Topical Intestinal Aminoimidazole Agonists of G-Protein-Coupled Bile Acid Receptor 1 Promote Glucagon Like Peptide-1 Secretion and Improve Glucose Tolerance

Article abstract of DOI:10.1021/acs.jmedchem.6b01873

The role of the G-protein-coupled bile acid receptor TGR5 in various organs, tissues, and cell types, specifically in intestinal endocrine L-cells and brown adipose tissue, has made it a promising therapeutical target in several diseases, especially type-2 diabetes and metabolic syndrome. However, recent studies have shown deleterious on-target effects of systemic TGR5 agonists. To avoid these systemic effects while stimulating glucagon-like peptide-1 (GLP-1) secreting enteroendocrine L-cells, we have designed TGR5 agonists with low intestinal permeability. In this article, we describe their synthesis, characterization, and biological evaluation. Among them, compound 24 is a potent GLP-1 secretagogue, has low effect on gallbladder volume, and improves glucose homeostasis in a preclinical murine model of diet-induced obesity and insulin resistance, making the proof of concept of the potential of topical intestinal TGR5 agonists as therapeutic agents in type-2 diabetes.

Full text of DOI:10.1021/acs.jmedchem.6b01873

Subscriber access provided by University of Colorado Boulder
Article
Topical intestinal aminoimidazole agonists of G-Protein-
Coupled Bile Acid Receptor 1 promote Glucagon Like
Peptide-1 secretion and improve glucose tolerance.
Manuel Lasalle, Vanessa Hoguet, Nathalie Hennuyer, Florence Leroux, Catherine
Piveteau, Loic Belloy, Sophie Lestavel, Emmanuelle Vallez, Emilie Dorchies,
Isabelle Duplan, Emmanuel Sevin, Maxime Culot, Fabien Gosselet, Rajaa Boulahjar,
Adrien Herledan, Bart Staels, Benoit Deprez, Anne Tailleux, and Julie Charton
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01873 • Publication Date (Web): 17 Apr 2017
Downloaded from http://pubs.acs.org on April 18, 2017
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Topical intestinal aminoimidazole agonists of Gꢀ
ProteinꢀCoupled Bile Acid Receptor 1 promote
Glucagon Like Peptideꢀ1 secretion and improve
glucose tolerance.
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
†
#
†
#
‡
†
Manuel Lasalle , Vanessa Hoguet , Nathalie Hennuyer , Florence Leroux , Catherine
†
‡
‡
‡
‡
Piveteau , Loïc Belloy , Sophie Lestavel , Emmanuelle Vallez , Emilie Dorchies , Isabelle
‡
§
§
§
†
Duplan , Emmanuel Sevin , Maxime Culot , Fabien Gosselet , Rajaa Boulahjar , Adrien
†
‡
†
‡◊
†◊
Herledan , Bart Staels , Benoit Deprez* , Anne Tailleux Julie Charton*
†
Univ. Lille, Inserm, Institut Pasteur de Lille, U1177 ꢀ Drugs and Molecules for living Systems,
Fꢀ59000 Lille, France.
‡
Univ. Lille ꢀ EGID, Inserm, CHU Lille, Institut Pasteur de Lille, U1011 ꢀ Nuclear receptors,
cardiovascular diseases and diabetes, Fꢀ59000 Lille, France.
§
Univ. Artois, EA 2465 ꢀ Bloodꢀbrain barrier laboratory (LBHE), Fꢀ62300 Lens, France.
#
These authors contributed equally to this work.
◊
These authors contributed equally to this work.
KEYWORDS. TGR5, Aminothioimidazole, GLPꢀ1, glucose tolerance
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 103
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
ABSTRACT. The role of the GꢀProteinꢀcoupled Bile Acid Receptor TGR5 in various organs,
tissues and cell types, specifically in intestinal endocrine Lꢀcells and brown adipose tissue, has
made it a promising therapeutical target in several diseases, especially typeꢀ2 diabetes and
metabolic syndrome. However, recent studies have shown deleterious onꢀtarget effects of
systemic TGR5 agonists. To avoid these systemic effects while stimulating Glucagon Like
Peptideꢀ1 (GLPꢀ1) secreting enteroendocrine Lꢀcells, we have designed TGR5 agonists with low
intestinal permeability. In this paper, we describe their synthesis, characterization, and biological
evaluation. Amongst them, compound 24 is a potent GLPꢀ1 secretagogue, has low effect on
gallbladder volume, and improves glucose homeostasis in a preclinical murine model of dietꢀ
induced obesity and insulinꢀresistance, making the proof of concept of the potential of topical
intestinal TGR5 agonists as therapeutic agents in typeꢀ2 diabetes.
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.
Bile acids have been known for over a decade to act not only as lipid solubilizing agents during
1
digestion, but also as signaling molecules, through two main receptors : the nuclear receptor
2
3,4
FXR (identified in 1999), and the GꢀProteinꢀCoupled Bile Acid Receptor TGR5 (aka GPꢀ
BAR1, GPR131, MꢀBAR identified in 2002). TGR5 acts as a sensor of the prandial state of the
organism, by detecting bile acids in various tissues, organs, and cells, such as enteroendocrine Lꢀ
cells, brown adipose tissue, skeletal striated muscles, but also gallbladder, immune cells and
neurons. Based on the existing functional and pharmacological data in preclinical models, TGR5
5
appears as an attractive target to treat metabolic diseases through different pathways . First,
TGR5 expressed in enteroendocrine Lꢀcells promotes the transcription of the proglucagon gene
expression and the secretion of the incretin Glucagonꢀlike peptide 1 (GLPꢀ1), which displays
ACS Paragon Plus Environment

Page 3 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
numerous beneficial effects on glucose and energy homeostasis (insulin secretion, beta cell
6
preservation, appetite regulation, gastric emptying) . Second, TGR5 activation in brown adipose
tissue (BAT) and muscle increases deiodinase 2 gene expression and consequently conversion of
the inactive T4 hormone into active thyroxin T3, and uncoupling proteinꢀ1 (UCPꢀ1) activation
leading to energy expenditure. Interestingly, the link between TGR5 activation in BAT and
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
7
8
energy expenditure firstly demonstrated in mouse has been recently shown also in humans .
Nevertheless, recent investigations in preclinical animal models have suggested that systemic
9,10
11
TGR5 agonists may trigger unwanted effects such as gallbladder swelling , itching , or more
1
2ꢀ14
recently cardiovascular issues
. In this context, TGR5 localization on enteroendocrine Lꢀcells
in the intestine epithelium offers a new opportunity. Indeed, the development of intestinalꢀ
targeted agonists may lead to GLPꢀ1 secretagogue compounds with a systemic bioavailability
low enough to get rid of onꢀtarget as well as offꢀtarget systemic side effects.
We aim to obtain compounds with low gastroꢀintestinal (GI) absorption that can exert their
1
5
biological activity locally in GI tract while minimizing systemic exposure. This strategy
seemed particularly suitable for the development of TGR5 agonists because it would enable
TGR5 agonists to stimulate the GLPꢀ1 release by enteroendocrine Lꢀcells without triggering any
16,17
other TGR5ꢀrelated effect
. To access such compounds, we decided to design our compounds
as chimeric molecules associating a “pharmacophore” that would bear the pharmacological
activity, and a “kinetophore”, that would control the kinetic properties of the whole molecule,
without modifying significantly its effect on the receptor. The kinetophore concept was
introduced in 2006 to describe a highly polar and/or large and hindered chemical moiety that is
tethered to a pharmacologically active structure, to drastically modify its pharmacokinetic
18
properties, especially by decreasing its oral absorption .
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 103
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
Our efforts firstly focused on optimizing the pharmacophore part and developing structureꢀ
activity relationships to determine where to link the kinetophore on the molecules. Analysis of an
inꢀhouse HTS hit family showed a similarity to a TGR5 agonist (CpdA) we had synthesized to
19
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
use as a reference, described in a patent by Exelixis . To increase the diversity of this triazole
family, a classical CꢀtoꢀN isosteric replacement strategy to the CpdA led to original compound 1
(scheme 1). Surprisingly, although very structurally close to CpdA compound 1 was completely
inactive on TGR5 showing SAR in our aminothiotriazole series contrasting with that of Exelixis.
Substitution of the extracyclic nitrogen by a methyl gave compound 2 that showed promising
activity on both human and murine TGR5 receptors. Switch of the central scaffold from a
triazole to an imidazole led to compound 3 with a drastic increase of potency (from 180 nM to
0
.8 nM on the murine receptor). Compound 3 was the starting point for the design of agonists
bearing several kinetophore moieties. In this paper, we describe SAR around 3, followed by the
design, synthesis and characterization of kinetophore bearing analogs in pharmacokinetic and
pharmacodynamics models.
ACS Paragon Plus Environment

Page 5 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1. Isosteric replacement strategy and scaffold switching.
N
N
S
N
N
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
O
CF3
O
O
F
F
In-house HTS hit
mEC50 = 1680 nM
hEC50 = 1310 nM
N
N
N N
O
O
N
S
N
S
N
N
Isosteric
replacement
H
NꢀMethylation
F
F
O
F
N
N
O
S
F
F
N
1
mEC50 > 10 000 nM
hEC50 > 10 000 nM
2
mEC50 = 180 nM
F
hEC50 = 4300 nM
F
Scaffold switching
TGR5 agonist CpdA
mEC50 = 21 nM
hEC50 = 470 nM
O
F
N
O
N
S
N
F
F
mEC50 = 0,8 nM
hEC50 = 35 nM
3
CHEMISTRY
3-amino-5-thio-[1,2,4]triazole series. Synthesis of the 3ꢀaminoꢀ5ꢀthioꢀ[1,2,4]triazole derivatives
was performed in four to five steps with moderate yield (scheme 2). The first step was the
2
0
synthesis of the core aminothiotriazole, by a oneꢀpot twoꢀsteps reaction . Selective substitution
of the sulfur atom was then carried out in classic nucleophilic substitution conditions to afford
the 5ꢀbenzylsulfanylꢀ4ꢀphenylꢀ4Hꢀ[1,2,4]triazolꢀ3ꢀylamine derivative. Access to compound 1
was enabled via a ChanꢀLam coupling, requiring protection and deprotection steps of the
exocyclic nitrogen by an acetyl group. Compound 2 was obtained through the same coupling
reaction after methylation of the nitrogen by reductive amination.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 103
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
Scheme 2. Synthetic route to 1ꢀ2.
F
H
N
N
N
N N
HN
a
b
NH2 .HCl
H N
SH
2
N
H N
S
2
N
NH2
F
1a
1b
F
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
F
c or e
O
F
N
N
F
N
N
O
d
R'
N
S
f
N
N
S
N
R
H
F
F
1
d R= COCH3
1
c R= COCH3
1
2
R = H
R= CH3
F
2c R= CH3
F
Reagents and conditions: (a) i ) 4ꢀfluorophenylisothiocyanate, DIEA, DMF, 50°C, 15h ii) NaOH
M, 50°C, 60h , 87% (b) 2ꢀbromomethylꢀ1,3ꢀdifluoroꢀbenzene, DIEA, CH Cl , room temp., 3h,
2
2
2
98% (c) i) MeONa, paraformaldehyde, MeOH, room temp., 16h ii) NaBH , MeOH, reflux, 30
4
min, 76% (d) ChanꢀLam coupling, 3,4ꢀdimethoxyphenylboronic acid, pyridine, Cu(OAc) , MS
2
4
Å, CH Cl , room temp., 24ꢀ48h, 11ꢀ50% (e) Ac O, CH Cl , room temp., 30 min, 66% (f) AcCl,
2 2 2 2 2
EtOH, reflux, 30 min, 57%.
5-amino-2-thio-imidazole series. 5ꢀaminoꢀ2ꢀthioꢀimidazoles derivatives were obtained through a
steps synthetic route, using simple building blocks such as aniline, aminoꢀacids or benzyl
6
derivatives to speed up the exemplification in this family. The first step was a monoꢀmethylation
of the desired aniline by reductive amination. The methylaniline or Nꢀmethylbenzylamine was
firstly acetylated by chloroacetyl chloride. Then a nucleophilic substitution of the chloride by
ammonia afforded the primary amine. This synthetic route was later optimized into a more
convenient and efficient route: 1ꢀpropylphosphonic acid cyclic anhydride acylation with a Boc
Nꢀprotected glycine followed by an acidic deprotection of the amino group. Use of either a
commercial isothiocyanate or an aniline and TCDI led to the thiourea intermediate. The sulfur
atom could then be selectively alkylated by a benzyl derivative to give the corresponding
isothiourea. Finally, we took advantage of the 1ꢀpropylphosphonic acid cyclic anhydride ꢀ
ACS Paragon Plus Environment

Page 7 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
21
mediated cyclodehydration method that we have developed previously to obtain the desired 5ꢀ
aminoꢀ2ꢀthioimidazole compounds from the isothiourea intermediates. (Schemes 3ꢀ5)
Scheme 3. Synthetic route to 3ꢀ6
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
O
O
O
O
O
O
O
a
b
O
c
O
O
+
O
NH
Cl
NH3
NH2
N
N
N
3b
3
a
3c
d
O
O
O
O
O
O
f
H
N
e
H
H
N
O
R
N
N
N
N
N
S
N
S
S
R
F
F
3ꢀ6
3eꢀ6e
3d
F
Reagents and conditions: (a) i) paraformaldehyde, MeONa, room temp., 16h ii) NaBH , MeOH,
4
reflux, 1ꢀ3h, 77% (b) Chloroacetyl chloride, DIEA, CH Cl , 0°C, 30 min (c) aq. NH , EtOH,
2
2
3
6
5°C, 1h, 75% (over two steps) (d) 4ꢀfluorophenylisothiocyanate, NEt , EtOH, room temp., 15
3
min, 76 % (e) R ꢀCl, NaI, K CO , CH CN, room temperature, 16h, 90ꢀ98% (f) 1ꢀ
2
3
3
propylphosphonic acid cyclic anhydride, DIEA, EtOAc, 150°C (µW), 10ꢀ40 min, or reflux
classical heating), 24h, 32ꢀ61%.
(
Scheme 4. Synthetic route to 7-12
O
O
O
b'
c'
H
N
O
N
O
10b
O
O
O
O
O
O
a
b
O
c
O
O
+
O
NH
Cl
NH3
X
O
NH2
N
N
N
3
b
-
-
3a
3c X =HCOO
d or d'
-
-
1
0c X =CF COO
3
O
O
O
O
O
F
f
O
e
H
N
H
H
N
N
N
N
O
N
R
R
N
S
N
S
S
R
F
7eꢀ12e
F
F
7dꢀ12d
7ꢀ12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 103
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
Reagents and conditions: (a) i) paraformaldehyde, MeONa, room temp., 16h ii) NaBH , MeOH,
4
reflux, 1ꢀ3h, 77% (b) Chloroacetyl chloride, DIEA, CH Cl , 0°C, 30 min (c) aq. NH , EtOH,
2
2
3
65°C, 1h, 75% (over two steps) (b’) BocꢀGlyꢀOH, 1ꢀpropylphosphonic acid cyclic anhydride,
DIEA, EtOAc, room temp., 30 minꢀ48h, 100% (c’) TFA (30 %) /DCM, room temp., 30 min,
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
1
00% (d) RꢀN=C=S, NEt , EtOH, room temp. or 60°C, 15 min to overnight, 30ꢀ53% (d’) aniline,
3
TCDI, NEt , dioxane, room tempꢀ60°C, overnight, 49% (e) 2ꢀbromomethylꢀ1,3ꢀdifluoroꢀ
3
benzene, NaI, K CO , CH CN, room temperature, 16h, 82ꢀ99% (f) 1ꢀpropylphosphonic acid
2
3
3
cyclic anhydride, DIEA, EtOAC, 150°C (µW), 10ꢀ40 min, or reflux (classical heating), 24h, 11ꢀ
3 %.
6
Scheme 5. Synthetic route to 13-18
O
H
b'
c'
R'
N
O
N
R
O
14bꢀ16b, 18b
O
a or a'
O
b
c
R'
R'
R'
NH2
NH
R
R'
Cl
N
NH2
N
R
13cꢀ18c
R
13b, 17b
1
3a,15a,16a,18a
d
F
O
N
H
N
O
R'
H
H
N
f
R'
N
S
e
N
S
N
R'
N
N
N
R
R
R
F
F
S
F
F
F
1
3ꢀ18
13eꢀ18e
13dꢀ18d
F
Reagents and conditions: (a) i) paraformaldehyde, MeONa, room temp., 16h ii) NaBH , MeOH,
4
reflux, 1ꢀ3h, 51ꢀ96% (a’) Allyl bromide, K CO , DMF, 80°C, 16h (b) Chloroacetyl chloride,
2
3
DIEA, CH Cl , 0°C, 30 min (c) aq. NH , EtOH, 65°C, 1h, 75% (b’) BocꢀGlyꢀOH, 1ꢀ
2
2
3
propylphosphonic acid cyclic anhydride, DIEA, EtOAc, room temp., 30 minꢀ48h, 79ꢀ100% (c’)
TFA (30%)/CH Cl , room temp., 30 min (d) 4ꢀfluorophenylisothiocyanate, NEt , EtOH, room
2
2
3
temp., 15ꢀ60 min, 90ꢀ100% (e) 2ꢀbromomethylꢀ1,3ꢀdifluoroꢀbenzene, NaI, K CO , CH CN,
2
3
3
ACS Paragon Plus Environment

Page 9 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
room temperature, overnight, 58ꢀ100% (f) 1ꢀpropylphosphonic acid cyclic anhydride, DIEA,
EtOAc, 150°C (µW), 10ꢀ40 min, or reflux (classical heating), 24h, 10ꢀ59%.
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
Compound 19 was prepared from 4ꢀfluorophenylisothiocyante and 2ꢀaminoacetonitrile. After
addition on the isothiocyanate, alkylation using 2ꢀbromomethylꢀ1,3ꢀdifluoroꢀbenzene on the
sulfur atom was followed by spontaneous cyclization into the 5ꢀaminoꢀ2ꢀthioimidazole 19b. The
amino group was then acylated using 3,4ꢀdimethoxybenzoic acyl chloride, prepared from the
corresponding carboxylic acid to give compound 19. (scheme 6)
Scheme 6. Synthetic route to 19
N
S
F
F
N
O
N
F
N
N
HN
F
SH
a
b
H N
S
c
N
S
2
N
N
O
H
F
F
O
F
F
19a
19b
19
Reagents and conditions: (a) 2ꢀaminoacetonitrile hydrochloride, NEt , DMF, room temp. to 0°C,
3
3
0 min, 80% (b) 2ꢀbromomethylꢀ1,3ꢀdifluoroꢀbenzene, DIEA, CH CN, room temp., 5 min (c)
3
3,4ꢀdimethoxybenzoic acid, thionyl chloride, CH Cl , THF, Pyridine, room temp., 4h, 31%.
2
2
To prepare compound 22, 4ꢀbromoꢀ2,6ꢀdifluorobenzyl alcohol was reacted with mesyl chloride
in presence of trimethylamine. (NMR analysis of the product revealed that the product was not
the expected mesylate, but rather the benzyl chloride). This benzyl derivative was then used to
afford the corresponding imidazole 20b in two steps (scheme 7). A Sonogashira coupling using
N,Nꢀdimethylpropargylamine was then performed on this imidazole to give 20. The tertiary
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 103
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
amine was finally methylated to give the corresponding quaternary ammonium, which was first
obtained as an iodide (21), and then as a formate (22) after purification by preparative HPLC in
alkaline buffer (ammonium formate buffer pH = 9.2).
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
Scheme 7. Synthetic route to 20-22
O
O
O
O
O
O
H
N
O
a
b
N
N
H
H
O
N
F
N
N
N
S
N
N
F
S
F
S
F
F
3
d
20a
F
Br
F
c
20b
Br
O
O
N
O
F
N
N
S
N
O
F
d
N
S
N
F
X
+
N
F
F
-
-
20
2
1 X = I
N
-
-
F
22 X = HCOO
Reagents and conditions: (a) RꢀCl, NaI, K CO , CH CN, 16 h, room temp. 95% (b) 1ꢀ
2
3
3
propylphosphonic acid cyclic anhydride, DIEA, EtOAc, reflux, 24h, 74% (c) N,Nꢀ
Dimethylpropargylamine, pyrrolidine, CuI, PdCl (dppf) , DMF, 80°C, 16h, 47% (d) MeI,
2
2
THF/Et O, room temp., 2ꢀ6h, 17ꢀ19%.
2
The compounds bearing a quaternary ammonium 23 or a sulfonate 24 were obtained through a
oneꢀstep synthesis from the chloropropyl derivative 23b. Synthesis of the corresponding benzyl
derivative 23ii was carried out starting with the 2,6ꢀdifluoroꢀ4ꢀhydroxybenzonitrile (scheme 8).
Alcaline hydrolysis of the cyano group was followed by esterification with methanol in acidic
conditions. The corresponding methyl ester was reduced into the benzyl alcohol using DIBAlH
to give compound 23i. Selective alkylation of the phenolic hydroxyl by 1ꢀbromoꢀ3ꢀ
ACS Paragon Plus Environment

Page 11 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chloropropane was then performed, followed by activation of the benzylic carbon using mesyl
chloride to give the compound 23ii.
Scheme 8. Synthetic route to intermediate 23ii
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
F
Cl
F
F
HO
N
b
a
O
OH
F
OH
F
F
Cl
23i
23ii
Reagents and conditions : (a) i) NaOH, H O, reflux, 24h, 94% ii) H SO , MeOH, reflux, 18h,
2
2
4
9
0% iii) DIBAlH, THF, 0°C, 1.5 h then NaKꢀtartrate aq (1M), room temp., 0.5 h, 80% (b) i)
Cl(CH ) Br, K CO , CH CN, reflux, 3h, 96% ii) MsCl, NEt , CH Cl , 0°C to room temp., 16h,
2
3
2
3
3
3
2
2
100%.
Intermediate 23ii was used to obtain the corresponding imidazole 23b in two steps (Scheme 9).
The sulfonate derivative 24 was readily synthesized from 23b using sodium sulfite and sodium
2
2
iodide in a dioxane/water mixture under microwave irradiation . Quaternary ammonium 23 was
obtained by reaction of 23b with 1,4ꢀdiazabicylco[2.2.2]octane (DABCO) in acetonitrile under
microwave irradiation.
Scheme 9. Synthetic route to intermediate 23-24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 103
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
O
O
O
O
O
O
O
H
H
H
N
a
b
N
N
N
N
F
N
N
O
F
N
S
S
S
N
F
F
F
O
3
d
F
2
3a
23b
F
Cl
O
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
Cl
c
d
O
O
N
O
N
F
S
O
F
N
N
N
S
N
F
O
O
F
F
O
F
S
+
N
O O
N
2
4
NH4+
23
HCOO-
Reagents and conditions: (a) 23ii, K CO , CH CN, room temp., 16h, 73% (b) 1ꢀ
2
3
3
propylphosphonic acid cyclic anhydride, DIEA, EtOAc, reflux, 28h, 96% (c) NaI, Na SO ,
2
3
dioxane/H O (1/1), µW, 130°C, 100 min, 36% (d) DABCO, CH CN, µW, 100°C, 30 min, 23%.
2
3
Intermediate 25iii was obtained in 3 steps starting from intermediate 23i. The phenolic oxygen
was first alkylated using 1,3ꢀdibromopropane to give intermediate 25i. Then a nucleophilic
substitution using tertꢀbutyl Nꢀtertꢀbutoxycarbonylcarbamate was carried out on the
bromopropyl derivative to afford intermediate 25ii. This benzylic alcohol was then converted
into the benzyl chloride derivative 25iii using mesyl chloride as describe above for intermediate
23ii.
ACS Paragon Plus Environment

Page 13 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 10. Synthetic route to intermediate 25iii
HO
Cl
HO
HO
F
F
F
F
F
F
F
F
a
b
c
O
O
O
O
O
OH
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
Br
5i
O
N
O
N
2
3i
2
25ii
O
O
O
O
25iii
Reagents and conditions: (a) 1,3ꢀdibromopropane, K CO , CH CN, reflux, 3h, 76% (b) tertꢀbutyl
2
3
3
Nꢀtertꢀbutoxycarbonylcarbamate, Cs CO , DMF, 70°C, 30min, 95% (c) NEt , MsCl, CH Cl ,
2
3
3
2
2
0°C to room temp., overnight, 87%.
Intermediate 25iii was used to obtain the corresponding imidazole 25b in two steps. After
deprotection, amine 25c was reacted with sodium 2ꢀchloroethanesulfonate in presence of sodium
iodide and DIEA in DMF under microwave irradiation (scheme 11).
Scheme 11. Synthetic route to 25.
F
O
O
O
O
F
O
O
O
H
N
N
O
O
N
O
a
H
H
N
N
F
b
N
O
N
S
O
O
N
N
N
F
S
S
F
F
3d
25a
25b
O
O
O
F
O
N
O
c
F
F
F
F
O
O
O
O
N
O
SO3-
SO3-
N
O
S
d
S
N
N
N
N
N
NH2
2
5
25c
F
F
Reagents and conditions: (a) 25iii, DIEA, CH CN, room temp., overnight, 35% (b) 1ꢀ
3
propylphosphonic acid cyclic anhydride, DIEA, EtOAc, reflux, 48h, 62% (c) TFA (10%),
CH Cl , room temp., 5h, 98% (d) Sodium 2ꢀchloroethanesulfonate, NaI, DIEA, DMF, µW,
2
2
100°C, 2h, 16%.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 103
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
Compounds 26-30 were prepared in a similar way replacing glycine by alanine, or correctly
protected lysine and glutamic acid (scheme 12).
Scheme 12. Synthetic route to 26-30.
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
O
O
O
O
CF COO-
O
O
3
O
O
O
O
b
c
O
H
a
+
H
N
O
NH3
N
N
N
N
NH
N
R
O
R
R
SH
F
3
a
2
6aꢀ28a
26bꢀ28b
26cꢀ28c
d
O
F
R'
O
F
O
O
R
N
O
F
f or g
e
N
H
O
S
O
N
N
N
N
N
N
S
N
F
R
S
F
F
F
F
F
2
7 R=(CH ) -NH
2 4 2
26 R=CH3
26d R=CH3
27d R=(CH ) -NHFmoc
h
29 R=(CH ) -COOH
2 2
2
7e R=(CH ) -NHFmoc
2 4
+
2 4
3
0 R=(CH ) -N (CH )
2 4 3 3
28
R=(CH ) -COOCH
28d R=(CH ) -COOCH
2 2 3
2
2
3
Reagents and conditions: (a) BocHNꢀCH(R)ꢀCOOH, 1ꢀpropylphosphonic acid cyclic anhydride,
DIEA, EtOAc, room temp., 30minꢀ1h, 66ꢀ97% (b) TFA (30%)/ CH Cl , 30 min, 100% (c) 4ꢀ
2
2
fluorophenylisothiocyanate, TEA, EtOH, 5 min, 70ꢀ79% (d) 2ꢀbromoethylꢀ1,3ꢀdifluoroꢀbenzene,
K CO , Acetonitrile, room temp., overnight, 84ꢀ89% (e) 1ꢀpropylphosphonic acid cyclic
2
3
anhydride, DIEA, EtOAc, reflux (classical heating), 48h, 6ꢀ32% (f) NaOH 1N, MeOH, room
temp., overnight, 93% (g) piperidine, EtOAc, room temp., 60% (h) MeI, DIEA, Et O/THF (1/1),
2
room temp., overnight, 20%.
RESULTS AND DISCUSSION
All compounds were evaluated in vitro for their TGR5 agonist activity on both human and
murine receptors in a CREꢀdriven luciferase reporter assays in TGR5ꢀtransfected HEK293 cells.
Full agonist activity (efficacy > 80% measured relative to lithocholic acid 10 µM) was obtained
for all active compounds.
ACS Paragon Plus Environment

Page 15 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Pharmacophore optimization
TGR5 agonist activities on both human and murine receptors of compounds 1 and 2 are reported
in Table 1. Surprisingly, compound 1 was shown to be completely inactive on both receptors. In
contrast, compound 2 showed promising activity on the murine receptor.
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
Table 1. In vitro activity of compounds 1 and 2
O
N
N
O
F
N
S
N
F
R
F
hTGR5
EC (nM)
mTGR5
Example
R
H
EC (nM)
50
50
a
a
1
2
NA
NA
CH3
4300
[3800ꢀ4900]
180
[150ꢀ220]
EC values are reported with their 95% confidence intervals in brackets.
50
a
NA: not active (Emax < 10% at all concentrations up to 30µM).
Replacing the triazole by an imidazole in 3 gave an interesting 2 log improvement of potency on
both human and murine receptors (Table 2). We then decided to draw Structure Activity
Relationships of this series by exploring the role of the substituants on the aromatics rings. SAR
on the benzylic ring on the rightꢀhand side of the molecules are described in the Table 2. The
2
,6ꢀdifluoro substitution of this phenyl ring appeared to be important for activity, especially on
the murine receptor. Indeed, the hEC /mEC ratio dropped from 44 for compound 3 to 2 for
5
0
50
compound 5.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 103
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
Table 2. In vitro activity of compounds 2ꢀ6
O
X
N
O
R
N
S
N
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
F
hTGR5
EC (nM)
mTGR5
Example
2
X
R
EC (nM)
5
0
50
F
N
4300 [3800ꢀ4900]
35 [22ꢀ54]
180 0 [150ꢀ220]
0.8 [0.5ꢀ1.4]
F
F
3
CH
F
F
4
5
CH
CH
128 [81ꢀ200]
156 [105ꢀ231]
12.0 [5ꢀ32]
68.0 [54ꢀ85]
6
CH
299 [267ꢀ335]
41.0 [27ꢀ63]
EC values are reported with their 95% confidence intervals in brackets.
50
Table 3 describes the SAR on the phenyl ring on the lower part of the molecule. As shown by the
drastic decrease in potency for compound 8 and 9, more hindered group than fluorine are not
tolerated at the para position, whether they are electrodonor or electroattractor. On the opposite,
adding ortho and meta substitution on the paraꢀfluorophenyl moiety showed interesting potency,
with even a slight improvement with the compound 12 compared to compound 3.
ACS Paragon Plus Environment

Page 17 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. In vitro activity of compounds 3, 7ꢀ12
O
F
N
O
N
S
N
F
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
hTGR5
EC (nM)
mTGR5
Example
3
EC (nM)
50
50
35 [22ꢀ54]
0.8 [0.5ꢀ1.4]
5.1 [4.2ꢀ6.1]
F
7
8
254 [207ꢀ312]
7745 [7135ꢀ8407]
939.0 [731ꢀ1207]
O
9
900 [804ꢀ1008]
35 [27ꢀ45]
366.0 [321ꢀ417]
2.0 [1.6ꢀ2.5]
CF3
1
0
Cl
F
F
F
11
47 [34ꢀ65]
20 [13ꢀ30]
4.8 [3.7ꢀ6.3]
0.8 [0.2ꢀ3.4]
1
2
O
F
EC values are reported with their 95% confidence intervals in brackets.
50
Exploration of the SAR on the phenyl ring of the leftꢀhand side of the molecule is described in
Table 4. The 4,5ꢀdimethoxy substitution appeared crucial for the compounds activity on TGR5,
as removal of one or both methoxy groups (compounds 14 and 15) lead to a drastic drop of the
potency (more than 100 fold on the murine receptor). 4ꢀchloroꢀ3ꢀmethoxy and 3,4ꢀdichloro
substitution lead to a decrease in potency, especially for the dichloro substitution on compound
16. Homologation of the 3,4ꢀdimethoxyphenyl to 3,4ꢀdimethoxybenzyl group lead to a less
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 103
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
active compound (17). Whereas modification of the phenyl substitution was generally poorly
tolerated, modification of the alkyl substitution of the extracyclic nitrogen showed only a
moderate decrease of potency (compound 18). It is worth noting that we have tried several
strategies to obtain the Nꢀdesmethyl analog of compound 3 but were unable to isolate it, because
of its high instability.
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
Table 4. In vitro activity of compounds 3, 13ꢀ19
F
N
N
S
N
F
F
R
hTGR5
mTGR5
Example
3
R
EC (nM)
EC (nM)
50
50
O
CH3
35 [22ꢀ54]
885 [840ꢀ932]
1106 [986ꢀ1240]
0.8 [0.5ꢀ1.4]
440 [37ꢀ52]
O
Cl
O
1
3
4
CH3
CH3
O
1
141.0 [120ꢀ165]
1
5
6
O
CH3
CH3
994 [915ꢀ1079]
750 [706ꢀ799]
91.0 [73ꢀ113]
Cl
1
507.0 [456ꢀ564]
Cl
O
1
7
CH3
326 [290ꢀ365]
86 [44ꢀ169]
40. 0 [24ꢀ66]
7.7. [6.6ꢀ9]
O
O
ꢀ
CH2ꢀ
CH=CH
18
O
2
O
O
a
1
9
H
NA
≥ 6000
O
EC values are reported with their 95% confidence intervals in brackets
50
a
NA: not active (Emax < 10% at all concentrations up to 10µM).
ACS Paragon Plus Environment

Page 19 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Kinetophore positioning and optimization.
Our strategy to obtain intestineꢀtargeted TGR5 agonists was to introduce on our potent agonists a
kinetophore group that would limit the intestinal absorption without effect on the TGR5 agonist
activity of such compounds. The keystone of this strategy is the identification of a “mute”
position where linking such groups would not impact the target interaction. Analysis of the SAR
around compound 3 has demonstrated that the dimethoxy substitution on the leftꢀhand side of the
molecule was crucial and that the bottom of the molecule does not tolerate paraꢀsubstitution by
bigger groups than fluorine. We then decided to explore para substitution of the benzyl on the
rightꢀhand side of the molecule to add our kinetophore groups. The kinetophores were selected
among structural groups making the compound large and/or polar.
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
In this paper, we will focus on ionic kinetophores, especially permanently ionized quaternary
ammonium and sulfonate groups, which cannot be protonated in vivo (pKa <ꢀ1).
Table 5 describes the TGR5 agonist activity of compound 3 and its analogs modified by
substitution on the para position of the benzyl ring. First, we show here that this position can
efficiently tolerate polar and hindered group, as all the compounds maintain a potency and full
efficacy comparable to compound 3. Both positive and negative charges are well tolerated with
better activities for the anionic compounds (24, 25).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 103
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
Table 5. In vitro activity of compounds 3, 20ꢀ25
O
N
O
R
N
S
N
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
F
hTGR5
mTGR5
Example
R
EC (nM)
EC (nM)
5
0
50
F
3
35 [22ꢀ54]
71. [62ꢀ81]
0.8 [0.5ꢀ1.4]
4.1. [1.7ꢀ9.8]
7.5. [6.2ꢀ9.1]
17.0 [11ꢀ27]
F
F
20
21
22
N
F
F
I
+
N
177 [146ꢀ214]
F
F
HCOO-
2
60 [234ꢀ288]
+
N
F
F
HCOO-
O
23
238 [210ꢀ268]
24 [16ꢀ37]
9.8 [8.6ꢀ11.2]
0.4 [0.3ꢀ0.6]
+
N
N
F
F
O
2
4
5
O
S
O O
+
NH4
F
F
F
O
S
O
O
O
+
NH4
2
N
77 [41ꢀ144]
0.3 [0.3ꢀ0.4]
O
+
NH4
S
O
O
EC values are reported with their 95% confidence intervals in brackets.
50
Finally, we explored the impact of a charged kinetophore moiety on the position 4 of the
imidazole ring on potency. As can be seen in Table 6, a drastic loss of potency was observed for
all compounds substituted on the position 4 of the imidazole (26ꢀ30).
ACS Paragon Plus Environment

Page 21 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 6: In vitro activity of compounds 3, 26-30
O
R
F
N
O
N
S
N
F
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
F
hTGR5
mTGR5
Example
3
EC (nM)
EC (nM)
5
0
50
R
ꢀH
ꢀCH3
35 [22ꢀ54]
337 [304ꢀ373]
> 4500
1113 [992ꢀ1249]
0.8 [0.5ꢀ1.4]
7 0 [6ꢀ8]
119 0 [98ꢀ145]
82 0 [64ꢀ104]
1935.0 [1483ꢀ2526]
254.0 [195ꢀ331]
26
27
28
ꢀ(CH ) ꢀNH
2
2
4
ꢀ(CH ) ꢀCOOMe
2 2
a
29
ꢀ(CH ) ꢀCOOH
NC
≥ 10000
2
2
+
3
0
ꢀ(CH ) ꢀN (CH )
2
4
3 3
EC values are reported with their 95% confidence intervals in brackets.
50
a
NC: not calculated (Emax < 20% at all concentrations up to 10µM).
The pharmocophore compound 3 and kinetophoreꢀlinked compounds 22, 23 and 24 showing
interesting in vitro TGR5 activity, were then evaluated for physicoꢀchemical and in vitro ADME
properties.
As can be seen in Table 7, the kinetophore part of compounds 22ꢀ24 profoundly impacts
physicoꢀchemical properties: aqueous solubility of compounds 22ꢀ24 compared to 3 is largely
improved and their lipophilicity decreased.
As polar surface area (PSA) has been described as a predictive parameter of the passive
molecular transport through membranes and compound absorption in the GI tract, PSA as well as
23ꢀ25
clogD and AlogP were calculated for compounds 22ꢀ24 using Pipeline Pilot™ (Accelrys)
.
In our case, the calculated (clogD ) was quite different from measured logD for all the
7
.4
7.4
evaluated compounds (neutral, cationic and anionic). As can be seen in table 7, PSA is not
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 103
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
impacted by the introduction of our cationic kinetophores (3 versus 22ꢀ23), because the charge is
buried in the tetraꢀsubstituted ammonium, and neither a lone pair nor a hydrogen atom remains
for hydrogen bonding. In contrast, introduction of the sulfonate group, a fully exposed charge
2
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
leads to a much higher PSA (140 Å described as a threshold value to be usually poorly
absorbed).
Cellular permeability was performed in Cacoꢀ2 monolayer permeability assay. In this assay, both
cationic (22, 23) and anionic (24) compounds exhibited a very low permeability, much lower
ꢀ6
ꢀ1
ꢀ6
than that of our pharmacophore compound 3 (Papp AꢀB≤ 0.08 10 cm.s and P AꢀB = 5.32 10
app
ꢀ1
cm.s respectively). In this series, cellular permeability is not correlated with PSA. Indeed,
compound 3 and cationic compounds 22 and 23 display the same PSA, but are highly different in
term of permeability.
Incubation with mouse liver microsomes in vitro revealed that our compounds are metabolically
instable (t1/2 = 1.2 min for compound 24). This result suggests a high first pass effect for these
compounds that will further lower systemic exposure in addition to a low intestinal permeability.
ACS Paragon Plus Environment

Page 23 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 7. In vitro ADME evaluation of kinetophoreꢀcoupled compounds
Compound
kinetophore
3
22
23
24
O
S
O O
+
+
N
N
N
ø
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
1
Calculated parameters
cLogD
6.63
65
5.93
65
5.36
73
4.15
140
2
PSA (Å )
Measured parameters
Aqueous Solubility (µM)
LogD (7.4)
6.3
>200
1.0
>200
0.9
>200
0.8
3.5
a
b
c
Papp AꢀB
5.32 ± 0.94
0.08 ± 0.02
0.016±0.005
0.031±0.011
80.1 ± 4.2 %
recovery (%)
84.4 ± 6.9 %
69.8 ± 8.7% 69.3 ± 1.2 %
a
b
c
Papp BꢀA
5.47 ± 1.29
0.24 ± 0.03
0.22±0.06
19±1.91
recovery (%)
106.4 ± 19.7 % 93.4 ± 0.7 % 99.6 ± 0.4 % 101.8 ± 3.5 %
c
Efflux ratio
1
3.1
14
636
1.2
Microsomal stability
t (min)
0
.7
10.6
N.D.
1
/2
Microsomal stability
1
287
264
N.D.
1211
Cl (µL/min/mg proteins)
int
1
a
TM
calculated parameters using Pipeline Pilot (Accelrys)
Cell membrane permeability assessed on a Cacoꢀ2 cell monolayer. “AꢀB” indicates the transport from
apical side to basolateral side “BꢀA” indicates the transport from basolateral side to apical side.
ꢀ6
ꢀ1
Compounds were tested at 10 µM. Permeability is expressed in 10 cm.s Permeability classification:
−
6
ꢀ1
−6
ꢀ1
low : P < 2 × 10 cm.s ; high : P > 20 × 10 cm.s .
app
app
b
c
Compound tested at 5 µM
Compound tested as an iodide (ie compound 21)
Compounds 22 and 24, bearing interesting in vitro activity and very low permeability were
chosen for further in vivo pharmacokinetic and pharmacodynamic studies in C57Bl6 mice.
The time required by the drug to reach ileum and colon containing the highest density of target
6
Lꢀcells after oral dosing was unknown. Therefore, compounds 22 and 24 were administered
orally at three consecutive timeꢀpoints (Tꢀ6h, Tꢀ4h, Tꢀ2h; 3 x 20 mg/kg) to ensure the complete
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 103
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
exposure of the GI tract. In order to test the ability of compound 22 to promote GLPꢀ1 secretion,
mice were treated with vehicle or compound 22 as described, prior to an oral glucose (2 g/kg) or
water load. Blood samples were collected 15 minutes later to measure active GLPꢀ1 plasma
concentrations. In addition, sitagliptin (25 mg/kg) was given orally 1 hour prior to blood sample
collection to prevent GLPꢀ1 degradation. As expected, glucose increased GLPꢀ1 concentrations
in vehicle treated mice (Figure 1, Veh+Water vs Veh+Glc). As expected, compound 22 further
enhanced the glucoseꢀinduced GLPꢀ1 secretion compared to vehicle (Figure 1, Veh+Glc vs
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
2
2+Glc). Interestingly, compound 22 induced a 4 fold increase in active GLPꢀ1 concentrations
even in absence of glucose challenge compared to vehicle treated mice (Figure 1, Veh+Water vs
22+Water). Thus, compound 22 by itself appears to be a potent GLPꢀ1 secretagogue in vivo even
in absence of glucose challenge.
In the same experimental setting, 24 lead to a 2.6 fold increase in GLPꢀ1 concentrations
compared to the vehicle treated group in absence of glucose challenge (Figure 2). To evaluate the
impact of the kinetophore in this approach, the effect of naked pharmacophore 3 was measured
in vivo at 15.5 mg/kg, i.e. the same molar quantity as 20 mg/kg of compound 24. No difference
with vehicle could be detected (data not shown). This contrasts interestingly with what has been
observed on kinetophore bearing compounds, and illustrates the necessity of controlling
absorption of our TGR5 agonists, to reach the distal section of the gut, where most of Lꢀcells are
located.
ACS Paragon Plus Environment

Page 25 of 103
Journal of Medicinal Chemistry
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
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
Figure 1. In vivo GLPꢀ1 secretion in C57Bl6 mice. Vehicle (white bars) or compound 22 (grey
bars) were administered to mice at 3 consecutive timeꢀpoints (Tꢀ6h, Tꢀ4h, Tꢀ2h, 3 x 20 mg/kg)
before challenge with a bolus of water or glucose (T0). All groups received sitagliptin (25
mg/kg) 45 min prior to blood sample collection (T+15 min). Mean ± SEM (n = 6/group). *
p<0.05, ** p<0.01 (ANOVA and HolmꢀSidak test). Veh = vehicle (Tween 0.1%). Glc = glucose
(
2g/kg). 22 = Compound 22 in Tween 0.1% (3 x 20 mg/kg).
Figure 2. In vivo GLPꢀ1 secretion in C57Bl6 mice. Vehicle (white bars) or compound 24 (grey
bars) were administered to mice at 3 timeꢀpoints (Tꢀ6h, Tꢀ4h, Tꢀ2h, 3 x 20 mg/kg) before blood
collection (T0) to measure active GLPꢀ1 plasma concentrations. All groups received sitagliptin
(
25 mg/kg) 45min prior to blood sampling. Mean ± SEM (n = 8/group). *** p<0.001 (Student’s t
test)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 103
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
Then, plasma AUC, as well as liver and gallbladder exposure were measured in mice dosed
orally with compounds 22 and 24 at 20 mg/kg. As expected from the low in vitro permeability of
compounds 22 and 24, fecal recovery was quantitative (Table 8). Micromolar concentrations
were reached after 4 hours in the gallbladder, while 100ꢀfold lower Cmax was measured in
plasma. Of note, the relatively high concentration in gallbladder results from the biliary excretion
of only a small fraction (<0.1% for 24) of administered compound.
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
Table 8. Pharmacokinetic parameters of kinetophoreꢀcoupled compounds 22 and 24 in C57Bl6
mice
Compound
22
24
[
C]plasmamax (nM)
100
58
102
63
[
C]plasmamax (ng/mL)
Tmax plasma (min)
AUC (ng.min/mL)
t1/2 (min)
15
45
2909
97
5774
317
385
Liver Cmax (nM)
nd
Gallbladder (tissue and
content) Cmax (nM)
9028
88
10980
100
a
Fecal recovery (%)
n=3 mice/time; male C57Bl6/J mice. Compounds were administered orally (20 mg/kg,
formulated in Tween 0.1%).
a
Feces collected 24 hours after compound dosing (20 mg/kg). Parent compound was extracted
with organic solvent and analyzed quantitatively using mass spectrometry.
ACS Paragon Plus Environment

Page 27 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Intestinal exposure was then measured for compound 24. Very interestingly, the results of the
exposure of the different sections of GI tract showed that the targeted distal sections (ileum and
colon) were maximally exposed by compound 24 8 hours after oral dosing (Figure 3).
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
1
500
000
500
0
2 hours
4
8
hours
hours
1
2
d
h
uod
4
e
h
nu
8
m
h
2 hj eju
4
n
h
um8h
2h ile
4
u
h
m
8h
l
2h co4 ho n 8h
duodenum
jejunum
ileum
colon
Figure 3. Timeꢀcourse of compound 24 concentrations in duodenum, jejunum, ileum and colon
of C57Bl6 mice, after a single oral administration (20 mg/kg). Mean ± SEM (n= 3 mice/time).
In light of the results obtained concerning the intestinal exposure by compound 24, we next
explored the effect of the time elapsed between oral dosing and the GLPꢀ1 secretion assay. As
maximal GLPꢀ1 secretion is expected when TGR5 is activated in distal sections of the GI tract,
we compared GLPꢀ1 secretion at two timeꢀpoints after oral load by compound 24. Consistent
with the pharmacokinetics observed in the intestine, a 3.6 fold and a 10 fold increase in active
GLPꢀ1 concentrations were observed respectively 4 hours and 7 hours after administration of 24
(Figure 4).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 103
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
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
Figure 4. In vivo GLPꢀ1 secretion in C57Bl6 mice. Vehicle (white bars) or compound 24 (grey
bars) were administrated to mice 4 hours (t +4h) or 7 hours (t +7h) before blood collection for
active GLPꢀ1 plasma concentration measurement. All groups received sitagliptin (25 mg/kg) 45
min prior to blood sampling. Mean ± SEM (n = 6/group). ** p<0.01 (Student’s t test). Vehicle =
Tween 0.1%. 24 = Compound 24 in Tween 0.1% (20 mg/kg).
Then, C57Bl6 mice were orally treated with increasing doses of 24 and plasma active GLPꢀ1
concentrations were measured 8 hours postꢀdosing. Compound 24 induced a doseꢀdependent
GLPꢀ1 secretion (Figure 5A). Gallbladder relaxation, described in rodent treated with both bile
26,27
acid derivatives and nonꢀsteroid small TGR5 agonist molecules
, was assessed by gallbladder
volume measurement in fasted mice 8 hours after a single oral administration of vehicle or 24 at
increasing dose. Interestingly, a high GLPꢀ1 secretion (6ꢀfold increase in active GLPꢀ1 plasma
concentrations) was obtained at a dose (5 mg/kg) that did not modify gallbladder volume,
suggesting that in these conditions gallbladder exposure is not sufficient to trigger the
pharmacological response of the gallbladder (Figure 5B). Gallbladder motility was then studied
by gallbladder volume measurement in fasted and reꢀfed conditions at the highest tested dose (20
mg/kg). In this experiment, we showed that gallbladder volume is not significantly increased by
ACS Paragon Plus Environment

Page 29 of 103
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
4 both in fasted and reꢀfed conditions. We showed moreover that compound 24 does not disrupt
food stimulated gallbladder emptying and motility (ejection fraction of 30 and 25 % respectively
for vehicle and compound 24). (Figure 5C) Thus the ratio between the lowest active dose (5
mg/kg) and the highest dose that does not modify gallbladder motility (20 mg/kg) defines a
therapeutic window of 4.
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