Reduction of hERG inhibitory activity in the 4-piperidinyl urea series of H3 antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.01.121

Structural features of the substituted 4-piperidinyl urea analogs 1, responsible for the H3 antagonist activity, have been identified. Structure-activity relationship of the H3 receptor affinity, hERG ion channel inhibitory activity and their separation is described. Preliminary pharmacokinetic evaluation of the compounds of the series is addressed.

Full text of DOI:10.1016/j.bmcl.2010.01.121

Bioorganic & Medicinal Chemistry Letters 20 (2010) 2359–2364
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Reduction of hERG inhibitory activity in the 4-piperidinyl urea series
of H3 antagonists
*
Michael Berlin , Yoon Joo Lee, Christopher W. Boyce, Yi Wang, Robert Aslanian, Kevin D. McCormick,
Steve Sorota, Shirley M. Williams, Robert E. West Jr., Walter Korfmacher
Chemical Research, Cardiovascular/Metabolic Diseases and CNS Discovery, and Drug Metabolism and Pharmacokinetics, Merck Research Laboratories, Kenilworth, NJ 07033, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Structural features of the substituted 4-piperidinyl urea analogs 1, responsible for the H3 antagonist
activity, have been identified. Structure–activity relationship of the H3 receptor affinity, hERG ion chan-
nel inhibitory activity and their separation is described. Preliminary pharmacokinetic evaluation of the
compounds of the series is addressed.
Received 10 December 2009
Revised 21 January 2010
Accepted 21 January 2010
Available online 28 January 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Histamine
H3 antagonists
N-Aryl-N0-benzyl-N-piperidinyl urea
hERG inhibition
Histamine acts biologically by interacting with four distinct his-
taminergic receptor subtypes, namely the H1, H2, H3, and H4
receptors. H3 receptors are Gi/o coupled, and are expressed primar-
ily in the CNS on histaminergic neurons, which project into all
main areas of the brain and part of the spinal cord. The H3 receptor
was first recognized as a presynaptic autoreceptor inhibiting hista-
mine synthesis and release via negative feedback.1 In addition, the
H3 receptor is involved in regulation of other neurotransmitter
systems, such as glutamate, acetylcholine, norepinephrine, dopa-
mine, GABA, and serotonin, owing to its expression on other types
of neurons as a presynaptic heteroreceptor,2 thus adding to the
potentially broad application of H3-based drug therapy. Evidence
from in vivo studies in animals suggests that H3 receptor ligands
may be useful in the treatment of CNS disorders, characterized
by deficient neurotransmitter signaling, such as ADHD, Alzheimer
disease, and sleep-related disorders.3 Additionally, given the
known role of central histamine in the control of appetite, H3
receptor ligands can be expected and have been reported active
in animal models of obesity.4 Finally, H3 receptor-mediated mod-
ulation of norepinephrine levels in the sympathetic nervous sys-
tem has been proposed as a means of maintaining vascular tone
and relieving nasal congestion associated with allergic rhinitis.5
While currently there are no marketed drugs acting at the H3
receptor, several companies have candidates in phase I and phase
II clinical trials for dementia, narcolepsy, neuropathic pain, ADHD,
schizophrenia, obesity, diabetes, and allergic rhinitis.3,6
Cardiovascular side effects—specifically, QT prolongation in
monkeys—have been previously quoted as a major reason for the
discontinued development of at least one promising clinical H3
candidate, ABT-239.7 Attempting to foresee this issue at the lead
optimization stage translates into the in vitro monitoring of hERG
channel inhibitory activity for the structural series of interest.8
Minimization of this activity as part of the ongoing H3 antagonist
lead optimization program in the series defined by the 4-piperidi-
nyl urea scaffold 1 is described.
R3
H
N
N
R1
O
N
R2
1
Work in the current series was initiated after N-aryl-N0-benzyl-
N-piperidinyl urea 2 was identified in screening as a structurally
novel lead with moderate H3 in vitro activity (Ki (hH3R) = 36 nM).
However, the high level of the accompanying hERG inhibition,
initially suggested by the high-throughput Rb efflux screen9 (see
Table 1), necessitated efforts to minimize the potential liability of
the series. Investigation of the three peripheral regions of the mol-
ecule, R1, R2, and R3, while maintaining the piperidinyl urea core of
1 intact, was undertaken.
The SAR trends are presented in Table 1. In addition to the Rb
efflux assay, most promising and/or prototypical compounds were
characterized in the medium-throughput hERG IonWorks assay.10
* Corresponding author. Fax: +1 908 740 7152.
E-mail addresses: michael.berlin@merck.com, michael.berlin@spcorp.com (M.
Berlin).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.01.121

2360
M. Berlin et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2359–2364
Table 1
H3 receptor affinity, Rb efflux, and hERG IonWorks SAR in the 4-piperidinyl urea series 1
% Rb efflux inhb (5
lg/mL)
hERG IonWorks IC50 (nM)
c
Structure
Kia (H3), nM
Br
Cl
H
N
2
3
4
5
6
7
36
93
61
N
Cl
O
O
N
N
Br
Cl
Cl
H
N
191
47
10
14
7
95
N
Br
N
MeO
H
N
78
O
N
Br
H
N
67
178
188
202
N
O
N
Br
NDd
H
N
N
O
N
Br
F
H
N
45
N
O
N
Br
Cl
Cl
8
9
23
83
42
15
H
N
N
O
N
Cl
F
F
7
6650
9190
H
N
N
O
O
N
Cl
N
10
41
H
N
N
N

M. Berlin et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2359–2364
2361
Table 1 (continued)
% Rb efflux inhb (5
lg/mL)
hERG IonWorks IC50 (nM)
c
Structure
Kia (H3), nM
F
F
11
447
18
H
N
N
O
O
N
N
F
H
N
N
12
13
>1000
13
OH
F
F
560
ND
H
N
N
N
O
O
N
N
Br
14
15
>1000
>1000
ND
H
N
H
N
F
9
H
N
N
N
O
N
OH
F
F
F
16
17
>1000
>1000
ND
H
N
N
O
O
O
N
Cl
1
H
N
N
NH
Cl
H
N
18
858
0
N
N
S
O O
Cl
F
F
19
20
86
67
11
14
12,700
H
N
N
O
O
N
OH
OH
Br
7880
H
N
N
N
(continued on next page)

2362
M. Berlin et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2359–2364
Table 1 (continued)
% Rb efflux inhb (5
lg/mL)
hERG IonWorks IC50 (nM)
c
Structure
Kia (H3), nM
Cl
F
21
22
23
>1000
4
H
N
N
OH
OH
O
O
N
N
Cl
F
F
H
N
>1000
5
N
H
N
O
Cl
>1000
33
H
N
N
O
N
N
Cl
F
F
F
F
24
25
26
27
28
16
624
3
46
28
94
2
3810
H
N
N
N
O
O
O
O
N
N
N
N
Cl
OH
H
N
N
N
Cl
384
NH2
N
H
N
N
Cl
49
12,000
H
N
N
N
N
Br
2
66
38
2310
H
N
N
N
O
O
N
N
Br
29
10
13,800
H
N
N
N
N
a
Recombinant human receptor binding assay. Values are means of two determinations. The range of duplicate values was typically less than threefold.
Rb efflux assay. Values are means of at least two experiments. The standard deviation was generally no more than twofold.
hERG IonWorks assay. Values are means of three experiments. The standard deviation was no more than 30%.
Not determined.
b
C
d

M. Berlin et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2359–2364
2363
13
e, d, f
R3
HN
a
b, c, d
e, d
R3NH2
2-7, 12
8-11, 14
N
BOC
g
e, d, b
d, b
16
15
18
h
k
23-25, 27-29
g
17
k, b
19, 20
26
i
j
21
22
Figure 1. Synthetic approaches to the compounds of Table 1. (a) 1-BOC-4-Piperidone, NaCNBH3, Ti(O-i-Pr)4, CH2Cl2; (b) TFA, CH2Cl2; (c) cyclopentanone, NaBH(OAc)3, CH2Cl2;
(d) R1–NCO, Et3N, THF, reflux; (e) 1-ethyl-4-piperidone, NaBH(OAc)3, CH2Cl2; (f) TBAF, THF; (g) BrCH2CH2OH, Et3N, THF, reflux; (h) CH3SO2Cl, Et3N, CH2Cl2; (i)
BrCH2CH(OH)CH2OH, K2CO3, DMF, 70 °C; (j) Et-NCO, Et3N, THF; (k) Ar-CHO, NaBH(OAc)3, CH2Cl2.
As discussed below, compounds from the series displayed similar
trends in both assays, while a favorable combination of data from
both assays was relied upon for compound selection. It was as-
sumed from the beginning that viability of the series depended
on the ability to maintain low nanomolar H3 receptor affinity. A
benzyl group on the monosubstituted urea nitrogen (R1) appeared
to be well positioned for preserving H3 potency, while a 4-fluoro
substituent was identified as a comparable alternative to hydro-
gen, both somewhat superior to the bulkier options (compounds
2–5, 7). Although some examples in which R2 is an ethyl group
showed a noticeable advantage over cyclopentyl analogs with re-
gard to H3 receptor affinity (e.g., 8 vs 3), overall these two substit-
uents appeared comparable and were used interchangeably. The
original 4-bromophenyl substituent, present in 2, could be re-
placed with 4-chlorophenyl (e.g., 9), but otherwise could not be
changed without drastic reduction in H3 receptor affinity. In par-
ticular, while a reduction in lipophilicity was achieved by transi-
tion to 4-fluorine (11), 4-hydrogen (12), or 4-hydroxymethyl (13)
and was counted on to reduce hERG inhibitory activity, a signifi-
cant drop in H3 activity was observed.
reduction in hERG inhibition. In fact, compounds 29 and 27
(clog P = 2.98) have the lowest clog P values among compounds
in Table 1, except for the completely H3-inactive 16. While this
clog P–IonWorks IC50 correlation is observed for compounds of
functional continuity—such as haloaryl analogs 7 (clog P = 5.27)
and 9 (clog P = 4.44))—clog P values do not seem to predict well
the effects of the noncontinuous functional changes, also previ-
ously referred to as ‘discrete structural modifications’.11 For exam-
ple, 2-aminopyridine 26 shows substantially higher activity in the
hERG IonWorks assay than 2-unsubstituted pyridine 24, despite
the lower clog P value (3.88 vs 4.20, respectively). Furthermore,
one could argue that the anticipated effect of hydroxy group in
19 is overpredicted by its lower clog P value (3.37) relative to the
deshydroxy analog 9 (clog P = 4.44), considering their comparable
IonWorks profiles. Overall, the link between the physicochemical
properties and hERG profiles of the compounds of the current ser-
ies appears traceable but tentative. Synthetic approaches to the
compounds of Table 1 are shown in Figure 1.
Preliminary pharmacokinetic evaluation revealed oral bioavail-
ability of 29 in the rat (AUC0–6 h = 3020 h nM (10 mg/kg, po)),13
although with a half-life just close to 1 h. PK limitations of 29 ap-
pear to be likely caused by rapid metabolism rather than inade-
quate absorption as suggested by high in vitro permeability and
solubility characteristics (Caco-2: 269 nm/s; aqueous kinetic solu-
The region of the molecule adjacent to the piperidine ring (R2)
was then considered for structural modifications. In view of the
strongly lipophilic nature of R1 and R3 possibly contributing to
the hERG activity, an overall increase in hydrophilicity, coupled
with the attenuation of basicity of the piperidine nitrogen, seemed
to be a prudent approach.11 Attention was primarily concentrated
on the 4-chlorophenyl subseries, derived from the piperidine 17,
with an anticipated advantage in hERG profile over the bromo ana-
logs, as suggested by IonWorks data on analog 9.
bility (pH 7.4) >250 lM). However, more in-depth pharmacoki-
netic or metabolic studies were not conducted at this point. The
profile of 29 clearly benefits from the increased polarity. Thus,
more lipophilic analogs 7 and 10 produced lower exposures after
the same oral dose in the rat (AUC 0–6 h of 250 and 340 h nM,
respectively), while still showing no sign of potential problem with
permeability (Caco-2: 116 and 72 nm/s, respectively). Compound
29 was selected for further profiling, results of which will be re-
ported in due course. Among other issues, in vivo implications of
the reduced hERG inhibitory activity in vitro would certainly be
of major interest.
In conclusion, a novel series of H3 antagonists, based on the 4-
piperidinyl urea core, has been identified. Excellent separation be-
tween H3 activity and hERG ion channel inhibition has been
achieved, and preliminary oral bioavailability in rats has been
demonstrated.
As can be seen from Table 1, H3 SAR of the R2 region remains
tight, similarly to the other parts of the molecule. Most of the sub-
stitutions, targeted to increase the hydrophilicity of the molecule
or reduce the pKa of the piperidine nitrogen, resulted in a drastic
reduction of H3 activity (e.g., 18, 21, 22). b-Hydroxyethyl substitu-
tion (19 and 20) was a notable exception. In a further effort to
improve H3 receptor affinity, select heterocycles—in particular,
4-aza-6-membered heteroaryls—were identified as the best option.
The extent of their substitution, however, was greatly limited by
the combined H3 and hERG SAR requirements, as evidenced by
compounds 25 and 26. While the basic 2-amino group in 26 re-
tained H3 activity, it negatively affected the hERG profile. Ulti-
mately, compound 29 appeared to offer the best balance
between H3 activity and reduced hERG inhibition, with the effect
of 4-bromophenyl substituent largely neutralized by the pyrida-
zine ring. Not surprisingly, hERG activity of the current series ap-
pears to correlate positively with the overall lipophilicity.11 As
References and notes
1. Arrang, J. M.; Garbarg, M.; Schwartz, J. C. Nature 1983, 302, 832.
2. Schlicker, E.; Kathmann, M. In The Histamine H3 receptor: A Target for New
Drugs; Leurs, R., Timmerman, H., Eds.; Elsevier, 1998; pp 13–26.
3. Sander, K.; Kottke, T.; Stark, H. Biol. Pharm. Bull. 2008, 31, 2163.
4. Malmlof, K.; Hastrup, S.; Wulff, S. B.; Hansen, B.; Peschke, B.; Jeppesen, C. B.;
Hohlweg, R.; Rimvall, K. Biochem. Pharmacol. 2007, 73, 1237.
gauged by clog P,
a drastic lipophilicity decrease from 2
(clog P = 6.56)12 to 29 (clog P = 2.98) is accompanied by a sharp

2364
M. Berlin et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2359–2364
5. McLeod, R. L.; Mingo, G. G.; Herczku, C.; DeGennaro-Culver, F.; Kreutner, W.;
Egan, R. W.; Hey, J. A. Am. J. Rhinol. 1999, 13, 391.
6. http://www.clinicaltrials.gov/.
7. Hancock, A. A. Biochem. Pharmacol. 2006, 71, 1103.
8. Lagrutta, A. A.; Trepakova, E. S.; Salata, J. J. Curr. Top. Med. Chem. 2008, 8, 1102.
9. Sorota, S.; Zhang, X. S.; Margulis, M.; Tucker, K.; Priestly, T. Assay Drug Dev.
Technol. 2005, 3, 47.
10. Bridgland-Taylor, M. H.; Hargreaves, A. C.; Easter, A.; Orme, A.; Hanthorn, D. C.;
Ding, M.; Davis, A. M., et al J. Pharmacol. Toxicol. Methods 2006, 154, 189.
11. Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G. J. Med. Chem. 2006, 49, 5029.
12. An algorithm from Biobyte, Inc. was used for log P calculations.
13. Dunn-Meynell, K.; Nardo, C.; Wainhaus, S.; Korfmacher. Abstracts, 38th Middle
Atlantic Regional Meeting of the American Chemical Society, Hershey, PA, Jun
4–7, 2006.