Evaluation of Amides, Carbamates, Sulfonamides, and Ureas of 4-Prop-2-ynylidenecycloalkylamine as Potent, Selective, and Bioavailable Negative Allosteric Modulators of Metabotropic Glutamate Receptor 5

Article abstract of DOI:10.1021/acs.jmedchem.8b01226

Negative allosteric modulators (NAMs) of the metabotropic glutamate receptor 5 (mGlu₅) hold great promise for the treatment of a variety of central nervous system disorders. We have recently reported that prop-2-ynylidenecycloalkylamine derivatives are potent and selective NAMs of the mGlu₅ receptor. In this work, we explored the amide, carbamate, sulfonamide, and urea derivatives of prop-2-ynylidenecycloalkylamine compounds with the aim of improving solubility and metabolic stability. In silico and experimental analyses were performed on the synthesized series of compounds to investigate structure-activity relationships. Compounds 12, 32, and 49 of the carbamate, urea, and amide classes, respectively, showed the most suitable cytochrome inhibition and metabolic stability profiles. Among them, compound 12 showed excellent selectivity, solubility, and stability profiles as well as suitable in vitro and in vivo pharmacokinetic properties. It was highly absorbed in rats and dogs and was active in anxiety, neuropathic pain, and lower urinary tract models.

Full text of DOI:10.1021/acs.jmedchem.8b01226

Subscriber access provided by University of Winnipeg Library
Article
Evaluation of amides, carbamates, sulfonamides and ureas of 4-
prop-2-ynylidenecycloalkylamine as potent, selective, and bioavailable
negative allosteric modulators of metabotropic glutamate receptor 5.
Davide Graziani, Silvia Caligari, Elisa Callegari, Carlo De Toma, Matteo Longhi,
Fabio Frigerio, Roberto Dilernia, Sergio Menegon, Luca Pinzi, Lorenza Pirona,
Valerio Tazzari, Anna Elisa Valsecchi, Giulio Vistoli, Giulio Rastelli, and Carlo Riva
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01226 • Publication Date (Web): 09 Jan 2019
Downloaded from http://pubs.acs.org on January 10, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Evaluation of amides, carbamates,
sulfonamides and ureas of 4-prop-2-
ynylidenecycloalkylamine as potent, selective,
and bioavailable negative
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
allosteric modulators of metabotropic
glutamate receptor 5.
a*
a
a
a
a
Davide Graziani , Silvia Caligari , Elisa Callegari , Carlo De Toma , Matteo Longhi ,
a
a
a
b
a
Fabio Frigerio , Roberto Dilernia , Sergio Menegon , Luca Pinzi , Lorenza Pirona ,
a
a
c
b
a
Valerio Tazzari , Anna Elisa Valsecchi , Giulio Vistoli , Giulio Rastelli and Carlo Riva .
a
Drug Discovery Department, Recordati S.p.A. Via M. Civitali 1, 20148 Milan (Italy).
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
b
Department of Life Sciences, University of Modena and Reggio Emilia, Via Campi
03, 41125 Modena (Italy).
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
Department of Pharmaceutical Sciences, Università degli Studi di Milano, Via
Mangiagalli 25, 20133 Milan, Italy.
Keywords
Metabotropic glutamate receptor 5, negative allosteric modulators, metabolism,
docking, solubility.
Abstract
Negative allosteric modulators (NAMs) of the metabotropic glutamate receptor 5
(mGlu ) hold great promise for the treatment of a variety of central nervous system
5
disorders. We recently reported that prop-2-ynylidenecycloalkylamine derivatives are
potent and selective NAMs of the mGlu receptor. In this work, we explored the amide,
5
carbamate, sulfonamide and urea derivatives of prop-2-ynylidenecycloalkylamine
ACS Paragon Plus Environment
2

Page 3 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds with the aim of improving solubility and metabolic stability. In silico and
experimental analyses were performed on the synthesized series of compounds to
investigate structure-activity relationships. Compounds 12, 32 and 49 of the
carbamate, urea, and amide classes, respectively, showed the most suitable
cytochrome inhibition and metabolic stability profiles. Among them, compound 12
showed excellent selectivity, solubility and stability profiles as well as suitable in vitro
and in vivo pharmacokinetic properties. 12 was highly absorbed in rats and dogs and
was active in anxiety, neuropathic pain and lower urinary tract models.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Introduction
The metabotropic glutamate 5 (mGlu ) receptor is a G protein-coupled receptor that is
5
highly expressed in the central nervous system (CNS). Indeed, it is mainly present in
the postsynaptic cortex, hippocampus, striatum, caudate nucleus, and nucleus
accumbens, usually colocalized with adenosine A , dopamine D , and it modulates
2A
2
the activity of NMDA and AMPA ionotropic glutamate receptors.1-5 Considering its
distribution in the central nervous system, the mGlu receptor has recently received
5
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
noteworthy attention as a promising drug target for a range of neurological disorders,
including anxiety, chronic pain, schizophrenia, levodopa-induced dyskinesia in
Parkinson’s disease, epilepsy and Fragile X syndrome.6-8
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The mGlu receptor, which belongs to class C of the G protein-coupled receptor family,
5
is characterized by a large extracellular N-terminal domain, named the “Venus flytrap
domain” (VFT), where the endogenous substrate (glutamate) interacts with the
orthosteric binding site. The allosteric binding site is located in the domain containing
the seven-transmembrane helices9,10, whose crystal structures (4OO9, 5CGD, 6FFH
and 6FFI), in complex with the negative allosteric modulators mavoglurant, HTL14242,
fenobam and M-MPEP, have recently been solved.11,12,13 A cysteine-rich region, which
links the VFT to the heptahelical transmembrane domains, is crucial for the
intramolecular conformational rearrangements induced by the mGlu -glutamate
5
interaction and hence for the enzymatic activity of the protein.14
Upon the interaction of glutamate with the orthosteric binding site, the mGlu receptor
5
mediates a plethora of important signaling pathways, such as inositol-1,4,5-
ACS Paragon Plus Environment
4

Page 5 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
triphosphate (IP ) and DAG messengers coupled with G
proteins and ERK 1/2,
3
q/11
cAMP and mTOR/PI3K signaling.15
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Ligands that interact with the mGlu allosteric binding site are able to modulate the
5
activity and efficacy of the endogenous orthosteric ligand. The concomitant binding of
ligands at both the orthosteric and the allosteric binding sites may result in different
outcomes. More specifically, allosteric ligands that promote potentiation of the
endogenous signal (positive cooperativity) are classified as “positive allosteric
modulators” (PAMs), whereas ligands that lead to the attenuation or complete block
of the signal (negative cooperativity) are classified as “negative allosteric modulators”
(NAMs). In this context, it is also possible to identify a third class of ligands, known as
“neutral allosteric modulators” (NALs) or “silent allosteric modulators” (SAMs), which
bind to the allosteric site and show neutral cooperativity with the orthosteric agonist.16
A notable number of companies and institutions have been involved in the research
on new mGlu receptor ligands. Ligands that interact with the mGlu allosteric pocket
5
5
can be classified based on their structure into alkyne-based or non-alkyne-based
chemotypes. The presence and reactivity of the alkyne group has been associated
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
7
with the depletion of cellular glutathione (GSH) and consequent hepatotoxicity, while
some non-alkyne-based chemotypes, such as PF-06297470 and PF-06462894, were
reported to induce specific adverse events associated with delayed type IV skin
hypersensitivity.18 Despite several alkyne-based (ADX-48621, AFQ056, RO-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
19
20
21
4
917523) and non-alkyne-based (fenobam, RGH-618, VU0424238 ) candidates
(
Chart 1) progressing to clinical studies, no mGlu NAM is currently available for
5
clinical use. In particular, based on a consistent preclinical disease model,
mavoglurant (AFQ056) at Novartis and basimglurant (RO-4917523) at Roche were
both under development in clinical phase 2/3 for Fragile X syndrome treatment.
Unfortunately, both studies failed to meet the primary endpoints.22,23
In patients affected by Parkinson's disease, mavoglurant was investigated in phase 2
clinical trials for the attenuation of levodopa-induced dyskinesia, with inconclusive or
conflicting results.24,25 Basimglurant was studied at Roche in phase 2 clinical studies
as an adjunctive therapy for the treatment of major depression in patients with an
inadequate response to selective serotonin reuptake inhibitors (SSRIs) or serotonin-
norepinephrine reuptake inhibitors (SNRIs). Despite some positive secondary
ACS Paragon Plus Environment
6

Page 7 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
endpoints, basimglurant did not meet the primary endpoint on the Montgomery-Åsberg
26
Depression Rating Scale (MADRS). The analysis of previous failed clinical trials in
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Autistic Spectral Disorder (ASD) is ongoing and suggests an improved paradigm for
translational research in neurodevelopmental disorders, such as the preclinical use of
1) more than one genetic background, 2) additional knock-out animal models for
diseases and 3) results reported from at least two independent laboratories.27
28
Moreover, the discovery of biased negative allosteric modulators that preferentially
target the ERK cellular pathway could represent an opportunity to reduce adverse
events and offer advantages over the first generation of mGlu modulators, which may
5
be useful for the treatment of Fragile X syndrome.29
In a previous study,30 we reported several 3-nitropyridin-2-yl derivatives of 4-(3-
arylprop-2-ynylidene)piperidines to be potent NAMs of the mGlu receptor, including,
5
for example, compound 4. Moreover, we investigated whether substitutions on the
aromatic ring portion linked to the triple bond in the molecules can affect the binding
and the interaction of the molecule with the mGlu receptor. Compound 4 showed a
5
high binding affinity and functional activity but was affected by a potent interaction with
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
cytochrome isoforms CYP1A2 (IC50 < 0.74 µM) and CYP3A4 (IC50 = 4.5 µM), and
these characteristics and potential drug-drug interaction issues were common among
the entire class of 4-(3-arylprop-2-ynylidene)piperidines that were directly substituted
with aromatic derivatives. In this paper, we have explored the possibility of replacing
the nitro-substituted heteroaryl ring that interacts with the extracellular subpocket of
the allosteric site with different groups. To this aim, a series of carbonyl derivatives,
such as amides, carbamates and ureas, and sulfonamide derivatives of prop-2-
ynylidenecycloalkylamine, was investigated. Here, the synthesis, physicochemical
properties, in vitro pharmacological activity, pharmacokinetic and metabolic stability
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
data are reported. Ligand binding modes for the human mGlu receptor and human
5
CYP450 3A4 isoform are described. Potent and selective negative allosteric
modulators of the mGlu receptor, endowed with increased mGlu selectivity and
5
1
ameliorated drug-like properties, were obtained and compared with previously
described molecules. In particular, the (3E)-3-prop-2-ynylidenepyrrolidine derivatives
of the various investigated chemical classes (12, 32 and 49) were the compounds
endowed with high activity, drug-like in vitro profiles and were selected for preclinical
ACS Paragon Plus Environment
8

Page 9 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
investigations. Compound 12 was further studied and shown to be clean in preclinical
toxicological studies.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Results and Discussion
Chemical synthesis
MPEP (1), MTEP (2) and fenobam (3) (Chart 1) were purchased from Tocris
Bioscience. Compound 4 (Chart 1) and the key intermediates 2-methyl-6-[3-(piperidin-
4-ylidene)prop-1-yn-1-yl]pyridine (14a, Scheme 2) and compound 17 (Chart 2) were
30
prepared as described by Anighoro et al. The majority of the compounds described
in this paper belong to four different chemical classes: amides, carbamates,
sulfonamides and ureas.31 These are derivatives of either 4-(3-arylprop-2-
ynylidene)piperidine or 3-(3-arylprop-2-ynylidene)pyrrolidine. A small number of
molecules are N-alkyl derivatives of 4-(6-methylpyrid-2-ylprop-2-ynylidene)piperidine.
Compound 10 was synthesized (Scheme 1) from commercially available tert-butyl 3-
oxoazetidine-1-carboxylate, which was reacted with the ylide generated from diethyl
32
[
3-(trimethylsilyl)prop-2-ynyl]phosphonate to obtain the intermediate 10a. The latter
was converted into 10b by carrying out a Sonogashira-Heck cross-coupling reaction
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with 2-bromo-6-methylpyridine with concurrent removal of the trimethylsilyl group.
Afterwards, the BOC deprotection of 10b using TFA in chloroform resulted in 10c, and
then acylation of 10c with ethyl chloroformate afforded compound 10. Beginning with
tert-butyl 6-oxo-2-azaspiro[3.3]heptane-2-carboxylate and using an analogous
synthetic pathway, compound 11 was prepared. N-tert-Butoxycarbonyl-3-
pyrrolidinone was submitted to the same Horner-Wadsworth-Emmons (HWE)
procedure used to synthesize 10a and 11a after chromatographic separation of the
E:Z isomers (obtained with 2:1 ratio) tert-butyl-(3E)-3-[3-(trimethylsilyl)prop-2-yn-1-
ylidene]pyrrolidine-1-carboxylate (12a) and tert-butyl-(3Z)-3-[3-(trimethylsilyl)prop-2-
yn-1-ylidene]pyrrolidine-1-carboxylate (13a). The structure of the isomers was
elucidated by NOESY experiments (Supporting Information Figure S1). Afterwards,
alkynyl Heck coupling of 12a and 13a with 2-bromo-6-methylpyridine afforded the
compounds 12b and 13b, respectively, which, after the removal of the BOC protecting
group (12c and 13c), were acylated with ethyl chloroformate to give compounds 12
and 13. Compounds 14 and 16 were prepared (Scheme 2) by acylating the key
intermediate 14a with either methyl chloroformate or ethyl chloroformate. Ethyl-4-[3-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(trimethylsilyl)prop-2-yn-1-ylidene]piperidine-1-carboxylate (15a), an additional key
ACS Paragon Plus Environment
10

Page 11 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
intermediate, was prepared from 1-ethoxycarbonyl-4-oxopiperidine by the standard
HWE procedure with diethyl-[3-(trimethylsilyl)prop-2-ynyl]phosphonate. Then, a
Sonogashira-Heck coupling was carried out on 15a using the appropriate aryl or
heteroaryl halogenide to obtain compounds 15 and 21–24. The ureas 25–33, amides
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
34–49 and sulfonamides 50–57 were synthesized from the key intermediates 14a, 12c
and 47b according to the procedures shown in Scheme 3. The ureas 25–27 were
synthesized from 14a by reaction with the appropriate commercial isocyanate. 28–31
and 33 were prepared by the use of triphosgene, which was previously reacted with
the appropriate secondary amine in the presence of DIEA and then with 14a or 12c.
32 was prepared by a reaction between commercially available 3-methyl-1-
[
methyl(propan-2-yl)carbamoyl]-1H-imidazol-3-ium iodide and 12c. Amides 34, 40, 48
and 49 were synthesized by acylating 14a or 12c with the appropriate aroyl or acyl
chloride in the presence of TEA. 35, 36, 39, 41–43, and 46 were prepared from 14a
by condensation with the given carboxylic acid in the presence of EDCI and HOBt,
while 37, 38, 44, 45 and 47 were prepared from 14a or 47b by condensation with PS-
DCC in DCM. The intermediate 47b was synthesized (Scheme 2) by the standard
HWE procedure with diethyl [3-(trimethylsilyl)prop-2-ynyl]phosphonate led on tert-
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
butyl-4-[3-(trimethylsilyl)prop-2-yn-1-ylidene]piperidine-1-carboxylate to give the N-
BOC derivative 47a, which then underwent BOC removal to give 47b. Sulfonamides
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50–57 were obtained by condensing 14a with the proper sulfonyl chloride using TEA
as the base. The NaB(OAc) H reductive amination of 14a with either butyraldehyde or
3
benzaldehyde led to 58 or 59, respectively. 60 was obtained from 14a by reaction of
2-chlorocyclopentanone and DIPEA in DMA. A two-step one-pot synthesis of 61 was
carried out by N-alkylation of 14a with 2-bromoacetic acid and TEA, followed by
amidification with pyrrolidine and polymer-supported by dicyclohexylcarbodiimide as
the condensing agent. The full analytical characterization of compound 12 and its
mesylate salt 12s is available in the Supporting Information (Figures S10–S16).
Biological results
Binding and functional results
The newly synthesized compounds were evaluated for their ability to displace the
3
allosteric antagonist [ H]MPEP from the binding site and for their binding affinity for
the membranes of CHO T-REx cells that were stably transfected with human mGlu5
receptors. The binding affinities, expressed as inhibition constants (Ki), were obtained
ACS Paragon Plus Environment
12

Page 13 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
by three independent experiments performed in triplicate and are reported in Tables
1-4. Specifically, Tables 1 to 4 report the biological data of the carbamate, urea, amide
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
and sulfonamide derivatives, respectively. Table S5 (Supporting Information) reports
3
the biological data of the alkyl derivatives. The displacement curves of [ H]MPEP
induced by compounds 12, 32, 49 and 55, which were considered to be the
prototypical compounds of each class of molecules, are shown in Figure 1. The Ki
values of compounds 12, 32, 49 and 55 are 3.5 nM, 8.6 nM, 7.9 nM, and 2.7 nM,
respectively.
Moreover, Tables 1-4 and T5 show the results of the functional Ca2+ mobilization
assay measured in the CHO T-REx cells with cloned human mGlu receptors, which
5
was determined only for compounds with a Ki affinity for the mGlu receptor of less
5
than 20 nM. Figure 2 shows the ability of compounds 12, 32, 49 and 55 to inhibit the
2+
quisqualate-induced intracellular calcium concentration [Ca ] response of the human
i
mGlu receptor. Under these experimental conditions, compounds 12, 32, 49 and 55
5
displayed the following IC50 values: 4.1 nM, 15.1 nM, 6.4 nM, and 43.8 nM,
respectively.
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The rat mGlu binding affinity was measured using CHO T-REx cell membranes with
1
3
33
mGlu receptor stably transfected and as the [ H]R214127 radiolabel. The functional
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Ca2+ activity of the rat mGlu receptor using the same cell line was determined for
1
compounds that showed an mGlu /mGlu binding ratio lower than 50-fold and for
1
5
compounds endowed with an attractive overall in vitro profile. Despite the difference
34
of only four residues between the mGlu and mGlu binding pockets, none of the
5
1
molecules discussed in this article showed an mGlu /mGlu binding ratio lower than
1
5
50-fold. These results are in agreement with the interaction of the ligands in the lower
34
intracellular subpocket (Figure 5), which is specific to the mGlu receptor.
5
30
While the “ene-yne portion” was described in the previous article as one of the most
suitable spacers for high-affinity mGlu ligands, in this article, we have (i) investigated
5
the effects of modulating the size of the ring condensed to the terminal double bond,
(ii) explored the substituents on the nitrogen atom of the condensed ring, and (iii)
studied the substitution on the triple bond. The substituents at the meta position of the
aryl group linked to the triple bond have been reported to enhance the mGlu affinity
5
35
and activity of most mGlu NAMs. Consistently, we kept the 6-methylpyridine ring or
5
ACS Paragon Plus Environment
14

Page 15 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a meta-substituted phenyl ring as established substituents of the triple bond. Our work
was focused on exploring the substitution of the nitrogen in the heterocyclic ring via
derivatization with a variety of carbonyl group derivatives, i.e., carbamates (10–24,
Chart 2), ureas (25–33, Chart 3) and amides (34–49, Chart 4), instead of the previously
described 3-nitropyridine ring (compound 4). The carbonyl group was proposed to be
a putative hydrogen bond acceptor (HBA) replacement of the nitro group, which was
previously suggested to take part in hydrogen bonding interactions with Asn747.30
Furthermore, a series of sulfonamide derivatives (50–57, Chart 5) were synthesized
as bioisosteres of the carbonyl group. The first aim of our work was to define the most
suitable ring bonded to the -ene moiety; consequently, different sized rings containing
nitrogen, including pyrrolidine, piperidine and the more tensioned 4-membered ring
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
azetidine, were considered. The mGlu affinities of the N-ethoxycarbonyl compounds
5
10 (azetidine), 12 (E-pyrrolidine) and 16 (piperidine) were 38.3 nM, 3.5 nM and 2.5
nM, respectively. The Z-pyrrolidine geometric isomer 13 was 69 times less active than
the E isomer 12. The affinities lower in the 4-membered rings than in the 5-membered
and 6-membered rings. These affinities could be related to the distance between the
pyridine ring centroid and the carbonyl HBA group, which is lower in compound 10
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
than in compounds 12 (approximately 1.5 Å) and 16 (approximately 0.5-1.0 Å
depending on chair or twisted piperidine conformation), where the latter appears to be
optimal for establishing interactions with the binding pocket. Compound 11, obtained
from 2-azaspiro[3.3]heptan-6-one, showed a binding affinity comparable to 16 (3.6 nM
and 2.5 nM, respectively). The replacement of the ethyl group with tert-butyl (17, Ki =
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
11.9 nM), benzyl (19, Ki = 4.8) or phenylethyl (20, Ki = 2.5 nM) groups resulted in
NAMs that exhibited high mGlu affinity. The only relevant exception was the
5
cyclobutylmethyl analogue 18, which was inactive or poorly active (Ki > 1000 nM).
Molecules belonging to the carbamate series showed remarkable agreement between
2+
binding affinity and functional quisqualate-induced Ca activity in the human mGlu
5
receptors. The ratio of functional vs binding activity was lower in small O-substituents
of carbamates (methyl (14), ratio = 1.5; ethyl (16), ratio = 1.3; ethyl (21), ratio = 2.5
folds) and increased with bulkier substituents (benzyl (19), ratio = 51; phenylethyl (20),
ratio = 20). Ethyl carbamates 21–24, each carrying different meta-substituted phenyl
rings at the alkyne moiety, were all potent nanomolar ligands.
The urea analogues were less potent than the carbamates. A direct comparison
between the tert-butyl carbamate 17 (Ki = 11.9 nM) and the tert-butyl urea 25 (ki = 399
ACS Paragon Plus Environment
16

Page 17 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
nM) suggests that the presence of an HBD in ureas is detrimental to binding affinity.
This finding suggests that there may be repulsion between the HBD moieties of ligands
and Asn747 and further reinforces the requirement of an HBA for potent affinity. N-
Methylation restored the binding affinity, as shown by the comparisons of 26 (Ki = 92
nM) with 29 (Ki = 20.4 nM) and of 27 (Ki = 276 nM) with 30 (Ki = 4.3 nM). The best
affinity value was obtained by modulating the bulkiness of the alkyl substituents (31
vs. 25 and 28); the isopropylethylamino derivative (31) showed a Ki of 2.9 nM. The
pyrrolidino-analogues isopropylmethylamino 32 and isopropylethylamino 33 showed
high binding affinity (8.6 nM and 5.8 nM, respectively) together with promising in vitro
profile characterized by no interaction with the investigated cytochrome isoforms (see
Table 2).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The mGlu functional/binding activity ratios of 32 and 33 (1.7- and 5-fold, respectively)
5
were less than that of 31 (45.8-fold). Although the two compounds displayed binding
2+
affinities less than 31, they were significantly more active in the Ca functional assay
(15 and 29 nM).
Alkyl amides 34–39 showed moderate affinity (Ki > 20 nM); therefore, they were not
profiled further. Specifically, the 2-methoxyacetyl and the 2-pyrrolidinacetyl groups of
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
6 and 37 were detrimental to affinity. The 2-phenacetamide and 3,5-
difluorophenylacetamide derivatives 38 and 39 were less potent (83 nM and 194 nM),
while the benzamide analogue 40 exhibited high binding affinity (18.4 nM). The
presence of an aromatic ring directly bonded to the carbonyl group in compounds 40–
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
49 resulted in nanomolar affinity. A meta substitution on the phenyl ring (42, 44–48)
showed similar or further increased affinity for the mGlu receptor (46, Ki = 0.7 nM).
5
From the limited set of data, it seems that the ortho and especially para methoxy
substituents of amides (41 and 43) negatively affected binding affinity. Compounds 44
and 46 showed greater functional/binding activity ratios (122- and 70-fold,
respectively) because of their different binding affinities (Ki = 11.3 nM and 0.7 nM,
respectively) and Ca2+ functional assay results (IC50 = 1383 nM and 49.2 nM,
respectively). Sulfonamides were synthesized as bioisosteres of the carbonyl groups
that potentially interact with the Asn747 residue of the binding pocket. The presence
of additional oxygen and the different hybridization states of the carbonyl vs. sulfonyl
groups could play a role in the hypothesized bioisosteric effect.36 In general,
sulfonamides should have an H-bonding basicity (pKBHX value ranging from 1.1 to 1.4)
capable of establishing a strong H bond with Asn747. Considering that carbamates,
ACS Paragon Plus Environment
18

Page 19 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
amides and ureas have higher pKBHX values than sulfonamides (1.5–3.0) depending
on substitution and cyclization,37 the rank of affinities and activities obtained are in
agreement with pKBHX values. The secondary alkyl sulfonamides (50–52) exhibited
weak activity. The aromatic sulfonamides 53–57 showed a higher binding affinity, even
though we found the highest affinity with para substitution and not with meta
substitution, as was found for the amide derivatives (55, 56 and 57 versus 54).
Regarding structure-activity relationships (SAR), it is interesting to observe the
different behavior of amides and sulfonamides. With sulfonamides, the strongest
affinity was obtained with para-substituted compounds; however, for amides, the
strongest affinity was obtained with meta-substituted compounds. This could be due
to the different dihedral angles imposed by the sulfonamide (approximately 103
degrees) versus amide (approximately 120 degrees) groups. The last series of
molecules synthesized included several N-alkyl derivatives (Chart 6). The presence of
the carbonyl group of carbamates, ureas or amides was essential for affinity.
Compounds lacking the carbonyl group (Supporting Information Table S5) (see 58 and
59 vs 34 and 40) proved completely inactive, confirming the importance of an HBA to
interact with Asn747 for high binding affinity. Compound 60, in which the carbonyl
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
group was separated by an additional sp carbon compared to that of the amides,
showed high binding affinity (25.7 nM). On the other hand, 61, in which the carbonyl
group is not directly bonded to the piperidine nitrogen but instead is separated by an
additional sp3 carbon atom, resulted in a completely inactive class of compounds.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cytochrome inhibition
1
(MPEP) and 2 (MTEP), even though they are very selective for the mGlu receptors,
1
CYP2D6 and CYP3A4 (see Table 1), were described as moderate competitive
inhibitors of recombinant human CYP1A2 (Ki, 0.5–1 μM).38 The 3-nitropyridine
derivative 4 inhibited both the CYP1A2 and the CYP3A4 isoforms. Carbamates 10–24
were generally affected by the same problem, displaying a micromolar inhibition of
CYP1A2 sometimes coupled with more relevant micromolar inhibition of CYP3A4. The
replacement of the 3-nitropyridine ring of 4 with small alkyl carbamates (14 and 16)
led to a clean CYP3A4 profile (IC > 20 µM), while the compounds bearing bulkier
50
alkyl groups, such as benzyl (19) and phenylethyl (20), moderately inhibited
cytochrome CYP3A4 (IC 7.7 µM and 3.7 µM, respectively). Lipophilicity was reported
50
ACS Paragon Plus Environment
20

Page 21 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
to correlate with the P450 binding affinity of inhibitors and substrates of CYP3A4.39
The replacement of the pyridine moiety bonded to the alkyne with a meta-substituted
phenyl ring (21–23) increased lipophilicity, lowered solubility and negatively affected
the CYP3A4 inhibition profile. The polar 3-cyano substituent (24) was beneficial for
the CYP3A4 interaction and for solubility (CYP3A4 IC >20 µM and solubility = 544
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
µM). For generally small and planar lipophilic substrates and inhibitors, the interaction
with CYP1A2 was unaffected by these latter changes.40
Although, in general, moderately bulky alkyl ureas generated active compounds, this
substitution inhibited cytochromes and overall in vitro profiles. Compounds 29, 30 and
31 were weak inhibitors of the CYP3A4 isoform, showing a IC values of 13.4, 4.0
50
and 15.4 µM, respectively. Again, a more favorable cytochrome inhibition profile could
be attributed to the pyrrolidine ring derivatives (32 and 33). Benzamide 40 and the
meta halo-substituted benzamides (45 and 46) were characterized by a clean profile
versus the three cytochrome isoforms. The aromatic sulfonamides 55–57 were devoid
of interaction with CYP1A2 (IC > 20 µM), while they retained moderate inhibition of
50
CYP3A4.
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
None of the molecules described and investigated in the current article showed
relevant inhibition of CYP2D6. CYP3A4 inhibition can be modulated by introducing the
pyridine ring or the cyano group on the phenyl ring bonded to alkyne or by modulating
the lipophilicity of the alkyl substituents of carbamates or ureas. The pyrrolidine moiety
had a favorable impact on cytochrome inhibition, and compounds 12, 32 and 49 al
exhibited good pharmacokinetic in vitro profiles.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Metabolic stability results
Tables 1–4 show that most compounds display very limited in vitro metabolic stability
in rat microsomes, while they exhibit acceptable stability in human microsomes.
Although some of these compounds have a clean CYP inhibition profile and high
human microsome stability, the rat microsome instability prevents in vivo testing in
rodent models. With the aim of identifying possible metabolites and elucidating
metabolic pathways to design stable molecules to be tested in vivo, metabolite
generation in rat microsomes and identification by HPLC-MS/MS were carried out on
three selected compounds (16, 31 and 46) belonging to different chemical classes
ACS Paragon Plus Environment
22

Page 23 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
carbamates, ureas and amides, respectively). The results are shown in the
Supporting Information (Figure S2). Compounds 16 and 31, in rat microsomes,
undergo primary and fast metabolic attack on the methyl pyridine ring. Two different
metabolites were identified for 16. The fragment with m/z 255 is specific to the 2-
methyl-6-[3-(4-piperidylidene)prop-1-ynyl]pyridine portion. Both metabolites of
compound 16 have the same mass spectrometric transition m/z 301.2/255.1,
corresponding to the N-oxide formation or to the hydroxylation at two different
positions on the same portion of the molecule (e.g., on pyridine and at its methyl
substituent). Urea 31 first undergoes oxidation of the pyridine portion with transition
m/z 342.3/255.1 and then undergoes a further transition (m/z 314.3/255.1) compatible
with urea de-ethylation. Compound 46, in rat microsomes, undergoes fast oxidative
metabolic attack and exhibits metabolites with different mass transitions (m/z
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
67.2/139.3), which are compatible with a metabolic oxidation attack of the 2-methyl-
6-[3-(4-piperidylidene)prop-1-ynyl]pyridine scaffold. Further fragmentations produced
+
the 3-chloro-N,N-diethyl-benzamide oxidized ([M] = 228.2), 3-chloro-N,N-diethyl-
+
+
benzamide ([M] = 212.2) and chlorobenzoyl ([M] = 139.3) fragments. The fragments
+
+
[
M] = 228.2 and [M] = 212.2 are not present in the mass spectrum of compound 46
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
without metabolism and are probably due to the initial oxidation at an unknown position
of the scaffold followed by nuclear decomposition. Reported in vitro stability studies
were performed to direct the synthesis of metabolically stable compounds, and no
further investigations were conducted to further characterize the nature of metabolites.
The results of the structure and stability relationships and the metabolite studies by
HPLC-MS/MS of the investigated compounds are substantially in agreement. The
metabolic attack for most of the molecules is directed on the 6-[3-(4-
piperidylidene)prop-1-ynyl]pyridine scaffold, which was subjected to quick oxidation,
and when the pyridine was substituted with a phenyl ring, the metabolic stability
increased for carbamates (21–24, 29-64% versus 16, 2.1%) and amides (47, 64%
versus 46 <1%). Further indications could be derived by comparing equipotent
compounds 11 and 16. Compound 16 shows low metabolic stability in rats, while
compound 11 is stable in both rat and human microsomes (80% and >95%,
respectively). The carbonyl groups of both molecules tend to be coplanar to one of the
two C-N bonds of the ring, and higher dihedral angles showed a net increase of
energy. Minima conformation and dihedral contour plots of amide bond rotation are
reported in the Supporting Information (Figures S3–S6). While the double bond and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
24

Page 25 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
carbonyl group of 11 lie on two different orthogonal planes, in 16 they are on two
parallel planes. Therefore, the spiralization and the carbonyl orientation do not affect
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
mGlu activity but instead impair rat microsome metabolism. 2-Azaspiro[3.3]heptane
5
exhibits good structural overlap with piperidine and can be used as a bioisostere to
protect piperidine from metabolic attack. The E-pyrrolidine analogue 12 showed
increased metabolic stability in both rat and human microsomes in comparison to the
piperidine derivative 16. The higher stability of pyrrolidine versus piperidine analogues
in rat microsomes was also confirmed in the other synthesized chemical classes. In
fact, pyrrolidine compounds 32 and 33 were significantly more stable (57 and 20%,
respectively) than 31 (<1%). Moreover, compound 31, according to the HPLC-MS
studies, undergoes two fast and concurrent oxidations, the first being on the pyridine
ring and the other on the ethyl chain of the urea, leading to de-ethylation. The latter
degradation is also possible for compound 33. The methyl-substituted urea 32 showed
a slower N-oxidative dealkylation, proving a more stable compound. Pyrrolidine
exhibited higher stability than piperidine, which was also corroborated by the amide
analogues. Pyrrolidine derivative 48 shows 55% remaining, while the stability of
piperidine analogue 46 is almost zero.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The aromatic amides are potent ligands for the mGlu receptor (42, 45, 46).
5
Unfortunately, these compounds, even while showing an acceptable profile, were
clearly unstable in rat and human microsomes (e.g., 46) because of oxidative attack.
Substitution of the pyridine ring with a meta-chlorophenyl ring leads to compound 47,
which was particularly stable in both rat and human species but displayed cytochrome
interactions. As previously discussed, the five-membered pyrrolidine ring derivatives
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(48 and 49) were stable in both human and rat microsomes. In particular, 49 showed
a very appealing overall profile consisting of high solubility, high binding affinity (7.9
nM), well-correlated functional activity in the calcium assay (6.4 nM) and a clean profile
for the most important cytochromes.
The last group of molecules investigated here was the piperidine sulfonamides, which
demonstrated instability in rat and human microsomes (55–57). Alkyl piperidine
derivatives (58–61) were almost inactive, with the exception of ketone 60, which
showed high stability in both rat and human microsomes. The direct comparison of 34
(rat % rem. 1%) and 58 (rat % rem. 22%) suggests that the carbonyl group is
necessary to obtain ligands with high mGlu affinity and that it affects the rate of rat
5
metabolism.
ACS Paragon Plus Environment
26

Page 27 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In conclusion, we observed that the main metabolic attack to these classes of
molecules is on the pyridine ring, and the rate of metabolism is affected by HBA (e.g.,
carbonyl moiety belonging to carbamate or ureas groups) and by its torsion angles in
the constrained rings present on the opposite side of the molecule. Further support for
these findings can be confirmed by exchanging the pyridine rings of derivatives with
phenyl rings (21–24 versus 16 and 47 versus 46). In fact, the phenyl derivatives 21–
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
24 and 47 possess metabolic stability in rats ranging from 29% to 64% and are
invariably more metabolically stable than the pyridine derivatives 16 and 46 in the
same species. The percent of a group of selected molecules (16, 31, 32, 41 and 47)
remaining after 30 minutes in the human CYP3A4 isoform was determined, and the
results are in substantial agreement with the obtained percent remaining in the human
microsomes (Table 5). Although the measured metabolism is likely due to an extended
set of metabolizing enzymes, including hydrolytic enzymes the act on labile groups
such as carbamates and amides, the metabolic stability of the reported compounds
was further investigated by performing human CYP3A4 docking simulations. In fact,
this cytochrome plays a particularly relevant role; this is confirmed in Table 5, in which
an encouraging agreement between the extent of metabolism exerted by human
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 161
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
microsomes and by the CYP3A4 enzyme alone is reported. Moreover, it should be
noted that the resolved structures of the rat CYP3A isoforms are not yet available, and
thus, this docking analysis is focused on human CYP3A4 only and is aimed at
rationalizing the different metabolic stabilities of each proposed class that was
observed in the human microsomes. As detailed in Supporting Information Experiment
E1 (Docking simulations of molecules as possible substrates of the human CYP3A4
isoform), the obtained docking results are in satisfactory agreement with the reported
metabolic data for both metabolic stability, which is related to the general propensity
of a given substrate to approach the heme group, and for the detected metabolites,
which are in line with the computed binding modes (Supporting Information Figures
S8-S9). Indeed, and as observed in the monitored metabolic pathways, the pyridine
ring is the moiety that often contacts the heme group, and the computed complexes
appear to be largely influenced by few polar interactions stabilized by the
carbamate/urea functions. Compounds 12, 32, and 49, which were characterized by
high rat and human stability, were selected for further investigation.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Analysis of compound solubility
ACS Paragon Plus Environment
28

Page 29 of 161
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Carbamates 10–14 and 16 were significantly more soluble (600-939 µM) than the 3-
nitropyridine derivative 4 (28 µM). The bulky and lipophilic carbamates 18 and 20 had
increased cLogP by approximately 1 unit compared to ethyl carbamates (see Table
S1), and their solubilities were comparable to that of compound 4 (12 and 31 µM). The
introduction of a phenyl ring in 21–24 caused a decrease in solubility in comparison
with the solubility of pyridine derivatives. In particular, the m-Me (21) and m-Cl phenyl
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(23) groups were detrimental for solubility (4 and 15 µM, respectively), while the small
m-F phenyl (22) and the polar m-CN phenyl (24) groups were able to restore high
solubility. The two urea compounds 32 and 33 were very soluble (962 µM and 927 µM,
respectively). The solubility of the aromatic amide derivatives was influenced by the
ring substituent position. Ortho (41) and meta (42) methoxy derivatives showed high
solubility (468 and 386 µM, respectively). Pyrrolidine and piperidine m-Cl benzoyl
derivatives 46 and 48 were sufficiently soluble (58 µM and 157 µM, respectively). The
furoyl substituent (49) was one of the most suitable aromatic fragments, lowering the
cLogP value (LogPGALAS 2.4) and conferring high aqueous solubility (420 µM).
Aromatic sulfonamides were insoluble (52, 54 and 57), while the cyclohexylmethyl
derivative 51 maintained high solubility (49 µM). The solubility of most of the
ACS Paragon Plus Environment
29