Novel substituted isoxazole FXR agonists with cyclopropyl, hydroxycyclobutyl and hydroxyazetidinyl linkers: Understanding and improving key determinants of pharmacological properties

Article abstract of DOI:10.1016/j.bmcl.2016.05.070

Several isoxazole-containing series of FXR agonists have been published over the last 15 years, subsequent to the prototypical amphiphilic ‘hammerhead’-type structure that was originally laid out by GW4064, the first potent synthetic FXR agonist. A set of novel compounds where the hammerhead is connected to the terminal carboxylic acid-bearing aryl or heteroaryl moiety by either a cyclopropyl, a hydroxycyclobutyl or a hydroxyazetidinyl linker was synthesized in order to improve upon the ADME properties of such isoxazoles. The resulting compounds all demonstrated high potencies at the target receptor FXR but with considerable differences in their physicochemical and in vivo profiles. The structure–activity relationships for key chemical features that have a major impact on the in vivo pharmacology of this series are discussed.

Full text of DOI:10.1016/j.bmcl.2016.05.070

Accepted Manuscript
Novel substituted isoxazole FXR agonists with cyclopropyl, hydroxycyclobutyl
and hydroxyazetidinyl linkers: Understanding and improving key determinants
of pharmacological properties
Olaf Kinzel, Christoph Steeneck, Thomas Schlüter, Andreas Schulz, Christian
Gege, Ulrike Hahn, Eva Hambruch, Martin Hornberger, Adriana Spalwisz,
Katharina Frick, Sanja Perović-Ottstadt, Ulrich Deuschle, Michael Burnet,
Claus Kremoser
PII:
S0960-894X(16)30568-6
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.05.070
BMCL 23929
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
4 April 2016
19 May 2016
21 May 2016
Please cite this article as: Kinzel, O., Steeneck, C., Schlüter, T., Schulz, A., Gege, C., Hahn, U., Hambruch, E.,
Hornberger, M., Spalwisz, A., Frick, K., Perović-Ottstadt, S., Deuschle, U., Burnet, M., Kremoser, C., Novel
substituted isoxazole FXR agonists with cyclopropyl, hydroxycyclobutyl and hydroxyazetidinyl linkers:
Understanding and improving key determinants of pharmacological properties, Bioorganic & Medicinal Chemistry
Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.05.070
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

-
1
-
Novel substituted isoxazole FXR agonists with cyclopropyl, hydroxycyclobutyl and
hydroxyazetidinyl linkers: Understanding and improving key determinants of
pharmacological properties
Olaf Kinzela*, Christoph Steenecka, Thomas Schlütera, Andreas Schulza, Christian Gegea,
Ulrike Hahnb, Eva Hambrucha, Martin Hornbergera, Adriana Spalwisza, Katharina Fricka,
Sanja Perović-Ottstadta, Ulrich Deuschlea, Michael Burnetb, and Claus Kremosera,
aPhenex Pharmaceuticals AG, Waldhofer Str. 104, 69123 Heidelberg, Germany
bSynovo GmbH, Paul-Ehrlich Str. 15, 72076 Tübingen, Germany
*Corresponding author. Tel.: +49 6221 652823
E-mail address: olaf.kinzel@phenex-pharma.com
A R T I C L E I N F O
Article history:
Received xxx
Revised xxx
Accepted xxx
Available online xxx
___________________
Keywords:
Farnesoid X receptor
FXR agonist
GW4064
High fat diet
C57BL/6J mice
NAFLD
NASH
___________________
A B S T R A C T
_____________________________________________________________________________________________________
Several isoxazole-containing series of FXR agonists have been published over the last 15 years, subsequent to the prototypical
amphiphilic “hammerhead”-type structure that was originally laid out by GW4064, the first potent synthetic FXR agonist. A set of
novel compounds where the hammerhead is connected to the terminal carboxylic acid-bearing aryl or heteroaryl moiety by
either a cyclopropyl, a hydroxycyclobutyl or a hydroxyazetidinyl linker was synthesized in order to improve upon the ADME
properties of such isoxazoles. The resulting compounds all demonstrated high potencies at the target receptor FXR but with
considerable differences in their physicochemical and in vivo profiles. The structure-activity relationships for key chemical
features that have a major impact on the in vivo pharmacology of this series are discussed.

-
2
-
The Farnesoid X Receptor (FXR) is a member of the
nuclear hormone receptor superfamily and senses bile
acids such as chenodeoxycholic acid (CDCA) and their
taurine or glycine amide conjugates as its endogenous
ligands.1–3 FXR mRNA can be detected throughout the
entire gastrointestinal tract from the esophagus to the
rectum, and in other tissues with exposure to bile acids
such as liver and kidneys.4,5 FXR heterodimerizes with
RXR (Retinoid X-Receptor) and this dimer complex acts as
a
ligand-activated transcription factor to control the
expression of various target genes which are involved in
bile acid, cholesterol, triglyceride and lipoprotein
homeostasis in the liver and circulation.6,7 Furthermore,
FXR regulates complex biological processes beyond
metabolism such as liver regeneration and intestinal
barrier integrity, and has been shown to influence the
enterohepatic immune system through anti-inflammatory
effects.8-11
Thus, FXR has been implicated as a target for novel
pharmacotherapies that address metabolic diseases such
as dyslipidemias, Type 2 Diabetes or the related Metabolic
Syndrome.12 Recently, 6-Et-CDCA (synonyms INT-747 or
Figure 1
Representative FXR agonists of the isoxazole type.
Obeticholic Acid, OCA), which is
derivative of CDCA, that is about 80-fold more potent as a
human FXR demonstrated significant
agonist13,
a semi-synthetic
In general, these compounds are lipophilic carboxylic
acids whose amphiphilic properties limit their utility as drug
candidates with regard to aqueous solubility and
formulation for oral administration. Furthermore, off-target
effects with membrane proteins in general and anion
membrane transporters in particular are believed to
increase with increasing lipo- and amphiphilicity.27,28 Apart
from GW4064, whose limited plasma exposure upon oral
dosing has been described,33 very limited data addressing
in vivo pharmacological properties of isoxazole-type FXR
agonists have been published.
improvements in insulin sensitivity and other beneficial
metabolic effects in a phase IIa study in patients with Non-
Alcoholic Fatty Liver Disease (NAFLD).14 The same
compound also yielded a significant reduction in Alkaline
Phosphatase (a biomarker for the extent of liver
impairment) in patients with Primary Biliary Cirrhosis (PBC)
in phase IIb and phase III studies.15,16 Further phase IIb
data indicate histopathological improvements in patients
with Non-alcoholic Steatohepatitis (NASH) upon treatment
with OCA over 72 weeks.17
We have previously drawn attention to the similarity
between synthetic isoxazole-type FXR agonists and
natural bile acids by shape and polarity distribution, and
this similarity manifests not only in FXR potency but also in
recognition by Bile Acyl: Coenzyme A Synthase (BACS)
and Bile Acyl Coenzyme A: Amino Acid N-acyltransferase
(BAAT), enzymes that conjugate bile acids: taurine and
glycine amides were found as phase II metabolites of 6 in
vivo.23 These earlier isoxazoles turned out to be very
potent with regard to lipid lowering in a high fat diet mouse
model, and the challenge here was to find novel isoxazole-
type FXR agonists of similar or improved potency with the
additional requirements of increased polarity, reduced
amphiphilicity and consequently improved ADME
properties.
Here we describe the rationale and medicinal
chemistry program that resulted in a series of FXR
agonists
with
improved
physicochemical
and
pharmacological properties relative to earlier compounds.
The first potent synthetic FXR agonist, GW-4064,
established an isoxazole-containing prototype.18 Many
subsequent patents and publications sought to address the
liabilities of GW4064, such as limited bioavailability and
photolability.19–25 Taken together, several general features
have emerged in isoxazole-type FXR agonists that are
necessary for maintaining activity (see Figure 1). The
isoxazole core is preserved in all journal publications. In
the patent literature, however, isoxazole replacements
such as 1,2,3 triazoles or pyrazoles with similar
substitution patterns and similar FXR potency have been
described.26 The 2,6-dichlorophenyl substituent of the
GW4064 template is generally conserved as the most
potent motif at this position, but alkylene spacers between
this and the isoxazole ring can be introduced without
significant loss in FXR activity (e.g. 2, Figure 1).21 Another
approach to improve potency and reduce lipophilicity in
this region has been to incorporate dichloropyridines and -
pyridine N-oxides (e.g. 3, Figure 1) as phenyl
replacements, resulting in analogs with comparable
potency and improved permeability.20 However, in vivo
data were not provided. In contrast to the isopropyl moiety
at the 5-position of the isoxazole ring, which can be
replaced only by cyclopropyl but no other small alkyl
groups without loss in potency,23 there is wide tolerability
for structural variation in linking the oxymethylene
substituent at the 4-position of the isoxazole core to the
terminal acidic entity that seems to be a mandatory
hallmark of potent FXR agonists of this type (Figure 1, 4-6
as examples).21–26 A comprehensive review of isoxazole-
type FXR agonists has been published.31
The structures of the newly identified FXR agonists are
shown in Table 1. The syntheses of analogs containing
cyclopropyl as a linker element are depicted in Schemes
1–6. Racemic 7 and 8 were prepared in four steps.
Alkylation of 2-chloro-4-hydroxy-benzaldehyde with chloro-
methyl-isoxazole building block 7e was followed by a
trans-selective
Horner-Wadsworth-Emmons
(HWE)
reaction with phosphonates 7c or 8c to furnish stilbenes
7b’ and 8b’. Cyclopropanation was accomplished by
reaction with diazomethane under palladium catalysis,
whereby the trans geometry of the two phenyl substituents
was maintained. Derivatives 13, 15 and 16 were prepared
from 7 by standard procedures (see Scheme 1). Eutomer
13b was prepared from 7b by the same procedure as 13
was prepared from 7. The enantiomers of 7a’/8a’ were
separated by chiral HPLC prior to ester hydrolysis
(Scheme 2).32
Pyridine analog 9 was introduced via the 2-cyano-
pyridyl phosphonate 9d, following a similar route as for 7/8,
leading to 9a, which was either hydrolysed to carboxylic
acid 9 or transformed into tetrazole 14 (Scheme 3).
Pyridine analog 10 was prepared first by a HWE reaction
but with inverted functionalities (Scheme 4), followed by
phenol
deprotection,
alkylation
with
7e,
then
cyclopropanation and final ester hydrolysis.

-
3
-
selective ester hydrolysis and reduction (Schemes 5,6).
The eutomer of 11, 11b was prepared by chiral HPLC
separation of the racemic methyl ester of 11 and final ester
hydrolysis, analogous to Scheme 2. The indazole analog
12 was prepared by a Heck reaction between vinyl building
block 12b and bromo-indazole 12a, with the resultant
olefin being transformed into 12 by methods analogous to
those for 11.
The syntheses of the FXR agonists containing
cyclobutyl or azetidinyl as linker elements are depicted in
Schemes 7–11. Cyclobutanone intermediates 18c/19c
were generated by cycloaddition with ketene in modest
yields. Selective lithiation of bromo-chloro building block
17a and addition to 18c/19c at low temperature afforded
hydroxycyclobutyl intermediates 18b/19b in which the aryl
substituents were trans. Palladium-catalysed cyanation
and hydrolysis afforded analogs 18 and 19. Further
standard transformations were applied to obtain
derivatives 20 to 22 from 18 or 18a (Scheme 7). The des-
hydroxycyclobutyl analog 17 was prepared by reduction of
17b with sodium borohydride/TFA, affording 17c as a
mixture in which the cis isomer predominated (ca. 8:1).
After phenol deprotection and cyanation the isomers were
separated by preparative HPLC and then transformed into
analogs 17cis and 17trans via alkylation with 7e and final
hydrolysis (Scheme 8).
Scheme 1: Reagents and conditions: (a) SOCl2, benzotriazole, DCM, 0–25°C,
1.5 h, 80%; (b) K2CO3, DMF, 60°C, O/N, 90%; (c) 7c/8c, NaH, THF, 0°C, 30
min, then 7d, 0°C–rt, 3 h, 60–77%; (d) CH2N2, Pd(OAc)2, Et2O, -50-0°C, 4–8
h, 55–60%; (e) LiOH, THF, H2O (5:1 v/v), 50°C, O/N; (f) for 13: (COCl)2, DCM,
cat. DMF, 0°C, then MeSO2NH2, DIEA, DMAP, 9%; for 15: (I) HATU, DMF,
0°C, 30 min, then NH2CH2CO2Me, DIEA; (II) LiOH, THF, H2O (5:1, v/v), rt,
O/N, 28%; for 16: HATU, DMF, 0°C, 30 min, then NH2(CH2)2SO3H, rt, 18 h,
42%.
Scheme 2: Reagents and conditions: (a) prep. chiral HPLC, Chiralpak ODH,
hexane/IPA; (b) LiOH, THF, H2O (5:1 v/v), 50°C, O/N.
Scheme 7: Reagents and conditions: (a) DMF, Cl(CH2)2Cl, (TfO)2O, 0°C, 1 h,
then bromo-vinylbenzene, 2,4,6-collidine, reflux, O/N, 27%; (b) 17a, n-BuLi,
THF, -78°C, 1 h, then 18c or 19c, -78°C, 1 h, 37–39%; (c) DMF, Zn(CN)2,
Pd2(dba)3, Xantphos, 115°C, 10 h, microwave, 42%; (d) NaOH, EtOH, reflux,
O/N, 56–90%; (e) NaH, DMF, MeI for 20, EtI for 21, Br(CH2)2OTHP for 22,
50°C, O/N, 41%; (f) CH2N2 (2M in Et2O), THF, 0°C–rt, 1 h, 98%; (g) NaOH,
MeOH, reflux, O/N, 26–50%, (h) for 22: AcOH, THF, 40°C, 12 h, 10%.
Scheme 3: Reagents and conditions: (a) 9d, NaH, THF, 0°C, 30 min, then 9c ,
0°C–rt, 3 h, 37%; (b) CH2N2, Pd(OAc)2, Et2O, -50–0°C, 77%; (c) NaH, DMF,
0°C, 1 h, then 7e, rt, O/N, 44%; (d) for 9: KOH, EtOH/H2O 3:1 v/v, reflux, 4 h,
29%; for 14: NaN3, NH4Cl, DMF, 100°C, O/N, 12%.
Scheme 4: Reagents and conditions: (a) 10c, NaH, THF, 0°C, 30 min, then
10b , 0°–rt, 1 h, 54%; (b) BBr3, DCM, 70°C–rt, 1 h; (c) K2CO3, DMF, 60°C,
O/N, 27% over two steps; (d) CH2N2, Pd(OAc)2, Et2O, -50–0°C, 4 h, 37%.
Scheme 8: Reagents and conditions: (a) NaBH4, TFA, rt, O/N, 65%; (b)
Zn(CN)2, Pd(PPh3)4, DMF/H2O (10:1 v/v), 90°C, O/N, 76%; (c) BBr3, DCM, -
78°C, cis/trans isomer separation by HPLC, 13% trans isomer; (d) 7e, K2CO3,
NaI, DMF, 60°C, O/N, 82%; (e) NaOH, EtOH/H2O, reflux, O/N, 50%.
The synthesis of analog 23 started with 3-
methylenecyclobutanecarbonitrile (Scheme 9). After
reaction with methyl magnesium bromide the resultant
ketone underwent a Claisen condensation with diethyl
oxalate to afford diketone 23c. A cyclisation with hydrazine
hydrochloride in ethanol afforded the desired pyrazole
regioisomer 23b, and olefin oxidation then provided the
cyclobutanone intermediate 23a in good yield.
Organometallic coupling of 17a with 23a was best
achieved by preparation of the Grignard reagent from 17a
with iPrMgCl•LiCl, to which 23a was added.
Scheme 5: Reagents and conditions: (a) 11c, NaH, DMF, 0°C, 30 min, then 2-
iodopropane in DMF, 0°C–rt, 2.5 h, 18%; (b) KOH, MeOH, 20°C, 24 h, 62%;
(c) BH3•Me2S, THF, 0°C, rt, O/N, 70%; (d) (COCl)2, DMSO, DCM, -30°C, 20
min, then added product of (c), 1 h, NEt3, O/N, 69%.
Scheme 6: Reagents and conditions: (a) (I). NaBH4, EtOH, 3 h, rt; (II). PBr3,
Et2O, 30 min, 0°C; (III). P(OEt)3, 175°C, 3 h; (b) 11b‘‘, NaH, THF, 0°C, 30 min,
then 11a‘, 0°–rt, 3 h, 33%; (c) CH2N2, Pd(OAc)2, Et2O, -50–0°C, 4 h, 65–70%;
(d) LiOH, THF, H2O (5:1 v/v), 50°C, O/N, 17–48%; (e) n-BuLi, THF,
MeP(Ph)3+Br-, -60°C, 4 h, 7c, -60°C–rt; (f) 12b, 12a, tri(orthotoloyl)phosphine,
TEA, DMF, Pd2(dba)3, 100°C, 16 h, 23% over two steps.
The pyrazole containing analog 11 was prepared
following the same approach from aldehyde 11a’, which
was prepared from the pyrazole diester 11c by alkylation,

-
4
-
Scheme 9: Reagents and conditions: (a) MeMgBr, Et2O, 0°C–rt; (b) NaOEt,
EtOH, 15 min, then (CO2Et)2 added, 67°C, 4.5 h, rt, O/N, 56%, over two steps;
(c) EtOH, N2H4-HCl, rt, 3 h; (d) NaIO4, RuCl3 x H2O, EtOH/H2O (77:13, v/v), rt,
45 min, Na2S2O3, 78% over two steps; (e) iPrMgCl•LiCl, THF, 17a, 0°C–rt, 4 h,
then 23a, -10°C–rt, 90 min, 61%; (f) THF, MeOH, H2O, LiOH, rt, 4.5 h, 92%.
Scheme 11: Reagents and conditions: (a) for 25b: 3-azetidine-3-ol HCl-salt,
Cs2CO3, CuI, L-proline, DMSO, 90°C, 18 h, 66%, for 26b: 3-azetidine-3-ol,
Cs2CO3, BINAP, Pd(OAc)2, dioxane, 85 °C, 12 h, 13%; (b) Dess-Martin
periodinane, DCM, rt, 2 h (57–75%); (c) 17a, n-BuLi, THF, -78 °C, 1 h, then
25a or 26a, -78 °C, 1 h (32–37%); (d) LiOH, THF, H2O, 50°C, 12 h (23–58%).
Scheme 10: Reagents and conditions: (a) tributyl(vinyl)tin, Pd(Ph3)4, DMF,
90°C, 4 h, 88%; (b) DMF, Cl(CH2)2Cl, (TfO)2O, 0°C, 1 h, then 24b, 2,4,6-
collidine, reflux, O/N, 40%; (c) 17a, n-BuLi, THF, -78°C, 1 h, then 18c or 19c, -
78°C, 1 h, 40%; (d) THF, MeOH, H2O, LiOH x H2O, rt, 4.5 h, 45%.
The FXR agonistic activities of these analogs are
shown in Table 1. The replacement of the stilbene olefin in
GW4064 with a cyclopropyl linker element provided a
series (7-8b) whose potency was dependent upon both the
absolute stereochemistry of the cyclopropyl substituents,
and upon the point of attachment of the carboxylic acid.
Despite high lipophilicity (see Table 2), compounds 7 and
8 demonstrated pharmacokinetic parameters in the mouse
which were suitable to investigate their in vivo
pharmacology (see Table 3).
This procedure resulted in a higher yield compared to the
addition of lithiated 17a and could be run at room
temperature (Scheme 9). Analog 24 was prepared
following the same route as for 18/19 (Scheme 10).
Hydroxyazetidine analogs 25 and 26 were prepared as
depicted in Scheme 11, exploiting Cu- or Pd-catalysed
coupling reactions of aromatic halides 25b/26b with
azetidin-3-ol. After oxidation to the azetidinones 25a/26a
the syntheses followed the same route as for 18/19.
Table 1
Overview of new isoxazole-type FXR agonists.
FRET
EC50 [nM]
Eff
(% Ref.)
M1H
EC50 [nM]
Eff
(% Ref.)
Cpd
1
R1
L
Structure see Figure 1
20
100
100
35
76
100
85
7
4-carboxy-phenyl
131
7a
4-carboxy-phenyl
1094
80
100
99
186
47
71
89
, (+)-rotamer
, (-)-rotamer
7b
4-carboxy-phenyl
8
3-carboxy-phenyl
3-carboxy-phenyl
32
102
101
34
90
87
88
8a
219
, (+)-rotamer
, (-)-rotamer
8b
9
3-carboxy-phenyl
18
66
100
101
29
57
98
68
10
11
10
7
103
102
14
6
73
81

-
5
-
11b
12
4
101
104
4
79
84
, (-)-rotamer
13
14
13
90
100
44
88
13b
14
48
94
44
81
79
, (-)-rotamer
131
101
111
15
16
150
76
100
98
209
58
34
1372
17 trans
17 cis
3-carboxy-phenyl
3-carboxy-phenyl
24
77
100
99
15
92
89
88
18
19
20
21
22
3-carboxy-phenyl
4-carboxy-phenyl
3-carboxy-phenyl
3-carboxy-phenyl
3-carboxy-phenyl
29
100
105
93
25
79
71
75
89
68
24
63
79
102
220
377
277
373
94
97
23
24
9
100
99
110
50
78
82
19
25
26
3-carboxy-phenyl
54
7
97
86
94
71
102
552
FRET: Biochemical ligand dependent Nuclear Receptor-Cofactor peptide interaction assay, using the biotinylated SRC-1 peptide b-CPSSHSSLTERHKILHRLLQEGSPS-
COOH (0.4 µM) and purified FXRaa187–472-LBD fused to GST (2.5 ng) together with 200 ng streptavidine-allophycocyanine and 6 ng europium labeled-anti-GST as
reagents in 25 µL assay buffer (20 mM Tris/HCl at pH 7.5; 5 mM MgCl2; 60 mM KCl; 1 mM DTT; 0.9 g/L BSA). FRET values are given in nM and Eff (%) is the maximum
efficacy of the compound relative to GW4064 are means of at least 2 assays.
M1H: Mammalian one hybrid assay; A cellular transient transfection assay were the human FXRaa187–472-LBD is fused C-terminally to a Gal4 DNA-binding domain under
transcriptional control of the CMV promoter in pCMV-BD (Stratagene). This chimeric plasmid construct is co-transfected into HEK293 cells together with pFRluc
(Strategene) encoding a Gal4 promoter driven firefly luciferase. Cells were treated with serial dilutions of the test compound. EC50 values were calculated from at least
three experiments.
Eff: % Efficacy is the extrapolated maximum signal generated by the test compound in a dose response dilution. Efficacy of GW4064 in the respective assay is set as
100%.
* Absolute configuration not known.

-
6
-
A first SAR exploration of the left hand side aromatic
moiety identified several carboxy-bearing heteroaromatic
analogs with further improved potencies such as 10, 11,
and 12. The greater polarity of these heteroaromatic
moieties is reflected in a significant increase in aqueous
solubility (measured for 10 and 11, see Table 2) and an
increase in membrane permeability (accompanied by a
decrease in PAMPA membrane accumulation). Changing
the position of the carboxy group from 3 (rel. to the
cyclopropyl, 10) to 4 (9) resulted in a drop in potency,
highlighting the importance of the exact position of the
acidic moiety on the terminal aromatic ring. A similar trend
can be seen when comparing the corresponding phenyl
analogs 7 and 8.
Accordingly, FXR LBD cocrystal structures with similar
ligands (GW4064, GW8062 from Ref. 29) were used as a
template for docking studies with 17trans and 18 (which
differ only in the presence of the hydroxyl group) to search
for potential novel interactions. These experiments
generated structures in which conformational changes
revealed a beneficial H-Bond interaction with the side
chain of Serine329 (see Figure 2). An interaction with this
serine residue has previously been described for
a
structurally unrelated class of FXR agonists as well as for
bile acid derivatives.30
The carboxylic acid moiety can be replaced by the
acyl-sulfonamide (13) or tetrazole (14) isosteres without
major loss of in vitro activity, opening up possibilities for
minimizing phase II metabolism in vivo - indeed, 13 is
slightly more potent than the corresponding carboxylic acid
7.
When comparing the racemic mixtures 7, 8, 11 and 13
with their corresponding eutomers 7b, 8b, 11b and 13b a
similar increase in potency in the FRET assay is observed
(1.64 to 1.88 fold). Although the absolute configuration of
the eutomers have not been elucidated for 7b, 11b and
13b they exhibit the same sense of optical rotation,
pointing towards a same absolute configuration of the
cyclopropyl linker. (7b and 13b have the same absolute
configuration since 13b is prepared from 7b, see Scheme
1). Together, these findings indicate that the difference in
activity between distomers and eutomers is mainly caused
by a more favourable binding of the cyclopropyl linker in
the eutomer series ((-)-rotamers).
The glycine and taurine conjugates of 7 (15 and 16),
both of which have been detected in vivo in mice (data not
shown; see also ref 23), display similar potency compared
to the parent in the FRET assay. The activity of 16 in the
cellular assay is significantly reduced, likely due to lower
cellular permeability as a result of the strongly acidic
sulfonic acid. This effect is similar to that observed with
natural bile acids and their taurine conjugates.1,2
Figure 2
Ball and stick representation of compound 18 docked into the LBD of human
FXR. The hydroxyl substituent at the cyclobutyl moiety of 18 points to the
hydroxyl functionality of Ser329 with a calculated distance of 2.6 Å, suggesting
a H-bond interaction. For docking experiments we used ICM 3.8-4a modeling
software (MolSoft LLC, San Diego, CA).35 The complex of FXR with GW4064
(PDB:3DCT) was used as reference structure. During the docking simulation
standard parameters were applied and Ser329 were allowed to be flexible.
Building on these encouraging results, compounds 25
and 26 were synthesized as hydroxyazetidinyl analogs of
8/18 and 10 in which stereoisomers are absent. While 25
has a slightly reduced activity in the FRET and M1H assay
compared to 18, compound 26 is equipotent with 10 in the
FRET assay (7 vs. 10 nM EC50), although it shows
reduced activity in the cellular assay.
The calculated physicochemical parameter clogD of a
subset of these analogs is provided in Table 2. A gradual
increase in hydrophilicity from compound 7 to compound
26 can be observed, with introductions of heteroaromatic
carboxylic acid moieties, the hydroxy-cyclobutyl linker and
the hydroxy-azetidinyl linker representing the major
increments. The impact on solubility of the linker
modifications is comparable to that of the benzoic acid
replacements (compare 8 to 10 or 18). A further beneficial
effect for compounds that bear both linker and benzoic
acid replacements simultaneously can be seen in the
PAMPA assay: compounds 23 and 26 demonstrate the
highest membrane flux and the lowest membrane
accumulation in the set. The low membrane accumulation
of only 3 fold for compound 26 is remarkable, given the
general tendency of these types of amphiphilic compounds
to insert into artificial bilayer membranes.
Interestingly, the PAMPA membrane accumulation,
together with the clogD seem to correlate with the
observed shift from the biochemical to the cellular FXR
activity: Compounds 7, which has a high clogD (5.2) and a
high membrane accumulation (30 fold) is more active in
the M1H assay than in the FRET assay, while compound
26 with the lowest clogD (3.2) and the lowest membrane
accumulation (3 fold) is two orders of magnitude less
active in the cellular assay than in the biochemical assay.
There is no standard model which would explain why more
lipophilic and membrane-accumulating compounds tend to
be more potent at FXR compared to more hydrophilic
ones. However, a recent publication provides a hypothesis
to explain this phenomenon.34
Since the cyclopropyl linker element is chiral in its
nature and no reliable enantioselective synthesis to access
the single enantiomers in a cost efficient manner was
available, we considered achiral alternatives. The
introduction of a 1,4-disubstituted cyclobutyl linker led to
compounds 17trans and 17cis. Both compounds
displayed encouraging biochemical and cellular FXR
potency, with the trans isomer being 3-fold more potent in
the FRET assay and 6-fold more potent in the M1H assay.
Compound 18 was synthesized opportunistically from the
hydroxylated cyclobutyl intermediates generated in the
synthesis of 17 and was also tested for activity.
Surprisingly, it showed a very similar potency compared to
17trans, despite the substantial polarity increase that is
reflected in improved aqueous solubility. The presence of a
hydroxyl substituent in this area of isoxazole-type FXR
ligands is unprecedented. The suitability of the newly
identified hydroxybutyl moiety as a potent and polar achiral
linker element was further corroborated by the synthesis of
direct homologs to the cyclopropyl examples, yielding the
additional pairs 7/19, 11/23, and 12/24. In all cases, the
hydroxybutyl analogs retained FXR activity with increased
aqueous solubility, improved membrane permeability and
decreased PAMPA membrane accumulation.
The similar FXR activity of these pairs of ligands is
suggestive of a specific interaction of the hydroxyl group
with the FXR ligand binding domain (LBD), so as to
compensate for the energy penalty of desolvation of the
hydroxyl group upon binding. This hypothesis is further
supported by a small set of ether derivatives (20–22):
removal of the H-bond donor in the methyl ether 20
reduces FXR activity by approximately 3- to 4-fold, and
ethyl ether 21 and hydroxyethyl ether 22 show even further
reduced activity.
The two most polar analogs 23 and 26 were tested on
a panel of 20 nuclear receptors, including PXR and
showed no activity up to 5 µM.
Table 2

-
7
-
Physicochemical properties and PAMPA assay data for
selected FXR agonists.
A
aq.
solubility
[µM]
PAMPA
[fold memb.
enrich.]**
PAMPA
[%flux]*
Cpd
clogD
7
8
10
11
17trans
18
19
23
5.2
5.1
3.8
4.2
5.6
4.4
4.4
3.5
4.4
3.2
72
20
14
17
n.d.
29
0
24
21
46
27
35
30
33
n.d.
22
319
27
9
11
40
3
165
158
n.d.
192
192
171
n.d.
197
25
26
* = [Cacceptor/(Cdonor + Cacceptor)] x 100 x 2, after 16 h at rt; ** = fold enrichment
in the membrane compartment is determined by Cmembrane * 2 / (Cacceptor+Cdonor
with Cmembrane Cacceptor and Cdonor representing the concentrations of this
B
)
,
compound in the membrane phase (extracted into a volume equal to acceptor
or donor compartment) or the acceptor or donor compartment, respectively.
The pharmacokinetic profiles of a subset of analogs in
mice were also assessed (Table 3). Overall there was a
clear general trend that more hydrophilic compounds had
an increased plasma exposure, a reduced clearance and a
reduced volume of distribution (compare 8 with 18, 26; 11
with 23). In addition, the position of the carboxylic acid on
the aromatic ring had a dramatic impact on the plasma
exposure (compare 18 with 19). Analogs with the acid in
the 3-position relative to the linker element exhibited
improved properties compared to their 4-position isomers.
Notably, the difference was greater in the more polar
hydroxycyclobutyl series compared to the cyclopropyl
series (compare 7 with 8 and 18 with 19).
Table 3
In vivo pharmacokinetic properties of selected FXR
agonists in C57BL/6J mice.
Cl/F
AUC
[hr*ng/
mL]
Vd/F
[L/
Cmax
[ng/
mL]
[mL/
min/
kg]
T1/2
[h]
C
Cpd.
F%
kg]
7
8
430
800
110
8600
180
960
3.6
4.7
5.5
0.3
3.9
0,8
280
1060
160
85
0.5
1.0
0.8
0.6
1.1
0.6
44
53
11
49
41
18
55
83
11
18
19
23
9800
96
4.7
190
16
1470
Dose: 5mg/kg p.o., 1 mg/kg i.v. All parameters except F were calculated from
the p.o. arm.
To assess pharmacodynamic properties, compounds
7, 8, 11, 18, and 23 were administered at a dose of 10
mg/kg for 10 days to C57BL/6J mice pre-fed and
maintained on a high fat diet with 1% (w/w) cholesterol.
Despite the differences in PK profiles, each of these
showed significant improvements in plasma lipid profiles
(Figure 3A). Compounds 7 and 11 exhibited the highest
cholesterol reduction, followed by compounds 18 and 13.
Compound 8b turned out to be less active. Liver levels of
Figure 3
Changes in total plasma cholesterol and total plasma triglycerides upon 10
days of 10 mg/kg p.o. administration of the indicated compounds into
the compounds
4 hours after the final dose were
C57BL/6J mice prefed and maintained on
a high fat diet (60% kcal fat)
assessed, but there were no obvious relationships
between plasma or liver exposure and the lipid lowering
effects observed (Figure 3B and 3C).
enriched with 1% (w/w) cholesterol. Shown are the % changes to the vehicle
control group (3A). Liver levels (3B) and liver/plasma ratios (3C) of the
indicated compounds (parent only) upon final sacrification of the animals, 4 h
after the final gavage (* = statistical significance ranges; *p<0.05, **p<0.01,
***p<0.001).
In summary, the challenge of improving polarity,
reducing
amphiphilicity
and
thereby
improving
pharmacokinetic properties while maintaining potency of
GW4064-type isoxazole FXR agonists was successfully
addressed by introducing a hydroxyl-bearing 4-membered
ring (either cyclobutyl or azetidinyl) as a linker between the
middle and the terminal aryl rings. The hydroxyl group may
engage in an H-bond interaction with Ser329 of the FXR
LBD. This is an unprecedented interaction within the series
of isoxazole FXR agonists and might be further explored to

-
8
-
generate improved analogs within this class of
compounds.
818119),
see
also:
http://clinicaltrial.gov/ct2/show/NCT01265498
The data provided here show that there is no direct
correlation between physicochemical and in vivo
pharmacokinetic and pharmacodynamic properties.
However, the more polar isoxazole-type FXR agonists are
clearly improved in terms of general drug-likeness
(aqueous solubility, membrane permeability), which will
likely facilitate pharmaceutical development. These
analogs exhibit in vivo efficacies that are similar to those of
the cyclopropyl linker compounds and other similarly
lipophilic FXR agonists.23
18. Maloney, P. R.; Parks, D. J.; Haffner, C. D.; Fivush, A. M.;
Chandra, G.; Plunket, K. D.; Creech, K. L.; Moore, L. B.;
Wilson, J. G.; Lewis, M. C.; Jones, S. A.; Willson, T. M. J.
Med. Chem. 2000, 43, 2971.
19. Akwabi-Ameyaw, A.; Bass, J. Y.; Caldwell, R. D.;
Caravella, J. A.; Chen, L.; Creech, K. L.; Deaton, D. N.;
Jones, S. A.; Kaldor, I.; Liu, Y.; Madauss, K. P.; Marr, H.
B.; McFadyen, R. B.; Miller, A. B.; Iii, F. N.; Parks, D. J.;
Spearing, P. K.; Todd, D.; Williams, S. P.; Wisely, G. B.
Bioorg. Med. Chem. Lett. 2008, 18, 4339.
Clearly, further studies are required that elucidate the
mechanistic basis for the observed ADME properties and
their relationships to the pharmacodynamic properties of
this class of novel FXR agonists.
20. Feng, S.; Yang, M.; Zhang, Z.; Wang, Z.; Hong, D.;
Richter, H.; Benson, G. M.; Bleicher, K.; Grether, U.;
Martin, R. E.; Plancher, J. M.; Kuhn, B.; Rudolph, M. G.;
Chen, L. Bioorg. Med. Chem. Lett. 2009, 19, 2595.
21. Bass, J. Y.; Caldwell, R. D.; Caravella, J. A.; Chen, L.;
Creech, K. L.; Deaton, D. N.; Madauss, K. P.; Marr, H. B.;
McFadyen, R. B.; Miller, A. B.; Parks, D. J.; Todd, D.;
Williams, S. P.; Wisely, G. B. Bioorg. Med. Chem. Lett.
2009, 19, 2969.
Acknowledgements
We would like to thank Dr. William Watkins from
Gilead Sciences for critical reading and editing of the
manuscript.
22. Akwabi-Ameyaw, A.; Bass, J. Y.; Caldwell, R. D.;
Caravella, J. A.; Chen, L.; Creech, K. L.; Deaton, D. N.;
Madauss, K. P.; Marr, H. B.; McFadyen, R. B.; Miller, A. B.;
Navas, F. 3rd; Parks, D. J.; Spearing, P. K.; Todd, D.;
Williams, S. P.; Bruce Wisely, G. Bioorg. Med. Chem. Lett.
2009, 19, 4733.
23. Abel, U.; Schlüter, T.; Schulz, A.; Hambruch, E.; Steeneck,
C.; Hornberger, M.; Hoffmann, T.; Perović-Ottstadt, S.;
Kinzel, O.; Burnet, M.; Deuschle, U.; Kremoser, C. Bioorg.
Med. Chem. Lett. 2010, 20, 4911.
24. Bass, J.Y.; Caravella, J. A.; Chen, L.; Creech, K. L.;
Deaton, D. N.; Madauss, K. P.; Marr, H. B.; McFadyen, R.
B.; Miller, A. B.; Mills, W. Y.; Navas, F. 3rd; Parks, D. J.;
Smalley, T. L. Jr; Spearing, P. K.; Todd, D.; Williams, S.
P.; Wisely, G. B. Bioorg. Med. Chem. Lett. 2011, 21, 1206.
25. Akwabi-Ameyaw, A.; Caravella, J. A.; Chen, L.; Creech, K.
L.; Deaton, D. N.; Madauss, K. P.; Marr, H. B.; Miller, A. B.;
Navas, F. 3rd; Parks, D. J.; Spearing, P. K.; Todd, D.;
Williams, S. P.; Wisely, G. B. Bioorg. Med. Chem. Lett.
2011, 21, 6154.
References and notes
1. Parks, D. J.; Blanchard, S. G.; Bledsoe, R. K.; Chandra,
G.; Consler, T. G.; Kliewer, S. A.; Stimmel, J. B.; Willson,
T. M.; Zavacki, A. M.; Moore, D. D.; Lehmann, J. M.
Science 1999, 284, 1365.
2. Makishima, M.; Okamoto, A. Y.; Repa, J. J.; Tu, H.;
Learned, R. M.; Luk, A.; Hull, M. V.; Lustig, K. D.;
Mangelsdorf, D. J.; Shan, B. Science 1999, 284, 1362.
3. Forman, B. M.; Goode, E.; Chen, J.; Oro, A. E.; Bradley, D.
J.; Perlmann, T.; Noonan, D. J.; Burka, L. T.; McMorris, T.;
Lamph, W. W.; Evans, R. M.; Weinberger, C. Cell 1995,
81, 687.
4. Higashiyama, H.; Kinoshita, M.; Asano, S. Acta Histochem.
2008, 110, 86.
5. Bookout, A. L.; Jeong, Y.; Downes, M.; Yu, R.; Evans, R.
M.;
Mangelsdorf,
D.
J.
www.nursa.org/10.1621/datasets.02001
6. Kalaany, N. Y.; Mangelsdorf, D. J. Annu. Rev. Physiol.
2006, 68, 159.
7. Calkin, A. C.; Tontonoz, P. Nat. Rev. Mol. Cell Biol. 2012,
13, 213.
8. Inagaki, T.; Moschetta, A.; Lee, Y. K.; Peng, L.; Zhao, G.;
Downes, M.; Yu, R. T.; Shelton, J. M.; Richardson, J. A.;
Repa, J. J.; Mangelsdorf, D. J.; Kliewer, S. A. Proc. Natl.
Acad. Sci. U. S. A. 2006, 103, 392.
9. Gadaleta, R. M.; van Erpecum, K. J.; Oldenburg, B.;
Willemsen, E. C.; Renooij, W.; Murzilli, S.; Klomp, L. W.;
Siersema, P. D.; Schipper, M. E.; Danese, S.; Penna, G.;
Laverny, G.; Adorini, L.; Moschetta, A.; van Mil, S. W. Gut
2011, 60, 463.
10. Chen, W. D.; Wang, Y. D.; Meng, Z.; Zhang, L.; Huang, W.
Biochim. Biophys. Acta 2011, 1812, 888.
11. Modica, S.; Gadaleta, R. M.; Moschetta, A. Nucl. Recept.
Signal. 2010, 8, e005.
12. Porez, G.; Prawitt, J.; Gross, B.; Staels, B. J. Lipid. Res.
2012, 53, 1723.
13. Pellicciari, R.; Fiorucci, S.; Camaioni, E.; Clerici, C.;
Costantino, G.; Maloney, P. R.; Morelli, A.; Parks, D. J.;
Willson, T. M. J. Med. Chem. 2002, 45, 3569.
14. Mudaliar, S.; Henry, R. R.; Sanyal, A. J.; Morrow, L.;
Marschall, H. U.; Kipnes, M.; Adorini, L.; Sciacca, C. I.;
Clopton, P.; Castelloe, E.; Dillon, P.; Pruzanski, M.;
Shapiro, D. Gastroenterology 2013, 145, 574.
15. (a) Kowdley, K.; Hirschfield, G.; Chapman, R.; Vincent, C.;
Jones, D.; Pares, A.; Luketic, V.; Gordon, S.; Pencek, R.;
Marmon, T.; Hooshmand-Rad, R. J. Hepatol. 2014, 60,
S192.; (b) Nevens, F., Andreone, P., Mazzella, G.,
Strasser, S., Bowlus, C., Invernizzi, P., Drenth, J.,
Pockros, P., Regula, J., Hansen, B., Hooshmand-Rad, R.,
Sheeron, S., Shapiro, D. J. Hepatol. 2014, 60, S525.
16. Silveira, M. G.; Lindor, K. D. Expert. Opin. Pharmacother.
2014, 15, 365.
26. International Patent Applications WO/2007/140174,
WO/2007/140174 and WO/2009/012125.
27. International Transporter Consortium: Giacomini, KM,
Huang, SM, Tweedie, DJ, Benet, L. Z., Brouwer, K. L.,
Chu, X., Dahlin, A., Evers, R., Fischer, V., Hillgren, K. M.,
Hoffmaster, K. A., Ishikawa, T., Keppler, D., Kim, R. B.,
Lee, C. A., Niemi, M., Polli, J. W., Sugiyama, Y., Swaan, P.
W., Ware, J. A., Wright, S. H., Yee, S. W., Zamek-
Gliszczynski, M. J., Zhang, L. Nat. Rev. Drug. Discov.
2010, 9, 215.
28. Emami Riedmaier, A.; Nies, A. T.; Schaeffeler, E.;
Schwab, M. Pharmacol. Rev. 2012, 64, 421.
29. Mi, L. Z.; Devarakonda, S.; Harp, J. M.; Han, Q.; Pellicciari,
R.; Willson, T. M.; Khorasanizadeh, S.; Rastinejad, F. Mol.
Cell. 2003, 11, 1093 and structures 3DCT and 3DCU from
the
Protein
Data
Bank
(pdb)
http://www.rcsb.org/pdb/home/home.do
30. a) Richter, H. G.; Benson, G. M.; Bleicher, K. H.; Blum, D.;
Chaput, E.; Clemann, N.; Feng, S.; Gardes, C.; Grether,
U.; Hartman, P.; Kuhn, B.; Martin, R. E.; Plancher, J. M.;
Rudolph, M. G.; Schuler, F.; Taylor, S. Bioorg. Med. Chem.
Lett. 2011, 21, 1134. b) D’Amore, C.; Di Leva, F. S.; Sepe,
V.; Renga, B.; Del Gaudio, C.; D’Auria, M. V.; Zampella,
A.; Fiorucci, S.; Limongello, V., J. Med. Chem 2014, 57,
937.
31. Gege, C.; Kinzel, O.; Steeneck, C.; Schulz, A.; Kremoser,
C. Curr. Top. Med. Chem. 2014, 19, 2143.
32. The absolute configuration of 8a was determined by
crystallization and structural elucidation by X-ray of the
diastereomeric salt (S)-1-(4-bromophenyl)ethan-1-aminium
3-(2-(2-chloro-4-methoxyphenyl)cyclopropyl)benzoate,
single enantiomer, which was obtained from 8a by
chemical transformations.
33. Bass, J. Y.; Caravella, J. A..; Chen, L.; Creech, K. L.;
Deaton, D. N.; Madauss, K. P.; Marr, H. B.; McFadyen, R.
B.; Miller, A. B.; Mills, W. Y.; Navas III, F.; Parks, D. J.;
17. Press Release Intercept Pharmaceuticals and NIDDK
(http://ir.interceptpharma.com/releasedetail.cfm?releaseid=

-
9
-
35. Abagyan, R.; Totrov, M.; Kuznetsov, D., J. Computational
Chem. 1994, 15, 488.
Smalley Jr., T. L.; Spearing, P. K.; Todd, D.; Williams, S.
P.; Wisely, G. B. Bioorg. Med. Chem. Lett. 2011, 21, 1206.
34. Hambruch, E.; Kinzel, O., Kremoser, C., Nucl. Rec. Res.
2016, article in press

- 10 -