Discovery of 7-arylsulfonyl-1,2,3,4, 4a,9a-hexahydro-benzo[4,5]furo[2,3-c] pyridines: Identification of a potent and selective 5-HT6 receptor antagonist showing activity in rat social recognition test

Article abstract of DOI:10.1016/j.bmcl.2011.12.026

Serotoninergic neurotransmission has been implicated in modulation of learning and memory. It has been demonstrated that 5-hydroxytryptamine ₆ (5-HT₆) receptor antagonists show beneficial effect on cognition in several animal models. Based on a pharmacophore model reported in the literature, we have designed and successfully identified a 7-benzenesulfonyl-1,2,3,4-tetrahydro-benzo[4,5]furo[2,3-c]pyridine (3a) scaffold as a novel class of 5-HT₆ receptor antagonists. Despite good activity against 5-HT₆ receptor, 3a exhibited poor liver microsome stability in mouse, rat and dog. It was demonstrated that the saturation of the double bond of the tetrahydropyridine ring of 3a enhanced metabolic stability. However the resulting compound, 4a (7-phenylsulfonyl-1,2,3,4,4a,9a-hexahydro- benzo[4,5]furo[2,3-c] pyridine-HCl salt) exhibited ～30-fold loss in potency along with introduction of two chiral centers. In our optimization process for this series, we found that substituents at the 2 or 3 positions on the distal aryl group are important for enhancing activity against 5-HT₆. Separation of enantiomers and subsequent optimization and SAR with bis substituted phenyl sulfone provided potent 5-HT₆ antagonists with improved PK profiles in rat. A potent, selective 5-HT₆R antagonist (15k) was identified from this study which showed good oral bioavailability (F = 39%) in rat with brain penetration (B/P = 2.76) and in vivo activity in a rat social recognition test.

Full text of DOI:10.1016/j.bmcl.2011.12.026

Bioorganic & Medicinal Chemistry Letters 22 (2012) 1421–1426
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of 7-arylsulfonyl-1,2,3,4, 4a,9a-hexahydro-benzo[4,5]furo[2,3-c]pyri-
dines: Identification of a potent and selective 5-HT6 receptor antagonist
showing activity in rat social recognition test
⇑
Rabindranath Tripathy , Robert J. McHugh, Edward R. Bacon, Joseph M. Salvino, George C. Morton,
Lisa D. Aimone, Zeck Huang, Joanne R. Mathiasen, Amy DiCamillo, Mark J. Huffman, Beth A. McKenna,
Karla Kopec, Lily D. Lu, Jie Qian, Thelma S. Angeles, Thomas Connors, Chrysanthe Spais, Beverly Holskin,
Emir Duzic, Hervé Schaffhauser, Gerard C. Rossé
Worldwide Discovery Research, Cephalon, Inc., 145 Brandywine Parkway, West Chester, PA 19380, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Serotoninergic neurotransmission has been implicated in modulation of learning and memory. It has
been demonstrated that 5-hydroxytryptamine6 (5-HT6) receptor antagonists show beneficial effect on
cognition in several animal models. Based on a pharmacophore model reported in the literature, we have
designed and successfully identified a 7-benzenesulfonyl-1,2,3,4-tetrahydro-benzo[4,5]furo[2,3-c]pyri-
dine (3a) scaffold as a novel class of 5-HT6 receptor antagonists. Despite good activity against 5-HT6
receptor, 3a exhibited poor liver microsome stability in mouse, rat and dog. It was demonstrated that
the saturation of the double bond of the tetrahydropyridine ring of 3a enhanced metabolic stability. How-
ever the resulting compound, 4a (7-phenylsulfonyl-1,2,3,4,4a,9a-hexahydro-benzo[4,5]furo[2,3-c]pyri-
dine-HCl salt) exhibited ꢀ30-fold loss in potency along with introduction of two chiral centers. In our
optimization process for this series, we found that substituents at the 2 or 3 positions on the distal aryl
group are important for enhancing activity against 5-HT6. Separation of enantiomers and subsequent
optimization and SAR with bis substituted phenyl sulfone provided potent 5-HT6 antagonists with
improved PK profiles in rat. A potent, selective 5-HT6R antagonist (15k) was identified from this study
which showed good oral bioavailability (F = 39%) in rat with brain penetration (B/P = 2.76) and in vivo
activity in a rat social recognition test.
Received 26 October 2011
Revised 1 December 2011
Accepted 5 December 2011
Available online 13 December 2011
Keywords:
5-HT6 receptor antagonist
Cognition
Cognitive dysfunction
Pharmacophore model
Rat social recognition test
Ó 2011 Elsevier Ltd. All rights reserved.
Decline in cognitive performances such as learning and memory
are associated with aging process in elderly population and with
patients of Alzheimer’s disease (AD).1 Despite the fact that current
symptomatic treatment for memory deficits in AD includes cholin-
ergic and glutamatergic drugs, there is a need to discover new and
effective therapeutic based on novel mechanism of action. Among
the serotonin receptors, 5-hydroxytryptamine6 (5-HT6) receptor is
almost exclusively expressed in the central nervous system, espe-
cially in the areas associated with learning and memory.2 In recent
years several studies and reviews show that serotoninergic neuro-
transmission is implicated in modulation of learning and memory3
and particularly 5-HT6 receptor antagonists show beneficial effect
on cognition in several animal models.4 Therefore, design of 5-HT6
receptor (5-HT6R) antagonists has received considerable research
interest5 and some compounds have advanced to clinical trials.5a,b
Shown in Figure 1 are two important compounds in this area of
research. LY-433518 (SGS-518) was in phase I clinical trials for
cognition in schizophrenia which was discontinued.6 The second
compound, SB-742457 is in phase II clinical trial for AD.5a,b
In a program to utilize 5-HT6 antagonists for treating cognitive
dysfunction, we were particularly interested in the design and syn-
thesis of highly selective compounds with improved brain penetra-
tion, using lead literature compounds as references. A survey of the
literature reveals that majority of the 5-HT6R antagonists contain a
basic nitrogen tethered to an aromatic/hetero-aromatic segment.5
An interesting pharmacophore model based on 3D analysis of 45
structurally diverse 5-HT6 antagonists has been reported.7 The
model suggests (Fig. 1) a central aromatic/heterocyclic ring (AR)
flanked by a positive ionizable group (PI, i.e., a basic nitrogen)
and a strong, multiple hydrogen bond acceptor (HBA, cf. sulfonyl
group) attached to a hydrophobic group (HYD, such as phenyl
ring). Interestingly, a contrasting report shows exceptions to this
model with potent 5-HT6 antagonists bearing no basic nitrogen,8
which could be due to a novel mode of interaction with the recep-
tor. Our efforts to design 5-HT6R antagonist was based on the
⇑
Corresponding author. Tel.: +1 610 738 6137; fax: +1 610 738 6558.
E-mail address: rtripath@cephalon.com (R. Tripathy).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.12.026

1422
R. Tripathy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1421–1426
O
O
S
N
HYD
HBA
NH
N
AR
PI
N
O
S
5-HT6 Pharmacophore Model
O
O
N
PI = Positive ionizable group
AR = Aromatic or Het. scaffold
SB-742457
LY-483518
HBA = Hydrogen bond acceptor group
HYD = Hydrophobic site
NH2
NH
HO
O
NH.HCl
HO
O
H
S
N
N
H
1
O
H
R
2
3
NH.HCl
NH.HCl
O
O
O
S
O
O
O
H
S
3a
4a (cis, racemic)
h5-HT6R Ki = 83 nM
h5-HT6R Ki = 2.9 nM
LMS (t1/2 in min.) = <5 (m, r), 10 (d), >40 (h) LMS (t1/2 in min.) = >40 (m, r, d, h)
Oral BA = 7%
Oral BA = 15%
Figure 1. Lead 5-HT6 receptor antagonists, reported pharmacophoric model and our design strategy.
proposed pharmacophoric model nucleating from a tetrahydro-b-
carbolin (2) template.
H
H
a
NH
N-BOC
As shown in Figure 1, the template 2 is a serotonin (1) analog
where the basic nitrogen has been constrained into a cyclic struc-
ture to provide distinct conformational and binding properties. We
envisioned that appendage of an aryl/heteroaryl sulfone unit
(HBA + HYD) to the constrained template 2 (AR + PI) and concom-
itant replacement of the indole NH by an oxygen atom (to improve
brain penetration) would generate novel, potent and selective
5-HT6 antagonists. The proposed structural class is represented
by the chemotype 3 (Fig. 1). The specific attachment point of the
sulfone motif could be varied to fine-tune the relative orientation
of the distal aromatic group with respect to the basic nitrogen.
Subsequent synthesis and optimization showed that the 7-position
of the template 3 was optimal for 5-HT6 binding.9a The representa-
tive compound 3a was found to be extremely potent against hu-
man 5-HT6. However, 3a demonstrated unacceptable mouse (m),
rat (r) and dog (d) liver microsome stability (LMS) and also exhib-
ited poor pharmacokinetic (PK) results in rat (Fig. 1). Interestingly,
saturation of the double bond of the tetrahydropyridine ring re-
sulted in compound 4a (7-phenylsulfonyl-1,2,3,4,4a,9a-hexahy-
dro-benzo[4,5]furo[2,3-c] pyridine-HCl salt) which showed
dramatic improvement in LMS across the species (t1/2 >40 min)
including the human (h) microsomes. Oral bioavailability in rat
also improved (15%) with good plasma exposure. On the other
hand, apart from introduction of two chiral centers, the compound
4a also exhibited ꢀ30-fold loss in potency against human 5-HT6.
However, encouraged by the enhanced LMS data, 4a was selected
for further optimization to improve potency against 5-HT6. Herein,
we disclose the structure–activity relationships (SAR) of the distal
aromatic ring of 4a which resulted in generation of a highly potent
and selective 5-HT6R antagonist with an improved rat PK profile
and in vivo activity in rat social recognition model.9b
O
H
HO
O
H
O
(cis, racemic)
Not Isolated
7
6
72% in 3
steps
b
H
H
N-BOC
c
N-BOC
87%
O
H
TfO
O
H
HO
9
8
d
74%
H
H
N-BOC
N-BOC
e
O
60%
O
H
O
H
B
O
I
10
5
(cis, racemic)
f
H
H
N-BOC
N-BOC
and
O
H
O
H
I
I
5b
(Enantiomer 2)
5a
(Enantiomer 1)
Scheme 1. Reagents and conditions: (a) BBr3, DCM; (b) (i) Boc2O, NaOH (ii) KOH,
MeOH; (c) Tf2O, TEA, DMAP, DCM; (d) Pinacolborane, Pd(dppf)Cl2, Dppf, TEA,
Dioaxane, 100 °C; (e) NaI, Chloramine T, THF, H2O; (f) Chiralpak AD–H column
(21 ꢁ 250 cm), 35 °C and eluting with 30% MeOH:0.1% diethylamine (flow: 16 L/
min) and CO2 (36 mL/min by using a Berger Minigram SFC instrument.
The general synthesis of the target sulfones is outlined in
Schemes 1 and 2. The key iodo starting material 5 was prepared
as follows. 7-Methoxy-1,2,3,4,4a,9a-hexahydro-benzo[4,5]furo

R. Tripathy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1421–1426
1423
H
H
NH.HCl
N-BOC
b, c
a
5
O
H
S
90%
70%
O
H
S
O
O
4a
11
(cis, racemic)
(cis, racemic)
Scheme 2. Reagents and conditions: (a) Thiophenol, Argon, Na(OtBu), CuI, Neocuproine, Toluene, 115 °C; (b) mCPBA, DCM, 0 °C–rt; (c) 4 N HCl in dioxane/rt.
Table 1
5-HT6R binding and functional activity for the analogs of 4a
H
NH.HCl
R
O
S
H
O
O
Cis (racemic)
Compd
R
h5-HT6Ra (Ki, nM)
r5-HT6Ra (Ki, nM)
h5-HT6Rb cAMP-(IC50, nM)
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
H
2-F
2-Cl
2-Ph
3-F
3-Cl
3-Me
3-CF3
83
32
38
15
21
11
37
48
85
—
145
206
136
52
40
243
—
—
156
—
179
133
135
33
27
227
357
281
306
—
—
—
245
3-OMe
3-OCHMe2
3-O-(CH2)2-OCH3
3-O-(CH2)3-OCH3
4-F
23
562
203
385
64
—
—
236
4-Me
a
[3H]LSD binding in membranes prepared from cells expressing full length human or rat 5-HT6R. The Ki values are the mean of
duplicate values from three experiments.
b
Mean of three functional activity experiments from human astrocytoma cell line stably expressing the human 5-HT6R.
[2,3-c]pyridine (6)10 upon treatment with BBr3 provided the phe-
nol 7 which was treated in situ with BOC anhydride to generate
the N-protected compound 8. The target intermediate 10 was
made by converting 8 into a triflate (9) followed by a metal-cata-
lyzed boronylation reaction as shown in Scheme 1. The boronate
10 was converted to the iodo intermediate (5) by the treatment
of NaI in presence chloramines T as the oxidant.11 The enantiomers
of the iodo compound 5 were separated via chiral chromatography
on a chiralpak AD–H column using a Berger Minigram SFC instru-
ment (5a and 5b, Scheme 1).12 The target sulfones were then syn-
thesized in three steps. One example with thiophenol as the
coupling partner is shown in Scheme 2. Copper iodide-mediated
coupling of racemic 5 with thiophenol followed by oxidation of
the resulting sulfide (11) to the corresponding sulfone and subse-
quent deprotection provided the target 4a.
compound 4a. Moreover, 4e demonstrated <100 nM activities
against r5-HT6R and in the functional assay. As expected, the com-
pound 4f with a 3-chloro substituent was also found to be very
potent with a Ki of 11 nM against 5-HT6. Another notable example
among the substituents (4e–l) at 3-position is the 3-isoproxy
bearing analog 4j which showed comparable binding Ki (23 nM)
with respect to 4e. The 4-position (4m, 4n) appears to be of
Table 2
h5-HT6R binding and PK data for the enantiomers of 4e and 4j
H
NH.HCl
R
O
H
S
O
O
Table 1 lists the binding affinities (Ki) of the compounds at the
5-HT6 receptors which was determined in human or rat [3H] LSD
([N-methyl-3H]Lysergic acid, diethylamide) binding in membranes
prepared from cells expressing full-length human or rat 5-HT6.13
Also reported are IC50 values of functional activity experiments
from a human astrocytoma cell line stably expressing human
5-HT6.13 We first examined the effect of halogen and other
substituents at each positions of the phenyl-sulfone moiety. The
compound bearing a 2-fluoro group (4b) demonstrated >2-fold
enhancement in potency against h5-HT6 over the des-fluoro com-
pound (4a). Increasing the size to a 2-chloro (4c) or 2-phenyl (4d)
did not improve potency. On the other hand, 4e, with a 3-fluoro
substituent exhibited a fourfold increase in potency over the
Cis
A = Enantiomer 1
B = Enantiomer 2
a
Compd
R
A/B
h5-HT6 (Ki, nM)
Oral BA%
Oral AUCb
4e–1
4e–2
4j–1
4j–2
3-F
3-F
3-OCHMe2
3-OCHMe2
A
B
A
B
11
85
17
31
<1
2
9
0
434
637
310
26
[3H]LSD binding in membranes prepared from cells expressing full length
human 5-HT6R. The Ki values are the mean of duplicate values from three
a
experiments.
b
ng/h/mL.

1424
R. Tripathy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1421–1426
SH
NH2
NH2
a
b
O
HO
58%
O
O
36%
O
O
12
13
14
O
H
c, d
NH. HCl
5a
O
S
O
H
O
O
67% in 3 steps
15k
(from enantiomer 1)
Scheme 3. Reagents and conditions: (a) Isopropanol, DEAD, Ph3P, THF; (b) (i) NaNO2, HCl, 0 °C (ii) Potassium ethyl xanthate, water, 65 °C (iii) 1 N NaOH, EtOH, reflux; (c) CuI,
NaOtBu, Neocuproine, DMF, 90 °C; (d) (i) mCPBA, DCM, 0 °C–rt (ii) 4 N HCl in dioxane.
H
methane and BOC-deprotection with 4 N HCl in dioxane to the
sulfone 15k.
NBOC
a
b
Si SH
O
H
Si
S
Li
S
Alternatively, target compounds 15g could be generated via a
reverse coupling between a triisopropylsilyl (TIPS) protected thiol
intermediate 18 and an aromatic bromo/iodo compound via Cu
mediated coupling14 followed by oxidation of the resulting sulfide
(19) and subsequent deprotection (Scheme 4). Compound 18 was
prepared from the iodo intermediate 5a and lithium salt of the
TIPS-SH via a Pd(0) mediated coupling15 (Scheme 4).
Si
5a
18
80%
16
~0.5M
c
17
O
H
O
H
NH.HCl
NBOC
Table 3 shows the binding results of the bis-substituted com-
pounds. In contrast to our previous observations, the enantiomer
2 of the 3,5-difluoro compound (15b) was found to be about sixfold
more potent than the enantiomer 15a. A similar trend was main-
tained for the enantiomeric pairs of 15c and 15d. Highly potent
compounds were obtained with 2,3-dichloro compounds 15e and
15f with single digit nanomolar activity (5 and 2 nM, respectively).
For the enantiomer 15f, the IC50 value in the 5-HT6 functional assay
was found to be 6 nM. Despite high potencies, both of the com-
pounds (15e and 15f) were found to be have poor oral PK profiles
in rat (data not shown). Other enantiomeric pairs such as 15g/15h
and 15i/15j were found to be less active. The last pair of enantio-
mers 15k and 15l bearing 3-isopropoxy-5-methoxy substituents
demonstrated very interesting results. Unlike the trend observed
for the disubstituted compounds, the enantiomer 1 (15k) was
found to be more potent than the enantiomer 2 (15l). The com-
pound 15k showed good potencies against h5-HT6R (4 nM), r5-
HT6R (27 nM) and in the functional assay (67 nM). The opposite
enantiomer 15l was found to be less potent though in the function
assay the IC50 value was slightly better than 15k (Table 3).
The PK properties of the compounds 15k and 15l in rat are
shown in Table 4. After a single iv bolus dose of 1 mg/kg, com-
d
O
H
F
S
O
H
F
S
80% in 3 steps
O
O
15g
(Enantiomer 1)
19
Scheme 4. Reagents and conditions: (a) LiHMDS, Degassed dioxane; (b) Pd (PPh3)4,
Degassed dioxane, 60 °C; (c) 3-Fluoro-5-isopropoxy bromobenzene, CuI, Neocupr-
oine, CsF, DMF, 90 °C; (d) (i) 70% mCPBA, Dichloroethane (ii) 1 N HCl in dioxane.
limited tolerance for substitution and led to sharp decrease in
potency for 4m.
Using information gleaned from the above examples, we fo-
cused on the compounds 4e and 4j as representative compounds
for chiral separation and further evaluation. Each of the enantio-
mers of 5 (5a and 5b, Scheme 1) was coupled to the requisite
thiophenol. After subsequent oxidation and BOC deprotection
(via similar route as shown in Scheme 2 for the synthesis of race-
mic compounds), the enantiomerically pure compounds were
obtained. The binding and PK results are shown in Table 2. It was
interesting to find that the most potent enantiomers 4e–1 and
4j–1 (both made from 5a, the enantiomer 1 of 5) showed poor
rat PK results exhibiting <1% and 9% bioavailability, respectively.
Out of the two less potent enantiomers (4e–2 and 4j–2), the com-
pound 4j–2 showed good oral bioavailability in rat (Table 2).
The effect of an additional substituent on the sulfone-benzene
ring was investigated as an avenue to enhance potency and selectiv-
ity. The requisite coupling partners for the enantiomers of 5a or 5b
were made from the corresponding substituted benzene precursors
which in turn were converted to thiols via standard literature
protocols. One example is shown in Scheme 3. 3-Amino-5-meth-
oxy-phenol (12) was alkylated with isopropanol and the intermedi-
ate 3-isopropoxy-5-methoxy-phenylamine (13) was converted to
3-isopropxy-5-methoxy-benzenethiol (14) via diazotization fol-
lowed by thiol group introduction in three chemical steps. Finally,
5a was coupled with 14 in presence of NaOtBu, CuI and
neocuproine in DMF at 115 °C followed by the oxidation of the
corresponding sulfide with m-chloroperbenzoic acid in dichloro-
pound 15k exhibited a t of 2.5 h with moderate clearance rate
½
(41 mL/min/kg) and a high volume of distribution (4.3 L/kg). Oral
administration of a single 5 mg/kg dose showed oral bioavailability
of 39% based on 24 h time point.16 15k also showed good brain
penetration with a brain to plasma ratio (B/P) of 2.76. For compar-
ison, the clinical reference compound SB-742457 had an iv half life
of 0.7 h with comparable clearance (41 mL/min/kg) and volume of
distribution (5.7 L/kg). Oral bioavailability SB-742457 was found to
be 24% with B/P of 0.7. The other enantiomer 15l displayed a short
iv half life (0.4 h), with moderate plasma clearance (28 mL/min/kg)
and a high volume of distribution (4.0 L/kg). Oral administration of
a single 5 mg/kg dose of 15l showed a bioavailability of 11% with a
B/P of 2.29 (Table 4).
The serotonin subtype selectivity in the counter screen binding
assays for 15k was found to excellent demonstrating only 19%,

R. Tripathy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1421–1426
1425
Table 3
SAR of the disubstituted benze-sulfone analogs
R'
H
NH.HCl
O
H
S
R
O
O
(Cis)
A = Enantiomer 1
B = Enantiomer 2
Compd
R/R0
A/Ba
h5-HT6Rb (Ki, nM)
r5-HT6Rb (Ki, nM)
h5-HT6Rc cAMP-(IC50, nM)
15a
15b
15c
15d
15e
15f
15g
15h
15i
5-F/F
5-F/F
5-F/Cl
5-F/Cl
2-Cl/Cl
2-Cl/Cl
A
B
A
B
A
B
A
B
A
B
A
B
166
28
68
11
5
2181
174
—
—
—
—
—
—
—
1363
96
790
63
40
6
—
—
—
—
2
3-F/OiPr
3-F/OiPr
2-F/OiPr
2-F/OiPr
3-OMe /OiPr
3-OMe /OiPr
49
65
56
21
4
15j
15k
15l
—
27
55
67
45
12
a
b
c
Assignment of enantiomers 1 or 2 (A and B) was based the configuration of the precursors (5a or 5b) used. Enantiomer 1 (A) from 5a, Enantiomer 2 (B) from 5b.
[3H]LSD binding in membranes prepared from cells expressing full length human or rat 5-HT6R. The Ki values are the mean of duplicate values from three experiments.
Mean of three functional activity experiments from human astrocytoma cell line stably expressing the human 5-HT6R.
Table 4
Phamacokinetics data in rat for 15k and 15l
Compd
t
(iv)(h)a
Vd (L/kg)
CL (mL/min/kg)
F(po)b (%)
Brain/Plasma ratio (po)
½
15k
15l
2.5
0.4
4.3
1
41
28
39c
11
2.76
2.29
a
b
c
iv vehicle: 3% DMSO, 30% Solutol, 67% PBS.
po vehicle: Tween 80: propylene carbonate: propylene glycol (5:4:1).
Oral bioavailability was calculated based on 24 h time points. (see Ref. 16).
25%, 6%, 6% and 16% inhibition at 10 lM against human 1A, 3, 5A, 7
and rat 1B subtype receptors, respectively. Moreover, against
Table 5
Selectivity and drug-like properties of 15k
human serotonin receptor subtypes 2A, 2B and 2C, 15k showed
IC50 values of 3.7, 6.2 and 19 lM, respectively (Table 5). Based on
its excellent serotonin receptor subtype selectivity and overall PK
profile, compound 15k was selected for further characterization.
As shown in Table 5, it demonstrated excellent LMS profile
across the species (t1/2 >40 min. m, r, d, h). In the CEREP lead profile
testing, compound 15k was highly selective, showing >50% inhibi-
Solubility, g/mL (pH7.4, pH2.0
MDCK-MDR1 (Papp, A to B) cm/s. and PDR
Plasma protein binding % (r, h)
Liver microsome stability, t1/2, min. (m, r, d and h)
l
)
>130, >130
2.88 ꢁ 10ꢂ06, 3
97.3, 95.4
>40 all
>30, >30, >22.9, >30,
>5.9
CYP IC50
,
lM (1A2, 2C9, 2C19, 2D6, 3A4)
h5-HT Selectivity, % Inh. at 10
3, 5A, 7)
l
M (1A, 1Ba, 2A, 2B, 2C, 19, 16, 83, 62, 71, 25,
6, 6
h5-HT Selectivity, Ki,
hREG, IC50
lM (2A, 2B, 2C)
3.7, 6.2, 19
3.3
tion (apart from 5-HT 2A, 2B and 2C receptors) at 10
out of 95 tested targets (receptor IC50 values in
l
l
M for only 6
M: M1-2.2,
, lM
a
M2-7.9, M4-4.8, DA transporter-6.3, NA tranporter-4.4, Ca channel
L-3.9). The CYP IC50 was >20 M for all isoforms (1A2, 2C9, 2C19,
2D6) except for CYP3A4 (IC50 = 5.9 M). Compound 15k had good
Rat 5-HT1B receptor.
l
l
MDCK-MDR1 permeability with an apical (A) to basolateral (B) va-
lue of 2.88 ꢁ 10ꢂ6 cm/sec and permeability directional ratio (PDR)
of 3. 15k also showed good aqueous solubility (>130 and >130
also evaluated in the rat social recognition test for short term
memory. The social recognition test is based upon the ability of
an adult rat to recognize a younger conspecific during the second
of two sessions.1315k was evaluated at doses of 0.1–3.0 mg/kg fol-
lowing ip administration with a 120 min pretreatment time for
enhancement of short-term/working memory. Compound 15k
exhibited a dose dependent effect and the doses of 0.3, 1.0 and
3.0 mg/kg produced significant decreases (22%, 21% and 35%,
respectively; p < 0.05, ANOVA, Dunnett’s post hoc test) in the ratio
of investigation duration (RID) of a familiar juvenile rat (Fig. 2). The
control group that received 3.0 mg/kg 15k and presented with a
novel juvenile rat for trial 2 was not significantly different from
l
g/mL at pH7.4 and pH2.0 respectively) and plasma protein binding
(estimated to be 97% in rat and 95% in human). In a functional
patch clamp assay 15k showed a hERG IC50 of 3.3 M.
l
To estimate the degree of 5-HT6 receptor occupancy in vivo, a
receptor autoradiography binding assay13 was used to test the
ability of 15k to penetrate the blood brain barrier and bind to
the target receptor. After intraperitonial (ip) administration,
compound 15k produced a dose-dependent inhibition of ex vivo
binding of [125I] SB-258585 with an estimated IC50 value of
9.54 0.1.39 mg/kg ip (n = 5–6 animals for each dose). 15k was

1426
R. Tripathy et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1421–1426
3. (a) Buhot, M. C.; Martin, S.; Segu, L. Ann. Med. 2000, 32, 210; (b) Meneses, A.
Drug News Perspect. 2001, 14, 396; (c) Perez-Garcia, G.; Meneses, A. Behav. Brain
Res. 2008, 195, 17.
4. King, M. V.; Marsden, C. A.; Fone, K. C. Trends Pharmacol. Sci. 2008, 29, 482.
5. For reviews see: (a) Rossé, G.; Schaffhauser, H. Curr. Topics Med. Chem. 2010, 10,
207; (b) Ivachtchenko, A. V.; Ivanenkov, Y. A.; Tkachenko, S. E. Exper. Opin. Ther.
Patents 2010, 20, 1171; Also see: (c) Ivachtchenko, A. V.; Golovina, E. S.;
Kadieva, M. G.; Kysil, V. M.; Mitkin, O. D.; Tkachenko, S. E.; Okun, I. Bioorg.
Medchem. 2011, 19, 1482; (d) Reid, M.; Carlyle, I.; Caulfield, W. L.; Clarkson, T.
R.; Cusick, F.; Epemolu, O.; Gilfillan, R.; Goodwin, R.; Jaap, D.; O’Donnel, E. C.;
Presland, J.; Rankovic, Z.; Spinks, G.; Thomson, A. M.; Thomson, F.; Starin, J.;
Wishort, G. Bioorg. Medchem. Lett. 2010, 20, 3713; (e) Nirogi, R. V. S.;
Deshpande, A. D.; Kambhampati, R.; Badange, R. K.; Kota, L.; Daulatabad, A.
V.; Shinde, A. K.; Ahmad, I.; Kandikere, V.; Jayarajan, P.; Dubey, P. K. Bioorg. Med.
Chem. Lett. 2011, 21, 346; (f) Liu, K. G.; Lo, J. R.; Comery, T.; Zhang, G. M.; Zhang,
J. Y.; Kowal, D. M.; Smith, D. L.; Di, L.; Kerns, E. H.; Schechter, L. E.; Robichaud, A.
J. Bioorg. Medchem. Lett. 2009, 19, 1115; (g) Trani, G.; Baddeley, S. M.; Briggs, M.
A.; Chuang, T. T.; Deeks, N. J.; Johnson, C. N.; Khazaragi, A. A.; Mead, T. L.;
Medhurst, A. D.; Milner, P. H.; Quinn, L. P.; Ray, A. M.; Rivers, D. A.; Stean, T. L.;
Stemp, G.; Trail, B. K.; Witty, D. R. Bioorg. Medchem. Lett. 2008, 18, 5698; (h) Liu,
K. G.; Lo, J. R.; Comery, T.; Zhang, G. M.; Zhang, J. Y.; Kowal, D. M.; Smith, D. L.;
Di, L.; Kerns, E. H.; Schechter, L. E.; Robichaud, A. J. Bioorg. Medchem. Lett. 2009,
19, 3214; (i) Seong, C. M.; Park, W. K.; Park, C. M.; Kong, J. Y.; Park, N. S. Bioorg.
Medchem. Lett. 2008, 18, 738; (j) Ahmed, M.; Briggs, M. A.; Bromidge, S. M.;
Buck, T.; Campbell, L.; Deeks, N. J.; Garner, A.; Gorden, L.; Hamprecht, D. W.;
Holland, V.; Johnson, C. N.; Medhurst, A. D.; Mitchell, D. J.; Moss, S. F.; Powles,
J.; Seal, J. T.; Stean, T. O.; Stemp, G.; Thompson, M.; Trail, B.; Upton, N.;
Winborn, K.; Witty, D. R. Bioorg. Medchem. Lett. 2005, 15, 4867; (k) Fleuente, T.;
Martin-Fontecha, M.; Sallander, J.; Benhamu, B.; Camplillo, M.; Medina, R. A.;
Pellissier, L.; Claeysen, S.; Dumuis, A.; Pardo, L.; Lopez-Rodriguez, M. L. J. Med.
Chem. 2010, 53, 1357; (l) Hayder, S. N.; Yun, H.; Andrae, P. M.; Mattes, J.; Zhang,
J.; Kramer, J.; Smith, D. L.; Huselton, C.; Graf, R.; Aschmies, S.; Schechter, L. E.;
Comery, T. A.; Robichaud, A. J. J. Med. Chem. 2010, 53, 2521.
Figure 2. Results of rat social recognition test with compound 15k.
vehicle, suggesting the reduction in RID by 15k was relevant to an
enhancement of short-term/working memory.
In conclusion, a novel 7-arylsulfonyl-1,2,3,4-tetrahydro-ben-
zo[4,5]furo[2,3-c]pyridine scaffold (3) for 5-HT6 receptor antago-
nists was designed based on a reported phamacophore model.
We demonstrated that the saturation of the double bond of the tet-
rahydro-furopyridine ring of 3a enhanced metabolic stability of the
resulting compound 4a (7-phenylsulfonyl-1,2,3,4,4a,9a-hexahy-
dro-benzo[4,5]furo[2,3-c] pyridine-HCl salt) albeit with 30-fold
loss in potency and introduction of two chiral centers. In our at-
tempts to optimize this series (4a), we found that the 2 or 3 posi-
tion on the distal benzene ring are important for activity.
Separation of enantiomers and subsequent optimization and SAR
study with bis-substituted aryl-sulfones provided potent 5-HT6
antagonist with improved PK profiles in rat. A potent, selective
functional antagonist from this study, 15k showed good oral bio-
availability (F = 39%) with good brain penetration (B/P = 2.78) and
in vivo activity in rat social recognition test.
6. Neale, A. C.; Jenkins, H.; Amend, D.; Lesem, M. Neuropsychopharmacology 2005,
30, S54.
7. López-Rodróguez, M. L.; Bellinda Benhamú, B.; De la Fuente, T.; Sanz, A.; Pardo,
L.; Campillo, M. J. Med. Chem. 2005, 48, 4216.
8. 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. Bioorg. Med. Chem. Lett. 2010, 20, 2133.
9. (a) Unpublished results.; (b) The scope of the manuscript is limited to SAR on
the sulfone-benzene ring only. For a related work from Cephalon: Sundar, B. G.;
Bailey, T. R.; Dunn, D. D.; Bacon, E. A.; Salvino, J. M.; Morton, G. C.; Aimone, L.
D.; Huang, Z.; Mathiasen, J. R.; Dicamillo, A.; Huffman, M. J.; McKenna, B. A.;
Kopec, K.; Lu, L. D.; Brown, R.; Qian, J.; Angeles, T.; Connors, T.; Spais, C.;
Holskin, B.; Galinis, D.; Duzic, E.; Schaffhauser, H.; Rosse, G. C. Bioorg. Med.
Chem. Lett. (communicated).
10. (a) Morice, C.; Domostoj, M.; Briner, K.; Mann, A.; Suffert, J.; Wermuth, C.-G.
Tetrahedron. Lett. 2001, 42, 6499; (b) Bacon, E. R.; Bailey, T. R.; Dunn, D. D.;
Hostetler, G. A.; Mchugh, R. J.; Morton, G. C.; Rosse, G. C.; Salvino, J. M.; Sundar,
B. G.; Tripathy, R. WO 2011087712
Acknowledgements
11. (a) Kabalka, G. W.; Gooch, E. E. J. Org. Chem. 1981, 46, 2582; (b) Kabalka, G. W.;
Sastry, K. A. R.; Hsu, H. C.; Hylarides, D. M. J. Org. Chem. 1981, 46, 3113; (c)
Kabalka, G. W.; Akula, M. R.; Zhang J. Nucl. Med. Biol. 2002, 29, 841.
12. Absolute configurations of 5a and 5b have not been determined.
13. See Schaffhauser, H.; Mathiasen, J. R.; Dicamillo, A.; Huffman, M. J.; Lu, L. D.;
McKenna, B. A.; Qian, J.; Marino, M. J. Biochem. Pharmacol. 2009, 78, 1035.
14. Fernandez-Rodriguez, M. A.; Shen, Q.; Hartwig, J. F. J. Am. Chem. Soc. 2006, 128,
2180. A CuI mediated coupling was carried out inplace of Pd(OAc)2.
15. (a) Rane, A. M.; Miranda, E. I.; Soderquist, J. A. Tetrahedron Lett. 1994, 35, 3225;
(b) Miranda, E. I.; Diaz, M. J.; Rosado, I.; Soderquist, J. A. Tetrahedron Lett. 1994,
35, 3221; (c) Huang, Y.; Mahmood, K.; Simpson, N. R. J. Labelled Compd.
Radiopharm. 1998, 41, 9; (d) Arnould, J. C.; Didelot, M.; Cadilhac, C.; Pasquet, M.
J. Tetrahedron Lett. 1996, 37, 4523.
We thank Mr. Kurt Josef for his help with NMR experiments. We
also thank Mr. Mark Olsen for chiral separation of enantiomers and
Ms. Rebecca A. Brown for bio-analysis.
References and notes
1. (a) Cansino, S. Int. J. Psychophysiol. 2009, 71, 64; (b) Gazzaley, A.; Sheridan, M.
A.; Cooney, J. W.; D’Esposito, M. Neuropsychology 2007, 21, 532; (c) Germano,
C.; Kinsella, G. J. Neuropsychol. Rev. 2005, 15, 1; (d) Gold, C. A.; Budson, A. E.
Exper. Rev. Neurother. 2008, 8, 1879.
2. (a) Gerard, C.; Martres, M. P.; Lefevre, K.; Miquel, M. C.; Verge, D.; Lanfumey, L.;
Doucet, E.; Hamon, M. E.; Mestikawy, S. Brain Res. 1997, 746, 207; (b) Hamon,
M.; Doucet, E.; Lefevre, K.; Miquel, M. C.; Lanfumey, L.; Insausti, R.; Frechilla, D.;
Del, R. J.; Verge, D. Neuropsychopharmacology 1999, 21, 68S.
16. A 24 h time course was followed for the compound 15k due to slower oral
absorption at the usual 6 h data collection.