Nidufexor (LMB763), a Novel FXR Modulator for the Treatment of Nonalcoholic Steatohepatitis

Article abstract of DOI:10.1021/acs.jmedchem.9b01621

Farnesoid X receptor (FXR) agonists are emerging as important potential therapeutics for the treatment of nonalcoholic steatohepatitis (NASH) patients, as they exert positive effects on multiple aspects of the disease. FXR agonists reduce lipid accumulation in the liver, hepatocellular inflammation, hepatic injury, and fibrosis. While there are currently no approved therapies for NASH, the bile acid-derived FXR agonist obeticholic acid (OCA; 6-ethyl chenodeoxycholic acid) has shown promise in clinical studies. Previously, we described the discovery of tropifexor (LJN452), the most potent non-bile acid FXR agonist currently in clinical investigation. Here, we report the discovery of a novel chemical series of non-bile acid FXR agonists based on a tricyclic dihydrochromenopyrazole core from which emerged nidufexor (LMB763), a compound with partial FXR agonistic activity in vitro and FXR-dependent gene modulation in vivo. Nidufexor has advanced to Phase 2 human clinical trials in patients with NASH and diabetic nephropathy.

Full text of DOI:10.1021/acs.jmedchem.9b01621

Subscriber access provided by Queen Mary, University of London
Drug Annotation
Nidufexor (LMB763), a Novel FXR Modulator for the
Treatment of Nonalcoholic Steatohepatitis (NASH)
Donatella Chianelli, Paul V Rucker, Jason Roland, David C. Tully, John Nelson, Xiaodong Liu,
Badry Bursulaya, Eloy D Hernandez, Jane Wu, Mahavir Prashad, Thierry Schlama, Yugang Liu,
Alan Chu, James Schmeits, David J. Huang, Robert Hill, Dingjiu Bao, Jocelyn Zoll, Young Kim,
Todd Groessl, Peter McNamara, Bo Liu, Wendy Richmond, Ignatio Sancho-Martinez, Andrew
Phimister, H Martin Seidel, Michael K Badman, Sean B Joseph, Bryan Laffitte, and Valentina Molteni
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01621 • Publication Date (Web): 15 Jan 2020
Downloaded from pubs.acs.org on January 15, 2020
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 58
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 58
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
Nidufexor (LMB763), a Novel FXR Modulator for the
Treatment of Nonalcoholic Steatohepatitis (NASH)
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
†
†
±
§
†
Donatella Chianelli, Paul V. Rucker, Jason Roland, David C. Tully, John Nelson, Xiaodong
†
†
†
†
€
¤
Liu, Badry Bursulaya, Eloy D. Hernandez, Jane Wu, Mahavir Prashad, Thierry Schlama,
Yugang Liu,€ Alan Chu,† James Schmeits, David J. Huang, Robert Hill,† Dingjiu Bao,†
†
†
†
†
†
†
£
†
Jocelyn Zoll, Young Kim, Todd Groessl, Peter McNamara, Bo Liu, Wendy Richmond,
†
∥
⊥
‡
Ignacio Sancho-Martinez, Andrew Phimister, H. Martin Seidel, Michael K. Badman, Sean B.
#
¥
†
Joseph, Bryan Laffitte and Valentina Molteni*
†Genomics Institute of the Novartis Research Foundation (GNF), San Diego, California 92121,
United States
ABSTRACT: Farnesoid X receptor (FXR) agonists are emerging as important potential
therapeutics for the treatment of nonalcoholic steatohepatitis (NASH) patients as they exert
positive effects on multiple aspects of the disease. FXR agonists reduce lipid accumulation in
the liver, hepatocellular inflammation, hepatic injury and fibrosis. While there are currently no
approved therapies for NASH, the bile acid-derived FXR agonist obeticholic acid (OCA; 6-ethyl
1
ACS Paragon Plus Environment

Page 3 of 58
Journal of Medicinal Chemistry
chenodeoxycholic acid) has shown promise in clinical studies. Previously, we described the
discovery of tropifexor (LJN452), the most potent non-bile acid FXR agonist currently in clinical
investigation. Here, we report the discovery of a novel chemical series of non-bile acid FXR
agonists based on a tricyclic dihydrochromenopyrazole core from which emerged nidufexor
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
(LMB763), a compound with partial FXR agonistic activity in vitro and FXR-dependent gene
modulation in vivo. Nidufexor has advanced to Phase 2 human clinical trials in patients with
NASH and diabetic nephropathy.
INTRODUCTION AND BIOLOGICAL RATIONALE
Nonalcoholic steatohepatitis (NASH) is a progressive stage of nonalcoholic fatty liver
disease (NAFLD), a condition that represents the hepatic manifestation of the metabolic
1
syndrome. NASH is characterized by excessive accumulation of fat in the liver (steatosis)
leading to hepatic inflammation (steatohepatitis), hepatocyte injury (ballooning) and the
development of fibrosis, which can eventually lead to cirrhosis and hepatocellular
carcinoma (HCC).2-4 At this time, there are no approved pharmacological therapies and
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 58
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
NASH is projected to become the leading indication for liver transplantation in the near
future.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 farnesoid X receptor (FXR) is a bile acid (BA)-activated nuclear receptor highly
expressed in the liver, gall bladder, intestines and kidney that upon activation, regulates
bile acid production, conjugation, and transport.7-9 In concert with fibroblast growth factor
19 (FGF19; rodent ortholog FGF15), a naturally-occurring hormone released from
enterocytes in response to physiological FXR activation,10 FXR initiates homeostatic
responses to control bile acid levels by inducing genes involved in bile acid efflux,
conjugation, detoxification, and renal excretion.1,11 Furthermore, FXR reduces hepatic bile
acids by downregulating expression of genes involved in hepatic bile acid uptake and
synthesis through the regulation of small heterodimer partner (SHP) levels in the liver.12,13
Besides its role in bile acid metabolism and signaling, FXR also contributes to the
regulation of triglyceride metabolism and de novo lipogenesis. Moreover, FXR agonists
are able to reduce steatosis through inhibition of SREBP-induced lipogenesis and fatty
acid -oxidation,11,14-18 and have been shown to reduce fibrosis and hepatic inflammation
ACS Paragon Plus Environment
3

Page 5 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
in various preclinical models of liver disease.19-23 Therefore, due to its role as a master
metabolic regulator,11,24 FXR has become a promising therapeutic target for NASH.1-5,25
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 from the Phase 2 FLINT trial in 2015 showed that the bile acid-derived FXR
agonist OCA is effective in the treatment of NASH with liver fibrosis.26 An interim analysis
of the recent Phase 3 REGENERATE trial confirmed the results from the FLINT trial.27
However, it also confirmed that higher rates of pruritus were experienced by patients
treated with OCA compared to placebo-treated patients.26,28 These effects could
potentially be due to the activation of cell surface bile acid receptors (GPBAR1) by either
29
OCA or its metabolites. Additionally, in the FLINT trial, OCA adversely effected serum
lipid profiles, including elevated LDL cholesterol and a modest reduction of HDL. These
effects could be problematic in a patient population with an elevated risk for
cardiovascular disease.26,28,30,31
To address these concerns, we initiated efforts to discover safe and effective FXR
agonists for the treatment of NASH and other FXR dependent diseases. We focused on
the development of non-bile acid FXR agonists devoid of agonistic activity at GPBAR1,
the bile acid-activated GPCR. Furthermore, to mitigate the lipid profile effects seen with
ACS Paragon Plus Environment
4

Journal of Medicinal Chemistry
Page 6 of 58
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
OCA, which have the potential to be FXR-mediated,31,32 we explored and optimized both
full and partial agonists from various non-bile acid-derived scaffolds.
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
We previously disclosed the highly potent full agonist tropifexor (LJN452)23,33,34 (Figure
1) and describe here the discovery of a novel series of selective FXR modulators based
on a 1-methyl-1,4-dihydrochromeno[4,3-c]pyrazole scaffold core. The lead compound
from this series (1, LMB763) showed partial activation of FXR target genes in vitro and is
a potent and specific FXR-gene modulator in vivo, reducing steatosis, inflammation and
fibrosis in NASH murine models. Compound 1 is currently under investigation in Phase II
clinical trials for the treatment of patients with NASH and diabetic nephropathy.35
O
O
OH
S
N
OH
F3CO
N
H
N
H
H
O
F
O
HO
OH
H
Obeticholic acid (OCA, INT-747)
Tropifexor (LJN452)
O
N
F
Cl
Cl
Cl
F
N
O
OH
O
N
O
H
O
O
GW4064
Isopropyl turofexorate (WAY-362450)
Figure 1. Selected FXR agonists
ACS Paragon Plus Environment
5

Page 7 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
RESULTS AND DISCUSSION
To identify non-bile acid FXR agonists for the treatment of liver diseases, we first
focused on the optimization of a previously described potent FXR scaffold (GW4064),36-
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
38
33
which led to our first clinical candidate tropifexor (LJN452) (Figure 1). In parallel, to
identify FXR agonists with a novel chemical pharmacophore, we performed two different
HTS campaigns including the entire GNF compound collection (about 3 million
compounds). The two campaigns were designed by configuring the screens to measure
different aspects of FXR biology. We identified two critical steps in FXR signaling: ligand-
induced interaction with a coactivator protein and downstream transcriptional activation
of target genes. Accordingly, the first screening campaign was based on a FXR-HTRF
biochemical assay measuring the ligand-induced interaction between FXR and the
Steroid Receptor Coactivator-1 (SRC1), while the second HTS involved a cell-based FXR
BSEP-luciferase reporter assay designed to measure the transcriptional activity of full
length FXR over its direct target genes (e.g. BSEP). In this assay, FXR agonism leads to
the transcriptional activation of the BSEP promoter, which in turn drives the expression
ACS Paragon Plus Environment
6

Journal of Medicinal Chemistry
Page 8 of 58
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
of a luciferase reporter. Thus, measurements of luciferase signal directly reflect FXR
transcriptional activity.
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 novel substituted 1-methyl-1,4-dihydrochromeno[4,3-c]pyrazole compound 2
(Table 1), was identified as an overlapping hit from both HTS efforts. In the FXR HTRF
biochemical assay, compound 2 was determined to have modest potency with EC50
=
480 nM and 70% efficacy relative to GW4064, the commonly used comparator FXR
agonist (Table 1). This activity correlated well with that measured in the FXR BSEP-luc
reporter gene cellular assay, where it showed EC50 = 690 nM with partial agonistic
activation of 67% relative to GW4064. Despite low potency, this novel compound
presented a very attractive opportunity for optimization due to its relative low molecular
weight (MW = 404) and high aqueous water solubility. As in vitro metabolic stability in rat
and human liver microsomes (RLM, HLM) was poor, medicinal chemistry optimization
was initiated with the objective of identifying compounds with improved potency and
ADME/PK properties while maintaining a partial FXR agonistic profile.
Compound 2 showed some structural similarities to an early FXR clinical candidate,
39
WAY-362450 (FXR450) which also contained a tricyclic core (Figure 1). By examining
ACS Paragon Plus Environment
7

Page 9 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the X-ray structure of the FXR ligand binding domain of WAY-362450 (FLI.PDB), we built
a computational molecular model of 2 (Figure 2) to help identifying regions of the molecule
that could be modulated by medicinal chemistry SAR efforts. The model suggested that
the 1-methyl-1,4-dihydrochromeno[4,3-c]pyrazole core fits in a hydrophobic pocket
formed by amino acid residues F288, A295, L291, I356, I366, M332, S336, M369, Y373,
and M454. The core has a limited contact with helix 12 W473 via the methyl substituent.
In addition, the core oxygen forms a weak hydrogen bond with sidechain O of Y373 (O –
O distance 3.6 A). The benzyl ring fits into a lipophilic pocket formed by residues I273,
I277, L291, I339, L352, and I356. Other binding features are hydrogen bonds formed by
the central and the terminal amide oxygen atoms with amino acid residues S336 and
R335 respectively.
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
8

Journal of Medicinal Chemistry
Page 10 of 58
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
Figure 2. Model of 2 docked into FXR ligand binding domain from X-ray structure of WAY-
362450 (FLI.pdb)
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
To take advantage of the small lipophilic niche around the phenyl group of the tricyclic
core (formed by sidechains of residues F288, L291, I361, and M454), substitutions at the
8
-position (C8, R ) were explored (Table 1). Small lipophilic substituents led to a 3- to 30-
1
fold improvement in potency (3-8) compared to the unsubstituted original hit (2), whereas
polar substitutions such as amides, primary alcohols, diols, aldehyde, carboxylic acid
were not tolerated (data not shown). In nearly all cases, metabolic stability in HLM and
RLM was poor, while acceptable aqueous solubility was retained only with Cl (3) and Br
(4) substituents.
Starting with compounds 3 and 4, additional core modifications were explored in an
attempt to further improve potency and metabolic stability (Figure 3). Removal of the
pyrazole’s methyl substituent (17), as well as the regioisomeric pyrazole (18), each led to
inactive compounds. Replacement of the same methyl group with larger substituents was
not tolerated, as exemplified by the isopropyl analog 19. Also, inactive were analogs
ACS Paragon Plus Environment
9

Page 11 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
bearing a gem-dimethyl substitution on the chromene C4 4-position (20) as well as the
amine obtained by reduction of the amide in the 3-position (C3, 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
ACS Paragon Plus Environment
10

Journal of Medicinal Chemistry
Page 12 of 58
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
a
Table 1. R and R Optimization
1
2
1
N
N
O
O
R1
8
3
N
N
H
4
R2
O
FXR HTRF
FXR BSEP-Luc
Solubility
pH=6.8
Metabolic
cLogD
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
Cmpound
R
1
R2
EC50 (M)
EC50 (M)
stability RLM,
(
pH 6.8)
a
a
(
%Eff)
(%Eff)
FASSIF (µM)
HLM (ER)
-
-
-
OCA
LJN452
2
-
-
3.4
10 (96)
42 (63)
0.69 (89)
0.69 (67)
-
3.2b
2.5
0.540 (93)
0.480 (70)
110
0.51, <0.20
0.97, 0.71
>175
H
3
4
5
6
7
8
Cl
Br
3.4
3.5
3.8
4.5
3.5
3.2
0.171 (70)
0.038 (73)
0.013 (70)
0.027 (68)
0.071 (71)
0.145 (62)
0.150 (70)
0.050 (76)
0.024 (64)
0.023 (40)
0.186 (49)
0.280 (79)
>175
-
73
3
-
0.92, 0.75
I
0.84, 0.74
sec-butyl
cyclopropyl
vinyl
-
9
-
0.9, 0.74
-
9
2.6
0.6
0.5
2.0
4.0
3.7
>10 (2)
>10 (2)
>10 (3)
-
-
-
-
Cl
H
10
-
H
11
>10 (2)
-
-
-
H
Me
N
1
2
3
4.8 (59)
2.84 (53)
0.025 (82)
2.17 (62)
4.84 (54)
0.044 (76)
<2
-
-
Br
Cl
Br
1
-
F
14
15
<2
0.91, 0.79
F
F
3.7
3.7
0.016 (78)
0.53 (68)
0.039 (79)
0.52 (57)
80
41
0.89, 0.75
0.88, 0.71
Br
Br
16
ACS Paragon Plus Environment
11

Page 13 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
b
Values are the mean of three or more experiments. Measured LogD pH6.8 = 1.9.
Next, we examined the SAR at the benzyl group (R ), which fits in the larger lipophilic
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
pocket (defined by I273, I277, L291, I339, L352, and I356) (Table 1). Addition of a methyl
at the benzylic position (9), complete deletion of the benzyl group (10) or replacement of
the benzyl with a methyl group (11) were not tolerated changes. Replacement of the
phenyl with a pyridine (12) or a cyclohexyl (13) resulted in reduced potency. However,
potent analogs were obtained by adding a fluoro in the ortho- and meta-positions of the
phenyl group (14, 15 vs 4). In contrast, para-fluoro substitution resulted in a 10-fold
decrease in potency (16 vs 4). However, in nearly all examples tested, metabolic stability
remained poor.
HN
O
N
O
N
N
O
N
N
O
O
O
O
Br
Br
Br
N
N
N
N
N
N
H
H
H
O
O
17
FXR HTRF EC50 > 10 M (3%)
FL FXR BSEP-luc EC50 = 11 M (13%)
18
FXR HTRF EC50 = 6 M (22%)
FL FXR BSEP-luc EC50 > 20 M (20%)
19
FXR HTRF EC50 = 6 M (29%)
FL FXR BSEP-luc EC50 = 2.7 M (35%)
N
N
O
N
N
O
N
O
N
O
Br
N
N
NH2
N
N
N
H
H
O
O
O
20
21
22
FXR HTRF EC50 = 4.3 M (14%)
FL FXR BSEP-luc EC50 > 10 M (10%)
FXR HTRF EC50 > 10 M (5%)
FL FXR BSEP-luc EC50 > 10 M (1%)
FXR HTRF EC50 > 10 M (1%)
FL FXR BSEP-luc EC50 > 10 M (1%)
ACS Paragon Plus Environment
12

Journal of Medicinal Chemistry
Page 14 of 58
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
Figure 3. Additional core and sidechain modifications. Values are the mean of three or
more experiments.
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
To address the issue of poor metabolic stability, we focused our optimization efforts on
the region of the scaffold that occupies the polar pocket. The interaction between the
terminal amide’s oxygen and R335 corresponds to the similar interaction of the carboxylic
33
acid in tropifexor with R331. The importance of such an interaction was demonstrated
by removal of the terminal amide from hit compound 2, which led to complete loss of
activity (22) (Figure 3). This suggested that the ethylendiamine moiety functions as a
spacer to position the remainder of the molecule into an appropriate geometry to allow
for a productive interaction of the terminal amide’s oxygen to R335. Carboxylic acid
moieties can often help improving metabolic stability in lipophilic molecules by decreasing
lipophilicity and increasing polarity (reducing LogD). Therefore, we sought to replace the
N-ethyl acetamide moiety with a carboxylic acid to target the preferred salt-bridge
interaction with R335, exemplified by the binding position of tropifexor. Inspired by
modeling, we designed and synthesized a variety of compounds with different linkers of
ACS Paragon Plus Environment
13

Page 15 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
varying
length
and
geometry
connecting
the
N-benzyl-1-methyl-1,4-
dihydrochromeno[4,3-c]pyrazole-3-carboxamide core to a carboxylic acid. Three distinct
series stood out with attractive FXR agonist activity, and optimized analogs from each of
the three series are shown in 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
Compounds 23, 24 and 1 all displayed a partial agonist activity with potent EC values.
50
Each of these three compounds also displayed attractive pharmacokinetics in rodents
with low-to-moderate clearance, moderate Vss, and short half-life. The bioavailability in
rat from a solution formulation was acceptable (22-51%); however, compounds 24 and 1
exhibited a dramatic reduction in bioavailability from a suspension formulation, suggesting
that poor aqueous solubility was limiting absorption. In order to improve bioavailability
from suspension, we explored a variety of salts, including the tris salts which are shown
in Table 2. Tris salts did not improve the bioavailability of 23, which remained at about
15% from suspension at both 10 and 100 mg/kg, comparable to the bioavailability
achieved with the neutral compound. Compound 24 showed a drop in bioavailability at a
higher solution dose and was not tested in suspension. Benzylic derivative 1, as a tris
salt, showed good bioavailability at both 10 and 100 mg/kg, likely due to the higher
ACS Paragon Plus Environment
14

Journal of Medicinal Chemistry
Page 16 of 58
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
dissolution rate for the salt compared to the neutral compound (Figure 4). For this reason,
and because of other attractive drug-like properties (e.g. high permeability, low clearance,
MW < 500), compound 1 was selected for further PK characterization to establish its
potential for progression to toxicology studies. Dose escalation PK studies in rat showed
a marginally over-proportional increase in exposure in the dose range 3-300 mg/kg, with
bioavailability ranging from 52-72% and similar total concentrations in plasma and liver
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 5).
Table 2. Frontrunner compounds
2
3
24
1
COOH
N
F
N
N
F
N
N
N
O
O
O
O
COOH
COOH
Cl
F
F
N
N
N
N
N
H
H
O
O
O
F
F
FXR HTRF EC50 (μM) (%Eff)a
0.024 (79)
0.046 (65)
0.007 (83)
0.006 (67)
0.007 (57)
0.032 (50)
FXR BSEP-Luc EC50 (μM) (% Eff)a
cLogD PH 6.8
MW
3.2
536
1.5
550
86
2.7g
487
94
PAMPA (Calc %F)
RLM ER
96
0.8
0.55
<0.21
46
<0.18
<0.21
34
HLM ER
<0.21
>175
19.9
1.43
1.97
Sol pH=6.8 (μM)
CL (mL/min/kg)b
Vss (L/kg)b
T1/2 (h)b
9.6
7
1.6
0.8
3
3.6
ACS Paragon Plus Environment
15

Page 17 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
%
%
%
F from solution (10 mg/kg)c
F from suspension (10 mg/kg)d
F TRIS salt (10 mg/kg)e
22
17
15
36
5
51
10
52
82
56f
28f
% F TRIS salt (100 mg/kg)e
12
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
a
b
Values are the mean of three or more experiments; Rat. IV: 3 mg/kg - PEG 300/D5W,
c
d
e
3
0
1
:1, solution; PO: PEG 300/D5W, 3:1; PO: 0.5% CMC/0.5% Tween 80; Suspension:
f
g
.5% CMC/0.5% Tween 80; Solution: 20 PEG300, 10% solutol . Measured LogD
=
pH6.8
.9.
0
.08
Tris Salt
Neutral
0.06
0
0
0
.04
.02
.00
0
20
40
60
Time (min)
Figure 4. Dissolution of 1 in FaSSIF increases for the Tris salt versus the free base.
Dose (mg/kg)a
Tissue
b
AUC (h*nM) Tissue /Plasma
3
10
30
100
300
AUC, po (h*M) 9.2
Cmax, po (M) 4.5
25.4
12.4
5.7
52
76.5
28.1
6.3
52
398.7 1040.2
Plasma
Liver
24529
36700
1106
-
1.50
0.05
80.9
5.6
82
140.8
6.3
Brain
T
1/2 (h), po
F(%)
3.9
63
72
AUC/Dose
3058 2540 2548 3987 3467
Figure 5. a) Rat dose escalation PK, PO, Tris salt, suspension: 0.5% MC/ Tween80; b)
Tissue distribution in rat, PO, 10 mg/kg, Tris salt. Suspension: 0.5% MC/ Tween80.
ACS Paragon Plus Environment
16

Journal of Medicinal Chemistry
Page 18 of 58
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
To further characterize the biological activity of 1, the induction of FXR target genes
(
BSEP and SHP) was monitored by quantitative real-time PCR (qPCR) in rat primary
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
hepatocytes. Compound 1 induced BSEP mRNA levels ~ 3.7–fold compared to a DMSO
control. Consistent with a partial agonist effect of 1, the induction of BSEP was ~ 65%,
and the induction of SHP was ~ 24% compared to the full agonist tropifexor (Figure 5).
Compound 1 showed about 100-fold lower potency than tropifexor, consistent with what
was observed in the FXR BSEP-luc reporter gene assay (Table 1).
Figure 5. Induction of FXR target genes BSEP and SHP in vitro in rat primary hepatocytes
by compound 1 compared to LJN452 (tropifexor). Primary rat hepatocytes were treated
for 24 h with compound 1 or vehicle only, and gene expression was determined by real-
time quantitative PCR. ***p≤0.0001 compared to vehicle by one-way ANOVA.
ACS Paragon Plus Environment
17

Page 19 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Extensive preclinical profiling across a panel of enzymes, ion channels, nuclear
receptors and GPCRs demonstrated the high selectivity of compound 1 for FXR. In
particular, it is worthwhile mentioning that compound 1 did not display activation of human
GPBAR1, as determined by measuring the increase of intracellular cAMP levels (EC >
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
80 M, 13 % Efficacy). In addition there was no activation of other nuclear receptors such
as AR, ER, GR, PPAR, PR, PXR using a TR-FRET assay (EC > 30  and no activation
50
of RXR, LXR, GR, PPAR, ER using a cell-based agonist assay (EC > 10 .
50
Compound 1 and tropifexor were also compared for their ability to induce FXR target
genes in vivo in rats following 2 weeks of dosing. Both test compounds were administered
orally once daily using a range of six dose levels and compared to the vehicle control
group. Gene induction was measured at 1, 3 and 7 hours post-dose, and peak induction
for each gene at each dose was plotted as shown in Figure 6. Treatment of rats with
compound 1 resulted in the dose-dependent upregulation of FXR target genes and
showed a rightward shift in potency compared to tropifexor, as expected from the in vitro
profiling. Interestingly, whereas compound 1 showed partial activity for BSEP in liver and
FGF15 in the ileum, full gene activation was observed for SHP in both the liver and the
ACS Paragon Plus Environment
18

Journal of Medicinal Chemistry
Page 20 of 58
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
ileum. While we do not have a firm explanation for this behavior, we hypothesize that SHP
induction might be integrating FXR and FGF signals in vivo. In addition, SHP expression
induces a negative feedback loop leading to the repression of CYP7a1 and CYP8b1
expression, both of which are central to the synthesis of BAs from cholesterol and
therefore potentially explaining the full repression of CYP8b1. It has to be noted that
several studies have demonstrated that although both Cyp7a1 and Cyp8b1 can be
repressed by SHP in liver upon FXR activation, Cyp7a1 is regulated in a FGF15-
dependent mechanism.40,41
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
Samples from a different multi-dose study were evaluated to determine the effects of
compound 1 on serum FGF15 levels in rat. The direct FXR-dependent induction of
FGF15/19 along with FGF15/19’s anti-cholestatic properties make it a convenient and
relevant serum biomarker for detecting target engagement and potentially predicting
therapeutic efficacy. Figure 7 depicts the dose-dependent (and time-dependent) increase
in circulating FGF15 protein levels by 1. At a dose of 1.5 mg/kg, a statistically significant
(p<0.01) 3.5-fold increase in serum FGF15 protein levels was detected at 10 hours
following the final dose. Interestingly, at the same dose and time point, a 30-fold increase
ACS Paragon Plus Environment
19

Page 21 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
in FGF15 mRNA was observed in the ileum. Two distinct possibilities might explain these
results: i) the increase in serum FGF15 levels may require a higher compound dose than
that increasing FGF15 mRNA level in the ileum or ii) FGF15 produced in the ileum may
enter the portal circulation leading to hepatic clearance of FGF15 prior to appearance in
the serum. In addition, current antibodies might present limitations for detecting FGF15
in serum presumably due to its low stability.42
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 6. Han-Wistar rats were dosed by oral gavage once daily for 14 days with vehicle
control, 1 (red line) and LJN452 (blue line) at the indicated doses. Doses: 1, 0.1, 0.3, 1.5,
7.5, 25, 100 mg/kg; LJN452: 0.003, 0.01, 0.03, 0.1, 0.3, 1.0, 3.0. Vehicle: 0.5%
methylcellulose, 0.5% Tween80, 99% water, suspension. Animals were sacrificed for
ACS Paragon Plus Environment
20

Journal of Medicinal Chemistry
Page 22 of 58
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
terminal tissue collection at 1, 3 and 7 h post-dose on day 14 for analysis. Gene levels
were determined by Taqman quantitative RT-PCR. Data is graphed by indicating peak
fold induction of FXR target genes at each dose, calculated by dividing the expression
level at a given dose by the level of the vehicle control group. N = 3 animals per group
with a minimum of 2 technical replicates per animal. Individual graphs for 1 are reported
in Figure S1 and S2 in the Supporting Information; data for LJN452 was previously
published.33
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 7. Han-Wistar rats were dosed by oral gavage, once daily for 14 days with vehicle
control or 1 at the indicated doses. On day 13, serum FGF15 levels were determined at
the indicated time points. A) time dependence; B) 10 h. Values are represented as mean
±
SEM. N = 3 animals per group.
ACS Paragon Plus Environment
21

Page 23 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The serum exposure of 1 was measured on day 1 and day 13 (Figure 8 and Supporting
Information). On day 1 and 13, serum exposure of LMB763 increased approximately
dose-proportionally from 0.1 to 100 mg/kg. Exposure at 0.1 and 0.3 mg/kg was likely
underestimated on day 13. No significant accumulation was observed (<2-fold) over the
dosing period.
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 8. Han-Wistar rats were dosed by oral gavage, once daily for 14 days with vehicle
control or 1 at the indicated doses. Day 1 and Day 13 mean PK profiles. Average values
(n=3) are reported.
Next, we sought to investigate the potential of compound 1 to improve
pathophysiological features in a murine model of NASH. Unfortunately, there are no
animal models recapitulating all aspects of the complex biology underlying NASH.
ACS Paragon Plus Environment
22

Journal of Medicinal Chemistry
Page 24 of 58
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
Therefore, we leveraged the use of an insulin-deficient murine NASH model, the STAM
model, characterized by hepatic steatosis, fibrosis, inflammation and hepatocellular
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
43
ballooning (Figure 9). The STAM model has been used extensively to ascertain the
effect of drugs such as silymarin on steatosis and provides a reliable model mimicking
the natural progression from steatosis to fibrosis and, ultimately, hepatocellular
carcinoma. Nevertheless, it is also important to also consider the limitations of the model,
namely, the lack of weight gain and therefore, the inability to test the role of a given
44
compound on insulin sensitivity. In this model, diabetes is initially induced by injection
of streptozotocin (STZ; a cytotoxic glucose analogue that induces pancreatic beta cell
death) in neonatal C57BL/6 mice (2-day-old), followed by feeding with a 60% high-fat diet
from week 4 to week 9. Four weeks after STZ delivery and 2 weeks after HFD initiation,
STAM mice were treated with 1 (3, 10, 30 mg/kg), OCA (25 mg/kg) or the corresponding
vehicle controls for 3 weeks. In addition to normal animals (non-STAM; “Normal” control
group), a different group of STAM mice was left untreated to establish a disease baseline
for comparative purposes (“NASH control”) (Figure 9A). At 9 weeks of age, animals were
sacrificed and their livers analyzed. Hematoxylin and eosin (H&E)-stained sections were
ACS Paragon Plus Environment
23

Page 25 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
evaluated for NAFLD Activity Score (NAS) according to the criteria of Kleiner et al. (Figure
45
9
B and 9C). As expected, NASH control animals exhibited micro- and macrovesicular
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
fat deposition, hepatocellular ballooning and inflammatory cell infiltration, resulting in a
significant increase in NAS compared to the normal control group (NAS 5.3 ± 0.5
compared to 0.0 ± 0.0) (Figure 9B and 9C). Treatment with compound 1 led to the
reduction of the various NAS parameters in a dose-dependent manner with the 10 and
30 mg/kg doses demonstrating statistically significant reductions (Figure 9B).
Interestingly, treatment with OCA (25 mg/kg) did not lead to a significant reduction in NAS
in the STAM model, thereby highlighting superior beneficial pharmacodynamic effects of
the non-bile acid FXR agonist 1 (Figure 9B). Triglycerides were unaffected with both
compound 1 and OCA (Figure 9B), consistently with previous results.23
To further investigate the role of compound 1 on liver fibrosis in the context of NASH,
we next evaluated collagen deposition using Picro Sirius Red staining. Quantification of
Picro Sirius Red-stained areas, demonstrated the appearance of fibrosis in NASH control
and vehicle-treated animals compared to the normal control group (Figure 9B and 9D).
Further comparisons demonstrated the reduction of fibrosis in the livers of animals treated
ACS Paragon Plus Environment
24

Journal of Medicinal Chemistry
Page 26 of 58
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 compound 1. Animals treated with OCA showed a trend towards reduced fibrosis;
however, this was not statistically-significant (Figure 9B and 9D). Of note, the observed
disconnect between the NAFLD score and the Sirius red data is potentially due to the
presence of outlier animals as exemplified by the error bars. In addition, NALFD activity
scores include histopathological readouts other than only fibrosis, and represents the
aggregate of steatosis, lobular inflammation, ballooning and fibrosis scores.45
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
25

Page 27 of 58
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 9. Compound 1 reduces NASH activity scores and limits liver fibrosis. A)
Schematic representation of the STAM NASH murine model and treatment regimen; B)
Treatment with LMB763 significantly reduces NAFLD activity scores, triglyceride levels
ACS Paragon Plus Environment
26

Journal of Medicinal Chemistry
Page 28 of 58
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
and liver fibrosis; C) and D) Representative H&E and Sirius Red images demonstrating
the lack of histopathological NASH features (C) upon treatment with 1 and reduced liver
fibrosis (D). Values are represented as mean ± SEM. *P < 0.01, **P < 0.001, and ***P <
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
0.0001 compared to vehicle control by one‐way ANOVA with Dunnett's test. N = 3 animals
per group.
Finally, we sought to evaluate the mechanisms by which compound 1 might be exerting
its effects in the context of NASH. We therefore conducted preliminary transcriptomic
profiling experiments (RNA sequencing; RNAseq) in the livers of mice from the
experiment described above. First, we compared NASH and control mice groups to
determine the gene signature associated with NASH. As expected, NASH mice
demonstrated differential regulation of a large number of genes (1436) compared to
normal control animals (Figure 10A). Gene Ontology (GO) analyses revealed an over-
representation of pathways related to inflammation, morphogenesis and cell
differentiation, in line with the pathophysiological changes associated with NASH
(Supporting Information Figure S3A). Second, we determined NASH genes modulated by
ACS Paragon Plus Environment
27

Page 29 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compound 1 and OCA relative to their respective vehicle controls. Compound 1 treatment
led to 888 differentially expressed genes, including reported canonical FXR target genes
as Nr0b2, OSTb, Map3k13, Aldh1l2 and Cyp8b1. Venn diagram comparisons
demonstrated that compound 1’s beneficial effects were largely attributed to the reversion
of NASH-specific gene expression signatures. Indeed, compound 1 was shown to
modulate the expression of a total of 250 genes differentially regulated in the context of
NASH compared to 26 genes such genes modulated by OCA (Figure 10A).
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 addition, unsupervised cluster analyses highlighted the segregation of a compound
1
cluster independently of OCA and further removed from the NASH group (Figure 10B).
Interestingly, the highest dose of compound 1 was closer to OCA than lower doses. This
apparent dose-response discrepancy might be explained by compound 1 eliciting both
direct effects on FXR target genes as well as indirect effects on the genes directly
regulated by the bile acid sensor. In line with this possibility, comparison of the gene
signatures of compound 1 and the full FXR agonist tropifexor (Supporting Information
Figure S3B) demonstrated a broader response of tropifexor (1219 DEGs) compared to
compound 1 (888 DEGs) and OCA (179 DEGs). It is noteworthy to consider that these
ACS Paragon Plus Environment
28