Obesity is a major risk factor in the modern world and associated
with many serious diseases of the metabolic syndrome such as
type 2 diabetes, dyslipidemia, coronary heart disease, and stroke.1
The melanin-concentrating hormone (MCH), a cyclic 19-amino
acid polypeptide, has been in the focus of obesity research over
the recent years.2 It is expressed in the lateral hypothalamus of
the brain and the natural ligand for the seven-transmembrane G-
protein-coupled receptors MCH-R1 and MCH-R2. MCH-R1 is
involved in the regulation of feeding and energy homeostasis and
has therefore been considered an interesting target for the
treatment of obesity over the years. While MCH-R1 is expressed
in humans and rodents, less is known about the exact function of
MCH-R2 which is also expressed in humans and other species,
but not in rodents. Despite many efforts of the pharmaceutical
industry to develop MCH-R1 antagonists as potential anti-obesity
agents, only few compounds advanced to the phase 1 clinical
stage.3 Safety concerns such as hERG inhibition or
phospholipidosis4 were among the prominent reasons for the
discontinuation of preclinical research programs in the past.5,6
Thus, there remains a high therapeutic need for the development
of safe and effective MCH-R1 antagonists.
lipophilicity of the compounds a series of derivatives based on a
central pyridazine scaffold rather than the phenyl moiety was
synthesized (Scheme 2). Starting from pyridazine 17,
Sonogashira coupling and subsequent hydrogenation furnished
building block 18. 2-Chloro pyridazine 18 could be further
modified by either Suzuki couplings utilizing aryl boronic esters
or nucleophilic displacement reactions giving rise to Boc-
protected 3-pyridazinyl-propylamines 21. Alternatively, the order
of reactions can be reversed using triflates 199 or iodides 209 as
starting material. After deprotection, Buchwald coupling with
aryl bromides 22 yielded the final products 23 in acceptable to
good yield.10
Pyridazine 24 displays a significantly reduced calculated
lipophilicity and is still a highly potent and selective MCH-R1
binder (Table 2). Gratifyingly, no hERG inhibition was observed
for this compound at 1 µM. However, first phospholipidogenic
effects4 were observed for this compound at a concentration of
6.25 µM in a cellular test system.11 According to Ploemen at al.8
the phospholipidogenic potential can depend on the polarity and
basicity of the compounds. We therefore decided to synthesize
additional compounds with further reduced lipophilicity (as
judged by the clogP) and basicity. Modifying R3 (Table 2) led to
compounds 25-29 with improved polarity but in most cases
reduced potency. Only the 4-methoxy-substituted derivative 26
showed comparable activity and was used for further
explorations. As a next step we explored the SAR around the
basic moiety R1R2N with a special emphasis on improving the
polarity further. Replacing the piperidine moiety with smaller
residues such as pyrrolidine 30 or dimethyl amine 31 led to
reduced affinity. While neutral compounds did not show binding
to the MCH-R1, compounds with reduced basicity such as
morpholine 32 retained potency. We decided to keep 6-
membered amines as preferred residues and explored the effect of
ring substitution further. In combination with starting point 24
compounds 32-35 differ in up to 3 clogP units and represent a
remarkable example on how to improve compound
characteristics by property-based design. All compounds display
single-digit nanomolar potency and excellent selectivity over the
M1 and 5-HT2A receptors. By consequently reducing the clogP to
a value of 2.2 a reduction of the phospholipidogenic potential
into a concentration range of 100 µM could be achieved. We
reasoned that polarity is clearly the decisive factor influencing
phospholipidosis in our series of compounds rather than the
basicity (see Table 2). Using the 4-substituted piperidines 34 and
35 as a basis for further modifications, additional derivatives
such as 36, 37, and 38 with an overall excellent profile with
respect to potency, selectivity, hERG inhibition (IC50 >10 µM for
38) and phospholipidosis (first effects >100 µM for 38) could be
identified.
Cl
A
A
1
R1R2N
O
Cl
A
H
N
A
R1R2N
2
O
Figure 1. Design of the 4-atom linker series
As part of our ongoing research program aimed at the
identification of potent MCH-R1 antagonists7, compounds such
as 2 were designed with a modified topology that, compared to
compound 1, positions the left-hand side aryl moiety differently
(Figure 1). We reasoned that modifying the position of the aryl
moiety could help overcome problems with respect to hERG
inhibition and phospholipidosis (PL) discovered in series 1, as
the position of lipophilic moieties is often an integral part of the
hERG and PL pharmacophore8.
Compound 4 is a potent MCH-R1 binder, but insufficient
selectivity over M1 was an initial concern for this probe (Table
1). Several related 4-atom linkers were synthesized to screen for
improved lead compounds (Scheme 1). Reducing the triple bond
(compounds 5 and 7) as well as introducing additional hetero
atoms (compounds 8 and 10) into the linker led to reduced
affinity to the receptor. In addition, selectivity over M1 remained
poor for these derivatives and binding to 5-HT2A was identified
as additional off-target activity. Removing the carbonyl group
from 7 finally resolved the problem and furnished compound 3
with single-digit nanomolar potency and improved selectivity
over M1 and 5-HT2A as a new lead to the MCH-R1 program. Of
note, N-methylated derivatives such as 6 or 9 showed reduced
affinity and were not followed up further.
Parallel to the 3-pyridazinyl-propylamine series we explored
the SAR of the related 3-pyridazinyl-propylethers (Table 3). The
synthesis of the compounds is described in Scheme 3. Employing
propargylic ether 488 in the Sonogashira reaction/hydrogenation
sequence, central intermediates 49 could be derived. Introduction
of the basic moieties R1R2N was achieved by either mesylation of
the benzylic alcohol and subsequent nucleophilic substitution
reaction or by transformation into the aldehyde and reductive
amination furnishing compound series 50. SAR in the 3-
pyridazinyl-propylether series were comparable to those in the
propylamine series. Substituted piperidines such as 39 and 40
were identified as potent and selective binders to MCH-R1.
Modification of R3 delivered additional potent compounds such
as 42, 43, and 45-47, however the correct combination of optimal
left-hand and right-hand side is crucial for good potency (for
comparison see 41 and 44). Selectivity over M1 and 5-HT2A, as
Compound 3 displayed a rather unattractive clogP of 7.7 and
did show strong inhibition of the hERG channel at a
concentration of 1 µM (Table 2). In order to reduce the
Roth, Gerald J.
