Design and synthesis of a novel series of histamine H3 receptor antagonists through a scaffold hopping strategy

Article abstract of DOI:10.1016/j.bmc.2014.12.036

Lead compounds 5-fluoro-2-methyl-N-[2-methyl-4-(2-methyl-[1,3′]bipyrrolidinyl-1′-yl)-phenyl]-benzamide (1), tetrahydro-pyran-4-carboxylic acid [((2S,3′S)-2-methyl-[1,3′]bipyrrolidinyl-1′-yl)-phenyl]-amide (2), and 3,5-dimethyl-isoxazole-4-carboxylic acid

Full text of DOI:10.1016/j.bmc.2014.12.036

Bioorganic & Medicinal Chemistry 23 (2015) 429–438
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design and synthesis of a novel series of histamine H3 receptor
antagonists through a scaffold hopping strategy
b,
Zhongli Gao a, , William J. Hurst , Daniel Hall b, Ryan Hartung b,à, William Reynolds b, Jiesheng Kang c,
⇑
Raisa Nagorny b, James A. Hendrix b,§, Pascal G. George d
a LGCR SMRPD Chemical Research, Sanofi US, 153-1-130, 153 2nd Ave, Waltham, MA 02451, USA
b Sanofi US, 55C-420A, 55 Corporate Drive, Bridgewater, NJ 08807, USA
c Proteogenomics, Sanofi/Genzyme R&D Center, 49 New York Avenue, Framingham, MA 01701, USA
d Sanofi R&D, 1, Avenue Pierre Brossolette, Chilly-Mazarin 91385, France
a r t i c l e i n f o
a b s t r a c t
Lead compounds 5-fluoro-2-methyl-N-[2-methyl-4-(2-methyl-[1,30]bipyrrolidinyl-10-yl)-phenyl]-benzam-
ide (1), tetrahydro-pyran-4-carboxylic acid [((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-yl)-phenyl]-amide
(2), and 3,5-dimethyl-isoxazole-4-carboxylic acid [((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-yl)-phe-
nyl]-amide (3) discovered in our laboratory, displayed high histamine H3 receptor (H3R) affinity, good
selectivity and weak human Ether-à-go-go-Related Gene (hERG) channel affinity with desirable overall
physico-chemical and pharmacokinetic (PK) profiles. Herein, we describe the design and synthesis
of a novel series of H3R antagonists utilizing a scaffold hopping strategy. Further structure–activity
relationship (SAR) studies of the series culminated in the identification of ((2S,30S)-2-methyl-
[1,30]bipyrrolidinyl-10-yl)-naphthalene-2-carboxylic acid (tetrahydro-pyran-4-yl)-amide (4c) and -[4-
((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-yl)-phenyl]-N-(tetrahydro-pyran-4-yl)-acetamide (4d), which
exhibited good H3R affinity in vitro, good selectivity, and desirable PK properties. Compounds 4c and
4d were also assessed in cardiac safety experiments. In particular, the effects of the compounds on action
potentials recorded from ventricular myocytes isolated from guinea pigs were used to screen compounds
that not only displayed a low affinity towards hERG channel, but also had lower interference with other
Article history:
Received 25 September 2014
Revised 8 December 2014
Accepted 16 December 2014
Available online 23 December 2014
Keywords:
Histamine H3 receptor antagonist/inverse
agonist
Structure activity relationship
Scaffold hopping
Cardiac myocyte model
cardiac ion channels. Compound 4c did not alter the major parameters in this model system at 610 lM,
and no significant induction of any major haemodynamic effect when intravenously administered at
3 mg/kg dose to anaesthetized mongrel dogs. Compound 4c is a new promising lead as orally potent
and selective H3R antagonist belonging to a distinct structural class.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
norepinephrine, serotonin, glutamate and others (negative modu-
lation).8,9 A key aspect of the unresolved puzzle of H3R pharmacol-
Since the discovery of the histamine H3 receptor (H3R),1
a
ogy is that, in addition to possible roles in cognition6 such as
Alzheimer’s disease, as well as sleep disorders, H3R ligands have
therapeutic potential in diseases and disorders such as attention
deficit hyperactivity disorder,10–13 schizophrenia,14 obesity,15,16
epilepsy,17 neuropathic pain,18 myocardial dysfunction19 and
others. This diversity of indications is difficult to rationalize with
current ideas regarding uniform ligand–receptor–G-protein inter-
actions. However, increasing evidence2,12 supports the complexity
of the CNS histamine system, including histamine-containing
neurons, H3R, interactions with signal transduction pathways and
the diversity of neurotransmitters whose syntheses and/or release
are modulated by H3R. Taken together, these and other findings
suggest the potential utility of H3R antagonists for the treatment
of central nervous system (CNS) diseases. Much effort has focused
G-protein coupled receptor, a wide variety of its functions for
health and disease have been elucidated.2–7 The receptor is localized
predominantly in the central and peripheral nervous systems. When
activated, the H3R can suppress the release of neurotransmitters
that include not only histamine but also dopamine, acetylcholine,
⇑
Corresponding author. Tel.: +1 781 434 3635.
E-mail addresses: zhongli.gao@sanofi.com (Z. Gao), william.hurst@sanofi.com
(W.J. Hurst), jhendrix@alz.org (J.A. Hendrix).
Tel.: +1 908 981 3723.
Current address: Combinatorial Chemistry 1, Sanofi US, 2090 East Innovation Park
à
Drive, Oro Valley, AZ 8575, USA.
§
Present address: Alzheimer’s Association, 225 N. Michigan Ave., 17th Floor,
Chicago, IL 60601, USA. Tel.: +1 212 568 0365.
http://dx.doi.org/10.1016/j.bmc.2014.12.036
0968-0896/Ó 2014 Elsevier Ltd. All rights reserved.

430
Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
on understanding H3R physiology and the discovery of increasingly
selective and potent H3R ligands with therapeutic potential.7,20,21
H3C
O
A
few H3R antagonists have entered into clinical development (for
review, see Refs. 7,20–22). However, no H3R antagonist compound
is approved for clinical use. There is still an unmet need to discover
novel, structurally diverse and highly selective H3R antagonists
with characterized therapeutic utility.
N
H
N
N
O
H3C
Earlier work from our group23–26 revealed a series of H3R antag-
onists/inverse agonists that displayed diverse pharmacological
profiles (Fig. 1): compound 1 exhibited oral activity in ob./ob.
mouse food intake inhibition model; compound 2 demonstrated
oral awakening effect in rat; compound 3 showed oral activity in
rat forced swim test (FST). Following these successes, the focus of
our H3R program was shifted towards the discovery of a structur-
ally diverse novel series. The new lead compound in the novel ser-
ies have to be comparable with the leads 1–3 in certain aspects,
such as high H3R affinity, good selectivity, acceptable human
Ether-à-go-go-Related Gene (hERG) channel selectivity and desir-
able overall physico-chemical, as well as pharmacokinetic (PK)
profiles but belong to a distinct and novel structural class.
Previous structure activity relationship studies24,25,27 revealed
that the bipyrrolidine moiety with 2S,30S stereochemistry was pre-
ferred. It would be optimal to keep the bipyrrolidine moiety
unchanged for the present objective. The novel series could be
derived through a scaffold hopping strategy, for example, by using
an ‘inverse amide’ to link the left-handed heterocyclic moiety
(designated as ‘HC’ in Fig. 2, structure 4) to the ‘Central core’ as
indicated in structure 4. The HC moiety and the central core can
be optimized simultaneously (Fig. 2). The rationale for such a
design was stemmed from the understanding of the previous SAR
studies and our pharmacophore hypothesis. The H3R ligands
should consistent of three components: the right-handed basic
center which are essential for H3R activity; the right-handed
heterocyclic (designated as ‘HC’) which are substantially enhancing
H3R activity; the Central core with proper lipophilicity and a linker
which provides the scaffold to hold the right-handed and the
left-handed moieties together in preferred angle and distance so
that H3R recognition and binding could be optimal. The design,
synthesis, and the pharmacological assessment of this novel series
of H3R ligands are thus described herein.
2
O
Central
core
N
HC
N
H
N
H3C
4
Figure 2. Design strategy for deriving a novel series of H3R ligands.
acid 6 to obtain amide 7. Intermediate 7 was then condensed with
(2S,30S)-2-methyl-[1,30]bipyrrolidinyl (8) (for synthesis, see Ref.
27) by using the reaction conditions adapted from the Buchwald
protocol28,29 to yield the desired amide analogs 4a–4e.
Compound 4e was further derivatized to 4f–4h following the
synthetic route illustrated in Scheme 2.
Thus, the t-butyl carbamate in compound 4e was removed
under acidic conditions to give compound 4f. The secondary amine
4f was converted into tertiary amine 4g by standard reductive
amination with cyclopentane-carboxaldehyde.
Compound 4i was synthesized in a different route, as shown in
Scheme 3. para-Bromo-benzoic acid t-butyl ester 9 was condensed
with
8
to obtain (2S,30S)-2-methyl-[1,30]bipyrrolidinyl-benzoic
acid t-butyl ester 10. The t-butyl protection group was then
removed by acidic hydrolysis (HCl in dioxane) to yield acid 11,
which was subsequently coupled with various amines 5 to obtain
the desired amide analog 4i (Table 1).
Compound 4j was synthesized similarly, as shown in Scheme 4.
4-Bromo-benzaldehy was condensed with piperidine-4-carboxylic
acid methyl ester to benzylamine 12. Simple ester–amide exchange
reaction with gave rise to amide 13. Finally, condensation with
8 yielded the desired compound 4j.
2. Results and discussion
Compounds 4a–4e (Table 1) were synthesized as highlighted in
Scheme 1. The synthesis began with the commercially available
amine 5, which was coupled with the substituted aryl carboxylic
H3C
O
H3C
O
CH3
N
H
N
N
H
N
N
O
N
H3C
H3C
F
2
1
H3C
O
H3C
N
N
H
N
N
O
CH3
H3C
3
Figure 1. Structures of H3R antagonists.

Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
431
Table 1
Histamine-3 receptor H3R affinity and calculated physico-chemical properties
R
N
N
4a-4j
No.
R
H3R binding
clogP
clogD7.4
MW
PSA
Rhesus
Ki (nM)
Rat
Ki (nM)
Mouse
Ki (nM)
(Å2)
45
2
—
0.8
1.0
10.0
2.2
1.7
ꢀ0.2
ꢀ0.7
371.52
371.52
H
N
O
4a
1.5 0.3
2.0 0.1
3.0 0.5
19.7 5.8
12.1 5.3
45
45
45
O
H
N
O
O
4b
4c
15.0 10.3
9.8 1.2
10.1 13.7
19.6 4.0
1.5
2.2
ꢀ0.9
ꢀ0.3
357.49
407.55
O
H
N
O
O
H
N
4d
2.5 0.1
8.3 2.8
10.8 1.0
2.5
0.1
371.52
45
O
O
O
H
N
N
4e
4f
2.8 0.1
0.8 0.2
19.4 0.1
1.1 0.1
3.4
1.9
0.9
470.65
370.53
65
48
O
H
N
HN
ꢀ2.1
O
H
N
O
N
4g
0.7 0.5
1.3 0.1
4.2
0.4
452.68
39
O
O
O
H
N
N
4h
4i
1.6 0.1
1.2 0.8
8.3 0.1
2.6
2.1
0.2
442.59
371.52
65
45
O
H
N
19.1 7.1
40.3 18.8
ꢀ0.3
O
O
N
4j
0.47
0.93
4.4
2.2
ꢀ1.1
384.56
39
N
H
Analogs 4a–4j were then tested in H3R binding assays. A corre-
lation between human and rhesus H3R binding affinity (see Ref.
24,25) seemed apparent. The project team took advantage of this
observation by using rhesus monkey H3R binding data as the first
tier screening for all the synthesized compounds. The choice of rhe-
sus monkey, rat, and mouse H3R binding assay as the first tier
screening, instead of human H3R binding assay in parallel, was
mainly due to the limited resources and aggressive timelines. Fur-
thermore, the rhesus monkey data would be useful for pharmaco-
logical assessment of the selected compounds in acute in vivo
models. Therefore, all the compounds synthesized were evaluated
in membranes isolated from a CHO cell line stably transfected with
the rhesus monkey H3 receptors (rh-H3R), rat (r-H3R) and mouse
(m-H3R). The results are listed in Table 1. The calculated physico-
chemical properties (clogP, clogD7.4, PSA) (calculated using
ACD/Labs methods) were used as a guide to select the optimal sub-
stituents for the ‘HC’ and the ‘Central core’. The in vitro ADME data
are given in Table 2. The metabolic stability was determined with
an in vitro assay using S9 fractions of human, mouse, and rat liver
microsomes. Values are expressed as a percent of parent compound
metabolized after 15 min incubation at 37 °C. The hERG affinity was
performed using a patch-clamp technique in the whole-cell config-
uration on Chinese hamster ovary (CHO) cells. The inhibition of
in H3R binding assays by displacement of [3H]N-
a-methylhistamine

432
Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
R3
R2
R2
N
HO
H
n
R4
n
R3
a
+
N
O
O
Br
Br
R4
6
7
5
R3
N
R2
n
R4
b
O
N
N
H3C
4a - 4e
Scheme 1. Syntheses of analogs 4a–4e. Reagents and conditions: (a) N-methylmorpholine, HOBT, EDCꢁHCl, CH2Cl2, DMF, rt, 16 h, 100% yield; (b) sodium tert-butoxide,
tris(dibenzylideneacetone)-dipalladium-(0), BINAP, toluene, (2S,30S)-2-methyl-[1,30]bipyrrolidinyl dihydrochloride (8), 85 °C (external), 15 h, 70–72% yield.
H
H
N
H
N
N
a
b
N
N
O
N
O
O
N
N
HN
N
O
N
N
O
H3C
H3C
H3C
4g
4e
4f
c
H
N
N
O
N
O
N
O
H3C
4h
Scheme 2. Syntheses of analogs 4f–4h. Reagents and conditions: (a) 4 N HCl in dioxane, thiophene, rt, overnight, 100% yield; (b) sodium triacetoxyborohydride,
cyclopentane-carboxaldehyde, DCE, rt, overnight, 56% yield; (c) ethyl chloroformate, THF/water (4:1, v/v), K2CO3, rt, overnight, 69% yield.
O
N
Br
N
a
b
O
O
H3C
O
10
9
O
O
c
N
N
N
N
N
H
HO
O
H3C
H3C
11
4i
Scheme 3. Synthesis of analog 4i. Reagents and conditions: (a) sodium tert-butoxide, tris(dibenzylideneacetone)-dipalladium-(0), BINAP, toluene, (2S,30S)-2-methyl-
[1,30]bipyrrolidinyl dihydrochloride (8), 85 °C (external), 15 h, 60%; (b) HCl (4 M/dioxane), dioxane, rt, overnight, 100%; (c) tetrahydropyran-4-yl amine, EDCꢁHCl, HOBt
N-methylmorpholine, CH2Cl2, DMF, rt, 16 h, 75–86% yield.

Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
433
N
b
Br
NH
a
O
H3C
H
O
+
H3C
Br
O
O
O
12
N
c
N
N
N
H3C
N
N
Br
H3C
O
O
H3C
13
4j
Scheme 4. Synthesis of analog 4j. Reagents and conditions: (a) sodium triacetoxyborohydride, DCE, HCl; 92% yield; (b) MeNH2, dixone, 50 °C, overnight, 46% yield; (c) sodium
tert-butoxide, tris(dibenzylideneacetone)-dipalladium-(0), BINAP, toluene, (2S,30S)-2-methyl-[1,30]bipyrrolidinyl dihydrochloride (8), 85 °C (external), 15 h, 53%.
hERG currents is expressed as percent inhibitions at 1.0
10.0 M concentration of the compounds tested.
The reversal of the amide’s connectivity maintained H3R affinity
when compared to its close analog 2 (4a vs 2). The compound also
exhibited good metabolic stability and hERG channel selectivity
l
M and
the introduction of a second basic center (4j) resulted in lowering
clogD7.4 to ꢀ1.1, which is undesirable for permeability and espe-
cially for CNS penetration.30–32 Indeed, the permeability was
decreased dramatically (Caco-2/TC7 of 9.1 ꢂ 10ꢀ7 cm/s) compared
to other analogs in the series.
l
(hERG channel percent inhibitions of 4% @ 1
l
M and 20% @
The selected compounds (4b, 4c, and 4d) were further profiled.
The physico-chemical properties were first measured and the
results are presented in Table 3. Compounds 4b and 4d showed
good solubility, while that for 4c was much lower. LogP and logD7.4
data were in a range of low, medium and high.
All three compounds were highly selective towards 33 GPCR
and ion channel biological targets, with a percentage of inhibition
of 30% or lower at 10 lM in the ligand binding assay. They were
10 M). The similar analog, 4b, displayed very comparable results,
l
confirming that this type of structure was worth further
exploration.
As a result of these positive findings, changes to the central core
were then explored. When the central core was extended by two
carbons (4c) or one carbon (4d), H3R affinity dropped slightly,
but not dramatically. However, clogP increased versus 4a and 4b,
signaling potentially better CNS penetration. Introduction of an
additional basic center (4f and 4g vs 4d) restored the H3R affinity
with comparable (4f) or substantially increased clogP (4g). How-
ever, clogD7.4 for 4f was as low as ꢀ2.1, implying the possibility
of a brain penetration issue. When the secondary amine was
masked by a carbamate (4h), clogP, clogD7.4, and PSA were
brought within a desirable range (clogP: 1.5–2.5; clogD7.4: 0.5–
2.0; PSA: 45–65)30–32 and with good H3R affinity. Insertion of a
methylene between the amine and tetrahydropyran (4i) main-
tained the rh-H3R affinity (4i vs 4b). However, the mouse H3R
affinity dropped by 4-fold, indicating the importance of distance
requirements for the hydrogen bond forming elements in the
mouse H3R. Using the amine instead of the amide linker (4j),
resulted in one of the most potent H3R compounds. Unfortunately,
negative in the Ames II and in vitro micronucleus tests (MNT). Nei-
ther the three compounds displayed cytochrome P450 induction
(human hepatocytes, n = 4; 1–60
CYP3A4), nor cytochrome P450 inhibition (all IC50 values
>100 M), indicating a low potential for drug–drug interactions.
lM for CYP1A1, CYP1A2, and
l
The in vitro permeability of 4b, 4c, and 4d was assessed by Caco-
2/TC7, with values of 111.1, 205.5, 80.8 (ꢂ10ꢀ7 cm/s), respectively
that predicted good intestinal absorption for all three compounds.
They were stable in S9 fractions and in plasma from several species
(human, guinea pig, rat, mouse, dog, monkey, sheep and rabbit)
(<5% metabolized when incubated with plasmas for 4 h at 37 °C).
In order to assess the hepatic metabolic liability, compounds 4c
and 4d were incubated at 37 °C with fresh Wistar rat and CF-1
mouse liver homogenized tissues at a concentration of 5.0 lM. At
the incubation time of 0, 0.33 and 2.0 hours (h), liver homogenates
(n = 3) were analyzed with LC/MS/MS to determine the remaining
percentage of parent compounds. In 2 h less than 15% of the com-
pound was metabolized in both rat and mouse, indicating a low
rate of first pass metabolism.
Table 2
In vitro Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET)
properties
No.
Caco-2/TC7a
S9 fraction liabilityb (%)
hERG inhibitionc
When the profiling data obtained thus far for these three com-
pounds were compared side-by-side, compounds 4b and 4d were
relatively similar while 4c was quite different. For resources con-
sideration, the project team decided to select only one compound
from 4b to 4d to continue the further profiling. Thus, 4d (from
4b to 4d) and 4c were selected.
(ꢂ10ꢀ7 cm/s)
Human
Mouse
Rat
% @1
l
M
% @10 lM
4a
4b
4c
4d
4g
4h
4i
98.4
111.1
205.5
80.8
69.6
9.6
0
0
14
0
0
0
3
ND
4
6
30
3
2
0
11
ND
3
4
20
21
31
19
54
36
24
ND
10
13
13
83
23
17
ND
14
10
8
35
32
17
ND
Table 3
13.8
9.1
Physico-chemical properties of the selected compounds
4j
2
4b
4c
4d
ND = not determined.
a
Caco-2/TC7 (Papp, 10ꢀ7 cm/s) is the rate at which a compound crosses the Caco-
Solubility (water)a (mg/mL)
pKa
logP
logD7.4
1.1
9.1
1.31
0.15
1.13
8.5
2.05
0.92
0.004
8.2
3.15
2.30
0.60
8.5
1.75
0.5
2/TC7 monolayer under experimental conditions.
b
S9 liability is the percent of parent compound metabolized after 15 min incu-
bation with S9 fractions of human, mouse, and rat liver microsomes at 37 °C.
c
hERG inhibition was determined using a patch-clamp technique in the whole-
a
cell configuration on Chinese hamster ovary (CHO) cells. The inhibition of hERG
currents is expressed as the percent inhibition at 1.0 and 10.0 lM, respectively.
Solubility was determined by stirring the crystalline material in pure water,
quantified by HPLC.

434
Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
Table 4
Table 6
Pharmacokinetics of selected compounds in male CF1 mice
Heart concentrations of compound 4c after intravenous administration in anaesthe-
tized dogs
4c
4d
Male CF1 micea
Male CF1 micea
Dose (mg/kg)
Heart sample
Tissue concentration*
Plasma
Brain
Plasma
Brain
ng/g
lM
iv
AUC0–1 (ngꢁh/mL)
t1/2 (h)
140
0.19
14
210
0.13
—
570
1.0
3.5
3.3
230
0.75
—
3
10
Right and left ventricles
Right and left ventricles
7760 2250
21300 1550
19.0
52.3
Cl (ngꢁh/mL)
*
Limit of quantification: 15 ng/g; values are mean SD (n = 3).
Vd (L/kg)
3.4
—
—
po
AUC (ngꢁh/mL)
Cmax
tmax
240
403
0.17
0.98
36
220
266
0.17
0.52
1600
1180
0.17
1.3
56
0.40
520
261
1.0
not alter the major parameters at 610 lM. However, at 30 lM,
t1/2 (h)
F (%)
2.4
the compound prolonged the action potential duration at 50%
repolarization (APD50) by 9% and action potential duration at 90%
repolarization (ADP90) by 17%, likely due to the inhibition of IKr
B/P ratiob
1.5
a
Administration at 2 mg/kg iv and 10 mg/kg po; iv formulation: 50% 1-methyl-2-
currents. Compound 4d shortened APD50 and APD90 at 10
and 30 M, respectively. Therefore, compound 4d was de-
prioritized.
lM
pyrrolidinone in saline; concentration = 1.0 mg/mL, dosing 2 mL; po formulation: 5%
DMSO/0.5% MC/0.2% Tween80, concentration = 1.0 mg/mL, dosing 10 mL.
l
b
B/P ratio is ‘brain to plasma’ ratio calculated with iv AUC0–1 exposure measured
Compound 4c was further assessed in cardiac safety in mongrel
dogs. Compound 4c, intravenously administered at the 3 mg/kg
dose to anaesthetized mongrel dogs, did not induce any major hae-
modynamic effect. At 10 mg/kg, iv dosing, the compound slightly
decreased arterial blood pressure and heart rate but prolonged
QT interval measured either under cardiac pacing or corrected for
heart rate, and lengthened both auricular and ventricular refrac-
tory. This observation is consistent with the results observed by
using isolated guinea pigs ventricular myocytes model system dis-
cussed above.
for brain and plasma tissues.
Compounds 4c and 4d were dosed orally to CF1 mice to mea-
sure PK parameters. The results are in Table 4. Both 4c and 4d dis-
played low plasma clearance, desirable half-life (4c: t1/2 = 0.98 h;
4d t1/2 = 1.3 h, po), and acceptable exposure (4c: 240 ngꢁh/mL
(AUC); 4d: 1600 ngꢁh/mL (AUC), for a 10 mg/kg, po) with good oral
bioavailability (36% and 56% for 4c and 4d, respectively) (Table 4).
The brain AUC (ngꢁh/mL) values in mice were 220 and 520, respec-
tively, when dosed orally at 10 mg/kg. The corresponding brain to
plasma ratios were 0.90 and 0.35 for 4c and 4d, respectively.
The cardiac safety of the selected compounds 4c and 4d was
assessed next. In order to identify compounds that not only display
a low affinity towards the hERG channel, but also do not interfere
with other cardiac ion channels,33 we evaluated the effects of the
selected compounds on action potentials recorded from ventricular
myocytes which were isolated from guinea pigs using a method
modified from that described by Salata.34 The percent changes
The cardiac tissue exposure of 4c is listed in Table 6. Compound
4c displayed sufficiently high cardiac exposure.
Compound 4c was bound to plasma proteins by 51.1–66.8% and
was homogeneously distributed between plasma and erythrocytes.
When compared to lead 2, which had plasma protein binding of
only 30–37%, compound 4c showed an improvement in this aspect,
where protein binding almost doubled.
3. Conclusion
from the baseline (0 lM of the compound) of the action potential
parameters are presented in Table 5. In this model system, 4c did
In summary, we have discovered a novel series of H3R antago-
nists through a scaffold hopping strategy. Further SAR studies of
the series culminated in the identification of ((2S,30S)-2-
methyl-[1,30]bipyrrolidinyl-10-yl)-naphthalene-2-carboxylic acid
(tetrahydro-pyran-4-yl)-amide (4c) and -[4-((2S,30S)-2-methyl-
[1,30]bipyrrolidinyl-10-yl)-phenyl]-N-(tetrahydro-pyran-4-yl)-acet-
amide (4d), which exhibited good potency in H3R binding assays,
excellent selectivity, and desirable PK properties. Compounds 4c
and 4d were also assessed in cardiac safety studies. In particular,
the effects of the compounds on action potentials recorded from
ventricular myocytes isolated from guinea pigs were used to screen
compounds that not only displayed a low affinity towards the hERG
channel, but also had lower interference with other cardiac ion
channels. Both 4c and 4d exhibited low affinity for hERG channel,
only 4c did not alter the major parameters (APD50 and APD) at
Table 5
Electrophysiological effects on single ventricular myocytes isolated from guinea pigs
Parameters*
Compound
4c
4d
Concentration (lM)
% change from baseline (0 lM)
RMP (mV)
Control
1
3
10
30
0.0
0.0
0.1 0.4
ꢀ0.2 0.6
ꢀ0.5 0.6
ꢀ1.1 0.3
0.1 0.2
0.2 0.2
0.6 0.4
0.0 0.7
APA (mV)
Control
0.0
0.0
1
3
10
30
0.1 0.1
0.2 0.3
ꢀ0.1 0.4
ꢀ1.0 0.5
ꢀ0.5 0.4
ꢀ1.0 0.4
ꢀ1.5 0.5
ꢀ2.4 0.7
610 lM in this model system. Compound 4c did not induce any
major haemodynamic effect when intravenously administered at
the 3 mg/kg dose to anaesthetized mongrel dogs. Compound 4c is
a new promising lead as potent H3R antagonist with desirable PK
properties belonging to a distinct structural class.
APD50 (ms)
APD90 (ms)
Control
1
3
10
30
0.0
0.0
0.0 1.0
2.7 1.3
5.9 1.7
9.3 2.0
ꢀ2.3 0.7
ꢀ5.5 1.6
ꢀ10.3 2.1
ꢀ17.5 2.9
Control
1
3
10
30
0.0
0.0
4. Experimental
4.1. Chemistry
0.4 0.8
3.4 0.9
8.4 1.3
16.9 2.5
ꢀ1.9 0.5
ꢀ4.2 0.9
ꢀ7.8 1.2
ꢀ12.3 1.4
*
4.1.1. General
AP, action potential; RMP, resting membrane potential; APA, action potential
All solvents and reagents were obtained from commercial
suppliers and used without further purification unless otherwise
amplitude; APD50, action potential duration at 50% of repolarization; APD90, action
potential duration at 90% of repolarization.

Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
435
stated. All reactions involving dry solvents or air-sensitive agents
were performed under a nitrogen atmosphere with glassware
dried prior to use. Solvents were dried according to standard pro-
cedures. Reactions were monitored by analytical thin-layer chro-
matography (TLC) analysis and analytical High Pressure Liquid
Chromatography–Mass Spectrometry (LCMS). TLC analyses are
performed with EM Science silica gel 60 F254 plates with visualiza-
tion by UV irradiation. The compounds were detected as single
spots on TLC plates and visualized using UV light (254 nm) or with
different spraying reagents such as KMnO4, ninhydrin, and bro-
mocresol green, respectively. Flash chromatography is performed
using Alltech pre-packed silica gel cartridges on Analogix Flash
chromatographic system (automatic sample collection). Eluent
systems are given in volume/volume concentrations. 1H NMR spec-
tra were run at 300 MHz on a Varian Mercury 300 spectrometer
with an ASW 5 mm probe, and usually recorded at ambient tem-
perature in a deuterated solvent, such as D2O, DMSO-d6 or CDCl3
unless otherwise noted. Chemical shifts values (d) are indicated
in parts per million (ppm) with reference to tetramethylsilane
(TMS) as the internal standard and coupling constants (J) are given
in Hertz (Hz). The following abbreviations are used: br = broad,
s = singlet, d = doublet, dd = doublet of doublet, t = triplet, dt =
doublet of triplet, tt = triplet of triplet, q = quartet, dq = doublet of
quartet, m = multiplet.
LCMS experiments to determine retention times (RT) and
associated mass ions are performed using one of the following
methods: Mass Spectra (MS) are recorded using a Micromass mass
spectrometer. Generally, the method used was positive electro-
spray ionization, scanning mass m/z from 100 to 1000. Liquid
chromatography was performed on a Hewlett Packard 1100 Series
Binary Pump & Degasser; Auxiliary detectors used were: Hewlett
Packard 1100 Series UV detector, wavelength = 220 nm and Sedere
SEDEX 75 Evaporative Light Scattering (ELS) detector tempera-
ture = 46 °C, N2 pressure = 4 bar. LCT: Grad (Acetonitrile+0.05%
TFA)/(H2O+0.05% TFA) = 5:95 (0 min) to 95:5 (2.5 min) to 95:5
KOH was added and the suspension was stirred at rt for 1 h. The
suspension was filtrated and rinsed with DCM. The filtrate was
concentrated and re-dissolved in 60 mL of anhydrous toluene.
A 200 mL of round bottomed flask was charged with 4-bromo-2-
methyl-N-(tetrahydro-pyran-4-yl)-benzamide (2.12 g, 7.10 mmol,
1.0 equiv), sodium tert-butoxide (0.99 g, 10.30 mmol, 1.45 equiv),
tris(dibenzylideneacetone)-dipalladium-(0) (64 mg, 0.07 mmol,
0.01 equiv), and BINAP (131 mg, 0.21 mmol, 0.03 equiv). To this
flask was transferred, under nitrogen, a solution of (2S,30S)-2-
methyl-[1,30]bipyrrolidinyl (8) in 60 mL of toluene prepared above.
The flask was heated to 85 °C (external) with stirring until the start-
ing material was completely consumed as judged by TLC analysis
(4 h). The mixture was cooled to room temperature, diluted with
ethyl acetate (20 mL) and saturated aq sodium bicarbonate
(15 mL). The two layers were separated. The aqueous layer was
extracted with EtOAc (3 ꢂ 15 mL). The combined organic extracts
were washed with brine and dried (K2CO3), filtered, and concen-
trated. The crude product was then purified by flash chromatogra-
phy on 50-g silica gel column, eluted with 0–2.5% of MeOH
containing 7 N of NH3 in DCM to obtain 2.39 g (91% yield) of a tan
colored solid. Re-crystallization from methyl t-butyl ether (MTBE)
and DCM yielded a colorless crystalline material, mp 147 °C. LCMS:
RT = 1.41 min, MS: 372 (M+H+). 1H NMR (300 MHz, CDCl3) d 7.30 (m,
2H), 6.33 (m, 1H), 5.81 (d, J = 7.8 Hz, 1H), 4.20 (m, 1H), 3.99 (m, 2H),
3.66–3.20 (m, 7H), 3.02 (m, 1H), 2.79 (m, 1H), 2.54 (q, J = 8.4 Hz, 1H),
2.46 (s, 3H), 2.22–1.43 (m, 10H), 1.14 (d, J = 6.3 Hz, 3H).
4.1.3. 4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-N-
(tetrahydro-pyran-4-yl)-benzamide (4b)
Step 1. Synthesis of 4-bromo-N-(tetrahydro-pyran-4-yl)-
benzamide.
Following the above procedure for 4a step 1, 4-bromo-N-(tetra-
hydro-pyran-4-yl)-benzamide was prepared in an identical man-
ner by coupling of tetrahydropyran-4-yl amine (0.23 g, 2.3 mmol,
1.0 equiv) with 4-bromobenzoic acid (0.51 g, 2.5 mmol, 1.1 equiv)
to give 0.65 g (100% yield) as white solid. LCMS: RT = 2.57 min,
MS: 285 (M+H+).
(3 min). Column: YMC Jsphere 33 ꢂ 24
lM, 1 ml/min.
4.1.2. 2-Methyl-4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-yl)-
N-(tetrahydro-pyran-4-yl)-benzamide (4a)
Step 2. Synthesis of 4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-
yl)-N-(tetrahydro-pyran-4-yl)-benzamide.
Step 1. Synthesis of 4-bromo-2-methyl-N-(tetrahydro-pyran-4-
yl)-benzamide.
Following the above procedure for 4a, step 2, the title com-
pound was prepared in an identical manner by condensing
(2S,30S)-2-methyl-[1,30]bipyrrolidinyl dihydrochloride (8) (0.7 g,
3.1 mmol, 1.1 equiv) with 4-bromo-N-(tetrahydro-pyran-4-yl)-
benzamide (0.80 g, 2.8 mmol), to give 1.0 g (67% yield) as a solid.
Re-crystallization from MTBE and DCM yielded a colorless crystal-
line material, mp 180 °C. Elemental analysis: Theoretical: C 70.55;
4-Bromo-2-methylbenzoic acid (5.38 g, 25 mmol) was dissolved
in DCM (100 mL) and DMF (25 mL) and the solution was cooled to
an ice-water bath. To this solution was added a solution of tetrahy-
dropyran-4-yl amine (2.60 g, 25 mmol, 1.0 equiv) in 10 mL of DCM,
followed by N-methylmorpholine (5.75 g, 8 mL, 75 mmol,
3.0 equiv), 1-hydroxylbenzotriazole (HOBt) (4.40 g, 32.5 mmol,
1.3 equiv), sequentially, and finally EDCꢁHCl (6.25 g, 32.5 mmol,
1.3 equiv). The resultant clear light brown solution was stirred at
rt overnight. TLC (10% MeOH in DCM) and LCMS detected a product
peak with m/z of 299 (M+H+). The reaction was quenched with sat-
urated sodium bicarbonate aqueous solution (10 mL) and DCM
(10 mL). The two layers were separated, and the aqueous layer
was extracted with DCM (15 mL ꢂ 2). The combined DCM extracts
were washed with sodium bicarbonate aqueous solution (10 mL)
and brine (10 mL), and dried (anhydrous potassium carbonate), fil-
tered, and concentrated in vacuo to get 6.87 g (yield 92%) of the title
compound as a white solid, LCMS: RT = 2.60 min, MS: 299 (M+H+).
1H NMR (300 MHz, CDCl3) d 7.31 (m, 3H), 5.60 (br s, 1H), 4.10 (m,
1H), 4.01 (m, 2H), 3.53 (dt, J = 11.8, 2.1 Hz, 2H), 2.42 (s, 3H), 2.00
(m, 2H), 1.55 (m, 2H).
H
8.74; N 11.75; Found: C 70.30 H 8.90 N 11.63. LC/MS:
RT = 1.35 min, MS: 358 (M+H+). 1H NMR (300 MHz, CDCl3) d 7.62
(d, J = 8.7 Hz, 2H), 6.50 (d, J = 8.7 Hz, 2H), 5.81 (d, J = 7.8 Hz, 1H),
4.20 (m, 1H), 3.99 (m, 2H), 3.66–3.20 (m, 7H), 3.02 (m, 1H), 2.79
(m, 1H), 2.54 (q, J = 8.4 Hz, 1H), 2.22–1.43 (m, 10H), 1.14
(d, J = 6.3 Hz, 3H).
4.1.4. 6-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-
naphthalene-2-carboxylic acid (tetrahydro-pyran-4-yl)-amide
(4c)
Step 1. Synthesis of 6-bromo-naphthalene-2-carboxylic acid
(tetrahydro-pyran-4-yl)-amide.
Following the above procedure for 4a, step 1, 6-bromo-naphtha-
lene-2-carboxylic acid (tetrahydro-pyran-4-yl)-amide was synthe-
sized in an identical manner by coupling of tetrahydropyran-4-yl
amine (0.26 g, 2.5 mmol, 1.0 equiv) with 6-bromo-naphthalene-2-
carboxylic acid (0.63 g, 2.5 mmol, 1.0 equiv) to give 0.85 g (100%
yield) of the title compound as a white solid. TLC (5% MeOH in
DCM) Rf = 0.7. LCMS: RT = 2.85 min, MS: 335 (M+H+). 1H NMR
(300 MHz, CDCl3) d 8.24 (s, 1H), 8.04 (s, 1H), 7.82 (m, 3H), 7.62
Step 2. Synthesis of 2-methyl-4-((2S,30S)-2-methyl-[1,30]bipyrr-
olidinyl-10-yl)-N-(tetrahydro-pyran-4-yl)-benzamide.
(2S,30S)-2-Methyl-[1,30]bipyrrolidinyl
dihydrochloride
(8)
(2.06 g, 7.83 mmol, 1.1 equiv), was dissolved in DCM (60 mL) and
methanol (5 mL) with the aid of sonication. 1.14 g of powder

436
Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
(m, 1H), 6.13 (m, 1H), 4.03 (m, 2H), 3.56 (dt, J = 2.2, 11.7 Hz, 2H),
2.06 (m, 2H), 1.62 (m, 1H).
1.0 equiv), to give 3.14 g (94% yield) of the title compound as a
tan colored crystalline solid. RT = 2.20 min, MS: 471 (M+H+). 1H
Step 2. Synthesis of 6-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-
yl)-naphthalene-2-carboxylic acid (tetrahydro-pyran-4-yl)-amide.
Following the above procedure for 4a, step 2, the title com-
pound was prepared in an identical manner by condensing
(2S,30S)-2-methyl-[1,30]bipyrrolidinyl dihydrochloride (8) (0.6 g,
1.1 mmol), with 6-bromo-naphthalene-2-carboxylic acid (tetrahy-
dro-pyran-4-yl)-amide (0.57 g, 1.91 mmol), to give (64% yield) of
the title compound as a fine powder. RT = 1.88 min, MS: 408
(M+H+). 1H NMR (300 MHz, CDCl3) d 8.12 (s, 1H), 7.78–7.64 (m,
3H), 7.02 (m, 1H), 6.72 (m, 1H), 6.07 (m, 1H), 4.30 (m, 1H), 4.04
(m, 2H), 3.60–3.34 (m, 7H), 3.05 (m, 1H), 2.83 (m, 1H), 2.58 (m,
1H), 2.2–1.41 (m, 10H), 1.16 (d, J = 6.2 Hz, 3H).
NMR (300 MHz, CDCl3) d 7.07 (d, J = 8.4 Hz, 2H), 6.53 (d,
J = 8.4 Hz, 2H), 5.26 (m, 1H), 3.91 (m, 3H), 3.56–3.35 (m, 7H),
3.02 (m, 1H), 2.81 (m, 3H), 2.53 (q, J = 8.4 Hz, 1H), 2.1 (m, 3H),
1.81 (m, 4H), 1.53–1.43 (m, 10H), 1.3–1.06 (m, 5H).
4.1.7. Synthesis of 2-[4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-
10-yl)-phenyl]-N-piperidin-4-yl-acetamide (4f)
4-[2-[4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-phenyl]-
acetylamino]-piperidine-1-carboxylic acid tert-butyl ester (4e)
(2.92 g, 6.2 mmol) was dissolved in 5 mL of dioxane. To this solu-
tion was added 1.0 g of PhSH, followed by addition of 4 N HCl in
dioxane (6.2 mL, 24.8 mmol, 4 equiv). The solution was stirred at
rt overnight. HCl was removed by passing nitrogen gas through
the solution and the gaseous HCl was captured by a drying tube
filled with NaOH.pallets. The solvent was removed by reduced
pressure distillation. The residue was further dried under high
vacuum to get 2.70 g (100%) of the title compound as a hydrochlo-
ride salt. LCMS: RT = 2.11 min, MS: 371 (M+H+). 1H NMR
(300 MHz, CDCl3) d 11.20 (br s, 1H), 9.00 (br s, 1H), 8.21 (m, 1H),
7.12 (d, J = 8.7 Hz, 2H), 6.57 (d, J = 8.7 Hz, 2H), 3.8–3.38 (m, 7H),
3.21 (m, 3H), 2.91 (m, 3H), 2.42–2.13 (m, 5H), 1.99–1.55 (m,
8H), 1.46 (d, J = 6.6 Hz, 3H).
4.1.5. 2-[4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-phenyl]-
N-(tetrahydro-pyran-4-yl)-acetamide (4d)
Step 1. Synthesis of 2-(4-bromo-phenyl)-N-(tetrahydro-pyran-
4-yl)-acetamide.
The title compound was prepared in substantially the same way
as 4a, step 1, by coupling of tetrahydropyran-4-yl amine with (4-
bromo-phenyl)-acetic acid to give 0.65 g (100% yield) of the title
compound as a thick oil which solidified on standing. TLC (5%
MeOH in DCM) Rf = 0.6. RT = 2.57 min, MS: 299 (M+H+). 1H NMR
(300 MHz, CDCl3) d 7.49 (d, J = 6.0 Hz, 2H), 7.13 (d, J = 6.0 Hz, 2H),
5.24 (br s, 1H), 3.99 (m, 3H), 3.49 (s, 2H), 3.48 (m, 2H), 1.85 (m,
2H), 1.38 (m, 2H).
4.1.8. N-(1-Cyclopentylmethyl-piperidin-4-yl)-2-[4-((2S,30S)-2-
methyl-[1,30]bipyrrolidinyl-10-yl)-phenyl]-acetamide (4g)
Step 2. Synthesis of 2-[4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-
10-yl)-phenyl]-N-(tetrahydro-pyran-4-yl)-acetamide.
Following the above procedure for 4a, step 2, the title compound
was prepared in an identical manner by condensing (2S,30S)-2-
methyl-[1,30]bipyrrolidinyl dihydrochloride (8) (0.6 g, 1.1 mmol),
To
a
solution of cyclopentane-carboxaldehyde (138 mg,
1.4 mmol) and 2-[4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-yl)-
phenyl]-N-piperidin-4-yl-acetamide hydrochloride (4f) (0.52 g,
1.18 mmol) in DCE (10 mL) was added powder sodium triacetoxy-
borohydride slowly under N2 at rt. The yellowish milky solution
was stirred at rt overnight. LCMS showed a major peak with MS
of 453. The reaction was quenched with NaHCO3 (aq) (15 mL).
NaOH (aq) (1 N, 5 mL) was added thereafter. The two layers were
separated, and the aqueous layer was extracted with DCM
(10 mL ꢂ 2). The combined DCM extracts were washed with
sodium bicarbonate (10 mL), and brine (5 mL ꢂ 2), dried (anhy-
drous potassium carbonate), filtered, and concentrated in vacuo
with
2-(4-bromo-phenyl)-N-(tetrahydro-pyran-4-yl)-acetamide
(0.57 g, 1.91 mmol), to give 510 mg (72% yield) of the title compound
as a white solid. This was recrystallized from DCM (1 mL) and MTBE
(15 mL) to get 460 mg of the title compound as a colorless crystalline
material. Elemental analysis: Theoretical: C 71.12 H 8.95 N 11.31 O
8.61; Found: C 71.26 H 8.95 N 11.37 (KARL FISCHER = 0.22). Mp
155–157 °C. RT = 2.23 min, MS: 372 (M+H+). 1H NMR (300 MHz,
CDCl3) d 7.08 (d, J = 8.4 Hz, 2H), 6.54 (d, J = 8.4 Hz, 2H), 5.3 (br s,
1H), 3.96 (m, 1H), 3.85 (m, 2H), 3.45 (m, 6H), 3.26 (m, 3H), 3.02 (m,
1H), 2.78 (m, 1H), 2.55 (q, J = 8.7 Hz, 1H), 2.21–1.71 (m, 7H), 1.48
(m, 1H), 1.31 (m, 2H), 1.14 (d, J = 6.6 Hz, 3H).
to get 0.30 g (56%) of the title compound as
a tan solid.
RT = 1.34 min, MS: 453 (M+H+). 1H NMR (300 MHz, CDCl3) d 7.06
(d, J = 8.7 Hz, 2H), 6.52 (d, J = 8.7 Hz, 2H), 5.24 (m, 1H), 3.75 (m,
1H), 3.52 (m, 2H), 3.45 (s, 2H), 3.38–3.20 (m, 3H) 3.03 (m, 1H),
2.74 (m, 2H), 2.56 (d, J = 8.4 Hz, 1H), 2.21–1.93 (m, 8H), 1.86–
1.43 (m, 14H), 1.27 (m, 2H), 1.14 (d, J = 6.3 Hz, 3H).
4.1.6. Synthesis of 2-[4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-
10-yl)-phenyl]-acetylamino]-1,2-piperidine-1-carboxylic acid
tert-butyl ester (4e)
Step 1. Synthesis of 4-[2-(4-bromo-phenyl)-acetylamino]-piper-
idine-1-carboxylic acid tert-butyl ester.
4.1.9. 4-[2-[4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-
phenyl]-acetylamino]-piperidine-1-carboxylic acid ethyl ester
(4h)
The title compound was prepared in substantially the same way
as 4a, step 1, by coupling of (4-bromo-phenyl)-acetic acid (2.84 g,
13.2 mmol, 1.1 equiv) with 4-amino-piperidine-1-carboxylic acid
tert-butyl ester (2.4 g, 12 mmol, 1.0 equiv) to give 4.62 g (96.9%
yield) of the title compound as a very white crystalline material.
RT = 3.01 min, MS: 398 (M+H+). 1H NMR (300 MHz, CDCl3) d 7.49
(d, J = 8.4 Hz, 2H), 7.15 (d, J = 8.4 Hz, 2H), 5.20 (m, 1H), 3.94 (m,
3H), 3.49 (s, 2H), 2.83 (br t, J = 12.6, 2H), 1.85 (m, 2H), 1.44 (s,
9H), 1.20 (m, 2H).
2-[4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-phenyl]-N-
piperidin-4-yl-acetamide hydrochloride (4f) (0.51 g, 1.38 mmol)
was dissolved in THF (20 mL) and water (5 mL) to form a clear
solution. To this solution was added 0.30 g of ethyl chloroformate
(3.00 mmol, 2.0 equiv), followed by K2CO3 (0.76 g, 5.52 mmol,
4.0 equiv). The clear solution was stirred at rt overnight. TLC (5%
of 7 N NH3/MeOH in DCM) showed the starting material was con-
sumed completely. LCMS: a single peak with MS of 443 (M+H+)
was detected. The reaction was diluted with EtOAc (10 mL) and
water (15 mL). The two layers were separated and the aqueous
layer was extracted with EtOAc (2 ꢂ 10 mL). The combined organic
solution was washed with brine (2 ꢂ 10 mL), dried (K2CO3), fil-
tered, and concentrated. The crude product was purified on a 25-
g silica gel column eluted with 0–5% MeOH in DCM to get 0.42 g
(69%) of the title compound as a tan solid. LCMS: RT = 1.88 min,
MS: 443 (M+H+). 1H NMR (300 MHz, CDCl3) d 7.06 (d, J = 8.7 Hz,
2H), 6.53 (d, J = 8.7 Hz, 2H), 5.28 (m, 1H), 4.09 (q, J = 7.2 Hz, 2H),
Step 2. Synthesis of 2-[4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-
10-yl)-phenyl]-acetylamino]-1,2-piperidine-1-carboxylic acid tert-
butyl ester (4e).
Following the above procedure for 4a, step 2, the title com-
pound was prepared in an identical manner by condensing
(2S,30S)-2-methyl-[1,30]bipyrrolidinyl dihydrochloride (8) (2.1 g,
7.83 mmol, 1.1 equiv), with 4-[2-(4-bromo-phenyl)-acetylamino]-
piperidine-1-carboxylic acid tert-butyl ester (2.82 g, 7.1 mmol,

Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
437
3.94–3.52 (m, 4H), 3.40 (s, 2H), 3.44–3.18 (m, 6H), 3.03 (m, 1H),
2.88 (m, 2H), 2.78 (m, 1H), 2.54 (d, J = 8.4 Hz, 1H), 2.21–1.69 (m,
8H), 1.49 (m, 1H), 1.23 (t, J = 7.2 Hz, 3H), 1.14 (d, J = 6.2 Hz, 3H).
24 mmol, 1.2 equiv) and powder sodium triacetoxyborohydride
slowly under N2 at rt. The yellowish milky solution was stirred at
rt overnight. LCMS of the reaction mixture: RT = 2.43 min,
MS = 314 (77%) with impurity at 2.72 min, MS = 425 (23%).
The reaction was quenched with aq NaHCO3 solution. The two
layers were separated, and the aqueous layer was extracted with
DCM (10 mL ꢂ 2). The combined DCM extracts were washed with
sodium bicarbonate (10 mL), and brine (5 mL ꢂ 2), dried (anhy-
drous potassium carbonate), filtered, and concentrated in vacuo
to get 5.71 g (yield: 92%) of the ester (12) as a liquid.
4.1.10. 4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-N-
(tetrahydro-pyran-4-yl-methyl)-benzamide (4i)
Step 1. Synthesis of 4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-
yl)-benzoic acid tert-butyl ester (10).
(2S,30S)-2-Methyl-[1,30]bipyrrolidinyl dihydrochloride (8) (1.1 g,
4 mmol) was treated with NaOt-Bu (1.1 g, 2equiv) in MeOH
(30 mL), filtered, and concentrated. To this flask was added
tert-butyl-4-bromo-benzoate (9) (1.0 g, 3.89 mmol), sodium tert-
butoxide (538 mg, 5.6 mmol), tris(dibenzylideneacetone)-dipalla-
dium-(0) (35 mg, 0.039 mmol), BINAP (72 mg, 0.116 mmol), and
toluene (40 mL) under nitrogen. The reaction flask was heated to
80 °C (external) for 24 h with stirring until the starting material
was completely consumed as judged by TLC analysis. The mixture
was cooled to room temperature, taken up in DCM (50 mL),
filtered, and concentrated. The crude product was then purified
by a silica gel flash chromatography eluted with 0–5% of 7 N NH3/
MeOH in DCM to get 1.02 g (80%) of the title compound as a yellow
oil. LCMS: RT = 2.23 min, MS: 331 (M+H+). 1H NMR (300 MHz,
CDCl3) d 7.87 (d, J = 8.7 Hz, 2H), 6.49 (d, J = 8.7 Hz, 2H), 3.6–3.23
(m, 5H), 3.01 (m, 1H), 2.78 (m, 1H), 2.55 (q, J = 8.4 Hz, 1H), 2.22–
1.43 (m, 13H), 1.14 (d, J = 6.3 Hz, 3H).
Step 2. Synthesis of 1-(4-bromo-benzyl)-piperidine-4-carbox-
ylic acid methyl amide (13).
The ester (12) (0.5 g, crude as such obtained above, about
1.6 mmol) was dissolved in dioxane (5 mL). To this solution was
added aqueous MeNH2 (40% w/w, 2.5 mL, excess). The clear color-
less solution was stirred over an oil bath heating at 50 °C overnight.
LCMS did not detect the SM, but a product with MS of 312, along
with the impurity at 2.332 min with MS of 424.
The solvent was removed via reduced pressure distillation. The
residue was taken in 5 mL of DCM and 5 mL of aq NaHCO3. The two
layers were separated. The aqueous layer was extracted with DCM
(3 ꢂ 15 mL). The combined organic extracts were washed with brine
and dried (K2CO3), filtered, and concentrated. The crude product was
then purified by flash chromatography on 25-g silica gel column,
eluted with 0–2.5% of 7 N NH3/MeOH in DCM to obtain 0.23 g (46%
yield) of the title compound (13) as a white powder. RT = 1.41 min,
MS: 312 (M+H+). 1H NMR (300 MHz, CDCl3) d 7.44 (d, J = 8.4 Hz,
2H), 7.20 (d, J = 8.4 Hz, 2H), 5.45 (br s, 1H), 3.43 (s, 2H), 2.89 (m,
2H), 2.82 (d, J = 4.8 Hz, 3H), 2.13–1.92 (m, 3H), 1.84–1.70 (m, 4H).
Step 3. Synthesis of 1-[4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-
10-yl)-benzyl]-piperidine-4-carboxylic acid methyl amide (4j).
A round bottomed flask (100 mL) was charged with 1-(4-
bromo-benzyl)-piperidine-4-carboxylic acid methyl amide (13)
(230 mg, 0.74 mmol, 1 equiv), sodium tert-butoxide (102 mg,
1.07 mmol, 1.45 equiv), tris(dibenzylideneacetone)-dipalladium-
(0), and BINAP. To this flask was transferred a solution of the amine
(0.25 M, pre-made from (2S,30S)-2-methyl-[1,30]bipyrrolidinyl
dihydrochloride (8), 3.7 mL, 0.925 mmol, 1.2 equiv) in toluene
under nitrogen. The flask was heated to 85 °C (external) with stir-
ring until the starting material was completely consumed (6 h) as
judged by TLC analysis. LC/MS: RT = 1.703 min; MS = 385.
Step 2. Synthesis of 4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-
yl)-benzoic acid (11).
4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-benzoic acid tert-
butyl ester (10) (0.28 g, 0.8 mmol) was treated with 5 mL of 4 N HCl
in dioxane at rt overnight. The HCl was removed by passing nitro-
gen gas through the solution and the gaseous HCl was captured
by a drying tube filled with NaOH.pallets. The solvent was evapo-
rated to dryness and dried further under high vacuum to yield
0.37 g (HCl salt, ꢃ100%) of the title compound as a white powder.
TLC (5% MeOH in DCM) Rf = 0.6.
Step 3. Synthesis of 4-((2S,30S)-2-methyl-[1,30]bipyrrolidinyl-10-
yl)-N-(tetrahydro-pyran-4-yl-methyl)-benzamide (4i).
4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-benzoic acid (11)
(0.17 g, 0.625 mmol) was dissolved in DCM (4 mL) and DMF (2 mL)
and the solution was cooled to an ice-water bath. To this solution
was added
a solution of tetrahydropyran-4-yl amine (0.10 g,
0.86 mmol, 1 equiv) in 1 mL of DCM, followed by N-methylmorpho-
line (0.2 mL, 3 equiv), 1-hydroxylbenzotriazole (HOBt) (0.11 g,
1.3 equiv), sequentially, and finally EDCꢁHCl (104 mg, 1.3 equiv).
The resultant clear light brown solution was stirred at rt overnight.
TLC (10% MeOH in DCM) and LC/MS detected the product peak at
retention time of 2.15 min with MS of 372. The reaction was
quenched with saturated sodium bicarbonate aqueous solution
(10 mL) and 10 mL of DCM. The two layers were separated, and the
aqueous layer was extracted with DCM (15 mL ꢂ 2). The combined
DCM extracts were washed with sodium bicarbonate (10 mL), and
brine (10 mL), dried (anhydrous potassium carbonate), filtered, and
concentrated in vacuo to get 0.17 g (74%) of the titled compound as
a solid. LCMS: RT = 1.43 min, MS: 372 (M+H+). 1H NMR (300 MHz,
CDCl3) d 7.68 (d, J = 8.7 Hz, 2H), 6.52 (d, J = 8.7 Hz, 2H), 6.08 (s, 1H),
3.98 (m, 2H), 3.66–3.20 (m, 7H), 3.02 (m, 1H), 2.78 (m, 1H), 2.54 (q,
J = 8.4 Hz, 1H), 2.22–1.43 (m, 10H), 1.54–1.30 (m, 3H), 1.14 (d,
J = 6.3 Hz, 3H).
The mixture was cooled to room temperature, diluted with
ethyl acetate (20 mL) and aq sodium bicarbonate (15 mL). The
two layers were separated. The aqueous layer was extracted with
EtOAc (3 ꢂ 15 mL). The combined organic extracts were washed
with brine and dried (K2CO3), filtered, and concentrated. The crude
product was then purified by flash chromatography on 25-g silica
gel column, eluted with 0–2.5% of 7 N NH3/MeOH in DCM to obtain
150 mg (53% yield) of the title compound as an oil. This was dis-
solved in 1 mL of DCM, added 0.5 mL of 1 N HCl in ether. The white
precipitate formed was collected and further dried to get 155 mg of
the title compound as a HCl salt. LCMS: RT = 1.77 min, MS = 385
(M+H+). 1H NMR (300 MHz, DMSO) d 7.89 (m, 1H), 7.44 (d,
J = 8.4 Hz, 2H), 7.20 (d, J = 8.4 Hz, 2H), 4.2–3.89 (m, 3H), 3.76–
3.17 (m, 5H), 3.43 (s, 2H), 2.89 (m, 2H), 2.56 (d, J = 5.1 Hz, 3H),
2.13–1.92 (m, 6H), 1.84–1.70 (m, 7H), 1.46 (d, J = 6.6 Hz, 3H).
4.2. Pharmacology
4.1.11. 1-[4-((2S,30S)-2-Methyl-[1,30]bipyrrolidinyl-10-yl)-benzyl]-
piperidine-4-carboxylic acid methylamide hydrochloride (4j)
Step 1. Synthesis of 1-(4-bromo-benzyl)-piperidine-4-carbox-
ylic acid methyl ester (12).
4.2.1. H3R binding inhibition constant (Ki) determination
The H3R radio ligand binding assay was performed using H3
receptor membranes prepared from the Flp-In T-REx 293 Cell Line
(Invitrogen) stably transfected with pcDNA5/FRT/TO (Invitrogen)
containing the human or rhesus monkey (Macaca mulatta) or rat
(Rattus norvegicus) 445 amino acid H3 receptor. [3H]-methylhista-
mine is purchased from Perkin Elmer and wheat germ agglutinin
To
a solution of 4-bromo-benzaldehyde (3.88 g, 21 mmol,
1.05 equiv) in DCE (60 mL) was added piperidine-4-carboxylic acid
methyl ester (2.86 g, 20 mmol), followed by 4 N HCl (6 mL,

438
Z. Gao et al. / Bioorg. Med. Chem. 23 (2015) 429–438
scintillation proximity assay (WGA SPA) beads are from Amer-
sham. The assay was performed in 96-well opti-plates (Packard).
contributions, the Sanofi analytical department for their support
utilizing analytical studies to confirm compound structures, per-
form solid state characterizations, formulation support and the
Sanofi DMPK and Drug Safety department for PK and in vitro/
in vivo drug safety assessments.
Each reaction contained 50
protein), 50 l WGA SPA beads (0.1
[3H]-methylhistamine (final concentration 2 nM) and 50
l
l H3R membranes (20–30
g) and 50 l of 83 Ci/mmol
l of
lg total
l
l
l
l
tested compound. The exemplified compounds and/or vehicle were
diluted with binding buffer from 10 mM DMSO solutions. Assay
plates were sealed with TopSeal (Perkin Elmer) and mixed on sha-
ker (25 °C, 1 h). Assay plates were read on TopCount scintillation
counter (Packard). Results were analyzed by Hill transformation
and Ki values were determined by Cheng–Prusoff equation.
References and notes
1. Arrang, J. M.; Garbarg, M.; Schwartz, J. C. Nature 1983, 302, 832.
2. Singh, M.; Jadhav, H. R. Mini-Rev. Med. Chem. 2013, 13, 47.
3. Berlin, M.; Boyce, C. W.; Ruiz Mde, L. J. Med. Chem. 2011, 54, 26.
4. Lazewska, D.; Kiec-Kononowicz, K. Expert Opin. Ther. Patents 2010, 20, 1147.
5. Gemkow, M. J.; Davenport, A. J.; Harich, S.; Ellenbroek, B. A.; Cesura, A.; Hallett,
D. Drug Discovery Today 2009, 14, 509.
6. Esbenshade, T. A.; Browman, K. E.; Bitner, R. S.; Strakhova, M.; Cowart, M. D.;
Brioni, J. D. Br. J. Pharmacol. 2008, 154, 1166.
7. Wijtmans, M.; Leurs, R.; de Esch, I. Expert Opin. Invest. Drugs 2007, 16, 967.
8. Leurs, R.; Blandina, P.; Tedford, C.; Timmerman, H. Trends Pharmacol. Sci. 1998,
19, 177.
9. Alves-Rodrigues, A.; Timmerman, H.; Willems, E.; Lemstra, S.; Zuiderveld, O. P.;
Leurs, R. Brain Res. 1998, 788, 179.
10. Weisler, R. H.; Pandina, G. J.; Daly, E. J.; Cooper, K.; Gassmann-Mayer, C.;
Investigators, A. T. T. S. CNS Drugs 2012, 26, 421.
11. Cowart, M.; Faghih, R.; Curtis, M. P.; Gfesser, G. A.; Bennani, Y. L.; Black, L. A.;
Pan, L.; Marsh, K. C.; Sullivan, J. P.; Esbenshade, T. A.; Fox, G. B.; Hancock, A. A. J.
Med. Chem. 2005, 48, 38.
12. Vohora, D.; Bhowmik, M. Front. Syst. Neurosci. 2012, 6, 72.
13. Herring, W. J.; Wilens, T. E.; Adler, L. A.; Baranak, C.; Liu, K.; Snavely, D. B.;
Lines, C. R.; Michelson, D. J. Clin. Psychiatry 2012, 73, e891.
14. Haig, G. M.; Bain, E.; Robieson, W.; Othman, A. A.; Baker, J.; Lenz, R. A. Schizophr.
Bull. 2014, 40, 1433.
4.3. Drug safety
4.3.1. Patch-clamp recording from CHO cells
Patch-clamp technique was used to evaluate the hERG channel
affinity of compounds as described previously.35 Chinese hamster
ovary (CHO) cells stably expressing the hERG cardiac potassium
channels were seeded on glass coverslips 12–48 h prior to patch-
clamp recording. Whole-cell configuration of the patch-clamp
technique was used to record hERG currents from CHO cells on a
coverslip. Electrodes were filled with pipette solution containing
120 mM potassium aspartate, 20 mM KCl, 4 mM Na2ATP, 5 mM
HEPES, 1 mM MgCl2, pH 7.2 with KOH (resistance: 2–4 M
X). The
recording chamber was perfused with external solution containing
130 mM NaCl, 5 mM KCl, 2.8 mM sodium acetate, 1.0 mM MgCl2,
10 mM HEPES, 10 mM glucose, 1.0 mM CaCl2, pH 7.4 with NaOH.
Currents were recorded at room temperature using an Axopatch
200B amplifier (Molecular Devices LLC, Sunnyvale, CA) and ana-
lyzed using the pCLAMP software (Molecular Devices LLC.).
15. Esbenshade, T. A.; Fox, G. B.; Cowart, M. D. Mol. Interventions 2006, 6, 77.
16. Hancock, A. A.; Diehl, M. S.; Fey, T. A.; Bush, E. N.; Faghih, R.; Miller, T. R.;
Krueger, K. M.; Pratt, J. K.; Cowart, M. D.; Dickinson, R. W.; Shapiro, R.;
Knourek-Segel, V. E.; Droz, B. A.; McDowell, C. A.; Krishna, G.; Brune, M. E.;
Esbenshade, T. A.; Jacobson, P. B. Inflamm. Res. Official J. Eur. Histamine Res. Soc.
2005, 54, S27.
17. Bhowmik, M.; Khanam, R.; Vohora, D. Br. J. Pharmacol. 2012, 167, 1398.
18. McGaraughty, S.; Chu, K. L.; Cowart, M. D.; Brioni, J. D. J. Pharmacol. Exp. Ther.
2012, 343, 13.
19. Levi, R.; Seyedi, N.; Schaefer, U.; Estephan, R.; Mackins, C. J.; Tyler, E.; Silver, R.
B. Biochem. Pharmacol. 2007, 73, 1146.
20. Kuhne, S.; Wijtmans, M.; Lim, H. D.; Leurs, R.; de Esch, I. J. Expert Opin. Invest.
Drugs 2011, 20, 1629.
21. Celanire, S.; Wijtmans, M.; Talaga, P.; Leurs, R.; de Esch, I. J. Drug Discovery
Today 2005, 10, 1613.
4.3.2. Action potential recording from guinea pig myocytes
Single ventricular myocytes were isolated from guinea pigs with
procedures and protocols approved by the Sanofi Institutional Ani-
mal Care and Use Committee (Bridgewater, New Jersey). Myocyte
isolation, action potential recording, and data analysis were per-
formed as described previously.35 Briefly, male guinea pigs (Hart-
ley) were anesthetized and a thoracotomy was performed. The
heart was quickly excised and perfused retrogradely at 10 mL/
min through the aorta with modified Tyrode’s solution (containing
132 mM NaCl, 4 mM KCl, 1.2 mM MgCl2, 10 mM HEPES, and 10 mM
glucose, pH adjusted to 7.4 with 10 M NaOH) for 4 min followed by
an 8-min perfusion with modified Tyrode’s solution containing
102 units/ml Type 2 collagenase (271 units/mg, Worthington Bio-
chemical Corporation, Lakewood, NJ) and 0.6 units/ml Type XIV
protease (4 units/mg, Sigma Chemical Co., St. Louis, MO). The left
ventricle was cut into small pieces and gently bubbled with 100%
O2 at room temperature for ꢃ2 min to disperse single myocytes,
which were maintained at room temperature for electrophysiolog-
ical recording, usually within 5 h of isolation.
22. Schwartz, J. C. Br. J. Pharmacol. 2011, 163, 713.
23. Gao, Z.; Hurst, W. J.; Czechtizky, W.; Hall, D.; Moindrot, N.; Nagorny, R.; Pichat,
P.; Stefany, D.; Hendrix, J. A.; George, P. G. Bioorg. Med. Chem. Lett. 2013, 23,
6269.
24. Gao, Z.; Hurst, W. J.; Guillot, E.; Czechtizky, W.; Lukasczyk, U.; Nagorny, R.;
Pruniaux, M. P.; Schwink, L.; Sanchez, J. A.; Stengelin, S.; Tang, L.; Winkler, I.;
Hendrix, J. A.; George, P. G. Bioorg. Med. Chem. Lett. 2013, 23, 3421.
25. Gao, Z.; Hurst, W. J.; Guillot, E.; Czechtizky, W.; Lukasczyk, U.; Nagorny, R.;
Pruniaux, M. P.; Schwink, L.; Sanchez, J. A.; Stengelin, S.; Tang, L.; Winkler, I.;
Hendrix, J. A.; George, P. G. Bioorg. Med. Chem. Lett. 2013, 23, 3416.
26. Gao, Z.; Hurst, W. J.; Czechtizky, W.; Francon, D.; Griebel, G.; Nagorny, R.;
Pichat, P.; Schwink, L.; Stengelin, S.; Hendrix, J. A.; George, P. G. Bioorg. Med.
Chem. Lett. 2013, 23, 6141.
27. Gao, Z.; Hurst, W. J.; Guillot, E.; Nagorny, R.; Pruniaux, M. P.; Hendrix, J. A.;
George, P. G. Bioorg. Med. Chem. Lett. 2013, 23, 4044.
28. Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1144.
29. Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J.; Buchwald, S. L. J. Org. Chem. 2000,
65, 1158.
30. Vilar, S.; Chakrabarti, M.; Costanzi, S. J. Mol. Graphics Modell. 2010, 28, 899.
31. Summerfield, S. G.; Read, K.; Begley, D. J.; Obradovic, T.; Hidalgo, I. J.;
Coggon, S.; Lewis, A. V.; Porter, R. A.; Jeffrey, P. J. Pharmacol. Exp. Ther. 2007,
322, 205.
32. Ravula, S. B.; Yu, J.; Tran, J. A.; Arellano, M.; Tucci, F. C.; Moree, W. J.; Li, B. F.;
Petroski, R. E.; Wen, J.; Malany, S.; Hoare, S. R.; Madan, A.; Crowe, P. D.; Beaton,
G. Bioorg. Med. Chem. Lett. 2012, 22, 421.
33. Roden, D. M.; Balser, J. R.; George, A. L., Jr.; Anderson, M. E. Annu. Rev. Physiol.
2002, 64, 431.
34. Salata, J. J.; Wasserstrom, J. A. Circ. Res. 1988, 62, 324.
35. Kang, J.; Chen, X. L.; Wang, H.; Ji, J.; Cheng, H.; Incardona, J.; Reynolds, W.;
Viviani, F.; Tabart, M.; Rampe, D. Mol. Pharmacol. 2005, 67, 827.
Ventricular myocytes in a recording chamber with the temper-
ature set at 37 1 °C were perfused with the modified Tyrode’s
solution containing 1.8 mM CaCl2. Action potentials were initiated
from a myocyte at 1 Hz current pulses using glass microelectrodes
(filled with 3 M KCl, Resistance = 20–45 M
X). Action potential sig-
nals were amplified using an Axoclamp 2B amplifier (Molecular
Devices LLC.) and stored for analysis with the pCLAMP software
(Molecular Devices, LLC.).
Acknowledgements
The authors greatly appreciate Sanofi R&D management for
their strong support, the H3R project team members for their