N′-(Arylsulfonyl)pyrazoline-1-carboxamidines as novel, neutral 5-hydroxytryptamine 6 receptor (5-HT6R) antagonists with unique structural features

Article abstract of DOI:10.1021/jm200466r

The 5-HT₆ receptor (5-HT₆R) has been in the spotlight for several years regarding CNS-related diseases. We set out to discover novel, neutral 5-HT₆R antagonists to improve off-target selectivity compared to basic amine-con

Full text of DOI:10.1021/jm200466r

ARTICLE
pubs.acs.org/jmc
N0-(Arylsulfonyl)pyrazoline-1-carboxamidines as Novel, Neutral
5-Hydroxytryptamine 6 Receptor (5-HT6R) Antagonists with Unique
Structural Features
Arnold van Loevezijn,* Jennifer Venhorst, Wouter I. Iwema Bakker, Cor G. de Korte, Wouter de Looff,
Stefan Verhoog, Jan-Willem van Wees, Martijn van Hoeve,† Rob P. van de Woestijne,
Martina A. W. van der Neut, Alice J. M. Borst, Maria J. P. van Dongen, Natasja M. W. J. de Bruin,
Hiskias G. Keizer,‡ and Chris G. Kruse
Abbott Healthcare Products B.V. (formerly Solvay Pharmaceuticals B.V.), C. J. van Houtenlaan 36, 1381 CP Weesp, The Netherlands
S Supporting Information
b
ABSTRACT: The 5-HT6 receptor (5-HT6R) has been in the
spotlight for several years regarding CNS-related diseases.
We set out to discover novel, neutral 5-HT6R antagonists to
improve off-target selectivity compared to basic amine-containing
scaffolds dominating the field. High-throughput screening
identified the N0-(sulfonyl)pyrazoline-1-carboxamidine scaffold
as a promising neutral core for starting hit-to-lead. Medicinal
chemistry, molecular modeling, small molecule NMR and X-ray
crystallography were subsequently applied to optimize the leads into antagonists (compounds 1ꢀ49) displaying high 5-HT6R
affinity with optimal off-target selectivity. Unique structural features include a pseudoaromatic system and an internal hydrogen
bond freezing the bioactive conformation. While physicochemical properties and CNS availability were generally favorable,
significant efforts had to be made to improve metabolic stability. The optimized structure 42 is an extremely selective, hERG-free,
high-affinity 5-HT6R antagonist showing good human in vitro metabolic stability. Rat pharmacokinetic data were sufficiently good
to enable further in vivo profiling.
’ INTRODUCTION
synaptic plasticity.18ꢀ20 Such additional effects may make 5-HT6R
antagonists a potential procognitive approach that goes beyond a
purely symptomatic treatment and may lead to prolonged efficacy
in patients.
For over a decade, the 5-HT6R has been in the spotlight for
central nervous system (CNS) related diseases.1,2 One apparent
reason for this is the observation that the 5-HT6R is mainly (though
not exclusively) expressed in the brain.3,4 Here it indirectly regulates
a variety of cholinergic, monoaminergic, and excitatory amino acid
neurotransmitters in a brain-region-specific manner.3,5,6 Although
maturing, our understanding of the complex field of 5-HT6R
neuropharmacology is neither complete nor fully consistent.7 With
the ambiguous functional role of the 5-HT6R, modulation of this
target has been associated with several indications for which unmet
medical needs exist.
One of the strongest interests in the 5-HT6R related to CNS
disorders today concerns cognitive functions of the brain. In
particular, beneficial effects of 5-HT6R antagonism in cognitive
impairment (learning and memory deficits) associated with
schizophrenia (CIAS) and Alzheimer’s disease (AD) are vigor-
ously investigated.8ꢀ17 The downstream signaling pathway(s)
and mechanisms involved in cognitive enhancement are yet to be
fully elucidated. Acute modulation by the 5-HT6R is believed to
occur at least in part through indirect pathways such as disin-
hibition of GABAergic neurons, leading to brain-region-specific
changes in levels of neurotransmitters such as acetylcholine,
glutamate, dopamine, and noradrenaline.3,5,9,15 Additionally, (sub)-
chronic treatment with 5-HT6R antagonists has shown effects on
In clinical studies, 5-HT6R antagonists are evaluated against
the currently established treatment of cognitive symptoms in AD,
dominated by the acetylcholinesterase inhibitors (AChEIs). The
latter compounds are known to lose their efficacy with disease
progression, and tolerability is sometimes limited because of
cholinergic side effects.21 The first disclosed results of clinical
trials in AD suggest that 5-HT6R antagonist monotherapy in the
initial stages of treatment has similar efficacy compared to the
AChEI donepezil22 and seems to be well tolerated. To distin-
guish 5-HT6R antagonists from established symptomatic AD
treatment, it will be interesting to learn whether mechanisms
additional to cholinergic modulation will lead to increased or
prolonged procognitive efficacy in the various disease stages.
Moreover, it remains to be seen whether positive preclinical
results in anxiolytic and antidepressive models23 will translate to
additional clinical benefits in these two noncognitive aspects
associated with AD. Since the mechanism of action on the
cholinergic system is different for 5-HT6R antagonists and AChEIs,
Received: April 18, 2011
Published: August 25, 2011
r
2011 American Chemical Society
7030
dx.doi.org/10.1021/jm200466r J. Med. Chem. 2011, 54, 7030–7054
|

Journal of Medicinal Chemistry
ARTICLE
Figure 1. Disclosed structures of 5-HT6R ligands that are or have been under active clinical development (data from Thomson Reuters Integrity).
adjunctive therapy may also be a valuable option. Indeed
preclinical data have suggested an additive or synergistic effect,24ꢀ26
and this finding is also receiving clinical follow-up.27 It may ultimately
result in improved efficacy or, in case subefficacious doses of one
or both of the individual compounds are used, lead to better
tolerability.
compounds may also inhibit this receptor, albeit with apparent
lower efficacy.4 Soon after completion of our 5-HT6R hit find-
ing activities, a higher affinity nonbasic compound was pub-
lished,45 but further development of this distinct novel chemo-
type has not been reported. Only very recently, a few additional
publications have appeared that disclose more potent 5-HT6R
antagonists lacking a basic amine functionality.46ꢀ50
In contrast to AD, CIAS is unresponsive to treatment with
AChEIs28 and as such represents a clear unmet medical need.29
Rather than cholinergic deficits, disturbed dopaminergic and
glutamatergic neurotransmission is believed to more strongly
underlie CIAS.30 With the emerging evidence that 5-HT6R
antagonism affects multiple neurotransmitters including dopa-
mine and glutamate, this may be a more effective treatment for
cognitive symptoms in schizophrenic patients, as also indicated by
preliminary clinical studies.31 However, 5-HT6R antagonists can-
not treat the full spectrum of positive and negative symptoms of
schizophrenia. Therefore, treatment of CIAS needs to exist as
adjunctive therapy with atypical antipsychotics32 that have a
critically balanced target profile. In such a combination treatment,
a highly selective 5-HT6R antagonist could be considered pre-
ferred. Adjunctive therapy with 5-HT6R antagonists in schizo-
phrenia may have additional benefits beyond cognition. First,
schizophrenic patients are known to be more prone to addiction,33
and 5-HT6R antagonism has been shown to be associated with
antiaddictive behavior in several studies.34ꢀ37 Second, atypical
antipsychotics are frequently associated with body weight gain,38
and 5-HT6R modulators have been shown to reduce food in-
take and body weight in diet-induced obese rodent models, pre-
sumably through a centrally mediated effect on satiety.39,40
All 14 5-HT receptor family members, except 5-HT3, bear the
G-protein-coupled receptor (GPCR) fold with seven character-
istic transmembrane helices. The 5-HT6R is further classified as a
biogenic amine GPCR because of a conserved aspartic acid in the
third transmembrane helix (TM3). In line with the presence of
this residue, virtually all 5-HT6R ligands described at the start of
our research program contain an interaction counterpart in the
formof a basic amine.41ꢀ43 Thisholds true for all selective 5-HT6R
antagonists that are still, or have been, under active clinical develop-
ment as far as their structures have been disclosed44 (Figure 1).
Few exceptions to this rule hinted at the possibility that neutral
Although a protonated amine moiety can demonstrably
promote potency of 5-HT6R ligands, its presence may lead to
adverse effects in the development stage. Nonselective profiles
with respect to both serotonergic subfamily members and other
biogenic amine GPCRs are not uncommon.3 Another liability
is the high potential of positively charged ligands for binding to
the human ether-a-go-go-related gene (hERG),51ꢀ53 which may
translate to QT prolongation and the associated risk of poten-
tially fatal cardiac arrhythmias. This is further heightened by the
fact that 5-HT6R ligands generally contain aromatic groups, which
represents another pharmacophoric feature of hERG inhibitors.
In fact, the 5-HT6R antagonistic framework described in various
papers details the presence of two hydrophobic/aromatic groups,
linked by a (double) hydrogen bond acceptor moiety and
carrying a substituent with a positively ionizable atom.41,42 Drugs
targeting monoaminergic GPCRs in general have an above-
average incidence of hERG affinity. For instance, several atypical
antipsychotics (targeting a relatively broad spectrum of this class
of targets) have been associated with blockade of the hERG
channel.54,55 Combination therapy with 5-HT6R antagonists that
would share this liability might increase the incidence of transla-
tion to severe cardiac safety risks.
Especially in the light of foreseen adjunctive treatments in
CIAS and AD, a high selectivity toward off-targets is of impor-
tance to limit disturbance of concurrent therapies and maximize
safety. In this study we therefore set out to discover novel,
preferably neutral 5-HT6R antagonists to potentially improve
off-target selectivity compared to the basic 5-HT6R scaffolds
dominating the field. High-throughput screening (HTS) identi-
fied the N0-(sulfonyl)pyrazoline-1-carboxamidine scaffold as a
promising core motif. Here we describe how an integrated
approach of medicinal chemistry, molecular modeling, nuclear
magnetic resonance (NMR) techniques, and X-ray crystallography
7031
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Table 1. In Vitro Pharmacological Results of Hit Compound 1 and Close Analogues
a
compd
R1
R2
R4
R6
R7
R8
AlogP
Ki (5-HT6),b nM
pA2 (5-HT6) c
1
H
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
C2H5
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Ph
C2H5
Bn
H
2-Cl-phenyl
2-Cl-phenyl
2-Cl-phenyl
2-Cl-phenyl
Phenyl
2.08
3.32
3.95
2.46
1.42
2.73
3.25
1.73
1.53
2.27
1.16
1.24
1.94
3.42
27.8 ( 7.7
92.0 ( 22.2
>1000
8.2 ( 0.3
<6.0
2
H
H
3
H
Ph
H
<6.0
4
H
i-C3H7
C2H5
C2H5
C3H7
CH3
H
H
>1000
<6.0
5
H
H
33.9 ( 8.8
7.9 ( 4.1
14.6 ( 3.5
17.8 ( 4.0
24.2 ( 4.6
8.0 ( 0.1
33.9 ( 23.0
231.1 ( 55.9
>1000
7.8 ( 0.2
8.3 ( 0.1
7.3 ( 0.2
8.2 ( 0.4
7.4 ( 0.2
8.1 ( 1.1
7.1 ( 0.1
6.1 ( 0.3
<6.0
6
H
H
3-Cl-phenyl
3-Cl-phenyl
3-Cl-phenyl
3-Cl-phenyl
3-Cl-phenyl
3-Cl-phenyl
3-Cl-phenyl
3-Cl-phenyl
3-Cl-phenyl
7
H
H
8
H
H
9
H
H
10
11
12
13
14
H
C2H5
C2H5
C2H5
CH3
C2H5
H
H
H
CH3
H
H
H
C2H5
H
CH3
H
H
23.9 ( 5.4
7.7 ( 0.1
a AlogP was calculated with Pipeline Pilot (Accelrys, Inc.; San Diego, CA, U.S.). b Displacement of specific [3H]N-methyllysergic acid diethylamide
([3H]LSD) binding in Chinese hamster ovary (CHO) cells stably expressing human 5-HT6 receptor, expressed as Ki ( SEM (nM). c Luminescence
measurement from a CHO-human-5-HT6-aequorin assay, expressed as pA2 ( SEM.
was applied to further develop the initial hit structure into a class
of high-affinity, CNS druglike 5-HT6R antagonists displaying
optimal selectivity toward off-targets.
’ RESULTS AND DISCUSSION
Hit Identification and Hit-to-LeadActivities. A high-through-
put screening campaign was initiated to identify novel antagonists
of the 5-HT6R, preferably lacking a basic nitrogen. Screening of a
diverse cross-section of our internal compound stock using an in-
Figure 2. Analogability of identified hit class.
house 5-HT6R functional assay resulted in a variety of interesting
hits belonging to distinct structural classes. Among them, com-
pound 1 sparked immediate interest because of its neutral con-
stituency, leadlike lipophilicity for a CNS drug (AlogP = 2.08) and
favorable functionalactivity, confirmed by a promisingresult in the
receptor binding assay (Ki = 27.8 nM, Table 1) after resynthesis.
The identified N0-(arylsulfonyl)pyrazoline-1-carboxamidine hit
class, sharing its core structure with one of our main classes of
CB1 cannabinoid receptor antagonists,56 provided ample oppor-
tunity for structural variation to explore the structureꢀactivity
relationships (SARs) (Figure 2).
Initial follow-up on this hit was conducted by exploration of
close analogues (Table 1), synthesized according to Scheme 1.
Pyrazoline building blocks 60, with up to one substituent on the
C3, C4, and C5 positions, were synthesized from α,β-unsatu-
rated aldehydes or ketones 59 by reaction with hydrazine.57
Following literature examples56 these were coupled with N-(bis-
methylsulfanylmethylidene)arylsulfonamides 62, prepared from
their corresponding arylsulfonamides 61, CS2, and methyl io-
dide, to furnish intermediates 63. Nucleophilic attack of 63 with
amines 64 gave target compounds 1ꢀ14.
Initial medicinal chemistry efforts focused on varying the R6
and R7 N-substituents on the carboxamidine moiety. As evi-
denced by compounds 2, 3, and 4, substitution of the R6 ethyl
group in 1 by more bulky groups is detrimental for 5-HT6R
functional activity. Discrete modifications were subsequently
introduced at the R8 aryl group connected to the sulfonyl moiety.
Removal of the 2-chloro atom from the phenyl ring (compound
5, AlogP reduced to 1.48) only has a marginal effect on the
activity and affinity compared to 1. In contrast, the 3-chlorophe-
nyl substitution in compound 6 (AlogP = 2.73) showed a more
than 3-fold increase in affinity. In combination with this R8
substituent, additional R6 variations with both elongation (7)
and shortening (8 and 9) of the N-ethyl group were investigated.
Neither strategy resulted in improved 5-HT6R antagonism.
Subsequently, the effects of differential substitution of the
pyrazoline ring were studied. In the obtained racemic mixtures,
reducing the ethyl R2 substituent of 6 to a methyl in 10 only had a
marginal effect. However, complete eradication of this group
(11) resulted in a sharp drop in both receptor binding and
functional activity. Introduction of even a small methyl R1
7032
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Scheme 1. Synthesis of Initial Hit Exploration Compoundsa
a Reagents and conditions: (a) H2NNH2 H2O, MeOH or CH3CN, 0 °C to room temp [R4 = H] or H2NNH2 H2O, t-BuOH, reflux [R4 = Ph]; (b) CS2,
3
3
KOH, DMF/H2O, 10 °C; (c) MeI, 10 °C to room temp; (d) 60, pyridine, reflux; (e) R6R7NH [64, R6,R7 ¼ Ar], MeOH, room temp or R6R7NH [64,
R6 = Ph, R7 = H], NaHMDS, THF, 0 °C to room temp.
Scheme 2. Alternative Route Used for Synthesis of N0-(Acyl)pyrazoline-1-carboxamidine Analogue 15a
a Reagents and conditions: (a) MeI, EtOH, 0 °C to room temp; (b) 60, DiPEA, toluene, reflux; (c) HCl, EtOH, room temp; (d) R8COCl, DiPEA,
CH2Cl2, room temp.
substituent, evidenced by compound 12, resulted in a marked
loss of activity. Compared to 8, replacement of the R7 hydrogen
by a methyl substituent (13) led to a dramatic decrease in
activity. Finally, the so far untouched C5-position of the pyrazo-
line ring was probed by introduction of an aryl functionality
(R4 = Ph). For the racemic mixture (14) this resulted in
interesting pharmacodynamic properties and as such identifica-
tion of another subseries with potential for further lead optimiza-
tion, albeit with a rather high AlogP (3.42).
To conclude the initial SAR studies, a single attempt was made to
replace the sulfonyl moiety in lead optimization candidate 6 by a
carbonyl group. Synthesis was achieved via an alternative route,
based on literature precedent56 with slight modifications, as depicted
in Scheme 2. Thiourea 65a was methylated to furnish a suitable
leaving group in S-methylisothiourea 66a, which could be sub-
stituted with pyrazoline building block 60a to give the pyrazoline-
1-carboxamidine core 67a. Regioselective acylation of the 67a with
3-chlorobenzoyl chloride gave target compound 15. Its lack of
5-HT6R affinity (Ki > 1000 nM) and functional activity (pA2 < 6)
demonstrated the importance of the sulfonyl linker, as seen in
numerous 5-HT6R ligands.
in the hit exploration study, some additional profiling of
these compounds was performed. As anticipated because of
the absence of a basic nitrogen,58 both are devoid of hERG
affinity with pIC50 < 5 measured at Zenas.59 Screening against a
panel of 40 targets at Cerep,60 dominated by monoaminergic
GPCRs including homologous human 5-HT receptor subtypes
1A, 1D, 2A, 2C, and 7, and the human D2 receptor, additionally
showed that these lead compounds are highly specific for the
5-HT6R with all measured off-target Ki > 1000 nM (see Supporting
Information for full data). Notably, despite an identical core
structure and substitution on the arylsulfonyl and guanidine N
highly similar to our previously disclosed CB1 antagonists,56 high
selectivity against the CB1 receptor was also observed. This is
due to the fact that for CB1 affinity aryl substituents on both the
pyrazoline C3 (R1) and C4 (R2) positions are a prerequisite.
The high membrane passage of 6 and 14 (39.8 and 38.5%,
respectively, likely aided by their nonionizable character) and
their low P-glycoprotein mediated efflux (P-gp factor 1.0 and
1.4, respectively) were also favorable. However, both com-
pounds were found to have quite low metabolic stability in both
human (Clint of 53 and 73 (μL/min)/106 cells, respectively)
and rat (Clint of 92 and 116 (μL/min)/106 cells, respectively)
hepatocytes. As such, improving this was an issue requiring
rigorous investigation in follow-on lead optimization.
As compounds 6 (pyrazoline 4-alkyl subseries) and 14
(pyrazoline 5-aryl subseries) elicited the most favorable
profiles in terms of 5-HT6R affinity and antagonistic potency
7033
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Figure 3. Docking poses of 5-HT6R antagonists: (A) compound 6; (B) (S)-isomer of compound 14; (C) compound 50 (GSK-742457); (D)
compound 47. Conserved interactions featured by all ligands involve hydrogen bonds (shown in yellow dotted lines) with S193 and N288. The basic
antagonists 50 and 47 furthermore engage in a salt bridge with D106, which is conserved in biogenic GPCRs. The neutral ligands in contrast display π-
stacking and/or hydrophobic interactions with residues in the hydrophobic pocket, which is formed when D106 adopts the gauche+ conformation. In
this conformation D106 is “neutralized” by Y310. Notably, the latter interaction (including a ligand salt bridge with the conserved aspartate) has also
been observed for corresponding residues in crystallized GPCR complexes with basic compounds, e.g., carazolol and timolol bound to the β2-adrenergic
receptor (PDB codes 2RH1 and 3D4S).61 Ballesteros numbering of the featured residues is the following: D106/D3.32; S193/S5.43; F197/F5.46;
F284/F6.51; F285/F6.52; N288/N6.55; Q291/Q6.58; Y310/Y7.43.
Design. In the next step we strived to better understand the
limited SAR around the N0-(sulfonyl)pyrazoline-1-carboxamidine
core obtained thus far. To take optimal advantage of unexplored
optimization opportunities, we engaged in docking studies using a
5-HT6R homology model (see Experimental Section for details).
Particular points of interest to address included the site and mode
of binding compared to ionizable reference structures and the
nature of the interactions resulting in the somewhat unexpected
high binding affinity observed for this neutral class.
Docking studies of representatives from the series showed
striking features, illustrated by compounds 6 and 14 in Figure 3A
and Figure 3B. The binding site of the N0-(arylsulfonyl)pyrazoline-
1-carboxamidines almost entirely overlaps with that of ionizable
5-HT6R antagonists such as 3-benzenesulfonyl-8-piperazin-1-yl-
quinoline (GSK-742457, 50) currently in clinical phase II (Figure 3C).
Common interaction residues include serine-193 (S193) and
asparagine-288 (N288). Both residues hydrogen-bond to the
sulfonyl moiety of 50. In the case of 6 and 14, S193 similarly
hydrogen-bonds to the sulfonyl group. Residue N288, however,
is able to interact with both the sulfonyl oxygen and amide lone
pair of the compounds. Interestingly, the 3-chloro substituent
in 6 is able to engage in polar interaction with glutamine-291
(Q291), which in turn stabilizes N288. This is nicely in line with
the observed decrease in Ki of 6 compared to the unsubstituted
and 2-chloro substituted phenyl analogues 5 and 1. This polar
interaction is not present in the 5-aryl substituted pyrazoline,
compound 14. The introduction of the bulky R4 group requires
small reorientations of the structure in the binding pocket
compared to 6, thus moving the 3-chloro substituent on the R8
phenyl ring further away from Q291. The absence of this polar
interaction is however compensated by π-stacking interactions of
the R4-aryl group with phenylalanine-284 (F284). This was only
observed for the (S)-isomer; the (R)-isomer could not be docked
because of steric restrictions. In contrast, compound 6 with its R2
ethyl substituent on the pyrazoline 4-position could be docked in
both configurations without preference.
A notably conserved interaction is the π-stacking interaction
with phenylalanine-285 (F285), held in place by phenylalanine-
197 (F197). Where 50 optimally positions its quinoline moiety
for stacking interactions, a pseudoaromatic system is formed by
the conjugated pyrazoline-1-carboxamidine moiety of 6 and 14,
satisfying the same molecular interactions. Another prominent
feature of active members from the N0-(arylsulfonyl)pyrazoline-
1-carboxamidine structural class, deduced from docking studies,
7034
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Figure 4. Change in chemical shift of the carboxamidine NH protons, and a reference proton, as a result of titration with DMSO. Illustrated are the Δδ
values observed for 9 and 6 of the para proton of the phenylsulfonyl group (blue), the carboxamidine NH hydrogen (R7) bonded to the sulfonyl group
(green), and a free carboxamidine NH (R6) in 9 (red).
is the presence of an internal hydrogen bond in a six-membered
ring topology62 between the R7 hydrogen substituent and one of
the sulfonyl oxygens. This intramolecular interaction stabilizes
the bioactive conformation, and the resulting favorable entropy
contribution adds to the high affinity of this structural class
despite the absence of a basic nitrogen. The latter, when present,
interacts with aspartate-106 (D106) through a salt bridge as
illustrated in Figure 3C for 50. In the case of the neutral
N0-(arylsulfonyl)pyrazoline-1-carboxamidines, D106 has chan-
ged rotamers from the trans to the gauche+ conformation. In this
conformation the negative charge is facing away from the
antagonist and toward tyrosine-310 (Y310), with which it forms
an alternative hydrogen bond. A hydrophobic pocket is thus
shaped to accommodate apolar groups from the ligands. As
illustrated by 11, 10, and 6, increasing the size of the aliphatic R2
substituent indeed provides an opportunity to increase ligand
affinity via additional hydrophobic interactions. The low affinity
of compound 13 is also in line with modeling efforts, as the
presence of both an R6 and an R7 alkyl substituent precludes
intramolecular hydrogen bond formation with the sulfonyl group
that stabilizes the bioactive conformation.
Validation of the Binding Mode Hypothesis by Ligand-
Based NMR and X-ray Diffraction. The characteristic intramo-
lecular hydrogen bond featured in the binding poses of the
N0-(arylsulfonyl)pyrazoline-1-carboxamidines prompted us to
investigate whether this interaction can be measured in solution
by NMR. If detectable, it would indicate this internal interaction
to be highly favorable and to result in a low energy conformation.
Several representative compounds were used to study the
change in shift (Δδ) of an NH peak upon addition of dimethyl-
sulfoxide (DMSO).63 Any type of hydrogen bonding of the
studied compound with DMSO will result in deshielding of the
bonded proton. Therefore, a downfield shifted proton resonance
upon titration with DMSO indicates increased hydrogen bond-
ing character of the compound NH with DMSO. A strongly
hydrogen-bonded NH, such as in the hypothesized internal
hydrogen bond, will already show the upfield shift and will resist
solvation by DMSO. An observed small Δδ for the compound’s
NH thus confirms the existence of an internal hydrogen bond. As
shown in Figure 4, compound 9 contains two NH peaks. One is
solvatable, showing a large downfield shift upon DMSO titration,
while the other is clearly bound. For comparison, the shift of the
para-phenyl proton is shown as well. Compound 6 also displays
little solvation, indicating its NH to be already engaged in quite a
strong hydrogen bonding interaction. Additionally studied ana-
logues with a single carboxamidine NH proton displayed the
same pattern (data not shown). One may argue that the internal
hydrogen bond can alternatively involve the N atom at position 2
in the pyrazoline ring instead of the sulfonyl oxygen. To exclude
this possibility, an analogue lacking this possible interaction was
also studied (pyrazoline replaced by 2-aza-bicyclo[2.2.2]octane).
This again resulted in a similar pattern (data not shown),
confirming that the hydrogen bond interaction occurs between
the sulfonyl oxygen and carboxamidine NH.
The homology model predicted only the (S)-enantiomer of 14
to be active. As an additional check for the binding mode
hypothesis, racemic 14 was separated by chiral preparative high
performance liquid chromatography (HPLC) into its two en-
antiomers 16 and 17 and tested in the 5-HT6 receptor binding
and functional assays (Table 2). The (ꢀ)-isomer 16 was shown
to be active, whereas the (+)-isomer 17 was inactive. Crystal-
lization attempts were carried out for both enantiomers, and the
crystals of the inactive (+)-enantiomer 17 were of suitable quality
for X-ray diffraction. Both the Flack x parameter64 value
(ꢀ0.02 ( 0.004) and the independent analysis of the data in
terms of the Hooft parameter65 (ꢀ0.002 ( 0.012) indicated an
enantiomerically pure structure of 17 with a high level of confidence.
As depicted in Figure 5, the chiral carbon in 17 proved to have the
(R)-configuration. This means that, in line with the predictions from
the protein model, the active (ꢀ)-enantiomer 16 has the (S)
absolute stereochemistry. Interestingly, the crystal structure of 17
features the same internal hydrogen bond found by NMR and
docking studies, again indicating this to be an abundantly populated
conformation for this structural class.
Weak Spot Analysis. Metabolic stability was identified as an
issue for further optimization, based on the early ADME
(absorption, distribution, metabolism, and excretion) properties
of leads 6 and 14. Therefore, weak spot predictions were used to
steer compound optimization using the application MetaSite.66
First, both compounds were profiled in a cytochrome P450
(CYP) reaction phenotyping assay at Cyprotex67 to identify
which of the main CYP isoforms were major metabolizers.
7035
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Table 2. In Vitro Pharmacological Results of the 5-Aryl Substituted Pyrazolines
compd
14 racemate
R4
AlogP a
Ki (5-HT6),b nM
pA2 (5-HT6),c
Clint, human,d
Clint, rat,d
phenyl
3.42
3.42
3.42
2.27
2.06
2.98
23.9 ( 5.4
9.3 ( 3.5
>1000
7.7 ( 0.1
8.0 ( 0.2
<6.0
73
116
116
107
87
16 (S)-(ꢀ)-enantiomer
17 (R)-(+)-enantiomer
18 racemate
phenyl
107
ND
50
phenyl
pyridin-4-yl
furan-2-yl
4-F-phenyl
321.8 ( 42.5
8.2 ( 2.5
138.1 ( 43.1
6.9 ( 0.4
8.1 ( 0.1
6.4 ( 0.1
19 racemate
92
198
126
20 racemate
107
a AlogP was calculated with Pipeline Pilot (Accelrys, Inc.; San Diego, CA, U.S.). b Displacement of specific [3H]N-methyllysergic acid diethylamide
([3H]-LSD) binding in CHO cells stably expressing human 5-HT6 receptor, expressed as Ki ( SEM (nM). c Luminescence measurement from a CHO-
human-5-HT6-aequorin assay, expressed as pA2 ( SEM. d In vitro intrinsic clearance from hepatocyte assay [(μL/min)/106 cells].
Figure 6. Prominent metabolic weak spots of 6 and 14 predicted by
MetaSite’s 2C19 and 3A4 models.
Figure 5. X-ray structure of 17, the (R)-isomer of racemic 14, featuring
the internal hydrogen bond indicated by docking to be present in the
bioactive conformation of this compound.
felt that screening of racemic mixtures would suffice. In order to
investigate whether modifications in the pyrazoline C5-position
would produce more stable and/or more active compounds,
several analogues with the R4 phenyl substituent replaced by a
In either case, this proved to be CYP 2C19 and CYP 3A4. We
five- or six-membered heteroaryl group were prepared. The latter
analogues are represented in Table 2 by compounds 18 (AlogP =
2.27) and 19 (AlogP = 2.06), which have reduced lipophilicity
compared to 14 (AlogP = 3.42). In 18, one of the highly ranked
weak spots by MetaSite (the para position of the phenyl ring) is
directly modified by replacing the CH for an N. In addition, we
tried to block this predicted weak spot by introducing a fluoro
substituent on the ring (compound 20, AlogP = 2.98). These test
compounds were synthesized through the intermediates depicted
in Scheme 3. The followed route is analogous to the general one
depicted in Scheme 1. If the required α,β-unsaturated aldehyde
substrate 59 (R1 = R2 = H) was not commercially available, it was
synthesized from the R4-aldehyde 68 via a Wittig-type reaction.68
Both replacement of the phenyl by a 4-pyridyl or a 4-fluoro-
phenyl ring had a strong detrimental impact on 5-HT6R affinity
and activity, whereas metabolic stability did not improve sig-
nificantly. The phenyl to 2-furyl replacement in compound 19
resulted in retained 5-HT6R activity, although without any sign of
improved resistance despite a marked reduction in lipophilicity.
This series was therefore abandoned, and attention was refo-
cused on modifications in the guanidine N-alkyl, arylsulfonyl, and
pyrazoline C4 substituents.
thus focused on MetaSite’s rankings for these specific isoforms.
The most highly ranked weak spots from the MetaSite prediction
are depicted in Figure 6.
For 6 the weak spots amenable for further optimization included
the R8 aryl substituent on the sulfonyl moiety, the R6 carboxamidine
N-ethyl group, the C4-position of the pyrazoline ring, and the R2
alkyl substituent attached to this position. According to the docking
studies, ample room for additional substitution is available at the
R8 and R2 positions. These positions thus warranted extensive
further exploration while trying to avoid rigorous rises in lipophi-
licity (6, AlogP = 2.73) and polar surface area (6, PSA = 82 Å2) that
would endanger CNS availability. Given that both initial SAR and
docking studies indicated limited substitution possibilities in groups
at the R6 position, only discrete variations were considered here. But
first, the 5-aryl substituted analogue 14 (AlogP = 3.42, PSA = 82 Å2),
with its bulkier R4 substitution at the pyrazoline C5 position, was
further explored.
Lead Optimization: Pyrazoline 5-Aryl Substitution. Chiral
separation of 14 showed that the active (S)-(ꢀ)-enantiomer 16
displayed good activity toward the 5-HT6R with a Ki of 9.3 nM
and a pA2 of 8.0, whereas the (R)-(+)-enantiomer 17 was inactive
(Table 2). Moreover, no significant difference in metabolic
stability between the racemate and its two pure enantiomers
was observed. Therefore, to evaluate the R4 modifications, it was
Lead Optimization: N-Alkyl Substitutions. Early SAR stud-
ies involving R6 demonstrated that the carboxamidine N-ethyl
7036
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Scheme 3. Synthesis of Pyrazoline C5 (Hetero)aryl Analoguesa
a Reagents and conditions: (aa) Ph3PdCHꢀCHO, DMF, room temp; (a) H2NNH2 H2O, t-BuOH, reflux [R4 = 2-furyl] or H2NNH2 H2O, Et2O or
3
3
THF, ꢀ10 °C to room temp [R4 = 4-piperidyl or 4-F-Ph]; (b) CS2, KOH, DMF/H2O, 10 °C; (c) MeI, 10 °C to room temp; (d) 60, pyridine, 60 °C or
reflux; (e) EtNH2 [64, R6 = Et, R7 = H], MeOH, room temp.
Scheme 4. Synthesis of Carboxamidine NH-R6 Analoguesa
A wide range of variations was introduced with a representative
selection listed in Table 3. These include several bicyclic
(hetero)aromatic ring systems that could be accommodated by
our 5-HT6R homology model. As set out in Scheme 5, these
compounds are accessible via two distinct routes. The first option
is essentially the strategy as followed in Scheme 1, in which the R8
arylsulfonyl substituent is introduced early in the route. Options
are left for diversification with different pyrazoline and amine
building blocks in a later stage. This strategy was for example
chosen to evaluate arylsulfonyl substituents present in basic
competitor 5-HT6R modulators, such as SB-271046 (55) and
a Reagents and conditions: (a) CF3CH2NH2, MeOH, 60 °C (sealed
WAY-181187 (52) leading to analogues 35 and 36, respectively.
tube); (b) CH3ONH2 HCl, Et3N, MeOH, 50 °C.
3
However, in this case a more prominent need to introduce
diversity in the R8 substituent as late as possible in the route
existed. We therefore also developed a suitable parallel synthesis
approach complying with these criteria. Analogous to the acyla-
tion reaction set out in Scheme 2, it was found that reaction of
67a with arylsulfonyl chlorides under mild conditions led to
formation of the target compounds in a highly regiospecific
manner. The neutral character of the products allowed a suitable
parallel workup procedure. Extraction of the reaction mixture
with 2 N aqueous NaOH quenched any remaining sulfonyl
chloride, removing the sulfonic acids into the aqueous layer.
Subsequent SPE purification of the organic phase over strong
cation exchange (SCX) columns caught any remaining basic core
67a while eluting the neutral products. With a few exceptions, the
target compounds were directly obtained from evaporation
without the need for further purification.
Overall it can be concluded that quite substantial variations in
R8 are tolerated, without dramatic loss or gain of 5-HT6R activity.
This concurs with docking studies, as does the observation that
the 3-chloro atom of 6 fulfills a particular role in 5-HT6R binding
(Figure 3A). All other monosubstitutions introduced at the
phenylsulfonyl 3-position (24ꢀ27) exhibited lower activity
toward the 5-HT6R and at best slightly improved human but
similar rat metabolic stability. Moving the chloro substituent
from the 3- to the 4-position (23) was allowed but without
obvious benefit. Introduction of an additional chloro substituent
onto 6 only had a positive effect on 5-HT6R activity when placed
at the 4-position (28, rise of AlogP to 3.39) but rendered the
substituent was most favorable for 5-HT6R affinity and potency
(Table 1). At the same time, however, this group constitutes a
weak spot. Hydroxylation at the α-position as predicted by MetaSite
(Figure 6) would lead to N-dealkylation. Indeed incubation of 6
with human hepatocytes led to clearly detectable levels of the N-
desethyl analogue 9 (data not shown). Although this metabolite is
active, it is less potent than the parent compound. We therefore
attempted some final modifications of the α- and β-positions of the
N-ethyl substituent to modulate the rate of metabolism (Scheme 4).
Substitution of the terminal carbon atom with three fluorines
(21, AlogP = 2.74) resulted in a marked reduction in receptor
binding and functional activity (Ki = 189.4 ( 30.9 nM; pA2 =
7.0 ( 0.1), whereas compared to 6, no improvement in metabolic
stability was observed (Clint of 43 and 99 (μL/min)/106 cells in
human and rat hepatocytes, respectively). Modification of the α-
position by going to the N-methoxy derivative 22 (AlogP = 2.22)
was found to be detrimental for 5-HT6R activity (Ki = 297.5 (
137.7 nM; pA2 < 6). With our attempts being unsuccessful and
further options being limited because of steric restraints, this line
of SAR exploration was terminated.
Lead Optimization: Arylsulfonyl Substitutions. As com-
pound 6 still proved to be the best lead so far, this structure was
used as the basis for further optimization focused on the arylsulfonyl
moiety. Several analogues were specifically synthesized to address
weak spots on the R8 phenyl ring, whereas others were part of a
broader SAR exploration also covering heteroaromatic moieties.
7037
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Table 3. In vitro Pharmacological Results of the Arylsulfonyl SAR and Metabolic Stability studies
compd
R8
AlogP a
Ki (5-HT6),b nM
pA2 (5-HT6),c
Clint, human,d
Clint, rat,d
1
2-Cl-phenyl
2.08
2.73
2.73
1.62
2.36
1.90
1.40
3.39
2.75
2.75
2.75
2.29
2.97
2.32
4.20
1.95
1.66
0.81
0.94
27.8 ( 7.7
7.9 ( 4.1
18.5 ( 7.0
17.0 ( 4.4
17.9 ( 7.4
27.4 ( 7.5
26.6 ( 6.6
7.9 ( 0.0
21.9 ( 9.1
119.4 ( 6.5
196.7 ( 32.2
2.7 ( 0.2
5.6 ( 1.5
17.2 ( 7.2
8.5 ( 2.2
7.9 ( 4.2
4.1 ( 2.0
510.1 ( 67.4
<6.0
8.2 ( 0.3
8.3 ( 0.1
8.2 ( 0.3
7.9 ( 0.1
7.5 ( 0.3
7.1 ( 0.2
7.0 ( 0.3
8.9 ( 0.1
7.6 ( 0.2
7.0 ( 0.0
6.8 ( 0.1
8.1 ( 0.2
8.6 ( 0.3
8.1 ( 0.1
8.1 ( 0.1
8.0 ( 0.2
8.5 ( 0.2
6.9 ( 0.2
>1000
39
53
40
33
27
51
46
92
139
ND
ND
36
77
41
82
20
39
11
4
99
6
3-Cl-phenyl
92
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
4-Cl-phenyl
99
3-F-phenyl
116
116
116
92
3-CF3-phenyl
3-Me-phenyl
3-MeO-phenyl
3,4-di-Cl-phenyl
3,5-di-Cl-phenyl
2,5-di-Cl-phenyl
2,6-di-Cl-phenyl
2-F-3-Cl-phenyl
1-naphthyl
198
126
116
126
116
154
139
154
116
126
53
2-naphthyl
5-chloro-3-methylbenzo[b]thiophen-2-yl
6-chloroimidazo[2,1-b]thiazol-5-yl
5-chlorothiophen-2-yl
5-chloro-1,3-dimethyl-1H-pyrazol-4-yl
4-methylsulfonylphenyl
20
a AlogP was calculated with Pipeline Pilot (Accelrys, Inc.; San Diego, CA, U.S.). b Displacement of specific [3H]N-methyllysergic acid diethylamide
([3H]LSD) binding in CHO cells stably expressing human 5-HT6 receptor, expressed as Ki ( SEM (nM). c Luminescence measurement from a CHO-
human-5-HT6-aequorin assay, expressed as pA2 ( SEM. d In vitro intrinsic clearance from hepatocyte assay [(μL/min)/106 cells].
Scheme 5. Synthesis of R8 Arylsulfonyl Analogues via Two Approachesa
a Reagents and conditions: (a) (i) R8-SO2Cl, DiPEA, DCM, room temp; (ii) 2 M aqueous NaOH extraction; (iii) SCX purification; (b) CS2, KOH or
NaOH, DMF/H2O, 10 °C; (c) MeI, 10 °C to room temp; (d) 60a, pyridine, reflux; (e) EtNH2 [64, R6 = Et, R7 = H], MeOH, room temp.
structure even more vulnerable to metabolism. Despite being a
predicted weak spot, substitution of the 5-position in 6 did not
improve metabolic stability, as evidenced by 29. On an R8 phenyl
substituent carrying two hydrophobic substituents, a 2,3- or 3,4-
disubstitution pattern seemed most favorable for 5-HT6R affinity
(32, 33, 34), but this did not lead to the desired increase in
metabolic stability. Strong negative effects on activity were seen
when (di)substitution close to the sulfonyl linker forced the
(hetero)aryl ring in distinct conformations (30, 31, and 38).
Monocyclic heteroaryl groups with properly positioned substit-
uents are well tolerated, exemplified by the less lipophilic
analogue 37 (AlogP = 1.66). Although human metabolic stability
of the latter increased relative to 6, it actually decreased in rat,
which would imply a serious limitation for preclinical research.
The 4-methylsulfonylphenyl substituent in 39 relinquished
5-HT6R activity altogether, indicated to be due to steric hindrance
in the pocket by docking. Notably, this even more hydrophilic
analogue (AlogP = 0.94) is the only compound demonstrating
7038
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Table 4. In Vitro Pharmacological Results of the Pyrazoline C4 SAR and Metabolic Stability Studies
compd
6 racemate
R2
R3
R8
AlogP a
Ki (5-HT6),b nM
pA2 (5-HT6),c
Clint, human,d
Clint, rat,d
C2H5
C2H5
C2H5
CH3
H
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
3-Cl-Ph
Ph
2.73
2.73
2.73
2.69
2.95
2.58
3.68
1.92
1.00
1.74
2.02
7.9 ( 4.1
8.3 ( 0.1
8.0 ( 0.2
7.5 ( 0.2
7.8 ( 0.1
7.0 ( 0.2
8.2 ( 0.03
8.4 ( 0.3
8.0 ( 0.4
8.7 ( 0.1
<6.0
53
59
49
14
34
26
48
5
92
107
126
77
40 (+)-enantiomer
H
2.8 ( 0.5
41 (ꢀ)-enantiomer
H
16.4 ( 3.8
24.1 ( 5.2
49.9 ( 26.5
6.4 ( 3.1
42
43
44
45
46
47
48
49
CH3
C2H5
C2H5
116
116
139
82
(CH2)4
(CH2)5
5.2 ( 3.6
(CH2)2O(CH2)2
71.1 ( 14.9
7.2 ( 0.7
(CH2)2NH(CH2)2
3
25
CH2F
CH2F
92.0 ( 22.2
146.6 ( 42.8
11
63
CH3
CH3
7.2 ( 0.2
1
41
a AlogP was calculated with Pipeline Pilot (Accelrys, Inc.; San Diego, CA, U.S.). b Displacement of specific [3H]N-methyllysergic acid diethylamide
([3H]LSD) binding in CHO cells stably expressing human 5-HT6 receptor, expressed as Ki ( SEM (nM). c Luminescence measurement from a CHO-
human-5-HT6-aequorin assay, expressed as pA2 ( SEM. d In vitro intrinsic clearance from hepatocyte assay [(μL/min)/106 cells].
significantly lowered intrinsic clearance in both human and rat
hepatocytes. Finally, the competitor-derived bicyclic heteroaro-
matic arylsulfonyl substituents in 35 (rise of AlogP to 4.20 and
PSA to 110 Å2) and 36 (more attractive AlogP of 1.95) gave
activities similar to that of the lead structure. The latter showed
improved human metabolic stability but somewhat lower rat
metabolic stability. Another drawback of 36 was its affinity for the
P-gp transporter (P-gp factor of 1.8), which is absent for other
active compounds in Table 3. Moreover, its high calculated
molecular polar surface area (PSA) of 128 Å2 (vs 82 Å2 for 6)
would not be beneficial either for CNS availability.69
In conclusion, none of the compounds evaluated resulted in an
improved overall profile compared to 6 in terms of 5-HT6R
affinity and activity, metabolic stability, and molecular properties.
Especially improvements in rat metabolic stability appeared to be
challenging for this compound class. Attempts to tackle the
metabolic issue were therefore reverted to variations at the
pyrazoline ring.
Lead Optimization: Pyrazoline 4-Substitutions. As sup-
ported by modeling efforts, chiral separation of the racemic lead
6 showed that indeed both enantiomers are active on the
5-HT6R, albeit with a moderate preference for the isomer with
(+) optical rotation 40. No significant difference was seen in
intrinsic clearance of the (+)-enantiomer 40 and the (ꢀ)-
enantiomer 41. However, disubstitution of the pyrazoline C4
position would address the predicted weak spot on that position
and seemed a viable opportunity given the lack of clear pre-
ference for any of the two enantiomers. As docking studies
indicated that larger substituents at the pyrazoline C4-position
may also be accommodated, di- and spiro-substituted analogues
were synthesized, preferably with symmetrical substitution to
obtain achiral compounds (Table 4).
We chose for base-assisted alkylation of malononitrile (69): for
the acyclic disubstituted compounds two subsequent cycles of
deprotonation followed by attack onto R2-X or R3-X (X = leaving
group), and for the anticipated spiro analogues reaction with a
linear molecule X-R2-Y-R3-X bearing a leaving group X at both
termini in the presence of 2 equiv of base.70,71 The nitrile groups
of the formed dialkylated malononitriles 70 were fully reduced to
the corresponding diamines 71 using LiAlH472 or BH3.73 Oxida-
tive ring closure of the diamines74 furnished the 4,4-disubstituted
4,5-dihydro-3H-pyrazoles 72 (double bond between both
nitrogens), which can tautomerize to the thermodynamically
more stable 4,4-disubstituted 4,5-dihydro-1H-pyrazoles 73
(double bond between N2 and C3) under subsequent reaction
conditions (or upon prolonged storage) to be incorporated into
the target molecules according to the previously described routes
through either intermediate 74 or 75 (Scheme 6).
For the only basic compound from this series, 47, the nitrogen
atom ending up in the spiro piperidyl ring was protected with a
benzyl group throughout the whole synthesis and deprotected
with 1-chloroethyl chloroformate (ACE-Cl) in the final step (to
prevent concomitant reduction of the chloro substituent on the
phenylsulfonyl moiety, which would take place upon catalytic
hydrogenation). The symmetrical diamine intermediate 71g,
used to synthesize the pyrazoline 72g incorporated in compound
48 (R2 = R3 = CH2F), was unknown in literature to the best of
our knowledge and could not be prepared via the dialkylation
strategy described above. It was obtained via an alternative,
seven-step procedure that is fully described in the Supporting
Information.
With a highly similar AlogP, the achiral 4,4-dimethyl substi-
tuted pyrazoline analogue 42 showed a significant improvement
in human metabolic stability compared to the 4-monoethyl
substituted lead 6, although rat metabolic stability only improved
slightly. Less optimal filling of the hypothesized hydrophobic
binding pocket (formed when D106 adopts the gauche+ con-
formation) is a likely explanation for its somewhat reduced
Synthesis of the required 4,4-disubstituted pyrazoline building
blocks required another strategy, since ring closure of α,
β-unsaturated aldehydes or ketones with hydrazine (as set out in
Scheme 1) limits the number of C4 substituents to one group R2.
7039
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Scheme 6. Synthesis of 4,4-Disubstituted Pyrazoline Analoguesa
a Reagents and conditions: (a) R2-X + R3-X or X-R2-Y-R3-X, KO-t-Bu, TBAB, DMSO (optional), 0 °C to room temp or DBU, DMF, 0 to 80 °C or
K2CO3, DMF, 65 °C [X = halogen]; (b) LiAlH4, Et2O or CPME, 0 °C to room temp or BH3 THF, THF, ꢀ10 to 60 °C; (c) NaClO, H2O2, MeOH/
3
H2O, 0 °C to room temp; (d) 72, pyridine, reflux; (e) EtNH2, MeOH, room temp; (f) 72a, pyridine, reflux; (g) HCl, i-PrOH/EtOAc; (h) R8-SO2Cl,
DiPEA, DCM, room temp [R8 = Ph].
5-HT6R affinity. When going to the 4,4-diethyl substituted
analogue 43 (slight rise of AlogP to 2.95), some of the gain in
human metabolic stability (and all in rat) was lost but also the
affinity showed a drop rather than the anticipated gain. However,
by rigidification of its substituents into spiro five-membered ring
analogue 44 (AlogP = 2.58), apparently a more optimal fit in the
hydrophobic pocket was achieved given the clear improvement
in measured Ki and pA2 values. Human metabolic stability was
still better than that of lead 6, whereas rat metabolic stability was
already somewhat negatively affected. Growing the size of the
hydrophobic spiro substituent to a six-membered ring in 45
(significant rise of AlogP to 3.68) further improved 5-HT6R
affinity but again at the cost of metabolic stability, ending with
similar human and higher rat intrinsic clearance compared to lead
6. By introduction of an oxygen atom in the ring (46, AlogP =
1.92), a reduced metabolic rate was observed that was more
pronounced in human than in rat hepatocytes. The lowered
hydrophobicity of the spiro ring likely contributes to this but at
the same time results in a clear reduction in 5-HT6R affinity. The
heteroatom variation made in 47 (AlogP = 1.00, PSA = 94 Å2)
was primarily aimed at confirming our hypotheses and support-
ing the followed strategy. This analogue is the only one with a
basic residue, according to docking studies, well-posed for
interaction with D106 when changed into the trans conformation
(Figure 3D). This is the same interaction observed for “classical”
5-HT6R ligands containing a positive charge, such as 50
(Figure 3C). As expected 47 shows favorable receptor binding
and functional activity data, although the gain compared to 45 is
marginal. Whereas a salt bridge with 47 should give a stronger
enthalpy gain than the hydrophobic interaction with 45, a
desolvation penalty for the more polar 47 will likely in part
counterbalance this difference. Again, it is seen that the presence
of a heteroatom in the spiro ring is clearly beneficial for metabolic
stability, in this case also with a marked improvement in rat.
Although the foregoing makes 47 appear an ideal optimized
lead, off-target selectivity was a concern. Profiling in a panel of
118 receptor, ion channel, and transporter assays and 42 enzyme
assays at Cerep60 showed, contrary to nonbasic lead structure 6,
some affinity toward other monoaminergic GPCRs. Nonethe-
less, with pKi between 5.0 and 6.1 measured in the 5-HT2A,
5-HT2B, 5-HT3, M1, M3, M4, M5, and nonselective σ receptor
binding assays, a relatively large window to 5-HT6R affinity still
exists (see Supporting Information for full data). Membrane
passage of 47 was, with 14.7%, lower than that of 6 but not
critical. However, a clear increase in P-gp affinity was seen (P-gp
factor of 18.9). Even more important, screening of 47 on hERG
confirmed that a basic nitrogen in this class contributes to affinity
for this antitarget. Although the hERG affinity was relatively low
(pIC50 of 5.1 (Zenas)59), the low plasma protein binding of
this compound (66% in human and 59% in rat (Cyprotex)75)
likely accounted for sufficient peripheral exposure of free 47 to
cause prolongation of the QT interval of 25 ms at a dose of
50 mg/kg per os (po) in our telemetered conscious freely
moving guinea pig QTc model.76 No significant QTc effect was
seen at a lower dose of 10 mg/kg po.
As such we reverted to compound 42, providing the best
compromise between 5-HT6R affinity and functional activity,
human and rat metabolic stability, CNS druglike properties, and
off-target selectivity. Two final modifications of this compound
were made to complete our SAR picture. With an apparent
tendency for metabolism in the alkyl groups on the pyrazoline C4
position, we investigated whether introduction of a fluoro
substituent on both methyl groups in 42 would lead to a lower
intrinsic clearance. Regretfully, analogue 48 (AlogP = 1.74) only
showed a marginal gain in both human and rat metabolic
stability, whereas receptor binding and functional activity were
seriously affected. Removal of the 3-chloro substituent in 42 to
get to analogue 49 (AlogP = 2.02), bearing the unsubstituted
7040
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
phenylsulfonyl residue as in 50, resulted in a clear improvement
in metabolic stability. However, in line with the proposed polar
interaction of the 3-chloro substituent with Q291, the 5-HT6R
affinity of 49 is significantly reduced compared to 42, rendering it
uninteresting from a pharmacological point of view.
druglike. The absence of a basic nitrogen, in all target compounds
1ꢀ49 except 47, contributed to exceptional selectivity toward
off-targets (like other monoaminergic GPCRs) and antitargets
(like hERG). This may be especially beneficial in envisaged
scenarios where 5-HT6R antagonists will be developed as adjunc-
tive therapies. Inthis class ofnonbasic N0-(arylsulfonyl)pyrazoline-
1-carboxamidines, the absence of a salt bridge interaction with the
receptor is compensated by a combination of two factors. First,
there is a favorable entropy contribution due to an internal
hydrogen bond that stabilizes the bioactive conformation. Second,
an enthalpy gain is possible due to hydrophobic interactions with a
pocket that becomes available when, in the absence of a proto-
nated basic interaction partner, D106 changes conformation to
interact with Y310 in the receptor. During the phases of lead
optimization described here, we came across an apparent trade-off
between potency and metabolic stability while varying different
parts of the structure. While we were sometimes successful in
improving human in vitro intrinsic clearance, it proved to be
particularly challenging to significantly improve rat metabolic
stability, especially when simultaneously trying to keep other
properties compliant with CNS druglike criteria. A compound
within this series featuring a well-suited overall balance was 42.
Being profiled as potent, selective, hERG-free, and CNS available,
it appears to be a suitable candidate for further in vivo pharmaco-
logical testing. On the basis of in vitro intrinsic clearance data
obtained from hepatocytes, predictions for the human PK profile
are more favorable than for rat.
Profiling of Compound 42. On the basis of the necessary
trade-off between potency and metabolic stability emerging from
the lead optimization efforts described, achiral compound 42 was
considered to be an appropriate frontrunner candidate. With a
molecular weight of 343, an AlogP of 2.69, a PSA of 82 Å2, one
H-bond donor, six H-bond acceptors, and five rotational bonds,
its calculated molecular properties are CNS druglike. Membrane
passage is excellent (41%), and it has no affinity for the P-gp
transporter (P-gp factor of 1.0). When profiled comprehensively
at Cerep60 in a panel of 86 receptor, ion channel, and transporter
assays and 27 enzyme assays, the only off-target affinities
measured were a pKi of 5.8 for the peripheral benzodiazepine
receptor and a pKi of 5.1 for the 5-HT2B receptor (see Support-
ing Information for full data). Significant CYP inhibition was not
observed, with pIC50 < 5 measured for all CYPs tested (1A2,
2C9, 2C19, 2D6, and 3A4). Most importantly, 42 is devoid of a
basic moiety and was shown to be free of hERG affinity (pIC50
<
5 with a PI of 15 at 10ꢀ5 M measured at Zenas59). Despite a high
free fraction in plasma (plasma protein binding 63% measured at
Cyprotex75), the compound was confirmed to be free of QTc
effects at doses up to 100 mg/kg sc in guinea pig.76 On the basis
of in vitro intrinsic clearance in hepatocytes, human metabolic
stability appears to be acceptable whereas it is suboptimal in rat.
The latter is reflected by the rat in vivo pharmacokinetic (PK)
results. Absorption was very fast (tmax = 0.5 h), likely aided by the
nonionizable character of the molecule and the high membrane
permeability. Oral bioavailability in rat was low (∼7%), which
was clearly confirmed to be due to a high first-pass effect in an
accelerated infusion study (significantly higher exposures in the
intravenous (iv) curve compared to the nearly parallel intraduo-
denal (id) and intravenoportal (ivp) curves; data not shown).
However, with intraperitoneal (ip) administration a rat bioavail-
ability of 45% could be achieved. Calculation of the clearance
after iv administration indicated it to be at the liver blood flow. A
plasma half-life of just under 2 h was determined both after iv and
po administration. With a CNS/plasma ratio of 1.0, the com-
pound was shown to be brain-penetrable, and the relatively high
in vitro fraction unbound in brain homogenate77 of 0.10 will
contribute to sufficient exposure of free drug in the brain.
Altogether, this rat PK/PD profile was considered good enough
to investigate 42 as a frontrunner from this unique structural class
in rat in vivo models predicted to be relevant for indications
associated with the 5-HT6R. A study to compare the potential of
42, 47, and 50 to improve scopolamine-induced deficits in rat
models for episodic memory is published elsewhere.78
’ EXPERIMENTAL SECTION
Chemistry. Unless stated otherwise, starting materials, reagents and
solvents were purchased as high-grade commercial products and were
used without further purification. Reactions sensitive to moisture and/or
oxygen were carried out under an inert atmosphere of anhydrous
nitrogen, with solvents dried by the Pure-Solv EN solvent purification
system (Innovative Technology, Inc.). Yields refer to isolated pure pro-
ducts and were not maximized.
Analytical thin-layer chromatography (TLC) was performed on
Merck precoated 60 F254 plates, and spots were visualized with UV
light (254 nm), I2, 10% phosphomolybdic acid solution in ethanol,
ninhydrin solution, permanganate solution, anisaldehyde solution, bro-
mocresol green solution, or 2,4-dinitrophenylhydrazine solution. Flash
chromatography was performed on glass column using silica gel 60
(0.040ꢀ0.063 mm, Merck) or on a B€uchi Sepacore automated flash
system equipped with a Supelco VersaFlash station employing Supelco
VersaPak silica cartridges. Column chromatography was performed
using silica gel 60 (0.063ꢀ0.200 mm, Merck).
1H and 13C nuclear magnetic resonance (NMR) spectra were
recorded using the standard methods provided by the manufacturer at
room temperature, unless indicated otherwise, on a Bruker Avance 400
instrument (1H, 400 MHz), a Varian UN400 instrument (1H, 400 MHz),
or a Varian VXR200 instrument (1H, 200 MHz) in the indicated
deuterated solvent (i.e., DMSO-d6 or CDCl3) with tetramethylsilane as
an internal standard. Chemical shifts (δ scale) are given in parts per
million (ppm) downfield from tetramethylsilane. Coupling constants (J)
are expressed in hertz (Hz). The following abbreviations are used to
describe peak patterns: s (singlet), d (doublet), dd (double doublet),
ddd (double double doublet), t (triplet), dt (double triplet), q (quartet),
dq (double quartet), m (multiplet) and br (broad).
’ CONCLUSIONS
A broad scope of medicinal chemistry tools have enabled us to
gain proper knowledge of molecular properties and targetꢀ
ligand interactions of a unique, nonbasic structural class of
5-HT6R antagonists identified by HTS hit 1 and explored with
hit-to-lead analogues 2ꢀ15. This has allowed us to efficiently
narrow the large diversity possible around the identified core
structure 58 and to design lead optimization programs around
distinct substituents in order to improve 5-HT6R potency and/or
metabolic stability while trying to keep the molecules CNS
Melting points were recorded on a B€uchi B-545 melting point
apparatus and are uncorrected.
Mass spectra were recorded on a Micromass QTOF-2 (quadrupole
time-of-flight) instrument with MassLynx application software for
7041
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Table 5a
Table 6a
step total time (min) flow (μL/min) solvent A (%) solvent B (%)
step total time (min) flow (μL/min) solvent A (%) solvent B (%)
0
1
2
3
4
0
2000
2000
2000
2000
2000
95
0
5
100
100
5
0
1
2
3
4
5
0.2
1600
1600
1600
1600
1600
500
90
0
10
100
100
10
1.8
2.5
2.7
3.0
2.5
0
2.8
0
95
95
2.9
90
90
90
5
3.10
3.11
10
a A = 100% water with 0.1% HCOOH, pH ≈ 3. B = 100% acetonitrile
10
a A = 100% water with 0.2% HCOOH. B = 100% acetonitrile with
0.2% HCOOH.
with 0.1% HCOOH.
acquisition and reconstruction of the data. Exact mass measurement by
HRMS was done of the quasimolecular ion [M + H]+.
Optical rotations ([α]D) were measured on an Optical Activity
polarimeter. Specific rotations are given as deg/dm, and the concentra-
tions are reported as g/100 mL of the specified solvent and were recorded
at 23 °C.
(85%) of 1 as a colorless oil. 1H NMR (400 MHz, CDCl3) δ 0.96 (t, J = 7.5
Hz, 3H), 1.16 (t, J = 7.3 Hz, 3H), 1.41ꢀ1.70 (m, 2H), 3.03ꢀ3.17 (m, 1H),
3.44ꢀ3.58 (m, 2H), 3.71 (dd, J = 11.4 and 7.5 Hz, 1H), 4.11 (t, J = 11.2 Hz,
1H), 6.90 (br s, 1H), 6.93 (d, J= 1.5 Hz, 1H), 7.30ꢀ7.40 (m, 2H), 7.45 (dd,
J = 7.8 and 1.6 Hz, 1H), 8.15 (dd, J = 7.6 and 2.0 Hz, 1H).
N0-(2-Chlorophenylsulfonyl)-N-benzyl-4-ethyl-4,5-dihy-
dro-1H-pyrazole-1-carboxamidine (2). Following the same
procedure as described for 1, reaction of 63aa (700 mg, 2.02 mmol)
with benzylamine (3 mL, excess) and purification by flash chroma-
tography (Et2O) gave 700 mg (85%) of 2 as a colorless oil. 1H NMR
(400 MHz, CDCl3) δ 0.96 (t, J = 7.4 Hz, 3H), 1.45ꢀ1.68 (m, 2H),
3.07ꢀ3.20 (m, 1H), 3.79 (dd, J = 10.7 and 7.7 Hz, 1H), 4.21 (t, J =
10.7 Hz, 1H), 4.64 (d, J = 5.5 Hz, 2H), 6.93 (d, J = 1.6 Hz, 1H), 7.07
(br s, 1H), 7.20ꢀ7.31 (m, 5H), 7.33 (dd, J = 7.4 and 1.3 Hz, 1H), 7.39
(td, J = 7.6 and 1.7 Hz, 1H), 7.46 (dd, J = 7.8 and 1.4 Hz, 1H), 8.14 (dd,
J = 7.9 and 1.8 Hz, 1H).
N0-(2-Chlorophenylsulfonyl)-N-phenyl-4-ethyl-4,5-dihy-
dro-1H-pyrazole-1-carboxamidine (3). 63aa (1.00 g, 2.89 mmol)
and aniline (0.26 mL, 2.89 mmol) were dissolved in 10 mL of dry
tetrahydrofuran (THF), and the solution was cooled in an ice bath.
Subsequently, 3.18 mL of a 1 M solution of NaHMDS in THF was added
dropwise, and stirring was continued for 20 min at room temperature.
The mixture was concentrated under reduced pressure. Water was
added, and the mixture was extracted twice with EtOAc. The combined
organic layers were dried over Na2SO4, filtered, and concentrated under
reduced pressure. The crude product was purified by flash chromatog-
raphy (gradient Et2O/petroleum ether (PE) = 1:1 to Et2O). The residue
was triturated with diisopropyl ether, filtered off, and dried in vacuo to
give 180 mg (16%) of 3 as white crystals, mp 128ꢀ130 °C. 1H NMR
(400 MHz, CDCl3) δ 0.97 (t, J = 7.4 Hz, 3H), 1.45ꢀ1.69 (m, 3H),
3.07ꢀ3.19 (m, 1H), 3.82 (dd, J = 11.8 and 7.2 Hz, 1H), 4.21 (t, J = 11.3
Hz, 1H), 6.87 (d, J = 1.6 Hz, 1H), 7.01 (d, J = 7.7 Hz, 2H), 7.05ꢀ7.11
(m, 1H), 7.19 (t, J = 7.7 Hz, 2H), 7.31 (td, J = 7.8, 7.3, and 1.5 Hz, 1H),
7.40 (td, J = 7.6 and 1.8 Hz, 1H), 7.45 (dd, J = 8.0 and 1.4 Hz, 1H), 8.12
(dd, J = 7.8 and 1.7 Hz, 1H), 8.68 (br s, 1H).
N0-(2-Chlorophenylsulfonyl)-N-isopropyl-4-ethyl-4,5-di-
hydro-1H-pyrazole-1-carboxamidine (4). Following the same
procedure as described for 1, reaction of 63aa (700 mg, 2.02 mmol) with
isopropylamine (3 mL, excess) and purification by flash chromatography
(Et2O) gave 620 mg (86%) of 4 as a pale yellow oil. 1H NMR (400 MHz,
CDCl3) δ 0.95 (t, J = 7.5 Hz, 3H), 1.11 (d, J = 6.5 Hz, 3H), 1.12 (d, J =
6.5 Hz, 3H), 1.43ꢀ1.68 (m, 2H), 3.09 (m, 1H), 3.75 (dd, J = 11.4 and
7.5 Hz, 1H), 4.19 (m, 2H), 5.23 (br s, 1H), 6.92 (d, J = 1.5 Hz, 1H), 7.33
(td, J = 7.7 and 1.8 Hz, 1H), 7.38 (td, J = 7.6 and 1.8 Hz, 1H), 7.47 (dd,
J = 7.7 and 1.8 Hz,1H), 8.15 (dd, J = 7.7 and 1.8 Hz, 1H).
N0-(Phenylsulfonyl)-N,4-diethyl-4,5-dihydro-1H-pyrazole-
1-carboxamidine (5). Following the same procedure as described for
1, reaction of 63ab (2.99 g crude, max 8.63 mmol) with ethylamine
(5 mL of a 70% solution in water, excess) and purification by flash
chromatography (Et2O) followed by crystallization from methyl tert-butyl
Spectroscopic data of all described compounds were consistent with
the proposed structures. The purity of the target compounds 1ꢀ49 was
established as g95%, based on liquid chromatographyꢀmass spectro-
metry (LCꢀMS) analysis (average from ultraviolet (UV) detection and
evaporative light scattering detector (ELSD)) by one of the two systems
described below, except for 35 (94%), 36 (92%), and 38 (94%).
The first LCꢀMS system consisted of two Perkin-Elmer series 200
micropumps connected to each other by a 50 μL tee mixer, with a Gilson
215 autosampler connected. The autosampler had a 2 μL injection loop
and was connected to a Waters Atlantis C18 30 mm ꢁ 4.6 mm column
with 3 μm particles, thermostated in a Perkin-Elmer series 200 column
oven at 40 °C. The column was connected to a Perkin-Elmer series 200
UV meter with a 2.7 μL flow cell. The wavelength was set to 254 nm. The
UV meter was connected to a Sciex API 150EX mass spectrometer with
the following parameters: scan range, 150ꢀ900 amu; polarity, positive;
scan mode, profile; resolution Q1, UNIT; step size, 0.10 amu; time per
scan, 0.500 s; NEB, 10; CUR, 10; IS, 5200; TEM, 325; DF, 30; FP, 225;
EP, 10. The evaporative light scattering detector was a Sedere Sedex 55
operating at 50 °C and 3 bar of N2, connected to the Sciex API 150. The
complete system was controlled by a G3 Power Mac. Table 5 shows the
gradient elution method used as a standard.
The second LCꢀMS system consisted of Waters 1525μ pumps,
connected to a Waters 2777 autosampler. The autosampler had a 10 μL
injection loop, and the injection volume was 10 μL. The autosampler
was connected to a Waters Sunfire C18 30 mm ꢁ 4.6 mm column with
2.5 μm particles, thermostated at 23 °C. The column was connected to a
Waters 2996 PDA, and the wavelength was scanned from 240 to 320 nm
with 1.2 nm resolution and 20 Hz sampling rate. After the PDA the flow
was split 1:1 and connected to a Waters 2424 ELSD and a Waters ZQ
mass detector. The ELSD operated with the following parameters: gas
pressure, 40 psi; data rate, 20 points/s; gain, 500; time constant, 0.2 s;
nebulizer mode, cooling; drift tube, 50 °C. The mass spectrometer
operated with the following parameters: scan range, 117ꢀ900 amu;
polarity, positive; data format, centroid; time per scan, 0.500 s; interscan
time, 0.05 s; capillary, 2.5 kV; cone, 25 V; extractor, 2 V; RF lens, 0.5 V;
source temp, 125 °C; desolvation temp, 400 °C; cone gas, 100 L/h;
desolvation gas, 800 L/h; LM 1 resolution, 15; HM 1 resolution, 15; ion
energy, 0.5; multiplier, 500 V. The complete system was controlled by
Masslynx 4.1. Table 6 shows the gradient elution method used as a standard.
Syntheses. N0-(2-Chlorophenylsulfonyl)-N,4-diethyl-4,5-di-
hydro-1H-pyrazole-1-carboxamidine (1). 63aa (1.30 g, 3.76 mmol)
was dissolved in 10 mL of MeOH. A 70% solution of ethylamine in water
(5 mL, excess) was added, and the mixture was stirred for 1 h at room
temperature. The mixture was concentrated under reduced pressure and the
crude product was purified by flash chromatography (Et2O) to give 1.09 g
7042
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
ether (MtBE) gave 1.66 g (53% over two steps) of 5 as white crystals, mp
43ꢀ45 °C. 1H NMR (400 MHz, CDCl3) δ 0.94 (t, J = 7.4 Hz, 3H), 1.12
(t, J = 7.3 Hz, 3H), 1.41ꢀ1.67 (m, 2H), 3.00ꢀ3.14 (m, 1H), 3.40ꢀ3.51
(m, 2H), 3.68 (dd, J = 11.4 and 7.4 Hz, 1H), 4.07 (t, J = 11.7 Hz, 1H), 6.86
(br s, 1H), 6.89 (d, J = 1.6 Hz, 1H), 7.35 (br s, 1H), 7.39 (t, J = 7.8 Hz,
1H), 7.38ꢀ7.48 (m, 3H), 7.89ꢀ7.95 (m, 2H).
1H), 3.42ꢀ3.51 (m, 2H), 3.71 (dd, J = 11.4 and 7.4 Hz, 1H), 4.01 (t, J =
10.0 Hz, 1H), 6.82 (br s, 1H), 7.02 (t, J = 1.7 Hz, 1H), 7.38 (t, J = 7.9 Hz,
1H), 7.43 (ddd, J = 8.0, 1.9, and 1.2 Hz, 1H), 7.81 (ddd, J = 7.6, 1.5, and
1.4 Hz, 1H), 7.92 (t, J = 1.7 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-ethyl-3-methyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (12). Following the same procedure
as described for 1, reaction of 63dc (590 mg crude, max 1.78 mmol) with
ethylamine (1 mL of a 70% solution in water, excess) and purification by
flash chromatography (DCM/acetone = 98:2 to 95:5) gave 340 mg (22%
over two steps) of 12 as a colorless oil. 1H NMR (400 MHz, CDCl3) δ1.14
(t, J = 7.3 Hz, 3H), 2.04 (s, 3H), 2.82 (t, J = 9.8 Hz, 2H), 3.39ꢀ3.48 (m,
2H), 4.08 (t, J = 9.8 Hz, 2H), 6.65 (br s, 1H), 7.36 (t, J = 7.8 Hz, 1H),
7.40ꢀ7.44 (m, 1H), 7.79ꢀ7.83 (m, 1H), 7.92 (t, J = 1.8 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N,N-dimethyl-4-ethyl-4,5-di-
hydro-1H-pyrazole-1-carboxamidine (13). Following the same
procedure as described for 1, reaction of 63aa (700 mg, 2.02 mmol) with
dimethylamine (excess, bubbled through the reaction mixture) and
purification by crystallization from MtBE gave 270 mg (39%) of 13 as
white crystals, mp 94ꢀ95 °C. 1H NMR (400 MHz, CDCl3) δ 0.92 (t, J =
7.5 Hz, 3H), 1.41ꢀ1.63 (m, 2H), 2.93ꢀ3.06 (m, 1H), 3.15 (s, 6H), 3.52
(dd, J = 11.4 and 7.0 Hz, 1H), 3.90 (t, J = 11.0 Hz, 1H), 6.95 (d, J = 1.5
Hz, 1H), 7.30ꢀ7.40 (m, 2H), 7.45 (dd, J = 7.8 and 1.6 Hz, 1H), 8.18 (dd,
J = 7.6 and 2.0 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-ethyl-5-phenyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (14). Following the same procedure
as described for 1, reaction of 63ec (1.00 g, 2.54 mmol) with ethylamine
(3 mL of a 70% solution in water, excess) and purification by crystallization
from EtOH/water = 4:1 followed by drying in vacuo gave 700 mg (71%) of
14 as white crystals, mp 124ꢀ126 °C. 1H NMR (400 MHz, CDCl3) δ 1.21
(t, J = 7.1 Hz, 3H), 2.73 (ddd, J = 18.7, 6.6, and 1.8 Hz, 1H), 3.37 (ddd, J =
18.7, 11.8, and 1.7 Hz, 1H), 3.56ꢀ3.72 (m, 2H), 5.44 (dd, J = 11.8 and 6.6
Hz, 1H), 6.98 (br s, 1H), 6.99 (d, J = 1.5 Hz, 1H), 7.07 (t, J = 8.1 Hz, 1H),
7.19 (dd, J = 4.8 and 1.8 Hz, 1H), 7.28 (ddd, J = 7.9, 2.0, and 1.2 Hz, 1H),
7.39 (t, J = 1.8 Hz, 1H).
N0-(3-Chlorobenzoyl)-N,4-diethyl-4,5-dihydro-1H-pyrazole-
1-carboxamidine (15). 67a (200 mg, 0.98 mmol) was suspended in
5 mL of DCM. Diisopropylethylamine (0.36 mL, 2.06 mmol) and
3-chlorobenzoyl chloride (154 mg, 0.88 mmol) were added, and the
mixture was stirred overnight at room temperature. The mixture was
washed with 5% aqueous NaHCO3 followed by 2 M NaOH and the
organic layer was dried over Na2SO4, filtered, and concentrated under
reduced pressure to give 200 mg (74%) of 15 as a yellow oil. 1H NMR
(400 MHz, CDCl3) δ 0.99 (t, J = 7.5 Hz, 3H), 1.24 (t, J = 7.2 Hz, 3H),
1.49ꢀ1.71 (m, 2H), 3.08ꢀ3.24 (m, 1H), 3.32ꢀ3.76 (m, 3H), 3.96ꢀ4.21
(m, 1H), 6.37 (br s, 1H), 6.93 (d, J = 1.5 Hz, 1H), 7.32 (t, J = 7.9 Hz, 1H),
7.38ꢀ7.42 (m, 1H), 8.03ꢀ8.06 (m, 1H), 8.16 (t, J = 1.8 Hz, 1H).
Compounds 16 and 17 from Chiral Separation of 14. The
enantiomers of 14 were separated by chiral preparative HPLC. The
chiral preparative HPLC system consisted of a 250 mm ꢁ 80 mm axial
compression column packed with CHIRALCELOD as stationary phase
and heptane/ethanol (80:20) as mobile phase. The flow rate used was
240 mL/min, and detection was done by using UV set at a wavelength of
230 nm. The racemate was injected as a solution of 4.5% m/v in
heptane:/ethanol (40:60).
N0-(3-Chlorophenylsulfonyl)-N,4-diethyl-4,5-dihydro-1H-
pyrazole-1-carboxamidine (6). Following the same procedure as
described for 1, reaction of 63ac (700 mg, 2.02 mmol) with ethylamine
(3 mL of a 70% solution in water, excess) gave 670 mg (97%) of 6 as a
pale yellow oil. 1H NMR (400 MHz, CDCl3) δ 0.95 (t, J = 7.4 Hz, 3H),
1.14 (t, J = 7.2 Hz, 3H), 1.43ꢀ1.67 (m, 2H), 3.04ꢀ3.17 (m, 1H),
3.40ꢀ3.51 (m, 2H), 3.68 (dd, J = 11.5 and 7.4 Hz, 1H), 4.08 (t, J = 11.3
Hz, 1H), 6.79 (br s, 1H), 6.92 (d, J = 1.6 Hz, 1H), 7.37 (t, J = 7.9 Hz,
1H), 7.42 (ddd, J = 8.0, 2.0, and 1.3 Hz, 1H), 7.80 (dt, J = 7.6 and 1.4 Hz,
1H), 7.91 (t, J = 1.8 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-propyl-4-ethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (7). Following the same procedure
as described for 1, reaction of 63ac (423 mg, 1.22 mmol) with pro-
pylamine (1.0 mL, 12.20 mmol) and purification by flash chromatog-
raphy (dichloromethane (DCM)) gave 430 mg (99%) of 7 as a colorless
oil. 1H NMR (400 MHz, CDCl3) δ 0.90 (t, J = 7.4 Hz, 3H), 0.98 (t, J =
7.4 Hz, 3H), 1.47ꢀ1.69 (m, 4H), 3.07ꢀ3.19 (m, 1H), 3.39 (q, J = 6.4
Hz, 2H), 3.72 (dd, J = 11.1 and 7.3 Hz, 1H), 4.12 (t, J = 11.1 Hz, 1H),
6.85 (br s, 1H), 6.94 (d, J = 1.5 Hz, 1H), 7.39 (t, J = 7.9 Hz, 1H),
7.43ꢀ7.46 (m, 1H), 7.81ꢀ7.85 (m, 1H), 7.94 (t, J = 1.8 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-methyl-4-ethyl-4,5-dihydro-1
H-pyrazole-1-carboxamidine (8). Following the same procedure
as described for 1, reaction of 63ac (500 mg, 1.45 mmol) with
methylamine hydrochloride (976 mg, 14.45 mmol) in the presence of
triethylamine (2.4 mL, 17.35 mmol) and purification by flash chroma-
tography (DCM) gave 470 mg (99%) of 8 as a colorless oil. 1H NMR
(400 MHz, CDCl3) δ 0.97 (t, J = 7.7 Hz, 3H), 1.47ꢀ1.69 (m, 2H), 3.04
(d, J = 5.1 Hz, 1H), 3.08ꢀ3.19 (m, 1H), 3.71 (dd, J = 11.1 and 7.3 Hz,
1H), 4.11 (t, J = 11.1 Hz, 1H), 6.88 (br s, 1H), 6.94 (d, J = 1.1 Hz, 1H),
7.39 (t, J = 7.9 Hz, 1H), 7.43ꢀ7.47 (m, 1H), 7.81ꢀ7.86 (m, 1H), 7.95
(t, J = 1.8 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-4-ethyl-4,5-dihydro-1H-pyra-
zole-1-carboxamidine (9). Following the same procedure as de-
scribed for 1, reaction of 63ac (300 mg, 0.87 mmol) with ammonia
(excess, bubbled through the reaction mixture) and purification by tri-
turation with Et2O followed by filtration and drying in vacuo gave 220 mg
(81%) of 9 as white crystals, mp 140.5ꢀ141.7 °C. 1H NMR (400 MHz,
CDCl3) δ 0.97 (t, J = 7.7 Hz, 3H), 1.45ꢀ1.68 (m, 2H), 3.13ꢀ3.18 (m,
1H), 3.54 (dd, J = 11.4 and 7.4 Hz, 1H), 3.93 (t, J = 11.7 Hz, 1H), 6.23 (br
s, 1H), 6.94 (d, J = 1.6 Hz, 1H), 7.35 (br s, 1H), 7.39 (t, J = 7.8 Hz, 1H),
7.45ꢀ7.48 (m, 1H), 7.80ꢀ7.85 (m, 1H), 7.92 (t, J = 1.8 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-ethyl-4-methyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (10). Following the same procedure
as described for 1, reaction of 63bc (500 mg crude, max 1.50 mmol) with
ethylamine (1 mL of a 70% solution in water, excess) and purification by
flash chromatography (gradient Et2O/PE = 1:1 to Et2O) gave 410 mg
(48% over two steps) of 10 as a colorless oil. 1H NMR (400 MHz, CDCl3)
δ 1.16 (t, J = 7.0 Hz, 3H), 1.20 (d, J = 7.0 Hz, 3H), 3.16ꢀ3.30 (m, 1H),
3.41ꢀ3.52 (m, 2H), 3.62 (dd, J = 11.3 and 7.5 Hz, 1H), 4.17 (t, J = 11.3 Hz,
1H), 6.80 (br s, 1H), 6.89 (d, J = 1.5 Hz, 1H), 7.38 (dd, J = 7.7 and 7.6 Hz,
1H), 7.43 (ddd, J = 8.0, 2.0, and 1.2 Hz, 1H), 7.82 (ddd, J = 7.6, 1.8, and 1.3
Hz, 1H), 7.92 (t, J = 1.8 Hz, 1H).
The first eluting enantiomer 16 was isolated in 89% yield with an ee of
>99; [α]D ꢀ37 (c 1, CHCl3). 1H NMR (400 MHz, CDCl3) data were
identical to the spectrum of the racemate (14).
The second eluting enantiomer 17 was isolated in 89% yield with an
ee of >99; [α]D +38 (c 1, CHCl3). 1H NMR (400 MHz, CDCl3) data
were identical to the spectrum of the racemate (14).
For both enantiomers, the amorphous materials obtained from
evaporation of the chiral preparative HPLC mobile phase were crystal-
lized from EtOH. Crystals of the (+)-enantiomer 17 were of suitable
quality for determination of the absolute configuration by X-ray.
N0-(3-Chlorophenylsulfonyl)-N-ethyl-4,5-dihydro-1H-pyr-
azole-1-carboxamidine (11). Following the same procedure as
described for 1, reaction of 63cc (500 mg, 1.57 mmol) with ethylamine
(1 mL of a 70% solution in water, excess) and purification by flash
chromatography (Et2O) gave 400 mg (81%) of 11 as a colorless oil. 1H
NMR (400 MHz, CDCl3) δ 1.16 (t, J = 7.2 Hz, 3H), 2.89 (t, J = 9.6 Hz,
7043
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Selected Crystallographic Data for 17. X-ray data were collected at the
Bijvoet Centre for Biomolecular Research, Utrecht University, The
Netherlands, with a Nonius KappaCCD diffractometer on a rotating
anode using Mo Kα radiation at 150 K. The structure was solved with
the program SHELXS97 (G. M. Sheldrick) and refined with SHELXL97
(G. M. Sheldrick). The program PLATON79 was used for the analysis of
the geometry, the illustrations, and the validation of the results. Crystal
system, monoclinic; space group, P21; unit cell dimensions, a = 8.8409 Å,
b = 11.2526 Å, c = 9.25158 Å; overall R-factor, 0.0254. The CIF has been
deposited at the Cambridge Crystallographic Data Centre (CCDC),
deposition number 817440.
N0-(3-Chlorophenylsulfonyl)-N-ethyl-5-(pyridin-4-yl)-4,5-di-
hydro-1H-pyrazole-1-carboxamidine (18). Following the same
procedure as described for 1, 63fc (1.76 g crude, max 4.4 mmol) was
reacted with ethylamine (2 mL of a 70% solution in water, excess). The
crude product was purified by flash chromatography (DCM, DCM/
MeOH = 99:1 to 96:4). Subsequently, the evaporation residue from the
combined product fractions was dissolved in MeOH and brought onto a
conditioned SCX ion exchange column, the column was washed with
MeOH, and the product was eluted with 1 M NH3 in MeOH. Evaporation
yielded 680 mg (21% over two steps) of 18 as a beige solid. 1H NMR (400
MHz, CDCl3) δ1.27(t,J=7Hz, 3H), 2.62ꢀ2.72(m, 1H), 3.33ꢀ3.45 (m,
1H), 3.60ꢀ3.79 (m, 2H), 5.34ꢀ5.45 (m, 1H), 6.88 (d, J = 6 Hz, 2H),
7.01(br s, 1H), 7.10ꢀ7.22 (m, 2H), 7.35 (d, J = 8 Hz, 1H), 7.45 (br s, 1H),
8.40 (d, J = 6 Hz, 2H).
pressure and the crude product was purified by flash chromatography
(EtOAc) to give 88 mg (22%) of 21 as a colorless oil. 1H NMR (400 MHz,
CDCl3) δ 0.97 (t, J = 7.5 Hz, 3H), 1.16 (t, J = 7.1 Hz, 3H), 1.46ꢀ1.69 (m,
2H), 3.07ꢀ3.19(m, 1H), 3.43ꢀ3.59(m, 2H), 3.67(dd, J=11.4and7.5Hz,
1H), 4.06 (t, J = 11.2 Hz, 1H), 6.97 (br s, 1H), 6.94 (d, J = 1.8 Hz, 1H),
7.52 (d, J = 8.4 Hz, 1H), 7.76 (dd, J = 8.4 and 2.0 Hz, 1H), 8.03 (d, J = 2.0
Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-methoxy-4-ethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (22). 63ac (500 mg, 1.45 mmol) was
dissolved in 10 mL of MeOH. O-Methylhydroxylamine hydrochloride (1.21 g,
14.5 mmol) and triethylamine (2.43 mL, 17.4 mmol) were added, and the
mixture was stirred overnight at 50 °C. The mixture was concentrated under
reduced pressure and the residue was purified by flash chromatography
(EtOAc/PE = 1:9 to 1:1) to give 160 mg (32%) of 22 as a pale yellow oil. 1H
NMR (400 MHz, CDCl3) δ 0.98 (t, J = 7.5 Hz, 3H), 1.47ꢀ1.70 (m, 2H),
3.08ꢀ3.18 (m, 1H), 3.54 (dd, J = 11.5 and 7.2 Hz, 1H), 3.81 (s, 3H), 3.94 (t,
J= 11.5 Hz, 1H), 7.00 (d, J= 1.4 Hz, 1H), 7.40 (t, J=7.8Hz,1H),7.44ꢀ7.48
(m, 1 H), 7.82ꢀ7.86 (m, 1H), 7.95 (t, J = 1.8 Hz, 1H), 9.16 (br s, 1H).
N0-(4-Chlorophenylsulfonyl)-N,4-diethyl-4,5-dihydro-1H-
pyrazole-1-carboxamidine (23). Following the same procedure as
described for 1, reaction of 63ad (750 mg, 2.17 mmol) with ethylamine
(3 mL of a 70% solution in water, excess) gave 690 mg (93%) of 23 as
pale yellow crystals, mp 84ꢀ86 °C. 1H NMR (400 MHz, CDCl3) δ 0.97
(t, J = 7.5 Hz, 3H), 1.15 (t, J = 7.1 Hz, 3H), 1.46ꢀ1.69 (m, 2H), 3.12
(dd, J = 10.9 and 7.1 Hz, 1H), 3.43ꢀ3.50 (m, 2H), 3.70 (dd, J = 11.4 and
7.5 Hz, 1H), 4.10 (t, J = 11.3 Hz, 1H), 6.81 (br s, 1H), 6.93 (d, J = 1.8
Hz, 1H), 7.42 (m, 2H), 7.88 (m, 2H).
N0-(3-Fluorophenylsulfonyl)-N,4-diethyl-4,5-dihydro-1H-
pyrazole-1-carboxamidine (24). Following the same procedure
as described for 1, reaction of 63ae (580 mg, 1.52 mmol) with
ethylamine (3 mL of a 70% solution in water, excess) gave 570 mg
(99%) of 24 as a yellow oil. 1H NMR (400 MHz, CDCl3) δ 0.96 (t, J =
7.7 Hz, 3H), 1.14 (t, J = 7.1 Hz, 3H), 1.45ꢀ1.68 (m, 2H), 3.04ꢀ3.18
(m, 1H), 3.41ꢀ3.51 (m, 2H), 3.70 (dd, J = 11.0 and 7.5 Hz, 1H), 4.10
(t, J = 11.3 Hz, 1H), 6.80 (br s, 1H), 6.92 (d, J = 1.6 Hz, 1H), 7.16 (tdd,
J = 8.4, 2.5, and 0.9 Hz, 1H), 7.42 (dt, J = 8.1 and 5.4 Hz, 1H), 7.63 (ddd,
J = 8.6, 2.4, and 1.5 Hz, 1H), 7.72 (ddd, J = 7.7, 2.4, and 0.9 Hz, 1H).
N0-(3-(Trifluoromethyl)phenylsulfonyl)-N,4-diethyl-4,5-
dihydro-1H-pyrazole-1-carboxamidine (25). Following the
same procedure as described for 1, reaction of 63af (500 mg, 1.32
mmol) with ethylamine (1 mL of a 70% solution in water, excess)
gave 470 mg (95%) of 25 as a yellow oil. 1H NMR (400 MHz, CDCl3)
δ 0.96 (t, J = 7.7 Hz, 3H), 1.14 (t, J = 7.7 Hz, 3H), 1.45ꢀ1.68 (m, 2H),
3.07ꢀ3.18 (m, 1H), 3.40ꢀ3.51 (m, 2H), 3.71 (dd, J = 11.4 and 7.4
Hz, 1H), 4.10 (t, J = 11.4 Hz, 1H), 6.77 (br s, 1H), 6.93 (d, J = 1.6 Hz,
1H), 7.58 (t, J = 7.9 Hz, 1H), 7.72 (d, J = 8.0 Hz, 1H), 8.12 (d, J = 7.9
Hz, 1H), 8.22 (s, 1H).
N0-(3-Chlorophenylsulfonyl)-N-ethyl-5-(furan-2-yl)-4,5-
dihydro-1H-pyrazole-1-carboxamidine (19). Following the
same procedure as described for 1, 63gc (1.30 g crude, max 3.39
mmol) was reacted with ethylamine (2 mL of a 70% solution in water,
excess). The mixture was concentrated under reduced pressure, 5%
aqueous NaHCO3 was added, and the mixture was extracted twice
with DCM. The combined organic layers were dried over Na2SO4,
filtered, and concentrated under reduced pressure. The crude
product was purified by flash chromatography (DCM, DCM/acet-
one = 99:1) and subsequently by reverse phase preparative HPLC to
give 420 mg (17% over three steps) of 19 as an orange amorphous
solid. 1H NMR (400 MHz, CDCl3) δ 1.16 (t, J = 7.1 Hz, 3H), 3.00
(ddd, J = 18.6, 6.4, and 1.7 Hz, 1H), 3.22 (ddd, J = 18.6, 12.0, and 1.6
Hz, 1H), 3.49ꢀ3.64 (m, 2H), 5.60 (dd, J = 12.0 and 6.4 Hz, 1H), 6.01 (d,
J = 3.2 Hz, 1H), 6.17 (dd, J = 3.2 and 1.7 Hz, 1H), 7.08 (br s, 1H), 7.03 (t,
J = 1.6 Hz, 1H), 7.21 (d, J = 1.4 Hz, 1H), 7.29 (t, J = 7.9 Hz, 1H),
7.38ꢀ7.42 (m, 1H), 7.56ꢀ7.60 (m, 1H), 7.69 (t, J = 1.8 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-ethyl-5-(4-fluorophenyl)-4,5-
dihydro-1H-pyrazole-1-carboxamidine (20). Following the same
procedure as described for 1, 63hc (8.40 g crude, max 11.87 mmol) was
reacted with ethylamine (9.56 mL of a 70% solution in water, excess). The
mixture was concentrated under reduced pressure, 5% aqueous NaHCO3
was added, and the mixture was extracted twice with EtOAc. The combined
organic layers were dried over MgSO4, filtered, and concentrated under
reduced pressure. The crude product was purified by flash chromatography
(DCM, DCM/MeOH = 95:5) and subsequently by reverse phase pre-
parative HPLC. The obtained material was triturated with Et2O/PE = 1:2,
and the solids were filtered off and dried in vacuo at 40 °C to give 0.93 g
(13% over two steps) of 20 as a yellow solid, mp 139ꢀ140 °C. 1H NMR
(400 MHz, CDCl3) δ 1.22 (t, J = 7.2 Hz, 3H), 2.70 (dd, J = 6.8 and 1.5 Hz,
1H), 3.36 (dd, J = 11.9 and 1.5 Hz, 1H), 3.55ꢀ3.74 (m, 2H), 5.42 (dd, J =
11.9 and 6.8 Hz, 1H), 6.85 (t, J = 8.5 Hz, 2H), 6.91ꢀ6.99 (m, 2H), 7.00 (t,
J = 1.6 Hz, 1H), 7.14 (t, J = 7.7 Hz, 1H), 7.24 (br s, 1H), 7.31ꢀ7.36 (m,
1H), 7.36ꢀ7.41 (m, 1H).
General Procedure for Parallel Synthesis of N0-(Arylsulfonyl)-
N,4-diethyl-4,5-dihydro-1H-pyrazole-1-carboxamidines for
Compounds 26, 27, 29ꢀ32, 37, and 39. 67a (205 mg, 1.0 mmol)
was taken up in DCM (5.0 mL), and N,N-diisopropylethylamine
(DiPEA) (0.366 mL, 2.1 mmol) was added. To this suspension, a stock
solution of the arylsulfonyl chloride (0.9 mmol in 5.0 mL of DCM) was
added, and the mixture was agitated overnight at 30 °C. The mixture was
extracted against a 2 M aqueous NaOH solution (10 mL), and the
organic phase was separated and evaporated to dryness. The residue was
taken up in 10 mL of MeOH and sampled onto a preconditioned SCX
column (5 g of IST Isolute SCX-3 material of ∼0.6 mequiv/g capacity,
prepacked in a 25 mL cartridge). The column was eluted with 15 mL of
MeOH, and the combined fractions collected during sampling and
elution were evaporated to dryness.
N0-(3-Chlorophenylsulfonyl)-N-(2,2,2-trifluoroethyl)-4-ethyl-
4,5-dihydro-1H-pyrazole-1-carboxamidine (21). 63ac (350 mg,
1.01 mmol) was dissolved in 20 mL of MeOH, and 2,2,2-trifluoroethyla-
mine (805 μL, 10.1 mmol) was added. The mixture was stirred overnight at
60 °C in a sealed tube. The mixture was concentrated under reduced
N0-(m-Tolylsulfonyl)-N,4-diethyl-4,5-dihydro-1H-pyrazole-1-
carboxamidine (26). By use of 3-methylbenzenesulfonyl chloride as
reagent in the general procedure described above, 214 mg (66%) of 26 was
7044
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
isolated. 1HNMR(400MHz, DMSO-d6) δ0.87 (t, J= 7.5 Hz, 3H), 0.96 (t,
J= 7.2 Hz, 3H), 1.37ꢀ1.60 (m, 2H), 2.37 (s, 3H), 3.16ꢀ3.30 (m, 3H), 3.55
(dd, J= 11.1 and 7.0 Hz, 1H), 3.96 (t, J= 10.9 Hz, 1H), 7.30ꢀ7.35 (m, 2H),
7.38 (t, J = 7.6 Hz, 1H), 7.59 (d, J = 7.6 Hz, 1H), 7.61 (s, 1H), 7.67
(br s, 1H).
N0-(3-Methoxyphenylsulfonyl)-N,4-diethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (27). By use of 3-methoxyben-
zenesulfonyl chloride as reagent in the general procedure described
above, 224 mg (66%) of 27 was isolated. 1H NMR (400 MHz, CDCl3)
δ 0.97 (t, J = 7.5 Hz, 3H), 1.15 (t, J = 7.1 Hz, 3H), 1.43ꢀ1.73 (m, 2H),
3.04ꢀ3.16 (m, 1H), 3.42ꢀ3.52 (m, 2H), 3.71 (dd, J = 11.4 and 7.5 Hz,
1H), 3.85 (s, 3H), 4.11 (t, J = 11.4 Hz, 1H), 6.85 (br s, 1H), 6.92 (d, J =
1.5 Hz, 1H), 6.99ꢀ7.03 (m, 1H), 7.35 (t, J = 7.9 Hz, 1H), 7.47ꢀ7.49
(m, 1H), 7.51ꢀ7.55 (m, 1H).
3.33ꢀ3.43 (m, 2H), 3.64 (dd, J = 11.4 and 7.4 Hz, 1H), 3.99 (t, J = 11.4
Hz, 1H), 6.88 (br s, 1H), 6.87 (d, J = 1.6 Hz, 1H), 7.49 (dd, J = 8.1 and
7.4 Hz, 1H), 7.54 (ddd, J = 8.1, 6.8, and 1.3 Hz, 1H), 7.61 (ddd, J = 8.4,
6.8, and 1.4 Hz, 1H), 7.89 (d, J = 8.1 Hz, 1H), 7.97 (d, J = 8.2 Hz, 1H),
8.27 (dd, J = 7.3 and 1.3 Hz, 1H), 8.88 (d, J = 8.5 Hz, 1H).
N0-(Naphthalen-2-ylsulfonyl)-N,4-diethyl-4,5-dihydro-1H-
pyrazole-1-carboxamidine (34). Following the same procedure as
described for 1, reaction of 63ai (400 mg, 1.11 mmol) with ethylamine
(1 mL of a 70% solution in water, excess) gave 397 mg (100%) of 34 as a
yellow oil. 1H NMR (400 MHz, CDCl3) δ 0.95 (t, J = 7.7 Hz, 3H), 1.14
(t, J = 7.7 Hz, 3H), 1.43ꢀ1.62 (m, 2H), 3.02ꢀ3.13 (m, 1H), 3.40ꢀ3.55
(m, 2H), 3.71 (dd, J = 11.2 and 7.3 Hz, 1H), 4.10 (t, J = 11.2 Hz, 1H),
6.90 (br s, 1H), 6.90 (d, J = 1.6 Hz, 1H), 7.53ꢀ7.60 (m, 2H), 7.84ꢀ7.98
(m, 4H), 8.47 (d, J = 1.4 Hz, 1H).
N0-(3,4-Dichlorophenylsulfonyl)-N,4-diethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (28). Following the same proce-
dure as described for 1, reaction of 63ag (500 mg, 1.31 mmol) with
ethylamine (1 mL of a 70% solution in water, excess) gave 500 mg
(100%) of 28 as a yellow oil. 1H NMR (400 MHz, CDCl3) δ 0.97 (t, J =
7.5 Hz, 3H), 1.16 (t, J = 7.1 Hz, 3H), 1.46ꢀ1.69 (m, 2H), 3.07ꢀ3.19 (m,
1H), 3.43ꢀ3.59 (m, 2H), 3.67 (dd, J = 11.4 and 7.5 Hz, 1H), 4.06 (t, J =
11.2 Hz, 1H), 6.97 (br s, 1H), 6.94 (d, J = 1.8 Hz, 1H), 7.52 (d, J = 8.4
Hz, 1H), 7.76 (dd, J = 8.4 and 2.0 Hz, 1H), 8.03 (d, J = 2.0 Hz, 1H).
N0-(3,5-Dichlorophenylsulfonyl)-N,4-diethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (29). By use of 3,5-dichloro-ben-
zenesulfonyl chloride as reagent in the general procedure described
above, 239 mg (63%) of 29 was isolated. 1H NMR (400 MHz, DMSO-
d6) δ 0.87 (t, J = 7.5 Hz, 3H), 0.97 (t, J = 7.2 Hz, 3H), 1.38ꢀ1.62 (m,
2H), 3.16ꢀ3.30 (m, 3H), 3.54 (dd, J = 11.1 and 7.0 Hz, 1H), 3.96 (t, J =
10.9 Hz, 1H), 7.37 (s, 1H), 7.74 (d, J = 1.6 Hz, 2H), 7.82 (t, J = 1.6 Hz,
1H), 7.96 (br s, 1H).
N0-(2,5-Dichlorophenylsulfonyl)-N,4-diethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (30). By use of 2,5-dichloroben-
zenesulfonyl chloride as reagent in the general procedure described
above, 191 mg (51%) of 30 was isolated. 1H NMR (400 MHz, DMSO-
d6) δ 0.87 (t, J = 7.5 Hz, 3H), 0.95 (t, J = 7.2 Hz, 3H), 1.38ꢀ1.61 (m,
2H), 3.17ꢀ3.30 (m, 3H), 3.52ꢀ3.60 (m, 1H), 3.98 (t, J = 10.9 Hz, 1H),
7.37 (s, 1H), 7.59ꢀ7.66 (m, 2H), 7.89 (br s, 1H), 7.95 (d, J = 1.9 Hz,
1H).
N0-(5-Chloro-3-methylbenzo[b]thiophen-2-ylsulfonyl)-
N,4-diethyl-4,5-dihydro-1H-pyrazole-1-carboxamidine (35).
Following the same procedure as described for 1, reaction of 63aj (500 mg,
1.20 mmol) with ethylamine (1 mL of a 70% solution in water, excess)
1
gave 410 mg (83%) of 35 as an off-white foam. H NMR (400 MHz,
CDCl3) δ 0.98 (t, J = 7.4 Hz, 3H), 1.14 (t, J = 7.7 Hz, 3H), 1.48ꢀ1.70
(m, 2H), 2.69 (s, 3H), 3.08ꢀ3.19 (m, 1H), 3.47ꢀ3.59 (m, 2H), 3.78
(dd, J = 11.2 and 7.6 Hz, 1H), 4.10 (t, J = 11.2 Hz, 1H), 6.84 (br s, 1H),
6.94 (d, J = 1.6 Hz, 1H), 7.37 (dd, J = 8.6 and 2.0 Hz, 1H), 7.71 (d, J =
8.6 Hz, 1H), 7.73 (d, J = 2.0 Hz, 1H).
N0-(6-Chloroimidazo[2,1-b]thiazol-5-ylsulfonyl)-N,4-diethyl-
4,5-dihydro-1H-pyrazole-1-carboxamidine (36). Following the
same procedure as described for 1, reaction of 63ak (380 mg, 0.97 mmol)
with ethylamine (1 mL of a 70% solution in water, excess) gave 390 mg
(100%) of 36 as a yellow oil. 1H NMR (400 MHz, CDCl3) δ0.96 (t, J = 7.5
Hz, 3H), 1.17 (t, J = 7.1 Hz, 3H), 1.46ꢀ1.69 (m, 2H), 3.04ꢀ3.16 (m, 1H),
3.48ꢀ3.57 (m, 2H), 3.70 (dd, J = 11.4 and 7.5 Hz, 1H), 4.10 (t, J = 11.3 Hz,
1H), 6.77 (br s, 1H), 6.96 (d, J = 1.5 Hz, 1H), 7.52 (d, J = 4.6 Hz, 1H), 7.98
(d, J = 4.6 Hz, 1H).
N0-(5-Chlorothiophen-2-ylsulfonyl)-N,4-diethyl-4,5-dihy-
dro-1H-pyrazole-1-carboxamidine (37). By the use of 5-chlor-
othiophene-2-sulfonyl chloride as reagent in the general procedure
described above and subsequent reverse phase preparative HPLC
1
purification, 82 mg (24%) of 37 was isolated. H NMR (400 MHz,
CDCl3) δ 0.99 (t, J = 7.5 Hz, 3H), 1.20 (t, J = 7.1 Hz, 3H), 1.48ꢀ1.71
(m, 2H), 3.08ꢀ3.21 (m, 1H), 3.46ꢀ3.60 (m, 2H), 3.74 (dd, J = 11.0
and 7.4 Hz, 1H), 4.14 (t, J = 11.0 Hz, 1H), 6.77 (br s, 1H), 6.83 (d, J =
3.9 Hz, 1H), 6.95 (d, J = 1.4 Hz, 1H), 7.35 (d, J = 3.9 Hz, 1H).
N0-(5-Chloro-1,3-dimethyl-1H-pyrazol-4-ylsulfonyl)-N,4-
diethyl-4,5-dihydro-1H-pyrazole-1-carboxamidine (38). 67a
(300 mg, 1.47 mmol) was suspended in 10 mL of DCM. Diisopropy-
lethylamine (0.53 mL, 3.08 mmol) and 5-chloro-1,3-dimethyl-1H-
pyrazole-4-sulfonyl chloride (275 mg, 1.32 mmol) were added, and
the mixture was stirred overnight at room temperature. Because of low
conversion, acetonitrile (10 mL) was added, the DCM was allowed to
evaporate, and the mixture was refluxed overnight. After concentration
under reduced pressure, DCM was added and the organic phase was
washed with 5% aqueous NaHCO3 followed by 2 M NaOH. The organic
layer was dried over Na2SO4, filtered, and concentrated under reduced
pressure. The crude product was purified by flash chromatography
(DCM/acetone = 98:2 to 95:5) to give 130 mg (27%) of 38 as an orange
oil. 1H NMR (400 MHz, CDCl3) δ 0.98 (t, J = 7.5 Hz, 3H), 1.17 (t, J =
7.2 Hz, 3H), 1.45ꢀ1.71 (m, 2H), 2.46 (s, 3H), 3.02ꢀ3.18 (m, 1H),
3.45ꢀ3.57 (m, 2H), 3.65ꢀ3.74 (m, 1H), 3.79 (s, 3H), 4.11 (t, 1H), 6.89
(br s, 1H), 6.92 (br s, 1H).
N0-(2,6-Dichlorophenylsulfonyl)-N,4-diethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (31). By use of 2,6-dichloroben-
zenesulfonyl chloride as reagent in the general procedure described
above, 155 mg (41%) of 31 was isolated. 1H NMR (400 MHz, CDCl3)
δ 0.97 (t, J = 7.4 Hz, 6H), 1.16 (t, J = 7.1 Hz, 3H), 1.46ꢀ1.68 (m,
2H), 3.05ꢀ3.16 (m, 1H), 3.49 (q, J = 7.2 Hz, 2H), 3.73 (dd, J =
11.6 and 7.4 Hz, 1H), 4.15 (t, J = 11.3 Hz, 1H), 6.83 (br s, 1H), 6.92
(d, J = 1.5 Hz, 1H), 7.21 (dd, J = 7.6 and 8.4 Hz, 1H), 7.40 (d, J =
8.0 Hz, 2H).
N0-(3-Chloro-2-fluorophenylsulfonyl)-N,4-diethyl-4,5-di-
hydro-1H-pyrazole-1-carboxamidine (32). By use of 3-chloro-
2-fluorobenzenesulfonyl chloride as reagent in the general procedure
1
described above, 199 mg (55%) of 32 was isolated. H NMR (400
MHz, DMSO-d6) δ 0.87 (t, J = 7.5 Hz, 3H), 0.95 (t, J = 7.2 Hz, 3H),
1.38ꢀ1.62 (m, 2H), 3.15ꢀ3.30 (m, 3H), 3.55 (dd, J = 11.1 and 7.0 Hz,
1H), 3.98 (t, J = 10.9 Hz, 1H), 7.33 (t, J = 7.6 Hz, 1H), 7.38 (s, 1H),
7.77 (t, J = 7.2 Hz, 1H), 7.95 (br t, J = 5.4 Hz, 1H).
N0-(Naphthalen-1-ylsulfonyl)-N,4-diethyl-4,5-dihydro-1H-
pyrazole-1-carboxamidine (33). Following the same procedure as
described for 1, reaction of 63ah (440 mg, 1.22 mmol) with ethylamine
(1 mL of a 70% solution in water, excess) and purification by flash
chromatography (DCM/acetone = 98:2) followed by crystallization
from cyclohexane/DCM gave 360 mg (83%) of 33 as white crystals, mp
79.0ꢀ80.0 °C. 1H NMR (400 MHz, CDCl3) δ 0.90 (t, J = 7.7 Hz, 3H),
1.05 (t, J = 7.7 Hz, 3H), 1.38ꢀ1.61 (m, 2H), 2.95ꢀ3.06 (m, 1H),
N0-(4-(Methylsulfonyl)phenylsulfonyl)-N,4-diethyl-4,5-di-
hydro-1H-pyrazole-1-carboxamidine (39). By use of 4-metha-
nesulfonylbenzenesulfonyl chloride as reagent in the general procedure
described above, 229 mg (59%) of 39 was isolated. 1H NMR (400 MHz,
DMSO-d6) δ 0.97 (t, J = 7.6 Hz, 3H), 1.16 (t, J = 7.2 Hz, 3H),
7045
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
1.38ꢀ1.62 (m, 2H), 3.19ꢀ3.26 (m, 3H), 3.27 (s, 3H), 3.54ꢀ3.60 (m,
1H), 3.98 (t, J = 10.8 Hz, 1H), 7.35 (s, 1H), 7.89 (br s, 1H), 8.02ꢀ8.09
(m, 4H).
Hz, 1H), 7.45ꢀ7.48 (m, 1H), 7.81ꢀ7.85 (m, 1H), 7.93 (t, J = 1.6
Hz, 1H).
N-[Ethylamino-(2,3,8-triaza-spiro[4.5]dec-3-en-2-yl)methyl-
idene]-3-chlorobenzenesulfonamide (47). Following the same
procedure as described for 1, reaction of 74f (10.20 g, 19.2 mmol) with
ethylamine (15.3 mL of a 70% solution in water, 0.19 mol) and purification
by flash chromatography (DCM/MeOH = 99:1 to 98:2) gave 6.92 g
(75%) of N-[ethylamino-(8-benzyl-2,3,8-triaza-spiro[4.5]dec-3-en-2-yl)-
methylidene]-3-chlorobenzenesulfonamide as a yellow amorphous com-
pound. 1H NMR (400 MHz, CDCl3) δ 1.16 (t, J = 7.2 Hz, 3H), 1.58 (br d,
J = 13.5 Hz, 2H), 1.81 (ddd, J = 13.5, 10.5, and 3.8 Hz, 2H), 2.19 (t, J = 10.8
Hz, 2H), 2.66ꢀ2.76 (m, 2H), 3.43ꢀ3.53 (m, 4H), 3.84 (br s, 2H), 6.82 (s,
1H), 6.86 (br s, 1H), 7.24ꢀ7.35 (m, 5H), 7.40 (t, J = 7.9 Hz, 1H),
7.44ꢀ7.48 (m, 1H), 7.81ꢀ7.85 (m, 1H), 7.94 (t, J = 1.8 Hz, 1H).
N-[Ethylamino-(8-benzyl-2,3,8-triaza-spiro[4.5]dec-3-en-2-
Compounds 40 and 41 from Chiral Separation of 6. The
enantiomers of 6 were separated by chiral preparative HPLC. The chiral
preparative HPLC system consisted of a 250 mm ꢁ 80 mm axial
compression column packed with CHIRALCELOK as stationary phase
and heptane/ethanol (80:20) as mobile phase. The flow rate used was
215 mL/min, and detection was done by using UV set at a wavelength of
220 nm. The racemate was injected as a solution of 4.5% m/v in
heptane/ethanol (70:30).
The first eluting enantiomer 40 was isolated in 67% yield with an ee of
99.6; [α]D +98 (c 1, MeOH). 1H NMR (400 MHz, CDCl3) data were
identical to the spectrum of the racemate (6).
The second eluting enantiomer 41 was isolated in 78% yield with an
ee of 99.0; [α]D ꢀ95 (c 1, MeOH). 1H NMR (400 MHz, CDCl3) data
were identical to the spectrum of the racemate (6).
N0-(3-Chlorophenylsulfonyl)-N-ethyl-4,4-dimethyl-4,5-di-
hydro-1H-pyrazole-1-carboxamidine (42). Following the same
procedure as described for 1, reaction of 74a (300 mg, 0.87 mmol) with
ethylamine (1 mL of a 70% solution in water, excess) and purification by
flash chromatography (DCM, DCM/acetone = 99:1 to 98:2) gave
290 mg (98%) of 42 as a colorless oil. 1H NMR (400 MHz, CDCl3) δ
1.16 (t, J = 7.3 Hz, 3H), 1.23 (s, 6H), 3.42ꢀ3.51 (m, 2H), 3.79 (br s,
2H), 6.76 (s, 1H), 6.78 (br s, 1H), 7.39 (t, J = 7.8 Hz, 1H), 7.44 (ddd, J =
8.0, 2.1, and 1.3 Hz, 1H), 7.82 (ddd, J = 7.7, 1.6, and 1.3 Hz, 1H), 7.93
(t, J = 2.0 Hz, 1H).
yl)methylidene]-3-chlorobenzenesulfonamide (6.91 g, 14.4 mmol) was
dissolved in 50 mL of 1,2-dichloroethane and cooled in an ice bath.
Dropwise, 1-chloroethyl chloroformate (1.73 mL, 15.9 mmol) was
added and the mixture was stirred for 1 h, keeping the temperature
below 5 °C. The mixture was concentrated under reduced pressure and
coevaporated three times with toluene. The residue was taken up in
50 mL of MeOH, and the mixture was stirred at room temperature over
the weekend. The mixture was concentrated under reduced pressure.
EtOAc and 2 M NaOH were added to the residue for extraction, and the
organic layer was separated, dried over Na2SO4, filtered, and concen-
trated under reduced pressure. The crude product was purified by flash
chromatography (EtOAc/MeOH/25% NH4OH = 80:20:2 to 70:30:3
to 50:50:5) to give 4.30 g (77%) of 47 as an off-white amorphous
N0-(3-Chlorophenylsulfonyl)-N,4,4-triethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (43). Following the same pro-
cedure as described for 1, reaction of 74b (540 mg, 1.44 mmol) with
ethylamine (2 mL of a 70% solution in water, excess) and purification
by extraction (DCM/5% aqueous NaHCO3) gave 520 mg (97%) of
1
compound. H NMR (400 MHz, CDCl3) δ 1.17 (t, J = 7.3 Hz, 3H),
1.51ꢀ1.59 (m, 2H), 1.69ꢀ1.78 (m, 2H), 2.72ꢀ2.80 (m, 2H),
2.95ꢀ3.04 (m, 2H), 3.43ꢀ3.53 (m, 2H), 3.87 (br s, 2H), 6.85 (s,
1H), 6.87 (br s, 1H), 7.41 (t, J = 7.8 Hz, 1H), 7.44ꢀ7.48 (m, 1H),
7.81ꢀ7.85 (m, 1H), 7.94 (t, J = 1.6 Hz, 1H).
1
43 as a pale yellow oil. H NMR (400 MHz, CDCl3) δ 0.87 (t, J =
7.5 Hz, 6H), 1.15 (t, J = 7.3 Hz, 3H), 1.49ꢀ1.64 (m, 4H), 3.41ꢀ3.50
(m, 2H), 3.81 (s, 2H), 6.73 (s, 1H), 6.75 (br s, 1H), 7.39 (t, J = 7.8 Hz,
1H), 7.43ꢀ7.47 (m, 1H), 7.80ꢀ7.85 (m, 1H), 7.94 (t, J = 1.6 Hz, 1H).
N-[Ethylamino-(2,3-diaza-spiro[4.4]non-3-en-2-yl)methyl-
idene]-3-chlorobenzenesulfonamide (44). Following the same
procedure as described for 1, reaction of 74c (550 mg, 1.48 mmol) with
ethylamine (2 mL of a 70% solution in water, excess) and purification by
extraction (DCM/water) gave 540 mg (99%) of 44 as a pale yellow oil.
1H NMR (400 MHz, CDCl3) δ 1.16 (t, J = 7.2 Hz, 3H), 1.63ꢀ1.85 (m,
10H), 3.42ꢀ3.53 (m, 2H), 3.85 (s, 2H), 6.81 (br s, 1H), 6.83 (s, 1H),
7.39 (t, J = 7.8 Hz, 1H), 7.42ꢀ7.47 (m, 1H), 7.80ꢀ7.85 (m, 1H), 7.93 (t,
J = 1.6 Hz, 1H).
N-[Ethylamino-(2,3-diaza-spiro[4.5]dec-3-en-2-yl)-methyl-
idene]-3-chlorobenzenesulfonamide (45). Following the same
procedure as described for 1, reaction of 74d (300 mg, 0.78 mmol) with
ethylamine (2 mL of a 70% solution in water, excess) and purification by
flash chromatography (DCM, DCM/acetone = 99:1 to 98:2) gave
200 mg (67%) of 45 as a colorless oil. 1H NMR (400 MHz, CDCl3) δ
1.16 (t, J = 7.2 Hz, 3H), 1.32ꢀ1.70 (m, 12H), 3.44ꢀ3.53 (m, 2H), 3.80
(s, 2H), 6.83 (br s, 1H), 6.83 (s, 1H), 7.40 (t, J = 7.8 Hz, 1H), 7.43ꢀ7.47
(m, 1H), 7.81ꢀ7.85 (m, 1H), 7.94 (t, J = 1.6 Hz, 1H).
N0-(3-Chlorophenylsulfonyl)-N-ethyl-4,4-bis(fluoromethyl)-
4,5-dihydro-1H-pyrazole-1-carboxamidine (48). Following the
same procedure as described for 1, reaction of 74g (0.12 g, 0.31 mmol)
with ethylamine (0.25 mL of a 70% solution in water, excess) and
purification by flash chromatography (DCM, DCM/MeOH = 99:1) gave
0.11 g (93%) of 48 as a pale yellow oil. 1HNMR(400MHz,CDCl3) δ1.20
(t, J = 7.3 Hz, 3H), 3.48ꢀ3.56 (m, 2H), 3.94 (br s, 2H), 4.50 (dd, J = 46.6
and 9.5 Hz, 2H), 4.53 (dd, J = 47.0 and 9.5 Hz, 2H), 6.91 (s, 1H), 7.00 (br
s, 1H), 7.41 (t, J = 7.7 Hz, 1H), 7.45ꢀ7.49 (m, 1H), 7.79ꢀ7.83 (m, 1H),
7.92 (t, J = 1.7 Hz, 1H).
N0-(Phenylsulfonyl)-N-ethyl-4,4-dimethyl-4,5-dihydro-
1H-pyrazole-1-carboxamidine (49). 75 (500 mg, 2.44 mmol)
was suspended in 10 mL of DCM. DiPEA (0.88 mL, 5.13 mmol) and
benzenesulfonyl chloride (0.31 mL, 2.44 mmol) were added, and the
mixture was stirred overnight at room temperature. The mixture was
extracted with 5% aqueous NaHCO3 and subsequently with 2 M
NaOH, and the organic layer was dried over Na2SO4, filtered, and
concentrated under reduced pressure. The crude product was
purified by flash chromatography (DCM/MeOH = 99.5:0.5 to
1
99:1) to give 520 mg (69%) of 49 as an orange oil. H NMR (400
MHz, CDCl3) δ 1.15 (t, J = 7.3 Hz, 3H), 1.22 (s, 6H), 3.42ꢀ3.51 (m,
2H), 3.79 (br s, 2H), 6.74 (s, 1H), 6.84 (br s, 1H), 7.42ꢀ7.51 (m,
3H), 7.92ꢀ7.97 (m, 2H).
N-[Ethylamino-(8-oxa-2,3-diaza-spiro[4.5]dec-3-en-2-
yl)methylidene]-3-chlorobenzenesulfonamide (46). Fol-
lowing the same procedure as described for 1, reaction of 74e
(810 mg, 2.09 mmol) with ethylamine (1.2 mL of a 70% solution in
water, excess) and purification by flash chromatography (EtOAc)
followed by crystallization from diisopropyl ether gave 240 mg
(30%) of 46 as white crystals, mp 108ꢀ110 °C. 1H NMR (400
MHz, CDCl3) δ 1.17 (t, J = 7.3 Hz, 3H), 1.52ꢀ1.59 (m, 2H),
1.80ꢀ1.89 (m, 2H), 3.44ꢀ3.53 (m, 2H), 3.53ꢀ3.60 (m, 2H),
3.83ꢀ3.90 (m, 2H), 6.87 (s, 1H), 6.90 (br s, 1H), 7.41 (t, J = 7.8
3-Pyridin-4-ylpropenal (59f). Triphenylphosphoranylidene acet-
aldehyde (6.08 g, 20.0 mmol) was suspended in 10 mL of dry N,N-
dimethylformamide (DMF). Pyridine-4-carbaldehyde (1.93 mL, 20.0 mmol)
was added, and the mixture was stirred overnight at room temperature.
The mixture was taken up in EtOAc and washed four times with 5%
aqueous NaHCO3. The organic layer was dried over Na2SO4, filtered,
and concentrated under reduced pressure. The residue was taken up in
PE, and the suspension was filtered. The filtrate was concentrated under
7046
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
reduced pressure to give 1.17 g of 59f as a crude yellow oil. This material
was used in subsequent steps without further purification. 1H NMR (400
MHz, CDCl3) δ 6.85 (dd, J = 16 and 8 Hz, 1H), 7.39ꢀ7.46 (m, 3H),
8.72 (d, J = 6 Hz, 2H), 9.78 (d, J = 8 Hz, 1H).
Na2SO4, filtered, and concentrated under reduced pressure to give 5.6 g
of a yellow oil containing 45% of the anticipated title compound and
55% of the noncyclized hydrazone intermediate. Additional refluxing for
24 h in n-butanol gave, after similar workup, 5.3 g of 60g as a crude
brown oil. This material was used in subsequent steps without further
purification. 1H NMR (400 MHz, CDCl3) δ 2.92 (ddd, J = 17, 8, and
2 Hz, 1H), 3.02 (ddd, J = 17, 10, and 2 Hz, 1H), 4.76 (dd, J = 10 and
8 Hz, 1H), 5.78 (br s, 1H), 6.20 (d, J = 3 Hz, 1H), 6.32 (dd, J = 3 and 2 Hz,
1H), 6.87 (br s, 1H), 7.36 (m, 1H).
4-Ethyl-4,5-dihydro-1H-pyrazole (60a). Hydrazine hydrate
(58.0 mL, 1.19 mol) was dissolved in MeOH (300 mL) and cooled
in an ice bath. To this mixture, a solution of 2-ethylacrolein (100 g,
1.19 mol) in MeOH (100 mL) was added at such a rate that the tem-
perature was kept below 10 °C. The ice bath was removed and the mixture
was stirred overnight at room temperature, after which the MeOH was
evaporated under reduced pressure. The product was obtained by vacuum
distillation (70ꢀ80 °C, 20 mbar) to give 54.9 g (47%) of 60a as a colorless
5-(4-Fluorophenyl)-4,5-dihydro-1H-pyrazole (60h). Hydra-
zine hydrate (13.23 mL, 199.8 mmol) was added to 50 mL of THF. The
mixture was cooled to ꢀ10 °C. A solution of 4-fluorocinnamaldehyde
(3.00 g, 19.98 mmol) in 50 mL of Et2O was added dropwise. The
mixture was stirred overnight, slowly reaching room temperature. Water
was added, and the mixture was extracted twice with EtOAc. The
combined organic layers were dried over MgSO4, filtered, and concen-
trated under reduced pressure. The residue was dissolved in MeOH and
brought onto a conditioned SCX ion exchange column. The column was
washed with MeOH, and the product was eluted with 1 M NH31in
MeOH. Evaporation yielded 2.15 g (52%) of 60h as a yellow oil. H
NMR (400 MHz, CDCl3) δ 2.66 (ddd, J = 17.1, 9.0, and 1.6 Hz, 1H),
3.13 (ddd, J = 17.1, 10.7, and 1.6 Hz, 1H), 4.71 (dd, J = 10.7 and 9.0 Hz, 1H),
5.81 (br s, 1H), 7.02 (t, J = 8.7 Hz, 2H), 7.30 (dd, J = 8.7 and 5.5 Hz, 2H).
N-(Bis-methylsulfanylmethylidene)-2-chlorobenzenesul-
fonamide (62a). 2-Chlorobenzenesulfonamide (41.6 g, 0.22 mol)
and carbon disulfide (22 mL, 0.36 mol) were dissolved in DMF
(300 mL), and the mixture was cooled in an ice bath. A solution of
KOH (29.0 g 0.52 mol) in water (100 mL) was added dropwise at such a
rate that the temperature was kept below 10 °C. The mixture was stirred
for 30 min at 5ꢀ10 °C. Subsequently, methyl iodide (32 mL, 0.51 mol)
was added dropwise at such a rate that the temperature was kept below
10 °C. Then the mixture was allowed to warm to room temperature and
stirred for another 30 min. Water (250 mL) was added, and the formed
white precipitate was filtered off and washed with water followed by
EtOH. Drying in vacuo yielded 42.6 g (66%) of 62a as white crystals,
mp 129ꢀ130 °C. 1H NMR (200 MHz, CDCl3) δ 2.57 (s, 6H),
7.32ꢀ7.60 (m, 3H), 8.11ꢀ8.27 (br d, J = 7.5 Hz, 1H).
1
liquid. H NMR (400 MHz, CDCl3) δ 0.98 (t, J = 7 Hz, 3H), 1.43ꢀ
1.68 (m, 2H), 2.90ꢀ3.01 (m, 2H), 3.42ꢀ3.53 (m, 1H), 5.38 (br s, 1H),
6.77 (s, 1H).
4-Methyl-4,5-dihydro-1H-pyrazole (60b). Following the same
procedure as described for 60a, using CH3CN as solvent, reaction of
hydrazine hydrate (16.65 mL, 0.34 mol) with 2-methylacrolein (24.02 g,
0.34 mol) and purification by vacuum distillation (102ꢀ108 °C, 250
mbar) gave 7.0 g (24%) of 60b as a colorless liquid. 1H NMR (400 MHz,
CDCl3) δ 1.18 (d, J = 7 Hz, 3H), 2.90 (t, J = 9 Hz, 1H), 3.00ꢀ3.12 (m,
1H), 3.51 (t, J = 9 Hz, 1H), 5.48 (br s, 1H), 6.73 (br s, 1H).
3-Methyl-4,5-dihydro-1H-pyrazole (60d). Hydrazine hydrate
(29.2 mL, 0.60 mol) was dissolved in MeOH (50 mL). To this solution,
methyl vinyl ketone (50 mL, 0.60 mol) was added at such a rate that the
temperature was kept below 50 °C. The mixture was stirred for 1 h at
50 °C, after which the MeOH was evaporated under reduced pressure.
The product was obtained by vacuum distillation (68ꢀ82 °C, 20 mbar)
to give 11.8 g (23%) of 60d as a colorless liquid. 1H NMR (400 MHz,
DMSO-d6) δ 1.88 (s, 3H), 2.47 (t, J = 10 Hz, 2H), 3.15 (t, J = 10 Hz,
2H), 6.10 (br s, 1H).
5-Phenyl-4,5-dihydro-1H-pyrazole (60e). Hydrazine hydrate
(9.2 mL, 0.19 mol) was heated to reflux. A solution of cinnamaldehyde
(10.0 g, 75.8 mmol) in t-BuOH (20 mL) was added dropwise, and the
mixture was refluxed overnight. After concentration under reduced
pressure, water was added to the residue and the mixture was extracted
twice with DCM. The combined organic layers were washed with water,
dried over Na2SO4, and concentrated under reduced pressure to give
10.46 g (81%) of a yellow oil containing 85% of the desired product 60e,
Via a similar procedure, compounds 62bꢀk were synthesized. Full
experimental details are provided in the Supporting Information.
N-[(4-Ethyl-4,5-dihydro-1H-pyrazol-1-yl)methylsulfanyl-
methylidene]-2-chlorobenzenesulfonamide (63aa). 60a (6.7 g,
68.3 mmol) was dissolved in pyridine (100 mL). 62a (20.2 g, 68.3
mmol) was added, and the mixture was refluxed overnight. The mixture
was concentrated under reduced pressure and the crude product was
purified by flash chromatography (DCM, DCM/acetone = 99:1 to 95:5)
1
which was used in subsequent steps without further purification. H
NMR (200 MHz, CDCl3) δ 2.61ꢀ2.80 (m, 1H), 3.03ꢀ3.23 (m, 1H),
4.72 (dd, J = 10 and 8 Hz, 1H), 5.83 (br s, 1H), 6.83 (t, J = 1 Hz, 1H),
7.20ꢀ7.47 (m, 5H).
4-(4,5-Dihydro-1H-pyrazol-5-yl)pyridine (60f). Hydrazine
hydrate (4.27 mL, 87.9 mmol) was added to 20 mL of Et2O. The
emulsion was cooled in an ice/salt bath to ꢀ10 °C. A solution of 59f
(1.17 g crude, max 8.8 mmol) in 20 mL of Et2O was added dropwise.
The mixture was stirred overnight, slowly reaching room temperature in
the melting ice bath. Subsequently, 5% aqueous NaHCO3 was added
and the mixture was extracted five times with EtOAc. The combined
organic layers were dried over Na2SO4, filtered, and concentrated under
reduced pressure. The residue was dissolved in MeOH and brought onto
a conditioned SCX ion exchange column. The column was washed with
MeOH, and the product was eluted with 1 M NH3 in MeOH.
Evaporation yielded 1.23 g (33% over two steps) of 60f as a brown
oil. 1H NMR (400 MHz, CDCl3) δ 2.61ꢀ2.71 (m, 1H), 3.15ꢀ3.25 (m,
1H), 4.68ꢀ4.76 (m, 1H), 6.82 (br s, 1H), 6.83 (br b, 1H), 7.27(d, J = 6
Hz, 2H), 8.57 (d, J = 6 Hz, 2H).
1
to give 15.6 g (62%) of 63aa as a yellow oil. H NMR (400 MHz,
CDCl3) δ 1.02 (t, J = 7 Hz, 3H), 1.55ꢀ1.77 (m, 2H), 2.28 (s, 3H),
3.27ꢀ3.39 (m, 1H), 4.13 (dd, J = 12 and 7 Hz, 1H), 4.57 (t, J = 12 Hz,
1H), 7.16 (d, J =2 Hz, 1H), 7.39(dt, J = 7 and 2 Hz, 1H), 7.46 (dt, J = 7 and
2 Hz, 1H), 7.52 (dd, J = 7 and 2 Hz, 1H), 8.18 (dd, J = 7 and 2 Hz, 1H).
Via a similar procedure, compounds 63ab, 63ac, 63ad, 63ae, 63af,
63ag, 63ah, 63ai, 63aj, 63ak, 63bc, 63cc, 63dc, 63ec, 63fc, 63gc, and
63hc were synthesized. Full experimental details are provided in the
Supporting Information.
1-Ethyl-S-methylisothiourea Hydroiodide (66a). Ethylthiour-
ea (20.5 g, 0.20 mol) was dissolved in 100 mL of EtOH. The mixture was
cooled in an ice bath, and MeI (13.5 mL, 0.22 mol) was added dropwise.
The mixture was stirred for 1 h at room temperature and concentrated
under reduced pressure to give 48.3 g (100%) of 66a as a light yellow oil.
1H NMR (400 MHz, DMSO-d6) δ 1.17 (t, J = 7 Hz, 3H), 2.61 (s, 3H),
3.34 (q, J = 7 Hz, 2H), 9.10 (br s, 3H).
N0,4-Diethyl-4,5-dihydro-1H-pyrazole-1-carboxamidine
Hydrochloride (67a). 60a (19.36 g, 0.20 mol) was dissolved in
100 mL of toluene. 66a (48.5 g, 0.20 mol) and diisopropylethylamine
5-(Furan-2-yl)-4,5-dihydro-1H-pyrazole (60g). 3-(2-Furyl)
acrolein (5.0 g, 40.9 mmol) was dissolved in 25 mL tert-butanol.
Hydrazine hydrate (4.0 mL, 81.9 mmol) was added, and the reaction
mixture was refluxed for 24 h. The mixture was concentrated under
reduced pressure. The residue was taken up in DCM and washed
twice with 5% aqueous NaHCO3. The organic phase was dried over
7047
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
(33.8 mL, 0.20 mol) were added, and the mixture was refluxed for
48 h. After concentration under reduced pressure, 2 M NaOH was
added and the mixture was extracted three times with DCM. The
combined organic layers were dried over Na2SO4, filtered, and con-
centrated under reduced pressure to yield 32.7 g of a red oil. The oil
was dissolved in EtOH, and 1 M HCl in EtOH (194 mL) was added
dropwise. The mixture was stirred at room temperature for 30 min
and concentrated under reduced pressure. By crystallization from
acetonitrile/MtBE = 1:1, 11.52 g (29%) of 67a was isolated as a beige
solid. 1H NMR (400 MHz, DMSO-d6) δ 0.96 (t, J = 7 Hz, 3H), 1.16
(t, J = 7 Hz, 3H), 1.46ꢀ1.73 (m, 2H), 3.28ꢀ3.45 (m, 3H), 3.56 (dd,
J = 10 and 6 Hz, 1H), 3.98 (t, J = 10 Hz, 1H), 7.33 (br s, 1H), 7.98
(br s, 2H), 8.07 (t, J = 6 Hz, 1H).
2,2-Diethylmalononitrile (70b). Malononitrile (15.2 g, 0.23 mol)
was mixed with tetrabutylammonium bromide (3.0 g, 4 mol %) and
ethyl iodide (36.8 mL, 0.46 mol). After being stirred for 30 min at
room temperature, the mixture was cooled in an ice bath and potassium
tert-butoxide (51.6 g, 0.46 mol) was added portionwise. The ice bath was
removed, and the mixture was stirred for 30 min at room temperature.
Water was added, and the aqueous portion was extracted with DCM.
The organic layer was dried over Na2SO4, filtered, and concentrated
under reduced pressure. Purification by flash chromatography (DCM)
gave 20.4 g (73%) of 70b as an orange oil, which solidified upon
standing. 1H NMR (400 MHz, CDCl3) δ 1.29 (t, J = 7 Hz, 6H), 2.00
(q, J = 7 Hz, 4H).
CDCl3) δ 2.93 (t, J = 7.0 Hz, 4H), 3.50 (t, J = 7.0 Hz, 4H), 3.74 (s, 2H),
7.29ꢀ7.36 (m, 5H).
Malononitrile (12.96 g, 0.20 mol) was dissolved in 250 mL of DMF.
K2CO3 (29.82 g, 0.22 mol) was added, and the mixture was stirred at
65 °C for 2 h. A solution of benzyl-bis-(2-chloroethyl)amine (46.0 g,
0.20 mmol) in 50 mL of DMF was added dropwise at 65 °C, and the
mixture was stirred at 65 °C for another 3 h. After cooling to room
temperature, the mixture was diluted with EtOAc and washed three
times with 5% aqueous NaHCO3. The organic phase was dried over
Na2SO4, filtered, and concentrated under reduced pressure to give 25.7 g
(56%) of 70f as a colorless oil. 1H NMR (400 MHz, CDCl3) δ 2.24 (t,
J = 5.2 Hz, 4H), 2.64 (br s, 4H), 3.55 (s, 2H), 7.26ꢀ7.36 (m, 5H).
2,2-Diethylpropane-1,3-diamine (71b). A suspensionofLiAlH4
(4.66 g, 0.123 mol) in 100 mL of dry Et2O was cooled in an ice bath, and a
solution of 70b (5.0 g, 40.9 mmol) in 50 mL of dry Et2O was added
dropwise at such a rate that the temperature was kept below 20 °C. The
mixture was stirred overnight at room temperature, cooled in an ice bath,
and quenchedby adding water (5 mL), 2 M aqueous NaOH(10 mL), and
again water (5 mL). The suspension was filtered, the filter cake was
washed with Et2O, and the combined filtrates were concentrated under
reduced pressure to give 5.0 g (94%) of 71b as a clear, light-yellow liquid.
1H NMR (400 MHz, CDCl3) δ 0.80 (t, J = 8 Hz, 6H), 1.08 (br s 4H), 1.22
(q, J = 8 Hz, 4H), 2.52 (s, 4H).
Via a similar procedure, compounds 71c, 71d, and 71f were synthesized.
Full experimental details are provided in the Supporting Information.
(4-(Aminomethyl)tetrahydropyran-4-yl)methanamine
(71e). 70e (1.52 g, 11.2 mmol) was dissolved in 25 mL of dry THF
and cooled to ꢀ10 °C. Dropwise, a 1 M solution of BH3 in THF
(55.8 mL, 55.8 mmol) was added. The mixture was warmed to room
temperature and subsequently stirred for 6 h at 60 °C. The reaction
mixture was cooled in an ice bath, quenched by dropwise addition of
6 M aqueous HCl (24.2 mL, 145 mmol), warmed to room tempera-
ture, and stirred for 2 h. The mixture was neutralized with 2 M
aqueous NaOH and washed three times with DCM. The aqueous
layer was concentrated under reduced pressure, and the residue was
taken up in CHCl3. The solids were filtered off and the organic
phase was concentrated under reduced pressure to give 1.0 g (62%)
of 71e as a yellow oil. 1H NMR (400 MHz, CDCl3) δ 1.43 (br s, 4H),
1.46 (t, J = 5.6 Hz, 4H), 2.73 (s, 4H), 3.67 (t, J = 5.6 Hz, 4H).
4,4-Dimethyl-4,5-dihydro-3H-pyrazole (72a). 2,2-Dimethyl-
1,3-propanediamine (20.0 g, 0.196 mol) was dissolved in a mixture of
water (80 mL) and MeOH (20 mL) and cooled in an ice bath.
Simultaneously, 30% aqueous H2O2 (120 mL, 1.17 mol) and 10%
aqueous NaClO (350 mL, 0.47 mol) were added dropwise. The mixture
was stirred overnight at room temperature and extracted with DCM.
The organic layer was dried over Na2SO4, filtered, and concentrated
under reduced pressure. Vacuum distillation (102ꢀ105 °C, 250 mbar)
gave 11.4 g (59%) of 72a as a colorless viscous oil. 1H NMR (200 MHz,
CDCl3) δ 1.05 (s, 6H), 4.14 (s, 4H).
Cyclopentane-1,1-dicarbonitrile (70c). Malononitrile (15.0 g,
0.23 mol) was dissolved in 200 mL of dry DMF, and the mixture was
cooled in an ice bath. Subsequently, 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU) (75 mL, 0.50 mol) and 1,5-dibromobutane (29.6 mL, 0.25 mol)
were added dropwise. The ice bath was removed, extra dry DMF
(100 mL) was added to keep the mixture stirrable, and the mixture
was heated to 80 °C for 2 h. After cooling, the mixture was poured into
DCM and washed five times with 5% aqueous NaHCO3. The organic
layer was dried over Na2SO4 and concentrated under reduced pressure.
The crude product was purified by flash chromatography (PE/EtOAc =
1
9:1) to give 23.4 g (86%) of 70c as a colorless liquid. H NMR (400
MHz, CDCl3) δ 1.93ꢀ2.04 (m, 4H), 2.36ꢀ2.48 (m, 4H).
Cyclohexane-1,1-dicarbonitrile (70d). Following the same
procedure as described for 70c, reaction of malononitrile (15.0 g,
0.23 mol) with 1,5-dibromopentane (34 mL, 0.25 mol) in the presence
of DBU (75 mL, 0.50 mol) and purification by flash chromatography
(PE/EtOAc = 9:1) gave 25.7 g (84%) of 70d as white crystals. 1H NMR
(400 MHz, CDCl3) δ 1.48ꢀ1.61 (m, 2H), 1.70ꢀ1.82 (m, 4H), 2.12
(t, J = 6 Hz, 4H).
Tetrahydropyran-4,4-dicarbonitrile (70e). Malononitrile (5.0 g,
75.7 mmol) was dissolved in 5 mL of DMSO. Subsequently, bis(2-
bromoethyl) ether (9.49 mL, 75.7 mmol) and tetrabutylammonium
bromide (1.22 g, 5 mol %) were added, followed by portionwise addition
of potassium tert-butoxide (8.49 g, 75.7 mmol). The mixture was stirred
for 4 h at room temperature, taken up in DCM, and washed 3 times with
5% aqueous NaHCO3. The organic phase was dried over Na2SO4,
filtered, and concentrated under reduced pressure. The crude material
was purified by flash chromatography (PE/Et2O = 65:35) to give 2.49 g
(24%) of 70e as an off-white solid. 1H NMR (400 MHz, CDCl3) δ 2.24
(t, J = 5.1 Hz, 4H), 3.87 (t, J = 5.1 Hz, 4H).
1-Benzylpiperidine-4,4-dicarbonitrile (70f). Bis-(2-chloro-
ethyl)amine hydrochloride (150.0 g, 0.84 mol) was suspended in
500 mL of acetonitrile. K2CO3 (348.4 g, 2.52 mol) and benzyl bromide
(99.8 mL, 0.84 mol) were added, and the mixture was refluxed for 24 h.
After cooling to room temperature, the mixture was filtered and the
filtrate was concentrated under reduced pressure. Purification by flash
chromatography (PE, PE/Et2O = 95:5 to 90:10) gave 46.0 g (21%) of
benzyl-bis-(2-chloroethyl)amine as a colorless oil. 1H NMR (400 MHz,
Via a similar procedure, compounds 72bꢀg were synthesized. Full
experimental details are provided in the Supporting Information.
N-[(4,4-Dimethyl-4,5-dihydro-1H-pyrazol-1-yl)methylsul-
fanylmethylidene]-3-chlorobenzenesulfonamide (74a). Fol-
lowing the same procedure as described for 63aa, reaction of 72a (0.23 g,
2.37 mmol) with 62c (700 mg, 2.37 mmol) and purification by flash
chromatography (DCM, DCM/acetone = 99:1 to 98:2) gave 600 mg
1
(73%) of 74a as white crystals, mp 118.5ꢀ119.7 °C. H NMR (400
MHz, CDCl3) δ 1.31 (s, 6H), 2.23 (s, 3H), 4.21 (s, 2H), 6.98 (s, 1H),
7.43 (t, J = 8 Hz, 1H), 7.47ꢀ7.52 (m, 1H), 7.86 (dt, J = 8 and 2 Hz, 1H),
7.97 (t, J = 2 Hz, 1H).
Via a similar procedure, compounds 74bꢀg were synthesized. Full
experimental details are provided in the Supporting Information.
N0-Ethyl-4,4-dimethyl-4,5-dihydro-1H-pyrazole-1-carbox-
amidine Hydrochloride (75). 72a (12.0 g, 122 mmol) was dissolved
7048
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
in 100 mL of pyridine. A solution of 66a (30.0 g, 122 mmol) in 50 mL of
pyridine was added, and the mixture was refluxed for 20 h. The mixture
was cooled to room temperature and concentrated under reduced
pressure, and the residue was taken up in 120 mL of DCM. The organic
phase was extracted with 2 N NaOH (2 ꢁ 120 mL), washed with water
(120 mL), dried over Na2SO4, and evaporated under reduced pressure
to yield 16.3 g (max 97 mmol, 79%) of an orange oil. The oil (10.0 g, max
59 mmol) was taken up in 50 mL of EtOAc and heated to 60 °C. After
removal of the heat source, 20 mL of a 5ꢀ6 N solution of HCl in
isopropanol was dosed over 4 min. After the mixture was cooled to room
temperature, 50 mL of EtOAc was added over 4 min, and the mixture
was stirred at 20 °C for 90 min. The formed crystals were collected by
filtration, washed with 20 mL of EtOAc, and dried in vacuo at mild
heating to give 6.52 g (32 mmol, 54%) of 75 as a yellow solid. 1H NMR
(400 MHz, DMSO-d6) δ 1.08 (t, J = 7.3 Hz, 3H), 1.15 (s, 6H), 3.08 (q,
J = 7.3 Hz, 2H), 3.37 (s, 2H), 6.65 (s, 1H), 8.25 (s, 2H).
In Vitro Affinity for the Human 5-HT6R. Affinity for human
5-HT6 receptors was measured in a membrane preparation of CHO cells
transfected with human 5-HT6 receptors by binding studies using
[3H]N-methyllysergic acid diethylamide ([3H]LSD) as ligand. The
membrane preparation was prepared from cells supplied by Euro-
screen (Brussels, Belgium). CHO/Gα16/mtAEQ/h5HT6-A1 cells were
grown in T-flasks in CHO-S-SFM II medium (Gibco BRL), supple-
mented with 1% dialyzed FCS, 2 mM L-glutamine, Geneticin at 500
μg/mL, and Zeocin at 200 μg/mL. Cells were harvested using 0.25% trypsin
(1 mL/T175-flask), centrifuged, and subsequently suspended in CHO-
S-SFM II medium and frozen at ꢀ80 °C. After thawing, cells were
centrifuged at 4 °C for 3 min at 1500g. From the pellet, cell membranes
were prepared by two cycles of homogenization (Potter-Elvehjem 10
strokes, 600 rpm) and centrifugation (40000g for 15 min, 4 °C). The
assay was established to achieve steady state conditions and to optimize
specific binding. For the 5-HT6R, membranes from 5 ꢁ 105 cells were
incubated with 5.0 nM [3H]LSD and test compound at 37 °C for 30 min.
Nonspecific binding was determined using 10ꢀ5 M serotonin. Assays
were terminated by vacuum filtration through glass fiber filters (GF/B)
that had been pretreated with 0.5% polyethyleneimine. Total and bound
radioactivity was determined by liquid scintillation counting. Greater
than 80% specific binding was achieved in each of these assays.
Compounds were tested at a 4 log concentration range; all determina-
tions were performed as triplicates. IC50 values were determined by
nonlinear regression analysis using Hill equation curve fitting. The
inhibition constants (Ki) were calculated from the ChengꢀPrusoff
equation:
Compounds were dissolved in dimethylsulfoxide (DMSO) at 10 mM
and diluted in assay buffer to tested concentrations, with a maximal
content of 0.1% (v/v) DMSO. Compounds were tested at a 5 log
concentration range, and three independent experiments were per-
formed in duplicate. Agonistic effects of compounds are expressed as
pEC50. Antagonistic effects of compounds were determined as inhibi-
tion of 10ꢀ8 M α-methylserotonin induced luminescence, and the pA2
was calculated according to the ChengꢀPrusoff equation.
In Vitro Determination of Metabolic Stability in the Pre-
sence of Human/Rat Hepatocytes. Pooled human hepatocytes
(five males, five females) and pooled rat hepatocytes (male Spragueꢀ
Dawley rats) were purchased from Invitrotech (Germany) and Tebu-bio
(France), respectively, and were stored in liquid nitrogen prior to use.
To obtain an in vitro estimate of biological half-life (t1/2), hepatocytes
(50 000 cells per well, in 96-well plates) were incubated with test
compound (1 μg/mL) in Williams medium E containing 5 μg/mL
insulin at 37 °C in a water bath, under an atmosphere of oxygen
containing 4ꢀ7% CO2. Final volume per well was 100 μL, and final
DMSO concentration (from stock solution of test compound) was kept
at e0.1% to avoid toxic effects on hepatocytes. Control incubations were
included; for each compound tested an incubation without hepatocytes
was performed to check for chemical instability. In each test plate, an
incubation was performed without compound to check for interfering
background, and midazolam was included as reference compound with
known characteristics. All incubations were performed in duplicate for
each test compound. Compounds were incubated for 0, 10, 20, 40, and
60 min. Reactions were stopped by the addition of 100 μL of ice-cold
acetonitrile at the appropriate time points, and the plates were vortexed
and put on ice. Incubation plates were centrifuged at 2500 rpm for 5 min
at 4 °C to precipitate the protein. The supernatant from each well was
pipetted into a collection plate, covered with a rubber cover, and stored
at ꢀ80 °C until HPLCꢀMS analysis. After thawing, the samples were
mixed and centrifuged at 3500 rpm for 10 min at 4 °C. To measure
possible reduction of the concentration of test compounds, samples
were injected into a single quadrupole HPLCꢀMS system (Agilent
series 1100 LC-MSD), using a gradient in order to achieve chromato-
graphic separation. In the mass spectrometer, ionization was achieved by
ESI, followed by analysis of the formed ions by SIM. Eluent A consisted
of 0.77 g of ammonium acetate + 800 mL of water + 100 mL of methanol
+ 100 mL of acetonitrile. Eluent B consisted of 0.77 g of ammonium
acetate + 100 mL of water + 100 mL of methanol + 800 mL of
acetonitrile. Table 7 shows the pump gradient and other parameters.
The “areas under the curve” at the different incubation times were
integrated and plotted against (incubation) time, from which the t1/2
was derived. The intrinsic clearance Clint [(μL/min)/106 cells] was
calculated as follows: [ln(2)/(t1/2 in min)](well volume in μL)/
(number of cells per well ꢁ 10ꢀ6) = [0.693/(t1/2 in min)][100/0.05].
P-Glycoprotein Assay. The capability of the human MDR1
P-glycoprotein pump to translocate compounds over a cellular mono-
layer of PK1 LLC MDR cells was assessed. The transport method
essentially described in the literature80 was used. Compounds were
added at the start of the experiment at 1 μg/mL to one side of the cellular
layer. The bottom-to-top transport was measured as well as the top-to-
bottom transport. Compound detection was performed using a LC/MS
method. The P-glycoprotein (P-gp) factor was expressed as the ratio of
the bottom-to-top transport over the top-to-bottom transport; com-
pounds with a P-gp factor of g2 are considered to be P-gp substrates.
The membrane passage (MP) was expressed as the mean percentage of
compound transported from bottom-to-top and from top-to-bottom at
3 h after adding the compound; compounds with a membrane passage of
g15% are considered to be well permeable.
IC50
Ki ¼
L
1 þ
Kd
wherein L represents the concentration radioligand ([3H]LSD) in the
assay and Kd the affinity of the radioligand for the receptor. Results are
expressed as mean Ki ( SD from at least three separate experiments.
In Vitro Functional Activity ((Ant)Agonism) on the Human
5-HT6R. The CHO human 5-HT6R aequorin assay was bought from
Euroscreen, Brussels, Belgium (Euroscreen, technical dossier, human
recombinant serotonin 5-HT6-A1 receptor, DNA clone, and CHO
AequoScreen recombinant cell line, catalog no. ES-316-A, February
2003). Human-5-HT6R-aequorin cells express mitochondrial targeted
apo-aequorin. Cells have to be loaded with coelanterazine in order to
reconstitute active aequorin. After binding of agonists to the human
5-HT6R, the intracellular calcium concentration increases and binding of
calcium to the apo-aequorin/coelenterazine complex leads to an oxida-
tion reaction of coelenterazine, which results in the production of apo-
aequorin, coelenteramide, CO2, and light (λmax = 469 nm). This
luminescent response is dependent on the agonist concentration.
Luminescence is measured using the MicroBeta Jet (Perkin-Elmer).
Receptor Binding and Enzyme Inhibitory Studies. Selec-
tivity profiling of selected compounds was performed at CEREP (Celle
l’Evescault, France).60 Compounds were tested at 10 μM (in duplicate).
7049
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Table 7a
5-HT6R antagonists from public and proprietary sources were docked
in the binding pocket. This was an ongoing process. As the model was in
agreement with reported and observed SAR of reference/proprietary
compounds, as well as mutagenesis data from the public domain,43,50,85,86
it was employed for optimization efforts of the N0-(arylsulfonyl)pyrazoline-
1-carboxamidines. Representatives of novel structures were manually
docked in the 5-HT6R binding pocket and minimized with restraint
backbone. Close analogues were subsequently docked in the rigid pocket
with GOLD,87 version 4, and subjected to further minimization, again using
the OPLS2005 force field and restraint backbone.
time (min)
eluent A (%)
eluent B (%)
flow (mL/min)
0.00
3.60
100
0
0
100
100
0
1
1
1
1
1
7.20
0
8.50
100
100
11.00
0
a Column: Inertsil 5 ODS-3 100 mm ꢁ 3.0 mm (CP22234) equipped
with Chromsep Guard column SS 10 mm ꢁ 2 mm (CP28141)
thermostated at 25 °C. Well plate temperature: 4 °C. Injection volume:
20 μL. Splitter (post column), 1:4. Total run time: 11.0 min. Detection
SIM: (M + H)+, (M ꢀ H)ꢀ, obtained from full scan recording. ESI
(pos/neg) spray: 4.0 kV. Fragmentor: 70. Gain: 2.0. Dwell: 700 ms.
Nebulizer pressure: 42 psi. Drying gas temperature: 325 °C. Capillary
temperature: 325 °C.
NMR Studies. The presence of an internal hydrogen bond between
the compound’s guanidine NH and sulfonamide SdO was experimen-
tally confirmed using the method published by Jansma et al.63 Measure-
ments were performed on a Bruker DRX600 spectrometer, equipped
with a CPTCI triple resonance 5 mm cryoprobe and running the
Topspin 1.3 software. In short, to a 10 mM solution of each investigated
compound dissolved in 0.6 mL of CDCl3 (0.05% tetramethylsilane
(TMS), Cambridge Isotope Laboratories Inc.) progressive aliquots of
DMSO-d6 (Cambridge Isotope Laboratories Inc.) were added at 0, 5, 7,
10, 14, 20, 28, 40, 56, and 80 μL, making the final solution 30% v/v
For the assays in which no activity was detected (PS or PI of <40%), the
compound was tested at least two times at 10 μM. In assays in which the
compound was active, it was tested further at lower concentrations to
determine an IC50.
1
DMSO-d6. The H NMR spectrum was recorded at each step. After
correction of the TMS reference to 0 ppm, the chemical shifts of the NH
proton(s) and other relevant peaks were recorded as a function of the %
DMSO. Dilution effects were assessed to have a negligible effect on the
studied chemical shifts. Drying the solution on 4ꢀ5 Å molecular sieves
before the titrations also had no effect on the chemical shift, indicating
that water content of the sample in CDCl3 did not influence the
measurements.
Receptor/Ion Channel/Transporter Binding Studies. Affinity for the
investigated target was measured in a membrane preparation of cells or
tissue with the specific target by binding studies using [3H]ligand.
Compounds were dissolved in dimethylsulfoxide (DMSO) at 10 mM
and diluted in assay buffer to tested concentrations. Testing was at a 3
log concentration range around a predetermined IC50 for the respective
assay: e.g., 10, 1, 0.1, and 0.01 μM for IC50 of 0.3 μM and 300, 30, 3, and
0.3 nM for one with IC50 of 10 nM. All determinations were performed
as duplicates. The highest concentration tested for primes was 10 μM.
Results were expressed as percentage of total ligand binding and that of
nonspecific binding, per concentration of compound tested (in
duplicate). From the concentration displacement curves, IC50 values
were determined by nonlinear regression analysis using Hill equation
curve fitting. For receptor binding assays, the inhibition constants (Ki)
were calculated from the ChengꢀPrusoff equation Ki = IC50/(1 + L/Kd),
where L is the concentration of radioligand in the assay and Kd the
affinity of the radioligand for the receptor. Results were expressed as pKi.
Compounds with no significant affinity at 10 μM and higher were
concluded to be inactive, denoted by pKi of <5.0.
Enzyme Inhibition Studies. Inhibitory properties of compounds for
the investigated enzyme were measured in an enzyme preparation
following incubation of compound with the enzyme preparation and
the substrate at a specific time and temperature. The enzyme product
was measured by a specified method of detection. The results were
expressed as a percentage of control specific activity ((measured specific
activity/control specific activity) ꢁ 100) and as a percentage inhibition
of control specific activity (100 ꢀ ((measured specific activity/control
specific activity) ꢁ 100)) obtained in the presence of test compound).
From the concentration curves, IC50 values were determined by non-
linear regression analysis using Hill equation curve fitting.
Guinea Pig QTc Studies. Guinea pigs were, at least 7 days prior to
the study, equipped with a telemetry transmitter. Male DunkinꢀHartley
guinea pigs, weighing 350ꢀ450 g at arrival, were allowed to acclimatize
during, at least, 5 days. On the day of surgery, the animals were
anesthetized with O2/N2O/isoflurane (0.8 L/min/0.8 L/min/3% as
induction and 0.8 L/min/0.8 L/min/1.75% as maintenance). A tele-
metry transmitter (TA11CTA-F-40) was implanted into the peritoneal
cavity under sterile conditions. Two electrodes were subcutaneously
tunneled to the breast of the animals and attached to the muscle tissue in
a lead II configuration, one near the apex of the heart and a second one in
the right armpit. Just before and 24 h after surgery, the animals received a
subcutaneous dose (7.5 mg per animal) of Baytrill. On the day of the
study, three guinea pigs received a subcutaneous or oral dose of the test
compound for determining electrocardiogram (ECG) parameters (QT,
QTc, and HR). After the animals were transferred from the holding to
the test room, they were individually placed in the test cages and allowed
to acclimatize for at least 30 min. The test cages were placed on a Data
Science International receiver (RA1010), and data acquisition took place
by Spike2, version 6 (Cambridge Electronic Design Ltd., Cambridge,
U.K.). The ECG signal was checked for suitability after the acclimatization
period. After this acclimatization period, a predrug ECG was recorded for
at least 30 min. Following the stabilization period, the animals were dosed
with test compound or vehicle. During the dosing procedure, the animals
were shortly removed from the cages and no ECG signal was recorded.
Once the animals were placed back into the test cages, ECG recording
began for 120 min. Subsequently the animals were removed from the test
cages and returned to their holding cage.
Molecular Modeling. All modeling studies were performed on a
Linux workstation equipped with two 64-bit 2.2 MHz dual-core AMD
Opteron processors. As the majority of the modeling work presented
here was performed well before the release of high-resolution crystal
structures of human GPCRs,81 homology modeling was based on the
bovine rhodopsin X-ray structure. The initial homology model of the
5-HT6R was generously provided by the group of Prof. Leonardo Pardo
and differs from the bovine rhodopsin X-ray structure (1U19)82 by
simulated Pro, Ser, and Thr kinks in the α helices.83 This model was
further optimized with the Schr€odinger modeling package,84 version 9.0,
using the OPLS2005 force field. To this end, structurally diverse
Per animal, the ECG signal was analyzed for both the stabilization
period (30 min) and the compound or vehicle period (120 min). The
ECG parameters were evaluated at 300 s intervals. At each time point,
five RR intervals and five QT intervals were measured. The following
variables from the lead II ECG were determined: RR interval; QT
interval; QTc interval (QTc = QT/(RR)1/2, Bazett’s correction88);
heart rate (based on RR interval). The following calculations were
performed based on the ECG data: (1) Per 300 s interval time point the
mean of five consecutive ECG complexes (RR, QT, QTc, and HR) were
7050
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
calculated. (2) An overall mean predrug value was estimated per ECG
parameter. The overall mean predrug value was calculated using mean
values calculated per 5 min interval each based on 5 consecutive ECG
complexes. (3) Per 300 s interval time point the difference between QTc
and mean predrug was calculated (Δ QTc). (4) The effect of the test
substance on QTc was expressed as AUC120 and Emax. The AUC per
time point was determined by multiplying the mean of two consecutive
Δ QTc values by 5 (=300 s interval expressed as minute). The AUC120
was calculated by summing all AUCs. The Emax was the highest Δ QTc
value after test substance treatment. (5) For each time point, the HR was
expressed as percentage relative to the predrug value.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: +31-36-5368020. E-mail: arnold.vanloevezijn@online.
nl. Address: Salsastraat 179, 1326 LW Almere, The Netherlands.
Present Addresses
†Astra Tech Benelux B.V., Signaalrood 55, 2718 SG Zoetermeer,
The Netherlands.
‡Lipid Nutrition B.V., Hogeweg 1, 1521 AZ Wormerveer, The
Netherlands.
Rat Pharmacokinetics. Preliminary kinetic data in rats were
gathered by administration of 1 mg/kg intravenously and 10 mg/kg
orally or intraperitoneally to Wistar rats. A group of 10 animals received
1 mg/kg intravenously (2 mL/kg, formulated in 25% 2-hydroxypropyl-
β-cyclodextrin (HPCD)). Blood samples were taken at time points 10,
30, 60, 180, and 420 min after dosing, two animals per time point, and at
1440 min after dosing (same two animals sampled at 10 min). After
wash-out, the same 10 animals received 10 mg/kg orally (10 mL/kg,
formulated in 1% methylcellulose (MC)). Blood and brain samples were
taken at 30, 60, 180, 420, and 1440 min after dosing. Another group of 10
animals received 10 mg/kg intraperitoneally (2 mL/kg, formulated in
1% MC + 5% mannitol). Blood and brain samples were taken at 30,
60, 180, 420, and 1440 min after dosing. Plasma samples, after liquidꢀ
liquid extraction, were analyzed using a generic bioanalytical method on
a TurboFlow MS/MS system. Brain samples were analyzed on the same
system after homogenization and solid-phase extraction. Levels in the
samples were calculated from a concentration versus response curve
obtained from spiked blank matrix samples, which were processed and
analyzed in the same way as the samples (extracted calibration curve).
The CNS/plasma ratio was estimated using the plasma (ng/mL) over
brain (ng/g) concentration over the entire oral dosing time range.
Determination of Fraction Unbound Brain. A literature
procedure89 was followed with modifications. Rat brain was homoge-
nized 1:6 with phosphate-buffered saline (PBS) and spiked with 1 μM
test compound; 300 μL of the spiked brain homogenate was dialyzed
against 500 μL of PBS in the rapid equilibrium dialysis device for 6 h at
room temperature while shaking. After that time point, an aliquot of 60
μL was taken from each well; 60 μL of PBS was added to the wells
containing brain homogenate, and 60 μL of brain homogenate was
added to the wells containing PBS to adjust the matrices for analysis.
Ethanol (240 μL/well) was used for quenching, and samples were frozen
until HPLC/MSꢀMS analysis to measure the AUC of the parent. The
fraction of unbound brain was calculated as follows:
’ ACKNOWLEDGMENT
The authors thank Christiaan Salomons for LCꢀMS purity
analyses, Jan Hoogendoorn for chiral preparative HPLC separa-
tions, Gerard Vollebregt and Martin Haazelager for 5-HT6R
binding and functional assay data, Annemieke Rensink and Pieter
Spaans for hepatocyte assay data, Tiny Adolfs and Sander Vader
for guinea pig QTc data, Jessica Dijksman for rat PK data, Gisela
Backfisch and Renate Anthes (Abbott GmbH & Co. KG,
Ludwigshafen, Germany) for fraction unbound brain data, and
Huub Nievelstein for valuable pharmacological input.
’ ABBREVIATIONS USED
5-HT6R, 5-hydroxytryptamine (serotonin) 6 receptor;ACE-Cl, 1-
chloroethyl chloroformate;AChEI, acetylcholinesterase inhibitor;
AD, Alzheimer’s disease; ADME, absorption, distribution, meta-
bolism, and excretion; Bn, benzyl;Boc, tert-butoxycarbonyl;
CHO, Chinese hamster ovary; CIAS, cognitive impairment
associated with schizophrenia; CNS, central nervous system; CYP,
cytochrome P450; DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene;
DCM, dichloromethane; DiPEA, N,N-diisopropylethylamine;
DMF, N,N-dimethylformamide; DMSO, dimethylsulfoxide;
ECG, electrocardiogram; ELSD, evaporative light scattering
detector; GABA, γ-aminobutyric acid; GPCR, G-protein-cou-
pled receptor; hERG, human ether-a-go-go-related gene (Kv11.1
potassium ion channel); HPCD, 2-hydroxypropyl-β-cyclodextrin;
HPLC, high performance liquid chromatography; HTS, high
throughput screening; id, intraduodenal; ip, intraperitoneal; iv, in-
travenous; ivp, intravenoportal; LCꢀMS, liquid chromatographyꢀ
mass spectrometry; MC, methylcellulose; MtBE, methyl tert-
butyl ether; NMR, nuclear magnetic resonance; po, per os
(oral); PE, petroleum ether (boiling range 40ꢀ60 °C); PBS,
phosphate buffered saline; P-gp, P-glycoprotein; PK, pharmaco-
kinetics; PSA, polar surface area; QTOF, quadrupole time-of-
flight; SAR, structureꢀactivity relationship; SCX, strong cation
exchange; THF, tetrahydrofuran;TLC, thin layer chromatogra-
phy;TMS, tetramethylsilane; UV, ultraviolet
% free ¼ ðAUC buffer chamber=AUC brain homogenate chamberÞ
ꢁ 100
% bound ¼ 100 ꢀ % free
fu;homogenate ¼ 1 ꢀ ð% bound=100Þ
undiluted fu;brain ¼ ð1=DÞ=½ð1=fu;homogenate ꢀ 1Þ þ 1=Dꢂ
where D is the dilution factor for the brain homogenate.
’ REFERENCES
(1) Monsma, F. J., Jr.; Shen, Y.; Ward, R. P.; Hamblin, M. W.; Sibley,
D. R. Cloning and expression of a novel serotonin receptor with high affinity
for tricyclic psychotropic drugs. Mol. Pharmacol. 1993, 43, 320–327.
(2) Ruat, M.; Traiffort, E.; Arrang, J. M.; Tardivel-Lacombe, J.; Diaz,
J.; Leurs, R.; Schwartz, J. C. A novel rat serotonin (5-HT6) receptor:
molecular cloning, localization and stimulation of cAMP accumulation.
Biochem. Biophys. Res. Commun. 1993, 193, 268–276.
(3) Woolley, M. L.; Marsden, C. A.; Fone, K. C. 5-ht6 receptors.
Curr. Drug Targets: CNS Neurol. Disord. 2004, 3, 59–79.
(4) Davies, S.; Silvestre, J. S.; Guitart, X. Drug discovery targets:
5-HT6 receptor. Drugs Future 2005, 30, 479.
’ ASSOCIATED CONTENT
S
Supporting Information. Selectivity profiling of com-
b
pounds 6, 14, 42, and 47 (Cerep data); synthesis scheme,
experimental procedures, and analytical data for intermediate
71g; detailed experimental information for intermediates not
fully described in the Experimental Section. This material is
available free of charge via the Internet at http://pubs.acs.org.
7051
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
(5) King, M. V.; Marsden, C. A.; Fone, K. C. A role for the 5-HT1A
5-HT4 and 5-HT6 receptors in learning and memory. Trends Pharmacol.
Sci. 2008, 29, 482–492.
(6) Nichols, D. E.; Nichols, C. D. Serotonin receptors. Chem. Rev.
2008, 108, 1614–1641.
(7) Borsini, F.; Bordi, F.; Riccioni, T. 5-HT6 pharmacology incon-
sistencies. Pharmacol., Biochem. Behav. 2011, 98, 169–172.
(8) Roth, B. L.; Hanizavareh, S. M.; Blum, A. E. Serotonin receptors
represent highly favorable molecular targets for cognitive enhancement
in schizophrenia and other disorders. Psychopharmacology 2004, 174,
17–24.
(9) Mitchell, E. S.; Neumaier, J. F. 5-HT6 receptors: a novel target for
cognitive enhancement. Pharmacol. Ther. 2005, 108, 320–333.
(10) Upton, N.; Chuang, T. T.; Hunter, A. J.; Virley, D. J. 5-HT6
receptor antagonists as novel cognitive enhancing agents for Alzheimer’s
disease. Neurotherapeutics 2008, 5, 458–469.
,
(28) Keefe, R. S. E.; Malhotra, A. K.; Meltzer, H. Y.; Kane, J. M.;
Buchanan, R. W.; Murthy, A.; Sovel, M.; Li, C.; Goldman, R. Efficacy and
safety of donepezil in patients with schizophrenia or schizoaffective
disorder: significant placebo/practice effects in a 12-week, randomized,
double-blind, placebo-controlled trial. Neuropsychopharmacology 2008,
33, 1217–1228.
(29) Marder, S. R.; Fenton, W.; Youens, K. Cognition in schizo-
phrenia: the MATRICS initiative. Am. J. Psychiatry 2004, 161, 25.
(30) Dawson, L. A. The central role of 5-HT6 receptors in modulat-
ing brain neurochemsitry. Int. Rev. Neurobiol. 2011, 96, 1–26.
(31) Neale, A. C.; Jenkins, H.; Amend, D.; Lesem, M. A 14-day,
dose-escalation, double blind, randomized, placebo-controlled study of
SGS-518 in adult patients with schizophrenia. Neuropsychopharmacology
2005, 30, S54.
(32) Arnt, J.; Olsen, C. K. 5-HT6 receptor ligands and their
antipsychotic potential. Int. Rev. Neurobiol. 2011, 96, 141–161.
(33) Batel, P. Addiction and schizophrenia. Eur. Psychiatry 2000,
15, 115–122.
(34) Di Chiara, G.; Valentini, V.; Fenu, S. Behavioral and neuro-
chemical pharmacology of 5-HT6 receptors related to reward and
reinforcement. Int. Rev. Neurobiol. 2011, 96, 111–139.
(35) de Bruin, N. M. W. J.; van Loevezijn, A.; Wijnen, J.; Herremans,
A. H. J.; Kruse, C. G. Use of 5-HT6 Antagonists To Prevent Relapse into
Addiction. WO 2008/087123, 2008.
(36) van Gaalen, M. M.; Schetters, D.; Schoffelmeer, A. N. M.; de
Vries, T. J. 5-HT6 antagonism attenuates cue-induced relapse to cocaine
seeking without affecting cocaine reinforcement. Int. J. Neuropsycho-
pharmacol. 2010, 13, 961–965.
(37) de Bruin, N. M. W. J.; Kloeze, B. M.; McCreary, A. C. The
5-HT6 serotonin receptor antagonist SB-271046 attenuates develop-
ment and expression of nicotine-induced locomotor sensitization in
Wistar rats. Neuropharmacology 2011, 61, 451–457.
(11) Geldenhuys, W. J.; van der Schyf, C. J. Serotonin 5-HT6
receptor antagonists for the treatment of Alzheimer’s disease. Curr.
Top. Med. Chem. 2008, 8, 1035–1048.
(12) Fone, K. C. F. Anupdateontherole ofthe5-hydroxytryptamine6
receptor in cognitive function. Neuropharmacology 2008, 55, 1015–1022.
(13) Johnson, C. N.; Ahmed, M.; Miller, N. D. 5-HT6 receptor
antagonists: prospects for the treatment of cognitive disorders including
dementia. Curr. Opin. Drug Discovery Dev. 2008, 11, 642–654.
(14) Liu, K. G.; Robichaud, A. J. 5-HT6 antagonists as potential
treatment for cognitive dysfunction. Drug Dev. Res. 2009, 70, 145–168.
(15) Wilson, C.; Terry, A. V., Jr. Enhancing cognition in neurological
disorders: potential usefulness of 5-HT6 antagonists. Drugs Future 2009,
34, 969–975.
(16) Geldenhuys, W. J.; van der Schyf, C. J. The serotonin 5-HT6
receptor: a viable drug target for treating cognitive deficits in Alzheimer’s
disease. Expert Rev. Neurother. 2009, 9, 1073–1085.
(17) Rossꢀe, G.; Schaffhauser, H. 5-HT6 receptor antagonists as
potential therapeutics for cognitive impairment. Curr. Top. Med. Chem.
2010, 10, 207–221.
(38) Taylor, D.; McAskill, R. Atypical antipsychotics and weight
gain: a systematic review. Acta Psychiatr. Scand. 2000, 101, 416–432.
(39) Heal, D. J.; Smith, S. L.; Fisas, A.; Codony, X.; Buschmann, H.
Selective 5-HT6 receptor ligands: progress in the development of a novel
pharmacological approach to the treatment of obesity and related
metabolic disorders. Pharmacol. Ther. 2008, 117, 207–231.
(40) Heal, D.; Gosden, J.; Smith, S. The 5-HT6 receptor as a target
for developing novel antiobesity drugs. Int. Rev. Neurobiol. 2011, 96,
73–109.
(41) Holenz, J.; Mercꢁe, R.; Díaz, J. L.; Guitart, X.; Codony, X.;
Dordal, A.; Romero, G.; Torrens, A.; Mas, J.; Andaluz, B.; Hernꢀandez, S.;
Monroy, X.; Sꢀanchez, E.; Hernꢀandez, E.; Pꢀerez, R.; Cubí, R.; Sanfeliu, O.;
Buschmann, H. Medicinal chemistry driven approaches toward novel
and selective serotonin 5-HT6 receptor ligands. J. Med. Chem. 2005,
48, 1781–1795.
(18) Foley, A. G.; Murphy, K. J.; Hirst, W. D.; Gallagher, H. C.;
Hagan, J. J.; Upton, N; Walsh, F. S.; Regan, C. M. The 5-HT6 antagonists
SB-271046 and SB-399885 increase basal and learning induced hippo-
campal ncam polysialylation. Int. J. Neuropsychopharmacol. 2004, 7, 215.
(19) Foley, A. G.; Hirst, W. D.; Gallagher, H. C.; Barry, C.; Hagan,
J. J.; Upton, N.; Walsh, F. S.; Hunter, A. J.; Regan, C. M. The selective
5-HT6 receptor antagonists SB-271046 and SB-399885 potentiate
NCAM PSA immunolabeling of dentate granule cells, but not neuro-
genesis, in the hippocampal formation of mature Wistar rats. Neuro-
pharmacology 2008, 54, 1166–1174.
(20) de Foubert, G.; O’Neill, M. J.; Zetterstr€om, T. S. C. Acute onset
by 5-HT6-receptor activation on rat brain BDNF and Arc mRNA
expression. Neuroscience 2007, 147, 778–785.
(21) Johnson, C. N.; Roland, A.; Upton, N. New symptomatic
strategies in Alzheimer’s disease. Drug Discovery Today: Ther. Strategies
2004, 1, 13–19.
(42) Lꢀopez-Rodríguez, M. L.; Benhamꢀu, B.; de la Fuente, T.; Sanz,
A.; Pardo, L.; Campillo, M. A three-dimensional pharmacophore model
for 5-hydroxytryptamine6 (5-HT6) receptor antagonists. J. Med. Chem.
2005, 48, 4216–4219.
(22) Maher-Edwards, G.; Zvartau-Hind, M.; Hunter, A. J.; Gold, M.;
Hopton, G.; Jacobs, G.; Davy, M.; Williams, P. Double-blind, controlled
phase II study of a 5-HT6 receptor antagonist, SB-742457, in Alzheimer’s
disease. Curr. Alzheimer Res. 2010, 7, 374–385.
(43) Boess, F. G.; Monsma, F. J., Jr.; Sleight, A. J. Identification of
residues in transmembrane regions III and VI that contribute to the
ligand binding site of the serotonin 5-HT6 receptor. J. Neurochem. 1998,
71, 2169–2177.
(23) Wesozowska, A.; Jastrzebska-Wiesek, M. Behavioral pharma-
cology: potential antidepressant and anxiolytic properties. Int. Rev.
Neurobiol. 2011, 96, 49–71.
(24) Callahan, P. M.; Ilch, C. P.; Rowe, W. B.; Tehim, A. Character-
izationofthe selective5-HT6 receptorantagonist SB-271046 inbehavioral
models of cognition. Soc. Neurosci. Abstr. 2004, 776.19 (abstract).
(25) Comery, T. A.; Schechter, L. E. Method for the Treatment of
Cognitive Dysfunction. WO 2007/087151, 2007.
(44) Data from Thomson Reuters Integrity, http://www.thomson-
reutersintegrity.com/.
)
)
(45) Wu, Y.-J.; He, H.; Hu, S.; Huang, Y.; Scola, P. M.; Grant-Young,
K.; Bertekap, R. L.; Wu, D.; Gao, Q.; Li, Y.; Klakouski, C.; Westphal, R. S.
Identification of a potent and selective 5-HT6 antagonist: one-step synthesis
of (E)-3-(benzenesulfonyl)-2-(methylsulfanyl)pyrido[1,2-a]pyrimidin-4-
ylidenaminefrom2-(benzenesulfonyl)-3,3-bis(methylsulfanyl)acrylonitrile.
J. Med. Chem. 2003, 46, 4834–4837.
(46) Kim, H.-J.; Doddareddy, M. R.; Choo, H.; Cho, Y. S.; No, K. T.;
Park, W.-K.; Pae, A. N. New serotonin 5-HT6 ligands from common feature
pharmacophore hypotheses. J. Chem. Inf. Model. 2008, 48, 197–206.
(47) Ivachtchenko, A. V.; Golovina, E. S.; Kadieva, M. G.; Koryakova,
A. G.; Kovalenko, S. M.; Mitkin, O. D.; Okun, I. M.; Ravnyeyko, I. M.;
(26) Marcos, B.; Chuang, T. T.; Gil-Bea, F. J.; Ramirez, M. J. Effects
of 5-HT6 receptor antagonism and cholinesterase inhibition in models
of cognitive impairment in the rat. Br. J. Pharmacol. 2008, 155, 434–440.
(27) Mitchell, E. S. 5-HT6 receptor ligands as antidementia drugs.
Int. Rev. Neurobiol. 2011, 96, 163–187.
7052
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
Tkachenko, S. E.; Zaremba, O. V. Synthesis and biological study of
3-(phenylsulfonyl)thieno[2,3-e][1,2,3]triazolo[1,5-a]pyrimidines as po-
tent and selective serotonin 5-HT6 receptor antagonists. Bioorg. Med.
Chem. 2010, 18, 5282–5290.
(63) Jansma, A.; Zhang, Q.; Li, B.; Ding, Q.; Uno, T.; Bursulaya, B.;
Liu, Y.; Furet, P.; Gray, N.; Geierstanger, H. Verification of a designed
intromolecular hydrogen bond in a drug scaffold by nuclear magnetic
resonance. J. Med. Chem. 2007, 50, 5875–5877.
(48) Ivachtchenko, A. V.; Golovina, E. S.; Kadieva, M. G.; Koryakova,
A. G.; Kovalenko, S. M.; Mitkin, O. D.; Okun, I. M.; Ravnyeyko, I. M.;
Tkachenko, S. E.; Zaremba, O. V. Solution phase parallel synthesis of
substituted 3-phenylsulfonyl-[1,2,3]triazolo[1,5-a]quinazolines: selective
serotonin 5-HT6 receptor antagonists. J. Comb. Chem. 2010, 12, 445–452.
(49) Ivachtchenko, A. V.; Dmitriev, D. E.; Golovina, E. S.;Dubrovskaya,
E. S.; Kadieva, M. G.; Koryakova, A. G.; Kysil, V. M.; Mitkin, O. D.;
Tkachenko, S. E.; Okun, I. M.; Vorobiov, A. A. Synthesis of cycloalkane-
annelated 3-phenylsulfonyl-pyrazolo[1,5-a]pyrimidines and their evalua-
tion as 5-HT6 receptor antagonists. Bioorg. Med. Chem. Lett. 2010, 20,
2133–2136.
(50) Harris, R. N., III; Stabler, R. S.; Repke, D. B.; Kress, J. M.;
Walker, K. A.; Martin, R. S.; Brothers, J. M.; Ilnicka, M.; Lee, S. W.;
Mirzadegan, T. Highly potent, non-basic 5-HT6 ligands. Site mutagen-
esis evidence for a second binding mode at 5-HT6 for antagonism.
Bioorg. Med. Chem. Lett. 2010, 20, 3436–3440.
(51) Aronov, A. M. Predictive in silico modeling for hERG channel
blockers. Drug Discovery Today: BIOSILICO 2005, 10, 149–155.
(52) Farid, R.; Day, T.; Friesner, R. A.; Pearlstein, R. A. New insights
about HERG blockade obtained from protein modeling, potential
mapping, and docking studies. Bioorg. Med. Chem. 2006, 14, 3160–3173.
(53) Imai, Y. N.; Ryu, S.; Oiki, S. Docking model of drug binding to the
human ether-a-go-go potassium channel guided by tandem dimmer mutant
patch-clamp data: a synergistic approach. J. Med. Chem. 2009, 52, 1630–1638.
(54) Kongsamut, S.; Kang, J.; Chen, X.-L.; Roehr, J.; Rampe, D. A
comparison of the receptor binding and HERG channel affinities for a
series of antipsychotic drugs. Eur. J. Pharmacol. 2002, 450, 37–41.
(55) Silvestre, J. S.; Prous, J. R. Comparative evaluation of potassium
hERG channel blockade by antipsychotics. Methods Find. Exp. Clin.
Pharmacol. 2007, 29, 457–465.
(64) Flack, H. D. On enantiomorph-polarity estimation. Acta Crys-
tallogr., Sect. A 1983, 39, 876–881.
(65) Hooft, R. W. W.; Straver, L. H.; Spek, A. L. Determination of
absolute structure using Bayesian statistics on Bijvoet differences. J. Appl.
Crystallogr. 2008, 41, 96–103.
(66) MetaSite, version 2.7.5; Molecular Discovery Ltd.: Perugia, Italy,
2006. For more background, see the following: Cruciani, G.; Carosati,
E.; De Boeck, B.; Ethirajulu, K.; Mackie, C.; Howe, T.; Vianello, R.
MetaSite: understanding metabolism in human cytochromes from the
perspective of the chemist. J. Med. Chem. 2005, 48, 6970–6979.
(67) http://www.cyprotex.com/cloescreen/in-vitro-metabolism/
cytochrome-p450-reaction-phenotyping/.
(68) Carsky, P.; H€unig, S.; Stemmler, I.; Scheutzow, D. Two step
redox systems, XXVII. Vinylogous bipyridyls and biquinolyls; syntheses
and UV/VIS-spectra. Liebigs Ann. Chem. 1980, 2, 291–304.
(69) van de Waterbeemd, H.; Camenisch, G.; Folkers, G.; Chretien,
J. R.; Raevsky, O. A. Estimation of bloodꢀbrain barrier crossing of drugs
using molecular size and shape, and H-bonding descriptors. J. Drug
Targeting 1998, 6, 151–165.
(70) Tsai, T.-Y.; Shia, K.-S.; Liu, H.-J. A facile method for the
construction of highly substituted acetonitriles and olefins. Malononi-
triles as acetonitrile carbanion and alkylidene dianion equivalents. Synlett
2003, 97–101.
(71) Diez-Barra, E.; de la Hoz, A.; Moreno, A.; Sanchez-Verdu, P.
Phase transfer catalysis without solvent. Synthesis of cycloalkane-1,1-
dicarbonitriles and alkanetetracarbonitriles. J. Chem. Soc., Perkin Trans. 1
1991, 4, 2593–2596.
(72) Briggs, C. B. A.; Newington, I. M.; Pitt, A. R. Synthesis and
properties of some novel nonionic polyol surfactants. J. Chem. Soc.,
Chem. Comm. 1995, 3, 379–380.
(56) Lange, J. H. M.; Coolen, H. K. A. C.; van Stuivenberg, H. H.;
Dijksman, J. A. R.; Herremans, A. H. J.; Ronken, E.; Keizer, H. G.;
Tipker, K.; McCreary, A. C.; Veerman, W.; Wals, H. C.; Stork, B.;
Verveer, P. C.; den Hartog, A. P.; de Jong, N. M. J.; Adolfs, T. J. P.;
Hoogendoorn, J.; Kruse, C. G. Synthesis, biological properties, and
molecular modeling investigations of novel 3,4-diarylpyrazolines as
potent and selective CB1 cannabinoid receptor antagonists. J. Med.
Chem. 2004, 47, 627–643.
(57) Grosscurt, A. C.; van Hes, R.; Wellinga, K. 1-Phenylcarbamoyl-
2-pyrazolines, a new class of insecticides. 3. Synthesis and insecticidal
properties of 3,4-diphenyl-1-phenylcarbamoyl-2-pyrazolines. J. Agric.
Food. Chem. 1979, 27, 406–409.
(58) In an effort to measure the basicity and acidity of the sulfo-
nylguaninde core moiety in 6, no pKa could be determined experimen-
tally between pH 2 and pH 12, confirming calculations previously
carried out with hit structure 1. Consequently, the pH-metric log P of
6 could not be determined. Alternatively, the HPLC log D was deter-
mined to be 2.4 at Sirius (http://www.sirius-analytical.com), being in
reasonable agreement with the AlogP of 2.73 as calculated with Pipeline
Pilot (http://accelrys.com).
(59) Currents were measured by using the whole-cell variant of the
patch clamp method according to the method described in the follow-
ing: Crumb, W.; Llorca, P.-M.; Lancon, C.; Thomas, G. P.; Garay, R. P.;
Hameg, A. Effects of cyamemazine on hERG, INa, ICa, Ito, Isus and IK1
channel currents, and on the QTc interval in guinea pigs. Eur. J.
Pharmacol. 2006, 532, 270–278.
(73) Bitha, P.; Child, R. G.; Hlavka, J. J.; Lin, Y.-I. Platinum
Complexes of Amines with Dibasic Acids. U.S. Patent 4,675,336, 1987.
(74) Stroefer, E.; Rohr, W.; Rotermund, G. W.; Fischer, R.; Neudert,
R. Preparation of 1-Pyrazolines. U.S. Patent 4,788,293, 1988.
(75) http://www.cyprotex.com/cloescreen/protein_binding/plasma-
protein-binding/.
(76) The anesthetized guinea pig model is thought to be a predictive
model for ECG changes in man, particularly that on QTc interval
changes: (a) Hey, J. A.; Pardo, M. D.; Sherwood, J.; Kreutner, W.; Egan,
R. W. Comparative analysis of the cardiotoxic proclivities of second
generation antihistamines in an experimental model of adverse clinical
ECG effects. Arzneim. Forsch. 1996, 46, 153–158. (b) Hey, J. A.; Pardo,
M. D.; Kreutner, W.; Egan, R. W. Cardiotoxic and drug interaction
profile of the second generation antihistamines ebastine and terfenadine
in an experimental animal model of torsade de pointes. Arzneim. Forsch.
1996, 46, 159–163. In the meantime it has been shown that the use of
telemetry transducers enables investigation of potential test substance
effect in the absence of anesthesia in freely moving animals.
(77) Read, K. D.; Braggio, S. Assessing brain free fraction in early
drug discovery. Expert Opin. Drug Metab. Toxicol. 2010, 6, 337–344.
(78) de Bruin, N. M. W. J.; Prickaerts, J.; van Loevezijn, A.; Venhorst,
J.; de Groote, L.; Houba, P.; Reneerkens, O.; Akkerman, S.; Kruse, C. G.
Two novel 5-HT6 receptor antagonists ameliorate scopolamine-induced
memory deficits in the object recognition and object location tasks in
Wistar rats. Neurobiol. Learn. Mem. 2011, 96, 392–402.
(79) Spek, A. L. Single-crystal structure validation with the program
PLATON. J. Appl. Crystallogr. 2003, 36, 7–13.
(60) http://www.cerep.fr/Cerep/Users/index.asp.
(80) Schinkel, A. H.; Wagenaar, E.; van Deemter, L.; Mol, C. A.;
Borst, P. Absence of the mdr1a P-glycoprotein in mice affects tissue
distribution and pharmacokinetics of dexamethasone, digoxin, and
cyclosporin A. J. Clin. Invest 1995, 96, 1698–1705.
(81) Cherezov, V.; Rosenbaum, D. M.; Hanson, M. A.; Rasmussen,
S. G.; Thian, F. S.; Kobilka, T. S.; Choi, H. J.; Kuhn, P.; Weis, W. I.;
Kobilka, B. K.; Stevens, R. C. High-resolution crystal structure of an
(61) Topiol, S.; Sabio, M. X-ray structure breakthroughs in the
GPCR transmembrane region. Biochem. Pharmacol. 2009, 78, 11–20.
(62) For a comprehensive consideration of the effects of intramole-
cular hydrogen bonding on conformational restrictions in targetꢀligand
interactions and on ligand physicochemical properties see the following:
Kuhn, B.; Mohr, P.; Stahl, M. Intramolecular hydrogen bonding in
medicinal chemistry. J. Med. Chem. 2010, 53, 2601–2611.
7053
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054

Journal of Medicinal Chemistry
ARTICLE
engineered human β2-adrenergic G protein-coupled receptor. Science
2007, 318, 1258–1265.
(82) Okada, T.; Sugihara, M.; Bondar, A. N.; Elstner, M.; Entel, P.;
Buss, V. The retinal conformation and its environment in rhodopsin in
light of a new 2.2 Å crystal structure. J. Mol. Biol. 2004, 342, 571–583.
(83) Deupi, X.; Olivella, M.; Govaerts, C.; Ballesteros, J. A.; Campillo,
M.; Pardo, L. Ser and Thr residues modulate the conformation of pro-
kinked transmembrane alpha-helices. Biophys. J. 2004, 86, 105–115.
(84) http://www.schrodinger.com.
(85) Boess, F. G.; Monsma, F. J., Jr.; Meyer, V.; Zwingelstein, C.;
Sleight, A. J. Interaction of tryptamine and ergoline compounds with
threonine 196 in the ligand binding site of the 5-hydroxytryptamine6
receptor. Mol. Pharmacol. 1997, 52, 515–523.
(86) de la Fuente, T.; Martín-Fontecha, M.; Sallander, J.; Benhamꢀu,
B.; Campillo, M.; Medina, R. A.; Pellissier, L. P.; Claeysen, S.; Dumuis,
A.; Pardo, L.; Lꢀopez-Rodríguez, M. L. Benzimidazole derivatives as new
serotonin 5-HT6 receptor antagonists. Molecular mechanisms of recep-
tor inactivation. J. Med. Chem. 2010, 53, 1357–1369.
(87) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible docking.
J. Mol. Biol. 1997, 267, 727–748.
(88) Bazett, H. C. An analysis of the time relations of the electro-
cardiograms. Heart 1920, 7, 353–370.
(89) Liu, X.; Van Natta, K.; Yeo, H.; Vilenski, O.; Weller, P. E.;
Worboys, P. D.; Monshouwer, M. Unbound drug concentration in brain
homogenate and cerebral spinal fluid at steady state as a surrogate for
unbound concentration in brain interstitial fluid. Drug Metab. Dispos.
2009, 37, 787–793.
7054
dx.doi.org/10.1021/jm200466r |J. Med. Chem. 2011, 54, 7030–7054