G-protein-coupled bile acid receptor 1 (GPBAR1, TGR5) agonists reduce the production of proinflammatory cytokines and stabilize the alternative macrophage phenotype

Article abstract of DOI:10.1021/jm501052c

GPBAR1 (also known as TGR5) is a G-protein-coupled receptor (GPCR) that triggers intracellular signals upon ligation by various bile acids. The receptor has been studied mainly for its function in energy expenditure and glucose homeostasis, and there is little information on the role of GPBAR1 in the context of inflammation. After a high-throughput screening campaign, we identified isonicotinamides exemplified by compound 3 as nonsteroidal GPBAR1 agonists. We optimized this series to potent derivatives that are active on both human and murine GPBAR1. These agonists inhibited the secretion of the proinflammatory cytokines TNF-α and IL-12 but not the antiinflammatory IL-10 in primary human monocytes. These effects translate in vivo, as compound 15 inhibits LPS induced TNF-α and IL-12 release in mice. The response was GPBAR1 dependent, as demonstrated using knockout mice. Furthermore, agonism of GPBAR1 stabilized the phenotype of the alternative, noninflammatory, M2-like type cells during differentiation of monocytes into macrophages. Overall, our results illustrate an important regulatory role for GPBAR1 agonists as controllers of inflammation.

Full text of DOI:10.1021/jm501052c

Article
pubs.acs.org/jmc
G‑Protein-Coupled Bile Acid Receptor 1 (GPBAR1, TGR5) Agonists
Reduce the Production of Proinflammatory Cytokines and Stabilize
the Alternative Macrophage Phenotype
Klemens Hogenauer,*,† Luca Arista,† Niko Schmiedeberg,† Gudrun Werner,‡ Herbert Jaksche,‡
̈
Rochdi Bouhelal,§ Deborah G. Nguyen,∥,⊥ B. Ganesh Bhat,∥ Layla Raad,‡ Celine Rauld,‡
and Jose M. Carballido‡
́
‡
§
†Global Discovery Chemistry, Autoimmunity, Transplantation and Inflammation, and Center for Proteomic Chemistry, Novartis
Institutes for BioMedical Research, Novartis Campus, CH-4056 Basel, Switzerland
∥Genomics Institute of the Novartis Research Foundation, 10675 John Jay Hopkins Drive, San Diego, California 92121, United States
S
* Supporting Information
ABSTRACT: GPBAR1 (also known as TGR5) is a G-
protein-coupled receptor (GPCR) that triggers intracellular
signals upon ligation by various bile acids. The receptor has
been studied mainly for its function in energy expenditure and
glucose homeostasis, and there is little information on the role
of GPBAR1 in the context of inflammation. After a high-
throughput screening campaign, we identified isonicotinamides
exemplified by compound 3 as nonsteroidal GPBAR1 agonists.
We optimized this series to potent derivatives that are active
on both human and murine GPBAR1. These agonists inhibited
the secretion of the proinflammatory cytokines TNF-α and IL-12 but not the antiinflammatory IL-10 in primary human
monocytes. These effects translate in vivo, as compound 15 inhibits LPS induced TNF-α and IL-12 release in mice. The response
was GPBAR1 dependent, as demonstrated using knockout mice. Furthermore, agonism of GPBAR1 stabilized the phenotype of
the alternative, noninflammatory, M2-like type cells during differentiation of monocytes into macrophages. Overall, our results
illustrate an important regulatory role for GPBAR1 agonists as controllers of inflammation.
suppresses cytokine secretion via impaired NF-κB signaling.22
INTRODUCTION
■
In addition, a study in human macrophages reported that bile
Since the discovery of the G-protein-coupled bile acid
acids leave the production of antiinflammatory cytokine IL-10
receptor 1 GPBAR1 (also known as TGR5),1 there has been
unaffected and only inhibit the secretion of proinflammatory
an increased interest to pharmacologically modulate this target.
cytokines TNF-α, IL-6 and IL-12p40.23 This observation
GPBAR1 was found to be involved in lipid and glucose
indicates that GPBAR1 activation does not completely block
metabolism as well as energy expenditure, and in this context,
cytokine expression but rather shifts the pattern toward that of
receptor agonists are being pursued by various pharmaceutical
an antiinflammatory macrophage.24
companies to offer treatment options for conditions such as
We became interested in GPBAR1 in the course of our
obesity, hypercholesterolemia, hypertriglyceridemia, and non-
search for selective immunomodulatory principles capable of
alcoholic steatohepatitis.2 Most importantly, the finding that
rebalancing immune inflammatory responses without inducing
GPBAR1 agonism induces the secretion of clinically relevant
systemic immunosuppressive effects. The goal of our drug
glucagon-like peptide 1 (GLP-1) raised expectations of an
discovery program was to identify rodent cross-reactive
alternative therapeutic mechanism for the treatment of type 2
compounds suited to advance target validation and develop-
diabetes mellitus3 and triggered extensive research efforts.4−17
ment of orally efficacious antiinflammatory therapeutics.
The potential of this receptor in immunology and
inflammation was outlined when bile acids (BAs), acting as
RESULTS AND DISCUSSION
■
receptor agonists, showed reduced lipopolysaccharide (LPS)
Target Biology and Hit-Finding. We tested various
induced release of the proinflammatory cytokines TNF-α, IL-1,
conjugated and unconjugated bile acids in a Ca2+-mobilization
assay in Jurkat cells stably expressing human (hu) or murine
(mu) GPBAR1 after retroviral transduction. Although GPBAR1
IL-6, and IL-8 in rabbit alveolar macrophages.1 For a long
period, the link to immunology was the subject of very few
follow-up studies.18−20 Recently, it was shown that bile acids
drive dendritic cell (DC) differentiation from monocytes
toward a IL-12 hypo-producing phenotype via the GPBAR1
pathway.21 Studies in mice suggest that GPBAR1 agonism
Received: July 15, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
is coupled to adenylyl cyclase through the Gαs GTP binding
protein, it proved possible to monitor its activation through
endogenous Gα15 and Gα16 promiscuous protein in Jurkat cells.
This association enables coupling of the vast majority of Gαs
and Gαi coupled receptors to PLCβ and Ca2+-mobilization
following receptor activation. We found that taurolitocholic
acid (TLCA, 1, Figure 1) was one of the most potent and cross-
also inhibited the production of TNF-α and IL-12 in isolated
human monocytes (Table 2). Elevations in cellular cAMP have
been shown to inhibit production of TNF-α from myeloid cells,
further supporting a link between Gs-coupled GPBAR1
activation and modulation of inflammatory cytokine produc-
tion.25,26 Similar to reported observations with macrophages,23
secretion of the antiinflammatory cytokine IL-10 remained at
DMSO control levels. Selective inhibition of proinflammatory
cytokines has been suggested to be a major driver for monocyte
differentiation toward the M2 macrophage phenotype which is
associated with resolution of inflammatory processes.27 This
offers new therapeutic opportunities for diseases characterized
by an overshooting Th1/Th17 cell component (such as
multiple sclerosis, psoriasis, or type 1 diabetes).
We screened the Novartis compound archive in order to find
nonsteroidal GPBAR1 agonists in a Ca2+-mobilization assay
format. We identified a number of chemically diverse hits that
specifically activate GPBAR1, as they only showed significant
induction of Ca2+-mobilization in cells expressing GPBAR1.
Active hits (total number 1229) were also tested in an
orthogonal assay format (cAMP production), and compounds
that were inactive in this assay were not further pursued. For
compounds that were active in both assays (in total 384), a
reasonable correlation (R2 = 0.46) between EC50 values was
observed (Figure 2). Differences are explained by GPBAR1
kinetics, as different incubation times were employed in the two
assays. To ensure both proximal and downstream functional
response of our agonists, we decided to run both assays in
parallel during our optimization program. After this hit
validation, we became interested in a cluster around the hit
compounds 2 and 3 (Table 1). These comparatively small
derivatives showed considerable activity on the human
GPBAR1 receptor. The moderate lipophilicity of these
compounds (HT-logP = 3.4 for compound 2 and HT-logP =
3.2 for compound 3)28 made them attractive starting points for
lead optimization (hu cAMP LipE = 3.9 for compound 2 and
hu cAMP LipE = 3.6 for compound 3, based on measured
log P).29
Figure 1. Structure of taurolitocholic acid (TLCA, 1).
reactive bile acids (hu pEC50 = 7.27, mu pEC50 = 6.54) with
comparable activity on cAMP stimulation (Table 1). TLCA
Table 1. cAMP and Ca2+-Mobilization Assay Data Directed
the Optimization of the Carboxylic Acid Fragment of
Carboxamides
Synthesis. Derivatives of isoxazole 2 and isonicotinamide 3
were assembled in a straightforward manner forming the amide
bond in the final step. Noncommercially available secondary
amines were prepared using reductive amination. Unexpectedly,
the amide bond formation was quite problematic, especially
when combinations of hindered carboxylic acids and electron
deficient aniline building blocks were used. Many standard
coupling conditions tested (DCC, EDC, HOBt, PyBOP,
HATU) were either low-yielding or failed to give any product.
We then switched to preparing the carboxylic acid chlorides in
situ from the carboxylic acids. Either 1-chloro-N,N,2-
trimethylprop-1-en-1-amine30 or POCl3/pyridine gave accept-
able yields for this transformation. Scheme 1 shows the
synthesis of indole 15 as a representative example.
In Vitro SAR. Both compounds 2 and 3 were inactive in the
murine version of the Ca2+-mobilization assay and only showed
high micromolar activity in the murine cAMP assay. As one of
our goals was to show pharmacological effects in a rodent
model, our optimization efforts were geared toward establishing
cross-reactivity for the mouse receptor. When probing the SAR
of the carboxylic acid fragment (Table 1), we found that 3-
methylpyridine 5 combined a more than 10-fold potency
increase on the human receptor with at least measurable activity
on the mouse ortholog. Lipophilicity for this derivative is still in
a reasonable range (HT-logP = 3.2), resulting in an overall
a
pEC50 values are the mean of at least three runs performed in four
replicates with standard deviation in parentheses. For n = 2, individual
values are reported.
B
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 2. cAMP, Ca2+-Mobilization and Primary Cell Assay Data Guiding the Optimization of Isonicotinamides
a
pEC50 and pIC50 values are the mean of at least three runs performed in four replicates with standard deviation in parentheses. For n = 1 or n = 2,
individual values are reported.
improved LipE of 4.8. The position of the methyl group is
important, as shifting it into the 2-position of the pyridine
resulted in a significant activity drop in all assays (2-
methylpyridine 4). Altering the position of the nitrogen as in
pyridine isomer 6 led to complete activity loss. Adding another
nitrogen (pyridazine 7) was detrimental for mouse cross-
reactivity. We also tested a number of five-membered
heterocyclic carboxylic acid fragments. Whereas dimethylisox-
azole 2 was selective for human GPBAR1, methylimidazole 8
showed comparable cross-reactivity potential to 3-methylpyr-
idine 5.
Next, we turned to investigate the substitution pattern on the
amine moiety (Table 2). Since neither one-carbon homologues
of benzylanilide 5 (amides 9 and 10) nor replacements of the
benzyl group by alkyl substituents (e.g., isobutyl derivative 11)
led to an improved murine assay activity, we decided to keep
the N-benzylaniline framework and focus on optimization of
the substitution pattern. Considerable activity gain in the
mouse assays could be achieved by introducing ortho-
substituents on the aromatic ring of the benzylamine. The 2-
methoxybenzyl derivative 12 showed a ∼4-fold activity gain on
murine GPBAR1 compared to the unsubstituted benzylanilide
5. When we combined this modification with appropriate
substitution on the anilinic phenyl group, we observed a
significant activity increase on both human and mouse
receptors. Ortho-trifluoromethoxyanilide 13 and dichloroani-
lide 14 are subnanomolar human GPBAR1 agonists with
considerable potency in the mouse cAMP assay. These
derivatives were also tested for their ability to inhibit LPS
induced cytokine release in primary human monocytes. Indeed,
both carboxamides 13 and 14 potently inhibit the secretion of
TNF-α and the p40 subunit of IL-12 with in vitro activities 100-
fold higher than TLCA. Similar to our observations with TLCA,
IL-10 secretion remains unaffected, highlighting that GPBAR1
C
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. Structure of the recently disclosed GPBAR-1 agonist
piperidinecarboxamide 16.17
Table 3. Comparison of Cytokine Inhibition Profile in
Human PBMCs for Key Compounds
a
hu PBMCs pIC50
a
compd
hu GPBAR1 pEC50, cAMP
TNF-α
IL-10
1
7.70 (0.29)
9.42 (0.12)
8.52 (0.48)
<5.00
6.84
<5.00
<5.00
<5.00
Figure 2. Out of a total number of 1229 hits in the hu GPBAR1 Ca2+-
mobilization assay, 384 hits (shown as dots and triangles) were also
active in the hu cAMP assay. pEC50 values were determined as n = 1 in
a miniaturized version of the assays. Cluster representatives described
in the text are indicated as black triangles. The black line represents
least-squares regression line (R2 = 0.46).
15
16
7.04
a
pEC50 and pIC50 values are the mean of at least three runs performed
in four replicates with standard deviation in parentheses. For n = 1,
individual values are reported.
antagonist settings, the compound did not generate a response
up to a concentration of 10 μM).
agonists selectively inhibit proinflammatory cytokine produc-
tion. For our planned in vivo model in mice, we continued to
search for compounds with an improved Ca2+-mobilization
activity on murine cells without compromising activity on
human primary cells. We identified indole 15 as a compound
Role of GPBAR1 on Macrophage Differentiation.
Because GPBAR1 agonists effectively suppress proinflamma-
tory cytokine production by macrophages, we assessed whether
their presence during monocyte differentiation would also affect
lineage commitment toward proinflammatory (M1) or
proresolving (M2) macrophage types. The experiment involved
the differentiation of human primary monocytes in the
presence of macrophage colony-stimulating factor (M-CSF)
either alone or together with IFN-γ or IL-4 to drive neutral,
M1, or M2 differentiation, respectively.27 Surface and cytokine
analysis was performed using a mass cytometer (CyTOF),
which assesses the simultaneous presence of large arrays of
antibodies, conjugated to monoisotopic lanthanides, by
monitoring the time-of-flight of the different metal tags.31
Figure 4 shows that cell populations separated into three
distinct clusters after a 6-day differentiation period, based on a
spanning-tree progression analysis of density-normalized events
(SPADE)31 performed using CyToBank software. GPBAR1
agonist indole 15 did not change the fate of macrophage
differentiation imposed by the cytokine cocktails. Particularly,
indole 15 did not alter the expression of surface markers such
as CD38 or CD200R and CD206, which are classically
associated with M1 or M2 differentiation, respectively
(Supporting Information Figure 1). If at all, GPBAR1 agonism
enhanced the expression of overall co-stimulatory molecules on
both M1 and M2 cell populations and potentiated the lineage
associated markers on both M1 and M2 cell types (Supporting
Information Figure 1), suggesting that the presence of
with an attractive cellular profile (murine Ca2+-flux EC50
=
7.43). Compared to the initial hit compound 3, indole 15 is still
reasonably small and displays moderate lipophilicity (HT-logP
= 3.8),28 illustrating that we were able to optimize potency in a
lipophilicity efficient manner (hu cAMP LipE = 5.6). The
activity in human primary cells of indole 15 was also compared
to piperidine-3-carboxamide 16 (Figure 3), a recently published
potent GPBAR1 agonist from a different structural class
(chronologically, that series was discovered after the carbox-
amides described in the current manuscript).17 In human
peripheral blood mononuclear cells (PBMCs),17 secretion of
the key cytokine TNF-α was inhibited to a similar degree by
both compounds (Table 3). In addition neither compound
showed an effect on the antiinflammatory cytokine IL-10.
Presumably because of its high unspecific binding, TLCA 1 did
not show any inhibition up to 10 μM. In a safety panel screen
of 64 off-target assays (for a full list, see Supporting
Information), indole 15 showed only moderate inhibition of
human opiate κ (IC50 = 1.3 μM) and opiate μ receptors (IC50 =
5.9 μM) while all other targets tested displayed IC50 values of
>10 μM. In addition, indole 15 did not show any activity on
FXR, a nuclear hormone receptor responsive to bile acids (in a
cellular Gal4 reporter gene assay, run in both agonist and
Scheme 1. Reductive Amination/Amide Bond Formation Sequence Exemplified for the Synthesis of Indole 15
D
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. SPADE analysis on all different data files, including macrophages differentiated with the different combinations of cytokines in the presence
or absence of indole 15 (3 μM), generates three distinct cell clusters, labeled M0, M1, and M2. The intensity of expression of representative M1 and
M2 differentiation markers CD38 and CD206, respectively, on the different (M0, M1, and M2) clusters of macrophage populations differentiated
with the indicated cytokine cocktail in the absence of GPBAR1 agonists is indicated as a heat map. Each sphere represents a potential cell
subpopulation, and its size is proportional to the number of cells contained in such subgroup. Cytokine cocktails forced >85% of the differentiating
cells to group in one of the three major clusters.
Figure 5. Intracellular cytokine production by resting and LPS-stimulated human M2 macrophages. The presence of indole 15 (3 μM) during M2
cell differentiation primes macrophages for sustained reduction of inflammatory cytokine/chemokine production. Shown are results of intracellular
analysis of M2 cells gated on single, living, CD45 positive cells, expressing CD14 and CD64 markers, which were differentiated for 7 days with M-
CSF and IL-4 in the absence or presence of indole 15 (3 μM). Cells were left unstimulated or were activated with LPS for an additional 24 h in the
presence or absence of indole 15 (3 μM). Expression is visualized as a relative intensity heat map of independent CyTOF dual counts per second.
Values for the proinflammatory cytokines IFN-γ, TNF-α, and IL-6 and chemokine CCL18 are indicated inside the cells to enhance data visualization.
The CCL18 heat map is presented separately from the other cytokines because of its higher relative expression.
polarizing cytokines is determinant over GPBAR1 signaling in
inducing macrophage differentiation. M2 cells, previously
differentiated in the presence or absence of GPBAR1 agonist,
were harvested and recultured without differentiating factors for
an additional 24 h in control medium or medium containing
LPS in the presence or absence of indole 15. In the absence of
any polarizing cytokine, M2 cells maintained their overall M2
surface phenotype. However, the cells differentiated in the
presence of the GPBAR1 agonist maintained a higher
expression of co-stimulatory molecules (such as CD11c,
CD40, CD64, and HLA-DR) as well as higher expression of
the M2 lineage markers (CD200R and CD206) than their
counterparts derived from cultures devoid of indole 15 during
differentiation (Supporting Information Figure 2). These effects
were specific, since other low abundance M2 cell surface
molecules, such as CD38, were not increased. Cytokine
production was also simultaneously assessed at the single
level to determine whether chronic or acute exposure to
E
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
GPBAR1 agonism would change the frequencies of cytokine
producing cells. The direct effects of indole 15 on the
frequency of cytokine producing cells were marginal. In
contrast, both overall baseline and LPS induced proinflamma-
tory factor production were consistently lower in the cultures
exposed to indole 15 during the 6-day differentiation period
(Figure 5, Supporting Information Figures 3 and 4). In
summary, these results suggest that GPBAR1 agonism not only
suppresses the overall production of proinflammatory cytokines
but also contributes to the stabilization of the antiinflammatory
M2 cell phenotype (as observed by the larger frequency of cells
displaying M2 surface markers and reduced inflammatory factor
production).
In Vivo Characterization. During our optimization efforts,
we determined high in vitro clearance in microsomal
preparations (human and rodent) throughout the series. This
metabolic instability was log P independent and likely caused
by the presence of a number of oxidizable functionalities and
the potential for metabolic switching. As indole 15 also showed
prominent in vitro metabolism (t1/2 in mouse liver microsomes
of 2 min),32 we did not expect prolonged exposure in vivo.
Indeed, in vivo clearance in mice was high and terminal half-life
was short, yet the compound showed rapid absorption (early
Figure 6. Indole 15 (50 mg/kg) inhibits LPS induced TNF-α
production in WT but not in GPBAR1 KO mice. Box plot (generated
using a Microsoft Excel 2010 template, with bottom and top of the box
showing first and third quartiles, inside band showing median,
whiskers representing minimum and maximum of all data; outliers
significantly outside 1.5 times interquartile range were removed)
shows the results from a minimum of n = 8 individual mice. veh =
vehicle control, dex = dexamethasone (5 mg/kg).
pg/mL) to a mean of 578 pg/mL (95% CI = 965−191 pg/mL,
p = 0.005). The effect was modest compared to the
antiinflammatory glucocorticoid dexamethasone (5 mg/kg).34
In KO mice, treatment with indole 15 had no effect on TNF-α
levels, confirming that the effect observed in WT mice was
indeed GPBAR1 dependent. Figure 7 illustrates that indole 15
Tmax) and a moderate oral bioavailability (Table 4). In order to
a
Table 4. Mouse PK Parameters for Indole 15
route (dose in mg·kg−1
CL [mL·min−1·kg−1
VSS [L·kg−1
t1/2 term. [h]
)
iv (2.5)
po (10)
]
146
3.0
]
0.64
6.9
b
AUC dn [μg·min·mL−1
]
2.8
41
abs oral BAV [%]
b
Cmax dn [nM]
95
Tmax [h]
0.5
a
Male C57BL/6 mice; each time point from three single sacrificed
animals. iv: bolus administration of 2.5 mg·kg−1, dissolved in N-
methylpyrrolidone/mouse plasma = 10:90, v/v, sampling 5, 30, 60,
120, 240, 480, 1440 min after administration. po: 10 mg·kg−1,
suspended in 0.5% carboxymethyl cellulose and 0.5% polysorbate 80 in
water, sampling 15, 30, 60, 120, 240, 480, 1440 min after
b
Figure 7. Indole 15 (50 mg/kg) inhibits LPS induced IL-12p70
production in WT but not in GPBAR1 KO mice. Box plot (description
see Figure 6) shows results from a minimum of n = 8 individual mice.
administration. dn = dose normalized.
ensure adequate exposure, pharmacokinetic experiments were
also conducted at high dose. After some experimentation, we
found that the use of a microemulsion (45% Cremophor RH40,
36% corn oil glycerides, 9% propylene glycerol, 10% ethanol)33
resulted in both sufficient exposure and low inter-animal
variability. By use of this formulation, both total and unbound
mouse plasma levels (mouse PPB = 98%, determined by rapid
equilibrium dialysis) exceed the EC50 in the murine Ca2+-
mobilization assay in the first 5 h after dosing at a po dose of 50
mg/kg (Figure 8). As in the acute model of LPS induced
cytokine release the terminal readout is taken 3.5 h after dosing,
these parameters were considered sufficiently supportive for in
vivo efficacy evaluation of indole 15. It should be noted that the
previously mentioned series around piperidine-3-carboxamide
16, which combines a comparable cellular profile with better
pharmacokinetics, had not been discovered at that time.
markedly inhibits LPS induced IL-12 production from a mean
of 456 pg/mL (95% CI = 523−389 pg/mL) to a mean of 205
pg/mL (95% CI = 262−148 pg/mL, p = 7 × 10−5); the
reduction is similar to dexamethasone treatment. The GPBAR1
dependence of this effect is only observed as a trend, most
probably due to the low IL-12 production by the vehicle treated
KO group. Serum levels of IL-10, IL-1β, IL-6, mKC, and IFN-γ
were below detection limit (data not shown).
SUMMARY AND CONCLUSIONS
■
We have developed a mouse cross-reactive series of GPBAR1
agonists with high potency in both cAMP production and Ca2+-
mobilization assays. In agreement with early findings using bile
acids as tools, optimized compounds such as indole 15 are
potent inhibitors of the proinflammatory cytokines TNF-α and
IL-12 in isolated human monocytes, whereas levels of the
antiinflammatory cytokine IL-10 remain unaffected. Using
indole 15, we could demonstrate the contribution of
GPBAR1 agonism to the overall differentiation and stability
Vehicle treatment of both WT and KO animals resulted in
similar levels of TNF-α production (Figure 6). Indole 15 (50
mg/kg po) inhibited LPS induced TNF-α production in WT
animals from a mean of 1554 pg/mL (95% CI = 1955−1153
F
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
min. High resolution mass analyses (HRMS) were performed by using
electrospray ionization in positive mode after separation by liquid
chromatography. The elemental composition was derived from the
averaged mass spectra acquired at a high resolution of about 30 000 on
an LTQ Orbitrap XL mass spectrometer (Thermo Scientific). The
high mass accuracy below 1 ppm was obtained by using a lock mass.
N-Benzyl-3,5-dimethyl-N-phenylisoxazole-4-carboxamide
(2). N-Phenylbenzylamine (1.3 g, 7.02 mmol), EDC·HCl (1.66 g, 8.44
mmol), and triethylamine (1.18 mL, 8.42 mmol) were added to a
solution of 3,5-dimethylisoxazole-4-carboxylic acid (1.0 g, 7.01 mmol)
in dry dichloromethane (10 mL), and the resulting mixture was stirred
for 22 h at rt. After that time, the reaction mixture was diluted with
EtOAc (50 mL) and extracted with aqueous HCl solution (1 M, 20
mL), followed by a saturated aqueous NaHCO3 solution (30 mL).
The combined organic layers were dried (MgSO4), and the solvent
was removed under reduced pressure. The product was purified by
flash chromatography using a heptane/EtOAc gradient to yield the
Figure 8. Total and unbound plasma concentration of indole 15 in
Balb/C mice (50 mg/kg po, n = 3, same vehicle as for the in vivo
experiment depicted in Figures 6 and 7, mouse PPB = 98%). The
dotted line indicates the murine Ca2+-mobilization EC50. Error bars
show standard deviation.
1
title compound 2 (240 mg, 0.78 mmol, 11%). H NMR (400 MHz,
CDCl3) δ 7.10−7.36 (m, 8H), 6.85−6.95 (m, 2H), 5.10 (s, 2H), 2.16
(s, 3H), 2.10 (s, 3H). UPLC/MS: 1.07 min, 307.2 [M + H]+. HRMS
(ESI+) calcd for C19H19N2O2 [M + H]+ 307.144 10, found 307.144 32.
N-Benzyl-N-phenylisonicotinamide (3). Isonicotinic acid (500
mg, 4.06 mmol), PyBroP (4.10 g, 8.79 mmol), and triethylamine (1.13
ml, 8.08 mmol) were added to a solution of N-phenylbenzylamine
(619 mg, 3.34 mmol) in dry dichloromethane (10 mL), and the
resulting mixture was stirred for 22 h at rt. After that time, the reaction
mixture was filtered through a syringe filter, and the filtrate was
concentrated under reduced pressure. The residue was purified by
preparative HPLC (Waters Sunfire C18 OBD 30 mm × 100 mm, flow
30 mL/min, water + 0.1% TFA and acetonitrile + 0.1% TFA gradient)
of the antiinflammatory M2 cell phenotype. Macrophages are
very plastic cells, and they can differentiate and redifferentiate
back and forward into M1 and M2 cell subsets in vitro.27
Secondary bile acids are generated in the intestinal tract from
primary bile acids with the help of Clostidrium bacteria. In view
of the prominent role of the microbiome controlling immune
regulation, it is tempting to speculate that GPBAR1 agonists
mimic one of the most ancient immunoregulatory mechanisms
provided by evolution to sustain homeostatic antiinflammatory
cell differentiation.
Although indole 15 was shown to exhibit a short half-life in
mice, the achieved exposure was sufficient to demonstrate in
vivo activity in an acute model of LPS-induced cytokine release.
This compound reduced TNF-α and IL-12 production at 50
mg/kg in vivo, a dose where both total and unbound
compound concentration exceeded EC50 levels in the murine
Ca2+-mobilization assay for this compound over the whole time
of the experiment. Compared to the broad immunosuppressant
dexamethasone, the effects on TNF-α and IL-12 production
were moderate, yet selective for proinflammatory cytokines.
Indole 15 had no effect on cytokine levels of GPBAR1 KO
mice, thereby confirming GPBAR1 dependency for the
observed effects. Overall, we could confirm our initial
hypothesis that GPBAR1 is an appropriate target for the
treatment of inflammatory diseases. However, the modest in
vivo effects of indole 15 at comparatively high doses as well as
the significant ADME hurdles that needed to be overcome led
us to discontinue our efforts in this compound series.
1
to yield the title compound 3 (817 mg, 2.84 mmol, 85%). H NMR
(400 MHz, CD3OD) δ 8.43−8.33 (m, 2H), 7.36−7.12 (m, 10H),
7.06−6.95 (m, 2H), 5.19−5.09 (s, 2H). UPLC/MS: 0.94 min, 289.2
[M + H]+. HRMS (ESI+) calcd for C19H17N2O [M + H]+ 289.133 54,
found 289.133 51.
N-Benzyl-2-methyl-N-phenylisonicotinamide (4). 2-Methyli-
sonicotinic acid (44 mg, 0.32 mmol) was dissolved in 500 μL of a
freshly prepared 1-chloro-N,N,2-trimethylpropenylamine solution in
dry dichloromethane (1 g/ml), and the resulting mixture was stirred
for 2.5 h at rt. Next, a solution of N-phenylbenzylamine (30 mg, 0.162
mmol) and N,N-diisopropylethylamine (55 μL, 0.32 mmol) in 1 mL of
dry dichloromethane was added, and the reaction mixture was then
stirred for 1.5 h at rt. After that time, the reaction mixture was
concentrated under reduced pressure. The residue was purified by
preparative HPLC (Waters Sunfire C18 OBD 30 mm × 100 mm, flow
30 mL/min, water + 0.1% TFA and acetonitrile + 0.1% TFA gradient)
1
to yield the title compound 4 (51 mg, 0.17 mmol, 53%). H NMR
(600 MHz, DMSO-d6) δ 8.44 (s, 1H), 7.52 (br s, 1H), 7.13−7.33 (m,
11H), 5.11 (s, 2H), 2.47 (s, 3H). UPLC/MS: 0.93 min, 303.2 [M +
H]+. HRMS (ESI+) calcd for C20H19N2O [M + H]+ 303.149 19, found
303.149 21.
N-Benzyl-3-methyl-N-phenylisonicotinamide (5). POCl3 (2.37
mL, 25.7 mmol) and pyridine (9.4 mL, 117 mmol) were added to a
stirred solution of N-phenylbenzylamine (4.27 g, 23.3 mmol) and 3-
methylpyridine-4-carboxylic acid (3.20 g, 23.3 mmol) in dry
dichloromethane (6 mL) at rt. The reaction mixture was heated at
reflux temperature for 2 h. After that time, a saturated aqueous
NaHCO3 solution was added (50 mL), followed by the addition of
dichloromethane (50 mL). The aqueous layer was extracted three
times with dichloromethane. The combined organic layers were dried
(MgSO4), and the solvent was removed under reduced pressure. The
product was purified by flash chromatography using a heptane/EtOAc
EXPERIMENTAL SECTION
■
Chemistry. Unless otherwise stated, all reagents and solvents were
purchased from commercial suppliers and used without further
purification. TLCA (1) was purchased from Sigma (catalog no. T-
1
7515). H NMR spectra were recorded on a Bruker 400 MHz or a
Bruker 600 MHz NMR spectrometer. Chemical shifts are reported in
parts per million (ppm) relative to an internal solvent reference.
Recorded peaks are listed in the order multiplicity (s, singlet; d,
doublet; t, triplet; q, quartet; qt, quintet; m, multiplet; br, broad),
coupling constants, and number of protons. Final compounds were
purified to ≥95% purity as assessed by analytical liquid chromatog-
raphy using the following UPLC/MS method: column Acquity HSS
T3 1.8 μm, 2.1 mm × 50 mm at 50 °C; eluent A, water + 0.05% formic
acid + 3.75 mM ammonium acetate; eluent B, acetonitrile + 0.04%
formic acid; gradient, from 2% to 98% B in 1.4 min; flow rate, 1.2 mL/
1
gradient to yield the title compound 5 (6.14 g, 20.3 mmol, 87%). H
NMR (400 MHz, DMSO-d6) δ 8.31 (s, 1H), 8.19 (d, J = 4.41 Hz,
1H), 7.22−7.38 (m, 5H), 6.99−7.20 (m, 6H), 5.11 (br s, 2H), 2.27 (s,
3H). UPLC/MS: 0.95 min, 303.2 [M + H]+. HRMS (ESI+) calcd for
C20H19N2O [M + H]+ 303.149 19, found 303.149 11.
N-Benzyl-4-methyl-N-phenylnicotinamide (6). 1-Chloro-
N,N,2-trimethylpropenylamine (116 μL, 0.87 mmol) was added to a
solution of 4-methylnicotinic acid (80 mg, 0.583 mmol) in dry
G
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
dichloromethane (5 mL), and the resulting pale yellow suspension was
stirred for 10 min at rt. Next, a solution of N-phenylbenzylamine (128
mg, 0.7 mmol) and triethylamine (243 μL, 1.75 mmol) in dry THF (1
mL) was added, and the reaction mixture was stirred for 1 h at rt. After
that time, the reaction mixture was diluted with 30 mL of
dichloromethane, and the organic layer was extracted with a saturated
aqueous NaHCO3 solution (2 × 20 mL). The organic layer was dried
(Na2SO4) and concentrated under reduced pressure. The residue was
purified by preparative HPLC (Waters Sunfire C18 OBD 18 mm × 60
mm, flow 20 mL/min, water + 0.1% TFA and acetonitrile + 0.1% TFA
gradient) to yield the title compound 6 (133 mg, 0.32 mmol, 55%). 1H
NMR (400 MHz, CDCl3) δ 8.33−8.45 (m, 2H), 7.08−7.43 (m, 9H),
6.79−6.91 (m, 2H), 5.12 (s, 2H), 2.55 (s, 3H). UPLC/MS: 0.96 min,
303.2 [M + H]+. HRMS (ESI+) calcd for C20H19N2O [M + H]+
303.149 19, found 303.149 11.
mL of dichloromethane, and the organic layer was extracted with a
saturated aqueous NaHCO3 solution (2 × 5 mL). The organic layer
was dried (Na2SO4) and concentrated under reduced pressure. The
residue was purified by preparative HPLC (Waters Sunfire C18 OBD
18 mm × 60 mm, flow 20 mL/min, water + 0.1% TFA and acetonitrile
+ 0.1% TFA gradient) to yield the title compound 8 (73 mg, 0.24
1
mmol, 40%). H NMR (400 MHz, DMSO-d6) δ 7.51 (s, 1H), 7.02−
7.35 (m, 10H), 5.13 (s, 2H), 3.67 (s, 3H), 1.64 (s, 3H). UPLC/MS:
0.72 min, 306.2 [M + H]+. HRMS (ESI+) calcd for C19H20N3O [M +
H]+ 306.160 09, found 306.160 19.
3-Methyl-N-phenethyl-N-phenylisonicotinamide (9). 3-Meth-
ylisonicotinic acid (137 mg, 1 mmol) was dissolved in of a freshly
prepared solution of 1-chloro-N,N,2-trimethylpropenylamine in dry
dichloromethane (1 g/mL, 1.5 mL), and the resulting mixture was
stirred for 2.5 h at rt. Next, a solution of N-phenethylaniline (99 mg,
0.5 mmol) and N,N-diisopropylethylamine (171 μL, 1 mmol) in dry
dichloromethane (1 mL) was added and stirring was continued for 2 h
at rt. After that time, the reaction mixture was concentrated under
reduced pressure. The residue was purified by preparative HPLC
(Waters Sunfire C18 OBD 30 mm × 100 mm, flow 30 mL/min, water
+ 0.1% TFA and acetonitrile + 0.1% TFA gradient) to yield the title
compound 9 (39 mg, 0.12 mmol, 12%). 1H NMR (600 MHz, DMSO-
d6) δ 8.44 (s, 1H), 8.32 (br s, 1H), 7.13−7.32 (m, 11H), 4.16 (t, J =
7.3 Hz, 2H), 2.91 (t, J = 7.3 Hz, 2H), 2.21 (s, 3H). UPLC/MS: 1.01
min, 317.2 [M + H]+. HRMS (ESI+) calcd for C21H21N2O [M + H]+
317.164 84, found 317.164 82.
Lithium 3-Methylpyridazine-4-carboxylate. A solution of 2-
acetylpent-4-enoic acid ethyl ester (3.85 g, 22.6 mmol) and Sudan III
(0.5 mg) in dry dichloromethane (30 mL) and methanol (3 mL) at
−78 °C was subjected to a stream of O3 in O2 until the color change to
blue was persistent. Next, an amount of 17.5 g of PL-TPP (polymer
bound triphenylphosphine, loading 1.42 mmol/g, 24.8 mmol) was
added, and the cooling bath was removed. After 1 h at rt, the reaction
mixture was filtered and concentrated under reduced pressure, yielding
3-oxo-2-(2-oxoethyl)butyric acid ethyl ester (3.89 g, 22.6 mmol,
100%) which was used in the next step without further purification.
A solution of 3-oxo-2-(2-oxoethyl)butyric acid ethyl ester (3.64 g,
21.1 mmol) in ethanol (35 mL) was slowly treated with 724 μL of
hydrazine hydrate in 10 mL of ethanol at 0 °C. The cooling bath was
removed, and the reaction mixture was stirred for 2.5 h at rt. Next, a
solution of sodium nitrite (2.21 g, 31.7 mmol) in 1 mL of H2O
followed by AcOH (7 mL) was added. After 1 h, a saturated aqueous
solution of NaHCO3 was added dropwise until gas formation stopped
completely. The mixture was extracted with EtOAc (3 × 50 mL). The
combined organic layers were dried (Na2SO4) and concentrated under
reduced pressure. The product was purified by flash chromatography
using a heptane/EtOAc gradient to give ethyl 3-methylpyridazine-4-
N,N-Dibenzyl-3-methylisonicotinamide (10). POCl3 (75.0 μL,
0.82 mmol) and pyridine (202 μL, 2.50 mmol) were added to a stirred
solution of dibenzylamine (96 μL, 0.50 mmol) and 3-methylpyridine-
4-carboxylic acid (68.5 mg, 0.50 mmol) in dry dichloroethane (2 mL)
at rt. The reaction mixture was heated to reflux in a microwave reactor
for 10 min. After that time, a saturated aqueous NaHCO3 solution was
added (10 mL), followed by the addition of dichloromethane (10 mL).
The aqueous layer was extracted with dichloromethane (3 × 10 mL).
The combined organic layers were dried (MgSO4), and the solvent
was removed under reduced pressure. The residue was purified by
preparative HPLC (Waters Sunfire C18 OBD 18 mm × 60 mm, flow
20 mL/min, water + 0.1% TFA and acetonitrile + 0.1% TFA gradient)
1
carboxylate (1.0 g, 6.04 mmol, 28%). H NMR (400 MHz, CDCl3) δ
9.24 (d, J = 5.1 Hz, 1H), 7.83 (d, J = 5.1 Hz, 1H), 4.43 (q, J = 7.1 Hz,
2H), 2.99 (s, 3H), 1.42 (t, J = 7.1 Hz, 3H). ESI-MS 189 (M + Na).
A solution of of ethyl 3-methylpyridazine-4-carboxylate (704 mg,
4.23 mmol) in THF (2 mL) was treated with an aqueous solution of
LiOH (2 M, 2.2 mL) and stirred for 1.5 h at rt. The reaction mixture
was concentrated under reduced pressure yielding crude lithium 3-
methylpyridazine-4-carboxylate which was used in the next step
without further purification.
1
to yield the title compound 10 (114 mg, 0.36 mmol, 72%). H NMR
(400 MHz, DMSO-d6) δ 8.48 (s, 1H), 8.42 (d, J = 4.89 Hz, 1H),
7.26−7.42 (m, 8H), 7.23 (d, J = 4.89 Hz, 1H), 7.09 (br s, 2H), 4.69
(br s, 2H), 4.29 (br s, 2H), 2.19 (s, 3H). UPLC/MS: 1.01 min, 317.2
[M + H]+. HRMS (ESI+) calcd for C21H21N2O [M + H]+ 317.164 84,
found 317.164 89.
N-Isobutyl-3-methyl-N-phenylisonicotinamide (11). POCl3
(68.2 μL, 0.74 mmol) and pyridine (269 μL, 3.35 mmol) were
added to a stirred solution of N-(2-methylpropyl)aniline (100 mg, 0.67
mmol) and 3-methylpyridine-4-carboxylic acid (92 mg, 0.67 mmol) in
dry dichloromethane (2 mL) at rt. The reaction mixture was heated at
reflux temperature for 1.5 h. After that time, a saturated aqueous
NaHCO3 solution was added (10 mL), followed by the addition of
dichloromethane (10 mL). The aqueous layer was extracted with
dichloromethane (3 × 10 mL). The combined organic layers were
dried (MgSO4), and the solvent was removed under reduced pressure.
The residue was purified by preparative HPLC (Waters Sunfire C18
OBD 18 mm × 60 mm, flow 20 mL/min, water + 0.1% TFA and
acetonitrile + 0.1% TFA gradient) to yield the title compound 11 (31
N-Benzyl-3-methyl-N-phenylpyridazine-4-carboxamide (7).
POCl3 (42.3 μL, 0.45 mmol) and pyridine (126 μL, 117 mmol)
were added to a stirred solution of N-phenylbenzylamine (4.27 g, 23.3
mmol) and lithium 3-methylpyridazine-4-carboxylate (50 mg, 0.352
mmol) in dry dichloromethane (5 mL) at rt. The reaction mixture was
stirred for 2 h at rt. After that time, a saturated aqueous NaHCO3
solution was added (20 mL), followed by the addition of
dichloromethane (20 mL). The aqueous layer was extracted with
dichloromethane (3 × 20 mL). The combined organic layers were
dried (MgSO4), and the solvent was removed under reduced pressure.
The product was purified by flash chromatography using a heptane/
EtOAc gradient to yield the title compound 7 (30 mg, 0.10 mmol,
1
1
28%). H NMR (400 MHz, CDCl3) δ 8.83 (d, J = 5.07 Hz, 1H),
mg, 0.12 mmol, 17%). H NMR (400 MHz, DMSO-d6) δ 8.29 (s,
7.22−7.35 (m, 8H), 7.10−7.17 (m, 1H), 6.98 (d, J = 5.07 Hz, 1H),
6.83 (dd, J = 3.20, 6.28 Hz, 1H), 5.11 (s, 2H), 2.75 (s, 3H). UPLC/
MS: 0.89 min, 304.2 [M + H]+. HRMS (ESI+) calcd for C19H18N3O
[M + H]+ 304.144 44, found 304.144 56.
N-Benzyl-1,4-dimethyl-N-phenyl-1H-imidazole-5-carboxa-
mide (8). 1-Chloro-N,N,2-trimethylpropenylamine (138 μL, 1.02
mmol) was added to a solution of 3,5-dimethyl-3H-imidazole-4-
carboxylic acid (101 mg, 0.72 mmol) in dry dichloromethane (4 ml),
and the resulting suspension was stirred for 1 h at rt. Next, N-
phenylbenzylamine (110 mg, 0.60 mmol) and triethylamine (252 μL,
1.80 mmol) were added, and the resulting reaction mixture was stirred
for 2.5 h at rt. After that time, the reaction mixture was diluted with 5
1H), 8.15 (d, J = 4.63 Hz, 1H), 7.12−7.28 (m, 5H), 7.01 (d, J = 4.85
Hz, 1H), 3.75 (d, J = 7.28 Hz, 2H), 2.24 (s, 3H), 1.65−1.78 (m, 1H),
0.93 (d, J = 6.62 Hz, 6H). UPLC/MS: 0.93 min, 269.2 [M + H]+.
HRMS (ESI+) calcd for C17H21N2O [M + H]+ 269.164 84, found
269.164 89.
N-(2-Methoxybenzyl)-3-methyl-N-phenylisonicotinamide
(12). 3-Methylisonicotinic acid (137 mg, 1 mmol) was dissolved in a
freshly prepared 1-chloro-N,N,2-trimethylpropenylamine solution in
dry dichloromethane (1 g/mL, 1.5 mL), and the resulting mixture was
stirred for 2.5 h at rt. Next, a solution of N-phenethylaniline (99 mg,
0.5 mmol) and N,N-diisopropylethylamine (171 μL, 1 mmol) in dry
dichloromethane (1 mL) was added. The resulting solution was then
H
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
stirred for 2 h at rt. After that time, the reaction mixture was
concentrated under reduced pressure. The residue was purified by
preparative HPLC (Waters Sunfire C18 OBD 30 mm × 100 mm, flow
30 mL/min, water + 0.1% TFA and acetonitrile + 0.1% TFA gradient)
NaHCO3 solution (2 × 30 mL). The organic layer was dried
(Na2SO4) and concentrated under reduced pressure. The product was
purified by flash chromatography using a cyclohexane/EtOAc gradient
to give 4-fluoro-2-methyl-N-((2-methyl-1H-indol-4-yl)methyl)aniline
(1.23 g, 3.90 mmol, 62%). 1H NMR (400 MHz, CDCl3) δ 7.95 (br s,
1H), 7.29−7.20 (m, 1H), 7.14−7.04 (m, 2H), 6.81 (ddd, J = 8.8, 6.1,
3.1 Hz, 2H), 6.62 (dd, J = 9.7, 4.8 Hz, 1H), 6.34−6.29 (m, 1H), 4.54
(s, 2H), 3.73 (br s, 1H), 2.47 (d, J = 1.0 Hz, 3H), 2.12 (s, 3H).
N-(4-Fluoro-2-methylphenyl)-3-methyl-N-((2-methyl-1H-
indol-4-yl)methyl)isonicotinamide (15). 3-Methylisonicotinic acid
(350 mg, 2.55 mmol) and pyridine (1.03 mL, 12.76 mmol) were
added to a solution of 4-fluoro-2-methyl-N-((2-methyl-1H-indol-4-
yl)methyl)aniline (1.21 g, 3.83 mmol) in dry THF (40 mL). At 0 °C, a
solution of POCl3 (509 mg, 3.32 mmol) in THF (1 mL) was added
dropwise. The cooling bath was removed, and the reaction mixture
was stirred for 1.5 h at rt. After that time, the reaction mixture was
diluted with EtOAc (50 mL), washed with a saturated aqueous
NaHCO3 solution (50 mL). The organic layer was dried (Na2SO4)
and concentrated under reduced pressure. The residue was purified by
preparative HPLC (Waters Sunfire C18 OBD 18 mm × 150 mm, flow
20 mL/min, water + 0.1% TFA and acetonitrile + 0.1% TFA gradient)
1
to yield the title compound 12 (40 mg, 0.12 mmol, 12%). H NMR
(600 MHz, DMSO-d6) δ 8.44 (1H), 7.52 (br s, 1H), 7.13−7.33 (10
H), 5.13 (s, 2H), 3.70 (s, 3H), 2.37 (s, 3H). UPLC/MS: 0.94 min,
333.2 [M + H]+. HRMS (ESI+) calcd for C21H21N2O2 [M + H]+
333.159 75, found 333.159 75.
N-(2-Methoxybenzyl)-2-(trifluoromethoxy)aniline. 2-
(Trifluoromethoxy)aniline (156 g, 0.88 mmol) and sodium triacetox-
yborohydride (328 mg, 1.47 mmol) were added to a solution of 2-
methoxybenzaldehyde (100 mg, 0.73 mmol) in dry dichloromethane
(1 mL), and the resulting mixture was stirred 2 h at rt. After that time,
a saturated aqueous NaHCO3 solution was added dropwise until gas
evolution had stopped. The mixture was diluted with EtOAc (5 mL)
and washed with saturated aqueous NaHCO3 solution. The organic
layer was dried (Na2SO4) and concentrated under reduced pressure.
The product was purified by flash chromatography using a heptane/
EtOAc gradient to give N-(2-methoxybenzyl)-2-(trifluoromethoxy)-
aniline (163 mg, 0.55 mmol, 75%). 1H NMR (400 MHz, DMSO-d6) δ
7.26−7.18 (m, 1H), 7.15 (dd, J = 7.9, 1.7 Hz, 2H), 7.09−7.02 (m,
1H), 7.00 (d, J = 8.2 Hz, 1H), 6.88−6.82 (m, 1H), 6.56 (td, J = 7.7, 1.4
Hz, 1H), 6.51 (dd, J = 8.2, 1.4 Hz, 1H), 6.15 (t, J = 6.2 Hz, 1H), 4.32
(d, J = 6.2 Hz, 2H), 3.85 (s, 3H).
N-(2-Methoxybenzyl)-3-methyl-N-(2-(trifluoromethoxy)-
phenyl)isonicotinamide (13). 1-Chloro-N,N,2-trimethylpropenyl-
amine (68 μL, 0.51 mmol) was added to a solution of 3-
methylisonicotinic acid (90 mg, 0.30 mmol) in dry dichloromethane
(3 mL), and the resulting mixture was stirred for 1 h at rt. Next, N-(2-
methoxybenzyl)-2-(trifluoromethoxy)aniline (49.8 mg, 0.36 mmol)
and N,N-diisopropylethylamine (127 μL, 0.91 mmol) were added, and
the resulting solution was then stirred for 2 h at rt. After that time, the
reaction mixture was diluted with dichloromethane (5 mL) and
washed with a saturated aqueous NaHCO3 solution. The organic layer
was dried (Na2SO4) and concentrated under reduced pressure. The
residue was purified by preparative HPLC (Waters Sunfire C18 OBD
18 mm × 60 mm, flow 20 mL/min, water + 0.1% TFA and acetonitrile
+ 0.1% TFA gradient) to give the title compound 13 (76 mg, 0.18
mmol, 60%). 1H NMR (400 MHz, DMSO-d6, 4:1 mixture of rotamers,
data given for major isomer) δ 8.37 (s, 1H), 8.16 (d, J = 4.85 Hz, 1H),
7.09−7.31 (m, 6H), 6.84−6.98 (m, 3H), 5.27 (d, J = 14.33 Hz, 1H),
4.76 (d, J = 14.33 Hz, 1H), 3.62 (s, 3H), 2.30 (s, 3H). UPLC/MS:
1.12 min, 417.2 [M + H]+. HRMS (ESI+) calcd for C22H20F3N2O3 [M
+ H]+ 417.142 05, found 417.141 88.
1
to yield the title compound 15 (370 mg, 0.91 mmol, 35%). H NMR
(400 MHz, CDCl3) δ 8.32 (s, 1H), 8.14 (d, J = 5.07 Hz, 1H), 7.95 (br
s, 1H), 7.22 (d, J = 8.16 Hz, 1H), 6.89−6.95 (m, 1H), 6.79 (d, J = 5.07
Hz, 1H), 6.64−6.73 (m, 2H), 6.38−6.46 (m, 3H), 5.81 (d, J = 13.67
Hz, 1H), 4.73 (d, J = 13.67 Hz, 1H), 2.46 (s, 3H), 2.36 (s, 3H), 2.08
(s, 3H). UPLC/MS: 1.00 min, 388.2 [M + H]+. HRMS (ESI+) calcd
for C24H23FN3O [M + H]+ 388.18197, found 388.18188.
Biology. All aspects of animal care, use, and welfare for research
animals used for in vivo studies were performed in compliance with
local governmental regulations. Furthermore, all animals were handled
in accordance with Novartis Animal Care and Use Committee
approved protocols and regulations.
Cell Culture. Stable Jurkat cell lines expressing GPBAR1 were
cultured as suspensions in T175 cc flasks at 37 °C, 5% CO2, and 95%
relative humidity in AIM-V serum-free medium (GIBCO BRL)
supplemented with 1% penicillin (stock solution 100IE/ml) and 1%
streptomycin (stock solution 100 μg/mL, Gibco BRL), 2 mM
glutamine, and 1 μg/mL puromycin (Sigma). Cells were passaged
every second day, and cell density was always kept below 1 × 106 cells
per mL of medium, in particular at harvesting before running the
calcium assay. Cells were split every 2−3 days with a ratio of 1:3 to
1:4.
GPBAR1 Cellular cAMP Assay. Jurkat cells overexpressing human
or mouse GPBAR1 were seeded at (50 000 cells/25 μL)/well in 384-
well plates (Greiner) in AIM V medium (Life Technologies)
supplemented with 1 mM IBMX (Sigma) and 10 mM HEPES.
Immediately after plating, an amount of 200 nL of compounds in
DMSO was transferred using the Pintool (GNF systems). After 1 h at
37 °C, cryptate and D2 conjugates from the cAMP HiRange HTRF kit
(Cisbio) were added (12.5 μL each). Plates were then incubated 1 h at
rt, and emission at 620 and 665 nm was read on the Envision
(PerkinElmer) as suggested by the manufacturer. Data were expressed
as a ratio of emission at 665/620 and then normalized to a DMSO
alone control. Dose−response data points were curve fitted using the
standard logistic regression model. The percent efficacy was then
determined based on N-(3,5-dichlorophenyl)-N-(1-(2-
methoxyphenyl)ethyl)-3-methylisonicotinamide35 as the internal con-
trol. The resultant EC50 values correspond to the compound
concentration that displaces 50% of the binding of the d2-labeled
cAMP.
N-(3,5-Dichlorophenyl)-N-(2-methoxybenzyl)-3-methylisoni-
cotinamide (14). 1-Chloro-N,N,2-trimethylpropenylamine (320 μL,
2.42 mmol) was added to a solution of 3-methylisonicotinic acid (150
mg, 1.09 mmol) in dry dichloromethane (10 mL), and the resulting
mixture was stirred for 45 min at rt. Next, 3,5-dichloro-N-(2-
methoxybenzyl)aniline (339 mg, 1.2 mmol) and N,N-diisopropylethyl-
amine (457 μL, 3.27 mmol) were added, and the resulting solution
was then stirred for 30 min at rt. After that time, the reaction mixture
was diluted with dichloromethane (5 mL) and washed with a saturated
aqueous NaHCO3 solution. The organic layer was dried (Na2SO4) and
concentrated under reduced pressure. The product was purified by
flash chromatography using a heptane/EtOAc gradient to yield the
1
title compound 14 (380 mg, 0.95 mmol, 87%). H NMR (400 MHz,
CD3OD) δ 8.37 (br s, 1H), 8.21−8.29 (m, 1H), 6.81−7.42 (m, 8H),
5.19 (br s, 2H), 3.74 (s, 3H), 2.37 (s, 3H). UPLC/MS: 1.16 min,
401.1 [M + H]+. HRMS (ESI+) calcd for C21H19Cl2N2O2 [M + H]+
401.081 81, found 401.081 70.
4-Fluoro-2-methyl-N-((2-methyl-1H-indol-4-yl)methyl)-
aniline. Dibutyltin dichloride (191 mg, 0.63 mmol) was added to a
solution of 2-methyl-1H-indole-4-carbaldehyde (1.00 g, 6.28 mmol)
and 4-fluoro-2-methylaniline (1.18 g, 9.42 mmol) in THF (15 mL),
and the resulting reaction mixture was stirred for 15 min at rt. Next,
phenylsilane (2.72 g, 25.1 mmol) was added, and stirring was
continued for 1 h at rt. After that time, the reaction mixture was
diluted with EtOAc (80 mL) and washed with a saturated aqueous
GPBAR1 Calcium Mobilization Assay. Experiments were run
using two different plate formats. The miniaturized 384-well format
was used in HTS, whereas the 96-well format was used from hit-to-
lead stage onward.
For the 96-well protocol, Jurkat cells expressing either human or
mouse GPBAR1 were harvested from tissue culture flasks by
centrifugation at 200g and resuspended in Fluo-4 AM dye loading
solution (3 mL per 1 × 107 cells containing 1 μM Fluo-4 AM, 2 mM
probenicid in complete RPMI medium buffered with 20 mM HEPES,
I
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
pH 7.4) on the day of the experiment. After incubation for 1 h at 37
°C, cells were washed twice with 10 mL of assay buffer containing 2
mM probenecid (centrifugation at 200g). Cells were then counted and
transferred into black/clear bottom 96-well plates at 2 × 105 cells per
well in 75 μL of assay buffer. Assay plates were centrifuged for 2 min at
200g before use on the FLIPR instrument (FLIPR I instrument,
Molecular Devices, Sunnyvale, CA, USA).
For the 384-well protocol, Jurkat cells expressing human GPBAR1
were harvested from cell culture flasks, counted, and adjusted to 2 ×
106 cells per mL of assay buffer. An aliquot of 20 μL per well was then
dispensed in 384-well assay plates (black with clear bottom, Corning,
USA) and plates were centrifuged for 3 min at 1200 rpm. After
incubation at 37 °C in a humidified atmosphere for 30 min, cells were
stained by dispensing 20 μL of loading solution containing Fluo-4 2-
fold concentrated into wells, and incubating plates for 1 h at 37 °C.
Subsequently, cells were washed using an EMBLA cell plate washer.
After the last wash, a volume of 20 μL of assay buffer was left over the
cell monolayer. Cells were then allowed to stabilize for 15−20 min
before starting measurements on the FLIPR II instrument (Molecular
Devices). Data were expressed as a normalized fluorescence ratio (ΔF/
F = (Fmax − Fb)/Fb) where Fb corresponds to f values prior to agonist
(TLCA) or compound injection and Fmax to the fluorescence at the
signal peak. Data were then normalized to TLCA values.
Human Monocyte Isolation, Differentiation, and Activation.
Human peripheral blood mononuclear cells (PBMCs) were isolated
from blood of healthy volunteers (Blutspendezentrum Basel) by
density gradient centrifugation over Ficoll-Paque PLUS (GE Health-
care) using 50 mL Leucosep tubes (Greiner Bio-One). Subsequently,
monocytes were isolated from PBMCs by magnetic removal of
complementary leukocyte lineages using an untouched monocyte kit
on a RoboSep instrument (both from Stemcell Technologies). Purified
monocytes were differentiated into macrophages by 6 days culture in
complete medium (CM) consisting of RPMI 1640 Glutamax medium
supplemented with 10% fetal calf serum (PAA Laboratories), 10 μg/
mL penicillin/streptomycin, 1 mM sodium pyruvate, 25 mM HEPES,
and 50 μM β-mercaptoethanol, all from Life Technologies. Isolated
monocytes were stimulated overnight with 10 ng/mL LPS (Sigma) in
the presence of the indicated compounds or control (DMSO), and
cytokine production was assessed by specific ELISA Ready-SET-Go
kits (Affymetrix-eBioscience).
For M0/M1/M2 phenotyping experiments, isolated monocytes
were differentiated into macrophages subsets by culturing them for 6
additional days at 2.5 × 106 cells/well in 6-well tissue culture plates
(Corning Costar) in the presence of 40 ng/mL M-CSF for M0
differentiation, 40 ng/mL M-CSF with 50 ng/mL IFN-γ for M1
differentiation, or 40 ng/mL M-CSF with 50 ng/mL IL-4 for M2
differentiation, cytokine cocktails (all from R & D Systems) in the
presence of indole 15 (3 uM) or control DMSO. At day 6, an aliquot
of 1 × 106 cells from each condition was used to evaluate surface
markers by mass cytometry using a CyTOF instrument (DVS-
Fluidigm).
anti-HLA-DR (DVS)/Yb174, and anti-CD303 (Miltenyi)/Lu175
.
Cisplatin solution (Enzo Life Sciences) was used at 25 mM final
concentration. DNA intercalator Ir191/193 (DVS Sciences) was used at
1:2000. CyTOF buffer consisted of PBS/CMF + 0.1% BSA + 2 mM
EDTA + 0.01% NaN3. Tru-Stain fcx (anti-human CD16/32,
BioLegend) was used at 5 μL per 1 × 106 cells in CyTOF buffer to
block unspecific binding of the antibodies. Permeabilization buffer for
cytokine staining was purchased from eBioscience.
Cell cultures were harvested into a 15 mL tube and spun down to
discard supernatants. Subsequently, they were washed twice with 1 mL
of CyTOF buffer and transferred separately into 1.5 mL Eppendorf
tubes, where they were resuspended and incubated with cisplatin
solution for 1 min. Cells were spun down again and exposed to 500 μL
of Fc block solution for 5 min at 4 °C. Subsequently, cells were stained
with antibody cocktails in 100 μL of CyTOF buffer by incubating
them during 30 min at 4 °C after gently vortexing the samples. After
two additional washes, the cells were fixed adding 100 μL per
condition fixation solution (Phosphotop, Roche) during 15 min at rt.
After two more washes, the cells were stained with intracellular
staining cocktail during 1 h at 4 °C. Following the incubation, the cells
were resuspended again, and 500 μL of intercalation solution was
added and incubated during 15 min at rt. Cells were washed once
more and incubated 15 min at rt with 70% methanol to achieve
dehydratation. After a final wash with CyTOF buffer, the cells were
spun down and the pelleted cells were adjusted at 1 × 106 per mL
concentration with CyTOF buffer immediately prior to CyTOF
acquisition.
Cells were analyzed on a CyTOF mass cytometer (DVS Sciences)
at an event rate of around 500 cells per second. The initial settings and
parameters of the instrument were set according to DVS process and
recommendations. Before gating of the cell subpopulations, generated
FCS files were normalized using a standard protocol provided by the
Nolan lab (Stanford University).36 Analysis, including cell density
plots, heat maps, and SPADE were created in Cytobank (http://www.
cytobank.org/).
LPS Induced Cytokine Release in Mice. WT animals were
obtained from Jackson laboratories (C57BL/6). GPBAR1 knockout
mice were obtained from Taconic (catalog no. TF0504, USA) and
back-crossed to get B6 background before use in experiments. Male
WT and KO mice (7−12 weeks old, 8−9 mice/group) were treated
with vehicle, dexamethasone (5 mg/kg), or indole 15 (50 mg/kg, 10
mL/kg, po) 2 h before the LPS challenge. Vehicle formulation used
was 45% Cremophor RH40, 36% corn oil glycerides, 9% propylene
glycerol, and 10% ethanol. Following drug treatment, all animals were
challenged with 40 μg/kg of LPS in saline ip. Whole blood was
collected 90 min after injection. All collected blood was stored at room
temperature until separation. Once centrifuged, the resulting serum
was drawn off into a labeled 96-well plate and stored at −80 °C until
analyzed. Sera were analyzed for TNF-α, IL12-p70, IL-10, IL-1β, IL-6,
mKC, and IFN-γ using MSD multiplex Elisa kit.
At day 7, the remaining differentiated macrophages were stimulated
during 2 h with 10 ng/mL LPS (Sigma) in the presence of test
compound or control (DMSO) and subsequently supplemented
overnight with brefeldin A (Affymetrix eBioscience) to avoid secretion
of produced cytokines. At day 8, the cells were analyzed for
extracellular and intracellular markers by mass cytometry using a
CyTOF instrument.
ASSOCIATED CONTENT
■
S
* Supporting Information
Surface analysis results, intensity heat maps, and a complete list
of safety panel assays. This material is available free of charge
via the Internet at http://pubs.acs.org.
Staining of Cell for Mass Cytometry. Metal-labeled antibodies
were prepared using MAXPAR antibody labeling kit (DVS Sciences)
according to the manufacturer’s protocol. The following antibody/
labels (100 μL/2.5 × 106 cells) were used: anti-CD25 (eBioscience)/
La139, anti-TNF-α (R&D)/Nd143, anti-IL-23p19 (R&D)/Sm147, anti-
CD40 (Biolegend)/Nd148, anti-IL-6 (Biolegend)/Sm149, anti-CD200R
(Biolegend)/Sm152, anti-IL-10 (R&D)/Eu153, anti-CD45 (DVS)/
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: klemens.hoegenauer@novartis.com. Phone:
0041613246225.
Present Address
Sm154, anti-CD33 (DVS)/Gd158 or anti-IFN-γ (Biolegend)/Gd158
,
⊥D.G.N.: Organovo Holdings Inc., 6275 Nancy Ridge Drive,
Suite 110, San Diego, CA 92121.
anti-CD11c (DVS)/Tb159, anti-CD14 (Biolegend)/Gd160, anti-IL-1β
(R&D)/Dy162, anti-CD38 (Biolegend)/Er167, anti-CD206 (Biole-
gend)/Er168, anti-CD71 (Biolegend)/Tm169, anti-CCL18 (Leinco)/
Notes
Er170, anti-CD64 (Biolegend)/Yb171, anti-IL-12p40 (R&D)/Yb172
,
The authors declare no competing financial interest.
J
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
A.; Hadcock, J. R.; Hepworth, D.; Herr, M.; Hinchey, T.; Janssen, A.
M.; Jennings, S. M.; Jiao, W.; Lavergne, S. Y.; Li, B.; Li, M.; Munchhof,
M. J.; Orr, S. T. M.; Piotrowski, D. W.; Roush, N. S.; Sammons, M.;
Stevens, B. D.; Storer, G.; Wang, J.; Warmus, J. S.; Wei, L.; Wolford, A.
C. Optimization of Triazole-Based TGR5 Agonists towards Orally
Available Agents. MedChemComm 2012, 4, 205−210.
(11) Piotrowski, D. W.; Futatsugi, K.; Warmus, J. S.; Orr, S. T. M.;
Freeman-Cook, K. D.; Londregan, A. T.; Wei, L.; Jennings, S. M.;
Herr, M.; Coffey, S. B.; Jiao, W.; Storer, G.; Hepworth, D.; Wang, J.;
Lavergne, S. Y.; Chin, J. E.; Hadcock, J. R.; Brenner, M. B.; Wolford, A.
C.; Janssen, A. M.; Roush, N. S.; Buxton, J.; Hinchey, T.; Kalgutkar, A.
S.; Sharma, R.; Flynn, D. A. Identification of Tetrahydropyrido[4,3-
d]pyrimidine Amides as a New Class of Orally Bioavailable TGR5
Agonists. ACS Med. Chem. Lett. 2013, 4, 63−68.
(12) Londregan, A. T.; Piotrowski, D. W.; Futatsugi, K.; Warmus, J.
S.; Boehm, M.; Carpino, P. A.; Chin, J. E.; Janssen, A. M.; Roush, N.
S.; Buxton, J.; Hinchey, T. Discovery of 5-Phenoxy-1,3-dimethyl-1H-
pyrazole-4-carboxamides as Potent Agonists of TGR5 via Sequential
Combinatorial Libraries. Bioorg. Med. Chem. Lett. 2013, 23, 1407−
1411.
ACKNOWLEDGMENTS
■
We thank R. Kuhr, P. Schmid, M. Boras, F. Amberger, R.
̈
Scheinecker, P. Nganga, Y. Regereau, and K.-S. Tucking
̈
(synthetic support); R. Feifel, S. Desrayaud, P. Guntz, L.
Hoffmann, G. Laue, Y. Chen, G. Welzel, H. Sullivan, L.
Tenaillon, A. Marcel, and J. Lehmann (in vitro and in vivo
studies); R. Embacher, S. Guldbrandsen, and C. Schwarzler
(CyTOF assistance and sample acquisition).
̈
ABBREVIATIONS USED
■
BA, bile acid; DC, dendritic cell; dn, dose normalized; FLIPR,
fluorometric imaging plate reader; GLP-1, glucagon-like
peptide 1; GPBAR1, G-protein-coupled bile acid receptor 1;
HT-logP, high-throughput log P; KO, knockout; LipE,
lipophilicity efficiency; LPS, lipopolysaccharide; PBMC,
peripheral blood mononuclear cell; rt, room temperature;
TGR5, Takeda G-protein-coupled receptor 5; TLCA, taur-
olitocholic acid; TLC, thin layer chromatography; WT, wild
type
(13) Martin, R. E.; Bissantz, C.; Gavelle, O.; Kuratli, C.; Dehmlow,
H.; Richter, H. G. F.; Obst Sander, U.; Erickson, S. D.; Kim, K.;
́
Pietranico-Cole, S. L.; Alvarez-Sanchez, R.; Ullmer, C. 2-Phenoxy-
REFERENCES
nicotinamides Are Potent Agonists at the Bile Acid Receptor GPBAR1
■
(1) Kawamata, Y.; Fujii, R.; Hosoya, M.; Harada, M.; Yoshida, H.;
Miwa, M.; Fukusumi, S.; Habata, Y.; Itoh, T.; Shintani, Y.; Hinuma, S.;
Fujisawa, Y.; Fujino, M. A G Protein-Coupled Receptor Responsive to
Bile Acids. J. Biol. Chem. 2003, 278, 9435−9440.
(2) Schaap, F. G.; Trauner, M.; Jansen, P. L. M. Bile Acid Receptors
as Targets for Drug Development. Nat. Rev. Gastroenterol. Hepatol.
2014, 11, 55−67.
(3) Pols, T. W. H.; Noriega, L. G.; Nomura, M.; Auwerx, J.;
Schoonjans, K. The Bile Acid Membrane Receptor TGR5 as an
Emerging Target in Metabolism and Inflammation. J. Hepatol. 2011,
54, 1263−1272.
(4) Evans, K. A.; Budzik, B. W.; Ross, S. A.; Wisnoski, D. D.; Jin, J.;
Rivero, R. A.; Vimal, M.; Szewczyk, G. R.; Jayawickreme, C.; Moncol,
D. L.; Rimele, T. J.; Armour, S. L.; Weaver, S. P.; Griffin, R. J.;
Tadepalli, S. M.; Jeune, M. R.; Shearer, T. W.; Chen, Z. B.; Chen, L.;
Anderson, D. L.; Becherer, J. D.; De Los Frailes, M.; Colilla, F. J.
Discovery of 3-Aryl-4-isoxazolecarboxamides as TGR5 Receptor
Agonists. J. Med. Chem. 2009, 52, 7962−7965.
(5) Pellicciari, R.; Gioiello, A.; Macchiarulo, A.; Thomas, C.;
Rosatelli, E.; Natalini, B.; Sardella, R.; Pruzanski, M.; Roda, A.;
Pastorini, E.; Schoonjans, K.; Auwerx, J. Discovery of 6α-Ethyl-23(S)-
methylcholic Acid (S-EMCA, INT-777) as a Potent and Selective
Agonist for the TGR5 Receptor, a Novel Target for Diabesity. J. Med.
Chem. 2009, 52, 7958−7961.
(6) Budzik, B. W.; Evans, K. A.; Wisnoski, D. D.; Jin, J.; Rivero, R. A.;
Szewczyk, G. R.; Jayawickreme, C.; Moncol, D. L.; Yu, H. Synthesis
and Structure−Activity Relationships of a Series of 3-Aryl-4-
isoxazolecarboxamides as a New Class of TGR5 Agonists. Bioorg.
Med. Chem. Lett. 2010, 20, 1363−1367.
(7) Herbert, M. R.; Siegel, D. L.; Staszewski, L.; Cayanan, C.;
Banerjee, U.; Dhamija, S.; Anderson, J.; Fan, A.; Wang, L.; Rix, P.;
Shiau, A. K.; Rao, T. S.; Noble, S. A.; Heyman, R. A.; Bischoff, E.;
Guha, M.; Kabakibi, A.; Pinkerton, A. B. Synthesis and SAR of 2-Aryl-
3-aminomethylquinolines as Agonists of the Bile Acid Receptor TGR5.
Bioorg. Med. Chem. Lett. 2010, 20, 5718−5721.
(8) Gioiello, A.; Rosatelli, E.; Nuti, R.; Macchiarulo, A.; Pellicciari, R.
Patented TGR5 Modulators: A Review (2006 − Present). Expert Opin.
Ther. Pat. 2012, 22, 1399−1414.
(TGR5). ChemMedChem 2013, 8, 569−576.
(14) Dehmlow, H.; Alvarez Sanchez, R.; Bachmann, S.; Bissantz, C.;
́
Bliss, F.; Conde-Knape, K.; Graf, M.; Martin, R. E.; Obst Sander, U.;
Raab, S.; Richter, H. G. F.; Sewing, S.; Sprecher, U.; Ullmer, C.;
Mattei, P. Discovery and Optimisation of 1-Hydroxyimino-3,3-
diphenylpropanes, a New Class of Orally Active GPBAR1 (TGR5)
Agonists. Bioorg. Med. Chem. Lett. 2013, 23, 4627−4632.
(15) Zhu, J.; Ning, M.; Guo, C.; Zhang, L.; Pan, G.; Leng, Y.; Shen, J.
Design, Synthesis and Biological Evaluation of a Novel Class of Potent
TGR5 Agonists Based on a 4-Phenyl Pyridine Scaffold. Eur. J. Med.
Chem. 2013, 69, 55−68.
(16) Zou, Q.; Duan, H.; Ning, M.; Liu, J.; Feng, Y.; Zhang, L.; Zhu,
J.; Leng, Y.; Shen, J. 4-Benzofuranyloxynicotinamide Derivatives Are
Novel Potent and Orally Available TGR5 Agonists. Eur. J. Med. Chem.
2014, 82, 1−15.
(17) Phillips, D. P.; Gao, W.; Yang, Y.; Zhang, G.; Lerario, I. K.; Lau,
T. L.; Jiang, J.; Wang, X.; Nguyen, D. G.; Bhat, B. G.; Trotter, C.;
Sullivan, H.; Welzel, G.; Landry, J.; Chen, Y.; Joseph, S. B.; Li, C.;
Gordon, W. P.; Richmond, W.; Johnson, K.; Bretz, A.; Bursulaya, B.;
Pan, S.; McNamara, P.; Seidel, H. M. Discovery of Trifluoromethyl-
(pyrimidin-2-yl)azetidine-2-carboxamides as Potent, Orally Bioavail-
able TGR5 (GPBAR1) Agonists: Structure−Activity Relationships,
Lead Optimization, and Chronic in Vivo Efficacy. J. Med. Chem. 2014,
57, 3263−3282.
(18) Keitel, V.; Donner, M.; Winandy, S.; Kubitz, R.; Haussinger, D.
̈
Expression and Function of the Bile Acid Receptor TGR5 in Kupffer
Cells. Biochem. Biophys. Res. Commun. 2008, 372, 78−84.
(19) Pols, T. W. H.; Nomura, M.; Harach, T.; Lo Sasso, G.;
Oosterveer, M. H.; Thomas, C.; Rizzo, G.; Gioiello, A.; Adorini, L.;
Pellicciari, R.; Auwerx, J.; Schoonjans, K. TGR5 Activation Inhibits
Atherosclerosis by Reducing Macrophage Inflammation and Lipid
Loading. Cell Metab. 2011, 14, 747−757.
(20) Pols, T. W. H. TGR5 in Inflammation and Cardiovascular
Disease. Biochem. Soc. Trans. 2014, 42, 244−249.
(21) Ichikawa, R.; Takayama, T.; Yoneno, K.; Kamada, N.; Kitazume,
M. T.; Higuchi, H.; Matsuoka, K.; Watanabe, M.; Itoh, H.; Kanai, T.;
Hisamatsu, T.; Hibi, T. Bile Acids Induce Monocyte Differentiation
toward Interleukin-12 Hypo-Producing Dendritic Cells via a TGR5-
Dependent Pathway. Immunology 2012, 136, 153−162.
(22) Wang, Y.-D.; Chen, W.-D.; Yu, D.; Forman, B. M.; Huang, W.
The G-Protein-Coupled Bile Acid Receptor, GPBAR1 (TGR5),
Negatively Regulates Hepatic Inflammatory Response through
Antagonizing Nuclear Factor Kappa Light-Chain Enhancer of
Activated B Cells (NF-κB) in Mice. Hepatology 2011, 54, 1421−1432.
(9) Duan, H.; Ning, M.; Chen, X.; Zou, Q.; Zhang, L.; Feng, Y.;
Zhang, L.; Leng, Y.; Shen, J. Design, Synthesis, and Antidiabetic
Activity of 4-Phenoxynicotinamide and 4-Phenoxypyrimidine-5-
carboxamide Derivatives as Potent and Orally Efficacious TGR5
Agonists. J. Med. Chem. 2012, 55, 10475−10489.
(10) Futatsugi, K.; Bahnck, K. B.; Brenner, M. B.; Buxton, J.; Chin, J.
E.; Coffey, S. B.; Dubins, J.; Flynn, D.; Gautreau, D.; Guzman-Perez,
K
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(23) Haselow, K.; Bode, J. G.; Wammers, M.; Ehlting, C.; Keitel, V.;
Kleinebrecht, L.; Schupp, A.-K.; Haussinger, D.; Graf, D. Bile Acids
̈
PKA-Dependently Induce a Switch of the IL-10/IL-12 Ratio and
Reduce Proinflammatory Capability of Human Macrophages. J.
Leukocyte Biol. 2013, 94, 1253−1264.
(24) Lewis, N. D.; Patnaude, L. A.; Pelletier, J.; Souza, D. J.; Lukas, S.
M.; King, F. J.; Hill, J. D.; Stefanopoulos, D. E.; Ryan, K.; Desai, S.;
Skow, D.; Kauschke, S. G.; Broermann, A.; Kuzmich, D.; Harcken, C.;
Hickey, E. R.; Modis, L. K. A GPBAR1 (TGR5) Small Molecule
Agonist Shows Specific Inhibitory Effects on Myeloid Cell Activation
in Vitro and Reduces Experimental Autoimmune Encephalitis (EAE)
in Vivo. PLoS One 2014, 9, e100883.
(25) Bailly, S.; Ferrua, B.; Fay, M.; Gougerot-Pocidalo, M. A.
Differential Regulation of IL 6, IL 1 A, IL 1β and TNF-α Production in
LPS-Stimulated Human Monocytes: Role of Cyclic AMP. Cytokine
1990, 2, 205−210.
(26) Foey, A. D.; Field, S.; Ahmed, S.; Jain, A.; Feldmann, M.;
Brennan, F. M.; Williams, R. Impact of VIP and cAMP on the
Regulation of TNF-α and IL-10 Production: Implications for
Rheumatoid Arthritis. Arthritis Res. Ther. 2003, 5, R317.
(27) Sica, A.; Mantovani, A. Macrophage Plasticity and Polarization:
In Vivo Veritas. J. Clin. Invest. 2012, 122, 787−795.
(28) Faller, B.; Grimm, H. P.; Loeuillet-Ritzler, F.; Arnold, S.; Briand,
X. High-Throughput Lipophilicity Measurement with Immobilized
Artificial Membranes. J. Med. Chem. 2005, 48, 2571−2576.
(29) Shultz, M. D. The Thermodynamic Basis for the Use of
Lipophilic Efficiency (LipE) in Enthalpic Optimizations. Bioorg. Med.
Chem. Lett. 2013, 23, 5992−6000.
(30) Haveaux, B.; Dekoker, A.; Rens, M.; Sidani, A. R.; Toye, J.;
Ghosez, L. α-Chloro Enamines, Reactive Intermediates for Synthesis:
1-Chloro-N,N,2-trimethylpropenylamine. Org. Synth. 1979, 59, 26.
(31) Bendall, S. C.; Simonds, E. F.; Qiu, P.; Amir, E. D.; Krutzik, P.
O.; Finck, R.; Bruggner, R. V.; Melamed, R.; Trejo, A.; Ornatsky, O. I.;
Balderas, R. S.; Plevritis, S. K.; Sachs, K.; Pe’er, D.; Tanner, S. D.;
Nolan, G. P. Single-Cell Mass Cytometry of Differential Immune and
Drug Responses across a Human Hematopoietic Continuum. Science
2011, 332, 687−696.
(32) Trunzer, M.; Faller, B.; Zimmerlin, A. Metabolic Soft Spot
Identification and Compound Optimization in Early Discovery Phases
Using MetaSite and LC-MS/MS Validation. J. Med. Chem. 2009, 52,
329−335.
(33) Fatouros, D. G.; Karpf, D. M.; Nielsen, F. S.; Mullertz, A.
Clinical Studies with Oral Lipid Based Formulations of Poorly Soluble
Compounds. Ther. Clin. Risk Manage. 2007, 3, 591−604.
(34) Elenkov, I. J.; Chrousos, G. P. Stress Hormones, Proin-
flammatory and Antiinflammatory Cytokines, and Autoimmunity. Ann.
N.Y. Acad. Sci. 2002, 966, 290−303.
(35) Prepared analogously to compound 14.
(36) Finck, R.; Simonds, E. F.; Jager, A.; Krishnaswamy, S.; Sachs, K.;
Fantl, W.; Pe’er, D.; Nolan, G. P.; Bendall, S. C. Normalization of Mass
Cytometry Data with Bead Standards. Cytometry, Part A 2013, 83,
483−494.
L
dx.doi.org/10.1021/jm501052c | J. Med. Chem. XXXX, XXX, XXX−XXX