Differentiation between partial agonists and neutral 5-HT(1B) antagonists by chemical modulation of 3-[3-(N,N-dimethylamino)propyl]-4- hydroxy-N-[4-(pyridin-4-yl)phenyl]benzamide (GR-55562)

Article abstract of DOI:10.1021/jm9702948

The synthesis and binding affinity at cloned h5-HT(1B), h5-HT(1D), and h5-HT(1A) receptors of 3-[3-(N,N-dimethylamino)propyl]-4-hydroxy-N-[4- (pyridin-4-yl)phenyl]benzamide (2, GR-55562) and four O-methylated analogs are described. The functional activity of these compounds was determined at the h5-HT(1B) receptor using a [³⁵S]GTPγS binding assay. The four analogs have been prepared in order to evaluate the influence of the alkylamino side chain conformation on binding and intrinsic activity. Whereas 2 and its derivatives display a similar binding affinity profile, major differences arise from analysis of the intrinsic activity data at h5-HT(1B) receptors. The O-methylated analog of 2, 3-[3-(N,N-dimethylamino)propyl]-4-methoxy-N- [4-(pyridin-4-yl)phenyl]benzamide (3a), and the (1Z)-3-(N,N- dimethylamino)prop-1-enyl derivative (3c) act as neutral and potent antagonists (in a similar way to 2), while the 3-(N,N-dimethylamino)-prop-1- ynyl (3b) and (1E)-3-(N,N-dimethylamino)prop-1-enyl (3d) analogs display nonnegligible agonist activity. Molecular modeling studies show that, when the common triaryl portions of the molecules are overlapped, the partial agonists and the neutral antagonists differ by a near-orthogonal orientation of the NH⁺ projection to the hydrogen-bond receptor site.

Full text of DOI:10.1021/jm9702948

3542
J . Med. Chem. 1997, 40, 3542-3550
Differ en tia tion betw een P a r tia l Agon ists a n d Neu tr a l 5-HT1B An ta gon ists by
Ch em ica l Mod u la tion of 3-[3-(N,N-Dim eth yla m in o)p r op yl]-4-h yd r oxy-
N-[4-(p yr id in -4-yl)p h en yl]ben za m id e (GR-55562)
Marie Lamothe,† Petrus J . Pauwels,‡ Karine Belliard,† Philippe Schambel,§ and Serge Halazy*,†
Medicinal Chemistry Division, Cellular and Molecular Biology Department, and Analytical Chemistry Department, Centre de
Recherche Pierre FABRE, 17 Avenue J ean Moulin, 81106 Castres Ce´dex, France
Received May 5, 1997X
The synthesis and binding affinity at cloned h5-HT1B, h5-HT1D, and h5-HT1A receptors of 3-[3-
(N,N-dimethylamino)propyl]-4-hydroxy-N-[4-(pyridin-4-yl)phenyl]benzamide (2, GR-55562) and
four O-methylated analogs are described. The functional activity of these compounds was
determined at the h5-HT1B receptor using a [35S]GTPγS binding assay. The four analogs have
been prepared in order to evaluate the influence of the alkylamino side chain conformation on
binding and intrinsic activity. Whereas 2 and its derivatives display a similar binding affinity
profile, major differences arise from analysis of the intrinsic activity data at h5-HT1B receptors.
The O-methylated analog of 2, 3-[3-(N,N-dimethylamino)propyl]-4-methoxy-N-[4-(pyridin-4-
yl)phenyl]benzamide (3a ), and the (1Z)-3-(N,N-dimethylamino)prop-1-enyl derivative (3c) act
as neutral and potent antagonists (in a similar way to 2), while the 3-(N,N-dimethylamino)-
prop-1-ynyl (3b) and (1E)-3-(N,N-dimethylamino)prop-1-enyl (3d ) analogs display nonnegligible
agonist activity. Molecular modeling studies show that, when the common triaryl portions of
the molecules are overlapped, the partial agonists and the neutral antagonists differ by a near-
orthogonal orientation of the NH+ projection to the hydrogen-bond receptor site.
In tr od u ction
are now termed 5-HT1B with distinction between human
(h5-HT1B) and rat (r5-HT1B) receptors, to account for
their pharmacological differences.
5-HT (serotonin), a neurotransmitter, is involved in
numerous physiological (e.g., thermoregulation, hemo-
dynamics, feeding, sleeping) and pathophysiological
(e.g., depression, hypertension, migraine, anxiety) pro-
cesses and interacts with various distinct membrane
receptors.1 These receptors have been divided in seven
classes: 5-HT1, 5-HT2, 5-HT3, 5-HT4, 5-HT5, 5-HT6, and
5-HT7.2 Among them, the 5-HT1B/1D receptors are the
most abundant 5-HT1 receptor subtypes in the mam-
malian central nervous system (CNS).
The 5-HT1B binding site was first characterized in rat
brain membranes and later on the 5-HT1D binding site
in bovine caudate nucleus.3 Molecular biological studies
demonstrated that the human 5-HT1D receptor site is
encoded by a family of two distinct genes termed
initially 5-HT1DR and 5-HT1Dâ
5-HT1B/1D receptors belong to the family of 7-TM
G-protein-coupled receptors (both coupled to adenylate
cyclase) and are implicated in important functional
activities:9 for example, in the vascular periphery,
activation of 5-HT1B/1D receptors on blood vessels leads
to vasoconstriction,10 and 5-HT1B/1D receptors are also
involved in the inhibition of protein extravasation,11 two
mechanisms relevant to the antimigraine activity of
5-HT1B/1D agonists such as sumatriptan, naratriptan,
zolmitriptan, and rizatriptan. In the CNS, presynaptic
5-HT1B/1D receptors on serotoninergic axon terminals
control the release of 5-HT.9 This function is likely to
be connected with the pathophysiology of depression,
since several lines of preclinical and clinical evidence
indicate that an enhancement of 5-HT-mediated neu-
rotransmission might underlie the therapeutic effect of
most antidepressant drugs. As a consequence, the
blockade of terminal 5-HT1B/1D receptors (and probably
more precisely 5-HT1B autoreceptors) by selective an-
tagonists has been proposed12,13 as a new approach
toward the design of potentially more efficient and fast-
acting antidepressant drugs since 5-HT1B/1D blockade
would in theory immediately elevate terminal 5-HT
release.
4
.
Binding studies have
previously shown that the human 5-HT1Dâ and rat
5-HT1B receptors are “pharmacologically different” as
demonstrated, for example, by studies with propanolol
which binds to the rodent 5-HT1B receptor with nano-
molar affinities but recognizes the 5-HT1Dâ binding site
only very poorly (IC50 > 1000 nM).5 It has now been
demonstrated that this difference in the pharmacologi-
cal profile is the result of a single amino-acid modifica-
tion.6,7 Since the amino-acid sequence, the function, and
the regional distribution of the rodent 5-HT1B receptor
and of the human 5-HT1Dâ receptor are nearly identical,
these two receptors are considered as species homologs.
The Serotonin Club Nomenclature Committee has re-
2′-Methyl-4′-(5-methyl[1,2,4]oxadiazol-3-yl)biphenyl-
4-carboxylic acid [4-methoxy-3-(4-methylpiperazin-1-yl)-
phenyl]amide (1, GR-127935; Chart 1) has proved most
valuable as a potent and selective 5-HT1B/1D receptor
antagonist, but it fails to discriminate between these
two receptor subtypes.14 Moreover, recent investiga-
tions at the level of cloned h5-HT1B/1D receptors have
shown that 1 acts as a weak but full agonist at 5-HT1D
sites and can be classified as a partial, “nonsilent”
antagonist at 5-HT1B sites.15,16 It has previously been
cently proposed8 a revised nomenclature for 5-HT1B
5-HT1DR, and 5-HT1Dâ receptor subtypes. 5-HT1DR re-
ceptors are now termed 5-HT1D, while 5-HT1Dâ receptors
,
† Medicinal Chemistry Division.
‡ Cellular and Molecular Biology Department.
§ Analytical Chemistry Department.
X Abstract published in Advance ACS Abstracts, October 1, 1997.
S0022-2623(97)00294-X CCC: $14.00 © 1997 American Chemical Society

Synthesis and Binding Affinity of GR-55562
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22 3543
Ch a r t 1
Sch em e 1a
a
Reagents: BBr3, CH2Cl2, 62%.
Sch em e 2
proposed that the failure of 1 to increase brain 5-HT
release in vivo may be related to the intrinsic activity
of this particular compound at 5-HT1B/1D sites.14 The
discovery of new, potent, neutral, and selective 5-HT1B
antagonists still represents a fascinating challenge to
permit the evaluation of the concept of 5-HT1B antago-
nism as a tool to improve 5-HT neurotransmission and,
hopefully, to identify new antidepressant drugs.
Interestingly enough, unlike 1, 3-[3-(N,N-dimethyl-
amino)propyl]-4-hydroxy-N-[4-(pyridin-4-yl)phenyl]-
benzamide (2, GR-55562) has recently been character-
ized as a neutral, potent, and more selective h5-HT1B
receptor antagonist.17 These conclusions were based on
functional studies (inhibition of forskolin-stimulated
adenosine 3′,5′-cyclic monophosphate production) in
HeLa cells stably transfected with 5-HT1B human recep-
tor subtypes.
dichloromethane (Scheme 1). The preparation of the
propyl analog 3a has been previously reported.20 It
involves the palladium-catalyzed coupling of 3-iodo-4-
methoxybenzoic acid and 1-(N,N-dimethylamino)-2-
propyne to give 5b followed by a complete reduction of
the alkyne function and the condensation of the result-
ing carboxylic acid 5a with 4-(4-pyridinyl)benzenamine
(Scheme 2). All other analogs (3b-d ) were obtained by
coupling of the appropriate carboxylic acid intermediate
5b-d with 4-(4-pyridinyl)benzenamine 9 using 1-[3-
(dimethylamino)propyl]-3-ethylcarbodiimide (EDCI) in
poor to moderate yields (Scheme 3).
The carboxylic acid intermediates 5b-d were pre-
pared from commercially available methyl 3-iodo-4-
methoxybenzoate (Scheme 4). A minor modification of
Sonogashira’s original methodology21 allowed the intro-
duction of the alkynyl chain of 4b in quantitative yield.
Compound 4b was then hydrolyzed under basic condi-
tions to give the first carboxylic acid intermediate 5b
in 94% yield. The Z intermediate 5c was obtained in
quantitative yield from 5b by catalytic hydrogenation
with 2% palladium(0) over charcoal. With such a small
quantity of catalyst the reaction was sufficiently slow
to allow the isolation of the (Z)-alkene. In order to
access the E intermediate, the hydrogenation was
performed under acidic conditions on methyl ester 4b.
However a mixture of the E and Z isomers as well as
traces of the fully hydrogenated propyl analog 4a was
obtained. Fortunately the E isomer 4d could be sepa-
rated from the mixture and was obtained in 61% yield.
A mild basic hydrolysis of 4d gave the carboxylic acid
intermediate 5d in 93% yield.
More recent investigations from our laboratory18 have
shown that [35S]GTPγS binding studies in C6-glial cells
stably transfected with 5-HT1B or 5-HT1D receptors
represent a highly sensitive and quantitative way to
characterize the intrinsic activity of 5-HT1B/1D ligands.
As a consequence this assay allows the differentiation
between full agonists, partial agonists, and neutral
antagonists. Importantly, the detection of signs of
residual agonist activity is particularly relevant to
identify neutral antagonists at 5-HT1B/1D receptor sub-
types. In this paper, we report on the synthesis and
binding at cloned h5-HT1B, 5-HT1D, and 5-HT1A recep-
tors of 2 and four O-methylated analogs (3a -d ). More-
over the functional activity of these compounds was
determined at the h5-HT1B receptor using a [35S]GTPγS
binding assay. These particular compounds have been
prepared in order to evaluate the influence of the basic
amino-side chain conformation on binding and intrinsic
activity at h5-HT1B receptors.
Resu lts
Ch em istr y. 219 was prepared from its methoxy
analog 3a in 62% yield by treatment with BBr3 in

3544 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22
Lamothe et al.
Sch em e 3a
a
Reagents: EDCI, Et3N, CH2Cl2, 4-DMAP, 45% for 3b, 25% for 3c, 20% for 3d .
Sch em e 4a
a
Reagents: (a) 2 equiv of pyrrolidine, 5% Pd(PPh3)4, 10% CuI, 1.7 equiv of 1-(N,N-dimethylamino)-2-propyne, THF, 100%; (b) LiOH,
THF, H2O, 94% for 5b, 93% for 5d ; (c) 2% Pd/C, EtOH, H2, 100%; (d) 2% Pd/C, MeOH, AcOH, H2, 61%.
Sch em e 5a
a
Reagents: (a) (Boc)2O, toluene, 70 °C, 100%; (b) i. n-BuLi, (i-PrO)3B, -78 °C to rt; ii. NaOH, 50%; (c) 1,2-dimethoxyethane, t-BuOK,
6% Pd(PPh3)4, H2O, 84 °C, 61%; (d) HCl, MeOH, 53%.
4-(4-Pyridinyl)benzenamine (9) was prepared accord-
ing to the general approach introduced by Terashima
and colleagues22 with some modifications that gave more
reproducible results and could be easily scaled up to a
few grams (Scheme 5). 4-Bromopyridine was first
converted into 4-pyridinylboronic acid (7)23 in 65% yield.
Attempts to couple 7 with 4-bromoaniline failed to give
any trace of 9. Therefore, 4-bromoaniline was first
protected as the Boc carbamate 6 in quantitative yield.
Then a modified Suzuki coupling24 between 6 and
boronic acid 7 gave 8 in 61% yield. Finally the depro-
tection of the aniline function was carried out with
hydrogen chloride in methanol to give the biaryl inter-
mediate 9 in 53% yield after purification over a short
column of alumina.
Biologica l Da ta . In a first series of experiments, the
binding affinities were measured for 2 and the four
derivatives 3a -d at recombinant h5-HT1A, h5-HT1B
,
and h5-HT1D receptors (Table 1). As previously re-
ported, 2 showed a selective binding affinity (50-70-
fold) for the h5-HT1B receptor subtype versus the h5-
HT1A and h5-HT1D receptor subtypes. The derivatives
3a -d showed also a preferential binding affinity for the
5-HT1B receptor subtype. h5-HT1B receptor affinities
were nearly identical regardless of the ligand displacing
either [3H]-5-CT or [3H]GR-125743; slight differences (3-
fold) were only apparent for 3a ,c. The functional
activity of these compounds was determined at the h5-
HT1B receptor using a [35S]GTPγS binding assay and a
membrane preparation from C6-glial cells stably ex-

Synthesis and Binding Affinity of GR-55562
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22 3545
Ta ble 1. Receptor Binding Affinities of 2 and Its Derivatives
for Recombinant h5-HT1A, h5-HT1D, and h5-HT1B Receptors
binding affinities (Ki, nM)a
b
b
b
b
h5-HT1A
h5-HT1D
h5-HT1B
h5-HT1B
[3H]-8-OH-DPATc [3H]-5-CTc [3H]-5-CTc [3H]GR-125743c
compd
HeLad
Cos-7d
C6-gliald
C6-gliald
2
842
>1000
>1000
633
700
540
1200
169
12.1 ( 0.9
70.1 ( 21.5
16.2 ( 4.9
19.0 ( 3.7
13.6 ( 3.0
14.0 ( 5.4
15.4 ( 1.9
22.3 ( 6.2
6.1 ( 1.4
3a
3b
3c
3d
>1000
560
18.8 ( 4.9
a
Values are presented as mean or mean (SD of 2-3 indepen-
dent experiments, each one performed in triplicate. Binding
affinities at 5-HT1A and 5-HT1D receptors are typically within
d
(10-20% of the mean. b Receptor subtype. c Radioligand. Trans-
fected cell type.
pressing h5-HT1B receptors.18 This preparation is sen-
sitive to stimulation of [35S]GTPγS binding by agonists
such as 5-HT and zolmitriptan. Both compounds are
efficacious agonists and stimulate [35S]GTPγS binding
to the same magnitude; a maximal [35S]GTPγS binding
response of 73 ( 5% was obtained. The compounds 2
and 3a ,c did not affect basal [35S]GTPγS binding;
neither stimulation nor inhibition of [35S]GTPγS binding
was obtained at micromolar concentrations. However,
both of them potently displaced the [35S]GTPγS binding
response of zolmitriptan (Figure 1A). Each compound
(1 µM) competitively antagonized the zolmitriptan-
mediated response. The derived KB values varied
between 19.4 and 22.7 nM in line with their respective
h5-HT1B binding affinities (Table 2). In contrast, com-
pounds 3b,d stimulated [35S]GTPγS binding at micro-
molar concentrations; their intrinsic activity repre-
sented respectively 23% and 32% at 1 µM compared to
a maximal response elicited by zolmitriptan (Figure 1B).
Both compounds (10 µM) partially displaced the zolmi-
triptan response (Figure 1C), compound 3d being more
potent than compound 3b in accordance with their
intrinsic activity.
Con for m a tion a l An a lysis. The goal of these com-
putations was to identify the common conformations of
3a -d and to rationalize the difference in intrinsic
activity observed with 3b,d at the 5-HT1B receptor site.
To begin with, the preferred conformations of the
common methoxybenzamide were determined through
computation of the torsion angles τ1, τ2, τ3, and τ4 (Chart
2). A single conformation for the molecule was obtained
(τ1 ) 56°, τ2 ) -41°, τ3 ) 151°, τ4 ) 0°) due to the rigidity
of the benzamide function and the steric hindrance
between the methoxy group and its ortho substituent.
Then the different alkylamino chains were introduced
(Table 3), and conformations that emerged from the
analysis were checked for further fitting studies with
the Disco module using the most constrained (3b) analog
as template restricted to its conformational families to
prevent redundant answers. The computation identi-
fied 14 matching points: five key atoms (three H-bond
acceptors, one H-bond donor, and 1 positive), six projec-
tions from the structure to the binding site (four H-bond
donors, one H-bond acceptor, and 1 negative), and three
ring centroids (Chart 3A). Finally, knowing that the
binding affinities of all four analogs for the 5HT1B
receptor were basically the same, an identical number
of matching points were chosen for all of them. Inter-
estingly enough, no common conformation was found
that included both the H-bond donor atom (NH+) and
its corresponding H-bond acceptor site. However when
F igu r e 1. Intrinsic activity and antagonist effects of 2 and
its derivatives on h5-HT1B receptor-mediated stimulation of
[35S]GTPγS binding to C6-glial cellular membranes. (A) Mem-
branes were preincubated with the indicated concentrations
of zolmitriptan in either the absence (b) or presence of 1 µM
2 (O), 3a (0), or 3c (4) before [35S]GTPγS binding was
performed. (B) Membranes were incubated with the indicated
concentrations of either zolmitriptan (b), 3b (O), or 3d (0) for
[35S]GTPγS binding. (C) Membranes were preincubated with
the indicated concentrations of zolmitriptan in either the
absence (b) or presence of 10 µM 3b (O) or 3d (0) before [35S]-
GTPγS binding was performed. Stimulation of [35S]GTPγS
binding is expressed as a percentage of that obtained with 10
µM 5-HT. Typical concentration binding curves are shown.
Intrinsic activity, EC50 values, and KB values of 2-4 indepen-
dent experiments are summarized in Table 2.
the restriction for the overlapping of the NH+ centers
was dropped, several solutions were found, the best
hypothesis having a tolerance of 0.7 Å. The molecules
superimposed two by two (3a and 3c, 3b and 3d ), with
a near orthogonality between the two projections of the
H-bond donor (NH+) to the binding site within the two
groups (Chart 3B). The energy difference above the
absolute minimum energy of the superimposed confor-
mation was low for both 3b (0.9 kcal/mol) and 3d (1.2
kcal/mol) and high for the two neutral antagonists 3a
(3 kcal/mol) and 3c (3.3 kcal/mol). In this hypothesis,
it was clearly impossible for one molecule of one class
to adopt the conformation of the other class model due

3546 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22
Lamothe et al.
Ta ble 2. Intrinsic Activity and Potencies of 2 and Its
Derivatives for Antagonism of h5-HT1B Receptor-Mediated
Stimulation of [35S]GTPγS Bindinga
types is another common property for these derivatives.
However, a totally different conclusion arises from
analysis of intrinsic activity data at h5-HT1B receptors:
the results reported in Table 2 clearly show that the
saturated derivative 3a together with the (Z)-alkene
derivative 3c act as neutral, potent antagonists (in a
very similar way to 2) while the alkyne derivative 3b
as well as the (E)-alkene derivative 3d show nonnegli-
gible intrinsic activity which is also reflected by their
poor ability to antagonize zolmitriptan.
h5-HT1B receptor
intrinsic activity
zolmitriptan
EC50 (nM)
compd
(1 µM, %b)
KB (nM)
2
-3.7 ( 3.8
4.0 ( 10.5
32.3 ( 2.9
2.3 ( 6.7
3000 ( 283
2850 ( 495
837 ( 612c
3300 ( 424c
4300 ( 424
21.3 ( 2.1
22.7 ( 4.0
3a
3b
3c
3d
19.4 ( 2.6
23.0 ( 0.0
a
By employing molecular modeling techniques the
difference in intrinsic activity between the two pairs of
O-methyl analogs could be visualized by conformational
analysis and superimposition. When the common tri-
aryl portions of the molecules are overlapped, the partial
agonists and the neutral antagonists differ by a near-
orthogonal orientation of the NH+ projection to the
hydrogen-bond acceptor site. In the saturated analog
3a and the Z-unsaturated analog 3c, the NH+ is
positioned above the benzene plane due to the steric
hindrance between the alkylamino chain and the ortho
OMe substituent. However, for the (E)-olefin 3d , this
steric hindrance is lowered and a favorable electronic
conjugation of the double bond with the aromatic ring
will orient the NH+ in the plane of the benzene. The
very rigid acetylenic analog 3a must keep its NH+ close
to the benzene plane.
The observation that compounds 3a -d have almost
the same affinity for the cloned h5-HT1B receptor
subtype suggests that all four derivatives bind in a very
similar way to the receptor. However, the results
obtained at the functional level by using the GTPγS
binding assay suggest that both compounds 3a ,c (which
appear as neutral antagonists) selectively stabilize an
inactive conformation of the receptor (G-protein inde-
pendent binding mode), according to the “allosteric
ternary complex model”.25 On the other hand, it can
be assumed that compounds 3b,d (characterized as
partial agonists) simultaneously stabilize both the
uncoupled (inactive) form of the receptor and the active
conformation which allows coupling to the G-protein.
This hypothesis would explain the partial agonist
properties of these two compounds.
Altogether, the results discussed above suggest that
the intrinsic activity, as measured by GTPγS binding,
of 5-HT1B ligands such as 3a -d is likely related to the
conformation of the basic amino-side chain. Interest-
ingly, replacing the piperazine ring of 1 with a more
flexible amino side chain has recently been reported to
modulate intrinsic activity at 5-HT1B receptors.26 For
example, the [(dimethylamino)ethoxy]phenyl derivative
10 (SB-216641; Chart 4) has been reported as a partial
agonist,27 while the [(dimethylamino)propyl]phenyl de-
rivative 11 (Chart 4) possesses negative efficacy at the
5-HT1B receptor and could be therefore characterized
as an inverse agonist.26 These results together with the
studies reported in this paper demonstrate the impor-
tance of the spatial orientation of this basic amino
residue when designing neutral 5-HT1B antagonists.
Values are presented as mean (SD of 2-4 independent
experiments, each one performed in triplicate. Antagonism of the
zolmitriptan-mediated response (EC50 ) 60 nM) was tested at 1
µM (2, 3a ,c) or 10 µM (3b,d ). Versus 10 µM 5-HT. c EC50 values
are corrected for intrinsic activity.
b
Ch a r t 2
Ta ble 3. Conformational Analysis of Compounds 3a -d
no. of
no. of
compd
torsion angles
conformations
energy range
3a
3b
3c
3d
4
2
3
3
68
14
44
44
8.4-13.4
10.6-11.7
12.8-16.9
12.9-16.5
to the steric hindrance of the ortho methoxy group which
induced different orientations for the alkylamino chain.
Discu ssion
Compound 2 and its derivatives display selective
binding affinity for the h5-HT1B receptor in contrast to
the highly potent but nonselective binding affinity of 1
for h5-HT1B and h5-HT1D receptors.14 The functional
GTPγS data clearly show that 2 did neither stimulate
nor inhibit [35S]GTPγS binding in the investigated
recombinant C6-glial cell line; it competitively antago-
nized the zolmitriptan-mediated response. Therefore,
2 can be defined as a neutral antagonist of h5-HT1B
receptors. Its antagonist potency correlates with its
binding affinity and the antagonist potency previously
determined in a cAMP-mediated assay in recombinant
HeLa cells.17 This suggests 2 differs from 1 inasmuch
as this latter compound has been shown to share mixed
agonist-antagonist properties.14 The reported phar-
macological properties of 2 and its O-methyl derivative
3a are very similar especially when considering intrinsic
activity at the h5-HT1B receptor subtype (Table 2): both
compounds have no intrinsic activity and thus can be
considered as neutral antagonists, and both of them
antagonize h5-HT1B receptor-mediated stimulation of
[35S]GTPγS binding by zolmitriptan with equal poten-
cies (KB of 21.3 and 22.7 nM, respectively). Considering
the comparison between the propylamino derivative 3a
with the conformationally restricted derivatives 3b-d ,
at the binding level, it can be concluded that all four
compounds bind with nearly the same affinity at the
h5-HT1B receptor subtype supported by the data ob-
tained with [3H]GR-125743 displacement. The low
affinity for cloned h5-HT1D and h5-HT1A receptor sub-
Exp er im en ta l Section
Ch em ica l Syn th esis. Tetrahydrofuran (THF) was distilled
from sodium benzophenone ketyl, and dichloromethane was
distilled from calcium hydride, both under nitrogen. All
reactions were performed under positive nitrogen atmosphere,
in anhydrous solvents (unless otherwise stated). Pyrrolidine

Synthesis and Binding Affinity of GR-55562
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22 3547
Ch a r t 3
(pentet), m (multiplet), br (broad), dd (doublet of doublets), and
dt (doublet of triplets). Infrared spectra were recorded on a
Nicolet 510P FT-IR spectrophotometer as solid dispersion in
Ch a r t 4
KBr or film and are reported in cm-1
. Elemental analyses
were performed on a Fison EA 1108 CHN analyzer, and
determined values are within 0.4% of theory. Mass spectra
were recorded on a Nermag R10-10B spectrometer at Toulouse
by the Laboratory of Mass Spectra at Paul Sabatier University
using the chemical ionization method (NH3) with the molecular
ion as (MH+). Melting points were measured on a Electro-
thermal 9200 instrument and are uncorrected.
Meth yl 3-[3-(N,N-Dim eth yla m in o)p r op -1-yn yl]-4-m eth -
oxyben zoa te (4b). Methyl 3-iodo-4-methoxybenzoate (5 g,
17.1 mmol), pyrrolidine (2.8 mL, 34.2 mmol), and tetrakis-
(triphenylphosphine)palladium(0) (990 mg, 0.85 mmol) were
dissolved in THF (34 mL). In a separate flask 1-(N,N-
dimethylamino)-2-propyne (3.1 mL, 29 mmol) was reacted with
copper iodide (320 mg, 1.7 mmol) in 34 mL of THF and then
transferred dropwise into the first solution. The reaction
mixture was heated to 60 °C for 1 h and brought back to 20
°C where the stirring was continued for 15 h. The solvent was
removed under reduced pressure, and the residue was purified
by flash chromatography (1:49:0.5 to 1:24:0.5 MeOH:CH2Cl2:
was distilled over KOH prior to use. All others chemicals were
used directly as received. Hydrogenations were carried out
in a Parr 3911 hydrogenation apparatus. Analytical thin layer
chromatography (TLC) was performed on Merck glass-backed
silica gel 60 F 254 plates. All flash chromatography was done
with SDS silica gel 60 ACC (35-70 µm, pH 7) using a gradient
eluent system. 1H NMR (200, 400 MHz) were recorded
respectively on Brucker AC-200 and Brucker DPX-400 spec-
trometers. When peak multiplicity is reported, the following
abbreviations are used: s (singlet), d (doublet), t (triplet), p

3548 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22
Lamothe et al.
NH4OH gradient) to give the title compound (4.28 g, 100%):
1H NMR (200 MHz, DMSO-d6) δ 8.0-7.85 (m, 2H), 7.17 (d,
1H, J ) 8 Hz), 3.89 (s, 3H), 3.82 (s, 3H), 3.46 (brs, 2H), 2,23
(s, 6H); Mass (MH+ ) 248); IR (film) 1723 (CO); Rf ) 0.45 (95:
5:0.5 CH2Cl2:MeOH:NH4OH). Anal. (C14H17NO3‚0.25H2O) C,
H, N.
mopyridine in ether was obtained upon washing of its hydro-
chloride salt with cold NaOH (1 M in ice-water) and extraction
with cold ether (0 °C). This solution can be stored for a few
days over magnesium sulfate under nitrogen in the freezer
(-10 °C). Its titer can be measured by gravimetric titration
following the procedure described by Murray and Langham.28
A solution of 4-bromopyridine (0.63 M in ether, 110 mL, 70
mmol) was cooled to -78 °C and added to a cold solution (-78
°C) of n-BuLi (1.6 M in hexane, 52.5 mL, 84 mmol) in ether
(225 mL). The reaction mixture was stirred a further 20 min
at -78 °C prior to the addition of triisopropyl borate (21 mL,
91 mmol) and warmed to 20 °C over 1 h. After 15 h at room
temperature, the reaction was quenched with 50 mL of water,
and the organic layer was extracted with 100 mL of NaOH
(0.5 M). The combined aqueous phases were washed with
ether and acidified carefully to pH 6 (using a pH meter) with
HCl (2 N). The title compound precipitated and was collected
by a simple filtration (5.69 g, 65% yield): 1H NMR (200 MHz,
DMSO-d6) δ 8.52 (brs, 2H), 7.63 (brs, 2H). Anal. (C5H6BNO2)
C, H, N.
N-(ter t-Bu toxycar bon yl)-4-(pyr idin -4-yl)an ilin e (8). The
reaction was carried out under argon. To a solution of 6 (17
g, 62 mmol) and 7 (7.67 g, 62 mmol) in 1,2-dimethoxyethane
(700 mL) was added a solution of potassium tert-butoxide (61.9
g, 552 mmol) and tetrakis(triphenylphosphine)palladium(0)
(4.33 g, 3.7 mmol) in water (276 mL). The reaction mixture
was stirred at 84 °C for 8 h, then cooled, and evaporated. The
residue was dissolved in dichloromethane, washed with NaOH
(0.1 N), dried over magnesium sulfate, filtered, and concen-
trated. The crude product was purified by flash chromatog-
raphy (1:99:0.4 to 1:49:0.4 MeOH:CH2Cl2:NH4OH gradient) to
give the title compound (10.2 g, 61%): 1H NMR (200 MHz,
DMSO-d6) δ 9.57 (s, 1H), 8.56 (brd, 2H, J ) 5.9 Hz), 7.73 (d,
2H, J ) 8.7 Hz), 7.65 (d, 2H, J ) 6.2 Hz), 7.58 (d, 2H, J ) 8.6
Hz), 1.47 (s, 9H); IR (KBr) 1719 (CO); Rf ) 0.3 (95:5:0.5 CH2-
Cl2:MeOH:NH4OH). Anal. (C16H18N2O2) C, H, N.
3-[3-(N,N-Dim eth yla m in o)p r op -1-yn yl]-4-m eth oxyben -
zoic Acid (5b). To a solution of 4b (1.5 g, 6.2 mmol) in THF
(13 mL) under normal atmosphere was added dropwise a 1 M
aqueous solution of NaOH (2 equiv, 12.4 mmol). The reaction
mixture was stirred until total consumption of 4b as measured
by TLC; then carefully the reaction was quenched with 1 N
HCl (2 equiv) and the mixture concentrated. Water was
coevaporated with toluene. The residue obtained was tritu-
rated in ethanol and filtered. This operation was repeated
with methanol to obtain the title compound as a solid in 100%
yield (1.49 g): 1H NMR (200 MHz, DMSO-d6) δ 13.87 (brs, 1H,
COOH), 11.25 (brs, 1H, NH+), 8.05-7.90 (m, 2H), 7.21 (d, 1H,
J ) 9 Hz), 4.31 (s, 2H), 3.91 (s, 3H), 2.83 (s, 6H); IR (KBr)
1716 (CO); Rf ) 0.1 (90:10:1 CH2Cl2:MeOH:NH4OH). Anal.
(C13H16NO3‚0.9H2O‚1.15HCl) C, H, N.
3-[(1Z)-3-(N,N-Dim et h yla m in o)p r op -1-en yl]-4-m et h -
oxyben zoic Acid (5c). A suspension of 5b (1.49 g, 6.38 mmol)
and palladium(0) (5% on charcoal, 0.02 equiv) in DMF (18 mL)
and ethanol (18 mL) was hydrogenated for 80 h at a starting
pressure of 50 psi of H2. The reaction mixture was filtered
over a Celite bed, and the solvents were removed under
reduced pressure. Crude 5c (1.94 g, oil) was used in the
remaining synthesis without further purification: 1H NMR
(200 MHz, DMSO-d6) δ 7.81-7.76 (m, 2H), 6.88 (d, 1H, J )
8.4 Hz), 6.57 (brd, 1H, J ) 11.8 Hz), 5.68 (dt, 1H, J ) 5.8,
11.8 Hz), 3.79 (s, 3H), 3.06 (brd, 2H, J ) 6 Hz), 2,11 (s, 6H).
Meth yl 3-[(1E)-3-(N,N-Dim eth yla m in o)p r op -1-en yl]-4-
m eth oxyben zoa te (4d ). A suspension of 4b (6.34 g, 25.6
mmol) and palladium(0) (5% on charcoal, 0.02 equiv) in
methanol (86 mL) and glacial acetic acid (2.2 mL) was
hydrogenated for 5 days at a starting pressure of 50 psi of H2.
The reaction mixture was filtered over a Celite bed, and the
solvents were removed under reduced pressure. The crude
product was purified by flash chromatography (5:95:0.5 MeOH:
CH2Cl2:NH4OH) to give the title compound (3.81 g, 59%, oil):
1H NMR (200 MHz, DMSO-d6) δ 8.03 (d, 1H, J ) 2.1 Hz), 7.87
(dd, 1H, J ) 2.1, 8.6 Hz), 7.13 (d, 1H, J ) 8.7 Hz), 6.77 (brd,
1H, J ) 16 Hz), 6.30 (dt, 1H, J ) 16, 6.5 Hz), 3.90 (s, 3H),
3.83 (s, 3H), 3.09 (brd, 2H, J ) 6.4 Hz), 2.21 (s, 6H), 1.92 (s,
AcOH); Mass (MH+ ) 250); Rf ) 0.3 (95:5:0.5 CH2Cl2:MeOH:
NH4OH). Anal. (C14H19NO3‚CH3COOH‚0.4CH2Cl2) C, H, N.
3-[(1E)-3-(N,N-Dim et h yla m in o)p r op -1-en yl]-4-m et h -
oxyben zoic Acid (5d ). To a solution of 4d (3.88 g, 15.6 mmol)
in THF (31 mL) under normal atmosphere was added dropwise
a 1 M aqueous solution of NaOH (2 equiv, 31 mmol). The
reaction mixture was stirred until total consumption of 4d as
measured by TLC; then the reaction was carefully quenched
with 1 N HCl (31 mL) and the mixture concentrated. Water
was coevaporated with toluene. The residue obtained was
triturated in ethanol and filtered. This operation was repeated
with methanol to obtain the title compound as a solid in 93%
yield (3.39 g): 1H NMR (200 MHz, DMSO-d6) δ 8.27 (d, 1H, J
) 1.9 Hz), 7.85 (dd, 1H, J ) 2.1, 8.6 Hz), 7.08 (d, 1H, J ) 8.7
Hz), 6.87 (brd, 1H, J ) 16 Hz), 6.35 (dt, 1H, J ) 16, 6.5 Hz),
3.88 (s, 3H), 3.35 (brd, 2H, J ) 6.6 Hz), 2.39 (s, 6H); Mass
(MH+ ) 236); Rf ) 0.1 (90:10:1 CH2Cl2:MeOH:NH4OH).
N-(ter t-Bu t oxyca r b on yl)-4-b r om oa n ilin e (6). 4-Bro-
moaniline (13.0 g, 76 mmol) and di-tert-butyl dicarbonate (19.8
g, 91 mmol) were dissolved in toluene (380 mL) and heated at
70 °C for 15 h. The solvent was removed under reduced
pressure, and the residue was dissolved in ethyl acetate and
washed successively with HCl (0.1 N) and brine. The organic
phase was dried over magnesium sulfate, filtered, and con-
centrated. The crude product was purified by flash chroma-
tography (5:95 to 10:90 EtOAc:petroleum ether (PE) gradient)
to give the title compound (20.5 g, 100%): 1H NMR (200 MHz,
DMSO-d6) δ 9.49 (s, 1H), 7.42 (s, 4H), 1.47 (s, 9H); IR (KBr)
1709 (CO); Rf ) 0.3 (95:5 PE:EtOAc). Anal. (C11H14BrNO2)
C, H, N.
4-(P yr id in -4-yl)a n ilin e (9). N-(tert-Butoxycarbonyl)-4-
(pyridin-4-yl)aniline (8; 4.1 g, 15 mmol) was treated with a
solution of 1 M HCl in methanol at 20 °C for 2 days. The
reaction mixture was concentrated, and the residue was
charged on
a short alumina column. Elution with 10%
methanol in dichloromethane allowed the removal of the
impurity. Then the column was washed with a more polar
eluent (9:90:1 MeOH:CH2Cl2:NH4OH) to give the title com-
pound as the free base (1.71 g, 57%): 1H NMR (200 MHz,
DMSO-d6) δ 8.46 (dd, 2H, J ) 1.6, 4.7 Hz), 7.63-7.49 (m, 4H),
6.61 (brd, 2H, J ) 8.6 Hz), 5.51 (s, 2H, NH2); Mass (MH+
171). Rf ) 0.3 (95:5:0.5 CH2Cl2:MeOH:NH4OH).
)
3-[3-(N,N-Dim et h yla m in o)p r op -1-yn yl]-4-m et h oxy-N-
[4-(p yr id in -4-yl)p h en yl]ben za m id e (3b). Carboxylic acid
5b (0.5 g, 1.85 mmol), aniline 9 (682 mg, 2.7 mmol), EDCI (389
mg, 2.1 mmol), triethylamine (785 µL, 5.6 mmol), and 4-(dim-
ethylamino)pyridine (0.1 equiv) were dissolved in THF (10 mL)
and dichloromethane (20 mL) and stirred at 20 °C for 72 h.
The solvents were removed under reduced pressure, and the
residue was dissolved in dichloromethane and washed with 1
M NaOH. The organic phase was dried over magnesium
sulfate, filtered, and concentrated. The crude product was
purified by flash chromatography (4:96:0.4 to 10:90:0.4 MeOH:
CH2Cl2:NH4OH gradient) to give the title compound (317 mg,
45%). The fumarate salt was prepared by addition of 1 equiv
of fumaric acid in hot methanol and precipitation of the salt
at 0 °C: 1H NMR (200 MHz, DMSO-d6) δ 10.35 (s, 1H), 8.61
(d, 2H, J ) 5.9 Hz), 8.06 (d, 1H, J ) 2.2 Hz), 8.02 (dd, 1H, J
) 2.3, 8.7 Hz), 7.95 (d, 2H, J ) 8.8 Hz), 7.84 (d, 2H, J ) 8.8
Hz), 7.72 (brd, 2H, J ) 5.5 Hz), 7.21 (d, 1H, J ) 8.7 Hz), 6.61
(s, 3H, fumarate), 3.92 (s, 3H), 3.55 (s, 2H), 2.31 (s, 6H); Mass
(MH+ ) 386); IR (film) 1716, 1677 (CO); Rf ) 0.36 (92:8:0.2
CH2Cl2:MeOH:NH4OH); mp ) 193 °C. Anal. (C24H23N3O2‚
1.5C4H4O4) C, H, N.
3-[(1Z)-3-(N,N-Dim et h yla m in o)p r op -1-en yl]-4-m et h -
oxy-N-[4-(p yr id in -4-yl)p h en yl]ben za m id e (3c). Carboxylic
acid 5c (1.5 g, 6.3 mmol), aniline 9 (1.8 g, 9 mmol), EDCI (1.34
g, 7 mmol), triethylamine (2.7 mL, 19 mmol), and 4-(dimethyl-
amino)pyridine (0.1 equiv) were dissolved in THF (30 mL) and
dichloromethane (30 mL) and stirred at 20 °C for 72 h. The
(P yr id in -4-yl)bor on ic Acid (7). A stock solution of 4-bro-

Synthesis and Binding Affinity of GR-55562
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22 3549
solvents were removed under reduced pressure, and the
residue was dissolved in dichloromethane and washed with 1
M NaOH. The organic phase was dried over magnesium
sulfate, filtered, and concentrated. The crude product was
purified by flash chromatography (4:96:0.4 to 10:90:0.4 MeOH:
CH2Cl2:NH4OH gradient) to give the title compound (621 mg,
25%). The fumarate salt was prepared by addition of 1 equiv
of fumaric acid in hot methanol and precipitation of the salt
at 0 °C: 1H NMR (200 MHz, DMSO-d6) δ 10.3 (s, 1H), 8.60 (d,
2H, J ) 5.9 Hz), 8.05-7.75 (m, 6H), 7.70 (d, 2H, J ) 6 Hz),
7.17 (d, 1H, J ) 8.8 Hz), 6.70 (d, 1H, J ) 11.9 Hz), 6.56 (s,
3H, fumarate), 5.86 (dt, 1H, J )11.9, 6 Hz), 3.87 (s, 3H), 3.36
(d, 2H, J ) 6 Hz), 2.33 (s, 6H); Mass (MH+ ) 388); Rf ) 0.25
pA2 ) log(concentration - 1) - log(antagonist concentration).
Radioligands were obtained from New England Nuclear (Les
Ulis, France) or Amersham (Les Ulis, France).
Molecu la r Mod elin g. Molecular modeling was performed
on a Silicon Graphics Indigo workstation using the Sybyl
Molecular Modeling Software, version 6.3, from Tripos Associ-
ates, Inc., St. Louis, MO. Optimization by energy minimiza-
tion was done with the Tripos force field including the
electrostatic term calculated from Gasteiger and Marsili atomic
charges. The BFGS method was used for minimization
(convergence at 0.001 kcal/mol). Compound 3b was submitted
to the Gridsearch option from Sybyl with stepwise rotation of
20° around the two alkylamine torsion angles which gave 361
conformations and then classified in 14 conformational families
using a cluster analysis method from family spl macro (written
by M. Neuwels and E. Vande Water from UCB SA). The other
analogs having three or four alkylamine torsion angles were
submitted to the Randomsearch option from Sybyl with
maximum cycles at 1000, energy cutoff at 70 kcal/mol, rms
threshold at 0.2 Å, and convergence threshold at 0.005. This
option located energy minima by randomly adjusting the
chosen torsion angles, minimizing the energy of the resulting
geometry, and eliminating duplicates (<5 kcal/mol above the
absolute minimum energy). The three-dimensional pharma-
cophore mapping was generated with the Disco module from
Sybyl with default values of disco file2.dat option. Searchs
were generated using 3b as the reference compound, detailed
by class with a minimum of 13 feature requirements to fulfill
the overall fitting (distance tolerance 0.5-1.0 Å, increment 0.1
Å).
(92:8:0.1 CH2Cl2:MeOH:NH4OH); mp
) 196 °C. Anal.
(C24H25N3O2‚1.5C4H4O4‚1.25H2O) C, H, N, H2O.
3-[(1E)-3-(N,N-Dim et h yla m in o)p r op -1-en yl]-4-m et h -
oxy-N-[4-(p yr id in -4-yl)p h en yl]ben za m id e (3d ). Carboxy-
lic acid 5d (3.36 g, 14.3 mmol), aniline 9 (3.8 g, 18.6 mmol),
EDCI (3 g, 15.7 mmol), triethylamine (6 mL, 43 mmol), and
4-(dimethylamino)pyridine (0.1 equiv) were dissolved in THF
(70 mL) and dichloromethane (70 mL) and stirred at 20 °C for
12 h. The solvents were removed under reduced pressure, and
the residue was dissolved in dichloromethane and washed with
1 M NaOH. The organic phase was dried over magnesium
sulfate, filtered, and concentrated. The crude product was
purified by flash chromatography (4:96:0.4 to 10:90:0.4 MeOH:
CH2Cl2:NH4OH gradient) to give the title compound (1.1 g,
20%). The fumarate salt was prepared by addition of 1 equiv
of fumaric acid in hot methanol and precipitation of the salt
at 0 °C: 1H NMR (200 MHz, DMSO-d6) δ 10.35 (s, 1H), 8.60
(brd, 2H, J ) 6.2 Hz), 8.15 (d, 1H, J ) 2 Hz), 8.00-7.80 (m,
5H), 7.71 (brd, 2H, J ) 6.2 Hz), 7.17 (d, 1H, J ) 8.7 Hz), 6.92
(d, 1H, J ) 16 Hz), 6.56 (s, 3H, fumarate), 6.45 (dt, 1H, J )
16, 6.6 Hz), 3.9 (s, 3H), 3.43 (d, 2H, J ) 6.7 Hz), 2.44 (s, 6H);
Mass (MH+ ) 388); IR (film) 1716, 1671 (CO); Rf ) 0.36 (92:
Ack n ow led gm en t. The authors thank Ms. S. Tardif
for excellent technical assistance with the binding
assays and Ms. E. Dupeyron for her help in the
preparation of this manuscript.
8:0.2 CH2Cl2:MeOH:NH4OH); mp
(C24H25N3O2‚1.5C4H4O4) C, H, N, H2O.
)
220 °C. Anal.
3-[3-(N,N-Dim eth yla m in o)p r op yl]-4-h yd r oxy-N-[4-(p y-
r id in -4-yl)p h en yl]ben za m id e (2). 3-[3-(N,N-Dimethylami-
no)propyl]-4-methoxy-N-[4-(pyridin-4-yl)phenyl]benzamide (3a ;
840 mg, 2.16 mmol) was dissolved in dichloromethane (14 mL)
and cooled to -78 °C. Boron tribromide (1 M in CH2Cl2, 8.6
mL, 8.6 mmol) was added, and the reaction was allowed to
warm progressively to 20 °C. A precipitate appeared. The
stirring was continued for 20 h, after which methanol (3 mL)
was added dropwise. The title compound was filtered and
washed thoroughly with dichloromethane to give the title
compound as a pale yellow solid (616 mg, 62%): 1H NMR (400
MHz, MeOD-d4) δ 8.79 (d, 2H, J ) 6.4 Hz), 8.36 (d, 2H, J )
6.5 Hz), 8.04 (s, 4H), 7.84 (d, 1H, J ) 1.8 Hz), 7.77 (dd, 1H, J
) 2, 8.3 Hz), 6.91 (d, 1H, J ) 8.4 Hz), 3.20 (m, 2H), 2.91 (s,
6H), 2.79 (t, 2H, J ) 7.6 Hz), 2.10 (p, 2H, J ) 7.6 Hz); Mass
(MH+ ) 375); Rf ) 0.2 (95:5:0.5 CH2Cl2:MeOH:NH4OH); mp
> 240 °C. Anal. (C23H25N3O2‚2HBr‚0.6H2O) C, H, N, HBr,
H2O.
Refer en ces
(1) Glennon, R. A. Serotonin receptors: clinical implications. Neu-
rosci. Biobehav. Rev. 1990, 14, 35-47.
(2) Hoyer, D.; Clarke, D. E.; Fozard, J . R.; Hartig, P. R.; Martin, G.
R.; Mylechorane, E. J .; Saxena, P. R.; Humphrey, P. P. Inter-
national union of pharmacology classification of receptors for
5-hydroxytryptamine (serotonin). Pharmacol. Rev. 1994, 46,
157-203.
(3) Hartig, P. R.; Branchek, T. A.; Weinshank, R. L. A subfamily of
5-HT1D receptor genes. Trends Pharmacol. Sci. 1992, 13, 152-
159.
(4) Weinshank, R. L.; Zgombick, J . M.; Macchi, M. J .; Branchek, T.
A.; Hartig, P. R. Human serotonin 1D receptor is encoded by a
subfamily of two distinct genes: 5-HT1DR and 5-HT1Dâ. Proc. Natl.
Acad. Sci. U.S.A. 1992, 89, 3630-3634.
(5) Glennon, R. A. The binding of propanolol at 5-HT1Dâ T355N
mutant receptors may involve formation of two hydrogen bonds
to asparagine. Mol. Pharmacol. 1996, 49, 198-206.
(6) Oksenberg, D. A single amino-acid difference confers major
pharmacological variation between human and rodent 5-HT1B
receptors. Nature (London) 1992, 360, 161-163.
Biologica l Meth od s. Radioligand binding experiments
were performed on membrane preparations of the HeLa/HA7
cell line for h5-HT1A receptors, Cos-7 cells transfected with h5-
HT1D receptors, and the C6-glial cell line stably transfected
with h5-HT1B receptors as previously described.18 [35S]GTPγS
binding to membrane preparations of C6-glial/h5-HT1B cells
was performed in glass tubes and consisted of 0.4 mL of
membrane preparation (20-120 µg of protein) and 0.05 mL of
compound in either the absence or presence of antagonist in
20 mM Hepes (pH 7.4) supplemented with 30 µM GDP, 100
mM NaCl, 3 mM MgCl2, and 0.2 mM ascorbic acid. After an
incubation period of 30 min at 25 °C, 0.05 mL of [35S]GTPγS
(500 pM) was added for an additional period of 30 min. The
reactions were stopped as previously described.18 Maximal
stimulation of [35S]GTPγS binding was defined in the presence
of 10 µM 5-HT. Emax values are expressed as a percentage of
the maximal response obtained with 10 µM 5-HT. EC50 values
are expressed as the concentration of compound at which 50%
of their own maximal stimulation was obtained. For antago-
nist studies, concentration ratios were calculated and used to
obtain estimates of pA2 values using the following equation:
(7) Metcalf, M. A.; MC Guffin, R. W.; Hamblin, M. W. Conversion
of the human 5-HT1Dâ serotonin receptor to the rat 5-HT1B
ligand-binding phenotype by Thr355Asn site directed mutagen-
esis. Biochem. Pharmacol. 1992, 44, 1917-1920.
(8) Hartig, P. R.; Hoyer, D.; Humphrey, P. P. A.; Martin, G. R.
Alignment of receptor nomenclature with the human genome:
classification of 5-HT1B and 5-HT1D receptor subtypes. Trends
Pharmacol. Sci. 1996, 17, 103-105.
(9) Pauwels, P. J . 5-HT1B/1D receptor antagonists. Gen. Pharmacol.
1997, in press.
(10) Ferrari, M. D.; Saxena, P. R. On serotonin and migraine:
a
clinical and pharmacological review. Cephalagia 1993, 13, 151-
165.
(11) Moskowitz, M. A. Neurogenic versus vascular mechanisms of
sumatriptan and ergot alkaloids in migraine. Trends Pharmacol.
Sci. 1992, 13, 307-311.
(12) Halazy, S.; Lamothe, M.; J orand-Lebrun, C. 5-HT1B/1D antago-
nists and depression. Exp. Opin. Ther. Pat. 1997, 7, 339-352.
(13) Briley, M.; Moret, C. Neurobiological mechanisms involved in
antidepressant therapies. Clin. Neuropharmacol. 1993, 16, 387-
400.
(14) Pauwels, P. J . Pharmacological properties of a putative 5-HT1B/1D
receptor antagonist GR-127935. CNS Drugs Rev. 1996, 2, 415-
428.

3550 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 22
Lamothe et al.
(15) Pauwels, P. J .; Colpaert, F. C. The 5-HT1D receptor antagonist
GR 127935 is an agonist at cloned human 5-HT1DR receptor sites.
Neuropharmacology 1995, 34, 235-237.
(23) Fisher, F. C.; Havinga, E. Thermal and photoinduced deborona-
tions of some pyridine- and benzeneboronate anions. Recl. Trav.
Chim. Pays-Bas 1974, 93, 21-24.
(24) Chan, K. S.; Zhang, H. Base effect on the cross-coupling of bulky
arylboronic acid with halopyridines. Tetrahedron Lett. 1996, 37,
1043-1044.
(25) Samama, P.; Cotecchia, S.; Costa, T.; Lefkowitz, R. J . A muta-
tion-induced activated state of the â2-adrenergic receptor. Ex-
tending the ternary complex model. J . Biol. Chem. 1993, 268,
4625-4636.
(26) Roberts, C.; Price, G. W.; Gaster, L.; J ones, B. J .; Middlemiss,
D. N.; Routledge, C. Importance of h5-HT1B receptor selectivity
for 5-HT terminal autoreceptor activity: an in vivo microdialysis
study in the freely-moving guinea-pig. Neuropharmacology 1997,
36, 549-557.
(27) Price, G. W.; Burton, M. J .; Roberts, C.; Watson, J .; Duckworth,
M.; Gaster, L. SB-216641 and BRL 15572 pharmacologically
discriminate between h5-HT1B and h-5HT1D receptors. Proc. Br.
Pharmacol. Soc. 1996, (Sept. 11-13, University of Dundee),
301P.
(28) Murray, A.; Langham, W. H. A synthesis of isonicotinic acid by
halogen-metal exchange and its application to the preparation
of isonicotinic-C14 acid hydrazide. J . Am. Chem. Soc. 1952, 74,
6289-6290.
(16) Watson, J . M.; Burton, M. J .; Price, G. W.; J ones, B. J .;
Middlemiss, D. N. GR-127935 acts as
a partial agonist at
recombinant human 5-HT1DR and 5-HT1Dâ receptors. Eur. J .
Pharmacol. 1996, 314, 365-372.
(17) Walsh, D. M.; Beattie, D. T.; Connor, H. E. The activity of 5-HT1D
receptor ligands at cloned human 5-HT1DR and 5-HT1Dâ receptors.
Eur. J . Pharmacol. 1995, 287, 79-84.
(18) Pauwels, P. J .; Tardif, S.; Palmier, C.; Wurch, T.; Colpaert, F.
C. How efficacious are 5-HT1B/1D receptors ligands at recombi-
nant human 5-HT1B and 5-HT1D receptors: an answer from
GTPγS binding studies with stably transfected C6-glial cell lines.
Neuropharmacology 1997, 36, 499-512.
(19) Mitchell, W. L.; Clitherow, J . W.; Glaxo Group Ltd. Brit. U.K.
Pat. Appl. GB 2,276,163; Chem. Abstr. 1995, 122, 9879p.
(20) Clitherow, J . W.; Scopes, D. I. C.; Skingle, M.; J ordan, C. C.;
Feniuk, W.; Campbell, I. B.; Carter, M. C.; Collington, E. W.;
Connor, H. E.; Higgins, G. A.; Beattie, D.; Kelly, H. A.; Mitchell,
W. L.; Oxford, A. W.; Wadsworth, A. H.; Tyers, M. B. Evolution
of a novel series of [(N,N-dimethylamino)propyl]- and pipera-
zinylbenzanilides as the first selective 5HT1D antagonists. J .
Med. Chem. 1994, 37, 2253-2257.
(21) Sonogashira, K.; Tohda, Y.; Hagihara, N. A convenient synthesis
of acetylenes: catalytic substitutions of acetylenic hydrogen with
bromoalkenes, iodoarenes and bromopyridines. Tetrahedron Lett.
1975, 4467-4470.
(22) Ishikura, M.; Ohta, T.; Terashima, M. A novel synthesis of 4-aryl-
and 4-heteroarylpyridines via diethyl(4-pyridyl)borane. Chem.
Pharm. Bull. 1985, 33, 4755-4763.
J M9702948