6.9
6.7
9.1
8.3
8.2
7.9
9.5
8.9
8.3
8.2
8.1
8.5
8.8
7.9
8.4
8.6
8.1
6.2
<5.9
<5.0
<5.8
<5.9
<5.9
6.2
6.2
<5.8
<5.5
<5.8
<5.7
<5
H
H
OCF3
NHAc
CH3
Cl
Cl
H
H
Cl
Cl
H
Pyrrolidine
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
7g
7h
7i
H
H
6.3
H
CF3
H
H
6.1
6.4
<6.1
6.0
H
F
7j
H
H
CF3
CH3
Br
H
<5.9
<6.0
6.8
<6.0
<6.0
6.7
7k
7l
H
H
H
H
7m
7n
7o
7p
7q
7r
7s
7t
Me
Me
Me
Me
Me
Me
Me
H
Cl
H
6.1
6.2
6.3
Cl
Cl
<5.8
<5.7
5.9
F
<5.9
6.6
H
F
H
CH3
CF3
Br
H
6.2
5.8
5.8
H
<5.0
<6.2
6.2
H
6.4
6.2
Cl
H
7u
H
Cl
6.0
<5.8

F. Micheli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3906–3912
3909
(pKi < 5.5) was detected on 5-HT1A, 5-HT1B, 5-HT1D
,
moderately high plasmatic clearance (59 ml/min/kg).
As the exploration was further detailed, it resulted quite
evident that the di-chloro substitution pattern on the
LH-aromatic moiety proved to be the best compromise
in terms of potency and metabolic stability.
5-HT6 and 5-HT7 receptors. No affinity was also
detected on Adr. b2 or DA D2 receptors, while poor
affinity was observed for DA D3 (pKi = 6.34) and Adr.
a1 (pKi = 5.74) receptors.
The different decoration of the scaffold, however, leads
to different values in potency. From the exploration, it
resulted quite clear that a methoxy group was favoured
in position R2, probably contributing to some specific
interaction within the receptor (es 7a). Actually, its
replacement with a trifluoromethoxy group (7c) led to
a completely inactive compound. The replacement with
a neutral group (Me, 7e) or with an electron-withdraw-
ing group (Cl, 7b) led to a significative decrease of
potency. The introduction of the acetamido group
(NHAc, 7d) where a possible H-bonding acceptor is still
present but in a slightly different position still leads to
almost 100-fold less potent compound.
Unfortunately, some of the selected RH resulted in
moderate to high blood clearance suggesting a possible
metabolic way of degradation of the scaffold. Accord-
ingly, to match the potency with the appropriate PK
parameters, a number of aniline intermediates were pre-
pared according to the Scheme 2 below reported, where
alkylation of the phenol 18 with substituted 1-(chloro-
acetyl)piperidines (or other amides) was followed by
reduction of the amide with boranes and completed with
the catalytic hydrogenation of the nitro group.
The results of this exploration are reported in Table 3.
The piperidine (7a) was clearly the best substituent as
far as potency was concerned because all the modifica-
tions led to slightly less potent compounds.
The above-reported drop of potency is consistent with
the decrease of electron-donating power of the substitu-
ents in agreement with the Swain–Lupton parameters,
as shown in Figure 4. A clear trend can actually be
observed between pKi on 5-HT2C receptor and Swain
and Lupton R parameter,15,16 a measure of the electron
donor resonance effect of substituents. Affinity towards
5-HT2C receptor increases as R decreases, that is, as
the electron donor resonance effect of the substituents
increases.
On the other hand, when considering the overall profile
of the molecule, including its PK properties, the intro-
NO2
NO2
R
N
R
(ii)
(i)
N
+
O
OH
O
O
O
Cl
O
Exploration of the left-handside aromatic portion was
well tolerated with respect to the potency and in partic-
ular led to very potent and selective compounds (entries
7g and 7k).
17
18
19
R= Me
R=4,4-diF
R=i-Pr
R=4,4 diMe
NO2
NH2
O
R
R
Nonetheless, the potency of a compound is not the only
important factor which leads a medicinal chemistry
exploration as other parameters are very important in
achieving a drug-like profile in a NCE.
(iii)
N
N
O
O
O
20
21
Considering the pharmacokinetic (PK)17 properties of
this class of lactams, it resulted quite clear that the
exploration was going in the right direction for a possi-
ble CNS drug. Compound 7a, the starting point, was
endowed with a good brain/blood ratio of 3.9 and
acceptable bioavailability in rats after oral administra-
tion (F = 11%), even if this was probably affected by a
Scheme 2. Reagents and conditions: (i) K2CO3, DMF, rt, 58–64%; (ii)
BH3, THF, reflux, then HCl 6N, 80 °C; (iii) Pd/C (10% Pd) (cat),
CH3OH, rt, H2, 1 atm.
Table 3. Affinity values for the class 7 where the right-hand-side chain
was modified
O
Cl
OMe
9.5
-OMe
9
N
5
R
8.5
8
Cl
Compound R5
O
pKi
7.5
-Cl
-CH3
7
6.5
6
-NHAc
-0.25
5-HT2C 5-HT2A 5-HT2B
7a
7v
Piperidine
Morpholine
Pyrrolidine
9.1
8.2
8.4
8.1
8.5
6.2
5.4
6.2
<5.8
<5.8
<5.3
<6.1
6.5
-OCF3
-0.05
5.5
7w
7x
7y
7z
6.2
-0.55
-0.45
-0.35
-0.15
0.05
4,4-DiF-piperidine
4-Me-Piperidine
<5.3
6.1
Swain and Lupton R
4,4-DiMe-piperidine 8.4
4-i-Pr-piperidine 7.9
<5.2
<6.1
Figure 4. Trend between 5-HT2C pKi and Swain and Lupton R
parameter.15,16
7aa
<6.0

3910
F. Micheli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3906–3912
Table 5. Affinity data for substituted lactams
duction of a group on the position 4 of the piperidine
(e.g., 7y) led to significative achievements. The slightly
reduced potency was actually matched with a reduced
plasma clearance in rat, leading to higher bioavailability
and to a more balanced profile as reported in Table 4.
Compound
R1
R2
pKi
5-HT2C
5-HT2A
5-HT2B
22
Cl
H
H
8.1
8.0
<6.2
5.9
<6.4
<6.0
<6.1
<5.1
22a
22b
22c
H
Cl
H
CH3
CH3
6.4
<5.8
<6.1
<5.9
A further step in the investigation of the unsaturated
lactams was related to the presence of the double bond
which was decorated as reported in Figure 5. As clearly
reported in Table 5, the introduction of the methyl
group on the double bond brings to a slightly reduced
potency at the receptor, with a complete abolishment
of the 5-HT2A and 5-HT2B activity for compound 22.
On the other hand, the PK profile was superimposable
with the unsubstituted system, and no further reduction
of the plasmatic clearance was achieved with the intro-
duction of the gem dimethyl group which was extremely
detrimental to the desired pharmacological activity (22b
and 22c).
Table 6. Affinity data for the saturated lactams class 6
4
O
R
R2
R1
N
3
R
O
N
Compound R1
R2
R3 R4
pKi
5-HT2C 5-HT2A 5-HT2B
6a
6b
6c
6d
6e
6f
H
OCH3 Cl Cl 7.9
Cl Cl Cl 7.0
6.3
<6.0
6.4
5.8
<5.8
5.7
H
As far as the saturated lactams 6 are concerned, it can
clearly be appreciated by Table 6, that the potency on
the desired target was not as good as the previously
described class. Moreover, when analyzed from the PK
point of view, a clear reduction of bioavailability was
observed.
CH3
H
OCH3 Cl Cl 7.9
OCF3 Cl Cl 6.1
NHAc Cl Cl 6.5
5.5
5.8
H
<5.5
<6.0
<5.6
<5.8
6.4
<5.8
<6.0
<5.3
<5.8
<6.0
<6.0
<5.8
H
OCH3 Cl
H
F
7.6
7.0
6g
6h
6i
H
OCH3
CH3
DiCH3 OCH3 Cl Cl 7.9
H
H
Cl Cl 6.4
Through studies of metabolite ID, it was possible to
clearly relate this to a specific metabolic pathway associ-
ated to the hydroxylation of the saturated lactam.
6j
6k
i-Pr
H
OCH3 Cl Cl 7.6
OCH3 7.0
6.4
<5.1
F
H
comparable PK profile. This might probably be linked
to some specific key interactions within the receptor;
currently both receptor modelling studies as well as
SDM studies are in progress to fully evaluate this point.
As illustrated in Table 7, the introduction of a a methyl
group was neutral in terms of pKi; nonetheless, the
working hypothesis was proved as compound 24 showed
a dramatic increase in bioavailability. Finally, to fully
complete the exploration, the corresponding ‘inverted’
lactams were prepared like in Scheme 1, but inverting
the acidic moieties with the anilines.
A pharmacophore model was generated to predict
antagonist recognition site on the 5-HT2C receptor.
New scaffolds were filtered through this newly de-
signed tool and ranked according to the overall qual-
ity of their fitting. The synthesis of the best fitting
compounds led to the identification of a new class
of potent and selective 5-HT2C antagonists. In partic-
ular, compound 7y represented the best compromise
between the potency/selectivity and the PK properties.
This compound has nanomolar affinity for the 5-HT2C
receptor. Its PK profile is encouraging to allow future
tests in different models of anxiety and depression.
Finally, further classes were identified through the
screening of the corporate database using the pharma-
cophore model and will the subject of further commu-
nications in due time.
As reported in Tables 8 and 9, compounds belonging to
these classes were endowed with similar, but not ‘super-
imposable’ profiles when compared with their counter-
parts 6 and 7, showing a reduced affinity and a
Table 4. Head to head PK comparison between compounds 7y and 7a
7y
11
4.0
4.5
52
7.2
1058
7a
Clp (ml/min/kg)
Vdss (L)
T1/2 (h)
59
10.7
2.8
11
F (%)
B/P
Brain concn 50 (ng/g)
3.9
320
1
1
R
O
O
N
R
O
O
N
N
N
O
O
Cl
Cl
22, 22a
Figure 5. Substituted double bond derivatives.
22b, 22c

F. Micheli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3906–3912
Table 7. Affinity data for the a substituted saturated lactams
3911
O
Cl
OMe
R1
N
Cl
O
N
Compound
R1
pKi
F (%)
B/P
Cl (ml/min/kg)
5-HT2C
5-HT2A
5-HT2B
23
24
H
CH3
8.3
8.3
6.2
<6.3
6.0
<6.1
13
50
5
1.8
65
20
Table 8. Affinity data for the inverted saturated lactams class 8
R3
OMe
O
R1
N
2
R
O
N
Compound
R1
R2
R3
pKi
5-HT2C
5-HT2A
5-HT2B
8a
8b
8c
8d
8e
8f
H
H
Cl
F
OCH3
OCH3
OCH3
Cl
7.0
7.7
7.8
8.0
8.0
7.7
<5.0
<5.7
<6.0
6.1
<5.0
5.8
H
H
<5.8
6.4
H
Cl
Cl
F
CH3
H
Cl
H
6.3
<5.9
<6.3
<6.1
Table 9. Affinity data for the inverted unsaturated lactams class 9
R2
R3
O
O
N
N
O
R1
Compound
R1
R2
R3
pKi
5-HT2C
5-HT2A
5-HT2B
9a
9b
9c
9d
9e
9f
H
Cl
H
Cl
F
Cl
8.3
8.0
8.3
8.2
8.2
6.7
6.9
5.9
<5.3
<5.2
<5.5
6.0
6.9
<5.7
<6.4
<5.8
6.3
H
OCH3
OCH3
OCH3
H
H
H
H
F
CH3
CH3
H
Cl
Cl
Cl
6.4
6.3
6.3
6.7
9g
Chem. Res. 1993, 3, 306; (c) Baxter, G.; Kennett, G.;
Blaney, F.; Backburn, T. Trends Pharmacol. Sci. 1995, 16,
105.
Supplementary data
Supplementary data associated with this article can be
found in the online version at doi:10.1016/
j.bmcl.2006.05.034.
3. Kahn, R. S.; Wetzler, S. Biol. Psychiatry 1991, 30, 1139.
4. Bromidge, S. M.; Dabbs, S.; Davies, D. T.; Davies, S.;
Duckworth, D. M.; Forbes, I. T.; Gaster, L. M.; Ham, P.;
Jones, G. E.; King, F. D.; Mulholland, K. R.; Saunders,
D. V.; Wyman, P. A.; Blaney, F. E.; Clarke, S. E.;
Blackburn, T. P.; Holland, V.; Kennett, G. A.; Lightowler,
S.; Middlemiss, D. N.; Trail, B.; Riley, G. J.; Wood, M. D.
J. Med. Chem. 2000, 43, 1123.
5. (a) Kennett, G. A.; Lightowler, S.; De Biasi, V.; Stevens,
N. C.; Wood, M. D.; Tulloch, I. F.; Blackburn, T. P.
Neuropharmacology 1994, 33, 1581; (b) Palvimaki, E. P.;
Roth, B. L.; Majasuo, H.; Laakso, A.; Kuoppamaki, M.;
References and notes
1. Hoyer, D.; Clarke, D. E.; Fozard, J. R.; Hartig, P. R.;
Martin, G. R.; Mylecharane, E. J.; Saxena, P. R.;
Humphrey, P. P. A. Pharmacol. Rev. 1994, 46, 157.
2. (a) Humphrey, P. P. A.; Hartig, P.; Hoyer, D. Trends
Pharmacol. Sci. 1993, 14, 233; (b) Nelson, D. L. Med.

3912
F. Micheli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3906–3912
Syvalahti, E.; Hietala, J. Psychopharmacology 1996, 126, 9. Gaster, L. M.; Wyman, P. A.; Mulholland, K. R.; Davies,
234; (c) Ni, Y. G.; Miledi, R. Proc. Natl. Acad. Sci. U.S.A.
1997, 94, 2036; (d) Quested, D. J.; Sargent, P. A.; Cowen,
P. J. Psychopharmacology 1997, 133, 305; (e) Kennaway,
D. J.; Moyer, R. W. Brain Res. 1998, 806, 257; (f) Moreau,
J.-L.; Boes, M.; Jenck, F.; Martin, J. R.; Mortas, P.;
Wichmann, J. Eur. Neuropsychopharmacol. 1996, 6, 169.
D. T. WO 9623769, 1996.
10. Gaster, L. M.; Wyman, P. A.; Mulholland, K. R.; Davies.,
D. T.; Duckworth, D. M.; Forbes, I. T.; Jones, G. E. WO
9623783, 1996.
11. Bromidge, S. M.; Lovell, P. J.; Moss, S. F.; Serafinowska,
H. T. WO 20010809, 2002.
6. (a) Di Giovanni, G.; De Deurwaerdere, P.; Di
Mascio, M.; Di Matteo, V.; Esposito, E.; Spampinato,
U. Neuroscience 1999, 91, 587; (b) Di Matteo, V.; Di
Giovanni, G.; Di Mascio, M.; Esposito, E. Neuro-
pharmacology 1998, 37, 265; (c) Millan, M. J.;
Dekeyne, A.; Gobert, A. Neuropharmacology 1998,
37, 953; (d) Reavill, C.; Kettle, A.; Holland, V.;
Riley, G.; Blackburn, T. P. Br. J. Pharmacol. 1999,
126, 572.
12. Bromidge, S. M.; Lovell, P. J.; Goodacre, C. WO
20030107, 2003.
13. Catalyst 4.6, Accelrys Inc.
14. Bonanomi, G.; Hamprecht, D.; Micheli, F.; Terreni, S.
WO 2004089897, 2004.
15. Swain, C. G.; Lupton, E. C. J. Am. Chem. Soc. 1968, 90,
4328.
16. Hansch, C.; Leo, A.; Unger, S. H.; Kim, K. H.; Nikaitani,
D.; Lien, E. J. J. Med. Chem. 1973, 16, 1207.
17. The research complied with national legislation and with
company policy on the Care and Use of Animals and with
related codes of practice.
7. Kalkman, H. O. Life Sci. 1994, 54, 644.
8. Fox, S. H.; Moser, Bill; Brotchie; Jonathan, M. Exp.
Neurol. 1998, 151, 35.