Novel 5-HT7 receptor inverse agonists. Synthesis and molecular modeling of arylpiperazine- and 1,2,3,4-tetrahydroisoquinoline-based arylsulfonamides.

Article abstract of DOI:10.1021/jm049743b

A series of arylpiperazine- and 1,2,3,4-tetrahydroisoquinoline-based arylsulfonamides was synthesized and evaluated for their interactions with the constitutively active 5-HT7 receptor. Effects on basal adenylate cyclase activity were measured using HEK-293 cells expressing the rat 5-HT7. All ligands produced a decrease of adenylate cyclase activity, indicative of their inverse agonism. Additionally, computational studies with a set of 22 inverse agonists, including these novel inverse agonists and inverse agonists known from literature, resulted in a pharmacophore model and a CoMFA model (R2 = 0.97, SE = 0.18). Docking of inverse agonists at the binding site of a model of the helical parts of the 5-HT7 receptor, based on the alpha carbon template for 7-TM GPCRs, revealed interesting molecular interactions and a possible explanation for observed structure-activity relationships.

Full text of DOI:10.1021/jm049743b

J . Med. Chem. 2004, 47, 5451-5466
5451
Novel 5-HT7 Recep tor In ver se Agon ists. Syn th esis a n d Molecu la r Mod elin g of
Ar ylp ip er a zin e- a n d 1,2,3,4-Tetr a h yd r oisoqu in olin e-Ba sed Ar ylsu lfon a m id es
Erik S. Vermeulen,*,§ Marjan van Smeden,§ Anne W. Schmidt,‡ J effrey S. Sprouse,‡ Ha˚kan V. Wikstro¨m,§ and
Cor J . Grol§
Department of Medicinal Chemistry, Center for Pharmacy, State University of Groningen,
A. Deusinglaan 1, NL-9713 AV Groningen, The Netherlands, and Pfizer Global R & D,
Department of Neuroscience, MS 8220-4178, Groton, Connecticut 06340
Received April 2, 2004
A series of arylpiperazine- and 1,2,3,4-tetrahydroisoquinoline-based arylsulfonamides was
synthesized and evaluated for their interactions with the constitutively active 5-HT7 receptor.
Effects on basal adenylate cyclase activity were measured using HEK-293 cells expressing the
rat 5-HT7. All ligands produced a decrease of adenylate cyclase activity, indicative of their
inverse agonism. Additionally, computational studies with a set of 22 inverse agonists, including
these novel inverse agonists and inverse agonists known from literature, resulted in a
pharmacophore model and a CoMFA model (R2 ) 0.97, SE ) 0.18). Docking of inverse agonists
at the binding site of a model of the helical parts of the 5-HT7 receptor, based on the R carbon
template for 7-TM GPCRs, revealed interesting molecular interactions and a possible
explanation for observed structure-activity relationships.
In tr od u ction
these inverse agonists have been used recently to derive
pharmacophore models for 5-HT7 receptor antago-
nism.26,27
In the process of characterization of the 5-HT7 recep-
tor, discovered over a decade ago,7,29,35,36,39 our attention
was drawn to the development ligands based on the
structure of 1,2,3,4-tetrahydroisoquinoline (THIQ) and
2-methoxyphenylpiperazine (2-MPP) linked with a spacer
to a number of divergent arylsulfonamides. The onset
to this approach was encouraged by the results obtained
from studies on the closely related 5-HT1A receptor.8,9,30-33
These studies revealed that replacement of the arylpip-
erazine moieties of known serotonergic drugs by the
THIQ nucleus resulted in potent inhibitors of subtypes
of the extensive family of 5-HT receptors. Simulta-
neously, the outcome of research projects in the area of
5-HT7 receptor antagonism suggested that aromatic
(sulfon-)amides with different alkyl chains as spacer
between this moiety and a positively charged nitrogen-
containing group proved to be potent ligands for this
receptor subtype.12,26,48 This supported the idea that
ligands could be developed based on both arylpipera-
zines and THIQs that would affect adenylate cyclase
activity mediated by the 5-HT7 receptor. Although its
(patho-)physiological role still remains unclear, it has
been demonstrated that the 5-HT7 receptor is constitu-
tively active, and that basal adenylate cyclase activity
was reduced by increased concentrations of 5-HT an-
tagonists (inverse agonists) in membranes of stable cell
lines expressing the human 5-HT7 receptors.22,21,44
Moreover, a number of recently developed ligands
showed a small apparent reduction in basal adenylate
cyclase activity in the absence of agonists during
pharmacological profiling as well,19,20,28,44 indicating
these compounds should be referred to as inverse
agonists instead of (neutral) antagonists. Several of
We wish to report the development and pharmaco-
logical characterization of novel 5-HT7 receptor inverse
agonist based on the structure of THIQ and 2-MPP and
present a pharmacophore model on the basis of inverse
agonists only. Additional computational studies (CoM-
FA, receptor modeling, docking of inverse agonist at the
binding site) suggested plausible explanations for ob-
served structure-activity relationships (SAR) and showed
interesting similarities with our recently published
pharmacophore model for 5-HT7 receptor agonism.47
Liga n d s. The geometry of the first pharmacophoric
hypothesis for 5-HT7 receptor antagonism26 was the
reference point for the development of two series of
ligands, based on 2-MPPs and THIQs. Additionally, a
small set of miscellaneous inverse agonists was collected
from previous in-house studies that were actually
designed as potential agonists, but unexpectedly de-
creased 5-HT7 receptor-mediated basal adenylate cy-
clase activity.
The chemical structure of 2-MPPs is well-known for
its affinity to several members of the family of seroton-
ergic receptors. Recently, it was demonstrated that
ligands possessing this structure interact with the 5-HT7
receptor as well.17,26,34 SAR studies with these ligands
revealed an optimal spacer length of four or five carbon
atoms to link the 2-MPPs with aryl-substituted ketones,
tetrahydrobenzindoles or naphtholactams and naph-
thosultams. However, aiming at ligands in which the
2-MPP moiety is linked to an aryl sulfonamide, it was
hypothesized that the spacer between the positively
ionizable nitrogen atom of the piperazine ring and the
sulfonamide moiety should consist of three carbon
atoms, matching the recently published selective 5-HT7
receptor ligands.12,13,15,28,45,46 Therefore 2-MPP-based
ligands 12 and 13a -c (Chart 1) were developed with
* To whom correspondence should be addressed. Tel: +31 50 311
8007, fax: +31 50 360 0390, e-mail: e.s.vermeulen@farm.rug.nl.
§ State University of Groningen.
‡ Pfizer Global R & D.
10.1021/jm049743b CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/23/2004

5452 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Vermeulen et al.
Ch a r t 1. Chemical Structure of the 5-HT7 Receptor Lig-
ands 1a -c, 2, 3a ,b, 9, 10a -c, 14a -h , 19, 20, 25, and 32
replacement for the arylpiperazine moiety in 5-HT1A
receptor ligands,8,9,30-33 and produce potent 5-HT7 re-
ceptor ligands when linked to tetrahydrobenzindoles as
well.18 Variation of the arylsulfonamide moiety coupled
to the THIQ nucleus by a C3 spacer would offer ad-
ditional information on the SAR of these ligands. In
addition, substitution of the sulfonamide nitrogen in
both series of novel 5-HT7 receptor ligands by a simple
ethyl substituent was investigated to mimic the steric
and lipophilic effect of the chiral pyrrolidine ring present
in the previously developed sulfonamides without the
necessity to separate enantiomers.
A small number of diverse ligands (19, 20, 25, 32,
Chart 1) from previous studies, which were tested for
their ability to affect 5-HT7 receptor-mediated adenylate
cyclase activity, appeared to decrease basal activity
levels as well. Finally, the highly rigid and potent 5-HT7
receptor inverse agonist (S)-methiothepin (2) and the
aporphine derivative 3b (Chart 1)25 served as templates
to select the active conformation of the far more flexible
ligands with long alkyl chains. Though at first sight
structurally not closely related, the results of the fitting
procedure indicate several striking similarities among
the chemical structures of the set of ligands used to
build the pharmacophore model.
Ch em istr y. The synthetic procedures leading to the
target compounds 12, 13a -c and 14a -h are illustrated
in Scheme 1. Synthesis of both series of sulfonamides
started off with either a simple nucleophilic substitution
of the secondary amines of 2-MPP or 6-methoxy-THIQ
(prepared by Pictet-Spengler cyclization of the 3-meth-
oxyphenethylamine with formaldehyde) with bromoac-
etonitrile or with a Michael addition with acrylonitrile.
While the methylene nitrile 6 was easily reduced to the
primary amine using only LiAlH4 in ether or THF,
application of the same conditions on the nitriles 7 and
8 obtained from Michael addition resulted in serious
differently substituted arylsulfonamides and a C3 (and
C2) alkyl chain connecting both moieties.
Analogously, a series of 5-HT7 receptor ligands was
developed containing the THIQ structure with a meth-
oxy substituent at the 6-position (14a -h , Chart 1). As
mentioned earlier, the structure had shown to be a good
Sch em e 1. Synthesis of Target Compounds 12, 13a -c and 14a -h a
a
Reagents and conditions: (a) Bromoacetonitrile, K2CO3, KI, acetonitrile, reflux. (b) Acrylonitrile, ethanol, reflux. (c) LiAlH4,
ether or THF, rt. (d) LiAlH4, AlCl3, ether, rt. (e) Ac2O, NaHCO3, (saturated aqueous solution), CH2Cl2. (f) LiAlH4, THF, reflux. (g)
ArSO2Cl, NaHCO3, (saturated aqueous solution), CH2Cl2.

Novel 5-HT7 Receptor Inverse Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5453
Sch em e 3. Synthesis of Target Compound 25a
Sch em e 2. Synthesis of Target Compounds 19 and 20a
a
Reagents and conditions: (a) NaCNBH3, AcOH. (b) Chloro-
a
acetyl chloride, K2CO3, CH2Cl2. (c) dimethylamine hydrochloride,
K2CO3, AcCN. (d) LiAlH4, AlCl3, THF.
Reagents and conditions: (a) Oxalyl chloride, diethyl ether.
(b) Dimethylamine hydrochloride, NaHCO3 (sat.)/CHCl3. (c) Li-
AlH4, THF. (d) KO-t-Bu, dimethyl oxalate, DMF, reflux. (e)
N-Benzylpiperidone, NaOCH3, CH3OH, reflux.
methoxyphenol was followed by ring closure under
alkaline conditions with 2-chloroacetyl chloride (Scheme
4). Subsequent reduction of 28 yielded the key inter-
mediate 29. Functionalization occurred via the same
pathway as applied for target compound 25. Again,
reduction in the final step turned out to be delicate, but
using the same modifications as before, the target
compounds could be obtained successfully.
amounts of cleavage, yielding primarily starting materi-
als. Therefore, the conditions were modified using 1
equiv of AlCl3 in combination with LiAlH4. Coupling of
the primary amine with the appropriate arylsulfonyl
chloride was straightforward. In case substitution of the
sulfonamide nitrogen was desired, acetylation and
subsequent reduction of the acetamide prior to coupling
with the arylsulfonyl chloride was preferred, due to
serious decrease of nucleophilicity of the sulfonamide
nitrogen. Coupling of the primary and secondary amines
with sulfonyl chlorides in an alkaline two-phase system
proceeded with complete conversion to the sulfonamide
end products. Overall yields of these reactions were in
the range of 60-75%.
The syntheses of indole-based target compounds 19
and 20 started with 5-methoxyindole (Scheme 2). A
classical tryptamine synthesis pathway41 was followed,
using oxalyl chloride and dimethylamine. Subsequent
reduction of the intermediate 17 yielded the 5-HT7
receptor agonist 18.47 Methylation of the indole-NH with
dimethyl sulfate under alkaline conditions gave the
target compound 19. Notably, when methylation of
5-methoxyindole was performed, followed by function-
alization with oxalyl chloride and dimethylamine, re-
duction of the intermediate 2-oxo-acetamide did not go
to completion, but stopped at the level of 2-hydroxy-
ethylamine, even at elevated temperatures and with a
large excess of LiAlH4.
3-D QSAR of 5-HT7 Recep tor In ver se Agon ists.
Using the set of ligands depicted in Chart 1, CoMFA
studies were performed to investigate three-dimensional
qualitative structure-activity relationships (3-D QSAR).
Therefore, the same methods were applied as recently
described.47 Full conformational analysis of the set of
5-HT7 receptor inverse agonists in their protonated form
was performed in MacroModel,2 and followed by a
pharmacophore identifying procedure through ligand
overlap using APOLLO.40 Since the nature of the set of
newly synthesized ligands is rather flexible due to the
presence of C2 and C3 spacers, the search for the active
conformation of ligands 1a -c, 12, 13a -c, 14a -h , 19,
20, 25, and 32 was guided by fitting these ligands with
the rigid structures of 2 and 3a ,b, and verified by the
recently published dimensions of the optimized phar-
macophore model for 5-HT7 receptor antagonism.
Both oxygen atoms of the sulfonamide containing
ligands were defined as H-bond-accepting moieties
(HBA). In case of absence of this moiety in 19, 20, 25,
and 32, the oxygen of the methoxy group at the
six-membered aromatic ring served this purpose. The
protonated, positively charged nitrogen atom present in
all ligands was defined as H-bond-donating (PI) moiety.
Centroids were defined for the aromatic ring next to the
sulfonamide moiety and the chiral five-membered het-
erocyclic rings (1a -c), while those ligands lacking this
moiety were equipped with a centroid through the six-
membered aromatic ring(s) of the indol(-in-)e (19, 20,
and 25), the benzoxazine (32), the aporphines (3a ,b),
and the (S)-methiothepin (2) core structure. In these
cases, the best fit of these centroids with the centroid
defined for the five-membered chiral pyrrolidine ring
of 1a -c was calculated. The superpositioned ligands
after the fitting procedure, and surrounded by their
mutual interaction points by means of water molecules,
are depicted in Figure 1. The set of superimposed
Ligand 20 was prepared in a single step using
N-benzylpiperidone in the presence of sodium methox-
ide.
Indoline 25 was prepared from 6-methoxyindole,
which was reduced with NaCNBH3 in glacial acetic acid
to the indoline (Scheme 3). Subsequent substitution of
the nitrogen atom with an ethylamine side chain
proceeded via the 2-chloroacetylated intermediate. Like
the reduction of nitriles 7 and 8, reduction of the amide
group at this stage required mild conditions using AlH3
(prepared from LiAlH4 and AlCl3) to prevent cleavage
of the amide bond.
Finally, the benzoxazine-based ligand 32 was syn-
thesized from the commercially available 2-nitro-4-
methoxyphenol in a straightforward process: catalytic
hydrogenation of the nitro group to the 2-amino-4-

5454 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Sch em e 4. Synthesis of Target Compound 32a
Vermeulen et al.
a
Reagents and conditions: (a) Pd/C, H2, EtOH. (b) 2-Chloroacetyl Chloride, K2CO3, AcCN, reflux. (c) LiAlH4, THF, rt. (d) 2-Chloroacetyl
chloride, pyridine (e) Amine, K2CO3, AcCN, rt. (f) LiAlH4, AlCl3, THF, rt.
F igu r e 1. Stereo picture of the best fit of ligands 1a -c, 2, 3a ,b, 12, 13a -c, 14a -h , 19, 20, 25 and 32. H-bonds with mutual
interaction points depicted in dotted lines. Nonessential hydrogen atoms have been omitted.
ligands was subsequently used for CoMFA computations
using the Sybyl molecular modeling software program3.
Recep tor Mod elin g: Dock in g of 5-HT7 Recep tor
In ver se Agon ists a t Bin d in g Site. The recently
published model of the helical parts of the 5-HT7
receptor,47 based on the template of R-carbon atoms of
the TM domains of GPCRs,5 was used to study the
molecular interactions of the inverse agonists at the
binding site of the model (see ref 47 for motivation).
Manual docking of the inverse agonist into the binding
site was guided by the 3-D orientation of the mutual
interaction points surrounding the set of ligands from
the APOLLO fitting procedure. The putative binding
site residue interacting with the HBA moieties of the
inverse agonists was oriented at the position of threo-
nine 244 (Thr5.43, numbering according to highly con-
served amino acid residues throughout family of GPCRs6)
of TM5, while the mutual interaction point of the
positively charged nitrogen atom (PI) was superposi-
tioned at aspartate 162 in TM3 (Asp3.32). Subsequent
minimization of the receptor-ligand complex resulted
in properly docked ligands and revealed several inter-
esting ligand binding interactions additional to the
calculated from the preceding alignment procedure by
APOLLO, were used to locate the ligand binding amino
acid residues in the receptor binding site. Careful
examination of this projection onto the inner sphere of
the binding site basically indicated the most distin-
guishing amino acid residues.
P h a r m a cology. All target compounds were evalu-
ated at Pfizer laboratories for their ability to compete
for [3H]5-CT binding to cloned rat 5-HT7 receptors
expressed in HEK-293 cells. Additionally, the effect on
adenylate cyclase activity relative to the effects of (S)-
methiothepin (2) was determined. Basal levels of cAMP
production equaled 24 pmol min-1 mg-1 protein. This
was reduced by 2 to 5 pmol min-1 mg-1 protein at a
ligand concentration of 1.0 × 10-5 M. This reduction
served as reference reduction level (100%) for all other
target compounds.
Resu lts a n d Discu ssion
Despite the heterogeneity of the chemical structures
of the 22 ligands used for the pharmacophore identifica-
tion for 5-HT7 receptor inverse agonism by APOLLO,
the calculations resulted in an ensemble of well-fitted
ligands surrounded by their mutual interaction points
mimicking the putative amino acid residues of the
binding site (Figure 1). Quite notable is the irregular
orientation of 20 with its benzyl-substituted tetrahy-
dropyridine ring out of line with the remainder of the
ligands. Additionally, the neighboring aromatic rings of
2 and 3a ,b appear to match well with the aromatic ring
of the sulfonamide containing ligands on one hand and
supposed interactions with Asp3.32 and Thr5.43
.
Recep tor Mod elin g: CoMF A Ma p p in g on to th e
5-HT7 Recep tor Bin d in g Site. The CoMFA model was
projected onto the binding site to identify conceivable
amino acid residues that could account for the contour
maps as computed during the comparative molecular
field analysis. For this purpose as well, the three-
dimensional orientation of mutual interaction points, as

Novel 5-HT7 Receptor Inverse Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5455
the chiral, five-membered heterocyclic ring of 1a -c on
the other hand. The variety in orientation of the
arylpiperazine and THIQ moieties and other 3D-QSAR
is reflected by the results of the CoMFA computations
discussed later in this section.
pharmacophore model that are respected by 1b, but
surpassed by 14e. With respect to these latter CoMFA
areas, it should be noted that the orientation of the
THIQ and arylpiperazine moieties is differing slightly
throughout the series and is influenced by the absence
or presence of an aliphatic substituent at the sulfona-
mide nitrogen atom. Those ligands that do not have
their hydrophobic substituent oriented precisely within
the boundaries of these areas (14c, 14e, and 14g)
significantly show lower affinity for the 5-HT7 receptor.
The lower value of the predictive power of this model
(Table 2, q2 ) 0.64, standard error of estimate ) 0.60
(cross-validated, set of ligands randomly divided into
two groups)) is most probably the result of the dis-
similarities of ligands 19, 20, 25, and 32 compared to
the large number of arylsulfonamide ligands.
Data listed in Table 1 clearly indicate that both the
1,2,3,4-tetrahydroisoquinoline and the arylpiperazine
target compounds can make good 5-HT7 receptor ligands
with potent inverse agonist characteristics. Generally,
the average values of the distances between the phar-
macophoric features of the ligands 1a -c, 2, 3a ,b, 12,
13a -c, 14a -h , 19, 20, 25, and 32 are close to the
corresponding values of recently published models.26,27
The data also indicate that the sulfonamides 12 and
13a -c act as potent inverse agonists with a C2 or a C3
alkyl chain bridging the arylpiperazine moiety and the
arylsulfonamide moiety. This latter geometry matches
well with the corresponding distance between HYD4
and PI of the initial pharmacophore model,26 but
contrasts with the optimum value of four to five carbon
atoms reported earlier.17,26,34
The mapping of the CoMFA areas onto the model of
the TM domains of the 5-HT7 receptor offers a practical
method of correlating both models (Figure 4). The yellow
areas surrounding the 2-MPP and THIQ moieties
indicate the boundaries of the lipophilic pocket hosting
HYD2 and HYD3 of the pharmacophore model and can
be superpositioned perfectly onto side chains of amino
Another important SAR that can be deducted from
the current series of ligands is the presence of the
hydrophobic substituent at the nitrogen of the sulfona-
mide (HYD1). Ligands lacking this feature (12, 13a ,
14c, 14e, and 14g) show considerable lower affinities
than their ethyl-substituted analogues. As for the dif-
ferent aryl groups of the arylsulfonamides, no large
differences in binding affinities can be detected, except
for 1b, which has a hydrogen bond (H-bond)-donating
group at the meta-position relative to the sulfonamide
moiety.
acid residues Ile159 (Ile3.29) of TM3 and Phe369 (Phe7.38
)
and Leu370 (Leu7.39) of TM7. Additionally, the red area
indicating the favorable effect of electronegative charge
in the vicinity of the oxygen atoms of the sulfonamide
moiety are projected nicely onto the side chain of the
H-bond-donating residue Thr244 of TM5 (Thr5.43), while
the green area indicating the favorable effect of bulk,
representing HYD1, is close to TM6 and surrounded by
Phe336 (Phe6.44) at the bottom and Trp340 (Trp6.48) at
the top of this lipophilic pocket.
In the case of the aporphine ligands 3a ,b, it was
hypothesized that the nitrogen atom of the nitrile
substituent could serve as H-bond-accepting moiety.
However, the different orientation of this substituent
and the correlated increased distance between this
moiety (HBA) and the positively charged nitrogen and
its substituent (PI and HYD2, respectively) in the case
of the more potent atropisomers 3b indicate this inter-
action is of no concern in this case, or the positive effect
of H-bond formation of 3a is canceled out by the methyl
substituent when oriented in the opposite direction.
In close-up, after minimization of the receptor-ligand
complex, the molecular interactions of 1b and the
binding site of the 5-HT7 receptor are depicted in Figure
5. The electrostatic interactions between the protonated
nitrogen atom and the negatively charged Asp162
(Asp3.32), the sulfonamide oxygen and the hydroxyl
group of Thr244 (Thr5.43), and in particular the hydroxyl
group at the meta position of the six-membered aromatic
ring of 1b and the side chain of Thr337 (Thr6.45) are
depicted in dotted lines. This latter H-bonding feature
was identified earlier by SAR studies in which the
hydroxyl-substituted 1b showed a substantial increase
in 5-HT7 receptor affinity compared to its methoxy-
substituted analogue.28 From the current molecular
modeling studies it could be suggested that the latter
methoxy-substituted analogue could accept a H-bond
donated by Thr6.45 as well, but the favorable effect of
this H-bond is most probably abolished by steric repul-
sions between the methyl group and TM6. Interestingly,
in case of the aporphine ligand 3b, it was observed after
minimization of the receptor-ligand complex that a
H-bond was formed as well between the nitrogen atom
of the nitrile substituent and Thr337 (Thr6.45) instead
of Thr5.43 (which forms a H-bond with the HBA groups
of all other ligands). This might explain the higher
binding affinity compared to its atropisomer 3a . The
conformation of 3a would be capable of forming a
H-bond with Thr5.43, but steric repulsion of the methyl
substituent with the relatively narrower inter helical
space between TM5 and TM6 eventually results in a
lower affinity for the 5-HT7 receptor than 3b.
The computational studies performed with the set of
ligands listed in Table 1 resulted in a CoMFA model
with a good correlation between actual and predicted
pKi values (Figure 3, R2 ) 0.97, standard error of
estimate ) 0.18 (not cross-validated)). The optimum
number of components equaled 6, and 52 out of 2808
columns were dropped during PLS analysis. The ratio
of steric and electrostatic effects equaled 0.60:0.40.
Stereo pictures of ligand 1b and 14e (Figure 2, upper
and lower panel respectively) are illustrative.Both
ligands have the oxygen atoms closely oriented toward
the areas indicative of the favorable effect of electro-
negative charge. Furthermore, the five-membered pyr-
rolidine ring of 1b perfectly occupies the area indicating
the favorable effect of the hydrophobic group (HYD1),
which is absent in case of 14e. An additional charac-
teristic of the low affinity ligand 14e is the orientation
of the substituted THIQ moiety far into the areas that
identify the negative influence of bulky groups in that
area. Obviously, these areas indicate the boundaries of
the lipophilic pocket hosting HYD2 and HYD3 of the

5456 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Vermeulen et al.

Novel 5-HT7 Receptor Inverse Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5457
F igu r e 2. Stereo pictures of ligand 1b (upper panel) and 14e (lower panel) surrounded by CoMFA fields. Nonessential hydrogen
atoms have been omitted.
role in receptor-ligand complex stabilization. This type
of interaction is known for its significant role in stabiliz-
ing local structures of proteins as well.42 Ligands 19,
20, 25, and 32 that miss an aromatic ring similar to
the arylsulfonamides and therefore are unable to form
stabilizing interactions with Phe5.47, show poor binding
affinities. The shortage of this type of interactions with
the binding site of the receptor is even more pronounced
as a result of the absence of lipophilic substituents
oriented near HYD2 and HYD3 of the pharmacophore
model. Although 20 possesses a benzyl group that could
achieve π-π stacking interactions with Phe7.38, the
orientation of this group is far from optimal due to the
conformations of the tetrahydropyridine ring structure
(Figure 1), resulting in serious repulsive interactions
with the boundaries of the lipophilic pocket instead.
In line with the results of this study, it is hypothesized
that the pharmacophore for 5-HT7 receptor inverse
agonism consists of four hydrophobic domains (HYD1-
4) instead of the three domains recently proposed by
Lopez-Rodriguez et al in the model for 5-HT7 receptor
antagonism.27 The fourth hydrophobic domain can host
the aromatic ring of the arylsulfonamide moiety present
in many inverse agonists (Chart 1). The dimensions of
the extended pharmacophore model for 5-HT7 receptor
inverse agonism are listed in Table 1 and graphically
illustrated in Figure 6.
F igu r e 3. Correlation between actual and predicted pKi
values based on CoMFA computations.
Ta ble 2. Partial Least Squares CoMFA Calculations
cross-validated run no.
(two groups)
optimum no. mean SE of
R2
components
predictions
1
2
3
4
0.64
0.59
0.61
0.63
4
3
4
3
0.60
0.66
0.62
0.63
non-cross-validated run 1
0.97
6
0.18
Additional stabilizing interactions are contributed to
Phe248 of TM5 (Phe5.47) by means of π-π stacking with
the six-membered aromatic ring of 5-HT7 receptor
inverse agonists. Similar stabilizing interactions can be
identified between Phe369 of TM7 (Phe7.38) and the six-
membered aromatic rings of THIQ and 2-methoxyphen-
yl piperazine moieties of the inverse agonists. In case
of 1b, it is suggested that C-H‚‚‚π interactions between
Phe7.38 and the piperidine ring plays a supplementary
Comparison of the currently extended pharmacophore
model for 5-HT7 receptor inverse agonism and the
recently published pharmacophore model for 5-HT7
receptor agonism47 reveals a number of interesting
similarities. The latter model (Figure 6, right panel)
consists of a hydrophobic domain hosting the six-
membered aromatic rings of indoles, 2-aminotetralins,
and ergolines (HYD1), a region hosting a H-bond-

5458 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Vermeulen et al.
F igu r e 4. Stereoviews of CoMFA projections onto binding site of 5-HT7 receptor with 1b as an example. Top view (upper panel)
and site view (lower panel). Nonessential hydrogen atoms have been omitted.
F igu r e 5. Close-up of molecular interactions between 1b and binding site of 5-HT7 receptor. Nonessential hydrogen atoms have
been omitted.
donating (positively charged) nitrogen atom (PI), and a
H-bond-accepting domain (HBA). SAR studies addition-
ally revealed an extra hydrophobic domain (HYD2) for
ligands that have an aromatic substituent instead of a
substituent capable of accepting a H-bond from the
hypothesized binding site residue. The dimensions of
both models turn out to be very similar. While even the
distance between HYD1 and HYD2 of the pharmaco-

Novel 5-HT7 Receptor Inverse Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5459
F igu r e 6. Pharmacophore models for 5-HT7 receptor inverse agonism (middle panel) and 5-HT7 receptor agonism (right panel).
Ligand 1b with projection of pharmacophore model for inverse agonism (left panel). Nonessential hydrogen atoms have been
omitted.
phore model for agonism on one hand, and the corre-
sponding distance between HYD1 and HYD4 of the
pharmacophore model for inverse agonism, are nearly
identical, the main difference between both models
appears to be the presence of the additional HYD2 and
HYD3 in the model for inverse agonism. This observa-
tion indirectly hints of the possibility of converting
agonists into inverse agonists by connection of the
protonatable nitrogen atom with substituents that can
occupy these hydrophobic domains. Likewise, it was
hypothesized that occupation of both the domains HBA
and HYD2 of the pharmacophore model for 5-HT7
receptor agonism could result in ligands with enhanced
receptor affinity as well, compared to agonists occupying
only either one of these domains. On the contrary,
simultaneous occupation of these domains could also be
(part of) the condition for 5-HT7 receptor inverse ago-
nism, since the majority of the ligands of the current
set of 22 ligands demonstrate this simultaneous occupa-
tion of the corresponding domains HBA and HYD4 of
the pharmacophore model for 5-HT7 receptor inverse
agonism.
Con clu sion
The synthesis and pharmacological evaluation of a
comprehensive set of novel inverse agonists has offered
additional information in the process of characterization
of the 5-HT7 receptor. Especially the N-ethyl-substituted
sulfonamides 13b, 14b, 14d , 14f, and 14h appeared to
be potent inhibitors of basal adenylate cyclase activity.
Computational studies on the basis of these novel
inverse agonists and a selection of previously published
ligands resulted in a CoMFA model with a good cor-
relation between experimental and predicted pKi values,
and a pharmacophore model that shows similarities
with recently published pharmacophore models for
5-HT7 receptor agonism and antagonism. In case of the
5-HT7 receptor, there are reasonable indications that
agonists and inverse agonists occupy the same binding
site, but additional pharmacophoric features give rise
to different conformational changes of the receptor
resulting in activation by agonists and deactivation by
inverse agonists.
Exp er im en ta l Section
The exact mechanism of action of ligands interacting
with the 5-HT7 receptor remains yet unrevealed. The
residues Asp162 (Asp3.32) and Thr244 (Thr5.43) are
important elements in binding the ligands to this site,
but the presence of additional substituents appears to
be conclusive with respect to the mode of action. It was
demonstrated that both 5-HT7 receptor agonists and
inverse agonists show interactions with aromatic resi-
dues of TM6 (Phe6.44, Trp6.48, Phe6.51, and Phe6.52). In
Molecu la r Com p u t a t ion s. Con for m a t ion a l An a lysis.
Calculations of physical properties (log P, pKa) were performed
with Pallas.1 All molecular computations were performed on
a Silicon Graphics O2 workstation running IRIX 6.3 or a
Silicon Graphics IRIS Indigo XS/4000 workstation running
IRIX 5.3. For conformational analyses, ligands were sketched
with the correct stereochemistry, if present, in their proto-
nated, positively charged state from standard fragments in
MacroModel.2 Structures were minimized with default options
prior to full conformational analysis. Conformational analyses
were performed using the Monte Carlo Multiple Minimum
(MCMM) search protocol10 with a minimum of 1500 steps via
the SUMM option.14 Minimizations were performed using the
Truncated Newton Conjugate Gradient (TNCG) minimization
method within the MM2 force field4,23,24 with simulation of a
distance dependent water continuum43 as implemented in
MacroModel. Per step, 250 iterations were performed until the
gradient reached the value of 0.01 kcal Å-1 mol-1. All confor-
mations within a range of 50 kJ /mol above the global minimum
were considered to be relevant. In case of nonchiral ligands,
mirror images were preserved using the NANT option. Ad-
ditional information needed for subsequent calculations with
the pharmacophore identifying software program APOLLO
was obtained through incorporation of DEBG options 3 and
28. Torsion bonds of n-propyl chains were fixed in their anti-
case of the dopamine D2 and the 5-HT2A receptor,38 it
16
has been hypothesized that this cluster plays an im-
portant role in the mechanism of activation. Molecular
interactions between a ligand, and this cluster would
cause a conformational change at the so-called proline
kink, effectuating TM6 to bend and giving the op-
portunity for TM6 and TM3 to move apart. Since this
cluster of residues is highly conserved throughout the
family of GPCRs, it is very likely that such a confor-
mational change also plays a role in activation of the
5-HT7 receptor by agonists and deactivation by inverse
agonists.

5460 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Vermeulen et al.
staggered conformation to reduce the number of low energy
conformations. Molecular dimensions and atomic distances
were determined by manual selection of the appropriate atomic
features in the Analyze mode.
dimensional orientation of mutual interaction points in the
most optimal manor. Minimization of the merged complex of
ligand and receptor (Tripos force field, Gasteiger-Hu¨ckel
charges, NB cutoff 8.0, dielectric constant 5.0, distance de-
pendent, conjugate gradient 0.1 kcal/mol/Å) resulted in prop-
erly docked ligands.
P h a r m a cop h or e Id en tifica tion . The VECADD module of
APOLLO40 was used to define extension vectors and centroids
to all low energy conformations of all ligands as calculated by
MacroModel in the previously described section. Extension
vectors were allocated to the lone-pairs of the oxygen atoms
of substituents (if present) of the six-membered aromatic ring
present in all ligands, and the protonated, positively charged
nitrogen atom also present in all ligands. The RMSFIT module
of APOLLO subsequently identified the conformations of the
ligands that exhibited the best overall least-squares fit with
respect to the specified extension vectors and centroids. This
calculation was performed in an energy dependent manor,
while all fitting points were considered equally important. In
case multiple solutions were calculated, the matches were
ranked with respect to conformational energies and root-mean-
square deviations. The best fit was extracted from the RMSFIT
output file using the MMDFIT module of APOLLO.
CoMF A Com p u ta tion s. Selected conformations of the
ligands were imported into a Sybyl3 database as pdb-files as
selected by the APOLLO molecular modeling and extracted
from the RMSFIT output file using the MMDFIT module.
Atom types were checked and adjusted if necessary, extension
vectors and centroids were deleted, and atomic charges were
calculated (Gasteiger-Hu¨ckel). On the basis of these database,
a new MSS was opened, and pKi values were entered manu-
ally. CoMFA parameters calculated with Sybyl standard
parameters and with steric and dielectric cutoff values of 15
and 10 kcal/mol, respectively (distance dependent dielectric
function), were inserted as separate column. Partial least
squares (PLS) calculations (non-cross-validated) resulted in the
described CoMFA model, which was subsequently used for
mapping onto the binding site of the 5-HT7 receptor to identify
conceivable amino acid residues.
Hom ology Mod elin g of th e Recep tor . The sequence of
449 amino acids of the human 5-HT7A receptor was taken from
the SwissProt database (entry P34969; http://us.expasy.org/
sprot/). On the basis of the identification of evolutionary highly
conserved amino acids within the rhodopsin-based family of
G-protein-coupled receptors, the helical parts of the receptor
were manually aligned with a template of R-carbon atoms of
the transmembrane domains.5 Initial minimization of the
helices separately (Tripos force field, Gasteiger-Hu¨ckel charges,
dielectric constant 5.0, distance dependent, conjugate gradient
0.1 kcal/mol/Å), and subsequent minimization of the ensemble
of the 7 transmembrane helices resulted in the model with a
total energetic value lower than the sum of energies of the
individually minimized helices.
Ma p p in g of CoMF A Mod el on t o R ecep t or Bin d in g
Site. The CoMFA model was projected onto the binding site
to identify conceivable amino acid residues that could account
for the contour maps as computed during the comparative
molecular field analysis. For this purpose, the three-dimen-
sional orientation of mutual interaction points, as calculated
from the preceding alignment procedure by APOLLO, were
used to locate the ligand binding amino acid residues in the
receptor binding site. The mutual interaction point forming
H-bonds with the HBA moieties of the ligands was superim-
posed onto Thr244 (Thr5.43), and the H-bond-accepting interac-
tion point forming H-bonds with the PI moieties of the ligands
was superimposed onto Asp162 (Asp3.32).
Ch em istr y. Gen er a l Rem a r k s. The purity of the target
compounds was established using multiple techniques. IR
spectra were recorded on an ATI-Mattson Genesis Series FT-
IR spectrophotometer. NMR data were recorded using Varian
Gemini-200 and Varian VXR-300 spectrometers (1H NMR at
200 or 300 MHz, 13C NMR at 50 or 75 MHz). Chemical shifts
are denoted in δ units (ppm) relative to the solvent (1H NMR
peaks: 7.26 for CDCl3 and 3.30 for CD3OD. 13C NMR peaks:
76.91 for CDCl3 and 49.50 for CD3OD). Electron impact (EI+)
mass spectra were recorded on a Unicam Automass mass
spectrometer in conjunction with a gas chromatograph. TLC
analyses were carried out on aluminum plates (Merck) pre-
coated with silica gel 60 F254 (0.2 mm). Visualization of spots
was performed with UV light and a commonly used alkaline
KMnO4 spray solution or Silica coated I2. Elemental analyses
(C, H, N) of target compounds were performed at the Univer-
sity of Groningen.
[4-(2-Met h oxyp h en yl)p ip er a zin -1-yl]a cet on it r ile (6).
2-Methoxyphenylpiperazin hydrochloride (0.23 g, 1.00 mmol)
was suspended in acetonitrile under nitrogen atmosphere.
Bromoacetonitrile (0.14 g, 1.10 mmol) and K2CO3 (0.35 g, 2.50
mmol) were added, followed by the addition of KI (cat. amount)
and DMF (1 drop). The mixture was heated to a gentle reflux
overnight. The cooled suspension was filtered and concentrated
in vacuo. The residue was partioned between CH2Cl2 (30 mL)
and water (10 mL). The organic layer was subsequently dried
(Na2SO4), filtered, and concentrated to yield a brown oil (0.23
g, 99% yield) that was purified by chromatography (Silica, CH2-
1
Cl2/CH3OH, 19:1). H NMR (300 MHz, CDCl3) δ (ppm): 2.74
(t, J ) 4.8 Hz, 4H), 3.07 (br, 4H), 3.49 (s, 2H), 3.80 (s, 3H),
6.80-6.87 (m, 3H), 6.93-6.97 (m, 1H). 13C NMR (75 MHz,
CDCl3) δ (ppm): 43.5, 47.7, 49.6, 53.0, 108.9, 112.4, 115.8,
118.5, 120.7, 138.3, 149.8. IR (neat, cm-1): 2940, 2826, 2231,
1675, 1593, 1501, 1452, 1382, 1302, 1239, 1140, 1012, 927, 861,
751. MS (EI+): 231, 191, 136, 92, 56 (100%).
3-[4-(2-Meth oxyph en yl)piper azin -1-yl]pr opion itr ile (7).
2-Methoxyphenylpiperazine (0.55 g, 2.86 mmol) was dissolved
in ethanol (10 mL) and acrylonitrile (0.50 g, 9.40 mmol). The
mixture was heated to reflux for 1 h. The volatiles were
removed under reduced pressure to yield the crude product
as colorless crystals (0.64 g, 97% yield). 1H NMR (300 MHz,
CDCl3) δ (ppm): 2.50 (t, J ) 7.0 Hz, 2H), 2.64-2.75 (m, 6H),
3.05 (br, 4H), 3.81 (s, 3H), 6.80-6.98 (m, 4H). 13C NMR (75
MHz, CDCl3) δ (ppm): 13.3, 48.0, 50.4, 51.0, 52.8, 108.7, 115.7,
118.5, 120.6, 148.5. IR (neat, cm-1): 2945, 2829, 2249, 1591,
1501, 1449, 1239, 1141, 1024, 935, 751. MS (EI+): 245, 205,
177, 120, 70 (100%). Mp (uncorrected): 83 °C.
2-[4-(2-Meth oxyp h en yl)p ip er a zin -1-yl]eth yla m in e (9).
Nitrile 6 (0.23 g, 1.00 mmol) was dissolved in THF (dry, 10
mL) and added dropwise to a freshly prepared suspension of
LiAlH4 (0.08 g, 2.10 mmol) in THF (dry, 10 mL) under nitrogen
atmosphere. The mixture was stirred at room temperature for
1 h. Careful workup by subsequent addition of water (1 equiv),
NaOH (1 N, 1 equiv) and water (4 equiv). Na2SO4 was added
and the suspension was then filtered over Celite. The clear
solution was concentrated in vacuo. The oil thus obtained (0.19
1
g, 81% yield) was used without further purification. H NMR
(300 MHz, CDCl3) δ (ppm): 1.41 (br, 2H), 2.40 (t, J ) 5.9 Hz,
2H), 2.56 (br, 4H), 2.73 (br, 2H), 3.00 (br, 4H), 3.75 (s, 3H),
6.75-6.93 (m, 4H). 13C NMR (75 MHz, CDCl3) δ (ppm): 48.2,
51.0, 52.8, 58.7, 108.6, 115.6, 118.4, 120.3, 138.8, 149.7. IR
(neat, cm-1): 2939, 2816, 1590, 1498, 1450, 1240, 1145, 1030,
749. MS (m/z): 236 (M + 1).
3-[4-(2-Methoxyphenyl)piperazin-1-yl]propylamine (10a).
The nitrile 7 (0.24 g, 1.00 mmol) was dissolved in diethyl ether
(dry, 10 mL) and added dropwise to a freshly prepared
suspension of AlCl3 (0.27 g, 2.00 mmol) and LiAlH4 (0.08 g,
2.10 mmol) in diethyl ether (dry, 15 mL) under nitrogen
Dock in g of Liga n d s in Bin d in g Site. Ligands were
manually docked into the binding site as described and
minimized to analyze the ligand binding interactions. This
docking procedure was guided by mapping the mutual interac-
tion points as calculated from the preceding alignment pro-
cedure by APOLLO onto the previously mentioned amino acid
residues. As a starting point for minimization, the conforma-
tions of the ligands used for superpositioning in APOLLO were
used, and the side chains of the hypothesized amino acid
residues were adjusted, if necessary, to adopt the three-

Novel 5-HT7 Receptor Inverse Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5461
atmosphere. The mixture was allowed to stir at room temper-
ature for 1 h. Careful addition of water decomposed the excess
of reactants. NaOH (50 mL, 0.5 N) was added and extracted
with CH2Cl2 (4 × 20 mL). The combined organic layers were
dried (Na2SO4), filtered, and concentrated in vacuo to yield 10a
as a colorless oil (0.21 g, 84% yield) which was used without
further purification. 1H NMR (300 MHz, CDCl3) δ (ppm): 1.35
(br, 2H), 1.54 (t, J ) 7.0 Hz, 2H), 2.33-2.38 (m, 2H), 2.54 (br,
4H), 2.62-2.66 (m, 2H), 2.98 (br, 4H), 3.73 (s, 3H), 6.71 (d, J
) 7.7 Hz, 1H), 6.77-6.89 (m, 3H). 13C NMR (75 MHz, CDCl3)
δ (ppm): 28.0, 38.3, 48.1, 51.0, 52.8, 54.1, 108.6, 115.6, 118.4,
120.3, 138.8, 149.7. IR (neat, cm-1): 3365, 3286, 2938, 2814,
1593, 1500, 1451, 1376, 1301, 1240, 1144, 1028, 927, 748. MS
(m/z): 250 (M + 1).
Method A: preparation of acetamides 10b and 11b, and
sulfonamides 12, 13a -c, and 14a -h : Primary amines were
dissolved in a vigorously stirred mixture of CH2Cl2 (generally
20 mL) and NaHCO3 (sat. solution, generally 10 mL). After
addition of the acetic anhydride or arylsulfonyl chloride,
stirring was applied for 30-300 min at room temperature.
Then, the layers were separated and the water layer was
washed with CH2Cl2. The combined organic layers were dried
(Na2SO4) and filtered. Evaporation of the volatiles yielded the
crude products.
(0.08 g, 0.26 mmol) was dissolved in THF (dry, 5 mL) and
added dropwise to a freshly prepared suspension of LiAlH4
(0.02 g, 0.53 mmol) in THF (dry, 5 mL) under nitrogen
atmosphere. The mixture was warmed to reflux for 2 h. Excess
of reactant was decomposed by addition of water (25 mL). The
water layer was extracted with CH2Cl2 (3 × 10 mL), and the
combined organic layers were dried over Na2SO4 and filtered.
Evaporation of volatiles yielded the intermediate amine 10c
as colorless oil (0.08 g, 99% yield). MS (EI+): 277, 262, 219,
190, 162, 120, 70, 58 (100%). The oil was dissolved in CH2Cl2
(5 mL), and p-toluenesulfonyl chloride (0.10 g, 0.53 mmol) was
added. NaHCO3 (5 mL, sat) was added, and the mixture was
stirred vigorously for 4 h at room temperature. The layers were
separated, and the water layer was washed with CH2Cl2 (5
mL). The combined organic layers were dried (Na2SO4),
filtered, and concentrated in vacuo. This yielded an oil that
was purified by chromatography (Silica, CH2Cl2/CH3OH, 98:
2) to yield pure 13b as an oil (0.09, 80% yield). The free base
was converted to its oxalate salt by treatment with oxalic acid
1
dihydrate in ethanol. H NMR (free base, 300 MHz, CDCl3) δ
(ppm): 1.03-1.08 (t, J ) 7.0 Hz, 3H), 1.74 (dt, J ) 7.3 Hz,
2H), 2.33-2.38 (m, 5H), 2.56 (br, 4H), 3.02 (br, 4H), 3.11-
3.21 (m, 4H), 3.80 (s, 3H), 6.80 (d, J ) 8.1 Hz, 1H), 6.85-6.96
(m, 3H), 7.23 (d, J ) 8.1 Hz, 2H), 7.64 (d, J ) 8.1 Hz, 2H). 13
C
NMR (75 MHz, CDCl3) δ (ppm): 11.5, 19.0, 23.8, 40.4, 43.2,
48.1, 50.9, 52.8, 53.1, 108.7, 115.6, 118.5, 120.4, 124.6, 127.1,
134.6, 138.8, 140.5, 149.7. IR (neat, cm-1): 3280, 2948, 2818,
1598, 1500, 1450, 1328, 1239, 1154, 1092, 1025, 830, 752.MS
(EI+): 431, 276, 205 (100%), 155, 120, 91 (100%), 56. Mp
(uncorrected): 178-180 °C. Anal. (C23H33N3O3S‚ C2H2O4) C,
H, N.
4-Meth oxy-N-{3-[4-(2-m eth oxyp h en yl)p ip er a zin -1-yl]-
p r op yl}-ben zen esu lfon a m id e (13c). Method A was applied,
using 10a (0.05 g, 0.20 mmol) and p-toluenesulfonyl chloride
(0.08 g, 0.35 mmol). This yielded an oil (0.07 g, 83% yield) that
was purified by chromatography (Silica, CH2Cl2/CH3OH, 99:
1) to yield pure 13c as oil. The free base was converted to its
oxalate salt by treatment with oxalic acid dihydrate in ethanol.
1H NMR (free base, 300 MHz, CDCl3) δ (ppm): 1.19 (br, 1H),
1.60 (dt, J ) 5.5 Hz, 2H), 2.40 (t, J ) 5.9 Hz, 2H), 2.54 (br,
4H), 3.01 (t, J ) 5.9 Hz, 6H), 3.79 (s, 6H), 6.79 (d, J ) 7.7 Hz,
1H), 6.86-6.95 (m, 5H), 7.73 (d, J ) 7.0 Hz, 2H). 13C NMR
(75 MHz, CDCl3) δ (ppm): 21.5, 41.7, 48.2, 50.8, 52.9, 53.1,
55.6, 108.6, 111.6, 115.9, 118.5, 120.6, 126.5, 129.4, 138.4,
149.7, 160.1. IR (neat, cm-1): 3281, 2944, 2820, 1596, 1499,
1454, 1327, 1241, 1155, 1095, 1026, 834, 749. MS (EI+): 419,
404, 205, 171, 140, 107, 77 (100%). Mp (uncorrected): 183-
186 °C. Anal. (C21H29N3O4S‚C2H2O4) C, H, N.
3-(6-Met h oxy-3,4-d ih yd r o-1H -isoq u in olin -2-yl)p r op i-
on itr ile (8). 6-Methoxy-1,2,3,4-tetrahydroisoquinoline hydro-
chloride (5,11 5.00 g, 25.1 mmol) was dissolved in ethanol (100
mL) and treated with triethylamine (2.54 g, 25.1 mmol) and
an access of acrylonitrile (4.00 g, 75.4 mmol). The mixture was
heated to reflux for 5 h. The volatiles were removed under
reduced pressure, and the residue was partioned between CH2-
Cl2 and water. The organic layer was washed with brine, dried
(Na2SO4), and filtered. Evaporation of the solvent yielded the
crude product, which could be purified by chromatography
(Silica, CH2Cl2/CH3OH, 99:1) to yield pure 8 as a slightly
yellow oil (5.21 g, 96% yield). 1H NMR (200 MHz, CDCl3) δ
(ppm): 0.99 (m, 1H), 1.26 (s, 1H), 1.43 (s, 1H), 1.63-1.82 (m,
1H), 2.59 (t, J ) 6.8 Hz, 2H), 2.66-2.79 (m, 2H), 2.87 (s, 1H),
3.57 (s, 1H), 3.76 (s, 3H), 6.60-6.70 (m, 2H), 6.91-6.98 (m,
1H). 13C NMR (50 MHz, CDCl3) δ (ppm): 28.2, 28.8, 39.3, 49.3,
53.7, 54.2, 55.1, 110.6, 111.6, 124.0, 126.0, 133.8, 156.5. IR
(neat, cm-1): 2930, 2834, 2247, 1610, 1506, 1272, 1146, 1037,
811. MS (m/z): 217 (M + 1).
N -{3-[4-(2-Me t h oxyp h e n yl)p ip e r a zin -1-yl]p r op yl}-
a cet a m id e (10b). Method A was applied, using crude 10a
(0.16 g, 0.65 mmol) and acetic anhydride (0.10 g, 1.00 mmol).
This yielded crude 10b (0.18 g, 95% yield) as a slightly yellow
oil that crystallized upon standing. 1H NMR (300 MHz, CDCl3)
δ (ppm): 1.66 (dt, J ) 6.2 Hz, 2H), 1.90 (s, 3H), 2.14 (br, 1H),
2.50 (t, J ) 6.2 Hz, 2H), 2.63 (br, 4H), 3.04 (br, 4H), 3.28-
3.34 (m, 2H), 3.81 (s, 3H), 6.81 (d, J ) 7.7 Hz, 1H), 6.86-6.88
(m, 1H), 6.90-6.99 (m, 1H), 7.15 (br, 1H). 13C NMR (75 MHz,
CDCl3) δ (ppm): 21.0, 22.4, 37.3, 48.3, 50.9, 52.9, 55.2, 108.7,
115.6, 118.5, 120.6, 138.5, 149.7, 167.3. IR (neat, cm-1): 3303,
2946, 2812, 1639, 1562, 1501, 1438, 1376, 1303, 1240, 1145,
1026, 743. MS (m/z): 292 (M + 1).
N-{2-[4-(2-Meth oxyp h en yl)p ip er a zin -1-yl]eth yl}-4-m e-
th ylben zen esu lfon a m id e (12). Method A was applied, using
9 (0.19 g, 0.80 mmol) and p-toluenesulfonyl chloride (0.17 g,
0.90 mmol). This yielded a yellow oil (0.27 g, 87% yield) that
was purified by chromatography (Silica, CH2Cl2/CH3OH, 99:
1) to yield pure 12 as oil. The free base was converted to its
oxalate salt by treatment with oxalic acid dihydrate in ethanol.
Recrystallization form ethanol yielded the pure compound. 1H
NMR (free base, 300 MHz, CDCl3) δ (ppm): 2.42-2.50 (m, 9H),
3.03-3.10 (m, 6H), 3.84 (s, 3H), 5.28 (br, 1H), 6.84-7.04 (m,
4H), 7.32 (d, J ) 8.6 Hz, 2H), 7.78 (d, J ) 8.3 Hz, 2H). 13C
NMR (75 MHz, CDCl3) δ (ppm): 20.1, 37.8, 49.0, 51.2, 53.9,
54.2, 109.7, 116.7, 119.5, 121.6, 125.7, 128.2, 135.1, 139.6,
141.9, 150.7. IR (neat, cm-1): 3300, 2954, 2810, 1596, 1499,
1449, 1326, 1235, 1154, 1091, 1024, 834, 747. MS (EI+): 389,
205 (100%), 190, 162, 120, 92, 70. Mp (uncorrected): 162-
164 °C. Anal. (C20H27N3O3S_C2H2O4) C, H, N.
N-{3-[4-(2-Met h oxyp h en yl)p ip er a zin -1-yl]p r op yl}-4-
m eth ylben zen esu lfon a m id e (13a ). Method A was applied,
using 10a (0.12 g, 0.48 mmol) and p-toluenesulfonyl chloride
(0.11 g, 0.60 mmol). This yielded a yellow oil (0.19 g, 98% yield)
that was purified by chromatography (Silica, CH2Cl2/CH3OH,
98:2) to yield pure 13a as oil. The free base was converted to
its oxalate salt by treatment with oxalic acid dihydrate in
ethanol. 1H NMR (free base, 300 MHz, CDCl3) δ (ppm): 1.55-
1.60 (m, 2H), 2.34 (s, 3H), 2.37-2.40 (t, J ) 5.9 Hz, 3H), 3.53
(br, 4H), 2.99-3.02 (m, 6H), 3.78 (s, 3H), 6.79 (d, J ) 7.7 Hz,
1H), 6.83-6.96 (m, 3H), 7.22 (d, J ) 8.1 Hz, 2H), 7.67 (d, J )
8.1 Hz, 2H). 13C NMR (75 MHz, CDCl3) δ (ppm): 19.0, 21.5,
41.7, 48.2, 50.8, 52.9, 55.5, 108.7, 115.9, 118.5, 120.6, 124.5,
127.1, 134.7, 138.4, 140.5, 149.7. IR (neat, cm-1): 3302, 2940,
2805, 1592, 1496, 1449, 1325, 1244, 1150, 1091, 1022, 827, 746.
MS (EI+): 403, 388, 205, 177, 136, 91 (100%), 56. Mp
(uncorrected): 171-173 °C. Anal. (C21H29N3O3S‚C2H2O4) C, H,
N.
3-(6-Meth oxy-3,4-d ih yd r o-1H-isoqu in olin -2-yl)p r op yl-
a m in e (11a ). 8 (3.70 g, 17.1 mmol) was dissolved in THF (dry,
10 mL) and added dropwise to a freshly prepared suspension
of AlCl3 (4.10 g, 30.8 mmol) and LiAlH4 (1.20 g, 31.6 mmol) in
THF (dry, 250 mL) under nitrogen atmosphere. The mixture
was allowed to stir at room temperature overnight. Workup
was initiated by careful subsequent addition of water (1 equiv),
N-Eth yl-N-{3-[4-(2-m eth oxyp h en yl)p ip er a zin -1-yl]p r o-
p yl}-4-m eth ylben zen esu lfon a m id e (13b). Acetamide 10b

5462 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Vermeulen et al.
NaOH (1 N, 1 equiv), and water (4 equiv) and filtration of the
salts thus formed over Celite. The clear solution was dried
(Na2SO4) and concentrated in vacuo. This yielded 11a as a
yellow-green oil (3.24 g, 86% yield), which was used for further
yield) that was purified by chromatography (Silica, CH2Cl2/
CH3OH, 96:4) to yield pure 14c as oil. The free base was
converted to its oxalate salt by treatment with oxalic acid
1
dihydrate in ethanol. H NMR (free base, 200 MHz, CDCl3) δ
1
reactions without purification. H NMR (200 MHz, CDCl3) δ
(ppm): 1.69-1.75 (m, 4H), 2.42 (s, 1H), 2.55 (t, J ) 10.3 Hz,
2H), 2.69 (t, J ) 5.7 Hz, 2H), 2.90 (t, J ) 5.7 Hz, 2H), 3.09 (t,
J ) 5.9 Hz, 2H), 3.49 (s, 3H), 3.80 (s, 3H), 6.65-6.76 (m, 4H),
6.92 (d, J ) 8.6 Hz, 1H), 7.26 (d, J ) 8.3 Hz, 1H) 7.65 (d, J )
8.3 Hz, 1H). 13C NMR (50 MHz, CDCl3) δ (ppm): 20.0, 23.2,
27.6, 42.4, 48.7, 53.8, 54.4, 55.9, 110.9, 111.7, 124.6, 125.5,
125.8, 133.6, 135.7, 141.5, 158.6. IR (neat, cm-1): 2932, 1612,
1505, 1455, 1328, 1159, 815. MS (m/z): 375 (M + 1). Mp
(uncorrected): 158-160 °C. Anal. (C20H26N2O3S‚C2H2O4) C, H,
N.
(ppm): 2.63 (t, J ) 6.3 Hz, 2H), 2.75-3.00 (m, 10 H), 3.66 (s,
2H), 3.81 (s, 3H), 6.81 (m, 3H). 13C NMR (50 MHz, CDCl3) δ
(ppm): 15.1, 27.7, 49.0, 51.4, 55.1, 55.3, 121.0, 121.9, 124.7,
126.7, 133.3, 156.6. IR (neat, cm-1): 3359, 2930, 1612, 1505,
1271, 1138, 1038, 812. MS (m/z): 221 (M + 1).
N-[3-(6-Meth oxy-3,4-d ih yd r o-1H-isoqu in olin -2-yl)p r o-
p yl]a ceta m id e (11b). Method A was applied, using crude 11a
(2.30 g, 10.5 mmol) and acetic anhydride (1.44 g, 15.0 mmol).
This yielded crude 11b (2.49 g, 91% yield) as a slightly brown
oil that was used without further purification. IR (neat, cm-1):
3290, 2930, 1824, 1650, 1505, 1447, 1373, 1271, 1138, 1036,
904, 813. MS (m/z): 263 (M + 1).
N-Eth yl-N-[3-(6-m eth oxy-3,4-d ih yd r o-1H-isoqu in olin -
2-yl)p r op yl]-4-m eth ylben zen esu lfon a m id e (14d ). Method
A was applied, using 11c (0.25 g, 1.35 mmol) and p-toluene-
sulfonyl chloride (0.11 g, 0.45 mmol). This yielded an oil (0.30
g, 55% yield) that was purified by chromatography (Silica, CH2-
Cl2/CH3OH, 95:5) to yield pure 14d as slightly yellow oil. The
free base was converted to its oxalate salt by treatment with
oxalic acid dihydrate in ethanol. 1H NMR (free base, 200 MHz,
CDCl3) δ (ppm): 1.13 (t, J ) 7.2 Hz, 2H), 1.82-1.90 (m, 5H),
2.42 (s, 1H), 2.53 (t, J ) 7.2 Hz, 2H), 2.69 (t, J ) 5.7 Hz, 2H),
2.88 (t, J ) 5.7 Hz, 2H), 3.18-3.30 (q, J ) 5.8 Hz, 3H), 3.54
(s, 3H), 3.78 (s, 3H), 6.64-6.73 (m, 4H), 7.29 (d, J ) 5.6 Hz,
1H), 7.68 (d, J ) 1.7 Hz, 1H) 7.72 (d, J ) 1.7 Hz, 1H). 13C
NMR (50 MHz, CDCl3) δ (ppm): 13.6, 20.0, 25.2, 27.9, 41.5,
44.4, 49.3, 53.8, 54.2, 110.5, 111.6, 125.6, 128.1, 134.0, 135.5,
141.5, 156.5. IR (neat, cm-1): 2931, 1611, 1505, 1465, 1336,
1156, 1037, 815, 729. MS (m/z): 402 (M + 1). Mp (uncor-
rected): 186-187 °C. Anal. (C22H30N2O3S‚C2H2O4) C, H, N.
Eth yl-[3-(6-m eth oxy-3,4-d ih yd r o-1H-isoqu in olin -2-yl)-
p r op yl]a m in e (11c). 11b (crude, 2.49 g, ca. 9.5 mmol) was
dissolved in THF (dry, 15 mL) and added dropwise to a freshly
prepared suspension of LiAlH4 (0.60 g, 15.8 mmol) in THF (dry,
75 mL) under nitrogen atmosphere. The mixture was refluxed
overnight. Careful workup by subsequent addition of water
(1 equiv), NaOH (1 N, 1 equiv) and water (4 equiv) and
filtration of the salts thus formed over Celite yielded a clear
solution that was dried (Na2SO4) and concentrated in vacuo.
The oil thus obtained (2.54 g, 99% yield) was purified by
1
chromatography (Silica, CH2Cl2/CH3OH, 99:1). H NMR (200
MHz, CDCl3) δ (ppm): 0.96 (t, J ) 6.4 Hz, 3H), 1.48-1.52 (m,
2H), 2.34-2.72 (m, 10 H), 3.24 (br, 1H), 3.60 (s, 2H), 3.76 (s,
3H), 6.62-6.74 (m, 2H), 6.92-6.98 (m, 1H). 13C NMR (50 MHz,
CDCl3) δ (ppm): 15.8, 28.8, 30.0, 41.2, 45.4, 51.2, 54.8, 56.0,
110.8, 112.2, 130.5, 131.2, 138.9, 156.8. IR (neat, cm-1): 2929,
1612, 1505, 1464, 1269, 1159, 1038, 803. MS (m/z): 249 (M +
1).
4-Meth oxy-N-[3-(6-m eth oxy-3,4-d ih yd r o-1H-isoqu in o-
lin -2-yl)p r op yl]-ben zen esu lfon a m id e (14e). Method A was
applied, using 11a (0.10 g, 0.45 mmol) and p-methoxybenze-
nesulfonyl chloride (0.28 g, 1.36 mmol). This yielded an oil
(0.17 g, 97% yield) that was purified by chromatography (Silica,
EtOAc/hexane, 2:3) to yield pure 14e as oil. The free base was
converted to its oxalate salt by treatment with oxalic acid
N-[3-(6-Meth oxy-3,4-d ih yd r o-1H-isoqu in olin -2-yl)p r o-
p yl]-ben zen esu lfon a m id e (14a ). Method A was applied,
using benzenesulfonyl chloride (0.18 g, 1.02 mmol) and 11a
(0.20 g, 0.93 mmol). This yielded an oil that was purified by
chromatography (Silica, CH2Cl2/CH3OH, 19:1) to yield pure
14a (0.27 g, 81% yield). The free base was converted to its
hydrochloric salt by treatment with HCl (1 N solution in
1
dihydrate in ethanol. H NMR (free base, 200 MHz, CDCl3) δ
(ppm): 1.28 (s, 1H), 1.93-2.12 (m, 5H), 2.52 (t, J ) 5.6 Hz,
2H), 2.58 (t, J ) 5.6 Hz, 2H), 2.89 (t, J ) 5.6 Hz, 2H), 3.53 (s,
1H), 3.79 (s, 3H), 3.89 (s, 3H), 6.67 (m, 5H), 6.98 (d, J ) 2.4
Hz, 1H), 7.97 (d, J ) 2.2 Hz, 1H). 13C NMR (50 MHz, CDCl3)
δ (ppm): 26.1, 28.0, 46.2, 49.3, 53.5, 54.2, 111.3, 111.6, 112.8,
125.5, 126.0, 127.9, 130.3, 134.2, 156.1, 162.2. IR (neat, cm-1):
2926, 1593, 1498, 1366, 1262, 1158, 804. MS (m/z): 391 (M
+ 1). Mp (uncorrected): 192-194 °C. Anal. (C20H26N2O4S‚
C2H2O4) C, H, N.
1
diethyl ether). H NMR (free base, 200 MHz, CDCl3) δ (ppm):
0.80-0.99 (m, 1H), 1.30 (s, 1H), 1.72 (s, 1H), 1.93-2.10 (m,
1H), 2.52 (t, J ) 6.8 Hz, 2H), 2.65 (t, J ) 5.8 Hz, 2H), 2.90 (t,
J ) 6.0 Hz, 2H), 2.99 (t, J ) 6.0 Hz, 2H), 3.72 (s, 3H), 6.60-
6.77 (m, 2H), 6.97 (d, J ) 8.5 Hz, 1H), 7.50-7.70 (m, 4H), 8.05
(d, J ) 1.0 Hz, 1H). 13C NMR (50 MHz, CDCl3) δ (ppm): 27.9,
28.2, 42.1, 46.3, 48.9, 53.8, 54.0, 125.5, 126.0, 127.6, 132.1,
132.3, 138.4. IR (neat, cm-1): 3411, 2937, 2485, 1614, 1509,
1448, 1371, 1167, 1084, 721, 551. MS (m/z): 361 (M + 1). Mp
(uncorrected): 126-128 °C. Anal. (C19H24N2O3S‚Cl) C, H, N.
N-E t h yl-4-m et h oxy-N-[3-(6-m et h oxy-3,4-d ih yd r o-1H -
isoqu in olin -2-yl)pr opyl]-ben zen esu lfon am ide (14f). Method
A was applied, using 11c (0.05 g, 0.18 mmol) and p-methoxy-
benzenesulfonyl chloride (0.04 g, 0.21 mmol). This yielded an
oil that was purified by chromatography (Silica, CH2Cl2/CH3-
OH, 99:1) to yield pure 14f as colorless oil (0.06 g, 80% yield).
The free base was converted to its oxalate salt by treatment
with oxalic acid dihydrate in ethanol. 1H NMR (free base, 200
MHz, CDCl3) δ (ppm): 1.02 (t, J ) 7.2 Hz, 2H), 1.19 (s, 1H),
1.28-1.42 (m, 5H), 3.02-3.22 (m, 6H), 3.49-3.59 (q, J ) 7.2
Hz, 3H), 3.73 (s, 3H), 3.81 (s, 3H), 6.63-6.81 (m, 5H), 6.93 (d,
J ) 8.8 Hz, 1H), 7.68 (d, J ) 8.8 Hz, 1H). 13C NMR (50 MHz,
CDCl3) δ (ppm): 7.2, 12.2, 22.2, 28.1, 42.0, 44.6, 48.0, 51.5,
53.9, 61.8, 112.5, 112.9, 117.0, 126.0, 126.5, 127.7, 130.4, 158.0,
161.4. IR (neat, cm-1): 2930, 1611, 1498, 1335, 1258, 1152,
1029, 804, 734. MS (m/z): 419 (M + 1). Mp (uncorrected):
188-190 °C. Anal. (C22H30N2O4S‚C2H2O4) C, H, N.
Na p h th a len e-1-su lfon ic Acid [3-(6-Meth oxy-3,4-d ih y-
d r o-1H-isoqu in olin -2-yl)p r op yl]a m id e (14g). Method A
was applied, using 11a (0.09 g, 0.42 mmol) and naphthalene-
1-sulfonyl chloride (0.11 g, 0.46 mmol). This yielded a slightly
yellow oil (0.17 g) that was purified by chromatography (Silica,
CH2Cl2/CH3OH, 99:1) to yield pure 14g (0.14 g, 81% yield).
The free base was converted to its oxalate salt by treatment
with oxalic acid dihydrate in ethanol. 1H NMR (free base, 200
N-Eth yl-N-[3-(6-m eth oxy-3,4-d ih yd r o-1H-isoqu in olin -
2-yl)p r op yl]-ben zen esu lfon a m id e (14b). Method A was
applied, using benzenesulfonyl chloride (0.07 g, 0.36 mmol)
and 11c (0.05 g, 0.18 mmol). This yielded an oil that was
purified by chromatography (Silica, CH2Cl2/CH3OH, 19:1) to
yield pure 14b (0.05 g, 39% yield). The free base was converted
to its oxalate salt by treatment with oxalic acid dihydrate in
1
ethanol. H NMR (free base, 200 MHz, CDCl3) δ (ppm): 1.12
(t, J ) 7.2 Hz, 2H), 1.31 (s, 1H), 1.85-1.92 (m, 4H), 2.58 (t, J
) 7.2 Hz, 2H), 2.74 (t, J ) 5.5 Hz, 2H), 2.89 (t, J ) 5.6 Hz,
2H), 3.21-3.28 (q, J ) 7.3 Hz, 3H), 3.56 (s, 1H), 3.78 (s, 3H),
6.64-6.74 (m, 2H), 6.94 (d, J ) 8.6 Hz, 1H), 7.49-7.59 (m,
4H), 7.84 (d, J ) 7.8 Hz, 1H). 13C NMR (50 MHz, CDCl3) δ
(ppm): 12.5, 25.0, 27.6, 28.2, 44.3, 49.2, 53.6, 53.8, 110.7, 111.7,
124.8, 125.5, 126.6, 127.5, 130.8, 133.6, 138.5, 156.6. IR (neat,
cm-1): 2927, 1690, 1611,, 1505, 1335, 1159, 1036, 802, 692.
MS (m/z): 389 (M + 1). Mp (uncorrected): 159-162 °C. Anal.
(C21H28N2O3S‚C2H2O4) C, H, N.
N-[3-(6-Meth oxy-3,4-d ih yd r o-1H-isoqu in olin -2-yl)p r o-
p yl]-4-m eth ylben zen esu lfon a m id e (14c). Method A was
applied, using 11a (0.13 g, 0.59 mmol) and p-toluenesulfonyl
chloride (0.12 g, 0.65 mmol). This yielded an oil (0.17 g, 77%

Novel 5-HT7 Receptor Inverse Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5463
MHz, CDCl3) δ (ppm): 0.80-1.00 (m, 2H), 1.28 (s, 1H), 1.71-
1.79 (m, 3H), 2.47 (t, J ) 5.7 Hz, 2H), 2.70 (t, J ) 5.9 Hz, 2H),
2.97 (t, J ) 5.3 Hz, 2H), 3.40 (s, 1H), 3.84 (s, 3H), 6.72-6.93
(m, 2H), 7.17-7.28 (m, 2H), 7.51 (t, J ) 7.8 Hz, 2H), 7.92 (d,
J ) 8.3 Hz, 1H), 8.04 (d, J ) 1.0 Hz, 1H), 8.24 (d, J ) 1.0 Hz,
1H), 8.37 (d, J ) 8.8 Hz, 1H). 13C NMR (50 MHz, CDCl3) δ
(ppm): 22.9, 27.4, 42.6, 48.1, 54.2, 56.1, 110.8, 111.8, 122.4,
124.8, 125.1, 125.8, 127.2, 128.1, 132.0, 132.5, 134.0, 156.6.
IR (neat, cm-1): 2925, 1612, 1505, 1320, 1159, 1135, 773. MS
(m/z): 411 (M + 1). Mp (uncorrected): 170-171 °C. Anal.
(C23H26N2O3S‚C2H2O4) C, H, N.
mmol) was dissolved in DMF under nitrogen atmosphere. After
addition of KO-t-Bu (0.10 g, 0.89 mmol) and dimethyl oxalate
(0.10 g, 0.85 mmol), the mixture was heated to reflux for 4 h.
The mixture was then allowed to cool to room temperature,
and ammonia (30 mL, 10% in water) was added. Extraction
with diethyl ether (3 × 15 mL), drying (Na2SO4) of the
combined organic layers, and subsequent evaporation of the
volatiles yielded a brown oil (0.17 g, 86% yield), which was
purified by chromatography (Silica, gradient CH2Cl2/CH3OH:
100/0-85/15). The free base was converted to its oxalate salt
1
by treatment with oxalic acid dihydrate in ethanol. H NMR
(300 MHz, CDCl3) δ (ppm): 2.30 (s, 6H), 2.57 (dd, J a ) 7.3
Hz, J b ) 8.4 Hz, 2H), 2.82-2.89 (m, 2H), 3.63 (s, 3H), 3.82 (s,
3H), 6.80 (s, 1H), 6.84 (dd, J a ) 2.2 Hz, J b ) 6.6 Hz, 1H), 7.00
(d, J ) 1.8 Hz, 1H), 7.12 (d, J ) 8.8 Hz, 1H). 13C NMR (75
MHz, CDCl3) δ (ppm): 21.2, 30.2, 43.0, 53.5, 58.0, 98.4, 107.5,
109.1, 109.7, 124.5, 125.6, 129.9, 151.1. MS (EI+): 232, 174,
131, 58 (100%). Anal. (C14H20N2O‚C2H2O4) C, H, N.
Na p h th a len e-1-su lfon ic Acid Eth yl-[3-(6-m eth oxy-3,4-
d ih yd r o-1H-isoqu in olin -2-yl)p r op yl]a m id e (14h ). Method
A was applied, using 11c (0.09 g, 0.36 mmol) and naphthalene-
1-sulfonyl chloride (0.09 g, 0.39 mmol). This yielded an oil (0.11
g, 70% yield) that was purified by chromatography (Silica, CH2-
Cl2/CH3OH, 99:1) to yield pure 14h as oil. The free base was
converted to its oxalate salt by treatment with oxalic acid
1
dihydrate in ethanol. H NMR (free base, 200 MHz, CDCl3) δ
3-(1-Ben zyl-1,2,3,6-tetr a h yd r op yr id in -4-yl)-5-m eth oxy-
1H-in d ole (20). In a three-necked bottle with cooler, 5-meth-
oxyindole (15, 0.50 g, 3.40 mmol) was dissolved in CH3OH (5
mL) under nitrogen atmosphere. N-Benzyl-4-piperidone (0.96
g, 5.08 mmol) and NaOCH3 (5 mL, 30% solution in CH3OH)
were added, and the mixture was heated to reflux for 4 h. The
reaction was allowed to cool to room temperature. The
precipitate that was formed, filtered off (0.90 g, 83% yield),
and dried in vacuo. Recrystallization from EtOAc yielded pure
(ppm): 1.11 (t, J ) 7.2 Hz, 2H), 1.26 (s, 1H), 1.74-1.81 (m,
5H), 2.33 (t, J ) 7.1 Hz, 2H), 2.54 (t, J ) 19.2 Hz, 2H), 2.79 (t,
J ) 4.9 Hz, 2H), 3.33-3.49 (m, 3H), 3.78 (d, J ) 7.6 Hz, 1H),
6.62-6.72 (m, 2H), 6.89 (d, J ) 8.6 Hz, 1H), 7.47-7.65 (m,
3H), 7.93 (d, J ) 7.6 Hz, 1H), 8.03 (d, J ) 8.5 Hz, 1H), 8.23 (d,
J ) 1.2 Hz, 1H), 8.65 (d, J ) 9.3 Hz, 1H). 13C NMR (50 MHz,
CDCl3) δ (ppm): 12.1, 24.3, 27.8, 40.0, 43.0, 49.1, 53.6, 53.7,
53.9, 110.5, 111.6, 122.6, 123.6, 125.3, 126.0, 126.5, 127.4,
128.2, 132.5, 133.8. IR (neat, cm-1): 2925, 1612, 1505, 1320,
1259, 1135, 773. MS (m/z): 439 (M + 1). Mp (uncorrected):
166-169 °C. Anal. (C25H30N2O3S‚C2H2O4) C, H, N.
1
20 as a white solid. H NMR (300 MHz, CDCl3) δ (ppm): 2.53
(d, J ) 1.5 Hz, 2H), 2.68-2.72 (m, 2H), 3.20 (dd, J a ) 2.6 Hz,
J b ) 2.9 Hz, 2H), 3.62 (s, 2H), 3.80 (s, 3H), 6.07-6.09 (m, 1H),
6.79-6.83 (m, 1H), 7.07 (d, J ) 2.6 Hz, 1H), 7.17-7.38 (m,
7H), 8.00 (br, 1H). 13C NMR (75 MHz, CDCl3) δ (ppm): 26.7,
47.6, 50.8, 53.5, 60.5, 100.5, 109.3, 109.7, 116.4, 119.5, 123.1,
124.6, 125.8, 126.8, 127.4, 151.9. IR (KBr, cm-1): 3032, 2925,
2854, 2794, 1650, 1614, 1601, 1479, 1447, 1380, 1260, 1211,
1125, 1034, 962, 880, 807, 736. MS (EI+): 318, 227, 198, 160,
128, 91 (100%). Anal. (C21H22N2O) C, H, N.
(5-Meth oxy-1H-in d ol-3-yl)oxoa cetyl Ch lor id e (16). Ox-
alyl chloride (1.00 g, 8.16 mmol) was dissolved in dry diethyl
ether (15 mL) and cooled to 0 °C. 5-Methoxyindole (15, 1.00 g,
6.80 mmol) was added portionwise. The bright orange pre-
cipitate thus formed was filtered after 30 min, dried in vacuo
(1.45 g, 75% yield), and used without further purification. IR
(KBr, cm-1): 3193, 1779, 1617, 1477, 1427, 1267, 1211, 991,
773, 705, 653.
6-Met h oxy-2,3-d ih yd r o-1H -in d ole (22). 6-Methoxy-1H-
indole (21, 1.50 g, 10.2 mmol) was dissolved in glacial acetic
acid (30 mL). NaCNBH3 (1.9 g, 30.6 mmol) was added
portionwise, while the temperature did not exceed 20 °C. After
stirring at room temperature for 3 h, the solvent was removed
under reduced pressure, and the residue was treated with
NaHCO3 (150 mL, sat.). The water layer was extracted with
ether (3 × 75 mL), and the combined organic phase was
washed with brine (30 mL), dried (Na2SO4), and filtered.
Evaporation of solvent yielded a colorless oil (1.48 g, 97%
yield), that could be purified by chromatography (Silica, CH2-
Cl2) to obtain pure 22. 1H NMR (200 MHz, CDCl3) δ (ppm):
2.95 (t, J ) 16.6 Hz, 2H), 3.55 (t, J ) 16.6 Hz, 2H), 3.75 (s,
3H), 3.88 (br, 1H), 6.26 (s, 1H), 6.30 (d, J ) 2.4 Hz, 1H), 7.00
(d, J ) 10.7 Hz, 1H). 13C NMR (50 MHz, CDCl3) δ (ppm): 27.5,
53.9, 95.1, 102.2, 120.2, 123.3, 151.1, 158.3. IR (neat, cm-1):
3375, 2947, 2847, 1614, 1500, 1286, 1196, 1150, 1030, 830. MS
(EI+): 149 (100%), 133, 117, 104, 91, 77, 63, 51.
2-Ch lor o-1-(6-m e t h oxy-2,3-d ih yd r oin d ol-1-yl)e t h a -
n on e (23). A solution of 6-Methoxy-2,3-dihydro-1H-indole (22,
2.20 g, 14.8 mmol) in CH2Cl2 (35 mL) was added dropwise to
a mixture of chloroacetyl chloride (3.35 g, 29.5 mmol) and K2-
CO3 (4.50 g, 32.6 mmol) in CH2Cl2 (75 mL). After addition,
the suspension was stirred at room temperature for 6 h. Water
(100 mL) was added and extracted with CH2Cl2 (3 × 50 mL).
The combined organic layers were washed with brine (20 mL),
dried (Na2SO4), and filtered. Evaporation of the solvent yielded
an off-white solid (2.95 g, 88% yield). Pure 23 was obtained
by recrystallization from diisopropyl ether. 1H NMR (200 MHz,
CDCl3) δ (ppm): 3.17 (t, J ) 16.4 Hz, 2H), 3.80 (s, 3H), 4.16
(s, 2H), 4.17 (t, J ) 16.6 Hz, 2H), 6.62 (d, J ) 10.5 Hz, 1H),
7.07 (d, J ) 8.1 Hz, 1H), 7.90 (s, 1H). 13C NMR (50 MHz,
CDCl3) δ (ppm): 25.9, 41.7, 47.2, 54.1, 101.6, 109.5, 123.2. IR
(KBr, cm-1): 2964, 1680, 1497, 1404, 1231, 1028, 842. Mp
(uncorrected): 126 °C
2-(5-Meth oxy-1H-in d ol-3-yl)-N,N-d im eth yl-2-oxoa ceta -
m id e (17). (5-Methoxy-1H-indol-3-yl)oxoacetyl chloride (16,
0.25 g, 1.05 mmol) was added portionwise to a solution of
dimethylamine (5 mL, 40% in water) at 0 °C. After 30 min,
CH2Cl2 (50 mL) was added. The layers were separated, and
the inorganic phase was washed with CH2Cl2. The combined
organic layers were dried (Na2SO4), filtered, and concentrated
in vacuo. The residue was crystallized from EtOAc/hexanes
1
to yield a white solid (0.23 g, 89% yield). H NMR (200 MHz,
CD3OD/CDCl3) δ (ppm): 2.80 (d, J ) 8.8 Hz, 6H), 3.64 (d, 8.8
Hz, 3H), 6.66 (s, 1H), 7.08 (d, J ) 8.1 Hz), 7.50 (s, 1H), 7.58 (s,
1H). IR (KBr, cm-1): 3090, 3038, 2894, 1605, 1518, 1490, 1437,
1265, 1210, 1149, 1030, 808.
[2-(5-Meth oxy-1H-in dol-3-yl)eth yl]-dim eth ylam in e (18).
In a three-necked bottle with cooler, LiAlH4 (0.11 g, 2.89 mmol)
was suspended in THF (10 mL) under nitrogen atmosphere.
2-(5-methoxy-1H-indol-3-yl)-N,N-dimethyl-2-oxoacetamide (17,
0.20 g, 0.81 mmol), dissolved in THF (10 mL) was added
dropwise at room temperature. The suspension was heated to
a gentle reflux for 2 h. After cooling to room temperature, the
excess of LiAlH4 was destroyed by careful addition of adequate
amounts of water and NaOH (4 N). The salts were filtered
over Celite and washed well (diethyl ether), and the residue
was concentrated in vacuo to yield a colorless oil (0.14 g, 79%
yield), which was pure 18 according to NMR and GC. 1H NMR
(200 MHz, CDCl3) δ (ppm): 2.31 (s, 6H), 2.61 (dd, J a ) 7.3
Hz, J b ) 8.8 Hz, 2H), 2.87 (dd, J a ) 7.3 Hz, J b ) 8.8 Hz, 2H),
3.81 (s, 3H), 6.80 (dd, J a ) 2.6 Hz, J b ) 6.2 Hz, 1H), 6.91, (s,
1H), 6.99 (d, J ) 2.2 Hz, 1H), 7.15 (d, J ) 8.8 Hz, 1H). 13C
NMR (75 MHz, CDCl3) δ (ppm): 21.2, 42.9, 53.5, 57.7, 98.2,
109.4, 109.5, 111.3, 120.0, 125.3, 129.0, 151.3. IR (free base,
neat, cm-1): 3410, 3145, 2930, 2800, 1623, 1580, 1485, 1215,
1171, 1066, 795. MS (EI+): 218, 160, 117, 89, 58 (100%).
[2-(5-Meth oxy-1-m eth yl-1H-in d ol-3-yl)eth yl]d im eth yl-
a m in e (19). In
a
three-necked bottle with cooler, [2-(5-
2-Dim eth yla m in o-1-(6-m eth oxy-2,3-d ih yd r oin d ol-1-yl)-
methoxy-1H-indol-3-yl)ethyl]dimethylamine (18, 0.18 g, 0.83
eth a n on e (24). Dimethylamine hydrochloride (0.18 g, 2.20

5464 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Vermeulen et al.
mmol) was added to a suspension of 2-chloro-1-(6-methoxy-
2,3-dihydroindol-1-yl)ethanone (23, 0.50 g, 2.20 mmol) and K2-
CO3 (0.61 g, 4.42 mmol) in acetonitrile (25 mL, cat. amount
DMF) under a nitrogen atmosphere. The mixture was stirred
overnight at 50 °C. After filtration, the volatiles were removed
in vacuo to yield the crude 24 as a semisolid (0.51 g, 99% yield),
which was used for reduction without further purification. 1H
NMR (200 MHz, CDCl3) δ (ppm): 2.35 (s, 3H), 3.05 (t, J )
16.6 Hz, 2H), 3.16 (s, 2H), 3.76 (s, 3H), 4.11 (t, J ) 16.8 Hz,
2H), 6.53 (d, J ) 10.5 Hz, 1H), 7.00 (d, J ) 8.1 Hz, 1H), 7.89
(s, 1H). 13C NMR (50 MHz, CDCl3) δ (ppm): 25.9, 44.1, 46.7,
54.0, 62.0, 101.4, 108.5, 123.1. IR (KBr, cm-1): 2933, 2827,
2772, 1654, 1492, 1283, 1157, 1029, 859. MS (EI+): 234, 58
(100%).
[2-(6-Met h oxy-2,3-d ih yd r oin d ol-1-yl)et h yl]d im et h yl-
a m in e (25). LiAlH4 (0.26 g, 6.84 mmol) was suspended in THF
(25 mL) under nitrogen atmosphere and cooled to 0 °C. AlCl3
(1.07 g, 8.04 mmol) was carefully added and stirred for 15 min.
Then, a solution of 2-(dimethylamino)-1-(6-methoxy-2,3-dihy-
droindol-1-yl)ethanone (24, 0.51 g, 2.20 mmol) in THF (10 mL)
was added dropwise. The reaction was stirred at room tem-
perature for 4 h. Careful workup of the reaction by addition
of water and NaOH (1 N) resulted in formation of salts, which
were filtered off over Celite. The mother liquor was concen-
trated in vacuo, and the remainder was partioned between
NaHCO3 (30 mL, sat.) and CH2Cl2 (50 mL). The organic phase
was dried (Na2SO4), filtered, and concentrated in vacuo.
Purification by chromatography (Silica, gradient CH2Cl2/CH3-
OH: 100/0-85/15) yielded pure 25 as a colorless oil (0.21 g,
43% yield). The free base was converted to its oxalate salt by
treatment with oxalic acid dihydrate in ethanol. 1H NMR (free
base, 200 MHz, CDCl3) δ (ppm): 2.26 (s, 6H), 2.49 (t, J ) 14.3
Hz, 2H), 2.84 (t, J ) 16.5 Hz, 2H), 3.13 (t, J ) 14.3 Hz, 2H),
3.35 (t, J ) 16.5 Hz, 2H), 3.71 (s, 3H), 6.03 (s, 1H), 6.11 (d, J
) 10.3 Hz, 1H), 6.88 (d, J ) 8.1 Hz, 1H). 13C NMR (50 MHz,
CDCl3) δ (ppm): 25.3, 43.3, 45.1, 51.6, 52.9, 54.3, 92.0, 98.7,
119.8, 121.8, 151.3, 157.6. IR (neat, cm-1): 2942, 2819, 2767,
1616, 1495, 1281, 1205, 1034, 818. MS (EI+): 220, 162, 58
(100%). Anal. (C13H20N2O‚C2H2O4) C, H, N.
2-Am in o-4-m eth oxyp h en ol (27). 4-Methoxy-2-nitrophenol
(26, 1.00 g, 5.9 mmol) was dissolved in ethanol (100 mL, absol)
with palladium (0.15 g, 10% on charcoal) and allowed to react
for 20 min at room temperature with hydrogen gas (35 psi) in
a Parr-shaker. The reaction mixture was filtered over Celite,
and the solvent was removed under reduced pressure. This
yielded pure 27 as an off-white solid (0.80 g, 98% yield). 1H
NMR (300 MHz, CDCl3) δ (ppm): 2.57 (br, 3H), 3.63 (s, 3H),
6.10 (dd, J a ) 2.6 Hz, J b ) 2.9 Hz, 1H), 6.25 (d, J ) 2.6 Hz,
1H), 6.55 (d, J ) 8.4 Hz, 1H). 13C NMR (75 MHz, CDCl3) δ
(ppm): 53.1, 100.3, 100.8, 113.1, 133.2, 135.9, 151.4. IR (KBr,
cm-1): 3383, 3292, 2960, 2832, 2591, 1600, 1514, 1465, 1310,
1234, 1201, 1032, 839, 802. MS (EI+): 139, 96 (100), 68.
6-Meth oxy-4H-ben zo[1,4]oxa zin -3-on e (28). 27 (0.25 g,
1.8 mmol) was dissolved in acetonitrile (15 mL) in a three-
necked bottle with reflux condenser under nitrogen. 2-Chlo-
roacetyl chloride (0.22 g, 2.0 mmol) was added dropwise,
followed by K2CO3 (0.65, 4.7 mmol). The suspension was
refluxed for 2 h, then cooled to room temperature and filtered.
The solution was concentrated in vacuo, and the residue was
partioned between CH2Cl2 (3 × 15 mL) and water (30 mL).
The combined organic layers were dried (Na2SO4), filtered, and
concentrated to yield 28 as a slightly pink solid (0.31 g, 87%
yield) that was recrystallized from methanol. 1H NMR (300
MHz, CDCl3) δ (ppm): 3.71 (s, 3H), 4.51 (s, 2H), 6.34 (d, J )
2.9 Hz, 1H), 6.46 (dd, J a ) 2.6 Hz, J b ) 6.2 Hz, 1H) 6.84 (d, J
) 8.8 Hz, 1H), 8.59 (br, 1H). 13C NMR (50 MHz, CDCl3) δ
(ppm): 53.3, 64.9, 99.6, 106.3, 114.7. IR (KBr, cm-1): 3197,
3106, 2898, 1712, 1610, 1518, 1464, 1398, 1215, 1048, 830. MS
(EI+): 179 (100), 109
LiAlH4 was decomposed by careful subsequent addition of
water (5 drops), NaOH (2 N, 5 drops), and water (ca 0.2 mL).
The salts thus formed were filtered over MgSO4, and the
solvent was evaporated to yield 29 (0.15 g, 91% yield) as
slightly green oil that crystallized upon standing and needed
no further purification. H NMR (300 MHz, CDCl3) δ (ppm):
3.32-3.35 (m, 2H), 3.66 (s, 3H), 3.73 (br, 1H), 4.14 (t, J ) 4.4
1
Hz, 2H), 6.12 (d, J ) 2.6 Hz, 1H), 6.17 (dd, J a ) 2.6 Hz, J b
)
2.9 Hz, 1H), 6.65 (d, J ) 8.8 Hz, 1H). 13C NMR (75 MHz,
CDCl3) δ (ppm): 38.6, 53.1, 62.5, 98.8, 101.0, 114.4, 131.6,
135.7, 151.9. IR (KBr, cm-1): 3389, 2947, 1620, 1514, 1204,
1167, 1047, 835, 782. MS (EI+): 165 (100), 150, 79, 68, 55.
[2-(6-Meth oxy-2,3-d ih yd r oben zo[1,4]oxa zin -4-yl)eth yl]-
d im eth yla m in e Hyd r och lor id e (32). 6-Methoxy-3,4-dihy-
dro-2H-benzo[1,4]oxazine (29, 0.17 g, 1.03 mmol), 2-chloro-
acetyl chloride (0.12 g, 1.07 mmol), and pyridine (0.15 mL)
were stirred at room temperature overnight under nitrogen.
The volatiles were evaporated, and the residue was partioned
between water and CH2Cl2. The organic layer was dried (Na2-
SO4), filtered, and concentrated in vacuo to yield 30 as a brown
oil (0.25 g, 100% yield). (MS (EI+): 241 (100%), 165, 150, 137,
109, 77). The brown oil obtained was dissolved in acetonitrile
(10 mL) under nitrogen. K2CO3 (0.28 g, 2.03 mmol), KI (1 small
crystal), and dimethylamine hydrogen chloride (0.15 g, 1.83
mmol) were added. The mixture was stirred at room temper-
ature for 20 h. The volatiles were removed under reduced
pressure, and the residue was partioned between acid (HCl, 1
M) and ether. The water layer was made alkaline (NaOH, 2
N, pH ) 8) and subsequently extracted with CH2Cl2. The
organic layer was dried (Na2SO4), filtered, and concentrated
in vacuo to yield crude 31 as a brown oil (0.13 g, 52% yield,
MS (EI+): 250, 58 (100%), IR (KBr): 2943, 2827, 2776, 1660,
1502, 1397, 1254, 1212, 1175, 1058, 860). This oil, dissolved
in THF (5 mL), was added dropwise to a suspension of AlCl3
(0.13 g, 0.98 mmol) and LiAlH4 (0.04 g, 1.05 mmol) in diethyl
ether. The reaction was allowed to stir at room temperature
for 4 h and was then carefully quenched with water and
neutralized with NaHCO3 (saturated aqueous solution). Ex-
traction with diethyl ether and subsequent drying (Na2SO4),
filtration, and evaporation of the organic layer yielded a brown
oil (0.11 g, 47% yield) that was purified by chromatography to
yield pure 32 as a colorless oil (Silica, CH2Cl2/CH3OH 98:2).
The compound was converted in its hydrochloride by treatment
of the free base in 2-propanol with HCl (1 N solution in diethyl
ether, 1 equiv). 1H NMR (free base, 300 MHz, CDCl3) δ
(ppm): 2.23 (s, 6H), 2.42-2.47 (m, 2H), 3.28-3.33 (m, 4H),
3.68 (s, 3H), 4.10 (t, J ) 4.4 Hz, 2H), 6.06-6.09 (m, 1H), 6.20
(d, J ) 2.6 Hz, 1H), 6.62 (d, J ) 8.8 Hz, 1H). 13C NMR (75
MHz, CDCl3) δ (ppm): 43.4, 45.0, 46.8, 53.0, 53.1, 61.8, 96.2,
98.1, 113.7, 133.1, 135.7, 152.3. IR (neat, cm-1): 2940, 2821,
2768, 1618, 1511, 1463, 1314, 1207, 1171, 1050, 823. MS
(EI+): 236, 178, 150, 58 (100%). Anal. (C13H20N2O2‚HCl_H2O)
C, H, N, O.
P h a r m a cology. 5-HT7 Recep tor Bin d in g Assa y. Binding
assays on membranes from HEK cells expressing rat 5-HT7
(r5-HT7 receptor obtained from Dr. David Sibley) receptors
were performed according to standard procedures. Briefly, cell
paste was homogenized in 50 mM Tris HCl buffer (pH 7.4)
containing 2.0 mM MgCl2 using a hand-held Polytron (setting
6 for 10 s) and spun in a centrifuge at 40 000 g for 10 min.
The pellet was resuspended in 50 mM Tris HCl buffer (pH
7.4) containing 0.5 mM EDTA, 10 mM MgSO4, 2 mM CaCl2,
10 µM pargyline, and 0.1% ascorbic acid. Incubations were
initiated by the addition of membranes (20 µg protein per well)
to 96-well plates containing test drugs and 0.3 nM [3H]5-CT
(final volume of 250 µL). Nonspecific binding was determined
by radioligand binding in the presence of a saturating con-
centration of 5-HT (10 µM). After a 2-h incubation period at
room temperature, assay samples were rapidly filtered through
Whatman GF/B filters and rinsed with ice-cold 50 mM Tris
buffer (pH 7.4) using a Skatron harvester (Molecular Devices).
Membrane-bound [3H]5-CT levels were determined by liquid
scintillation counting of the filters in BetaScint. The IC50 value
(concentration at which 50% inhibition of specific binding
6-Meth oxy-3,4-d ih yd r o-2H-ben zo[1,4]oxa zin e (29). In a
round-bottom flask LiAlH4 (0.10 g, 2.1 mmol) was suspended
in THF (25 mL, dry) under nitrogen atmosphere. 28 (0.18 g,
1.0 mmol) was added portionwise, and the reaction was
allowed to stir overnight at room temperature. The excess of

Novel 5-HT7 Receptor Inverse Agonists
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5465
(13) Forbes, I. T.; Douglas, S.; Gribble, A. D.; Ife, R. J .; Lightfoot, A.
P.; Garner, A. E.; Riley, G. J .; J effrey, P.; Stevens, A. J .; Stean,
T. O.; Thomas, D. R. SB-656104-A: a novel 5-HT(7) receptor
antagonist with improved in vivo properties. Bioorg. Med. Chem.
Lett. 2002, 12, 3341-3344.
(14) Goodman, J .; Still, W. C. An unbounded systematic search of
conformational space. J . Comput. Chem. 1991, 12, 1110-1117.
(15) Hagan, J . J .; Price, G. W.; J effrey, P.; Deeks, N. J .; Stean, T.;
Piper, D.; Smith, M. I.; Upton, N.; Medhurst, A. D.; Middlemiss,
D. N.; Riley, G. J .; Lovell, P. J .; Bromidge, S. M.; Thomas, D. R.
Characterization of SB-269970-A, a selective 5-HT7 receptor
antagonist. Br. J . Pharmacol. 2000, 130, 539-548.
(16) J avitch, J . A.; Ballesteros, J . A.; Weinstein, H.; Chen, J . A cluster
of aromatic residues in the sixth membrane-spanning segment
of the dopamine D2 receptor is accessible in the binding-site
crevice. Biochemistry 1998, 37, 998-1006.
(17) Kikuchi, C.; Ando, T.; Watanabe, T.; Nagaso, H.; Okuno, M.;
Hiranuma, T.; Koyama, M. 2a-[4-(Tetrahydropyridoindol-2-yl)-
butyl]tetrahydrobenzindole derivatives: new selective antago-
nists of the 5-hydroxytryptamine7 receptor. J . Med. Chem 2002,
45, 2197-2206.
occurs) was calculated by linear regression of the log concen-
tration-response data. Ki values were calculated according to
the Cheng-Prusoff equation, Ki ) IC50/(1 + (L/Kd)), where L
is the concentration of the radioligand used in the experiment
and the Kd value is the dissociation constant for the radioligand
(determined previously by saturation analysis).
5-HT7 Ad en yla te Cycla se Assa y. Effects on adenylate
cyclase activity were measured according to previously pub-
lished methods.37 HEK-293 cells expressing the rat 5-HT7 were
grown in Dulbecco’s modified Eagle’s media (DMEM) contain-
ing 10% fetal bovine serum, 400 µg of G418 (Geneticin), and 2
mM glutamine until flasks were confluent. Cells from confluent
flasks were harvested by replacing the media with phosphate-
buffered saline (PBS) containing 5 mM EDTA (pH 7.4). Cells
were homogenized in 5 mM HEPES buffer containing 1 mM
EGTA (pH 7.4) using a hand-held glass-Teflon homogenizer.
The homogenate was centrifuged in 35 000 g for 10 min. The
cell pellet was resuspended in 100 mM HEPES buffer contain-
ing 1 mM EGTA (pH 7.4). Membranes (30-40 µg protein) were
incubated at 37°C in a reaction medium containing 100 mM
HEPES (pH 7.4), 5.0 mM MgCl2, 0.5 mM ATP, 1.0 mM cAMP,
0.5 mM IBMX, 10 mM phosphocreatine, 0.31 mg/mL creatine
phosphokinase, 100 µM GTP, 1 µCi R-[33P]ATP per tube, and
test drugs (final volume of 100 µL). Incubations were termi-
nated after 15 min by adding 2% sodium dodecyl sulfate. After
separation of [33P]cAMP from [33P]ATP as described by
Salomon,37 [33P]cAMP levels were determined by liquid scintil-
lation counting, with the results expressed as picomoles per
minute per milligram of protein. EC50 and IC50 values were
calculated by linear regression analysis of the concentration-
response curves. Efficacy values were calculated as the
maximal effect of an agonist in terms of the maximal effect
produced by a known agonist such as 5-HT. Apparent Ki values
for antagonists were calculated as follows: Ki ) IC50/(1 + (C/
EC50)), where C is the concentration of the agonist used in the
experiment and EC50 is the EC50 for the agonist.
(18) Kikuchi, C.; Hiranuma, T.; Koyama, M. Tetrahydrothienopy-
ridylbutyl-tetrahydrobenzindoles: new selective ligands of the
5-HT(7) receptor. Bioorg. Med. Chem. Lett. 2002, 12, 2549-2552.
(19) Kikuchi, C.; Nagaso, H.; Hiranuma, T.; Koyama, M. Tetrahy-
drobenzindoles: Selective antagonists of the 5-HT7 receptor. J .
Med. Chem. 1999, 42, 533-535.
(20) Kikuchi, C.; Suzuki, H.; Hiranuma, T.; Koyama, M. New
tetrahydrobenzindoles as potent and selective 5-HT7 antagonists
with increased In vitro metabolic stability. Bioorg. Med. Chem.
Lett. 2003, 13, 61-64.
(21) Krobert, K. A.; Bach, T.; Syversveen, T.; Kvingedal, A. M.; Levy,
F. O. The cloned human 5-HT7 receptor splice variants:
A
comparative characterization of their pharmacology, function
and distribution. Naunyn S. Arch. Pharmacol. 2001, 363, 620-
632.
(22) Krobert, K. A.; Levy, F. O. The human 5-HT7 serotonin receptor
splice variants: constitutive activity and inverse agonist effects.
Br. J . Pharmacol. 2002, 135, 1563-1571.
(23) Lii, J .-H.; Allinger, N. L. Molecular mechanics. The MM3 force
field for hydrocarbons. 2. Vibrational frequencies and thermo-
dynamics. J . Am. Chem. Soc. 1989, 111, 8566-8575.
(24) Lii, J .-H.; Allinger, N. L. Molecular mechanics. The MM3 force
field for hydrocarbons. 3. The van der Waals' potentials and
crystal data for aliphatic and aromatic hydrocarbons. J . Am.
Chem. Soc. 1989, 111, 8576-8582.
Su p p or tin g In for m a tion Ava ila ble: Table of elemental
analyses. This material is available free of charge via the
Internet at http://pubs.acs.org.
(25) Linnanen, T.; Brisander, M.; Unelius, L.; Rosqvist, S.; Nordvall,
G.; Hacksell, U.; J ohansson, A. M. Atropisomeric derivatives of
2′,6′-disubstituted (R)-11- phenylaporphine: selective serotonin
5-HT(7) receptor antagonists. J . Med. Chem. 2001, 44, 1337-
1340.
(26) Lopez-Rodriguez, M. L.; Porras, E.; Benhamu, B.; Ramos, J . A.;
Morcillo, M. J .; Lavandera, J . L. First pharmacophoric hypoth-
esis for 5-HT7 antagonism. Bioorg. Med. Chem. Lett. 2000, 10,
1097-1100.
(27) Lopez-Rodriguez, M. L.; Porras, E.; Morcillo, M. J .; Benhamu,
B.; Soto, L. J .; Lavandera, J . L.; Ramos, J . A.; Olivella, M.;
Campillo, M.; Pardo, L. Optimization of the pharmacophore
model for 5-HT7R antagonism. Design and synthesis of new
naphtholactam and naphthosultam derivatives. J . Med. Chem.
2003, 46, 5638-5650.
(28) Lovell, P. J .; Bromidge, S. M.; Dabbs, S.; Duckworth, D. M.;
Forbes, I. T.; J ennings, A. J .; King, F. D.; Middlemiss, D. N.;
Rahman, S. K.; Saunders: D. V.; Collin, L. L.; Hagan, J . J .; Riley,
G. J .; Thomas, D. R. A novel, potent, and selective 5-HT7
antagonist: (R)-3-(2-(2-(4-methylpiperidin-1-yl)-ethyl)pyrrolidine-
1-sulfonyl)phenol (SB-269970). J . Med. Chem. 2000, 43, 342-
345.
(29) Lovenberg, T. W.; Baron, B. M.; de Lecea, L.; Miller, J . D.;
Prosser, R. A.; Rea, M. A.; Foye, P. E.; Racke, M.; Slone, A. L.;
Siegel, B. W. A novel adenylyl cyclase-activating serotonin
receptor (5-HT7) implicated in the regulation of mammalian
circadian rhythms. Neuron 1993, 11, 449-458.
(30) Mokrosz, J . L.; Bojarski, A. J .; Charakchieva-Minol, S.; Duszyn-
ska, B.; Mokrosz, M. J .; Paluchowska, M. H. Structure-activity
relationship studies of CNS agents, Part 23: N-(3-phenylpropyl)-
and N-[(E)-cinnamyl]-1,2,3,4-tetrahydroisoquinoline mimic 1-phen-
ylpiperazine at 5-HT1A receptors. Arch. Pharm. (Weinheim)
1995, 328, 604-608.
(31) Mokrosz, J . L.; Deren-Wesolek, A.; Tatarczynska, E.; Duszynska,
B.; Bojarski, A. J .; Mokrosz, M. J .; Chojnacka-Wojcik, E. 8-[4-
[2-(1,2,3,4-Tet r a h ydr oisoqu in olin yl]bu t yl-8-a za spir o[4.5]-
decane -7,9-dione: a new 5-HT1A receptor ligand with the same
activity profile as buspirone. J . Med. Chem. 1996, 39, 1125-
1129.
Refer en ces
(1) Pallas. [Version 1.2]. 1994. CompuDrug Chemistry Ltd.
(2) MacroModel. [Version 7.0]. 1999. Schrodinger Inc.
(3) Sybyl Molecular Modeling Software. [Version 6.8]. 2001. Tripos,
Inc.
(4) Allinger, N. L.; Yuh, Y. H.; Lii, J .-H. Molecular mechanics. The
MM3 force field for hydrocarbons. 1. J . Am. Chem. Soc. 1989,
111, 8551-8566.
(5) Baldwin, J . M.; Schertler, G. F. X.; Unger, V. M. An alpha-carbon
template for the transmembrane helices in the rhodopsin family
of G-protein-coupled receptors. J . Mol. Biol. 1997, 272, 144-
164.
(6) Ballesteros, J . A.; Weinstein, H. Integrated methods for the
construction of three-dimensional models and computational
probing of structure-function relations in G protein-coupled
receptors. Methods Neurosci. 1995, 25, 366-428.
(7) Bard, J . A.; Zgombick, J .; Adham, N.; Vaysse, P.; Branchek, T.
A.; Weinshank, R. L. Cloning of a novel human serotonin
receptor (5-HT7) positively linked to adenylate cyclase. J . Biol
Chem. 1993, 268, 23422-23426.
(8) Bojarski, A. J .; Mokrosz, M. J . 1,2,3,4-Tetrahydroisoquinoline
derivatives; lipophilicity evaluation vs 5-HT1A receptor affinity.
Pharmazie 1999, 54, 828-830.
(9) Bojarski, A. J .; Mokrosz, M. J .; Minol, S. C.; Koziol, A.; Weso-
lowska, A.; Tatarczynska, E.; Klodzinska, A.; Chojnacka-Wojcik,
E. The influence of substitution at aromatic part of 1,2,3,4-
tetrahydroisoquinoline on in vitro and in vivo 5-HT(1A)/5-HT-
(2A) receptor activities of its 1-adamantoyloaminoalkyl deriva-
tives. Bioorg. Med. Chem. 2002, 10, 87-95.
(10) Chang, G.; Guida, W. C.; Still, W. C. An internal-coordinate
Monte Carlo method for searching conformational space. J . Am.
Chem. Soc. 1989, 111, 4379-4386.
(11) Decker, H.; Becker, P. Liebigs Ann. Chem. 1913, 395, 342.
(12) Forbes, I. T.; Dabbs, S.; Duckworth, D. M.; J ennings, A. J .; King,
F. D.; Lovell, P. J .; Brown, A. M.; Collin, L.; Hagan, J . J .;
Middlemiss, D. N.; Riley, G. J .; Thomas, D. R.; Upton, N. (R)-
3,N-dimethyl-N-[1-methyl-3-(4-methyl-piperidin-1-yl)propyl]ben-
zenesulfonamide: The first selective 5-HT7 receptor antagonist.
J . Med. Chem. 1998, 41, 655-657.

5466 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Vermeulen et al.
(32) Mokrosz, M. J .; Bojarski, A. J .; Duszynska, B.; Tatarczynska,
E.; Klodzinska, A.; Deren-Wesolek, A.; Charakchieva-Minol, S.;
Chojnacka-Wojcik, E. 1,2,3,4-tetrahydroisoquinoline deriva-
tives: a new class of 5-HT1A receptor ligands. Bioorg. Med.
Chem 1999, 7, 287-295.
(33) Paluchowska, M. H.; Bojarski, A. J .; Charakchieva-Minol, S.;
Wesolowska, A. Active conformation of some arylpiperazine
postsynaptic 5-HT(1A) receptor antagonists. Eur. J . Med. Chem.
2002, 37, 273-283.
(34) Perrone, R.; Berardi, F.; Colabufo, N. A.; Lacivita, E.; Leopoldo,
M.; Tortorella, V. Synthesis and structure-affinity relationships
of 1-[ïmega-(4-aryl-1-piperazinyl)alkyl]-1-aryl ketones as 5-HT-
(7) receptor ligands. J . Med. Chem. 2003, 46, 646-649.
(35) Plassat, J . L.; Amlaiky, N.; Hen, R. Molecular cloning of a
mammalian serotonin receptor that activates adenylate cyclase.
Mol. Pharmacol. 1993, 44, 229-236.
(36) Ruat, M.; Traiffort, E.; Leurs, R.; Tardivel Lacombe, J .; Diaz,
J .; Arrang, J .; Schwartz, J . C. Molecular cloning, characteriza-
tion, and localization of a high-affinity serotonin receptor (5-
HT7) activating cAMP formation. Proc. Natl. Acad. Sci U.S.A.
1993, 90, 8547-8551.
(37) Salomon, Y. Adenylate cyclase assay. Adv. Cyclic Nucleotide Res.
1979, 10, 35-55.
(38) Shapiro, D. A.; Kristiansen, K.; Kroeze, W. K.; Roth, B. L.
Differential modes of agonist binding to 5-hydroxytryptamine-
(2A) serotonin receptors revealed by mutation and molecular
modeling of conserved residues in transmembrane region 5. Mol.
Pharmacol. 2000, 58, 877-886.
(39) Shen, Y.; Monsma, F. J ., J r.; Metcalf, M. A.; J ose, P. A.; Hamblin,
M. W.; Sibley, D. R. Molecular cloning and expression of a
5-hydroxytryptamine7 serotonin receptor subtype. J . Biol Chem.
1993, 268, 18200-18204.
(41) Speeter, M. E.; Anthony, W. C. The action of oxalyl chloride on
indoles: a new approach to tryptamines. J . Am. Chem Soc. 1954,
76, 6208-6210.
(42) Steiner, T.; Koellner, G. Hydrogen Bonds with p-Acceptors in
Proteins: Frequencies and Role in Stabilizing Local 3-D Struc-
tures. J . Biol. Chem. 2001, 305, 535-557.
(43) Still, W. K.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.
Semianalytical treatment of solvation for molecular mechanics
and dynamics. J . Am. Chem. Soc. 1990, 112, 6127-6129.
(44) Thomas David, R.; Gittins Susan, A.; Collin Lissa, L.; Middlemiss
Derek, N.; Riley Graham; Hagan J im; Gloger Israel; Ellis
Catherine, E.; Forbes Ian, T.; Brown Anthony, M. Functional
characterisation of the human cloned 5-HT7 receptor (long
form): Antagonist profile of SB-258719. Br. J . Pharmacol. 1998,
124, 1300-1306.
(45) Thomas, D. R.; Atkinson, P. J .; Hastie, P. G.; Roberts, J . C.;
Middlemiss, D. N.; Price, G. W. [3H]-SB-269970 radiolabels
5-HT7 receptors in rodent, pig and primate brain tissues.
Neuropharmacology 2002, 42, 74-81.
(46) Thomas, D. R.; Atkinson, P. J .; Ho, M.; Bromidge, S. M.; Lovell,
P. J .; Villani, A. J .; Hagan, J . J .; Middlemiss, D. N.; Price, G.
W. [H-3]-SB-269970 - A selective antagonist radioligand for
5-HT7 receptors. Br. J . Pharmacol. 2000, 130, 409-417.
(47) Vermeulen, E. S.; Schmidt, A. W.; Sprouse, J . S.; Wikstrom, H.
V.; Grol, C. J . Characterization of the 5-HT(7) receptor. Deter-
mination of the pharmacophore for 5-HT(7) receptor agonism
and CoMFA-based modeling of the agonist binding site. J . Med.
Chem. 2003, 46, 5365-5374.
(48) Wesolowska, A. In the search for selective ligands of 5-HT5,
5-HT6 and 5-HT7 serotonin receptors. Pol. J . Pharmacol. 2002,
54, 327-341.
(40) Snyder, J .P.; Rao, S.N.; Koehler, K.F.; Vedani, A.; Pelliciari, R.
APOLLO Pharmacophores and the pseudoreceptor concept. In
Trends in QSAR and Molecular Modeling 92; Wermuth, C. G.,
Ed.; ESCOM: Leiden, 1993; pp 44-51.
J M049743B