High-efficacy 5-HT1A agonists for antidepressant treatment: A renewed opportunity

Article abstract of DOI:10.1021/jm070714l

We report the discovery of novel 5-HT_1A receptor agonists and describe the process that led to the antidepressant candidate 9 (F 15599). 9 has nanomolar affinity for 5-HT_1A binding sites and is over 1000-fold selective with respect to the other 5-HT₁ receptor subtypes, 5-HT_2-7 receptor families, and also numerous GPCRs, transporters, ion channels, and enzymes. In a cellular model of signal transduction, 9 activates h5-HT_1A receptors with an efficacy superior to that of the prototypical 5-HT_1A agonist (±)-8-OH-DPAT and comparators undergoing clinical trials. After acute oral administration in rats, 9 totally reverses immobility in the forced swimming test and produces behaviors characteristic of 5-HT_1A receptor activation. However, these effects occurred at widely separated doses, suggesting that 9 discriminates between distinct populations of 5-HT_1A receptors. While the clinical relevance of these observations is still unknown, this opens new perspectives for the treatment of depressive disorders.

Full text of DOI:10.1021/jm070714l

5024
J. Med. Chem. 2007, 50, 5024-5033
High-Efficacy 5-HT1A Agonists for Antidepressant Treatment: A Renewed Opportunity
Jean Louis Maurel,† Jean-Marie Autin,† Philippe Funes,‡ Adrian Newman-Tancredi,§ Francis Colpaert,| and Bernard Vacher*,‡
Task Force Chemistry, Medicinal Chemistry 1 DiVision, Neurobiology 2 DiVision, and Toulouse ISTMT 2, Pierre Fabre Research Center,
17 aVenue Jean Moulin, 81106 Castres Cedex, France
ReceiVed June 19, 2007
We report the discovery of novel 5-HT1A receptor agonists and describe the process that led to the
antidepressant candidate 9 (F 15599). 9 has nanomolar affinity for 5-HT1A binding sites and is over 1000-
fold selective with respect to the other 5-HT1 receptor subtypes, 5-HT2-7 receptor families, and also numerous
GPCRs, transporters, ion channels, and enzymes. In a cellular model of signal transduction, 9 activates
h5-HT1A receptors with an efficacy superior to that of the prototypical 5-HT1A agonist (()-8-OH-DPAT
and comparators undergoing clinical trials. After acute oral administration in rats, 9 totally reverses immobility
in the forced swimming test and produces behaviors characteristic of 5-HT1A receptor activation. However,
these effects occurred at widely separated doses, suggesting that 9 discriminates between distinct populations
of 5-HT1A receptors. While the clinical relevance of these observations is still unknown, this opens new
perspectives for the treatment of depressive disorders.
Introduction
Previous reports from this laboratory1 described a novel series
a
of 5-HT1A agonists exhibiting unprecedented selectivity and
efficacy at the 5-HT1A receptor subtype. This research effort
Figure 1. Structure of F 13640.
culminated in the discovery of 1 (F 13640, Figure 1), which is
currently undergoing phase II clinical trials for the treatment
of severe, chronic pains.2 In preclinical models, the analgesia
produced by 1 excels that of available therapies in many
respects: (1) the magnitude of the analgesic effect equals that
of opioids with nociceptive pain and surpasses that of opioids,
antiepileptic, NMDA-antagonist, and antidepressant agents with
neuropathic pain;3 (2) its spectrum of activity is much broader,
encompassing acute and chronic, central and peripheral pains
of different origins;4 and (3) the time-course of the effect is
exceptional; the analgesia grows rather than decays upon chronic
administration (inverse-tolerance),5 and pain relief persisted for
up to 2 months after discontinuation of the treatment (curative-
like action).6 Clearly, compound 1 opens novel therapeutic and
research avenues in the field of analgesia.7
Despite a compelling pharmacological rationale gained over
more than two decades of intense research,8,9 only two poorly
selective, 5-HT1A partial agonists are marketed as anxiolytics,
that is, buspirone10 and tandospirone.11 Still, the results ac-
cumulated on 1 suggested that 5-HT1A agonists have indeed
therapeutic potential in domains other than anxiety. The enticing
prospect that the story of 1 in pain states might repeat itself in
other CNS disorders in which 5-HT1A transmission has been
implicated, such as depression12 and neuroprotection,13 prompted
ustothesearchfornovelhigh-efficacy5-HT1A agonists. Giventhat
compound 1 survived the evermore-demanding toxicity and
tolerance requirements, we focused our medicinal chemistry
effort on close structural analogues of 1. In practice, therefore,
we sought after compounds endowed with similar pharmaco-
dynamics but having altered pharmacokinetics relative to 1.14
Here, we describe the process through which we discovered
compound 9 (F 15599) and highlight some of the pharmacologi-
cal features that distinguish 9 from the parent molecule 1. We
also touch upon the early experimental cues that nurtured the
hypothesis that 9 interacts preferentially with certain populations
of 5-HT1A receptors rather than others. Such a profile, we
believe, renders 9 particularly adapted for exploring its antide-
pressant potential in human, a domain in which there is a
significant unmet need.15
Chemistry
Target compounds 2-21 were all prepared by reductive
amination either between an arylmethylamine of the type XII
or XIII and the cyanohydrine XIVa (Scheme 2), or between
an arylcarboxaldehyde of the type XVIII or XIX and the amine
XIVb (Scheme 3). The synthesis of intermediates XIVa and
XIVb has been described previously.16
Scheme 1 summarized the routes used to access the pyrim-
idylmethylamines of the type XII. From the commercially
available 2,4-dichloro-5-methylpyrimidine I, successive intro-
duction of a cyano group (II), then of a methylamino group,
led to III in which R1 ) CH3 and R ) NHCH3. On the other
hand, displacement of the 2-methylsulfonyl group in the known
substrate IV17 by cyanide anion18 yielded III in which R1 )
CO2CH3 and R ) H. Alternatively, the 2-chlorine on the Boc-
carbamate derived from V19 was first converted into a meth-
ylsulfonyl group, and then the cyano group was installed as
above to provide III in which R1 ) H and R ) NBocCH3.
Reduction of the cyano function in III by catalytic hydrogena-
tion furnished the corresponding amines XII. This reaction was,
however, more conveniently carried out in two steps: trapping
* Corresponding author. Telephone: (+33) 56371 4222. Fax: (+33)
56371 4299. E-mail: bernard.vacher@pierre-fabre.com.
† Task Force Chemistry.
‡ Medicinal Chemistry 1 Division.
§ Neurobiology 2 Division.
| Toulouse ISTMT 2.
a Abbreviations: HT, hydroxytryptamine; GPCRs, G-protein coupled
receptors; NMDA, N-methyl-D-aspartic acid; D, dopamine; Emax, maximum
effect; FPT, forepaw treading; FBP, flat body posture; FST, forced
swimming test; RGS, regulator of G-protein signaling; SSRIs, selective
serotonin reuptake inhibitors; SNRIs, serotonin and noradrenaline reuptake
inhibitors; TCAs, tricyclic antidepressants; MAOIs, monoamine oxidase
inhibitors; pEC50, negative logarithm of the dose that is effective in 50%
of test subject; MSD, minimum significant dose; Boc, tert-butoxycarbonyl;
DAST, (diethylamino)sulfur trifluoride.
10.1021/jm070714l CCC: $37.00 © 2007 American Chemical Society
Published on Web 09/06/2007

5-HT1A Agonists for Antidepressant Treatment
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20 5025
Scheme 1. Synthesis of Pyrimidine Intermediates of the Type XIIa
a Reagents and conditions: (i) NaCN, H2O, (CH3)3N; (ii) CH3NH2, EtOH; (iii) NaCN, DMSO; (iv) (Boc)2O, DMAP, THF; (v) CH3SNa, DMF; (vi)
mCPBA, CH2Cl2; (vii) H2 Pd/C, (Boc)2O, EtOAc; (viii) CH3OH, CH3ONa; 1,3-di-CO equivalent; (ix) NaBH4, EtOH; (x) DAST, CH2Cl2; (xi) DPPA, NEt3,
tBuOH; (xii) NaH, (CH3)2SO4; (xiii) H2 Pd/C, CH3OH; (xiv) HCl, iPrOH, EtOAc.
Scheme 2. Preparation of Ligands 7-18, 20, and 21a
Scheme 4. Preparation of Ligands 2-6, 19a
a Reagents and conditions: (i) NaBH(AcO)3, NEt3, C2H4Cl2.
a Reagents and conditions: (i) NaBH3CN, CH3OH, DABCO.
Scheme 3. Synthesis of Pyridine Intermediates XVIIIa
The diazine-type intermediates XIII were obtained from the
known 2-cyano-5-methylpyrazine24 and 3-cyano-6-methylpy-
ridazine.25 Thus, hydrogenation of 2-cyano-5-methylpyrazine
and 3-cyano-6-methylpyridazine on deactivated Pd0 catalyst
provided the 5-aminomethyl-2-methylpyrazine26 and the 6-ami-
nomethyl-3-methylpyridazine XIII, respectively (see Supporting
Information). Reductive amination between XIII and XIVa then
supplied compounds 7 and 8 (Scheme 2).
The aldehydes of the type XVIII were synthesized according
to Scheme 3. Among compounds XVIII, a few were known
(R1 ) Cl, CH2OH, and CH2F),27 and the others were prepared
from the ester XV.28 Thus, the alcohol function in XV was
protected as a MOM-ether, and then the C-5 ester function was
reduced to the alcohol XVI. Adjusting the oxidation state in
XVI to the aldehyde level followed by gem-difluoration with
an excess of DAST led to the derivative XVII.29 Cleavage of
the MOM-protecting group and then oxidation of the resulting
alcohols with Mn(IV) oxide completed the synthesis of alde-
hydes XVIII. The pyrimidine carboxaldehyde XIX was prepared
in three steps and modest overall yield from the known
compound 4-(fur-2-yl)-2-methyl-pyrimidine30 (see Supporting
Information).
a Reagents and conditions: (i) MnO2, CHCl3; (ii) NaH, MOMCl, THF;
(iii) LiAlH4, THF; (iv) DAST, CHCl3; (v) HCl, EtOH.
first the primary amine as a Boc-carbamate (VI), which can be
purified if needed, then releasing the amine just before the
reductive amination step. The synthesis of the intermediate XI
was achieved from the known 5-chloro-2-(methylsulfonyl)-4-
pyrimidinecarboxylic acid IX.20 Thus, a Curtius rearrangement21
from IX afforded the corresponding Boc-protected amine, which
was then N-methylated under basic conditions to give X. The
cyano function was incorporated by applying the same method
as that used for the conversion of IV into III. Some intermedi-
ates of the type VI (R ) H and R1 ) CH3, CH2CH3, and CHO)
were more conveniently accessed by constructing the pyrimidine
ring. To this end, the amidine VII22 was reacted with the
appropriate 1,3-dicarbonyl equivalent23 to deliver either VI (R
) H and R1 ) CH3 or CH2CH3) or VIII (R ) H and R1 )
CHO). The latter (VIII) served also as precursor for the
5-difluoro derivative VI in which R ) H and R1 ) CHF2.
Cleavage of the Boc-protecting group in VI produced the amines
XII, which were engaged in the reductive amination step to
give 9-18, 20, and 21 (Scheme 2).
Reductive amination between aldehydes XVIII or XIX and
amine XIVb led to the final products 3-6. Compound 2
required an extra step (i.e., hydrolysis of the 5-ester group) to
secure the corresponding 5-carboxylic acid (Scheme 4).
Results and Discussion
In humans and animals (rat and dog), the carboxylic acid 2
(Figure 2) constitutes the major metabolite of compound 1.31
Following the research direction defined above, we set out to
disrupt the metabolism of 1 by acting on the pathway(s)
responsible for the oxidation at the 5-methyl group.32 For this

5026 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20
Maurel et al.
Table 1. Activities of Compounds 1-9 and Reference at 5-HT1A Receptors in Vitro
5-HT1A efficacy
5-HT1A affinity
b
c
compd
X
Y
A
R1
CH3
CO2H
CH2OH
CH2F
CHF2
Cl
pKia
pEC50
Emax
1
2
3
4
5
6
7
8
9
CH
CH
CH
CH
CH
CH
N
CH
CH
CH
CH
CH
CH
CH
N
CH
CH
CH
CH
CH
CH
CH
CH
N
9.07 ( 0.05
<5
7.63 ( 0.13
74.7 ( 1.5
8.13 ( 0.01
8.92 ( 0.01
8.42 ( 0.05
9.16 ( 0.06
8.25 ( 0.07
7.18 ( 0.01
8.65 ( 0.11
8.85 ( 0.07
7.52 ( 0.13d
7.93 ( 0.02
8.73 ( 0.04
5.69 ( 0.03
6.64 ( 0.09
6.62 ( 0.02
7.16 ( 0.18
5.10 ( 0.34
80.0 ( 5.7
61.0 ( 5.2
96.4 ( 4.6
80.4 ( 10.1
89.7 ( 0.7
CH3
CH3
CH3
CH
CH
CH
6.41 ( 0.06
7.49 ( 0.02
7.03 ( 0.26
6.20 ( 0.03
6.64 ( 0.14
70.3 ( 1.3
35.9 ( 0.7
6.6 ( 1.6
72.0 ( 2
8-OH-DPAT
PRX-00023
flibanserin
5-HT
a Binding affinity values are expressed as means ( SEM of separate experiments, each performed in triplicate. b Concentration of agonist for 50% of
[35S]GTPγS binding in C6-glial cells expressing human 5-HT1A receptors; -log EC50 (pEC50) values were estimated using the mean values of three separate
c
experiments. Emax refers to maximal agonist effect; results expressed as % relative to 5-HT (100%). d Affinity reported in ref 35b: 17 nM.
Table 2. Induction of Elements of the Behavioral 5-HT Syndrome in
Rats by Selected Ligands
FPTa
ED50
FBPb
ED50
ratio
FBP/FPT
c
c
compd
1
4
5
6
9
0.38 (0.16-0.88)
6.10 (3.40-11.00)
2.50 (0.75-8.50)
3.80 (2.0-7.3)
1.50 (0.81-2.90)
40
1.40 (0.59-3.20)
16.00 (9.00-28.00)
4.20 (1.80-9.90)
4.90 (2.80-8.60)
8.30 (2.00-34.00)
>40
3.7
2.6
1.7
1.3
5.5
Figure 2. The major metabolite of F 13640.
purpose, we relied on two strategies: (1) the oxidation potential
at the 5-methyl group was altered either by substituting hydrogen
atom(s) for fluorine (cf., 4 and 5, Table 1) or by lowering the
electron density in the aromatic ring (cf., diazines 7-9); or (2)
the 5-methyl group was replaced by a non-oxidizable, isosteric
group (e.g., 6).33 Each one of these approaches provided ligands
that exhibited a varying degree of agonism at the 5-HT1A
receptor (e.g., 4, 5, 6, and 9). Of these, however, the pyrimidine
motif proved interesting enough to be investigated further.
Throughout this work, three compounds are used as refer-
ences: (()-8-OH-DPAT, as it is the prototypic 5-HT1A ago-
nist;34 N-{3-(4-(4-cyclohexylmethanesulfonylaminobutyl)piperazin-
1-yl)phenyl}acetamide (PRX-00023), which is claimed to be a
potent, full 5-HT1A agonist;35 and flibanserin, which is reported
as a preferential postsynaptic 5-HT1A agonist,36 although it
interacts also with 5-HT2 and dopaminergic D4 receptors.
All of the ligands examined in Table 1 had nanomolar affinity
for 5-HT1A binding sites, with the exception, however, of
compounds 2 and 8. Of interest, 2, the major metabolite of 1,
does not recognize the 5-HT1A receptor (nor any other monoam-
ine receptors) and, therefore, cannot contribute to the 5-HT1A
profile of 1. In C6-glial cells transfected with h5-HT1A receptors,
all of the derivatives listed in Table 1 exerted partial agonist
activities as evidenced by stimulation of [35S]GTPγS binding.
The magnitude of the responses (Emax) in this recombinant
system was similar for compounds 3, 5-7, and 9 and in the
range of that of 1. Because the potency of 3 and 7 was weak as
compared to that of 1, they were not investigated further. In
fact, the experimental conditions used in this cell-based
functional assay have been optimized to discriminate between
ligands with high intrinsic activity.37 Hence, 8-OH-DPAT was
found to act in this system as a weak agonist with a maximal
effect of only 35% of that of 5-HT. In these conditions, PRX-
00023 hardly stimulated the 5-HT1A receptor (6% relative to
8-OH-DPAT
PRX-00023
flibanserind
>40
>40
>40
20 (10-40)
a Forepaw treading induction scored at 60 min after treatment po. b Flat
body posture scored at 60 min after treatment po. c ED50 effective dose
values are mg/kg (95% confidence limit). d Observed 60 min after ip
administration.
5-HT), whereas flibanserin had a modest affinity but achieved
marked efficacy at the 5-HT1A receptor. Compounds 4, 5, 6,
and 9, which displayed in vitro characteristics similar to that of
1, were examined in vivo.
Direct and indirect 5-HT receptor agonists produce postural
movements in na¨ıve rats such as forepaw treading (FPT) and
flat body posture (FBP).38 These paradigms were used here38b
as the first-line screen for a central 5-HT1A agonistic activity.
Thus, upon acute, oral administration in rats, compounds 1, 4-6,
and 9 (Table 2) produced dose-dependent FPT and FBP,
indicating that they exerted 5-HT1A agonist action in vivo. There
was no simple correlation between affinity, functional activity,
and occurrence of these behavioral signs in rats. The results
shown in Table 2 revealed that the potency order of the ligands
was the same for both signs, except for compound 9. The latter
caused FPT in all animals from 10 mg/kg upward; in contrast,
FBP occurred in less than 60% of the animals, and the dose-
response curve was markedly shifted rightward. Thus, even
though 9 induces both stereotypic behaviors (FPT and FBP),
the responses were sign-dependent and the separation between
those was larger with 9 than with the other substrates (4-6).
In comparison, 8-OH-DPAT and flibanserin (even given ip)
performed poorly on these markers; flibanserin was also the
only agent for which FBP preceded FPT. Further, PRX-00023
evoked none of the elements of the 5-HT syndrome in rats
despite its alleged oral bioavailability.35b

5-HT1A Agonists for Antidepressant Treatment
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20 5027
Table 3. Activities of Compounds 10-21 at 5-HT1A Receptors in Vitro
5-HT1A efficacy
5-HT1A affinity
b
c
compd
R1
R2
pKia
pEC50
Emax
10
11
12
13
14
15
16
17
18
19
20
21
H
H
H
H
H
H
H
7.02 ( 0.03
6.66 ( 0.03
6.93 ( 0.16
7.25 ( 0.02
7.60 ( 0.10
8.47 ( 0.06
8.19 ( 0.01
8.32 ( 0.01
8.64 ( 0.03
9.37 ( 0.12
8.76 ( 0.10
9.13 ( 0.08
CO2CH3
CH2OH
CH2CH3
CHF2
Cl
H
H
H
H
5.44 ( 0.45
6.81 ( 0.04
6.35 ( 0.01
6.51 ( 0.03
7.05 ( 0.05
8.00 ( 0.03
7.07 ( 0.02
7.37 ( 0.03
82.5 ( 4.4
76.0 ( 2.7
75.5 ( 2.5
58.8 ( 2.6
63.4 ( 3.3
79.0 ( 0.1
64.1 ( 4.1
71.8 ( 2.2
CH3O
CH3NH
(CH3)2N
2-furyl
CH3NH
CH3NH
CH3
Cl
a Binding affinity values are expressed as means ( SEM of separate experiments, each performed in triplicate. b Concentration of agonist for 50% of
[35S]GTPγS binding in C6-glial cells expressing human 5-HT1A receptors; -log EC50 (pEC50) values were estimated using the mean values of three separate
c
experiments. Emax refers to maximal agonist effect; results expressed as % relative to 5-HT (100%).
Table 4. Antidepressant-like Activity and Elements of the 5-HT Syndrome in Rats of Selected Ligands
forced swimming testa
potency
ED50c (mg/kg)
efficacy
FPTb
ED50 (mg/kg)
compd
MSDd
MEDe
max effetf
1
4
9
15
19
21
0.06 (0.02-0.10)
0.17 (0.08-0.36)
0.05 (0.02-0.09)
0.03 (0.01-0.11)
>2.5
0.08
0.31
0.08
0.04
5
10
5
100
95
100
99
0.38g
6.10g
1.50g
2.8 (0.72-11)
5.0 (2.5-10.0)
5.0 (2.5-10.0)
40
0.32 (0.16-0.61)
0.75 (0.08-6.8)
>40
0.63
2.5
2.5
10
100
80
8-OH-DPAT
PRX-00023
flibanserin
>320
NSh
160
32
a All of the compounds were administered po 60 min before test. b Forepaw treading induction scored at 60 min after treatment po. c Based on the number
of animals affected. d Minimum significant dose (mg/kg). e The first dose at which the effects are maximal. f Percentage of animals affected. g See Table 2.
h NS: not significant.
It is clear that the ability to trigger FPT and FBP differs
substantially between 5-HT1A agonists, even between close
structural analogues. This aspect will be revisited in the context
of the forced swimming test (vide infra).
Compound 9 displayed marked binding affinity, high selec-
tivity,39 and functional activity at 5-HT1A receptor subtype. It
behaves as an agonist in vivo and was orally bioavailable in
rats.40 To gain further insight into the pharmacological conse-
quences of the change from a pyridine to a pyrimidine motif,
we initiated a limited SAR program around 9, guided by the
knowledge garnered in previous studies.1 The results obtained
are summarized in Table 3.
No doubt, the presence of a methyl group at C-5 on the
pyrimidine ring is beneficial (9 vs 10). In addition, any
substituent larger than a methyl (e.g., 9) or a chlorine (e.g., 15)
at the 5-position ruined the binding at the 5-HT1A sites (e.g.,
12-14). Such a trend, although reminiscent of that seen in the
pyridine series, was even clear-cut in the case of a ligand
containing a pyrimidine pharmacophore (e.g., 3, 5, Table 1 and
12, 14, Table 3), possibly underlying subtle differences in the
way both types of molecules dock in the active site of the
receptor. In contrast to the tight constraints imposed at C-5 on
the pyrimidine, the 6-position was quite tolerant to chemical
modulations, and groups of various sizes and electronic proper-
ties were all accommodated (e.g., 16-19 vs 10). Of these, the
2-furyl (19) was optimal. Combining a 5-methyl or -chlorine
with a group at C-6 improved affinity (20 and 21 vs 17).
Subsequently, however, such combinations were shown to have
a negative impact in the forced swimming test (vide infra). We
assumed that the contribution of the 6-substituent was mostly
lipophilic and, therefore, not productive in a drug-like sense.
Overall, the 5-HT1A receptor appeared to be less permissive
toward ligands containing a pyrimidine than toward those having
a pyridine pharmacophore.
With a set of promising compounds in hand, we probed their
antidepressant potential by means of the forced swimming test
(FST). The latter is the most widely used and the best-established
model of antidepressant-like efficacy.41 Further, activity in this
procedure is not restricted to a particular pharmacological
mechanism.42 FST data obtained on selected ligands are
summarized in Table 4. For comparative purposes, FPT data
are also included as it is the most sensitive component of the
5-HT syndrome in rats.
Upon single, oral administration, compounds 1, 4, 9, 15, and
21 dose-dependently (0.01-40 mg/kg) and completely decrease
immobility time relative to vehicle.43 Remarkably, the activity
of 9 and 15 in the FST was superimposable to that of 1 despite
their significant affinity and potency differences in vitro (Tables
1 and 3) and in vivo as regards syndrome-induction (Tables 2
and 4). Thus, 9 and 15 significantly shortened the latency to

5028 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20
Maurel et al.
swim at doses as low as 0.05 and 0.03 mg/kg, respectively,
whereas locomotor activation (FPT) only appeared at doses 30-
to 70-fold higher. Incidentally, it can be appreciated that, under
the experimental conditions used,42c motor effects did not
interfere with the response measured in the FST. What began
as a “routine” effort, merely directed toward finding a back-up
for compound 1, took an unforeseen and exciting turn.
marked affinity, specificity, and high efficacy for the 5-HT1A
receptor subtype. In vivo profiling of 9 revealed pronounced
differences in the 5-HT1A-mediated responses. Thus, 9 was
extremely potent in the FST, whereas its ability to produce
stereotypic behaviors (FPT and FBP) was much weaker. The
molecular mechanism(s) at the origin of this “FST-selectivity”
has not been elucidated yet, but it would appear that 9 stimulates
preferentially certain (postsynaptic) populations of 5-HT1A
receptors.
Conceptually, all anti-depressant medication that has been
(at least partly) successful addressed multiple targets, generally
through increased levels of endogenous neurotransmitters (e.g.,
SSRIs, SNRIs, TCAs, MAOIs).49 In sharp contrast, the present
work reinforces our previous finding12 that, at least in animals,
5-HT1A activation can produce maximally effective antidepres-
sant-like activity, provided the intrinsic activity of the agonist
is sufficient. In addition, we disclose preliminary evidence that
5-HT1A activation can be channeled toward the receptor
population(s) controlling the desired pharmacological effects,
promoting thus the idea that 5-HT1A agonist-directed trafficking
may be exploited.50 Further research is clearly needed to
progress on this topic.
Obviously, processes beyond G-protein signaling take part
in the control of in vivo outcomes. For instance, antidepressant-
like activity was favored over other behavioral effects with
compounds 9 and 15 but not with relatives (e.g., 19). Also, the
ED50 in FPT test increases with 9 and 15, whereas that in FST
does not (Table 4). This undermines any pharmacokinetic
contribution to this “in vivo selectivity” because it is unlikely
that these compounds distributed differently into the brain so
as to activate neural pathways unevenly. In contrast, the erosion
of all in vivo responses seen with 4 would fit in with a decline
in brain exposure as compared to that of 1.
In agreement with published data, flibanserin (even given ip)
was found inactive in the FST.44 Likewise, PRX-00023 exhibited
no antidepressant-like activity in that procedure, a result
consistent with the in vitro data we generated (cf., Table 1). It
thus would seem that the agonist properties of PRX-00023 at
the 5-HT1A receptor subtype have been largely overestimated.35
Indeed, we hope that the benefits seen in animals with
compound 9 will translate into clinical advantages. From FST
data, it is expected that 9 will be more effective51a and faster
acting51b than available antidepressant drugs. From syndrome
data, it is expected that 9 will be better tolerated insofar as
symptoms mediated by 5-HT1A receptors not implicated in the
antidepressant action might be spared at antidepressant doses.
The finding that the profiles of 9 and 15 in the FST do not
match those observed in other behavioral tests has several
implications: (1) the separation between antidepressant-like and
other 5-HT1A-mediated effects may (in theory) still be enhanced;
(2) the 5-HT syndrome can no longer be taken as a “quantita-
tive” index of 5-HT1A agonist activity in rats; and (3) further
mechanistic studies on compound 9 are warranted. In any case,
the heterogeneity in the in vivo responses must reflect hetero-
geneous activation of the 5-HT1A receptors regulating these
processes.45 It should be taken into account that, for each
compound, the data were all compared at 60 min after po dosing
to minimize pharmacokinetic influences. Hence, the picture that
begins to emerge is that FPT, FBP, and FST patterns are
controlled by distinct population(s) of 5-HT1A receptors, which,
themselves, are differentially regulated by these ligands.
Experimental Section
Chemistry. Melting points were determined on a Bu¨chi 530
1
melting point apparatus and were not corrected. H NMR spectra
were recorded on a Bruker Avance 400 spectrometer operating at
400 MHz for 1H and 100 MHz for 13C. Chemical shifts are reported
in δ value (ppm) relative to an internal standard of tetramethylsilane.
Infrared (IR) spectra were obtained on a Nicolet FT 510 P spectra
photometer. Microanalyses were obtained on a Fison EA 1108/
CHN analyzer. Mass spectra (TSQ 7000 Finnigan, Thermoelectron
Corp.) were determined by electron spay ionization (ESI); only
100% relative intensity peaks are given. Analytical thin-layer
chromatography was carried out on pre-coated plates (silicagel, 60
F 254 Merck).
General Method for the Reductive Amination between
Amines (XII, XIII) and XIVa. Preparation of 3-Chloro-4-
fluorophenyl-(4-fluoro-4-{[(heteroarylmethyl)-amino]-methyl}-
piperidin-1-yl)-methanone. To a solution of the chlorhydrate of
the amine XII or XIII (1 equiv) in methanol (7 mL per mmol)
were added successively XIVa (1.1 equiv), 1,4-diazabicyclo[2.2.2]-
octane (3 equiv), NaBH3CN (1.15 equiv), and molecular sieves 4
Å. The mixture was stirred at 50 °C for 6 h, and then the solid was
filtered out and the filtrate was concentrated under vacuum. The
residue was taken up in ethyl acetate, washed with water, brine,
dried over Na2SO4, filtered, and concentrated in vacuo.
Although we do not understand the molecular mechanism(s)
involved in the discrepancies highlighted above, this cannot be
explained by the conventional partition between pre- and
postsynaptic 5-HT1A receptors inasmuch as FPT, FBP, and anti-
immobility (FST) are predominantly linked to postsynaptic
actions of the agonist.46 We are currently exploring features
known to modulate signal transduction in GPCRs (e.g., coupling
with different G-proteins, RGS accessory proteins...) and
possibly involved in the original profile of 9.47,48
In a separate set of experiments, the anti-immobility effect
of 9 was blocked by pretreatment of the animals with the
selective 5-HT1A antagonist WAY-100635 (0.63 mg/kg sc),
confirming that the anti-immobility effect is attributable to
5-HT1A activation. Importantly, the antidepressant-like profile
of 9 was maintained upon repeated administrations, demonstrat-
ing that no tachyphilaxis developed to the antidepressant-like
response.
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(6-methylpyridazin-3-
ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (7). Reduc-
tive amination between XIIIa (7.3 g, 59 mmol) and XIVa (15.5 g,
49 mmol) and then purification by flash column chromatography
(silica gel, dichloromethane/methanol/ammonia, 97.5:2.25:0.25)
afforded 3.0 g (16%) of 7 as a yellow solid. Crystallization of 7 as
a chlorhydrate salt gave a white powder, mp ) 205 °C; HPLC
purity 98.3% (eluent, acetonitrile-water-KH2PO4, 250:750:6.8 g,
Conclusions
1
pH 4); IR (KBr) ν 3426, 1630, 1444 cm-1; H NMR (DMSO-d6)
Structural modifications of compound 1 with the objective
of altering its pharmacokinetic properties provided a novel series
of 5-HT1A agonists. SAR study then led us to the selection of
compound 9 as a candidate for development in depression. The
latter exhibited the desired in vitro 5-HT1A characteristics:
δ 1.71-2.02 (m, 4H), 2.69 (s, 3H), 2.96-3.19 (m, 1H), 3.21-
3.62 (m, 2H), 3.38 (d, J ) 19.70 Hz, 2H), 4.36-4.45 (m, 1H),
4.53 (s, 2H), 5.41 (s, 1H), 7.43-7.49 (m, 1H), 7.52 (t, J ) 8.60
Hz, 1H), 7.67 (d, J ) 6.57 Hz, 1H), 8.75 (d, J ) 8.59 Hz, 1H),
7.92 (d, J ) 8.59 Hz, 1H), 10.29 (s, 2H); 13C NMR (DMSO-d6) δ

5-HT1A Agonists for Antidepressant Treatment
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20 5029
20.3, 31.9, 32.2, 37.2, 42.7, 49.4, 52.8 (d, J ) 22.0 Hz), 92.5 (d,
J ) 176.0 Hz), 117.2, 119.8, 127.8, 129.2, 130.8, 131.5, 133.7,
153.7, 157.7 (d, J ) 249.1 Hz), 159.89, 166.19; MS (ESI) m/z )
494 [M+].
none (12). Reductive amination between XIIg (0.6 g, 3.4 mmol)
and XIVa (1.18 g, 3.7 mmol) and then purification by flash column
chromatography (silica gel, dichloromethane/methanol, 97:3) af-
forded 0.44 g (31%) of 12. Crystallization of 12 as a fumarate salt
gave an off-white powder, mp ) 166 °C; HPLC purity 99.8%
(eluent, acetonitrile-water-KH2PO4, 200:800:6.8 g, pH 4); IR
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-methylpyrazin-2-yl-
methyl)-amino]-methyl}-piperidin-1-yl)-methanone (8). Reduc-
tive amination between XIIIb (1.0 g, 8.1 mmol) and XIVa (2.51
g, 8.0 mmol) and then purification by flash column chromatography
(silica gel, dichloromethane/methanol/ammonia, 95:4.5:0.5) afforded
0.47 g (15%) of 8. Crystallization of 8 as a chlorhydrate salt gave
a white powder, mp ) 155 °C; HPLC purity 100% (eluent,
acetonitrile-water-KH2PO4, 250:750:6.8 g, pH 4); IR (KBr) ν
1
(KBr) ν 3083, 1683, 1425, 1274 cm-1; H NMR (DMSO-d6) δ
1.65-1.91 (m, 4H), 2.84 (d, J ) 21 Hz, 2H), 3.07-3.51 (m, 3H),
3.99 (s, 2H), 4.25 (s, 1H), 4.54 (s, 2H), 6.62 (s, 2H), 7.43-7.52
(m, 2H), 7.67 (d, J ) 6.9 Hz, 1H), 8.72 (s, 2H); 13C NMR (DMSO-
d6) δ 31.7, 32.6, 37.8, 43.0, 54.5, 55.5 (d, J ) 21.8 Hz), 58.7, 94.4
(d, J ) 171 Hz), 117.4, 120.0, 128.1, 129.6, 133.6, 134.2, 134.4,
156.1 (2C), 157.9 (d, J ) 247.5 Hz), 165.6, 166.4, 167.0; MS (ESI)
m/z ) 411 [MH+].
1
3426, 1637, 1442 cm-1; H NMR (DMSO-d6) δ 1.69-2.10 (m,
4H), 2.53 (s, 3H), 2.93-3.17 (m, 1H), 3.24 (d, J ) 20.0 Hz, 2H),
3.32 (s, 3H), 3.32-3.50 (m, 1H), 4.33 (s, 2H), 7.43-7.49 (m, 1H),
7.52 (t, J ) 8.8 Hz, 1H), 7.66 (d, J ) 6.8 Hz, 1H), 8.61 (s, 1H),
8.68 (s, 1H), 10.00 (s, 2H); 13C NMR (DMSO-d6) δ 20.9, 31.7,
32.2, 37.2, 42.7, 48.9, 52.6 (d, J ) 22.0 Hz), 92.5 (d, J ) 174.9
Hz), 117.2, 119.8, 127.8, 129.2, 133.7, 143.5, 144.4, 144.5, 153.6,
157.7 (d, J ) 249.2 Hz), 166.7; MS (ESI) m/z ) 394 [M+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-methylpyrimidin-2-
ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (9). Reduc-
tive amination between XIIe (8.21 g, 42.4 mmol) and XIVa (13.29
g, 43.2 mmol) and then purification by flash column chromatog-
raphy (silica gel, dichloromethane/methanol/ammonia, 95:4.5:0.5)
afforded 11.9 g (71%) of 9. Crystallization of 9 as a tosylate salt
gave a white powder, mp ) 195 °C; HPLC purity 99.8% (eluent,
acetonitrile-water-KH2PO4, 300:700:6.8 g, pH 5.5); IR (KBr) ν
3500-3300, 1622 cm-1; 1H NMR (CD3OD) δ 1.75-2.25 (m, 4H),
2.33 (s, 3H), 2.34 (s, 3H), 3.23-3.35 (m, 2H), 3.60-4.52 (m, 2H),
3.52 (d, J ) 19.50 Hz, 2H), 4.51 (s, 2H), 4.84 (s, 2H), 7.19 (d, J
) 7.95 Hz, 2H), 7.34 (dd, J ) 8.60 Hz, 1H), 7.43 (ddd, J ) 8.60,
4.60, 2.05 Hz, 1H), 7.61 (dd, J ) 7.02, 2.05 Hz, 2H), 7.64 (d, J )
7.95 Hz, 2H), 8.65 (s, 2H); 13C NMR (CD3OD) δ 15.4, 21.3, 33.3,
33.5, 38.9, 44.3, 52.4, 55.2 (J ) 20.8 Hz), 93 (J ) 175.8 Hz),
118.3 (J ) 21.8 Hz), 122.4 (J ) 18.2 Hz), 127, 128.8 (J ) 7.9
Hz), 129.8, 130.9, 132.3, 134.2 (J ) 4.2 Hz), 141.7, 143.6, 159.0,
159.4, 160.1 (J ) 251 Hz), 170.0; MS (ESI) m/z ) 395 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(pyrimidin-2-ylmethyl)-
amino]-methyl}-piperidin-1-yl)-methanone (10). Reductive ami-
nation between XIIk52 (0.77 g, 5.29 mmol) and XIVa (1.82 g, 5.79
mmol) and then purification by flash column chromatography (silica
gel, dichloromethane/methanol, 98:2) afforded 0.88 g (44%) of 10.
Crystallization of 10 as a fumarate salt gave an off-white powder,
mp ) 157 °C; HPLC purity 99.8% (eluent, acetonitrile-water-
KH2PO4, 200:800:6.8 g, pH 4); IR (KBr) ν 3500, 1702, 1625, 1287
cm-1; 1H NMR (DMSO-d6) δ 1.64-1.93 (m, 4H), 2.85 (d, J ) 21
Hz, 2H), 3.01-3.41 (m, 3H), 3.99 (s, 2H), 4.25 (s, 1H), 6.62 (s,
2H), 7.41 (t, J ) 4.8 Hz, 1H), 7.45-7.51 (m, 2H), 7.68 (d, J ) 6.4
Hz, 1H), 8.78 (d, J ) 4.8 Hz, 2H); 13C NMR (DMSO-d6) δ 31.8,
32.2, 38.8, 43.1, 55.1, 55.5 (d, J ) 21.6 Hz), 94.5 (d, J ) 170 Hz),
116.9, 119.6, 119.7, 127.6, 129.1, 133.7, 133.8, 157.1 (2C), 157.5
(d, J ) 247 Hz), 166.1, 166.6, 167.8; MS (ESI): m/z ) 381 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-carbomethoxypyri-
midin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (11).
Reductive amination between XIIb (1.12 g, 5.5 mmol) and XIVa
(1.74 g, 5.51 mmol) and then purification by flash column
chromatography (silica gel, dichloromethane/methanol, 97:3) af-
forded 0.24 g (10%) of 11. Crystallization of 11 as a fumarate salt
gave an off-white powder, mp ) 134 °C; HPLC purity 86.8%
(eluent, acetonitrile-water-KH2PO4, 300:700:6.8 g, pH 4); IR
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-ethylpyrimidin-2-yl-
methyl)-amino]-methyl}-piperidin-1-yl)-methanone (13). Reduc-
tive amination between XIIf (0.59 g, 3.4 mmol) and XIVa (1.18
g, 3.7 mmol) and then purification by flash column chromatography
(silica gel, dichloromethane/methanol, 97:3) afforded 0.88 g (63%)
of 13. Crystallization of 13 as a fumarate salt gave a white powder,
mp ) 172 °C; HPLC purity 99.7% (eluent, acetonitrile-water-
KH2PO4, 300:700:6.8 g, pH 4); IR (KBr) ν 2961, 2927, 1629, 1444,
1402 cm-1; 1H NMR (DMSO-d6) δ 1.20 (t, J ) 7.6 Hz, 3H), 1.65-
1.92 (m, 4H), 2.61 (q, J ) 7.6 Hz, 2H), 2.83 (d, J ) 21 Hz, 2H),
3.01-3.41 (m, 3H), 3.95 (s, 2H), 4.25 (s, 1H), 6.62 (s, 2H), 7.43-
7.51 (m, 2H), 7.67 (d, J ) 6.6 Hz, 1H), 8.66 (s, 2H); 13C NMR
(DMSO-d6) δ 14.8, 22.4, 31.9, 32.6, 37.5, 43.0, 54.7, 55.6 (d, J )
21.5 Hz), 94.9 (d, J ) 170.3 Hz), 116.9, 119.6, 127.7, 129.1, 133.8,
134.0, 134.3, 156.3 (2C), 157.5 (d, J ) 247.5 Hz), 165.4, 166.1,
166.6; MS (ESI) m/z ) 409 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-difluoromethylpy-
rimidin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-metha-
none (14). Reductive amination between XIIh (0.38 g, 1.6 mmol)
and XIVa (0.54 g, 1.7 mmol) and then purification by flash column
chromatography (silica gel, dichloromethane/methanol, 98:2) af-
forded 0.26 g (37%) of 14. Crystallization of 14 as a fumarate salt
gave a pale yellow powder, mp ) 142 °C; HPLC purity 98.8%
(eluent, acetonitrile-water-KH2PO4, 250:750:6.8 g, pH 4); IR
1
(KBr) ν 2964, 1620, 1438, 1405 cm-1; H NMR (DMSO-d6) δ
1.65-1.91 (m, 4H), 2.78 (d, J ) 21 Hz, 2H), 2.99-3.51 (m, 6H),
4.03 (s, 2H), 4.25 (s, 1H), 6.62 (s, 2H), 7.20 (t, J ) 55 Hz, 1H),
7.44-7.51 (m, 2H), 7.67 (d, J ) 7 Hz, 1H), 9.01 (s, 2H); 13C NMR
(DMSO-d6) δ 32.0, 32.4, 37.7, 43.3, 55.1, 55.5 (d, J ) 21.8 Hz),
95.1 (d, J ) 170 Hz), 112.9 (t, J ) 235 Hz), 116.9, 119.6, 125.7,
127.7 (2C), 129.1, 133.8, 133.9, 155 (t, J ) 6 Hz), 157.5 (d, J )
247.7 Hz), 166.0, 166.6, 170.5; MS (ESI) m/z ) 431 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-chloropyrimidin-2-
ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (15). Re-
ductive amination between XIIm53 (1.62 g, 8.0 mmol) and XIVa
(2.65 g, 8.4 mmol) and then purification by flash column chroma-
tography (silica gel, dichloromethane/methanol, 98:2) afforded 1.8
g (54%) of 15. Crystallization of 15 as a fumarate salt gave an
off-white powder, mp ) 152 °C; HPLC purity 99.4% (eluent,
acetonitrile-water-KH2PO4, 250:750:6.8 g, pH 4); IR (KBr) ν
3123, 1696, 1623, 1430 cm-1; 1H NMR (DMSO-d6) δ 1.64-1.91
(m, 4H), 2.65 (d, J ) 20.8 Hz, 2H), 3.09-3.50 (m, 3H), 4.01 (s,
2H), 4.25 (s, 1H), 6.63 (s, 2H), 7.42-7.52 (m, 2H), 7.67 (d, J )
7.2 Hz, 1H), 8.92 (s, 2H), 9.20 (s, 2H); 13C NMR (DMSO-d6) δ
31.8, 32.4, 39.3, 43.0, 54.4, 55.4 (d, J ) 21.6 Hz), 94.4 (d, J )
170.3 Hz), 116.9, 119.6, 127.6, 127.7, 128.8, 129.1, 133.8, 156.2
(2C), 157.5 (d, J ) 247.3 Hz), 165.9, 166.2, 166.6; MS (ESI) m/z
) 415 [M+].
1
(KBr) ν 2987, 1716, 1685, 1436, 1277 cm-1; H NMR (DMSO-
d6) δ 1.67-1.90 (m, 4H), 2.79 (d, J ) 22 Hz, 2H), 3.10-3.58 (m,
3H), 3.91 (s, 3H), 4.08 (s, 2H), 4.25 (s, 1H), 6.62 (s, 2H), 7.31-
7.49 (m, 2H), 7.61-7.67 (m, 1H), 9.21 (s, 1H); 13C NMR (DMSO-
d6) δ 30.6, 32.4, 37.5, 42.4, 52.5, 55.2, 55.5 (d, J ) 21.6 Hz), 95.1
(d, J ) 167 Hz), 116.9, 119.6, 121.9, 127.6, 129.0, 133.8, 134.0,
157.5 (2C), 157.5 (d, J ) 247.4 Hz), 163.8, 166.0, 166.6, 171.8;
MS (ESI) m/z ) 439 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(4-methoxypyrimidin-
2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (16). Re-
ductive amination between XIIn54 (0.73 g, 5.21 mmol) and XIVa
(2.0 g, 6.3 mmol) and then purification by flash column chroma-
tography (silica gel, dichloromethane/methanol, 95:5) afforded 1.1
g (51%) of 16. Crystallization of 16 as a fumarate salt gave a white
powder, mp ) 149 °C; HPLC purity 98.6% (eluent, acetonitrile-
water-KH2PO4, 250:750:6.8 g, pH 4); IR (KBr) ν 3054, 2077,
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-hydroxymethylpy-
rimidin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-metha-
1
1710, 1629, 1583 cm-1; H NMR (DMSO-d6) δ 1.68-1.99 (m,

5030 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20
Maurel et al.
5H), 2.87 (d, J ) 20.8 Hz, 2H), 3.08-3.91 (m, 3H), 3.88 (s, 2H),
3.92 (s, 3H), 4.27 (s, 1H), 6.62 (s, 2H), 6.82 (d, J ) 5.6 Hz, 1H),
7.45-7.52 (m, 2H), 7.67 (d, J ) 6.7 Hz, 1H), 8.48 (d, J ) 5.6 Hz,
1H), 9.30 (s, 2H); 13C NMR (DMSO-d6) δ 32.4, 32.9, 38.0, 43.5,
53.9, 55.1, 56.0 (d, J ) 21.6 Hz), 94.8 (d, J ) 170.4 Hz), 106.4,
117.4, 120.0, 128.1, 129.6, 134.2, 134.5, 157.9 (d, J ) 247.5 Hz),
157.9, 166.6, 167.0, 168.2, 169.4; MS (ESI) m/z ) 411 [MH+].
157.8 (d, J ) 248 Hz), 157.9, 164.5, 166.1, 166.6; MS (ESI) m/z
) 444 [M - H+].
General Method for the Reductive Amination between
Aldehydes (XVIII, XIX) and XIVb. To a solution of XIVb (1
equiv) and XVIII or XIX (1 equiv) in dichloroethane (14 mL per
mmol of XIVb) was added sodium triacetoxyborohydride (1.5
equiv), and the mixture was stirred at room temperature. Water
was added, the aqueous was extracted with dichloromethane, and
the combined organic layer was washed with brine, dried over Na2-
SO4, filtered, and concentrated in vacuo.
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(4-methylaminopyri-
midin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (17).
Reductive amination between XIIc (0.55 g, 3.98 mmol) and XIVa
(1.5 g, 4.8 mmol) and then purification by flash column chroma-
tography (silica gel, dichloromethane/methanol/ammonia, 90:9:1)
afforded 0.16 g (10%) of 17. Crystallization of 17 as an oxalate
salt gave a white powder, mp ) 180 °C; HPLC purity 98.8%
(eluent, acetonitrile-water-KH2PO4, 150:850:6.8 g, pH 2.5); IR
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(4-fur-2-yl-pyrimidin-
2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (19). Re-
ductive amination between XIX (0.8 g, 4.6 mmol) and XIVb (1.33
g, 4.6 mmol) and then purification by flash column chromatography
(silica gel, dichloromethane/methanol, 95:5) afforded 1.98 g (96%)
of 19. Crystallization of 19 as a hemi-fumarate salt gave a white
powder, mp ) 122 °C; HPLC purity 98.9% (eluent, acetonitrile-
1
(KBr) ν 3023, 1653, 1603, 1464 cm-1; H NMR (DMSO-d6) δ
1.52-2.03 (m, 4H), 2.85 (s, 3H), 2.98-3.45 (m, 3H), 3.28 (d, J )
20 Hz, 2H), 4.11 (s, 2H), 4.30 (s, 1H), 6.45 (d, J ) 6 Hz, 1H),
7.32-7.54 (m, 2H), 7.68 (d, J ) 7 Hz, 1H), 7.82 (s, 1H), 8.02 (s,
1H), 9.70 (s, 2H); 13C NMR (DMSO-d6) δ 31.3, 32.0, 36.1, 38.8,
42.7, 51.3, 53.6 (d, J ) 21.5 Hz), 92.7 (d, J ) 175 Hz), 103.6,
117.1, 119.6, 127.7, 129.1, 133.6, 156.3, 159.0 (d, J ) 249 Hz),
162.9, 163.6, 166.6; MS (ESI) m/z ) 410 [MH+].
water-KH2PO4, 300:700:6.8 g, pH 4); IR (KBr) ν 1631, 1445 cm-1
;
1H NMR (DMSO-d6) δ 1.70-1.93 (m, 4H), 2.85 (d, J ) 21.2 Hz,
2H), 3.08-3.51 (m, 3H), 3.97 (s, 2H), 4.25 (s, 1H), 6.61 (s, 1H),
6.75 (m, 1H), 7.42-7.51 (m, 3H), 7.62 (d, J ) 5.2 Hz, 1H), 7.67
(dd, J ) 7.2 Hz, 1H), 7.99 (s, 1H), 8.79 (d, J ) 5.3 Hz, 1H); 13C
NMR (DMSO-d6) δ 32.4, 32.9, 39.2; 45.0, 55.5, 56.0 (d, J ) 21.5
Hz), 95.4 (d, J ) 170.1), 113.0, 113.4, 117.3, 117.5, 120.2, 128.0,
129.6, 134.2, 134.5, 146.7, 151.3, 154.8, 157.9 (d, J ) 248 Hz),
158.4, 166.6, 167.0, 168.6; MS (ESI) m/z ) 447 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(4-dimethylaminopy-
rimidin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-metha-
none (18). Reductive amination between XIId (0.74 g, 4.8 mmol)
and XIVa (1.8 g, 5.7 mmol) and then purification by flash column
chromatography (silica gel, dichloromethane/methanol, 90:10)
afforded 0.66 g (33%) of 18. Crystallization of 18 as a fumarate
salt gave a white powder, mp ) 160 °C; HPLC purity 99.4%
(eluent, acetonitrile-water-KH2PO4, 250:750:6.8 g, pH 4); IR
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-carboxy-pyridin-2-
ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (2). Reduc-
tive amination between XVIIIa (0.9 g, 5.0 mmol) and XIVb (1.45
g, 5.0 mmol) and then purification by flash column chromatography
(silica gel, dichloromethane/methanol, 95:5) afforded 2.1 g (95%)
of the adduct, mp ) 177 °C. The ester was dissolved in a solution
of lithium hydroxide in water (25 mL) and ethanol (30 mL) and
stirred for 3 h at room temperature. Compound 2 was purified by
flash column chromatography (silica gel, dichloromethane/methanol,
75:25) to afford 0.5 g (24%) as a white powder, mp ) 184 °C;
HPLC purity 100% (eluent, acetonitrile-water-KH2PO4, 300:700:
1
(KBr) ν 3013, 1623, 1599, 1259 cm-1; H NMR (DMSO-d6) δ
1.71-1.94 (m, 4H), 2.87 (d, J ) 20.8 Hz, 2H), 3.06 (s, 6H), 3.00-
3.42 (m, 3H), 3.77 (s, 2H), 4.26 (s, 1H), 6.54 (d, J ) 5.2 Hz, 1H),
6.60 (s, 2H), 7.46-7.50 (m, 3H), 7.67, (d, J ) 5.6 Hz, 1H), 8.13
(d, J ) 5.2 Hz, 1H), 8.96 (s, 2H); 13C NMR (DMSO-d6) δ 32.3,
33.0, 36.8 (2C), 37.9, 43.4, 55.0, 55.9 (d, J ) 21.6 Hz), 94.7 (d, J
) 170.6 Hz), 101.4, 117.4, 120.1, 128.1, 129.6, 134.2, 134.6, 155.1,
157.9 (d, J ) 247.6 Hz), 159.2, 161.8, 166.2, 166.7, 167.0; MS
(ESI) m/z ) 424 [MH+].
1
6.8 g, pH 4); IR (KBr) ν 3665, 3350, 3201, 1616 cm-1; H NMR
(DMSO-d6) δ 1.65-1.91 (m, 4H), 2.74 (d, J ) 21.2 Hz, 2H), 3.06-
3.51 (m, 5H), 3.92 (s, 2H), 4.25 (s, 1H), 7.40-7.51 (m, 2H), 7.58
(dd, J ) 8 Hz, 1H), 7.66 (dd, J ) 6.4 Hz, 1H), 8.23 (d, J ) 6 Hz,
1H), 8.98 (s, 1H); 13C NMR (DMSO-d6) δ 32.1, 32.5, 37.6, 43.1,
54.7, 55.8 (d, J ) 21.8 Hz), 94.5 (d, J ) 170.1 Hz), 116.9, 119.7,
120.9, 127.7, 128.7, 129.2, 133.8, 137.0, 149.7, 157.6 (d, J ) 247.5
Hz), 162.6, 166.7, 167.2; MS (ESI) m/z ) 424 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(4-methylamino-5-me-
thylpyrimidin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-meth-
anone (20). Reductive amination between XIIa (0.63 g, 4.1 mmol)
and XIVa (1.45 g, 4.6 mmol) and then purification by flash column
chromatography (silica gel, dichloromethane/methanol, 90:10)
afforded 1.15 g (63%) of 20. Crystallization of 20 as a fumarate
salt gave a white powder, mp ) 157 °C; HPLC purity 99.1%
(eluent, acetonitrile-water-KH2PO4, 250:750:6.8 g, pH 4); 1H
NMR (DMSO-d6) δ 1.53-1.81 (m, 4H), 1.98 (s, 3H), 2.85 (d, J )
20.5 Hz, 2H), 2.86 (d, J ) 4.4 Hz, 3H), 3.02-3.41 (m, 4H), 3.73
(s, 2H), 4.25 (s, 1H), 6.61 (s, 4H), 6.87 (m, 1H), 7.42-7.51 (m,
2H), 7.67 (d, J ) 6.6 Hz, 1H), 7.84 (s, 1H); 13C NMR (DMSO-d6)
δ 13.3, 27.3, 32.0, 33.0, 37.2, 42.9, 54.5, 55.7 (d, J ) 21 Hz), 94.1
(d, J ) 171 Hz), 111.7, 116.9, 117.1, 127.7, 129.1, 133.8, 134.1,
151.8, 156.3, 158.8, 162.5 (d, J ) 270 Hz), 166.1, 166.6; MS (ESI)
m/z ) 424 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(4-methylamino-5-chlo-
ropyrimidin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-metha-
none (21). Reductive amination between XIIj (0.8 g, 4.6 mmol)
and XIVa (1.54 g, 4.9 mmol) and then purification by flash column
chromatography (silica gel, dichloromethane/methanol, 95:5) af-
forded 0.30 g (15%) of 21. Crystallization of 21 as a di-fumarate
salt gave a white powder, mp ) 92 °C; HPLC purity 90.0% (eluent,
acetonitrile-water-KH2PO4, 300:700:6.8 g, pH 4); IR (KBr) ν
3084, 1600, 1430, 1277 cm-1; 1H NMR (DMSO-d6) δ 1.67-1.91
(m, 4H), 2.90 (d, J ) 20 Hz, 2H), 2.91 (s, 3H), 3.07-3.41 (m,
3H), 3.81 (s, 2H), 4.26 (s, 1H), 6.61 (s, 4H), 7.43-7.51 (m, 3H),
7.67 (d, J ) 7 Hz 1H), 8.15 (s, 1H); 13C NMR (DMSO-d6) δ 27.4,
32.4, 32.5, 38.6, 42.9, 53.9, 55.2 (d, J ) 21.3 Hz), 94.2 (d, J )
170.5 Hz), 111.9, 117.0, 119.6, 127.6, 129.1, 133.8, 134.0, 151.6,
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-hydroxymethylpy-
ridin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (3).
Reductive amination between XVIIIc (0.28 g, 2.0 mmol) and XIVb
(0.59 g, 2.0 mmol) and then purification by flash column chroma-
tography (silica gel, dichloromethane/methanol, 95:5) afforded 0.47
g (57%) of 3. Crystallization of 3 as a fumarate salt gave a white
powder, mp ) 167 °C; HPLC purity 99.7% (eluent, acetonitrile-
water-KH2PO4, 200:800:6.8 g, pH 4); IR (KBr) ν 3350, 1697,
1616, 1263 cm-1; 1H NMR (DMSO-d6) δ 1.66-1.99 (m, 4H), 2.79
(d, J ) 20.8 Hz, 2H), 3.06-3.47 (m, 3H), 3.91 (s, 2H), 4.27 (s,
1H), 4.52 (s, 1H), 6.61 (s, 2H), 7.40-7.51 (m, 3H), 7.67 (d, J )
7 Hz, 1H), 7.70 (d, J ) 8 Hz, 1H), 8.46 (s, 1H); 13C NMR (DMSO-
d6) δ 32.4, 32.8, 39.2, 44.0, 54.4, 55.7 (d, J ) 22 Hz), 60.8, 95.1
(d, J ) 170.6 Hz), 117.3, 120.1, 121.9, 128.1, 129.6, 134.2, 134.3,
135.4, 136.3, 147.7, 157.7, 157.9 (d, J ) 247.5 Hz), 166.7, 167.0;
MS (ESI) m/z ) 410 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-fluoromethylpyridin-
2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (4). Re-
ductive amination between XVIIId (0.39 g, 2.8 mmol) and XIVb
(0.82 g, 2.8 mmol) and then purification by flash column chroma-
tography (silica gel, dichloromethane/methanol, 95:5) afforded 0.82
g (71%) of 4. Crystallization of 4 as a fumarate salt gave a white
powder, mp ) 160 °C; HPLC purity 99.9% (eluent, acetonitrile-
water-KH2PO4, 300:700:6.8 g, pH 4); IR (KBr) ν 2964, 1622, 1432
1
cm-1; H NMR (DMSO-d6) δ 1.56-2.04 (m, 4H), 2.75 (d, J )
20.8 Hz, 2H), 3.06-3.26 (m, 3H), 3.90 (s, 2H), 4.26 (s, 1H), 5.46

5-HT1A Agonists for Antidepressant Treatment
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20 5031
(d, J ) 48 Hz, 2H), 6.62, (s, 2H), 7.42-7.52 (m, 3H), 7.66 (dd, J
) 8.8 Hz, 1H), 7.85 (d, J ) 8 Hz, 1H), 8.57 (s, 1H); 13C NMR
(DMSO-d6) δ 32.3, 32.9, 37.9, 43.5, 54.6, 55.5 (d, J ) 21.7 Hz),
81.9 (J ) 160 Hz), 94.8 (d, J ) 170.4 Hz), 116.9, 119.6, 121.6,
127.7, 129.7, 133.7, 133.8, 136.5, 148.4, 157.5 (d, J ) 247 Hz),
160.1, 166.1, 166.6; MS (ESI) m/z ) 411 [M+].
an additional period of 30 min. The reactions were stopped by
adding 3 mL of ice-cold 20 mM HEPES containing 3 mM MgCl2
and rapid filtration over Whatman GF/B glass fiber filters using a
Brandel harvester. The filters were rinsed three additional times
with 3 mL of HEPES buffer, placed in scintillation vials, and the
radioactivity was extracted in 4 mL of Emulsifier-Safe. Nonspecific
binding was determined in the presence of 10 µM unlabeled GTPγS.
Maximal stimulation of [35S]GTPγS binding was defined in the
presence of 10 µM 5-HT. Bovine serum albumin was used as a
standard. Each compound was tested at six concentrations, and [35S]-
GTPγS binding values were expressed as a percentage of the
maximal response obtained with 10 µM 5-HT. Data were estimated
by means of nonlinear regression (sigmoidal model with unit slope;
Graphpad Prism). Emax value differences were analyzed statistically
by a one-way analysis of variance followed by paired comparisons
by Newman-Keuls tests; P values <0.05 were considered statisti-
cally significant.
Elements of the 5-HT Syndrome. The methods used were
similar to those described previously.38b,57 Behavioral observations
were made at two time points, from 10 to 20 min and from 55 to
65 min post-administration po. Four animals were observed
individually during each 10 min period; the rats were successively
observed during a 10 s period (i.e., one animal every 15 s) for the
presence (1) or absence (0) of forepaw treading (FPT). The behavior
was considered present if the animal showed uninterrupted sign
for at least 3 s. This cycle was repeated 10 times during a 10 min
period; thus, the incidence of a particular behavior could vary from
0 to 10 for any observation period. Flat body posture (FBP) was
scored present (1) if it occurred during the entire observation period,
otherwise, the score was 0. Compounds were dissolved in H2O and
administered in a volume of 10 mL/kg and tested in five animals
per dose. Doses refer to the weight of the free base and the order
of treatment with doses was unsystematic. Dose-response functions
were determined from the percentage of rats showing FBP, and
the percentage of rats showing FPT scores of 1 or more. These
criteria were based upon the incidence of each particular behavior
observed in control animals treated with saline. ED50 values and
their 95% confidence limits were calculated with the Litchfield and
Wilcoxon probit analysis.58
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-difluoromethylpy-
ridin-2-ylmethyl)-amino]-methyl}-piperidin-1-yl)-methanone (5).
Reductive amination between XVIIIb (0.69 g, 4.4 mmol) and XIVb
(1.28 g, 4.4 mmol) and then purification by flash column chroma-
tography (silica gel, dichloromethane/acetonitrile, 70:30) afforded
1.67 g (85%) of 5. Crystallization of 5 as a fumarate salt gave a
white powder, mp ) 164 °C; HPLC purity 99.9% (eluent,
acetonitrile-water-KH2PO4, 300:700:6.8 g, pH 4); IR (KBr) ν
3050, 2964, 1629, 1434 cm-1; 1H NMR (DMSO-d6) δ 1.66-1.95
(m, 4H), 2.77 (d, J ) 20.8 Hz), 2.81-3.42 (m, 3H), 3.96 (s, 2H),
4.28 (s, 1H), 6.62 (s, 2H), 7.14 (t, J ) 55.2 Hz), 7.43-7.51 (m,
2H), 7.62 (d, J ) 8 Hz, 1H), 7.67 (d, J ) 7 Hz, 1H), 7.99 (d, J )
8 Hz, 1H), 8.72 (s, 1H), 9.30 (s, 2H); 13C NMR (DMSO-d6) δ
32.4, 32.8, 37.9, 43.5, 54.7, 55.9 (d, J ) 21.7 Hz), 95.2 (d, J )
170.3 Hz), 114.4 (t, J ) 234.3 Hz), 117.4, 120.1, 122.2, 128.1,
128.4, 128.6, 129.6, 134.2, 134.4, 146.6, 157.9 (d, J ) 247.4 Hz),
163.0, 166.5, 167.0; MS (ESI) m/z ) 430 [MH+].
3-Chloro-4-fluorophenyl-(4-fluoro-4-{[(5-chloropyridin-2-yl-
methyl)-amino]-methyl}-piperidin-1-yl)-methanone (6). Reduc-
tive amination between XVIIIe (0.2 g, 1.4 mmol) and XIVb (0.4
g, 1.4 mmol) and then purification by flash column chromatography
(silica gel, dichloromethane/methanol, 95:5) afforded 0.30 g (51%)
of 6. Crystallization of 6 as a fumarate salt gave a white powder,
mp ) 175 °C; HPLC purity 99.8% (eluent, acetonitrile-water-
KH2PO4, 300:700:6.8 g, pH 4); IR (KBr) ν 3052, 1621, 1431 cm-1
;
1H NMR (DMSO-d6) δ 1.65-1.92 (m, 4H), 2.60 (d, J ) 21 Hz,
2H), 2.75-3.39 (m, 4H), 3.85 (s, 2H), 4.25 (s, 1H), 6.62 (s, 2H),
7.43-7.52 (m, 3H), 7.66 (d, J ) 7 Hz, 1H), 7.89 (dd, J ) 8.4, 2.4
Hz, 1H), 8.54 (s, 1H), 9.58 (s, 2H); 13C NMR (DMSO-d6) δ 32.1,
33.3, 38.0, 42.8, 53.9, 55.5 (d, J ) 21.6 Hz), 94.5 (d, J ) 170 Hz),
116.9, 120.2, 123.1, 127.6, 129.0, 129.1, 133.8, 133.9, 136.2, 147.1,
157.3 (d, J ) 250 Hz), 158.7, 166.0, 166.6; MS (ESI) m/z ) 414
[M+].
Radioligand Binding at the 5-HT1A Receptor. The membrane
preparations (rat cortex 10 mg/mL) and binding assays were
performed as described previously.55 At the end of the incubation
period (30 min at 23 °C, Tris-HCl buffer), the tube contents were
filtered under vacuum through GF/B filters with two 5 mL washes
of Tris-HCl buffer (50 mM, pH 7.4 at 25 °C). The radioactivity
retained on the filters was measured by scintillation spectroscopy
in 4 mL of scintillation fluid (Emulsifier safe, Packard). The results
of displacement experiments were analyzed using the nonlinear
regression program KELL RADLIG version 6 (Biosoft, Cambridge,
UK),56 and pKi values are given as means ( SEM of three
experiments, each comprising 6-7 concentrations differing by one
log unit interval. The -log of Ki values (pKi) were calculated from
the Cheng-Prusoff equation Ki ) IC50/(1 + [radioligand]/Kd). The
Kd value of 8-OH-DPAT is 3.1 nM. [3H]8-OH-DPAT (0.2 nM,
TRK.850: 160-240 Ci/mmol) was purchased from Amersham;
nonspecific binding was determined with 5-HT (10 µM) purchased
from Sigma.
[35S]GTPγS Binding in C6 Glial Cells. This was adapted from
the procedure described in detail by Koek.12 C6-glial cells were
collected in phosphate-buffered-saline (pH 7.4) and centrifuged for
20 min at 48 000g. The pellet was homogenized with a Polytron
in 20 mM HEPES containing 10 mM EDTA (pH 7.4) and
centrifuged for 10 min. The resulting pellet was washed twice in
10 mM HEPES (pH 7.4) containing 0.1 mM EDTA. The pellet
was stored at -80 °C in fractions of 600-750 µg of protein. The
pellet was thawed, diluted 20 times in 20 mM HEPES supplemented
with the 30 µM GDP, 100 mM NaC1, 3 mM MgC12, and 0.2 mM
ascorbic acid. Incubation mixtures were prepared in glass tubes
and consisted of 0.4 mL of membrane preparation (16-38 µg of
protein) and 0.05 mL of compound. After an incubation period of
30 min at 25 °C, 0.05 mL of [35S]GTPγS (500 pM) was added for
Forced Swimming Test. This was adapted from the procedure
described by Porsolt.42b A rat was placed in a Plexiglas cylinder
(height 45 cm, diameter 20 cm) containing 17 cm of water (at 25
°C) for 15 min on the first day of the experiment, and placed again
24 h later for 5 min. The duration of immobility during the 5 min
period was measured by an observer who was unaware of the
treatment conditions. Each animal was treated po (administration
volume: 10 mL/kg) with vehicle immediately after the first session,
and with a dose of the compound (n ) 5-12 per dose), or its vehicle
control, 1 h before the second session. Drug effects on the median
immobility time were analyzed statistically by means of a Kruskal-
Wallis nonparametric analysis of variance, followed, where ap-
propriate, by comparisons using Dunns test. Dose-response
functions were determined from the percentage of rats with
immobility times less than 138 s (a criterion based on the immobility
times obtained in 1500 controls over more than 10 years), and ED50
values and their 95% confidence limits were calculated with the
Litchfield and Wilcoxon probit analysis.58
Acknowledgment. We wish to thank Drs. M. B. Assie´, A.
Auclair, L. Bardin, C. Cosi, R. Depoortere, J. C. Martel, Ms.
C. Piedecoq, Mr. S. Brunel, and J. Lorenzio who contributed
to this project. We thank Dr. J. P. Ribet and Mr. P. Zalavari for
analytical support. Mr. L. Petitpas’ assistance with bibliographic
searches is also very much appreciated.
Supporting Information Available: Experimental and analyti-
cal data for intermediates II, III, VI, VIII, X-XIII, and XVI-
XIX. This material is available free of charge via the Internet at
http://pubs.acs.org.

5032 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20
Maurel et al.
(e.g., mean half-life elimination from plasma was 35 h). The aim
was not to improve over any particular PK parameter relative to 1
but rather to obtain a profile somewhat different. Indeed, extending
the patent life of these series was also a major incentive.
(15) Murray, C. J. L.; Lopez, A. D. Alternative projections of mortality
and disability by cause 1990-2020: Global Burden of Disease Study.
Lancet 1997, 349, 1498-1504. Andlin-Sobocki, P.; Jonsson, B.;
Wittchen, H.-U.; Olesen, J. Cost of disorders of the brain in Europe.
Eur. J. Neurol. 2005, 12 (supp. 1), 1-27.
(16) See ref 1b. Also: Vacher, B.; Bonnaud, B.; Koek, W. Pyridin-2-yl-
methylamine derivatives, method of preparing and application as
medicine. Pierre Fabre Medicament WO/9822459, 1998. Maurel, J.
L.; Bonnaud, B.; Ribet, J. P.; Vacher, B. Synthesis method and
intermediates of pyridin-2-yl-methylamine. Pierre Fabre Medicament
WO 02/064585, 2002. Vacher, B.; Bonnaud, B.; Maurel, J. L.;
Colpaert, F. Novel aryl-{4-halo-4-[heteroarylmethylamino)-methyl]-
piperidin-1-yl}-methanone derivatives, methods for production and
use thereof as medicaments. WO 03/106449, 2003.
(17) Methyl 2-(methylsulfonyl)pyrimidine-5-carboxylate, [38275-49-9].
(18) Ohta, K.; Kawachi, E.; Inoue, N.; Fukasawa, H.; Hashimoto, Y.; Itai,
A.; Kagechika, H. Retinoidal pyrimidinecarboxylic acids. Unexpected
diaza-substituent effects in retinobenzoic acids. Chem. Pharm. Bull.
2000, 48, 1504-1513. Liang, Y.-M.; Luo, S.-J.; Zhang, Z.-X.; Ma,
Y.-X. Efficient synthesis of a new pyrimidine derivative. Synth.
Commun. 2002, 32, 153-157.
(19) 2-Chloro-4-(methylamino)pyrimidine, [66131-68-8].
(20) 5-Chloro-2-(methylsulfonyl)-4-pyrimidinecarboxylic acid, [38275-
34-2].
References
(1) (a) Vacher, B.; Bonnaud, B.; Funes, P.; Jubault, N.; Koek, W.; Assie´,
M.-B.; Cosi, C. Design and synthesis of a series of 6-substituted-2-
pyridinylmethylamine derivatives as novel, high-affinity, selective
agonists at 5-HT1A receptors. J. Med. Chem. 1998, 41, 5070-5083.
(b) Vacher, B.; Bonnaud, B.; Funes, P.; Jubault, N.; Koek, W.; Assie´,
M.-B.; Cosi, C.; Kleven, M. Novel derivatives of 2-pyridinemethy-
lamine as selective, potent, and orally active agonists at 5-HT1A
receptors. J. Med. Chem. 1999, 42, 1648-1660.
(2) Colpaert, F. 5-HT1A receptor activation: New molecular and neu-
roadaptative mechanisms of pain relief. Curr. Opin. InVest. Drugs
2006, 7, 40-47.
(3) Colpaert, F. C.; Tarayre, J. P.; Koek, W.; Pauwels, P. J.; Bardin, L.;
Xu, X. J.; Wiesenfeld-Hallin, Z.; Cosi, C.; Carilla-Durand, E.; Assie´,
M. B.; Vacher, B. Large-amplitude 5-HT1A receptor activation: A
new mechanism of profound central analgesia. Neuropharmacology
2002, 43, 945-958. Bardin, L.; Tarayre, J. P.; Malfetes, N.; Koek,
W.; Colpaert, F. C. Profound, non-opioid analgesia produced by the
high efficacy 5-HT1A agonist F 13640 in the formalin model of tonic
nociceptive pain. Pharmacology 2003, 67, 182-194.
(4) Deseure, K. R.; Koek, W.; Colpaert, F. C.; Adriaensen, H. The
5-HT1A receptor agonist F 13640 attenuates mechanical allodynia in
a rat model of trigeminal neuropathic pain. Eur. J. Pharmacol. 2002,
456, 51-57. Kiss, I.; Degryse, A. D.; Bardin, L.; Alvarez Gomez de
Segura, I.; Colpaert, F. C. The novel analgesic, F 13640, produces
intra- and postoperative analgesia in a rat model of surgical pain.
Eur. J. Pharmacol. 2005, 523, 29-39.
(21) Pie, P. A.; Green, T. P.; Hennequin, L. F.; Curwen, J.; Fennell, M.;
Allen, J.; Lambert-van der Brempt, C.; Costello, G. Discovery of a
new class of anilinoquinazoline inhibitors with high affinity and
specificity for the tyrosine kinase domain of c-Src. J. Med. Chem.
2004, 47, 871-887.
(22) Lange, U. E. W.; Scha¨fer, B.; Baucke, D.; Buschmann, E.; Mack,
H. A new mild method for the synthesis of amidines. Tetrahedron
Lett. 1999, 40, 7067-7071. Bose, A. K.; Greer, F.; Gots, J. S.; Price,
C. C. Some derivatives of glycineamidine. J. Org. Chem. 1959, 24,
1309-1313.
(5) Deseure, K. R.; Koek, W.; Adriaensen, H.; Colpaert, F. C. Continuous
administration of the 5-HT1A agonist (3-chloro-4-fluoro-phenyl)-[4-
fluoro-4-{[(5-methyl-pyridin-2-ylmethyl)-amino]-methyl}piperidin-
1-yl]-methanone (F 13640) attenuates allodynia-like behavior in a
rat model of trigeminal neuropathic pain. J. Pharmacol. Exp. Ther.
2003, 306, 505-514.
(6) Colpaert, F. C.; Wu, W. P.; Hao, J. X.; Royer, I.; Sautel, F.;
Wiesenfeld-Hallin, Z.; Xu, X. J. High-efficacy 5-HT1A receptor
activation causes a curative-like action on allodynia in rats with spinal
cord injury. Eur. J. Pharmacol. 2004, 497, 29-33.
(23) Stolle, W. A. W.; Frissen, A. E.; Marcelis, A. T. M.; van der Plas,
H. C. Intramolecular Diels-Alder reactions of pyrimidines and a
computational study toward their structure and reactivity. J. Org.
Chem. 1992, 57, 3000-3007. Gupton, J. T.; Gall, J. E.; Riesinger,
S. W.; Smith, S. Q.; Bevirt, K. M.; Sikorski, J. A.; Dahl, M. L.;
Arnold, Z. An alternative preparation of the 2-dimethylaminometh-
ylene-1,3-bis(dimethylimmonio)propane salt from phosphonoacetic
acids and some applications in heterocyclic synthesis. J. Heterocycl.
Chem. 1991, 28, 1281-1285. Isoda, S.; Yamaguchi, H.; Satoh, Y.;
Miki, T.; Hirata, M. Medicinal chemical studies on antiplasmin drugs.
VI. Aza analogs of 4-aminomethylbenzoic acid. Chem. Pharm. Bull.
1980, 28, 1408-1414.
(7) Butera, J. A. Current and emerging targets to treat neuropathic pain.
J. Med. Chem. 2007, 50, 2543-46. Rice, A. S. C.; Hill, R. G. New
treatments for neuropathic pain. Annu. ReV. Med. 2006, 57, 535-
551. Mico, J. A.; Ardid, D.; Berrocso, E.; Eschalier, A. Antidepres-
sants and pain. Trends Pharmacol. Sci. 2006, 27, 348-354. Davies,
S. L. Rising to the pain challenge. Drug News Perspect. 2006, 19,
653-658.
(8) Pedigo, N. W.; Yamamura, H. I.; Nelson, D. L. Discrimination of
multiple [3H]5-hydroxytryptamine binding sites by the neuroleptic
spiperone in rat brain. J. Neurochem. 1981, 36, 220-226.
(9) Examples of 5-HT1A agonists that underwent clinical trials: zalospirone
[114298-18-9]; umespirone [107736-98-1]; S-15535 [146998-34-7];
repinotan [144980-29-0]; Org-13011 [142494-12-0]; lesopitron
[132449-46-8]; sunepitron [148408-65-5]; ipsapirone [95847-70-4];
flesinoxan [98206-10-1]; eptapirone [179756-85-5]; enilospirone
[59798-73-1]; AP-521 [151227-08-6]; alnespirone [138298-79-0];
adatanserin [127266-56-2]; VML-670 (Mealy, N. E. Drugs Future
2004, 29, 662); U-93385 [147145-16-2]; EMD-56551 [133109-87-
2]; CGP-50281 [100746-36-9]; Ebalzotan [149494-37-1].
(24) 2-Cyano-5-methylpyrazine, [98006-91-8].
(25) 3-Cyano-6-methylpyridazine, [49840-90-6].
(26) 5-Aminomethyl-2-methylpyrazine, [132664-85-8].
(27) (a) 5-Chloro-2-pyridinecarboxaldehyde, [31181-89-2]. (b) 5-(Hy-
droxymethyl)-2-pyridinecarboxaldehyde, [40749-33-5]. (c) 5-Fluo-
romethylpyridine-2-carboxaldehyde, [635323-74-9].
(28) Ethyl 6-(hydroxymethyl)nicotinate, [35005-81-3].
(29) Liu, Y.; Lien, I.-F. F.; Ruttgaizer, S.; Dove, P.; Taylor, S. D. Synthesis
and protection of aryl sulfates using the 2,2,2-trichloroethyl moiety.
Org. Lett. 2004, 6, 209-212.
(30) 4-(Fur-2-yl)-2-methyl-pyrimidine, [28565-17-5].
(31) Unpublished data.
(32) The bio-transformation of 1 into 2 is likely to proceed through the
alcohol 3. However, 3 could not be detected in vivo (neither any
phase 2 metabolite).
(33) The extra nitrogen in the diazines decreased the lipophilicity of the
compound (e.g., 9: log D ) 1.5, pKa ) 6.7) as compared to 1: log
D ) 2.6, pKa ) 6.4. This, in turn, influenced ADM parameters.
(34) 8-OH-DPAT: 8-hydroxy-2-dipropylaminotetralin, [76135-31-4].
(35) (a) de Paulis, T. Drug evaluation: PRX-00023, a selective 5-HT1A
receptor agonist for depression. Curr. Opin. InVest. Drugs 2007, 8,
78-86. Iyer, G. R.; Reinhard, J. F.; Oshana, S.; Kauffman, M;
Donahue, S. Tolerability, pharmacokinetics, and neuroendocrine
effects of PRX-00023, a novel 5-HT1A agonist, in healthy subjects.
J. Clin. Pharmacol. 2007, 47, 817-824. (b) Becker, O. M.; Dhanoa,
D. S.; Marantz, Y.; Chen, D.; Shacham, S.; Cheruku, S.; Heifetz,
A.; Mohanty, P.; Fichman, M.; Sharadendu, A.; Nudelman, R.;
Kauffman, M.; Noiman, S. An integrated in silico 3D model-driven
discovery of a novel, potent, and selective amidosulfonamide 5-HT1A
agonist (PRX-00023) for the treatment of anxiety and depression. J.
Med. Chem. 2006, 49, 3116-3135. This compound is in Phase IIb
for major depressive disorder.
(10) Buspirone: 8-(4-(4-(2-pyrimidinyl)-1-piperazinyl)butyl)-8-azaspiro-
(4.5)decane-7,9-dione, [36505-84-7].
(11) Tandospirone: 4,7-methano-1H-isoindole-1,3(2H)-dione, hexahydro-
2-(4-(4-(2-pyrimidinyl)-1-piperazinyl)butyl)-, (3aR,4â,7â,7aR)-, 2-hy-
droxy-1,2,3-propanetricarboxylate, [112457-95-1].
(12) Koek, W.; Vacher, B.; Cosi, C.; Assie´, M. B.; Patoiseau, J.-F.;
Pauwels, P. J.; Colpaert, F. C. 5-HT1A receptor activation and
antidepressant-like effects: F 13714 has high efficacy and marked
antidepressant potential. Eur. J. Pharmacol. 2001, 420, 103-112.
(13) Appert-Collin, A.; Duong, F. H. T.; Passilly Degrace, P.; Warter, J.
M.; Poindron, P.; Gies, J. P. MAPK activation via 5-hydrox-
ytryptamine 1A receptor is involved in the neuroprotective effects
of xaliproden. Int. J. Immunopathol. Pharmacol. 2005, 18, 21-31.
Kline, A. E.; Massucci, J. L.; Dixon, C. E.; Zafonte, R. D.; Bolinger,
B. D. The therapeutic efficacy conferred by the 5-HT(1A) receptor
agonist 8-hydroxy-2-(di-n-propylamino)tetraline (8-OH-DPAT) after
experimental traumatic brain injury is not mediated by concomitant
hypothermia. J. Neurotrauma 2004, 21, 175-185. Kline, A. E.; Yu,
J.; Horvath, E.; Marion, D. W.; Dixon, C. E. The selective 5-HT-
(1A) receptor agonist repinotan HCl attenuates histopathology and
spatial learning deficits following traumatic brain injury in rats.
Neuroscience 2001, 106, 547-555.
(14) In rat and dog, 1 (F 13640) has a bioavailability greater than 70%.
In human, its suitable pharmacokinetic properties were confirmed

5-HT1A Agonists for Antidepressant Treatment
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 20 5033
(36) Flibanserin: 1,3-dihydro-1-[2-[4-[3-(trifluoromethyl)phenyl]-1-pip-
erazinyl]ethyl]-2H-benzimidazol-2-one, [167933-07-5]. Invernizzi, R.
W.; Sacchetti, G.; Parini, S.; Acconcia, S.; Samanin, R. Flibanserin,
a potential antidepressant drug, lowers 5-HT and raises dopamine
and noradrenaline in the rat prefrontal cortex dialysate: Role of
5-HT1A receptors. Br. J. Pharmacol. 2003, 139, 1281-1288. Borsini,
F.; Evans, K.; Jason, K.; Rohde, F.; Alexander, B.; Pollentier, S.
Pharmacology of flibanserin. CNS Drug ReV. 2002, 8, 117-142.
Initially, this compound underwent clinical trial in depression but
was then switched to sexual dysfunction in women (Phase III).
(37) See ref 8. See also: Pauwels, P. J.; Tardif, S.; Wurch, T.; Colpaert,
F. C. Stimulated [35S]GTPγS binding by 5-HT1A receptor agonists
in recombinant cell lines: modulation of apparent efficacy by
G-protein activation state. Naunyn-Schmiedeberg’s Arch. Pharmacol.
1997, 356, 551-561.
(38) (a) Tricklebank, M. D.; Forler, C.; Fozard, J. R. The involvement of
sybtypes of the 5-HT1 receptor and of catecholaminergic systems in
the behavioural response to 8-hydroxy-2-(di-n-propylamino)tetraline
in the rat. Eur. J. Pharmacol. 1984, 106, 271-282. (b) Colpaert, F.
C.; Koek, W.; Lehmann, J.; Rivet, J. M.; Lejeune, F.; Canton, H.;
Bervoets, K.; Millan, M. J.; Laubie, M.; Lavielle, G. S 14506: A
novel, potent, high-efficacy 5-HT1A agonist and potential anxiolytic
agent. Drug DeV. Res. 1992, 26, 21-48.
471. Yocca, F. D.; Iben, L.; Meller, E. Lack of apparent receptor
reserve at postsynaptic 5-hydroxytryptamine1A receptors negatively
coupled to adenyl cyclase activity in rat hippocampal membranes.
Mol. Pharmacol. 1992, 41, 1066-1072. Lohse, M. J. G protein-
coupled receptors: too many dimers? Nat. Methods 2006, 3, 972-
973.
(47) Sullivan, N. R.; Crane, J. W.; Damjanoska, K. J.; Carrasco, G. A.;
D’Souza, D. N.; Garcia, F.; Van de Kar, L. D. Tandospirone activates
neuroendocrine and ERK (MAP kinase) signaling pathways specif-
ically through 5-HT1A receptor mechanisms in vivo. Naunyn-
Schmiedeberg’s Arch. Pharmacol. 2005, 371, 18-26. Newman-
Tancredi, A.; Cussac, D.; Marini, L.; Millan, M. J. Antibody capture
assay reveals bell-shaped concentration-response isotherms for h5-
HT1A receptor-mediated GRi3 activation: Conformational selection
by high-efficacy agonists, and relationship to trafficking of receptor
signaling. Mol. Pharmacol. 2002, 62, 590-601. Martinez, D.; Hwang,
D.-R.; Mawlawi, O.; Slifstein, M.; Kent, J.; Simpson, N.; Parsey, R.
V.; Hashimoto, T.; Huang, Y.; Shinn, A.; Van Heertum, R.; Abi-
Dargham, A.; Caltabiano, S.; Malizia, A.; Cowley, H.; Mann, J. J.;
Laruelle, M. Differential occupancy of somatodendritic and postsyn-
aptic 5HT(1A) receptors by pindolol: a dose occupancy study with
[11C]WAY 100635 and positron emission tomography in humans.
Neuropsychopharmacology 2001, 24, 209-229. Clarke, W. P.;
Yocca, F. D.; Maayani, S. Lack of 5-hydroxytryptamine1A-mediated
inhibition of adenyl cyclase in dorsal raphe of male and female rats.
J. Pharmacol. Exp. Ther. 1996, 277, 1259-1266.
(48) Mannoury la Cour, C.; El Mestikawy, S.; Hanoun, N.; Hamon, M.;
Lanfumey, L. Regional differences in the coupling of 5-hydrox-
ytryptamine-1A receptors to G proteins in the rat brain. Mol.
Pharmacol. 2006, 70, 1013-1021. Marazziti, D.; Palego, L.; Gi-
romella, A.; Mazzoni, M. R.; Borsini, F.; Mayer, N.; Naccarato, A.
G.; Lucacchini, A.; Cassano, G. B. Region-dependent effects of
flibanserin and buspirone on adenylyl cyclase activity in the human
brain. Int. J. Neuropsychopharmacol. 2002, 5, 131-140. Evans, K.
L. J.; Cropper, J. D.; Berg, K. A.; Clarke, W. P. Mechanisms of
regulation of agonist efficacy at the 5-HT(1A) receptor by phospho-
lipid-derived signaling components. J. Pharmacol. Exp. Ther. 2001,
297, 1025-1035. Rabiner, E. A.; Gunn, R. N.; Wilkins, M. R.;
Sargent, P. A.; Mocaer, E.; Sedman, E.; Cowen, P. J.; Grasby, P. M.
Drug action at the 5-HT(1A) receptor in vivo: autoreceptor and
postsynaptic receptor occupancy examined with PET and [carbonyl-
11C]WAY-100635. Nucl. Med. Biol. 2000, 27, 509-513.
(39) Compound 9 selectivity was over 1000-fold with respect to the other
serotonin subtypes, but also toward an extensive panel of other
receptors (e.g., dopaminergic, adrenergic...), transporters, ion chan-
nels, and enzymes.
(40) Initially, we estimated the bioavailability of these compounds from
the ratios of the doses inducing FPT at fixed intervals after ip and
po administrations. Subsequently, the absolute bioavailability in rat
of compound 9 was measured (F > 80%, unpublished data).
(41) See ref 8 and references therein.
(42) (a) Belozertseva, I. V.; Kos, T.; Popik, P.; Danysz, W.; Bespalov,
A. Y. Antidepressant-like effects of mGluR1 and mGluR5 antagonists
in the rat forced swim and the mouse tail suspension tests. Eur.
Neuropsychopharmacol. 2007, 17, 172-179. Cryan, J. F.; Valentino,
R. J.; Lucki, I. Assessing substrates underlying the behavioral effects
of antidepressants using the modified rat forced swimming test.
Neurosci. BiobehaV. ReV. 2005, 29, 547-569. Lucki, I. The forced
swimming test as a model for core and component behavioral effects
of antidepressant drugs. BehaV. Pharmacol. 1997, 8, 523-532. Detke,
M. J.; Lucki, I. Detection of serotonergic and noradrenergic
antidepressants in the rat forced swimming test: the effects of water
depth. BehaV. Brain Res. 1996, 73, 43-46. Borsini, F.; Meli, A. Is
the forced swimming test a suitable model for revealing antidepressant
activity? Psychopharmacology 1988, 94, 147-160. Porsolt, R. D.;
Bertin, A.; Blavet, N.; Deniel, M.; Jalfre, M. Immobility induced by
forced swimming in rats: effects of agents which modify central
catecholamine and serotonin activity. Eur. J. Pharmacol. 1979, 57,
201-210. (b) Porsolt, R. D.; Anton, G.; Blavet, N.; Jalfre, M.
Behavioral despair in rats: a new model sensitive to antidepressant
treatments. Eur. J. Pharmacol. 1978, 47, 379-391. (c) An important
experimental parameter in the FST that varies between laboratories
is the depth of the water in the cylinder.
(43) A previous study in this laboratory established that the tricyclic
antidepressant imipramine and the SSRI paroxetine reversed im-
mobility only weakly and partially upon repeated administration.
Koek, W.; Patoiseau, J. F.; Assie´, M.-B.; Cosi, C.; Kleven, M. S.;
Dupont-Passelaigue, E.; Carilla-Durand, E.; Palmier, C.; Valentin,
J. P.; John, G. W.; Pauwels, P. J.; Tarayre, J. P.; Colpaert, F. C.
F11440, a potent, selective, high efficacy 5-HT1A receptor agonist
with marked anxiolytic and antidepressant potential. J. Pharmacol.
Exp. Ther. 1998, 287, 266-283.
(49) Schildkraut, J. J. The catecholamine hypothesis of affective disor-
ders: A review of supporting evidence. Am. J. Psychiatry 1965, 122,
509-522.
(50) Gettys, T. W.; Fields, T. A.; Raymond, J. R. Selective activation of
inhibitory G-protein alpha-subunits by partial agonists of the human
5-HT1A receptor. Biochemistry 1994, 33, 4283-4290. Kenakin, T.
Inverse, protean, and ligand-selective agonism: matters of receptor
conformation. FASEB J. 2001, 15, 598-611.
(51) (a) Dremencov, E.; El Mansari, M.; Blier, P. Noradrenergic aug-
mentation of escitalopram response by risperidone: Electrophysi-
ologic studies in the rat brain. Biol. Psychiatry 2007, 61, 671-678.
(b) Blier, P. The pharmacology of putative early-onset antidepressant
strategies. Eur. Neuropsychopharmacol. 2003, 13, 57-66. Artigas,
F.; Celada, P.; Laruelle, M.; Adell, A. How does pindolol improve
antidepressant action? Trends Pharmacol. Sci. 2001, 22, 224-228.
Blier, P.; Bergeron, R.; De Montigny, C. Selective activation of
postsynaptic 5-HT-1A receptors induces rapid antidepressant re-
sponse. Neuropsychopharmacology 1997, 16, 333-338.
(52) 2-Pyrimidinemethanamine, [75985-45-4].
(53) 5-Chloro-2-pyrimidinemethanamine, [426266-77-5].
(54) 4-Methoxy-2-pyrimidinemethanamine, [909563-18-4].
(55) Newman-Tancredi, A.; Assie´, M. B.; Leduc, N.; Ormie`re, A.-M.;
Danty, N.; Cosi, C. Novel antipsychotics activate recombinant human
and native rat serotonin 5-HT1A receptors: affinity, efficacy and
potential implications for treatment of schizophrenia. Int. J. Neu-
ropsychopharmacol. 2005, 8, 341-356.
(56) McPherson, G. A. Analysis of radioligand binding experiments. A
collection of computer programs for the IBM PC. J. Pharmacol.
Methods 1985, 14, 213-228.
(57) Kleven, M.; Ybema, C.; Carilla, E.; Hamon, M.; Koek, W. Modifica-
tion of behavioral effects of 8-hydroxy-2-(di-n-propylamino)tetralin
following chronic ethanol consumption in the rat: evidence for the
involvement of 5-HT1A receptors in ethanol dependence. Eur. J.
Pharmacol. 1995, 281, 219-228.
(58) Tallarida, R. J.; Murray, R. B. Manual of Pharmacologic Calculations
with Computer Programs, 2nd ed.; Springer-Verlag: New York,
1987.
(44) Borsini, F.; Cesana, R. Further characterization of potential antide-
pressant action of flibanserin. Psychopharmacology 2001, 159, 64-
69.
(45) The progress made in understanding the mechanism of action of 9
will be reported in due course. For abstracts, see: Newman-Tancredi,
A.; Assie´, M. B.; Cosi, C.; Bruins-Slot, L.; Cussac, D.; Martel, J. C.
F15599 a highly selective serotonin 5-HT1A receptor agonist: in vitro
affinity and efficacy profile at native rat and recombinant human
5-HT1A receptors. Eur. Neuropsychopharmacol. 2006, 16, S232.
Assie´, M. B.; Bardin, L.; Depoortere, R.; Carilla-Durand, E.;
Newman-Tancredi, A. F15599 a highly selective serotonin 5-HT1A
receptor agonist: in vivo profile in neurochemical and behavioural
rat models of serotoninergic activity. Eur. Neuropsychopharmacol.
2006, 16, S233.
(46) Hjorth, S.; Magnusson, T. The 5-HT1A receptor agonist 8-OH-DPAT
preferentially activates cell body 5-HT autoreceptors in rat brain in
vivo. Naunyn-Schmiedeberg’s Arch. Pharmacol. 1988, 338, 463-
JM070714L