Novel FXR (farnesoid X receptor) modulators: Potential therapies for cholesterol gallstone disease

Article abstract of DOI:10.1016/j.bmc.2016.06.039

Metabolic disorders such as diabetes are known risk factors for developing cholesterol gallstone disease (CGD). Cholesterol gallstone disease is one of the most prevalent digestive diseases, leading to considerable financial and social burden worldwide. Ursodeoxycholic acid (UDCA) is the only bile acid drug approved by FDA for the non-surgical treatment of gallstones. However, the molecular link between UDCA and CGD is unclear. Previous data suggest that the farnesoid X receptor (FXR), a bile acid nuclear receptor, may protect against the development of CGD. In studies aimed at identifying the role of FXR, we recently identify a novel chemical tool, 6EUDCA (6-αethyl-ursodeoxycholic acid), a synthetic derivative of UDCA, for studying FXR. We found that 6EUDCA binds FXR stronger than UDCA in a TR-FRET binding assay. This result was supported by computational docking models that suggest 6EUDCA forms a more extensive hydrogen bound network with FXR. Interestingly, neither compound could activate FXR target genes in human nor mouse liver cells, suggesting UDCA and 6EUDCA activate non-genomic signals in an FXR-dependent manner. Overall these studies may lead to the identification of a novel mechanism by which bile acids regulate cell function, and 6EUDCA may be an effective targeted CGD therapeutic.

Full text of DOI:10.1016/j.bmc.2016.06.039

Accepted Manuscript
Novel FXR (Farnesoid X receptor) Modulators: Potential Therapies for Cho-
lesterol Gallstone Disease
Donna D. Yu, Sreenath S. Andrali, Hongzhi Li, Min Lin, Wendong Huang,
Barry M. Forman
PII:
S0968-0896(16)30462-X
http://dx.doi.org/10.1016/j.bmc.2016.06.039
BMC 13095
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
5 April 2016
16 June 2016
18 June 2016
Please cite this article as: Yu, D.D., Andrali, S.S., Li, H., Lin, M., Huang, W., Forman, B.M., Novel FXR (Farnesoid
X receptor) Modulators: Potential Therapies for Cholesterol Gallstone Disease, Bioorganic & Medicinal
Chemistry (2016), doi: http://dx.doi.org/10.1016/j.bmc.2016.06.039
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.

Novel FXR (Farnesoid X receptor) Modulators: Potential Therapies for Cholesterol
Gallstone Disease
Donna D. Yu,* Sreenath S. Andrali,* Hongzhi Li, Min Lin, Wendong Huang, and Barry
M. Forman
Department of Diabetes and Metabolic Diseases Research, Beckman Research Institute,
City of Hope National Medical Center, Duarte, CA 91010, USA
ABSTRACT:
Metabolic disorders such as diabetes are known risk factors for developing cholesterol
gallstone disease (CGD). Cholesterol gallstone disease is one of the most prevalent
digestive diseases, leading to considerable financial and social burden worldwide.
Ursodeoxycholic acid (UDCA) is the only bile acid drug approved by FDA for the non-
surgical treatment of gallstones. However, the molecular link between UDCA and CGD
is unclear. Previous data suggest that the farnesoid X receptor (FXR), a bile acid nuclear
receptor, may protect against the development of CGD. In studies aimed at identifying
the role of FXR, we recently identify a novel chemical tool, 6EUDCA (6-α ethyl-
ursodeoxycholic acid), a synthetic derivative of UDCA, for studying FXR. We found that
6EUDCA binds FXR stronger than UDCA in a TR-FRET binding assay. This result was
supported by computational docking models that suggest 6EUDCA forms a more
extensive hydrogen bound network with FXR. Interestingly, neither compound could
activate FXR target genes in human nor mouse liver cells, suggesting UDCA and
6EUDCA activate non-genomic signals in an FXR-dependent manner. Overall these
1

studies may lead to the identification of a novel mechanism by which bile acids regulate
cell function, and 6EUDCA may be an effective targeted CGD therapeutic.
ABBREVIATIONS USED
FXR, Farnesoid X Receptor; CDCA, chenodeoxycholic acid; UDCA, Ursodeoxycholic
acid; 6EUDCA, 6-αethyl-ursodeoxycholic acid; qPCR, quantitative polymerase chain
reaction; TGR5, G-protein-coupled receptor.
1. Introduction
Metabolic disorders such as diabetes are known high risk factors for developing
cholesterol gallstone disease (CGD). CGD is more prevalent in people who consume
western diets.1 Gallstones are formed in the gallbladder when a super saturation of bile
cholesterol precipitates as crystals. In the presence of mucin, cholesterol crystals nucleate
and grow to form gallstones, which can cause severe abdominal pain and gallbladder
inflammation. Currently, the treatment for gallstones is laproscopic cholecystectomy,
which is one of the most commonly preformed surgical procedures worldwide. However,
cholecystectomy is invasive and can cause surgical complications in terms of morbidity
and mortality, and not all patients with symptomatic gallstones are candidates for
surgery. Earlier studies have provided evidence for dissolution of cholesterol gallstones
through the use of ursodeoxycholic acid (UDCA),2 which is the only bile acid drug that
has been approved by the US FDA for treating liver diseases, including primary biliary
cirrhosis and CGD.3
2

UDCA (Figure 1) is a naturally occurring bile acid, derived from cholesterol. It has been
shown to alter lipid, glucose, and bile acid metabolism.3 Indeed, for centuries, bile acids
have played important roles in elucidating the mechanisms of metabolic disorders.4
However, the mechanism by which UDCA works on cholesterol gallstones is unknown.
There is, therefore, an urgent need for research defining the mechanism of UDCA and
exploring therapies that may have distinct mechanisms of action from past therapies.
Our preliminary and recently published data indicate that the farnesoid X receptor (FXR,
NR1H4) plays a role in protecting against the development of gallstone disases.5
Consequently, ligands that bind and activate FXR may be potential therapeutic agents.
FXR is a bile acid nuclear receptor that is highly expressed in the liver and gut.6 Studies
from several laboratories have demonstrated FXR serves as a key metabolic integrator
that regulates cholesterol, triglyceride, and glucose homeostasis.7 Bile acids are
endogenous signals that can act as FXR ligands 8 and have been shown to reduce
adiposity and plasma triglycerides as well as to promote insulin sensitization.9 The bile
acid chenodeoxycholic acid (CDCA) (Figure 1) is one of the most potent endogenous
agonists for FXR (EC50 ~ 13 M). However, CDCA is poorly effective as a therapeutic
agent because of its toxicity and its promiscuity in activating the intestine expressed G-
protein-coupled receptor TGR5, a G-protein-coupled bile acid receptor that mediates
cellular responses to bile acids.10 In contrast, the hydrophilic bile acid UDCA is a widely
used therapeutic agent that has minimal side effects.11
The available data suggest that UDCA can treat and protect against gallstones, but its
molecular target is unknown. Recent studies indicated that UDCA could be working
3

through a nuclear receptor, the glucocorticoid receptor (GR), but these studies did not
show that UDCA directly binds to GR.11 Others have shown that UDCA binds to FXR
expressed in CHOK1 cells, and UDCA may regulate constitutively expressed IBABP
(ileal bile acid binding protein) in Caco-2 cells. This suggests that UDCA acts as a partial
agonist through an FXR-mediated mechanism.12 Because activation of FXR through its
ligands could control the formation of gallstones,13 it follows that UDCA may have a key
role in treating gallstones with distinct molecular signatures via FXR. However, UDCA is
a weak modulator of FXR and has limited clinical efficacy, which is not an ideal clinical
tool for testing the FXR-mediated mechanism. There is therefore a need to develop more
potent and higher affinity small molecules that can act as targeted therapeutic agents for
the treatment of CGD. Building upon the successful synthesis strategy for 6-ECDCA (an
ethyl at C-6 on the CDCA ring B),14 we designed and synthesized 6-αethyl-
ursodeoxycholic acid (6EUDCA), an α-ethylated derivative of UDCA. We then identified
6EUDCA as a potent, non-genomic FXR activator that binds FXR stronger than UDCA
and shows excellent properties for development as a potential treatment for CGD and
related metabolic disorders.
4

Figure 1. Chemical structures of the naturally-occurring bile acids CDCA and UDCA, as
well as the previously disclosed FXR synthetic agonists, 6EUDCA, GW4064, and
DY246.
2. Results and discussion
2.1. Chemistry
Previously, we developed an improved synthetic route for large scale production of
6ECDCA (6-αethyl-chenodeoxycholic acid) using the cost-effective and readily-available
starting material CDCA.15 We noted the possibility of using an improved strategy for the
synthesis of 6EUDCA from CDCA, which could serve as the key starting material for the
production of kilogram quantities of 6EUDCA. 6EUDCA is a novel derivative of UDCA
whose preparation was reported in a 2011 U.S. patent application.16 However, that
synthesis methodology was complicated, involving multiple steps, and resulted in a low
yield.
5

To convert CDCA to 6EUDCA, we first evaluated the synthetic procedures for UDCA.
The existing methodologies for synthetic UDCA start from the major component of bile
acids, cholic acid (CA).17 In 1954, Kanajawa reported the first synthetic method for the
preparation of UDCA from CA in 7-steps (<9% yield).18 This procedure is still used for
the industrial production of UDCA. Several similar synthetic methods starting from CA,
including the radical reduction in the last step with an alkaline metal (Na or K) in
refluxing alcohol, have been reported recently.19-21 However, it is difficult to control
these vigorous reaction conditions, and the use of concentrated metal bases under high
pressure and temperature can easily lead to explosions. With this in mind, several
modifications have been reported. In 1980, Faba reported a UDCA synthesis that begins
with CA. CA was converted to a methyl ester intermediate and oxidized by Jones reagent
with the final step using Zn-EtOH to reduce 7-keto to get UDCA.22 In 1982, Iida and
Chang reported an improved synthetic method that described the four possible 3, 7-
dihydroxy acids.23 This strategy involved a key step for reducing 12-oxo
tosylhydrazones. In addition, in 1981, Calzda reported a 6-step synthesis for producing
UDCA starting from CA; this method used aluminum alkoxide (Al(OiPr)3/iPrOH) instead
of alkaline metal in the final reduction step.24 However, these procedures for producing
UDCA and its analogs are long, costly, and complicated and have low yields.
To simplify the production of UDCA, we describe here an improved facile synthesis of
UDCA from the readily available starting material CDCA (Figure 2). Key synthetic
intermediate 1 was converted from commercially available CDCA according to our
previously reported synthetic approach.15 In this synthetic step, not surprisingly, partially
regioselective oxidation proved to be very challenging; the desired corresponding
6

intermediate 1 from CDCA with pyridinium chlorochromate (PCC) was obtained with
82% yield. Thus, the regioselective installation of carbonyl functionality in the 7-position
of CDCA was well done. In contrast to the current regioselective oxidation work, earlier
work included a partial oxidation of CDCA,25 a laborious potassium chromate procedure
and a modified synthetic route via CA.26 These methods were complicated and restricted,
clearly suggesting the current regioselective oxidation methodology we improved is
facile to carry out for the large scale production of UDCA.
The only chemical difference between UDCA and CDCA is the orientation of the
hydroxyl group at the C7 position. In UDCA, the hydroxyl is oriented equatorially (7β-
OH), and in CDCA, the hydroxyl is oriented axially (7-OH). To achieve a 7β-OH at the
C7 position, intermediate 1 in Figure 2 was converted to UDCA by Meerwein-Ponndorf-
Verley (MPV) type reduction.27 In particular, by modification of published methods,11
intermediate 1 was reduced stereoselectively by Al-catalytic reduction with aluminum
alkoxide (Al(OiPr)3/iPrOH), giving the expected UDCA with the equatorially oriented
hydroxyl (7β-OH) at an 82% yield that was highly reproducible. Thus, UDCA was
successfully synthesized in 2 steps from CDCA by an improved stereoselective procedure
giving a 60-64 % overall yield.
Prompted by production of UDCA from CDCA, we directed our efforts to the synthesis
of 6EUDCA from CDCA. The synthesis of the protected 3-hydroxyl of 2 started with the
previously reported 6ECDCA preparation.15 A critical step in the synthesis of 6EUDCA
employed selective MPV reduction. In initial experiments, intermediate 2 was treated
with base LDA, followed by deprotection with PPTS.28 This gave compound 3 in a
highly efficient manner. We note that a more efficient way to insert an ethyl in the C6
7

position of CDCA was recently described.29 However, those procedures for producing an
ethyl insertion in the C6 position of CDCA are long and costly, even though they give
higher yields.
It is worth noting that the configuration at C-6 on intermediate 3, the 6α-ethyl derivative,
was the only compound formed, this is consistent with previous research by Pellicciari et
al.6 MPV selective reduction of compound 3 with aluminum alkoxide (Al(OiPr)3/iPrOH)
in the presence of isopropanol led to the formation of 6EUDCA. The stereoselective
reduction on substrate 4 is in agreement with the traditional MPV mechanism, which
typically yields the more thermodynamically stable, equatorial alcohol product.27 Thus,
our 4-step conversion of CDCA to 6EUDCA was achieved with approximately 15%
overall yield (Figure 2).
8

Figure 2. Highly stereoselective synthesis of UDCA and 6EUDCA from CDCA.
The determination of the C7 stereochemistry of 6EUDCA was made by 1H NMR using
CDCl3 as a solvent. 6EUDCA and 6ECDCA are epimers, differing only in the
configuration of the chiral carbon at C7.30 The C3 and C7 methine proton peaks of
6EUDCA are easily identified. H7 in 6EUDCA is a multiplet at about 3.36 ppm (J = 8.18
Hz) in CDCl3 whereas the corresponding signal in 6ECDCA is a singlet at about 3.70
ppm.
2.2. Biological evaluations
2.2.1. Time-resolved fluorescence resonance energy transfer binding assay (TR-
FRET)
Next, we compared the binding affinities of synthesized 6EUDCA and UDCA to FXR
using a time-resolved fluorescence resonance energy transfer binding assay (TR-FRET).
Positive and negative controls are described in Experimental Information. Both UDCA
and 6EUDCA bound to FXR; however, UDCA bound FXR with a very low affinity and
only at high concentrations. In contrast, 6EUDCA bound FXR with higher affinity as
indicated by the ability of low concentrations of ligand to produce signals in the TR-
FRET assay (Figure 3). These data suggest that 6EUDCA may be a better therapeutic
ligand for FXR as compared to UDCA. Additional TR-FRET studies are required to
explore and confirm this suggestion.
9

Figure 3. Binding of UDCA and 6EUDCA to FXR in a TR-FRET assay. (A and B)
UDCA at high concentrations binds to FXR as indicated by the reduction in fluorescence
signal of DY24631 Shown are data from 15 min (A) and 25 min (B) incubations.
6EUDCA binds strongly to FXR at concentrations as low as 10 μM, as indicated by
reduced fluorescence signal of DY246 (C, 15 min incubation and D 25 min incubation).
Apparently, 6EUDCA is more potent in binding and activating FXR and has greatly
increased potency relative to UDCA. DY246 was chosen as a positive control in the TR-
FRET assay because it is a derivative of GW4064, which is a potent FXR agonist.32 In
addition, DY246 is a potent FXR agonist itself, with an EC50 of 550 nM and has
10

successfully been used as a fluorescent probe in a high throughput screening campaign to
identify FXR antagonists.30
2.2.2. Biological effects of UDCA and 6EUDCA on the human liver cell line HepG2
and on primary mouse hepatocytes
Next, we investigated the biological effects of UDCA and 6EUDCA on the human liver
cell line HepG2 and on primary mouse hepatocytes. Initially, we tested whether UDCA
or 6EUDCA could affect FXR in HepG2 cells. Our qPCR data indicate that neither
UDCA nor 6EUDCA could induce expression of the FXR target gene SHP (Figure 4).
GW4064, a known agonist of FXR, was used as a positive control in all our experiments.
Figure 4. QPCR analysis of the FXR target gene SHP in liver cells. Serum starved
HepG2 cells were treated either with DMSO, 10 μM GW4046, 100 μM UDCA, or 10 μM
6EUDCA overnight. Changes in the expression of the SHP gene were analyzed and
normalized with the internal standard 36b4. Data are presented as mean  S.D. ANOVA
and Dunnett’s multiple comparison tests were applied to determine the significance of the
11

differences between DMSO and FXR ligands. Asterisks (***) indicate a significant
difference in SHP expression (p<0.0001) compared to DMSO.
2.2.3. Effect of UDCA and 6EUDCA on AKT and its downstream target GSK3beta
The UDCA result is consistent with previous literature suggesting that UDCA is not an
agonist for FXR. Subsequently, based on previous studies in our laboratory, we tested
the effect of UDCA and 6EUDCA on AKT and its downstream target GSK3beta. Serum
starved HepG2 cells were treated with or without UDCA or 6EUDCA for 10 min.
Proteins in cell extracts were separated by SDS-PAGE, and Western blot analysis was
performed using specific antibodies against phosphorylated AKT and phosphorylated
GSK3beta. Both UDCA and 6EUDCA were effective in inducing AKT phosphorylation
in HepG2 cells (Figure 5).
Figure 5. UDCA and 6EUDCA can induce phosphorylation of AKT in liver cells.
Serum-starved HepG2 cell extracts were prepared after incubation with DMSO,
GW4064, UDCA, or 6EUDCA for 10 min. Extracts were analyzed by Western blot using
specific antibodies against pAKT S473 (left panel) or pGSK3beta (right panel).
12

To further investigate the dose dependence of AKT phosphorylation, we treated HepG2
cells with increasing concentrations of 6EUDCA. Western blot analysis indicated that
changes brought about in phosphorylation levels of AKT by 6EUDCA were dose
dependent (Figure 6).
Figure 6. Induction of phosphorylation of AKT by 6EUDCA is concentration dependent
in liver cells. Serum-starved HepG2 cell extracts were prepared after incubation with
various concentrations of 6EUDCA (DMSO/0 μM, 1 μM, 3 μM, 10 μM, or 30 μM) for
10 min. Extracts were analyzed by Western blot using specific antibodies against pAKT
S473 (left panel) or pGSK3beta (right panel). This data shows that much lower
concentrations of 6EUDCA were required to induce AKT phosphorylation as compared
to UDCA indicating a higher potency for 6EUDCA.
We next tested if UDCA- and 6EUDCA-dependent phosphorylation of AKT could be
mediated by FXR. Serum-starved mouse primary hepatocytes isolated from either wild-
type or fxr-/- mice were treated with or without UDCA or 6EUDCA for 10 min. Cell
13

extracts were prepared and immunoblotted for phosphorylated AKT or phosphorylated
GSK3beta. Western blot analysis indicated that UDCA and 6EUDCA induced
phosphorylation of AKT in wild-type hepatocytes but not in fxr-/- hepatocytes (Figure 7).
Figure 7. Phosphorylation of AKT and GSK beta is mediated by FXR in liver cells.
Serum-starved primary hepatocyte extracts from either wild-type or fxr-/- mice were
prepared after incubation with FXR ligands (10 min). Extracts were analyzed by Western
blot using specific antibodies against pAKT S473 (A and B) or pGSK3beta (C and D).
This result strongly suggests that UDCA and 6EUDCA act via FXR in liver cells to
induce phosphorylation of AKT. This result also suggests a new non-genomic function
for FXR apart from its role as a transcription factor in liver cells.
2.3. Computational modeling of FXR and binding of UDCA and 6EUDCA to FXR
14

Finally, to explore the binding mode of 6EUDCA and UDCA to FXR, our in-house All-
Around-Docking (AAD) methodology 33 was used to predict the best docking site and
docking pose. AAD uses Glide software (Schördinger) and allows a small molecule to
search the whole surface of a target protein for the docking site with the lowest docking
score. The structure of the FXR protein bound to GW4064 was used as the docking target
(PDB ID: 3DCT). 34 Using this approach, 6EUDCA, UDCA, and GW4064 were
predicted to share the same FXR binding pocket (Figure 8A and 8B), which is formed by
M265, M290, R331, H447, W469, etc. 6EUDCA contacts more closely with the surface
of FXR than that of UDCA. For example, there are 10 residues within 2Å of 6EUDCA;
these residues are M265, A291, H294, V325, M328, F329, R331, S332, H447, and W454.
In contrast, there are only 3 residues (A291, R331 and S332) within 2Å of UDCA.
Our computational modes show that 6EUDCA forms a larger network of hydrogen bonds
with FXR as compared to UDCA. As depicted in Figure 8C, two hydrogen bonds are
formed between the UDCA molecule and FXR protein. On the other hand, eight
hydrogen bonds are formed between 6EUDCA and FXR (Figure 8D). Furthermore, the
6EUDCA ethyl can form a strong hydrophobic interaction with the W454 residue (Figure
8D). The calculated binding energies predict that FXR associates more strongly with
GW4064 (estimated -17.19 kcal/mol binding energy) than UDCA (estimated -9.24
kcal/mol) or 6EUDCA (-10.85 kcal/mol). Based on these results, 6EUDCA is predicted
to bind more strongly than UDCA.
15

GW4064
UDCA
B
D
A
C
R331
M328
M290
S332
F329
V325
H447
M365
H447
W454
Figure 8. Binding modes of 6EUDCA and UDCA to protein FXR. A, All-Around
Docking poses of UDCA (pink sticks) and GW4064 molecule (grey sticks) on FXR. B,
Binding pocket of FXR bound by 6EUDCA. All three compounds above are predicted to
bind in the same pocket. C, Two hydrogen bonds (red dots) are formed between UDCA
and residues H447 and M365 on FXR. The UDCA carbon atom that corresponds to
6EUDCA ethyl is shown in blue. D, Eight hydrogen bonds (red dots) are formed between
6EUDCA and residues S332, R331, V325, F329, M328, H447, and M290. The 6EUDCA
ethyl is shown in blue and has strong hydrophobic interactions with the W454 residue.
16

3. Conclusions
In summary, we have described a new chemical synthesis of a modified derivative of
UDCA, 6EUDCA, and characterized its biologic effects. As expected based on previous
results, neither compound was able to induce expression of the well known FXR target
gene SHP. However, both compounds are apparently able to induce increased
phosphorylation of AKT and GSK3beta in HepG2 cells, and in primary wild type mouse
hepatocytes, but not those from FXR null mice, suggesting a role for UDCA and
6EUDCA, in modulating cell signaling in an FXR-dependent manner. This may lead to
the identification of a novel mechanism by which bile acids regulate cell function, and
6EUDCA may be an effective targeted CGD therapeutic. Detailed investigation of the in
vivo biological effects of 6EUDCA in mouse model of CGD is under way.
4. Experimental section
4.1. Chemistry
Organic reagents were purchased from commercial suppliers unless otherwise noted and
were used without further purification. All solvents were analytical or reagent grade. All
reactions were carried out in flame-dried glassware under argon or nitrogen. Melting
points were determined and reported automatically by an optoelectronic sensor in open
1
capillary tubes and were uncorrected. H NMR and 13C NMR spectra were measured at
500 MHz and 125 MHz respectively, and using CDCl3 or CD3OD as the solvents and
tetramethylsilane (Me4Si) as the internal standard. Flash column chromatography was
performed using Sigma-Aldrich silica gel 60 (200-400 mesh), carried out under moderate
pressure by using columns of an appropriate size packed and eluted with appropriate
17

eluents. Silica gel chromatography was performed on a Biotage flash column gradient
pump system using 15 cm long columns. All reactions were monitored by TLC on
precoated plates (silica gel HLF). TLC spots were visualized either by exposure to iodine
vapors or by irradiation with UV light. Organic solvents were removed in vacuum by
rotary evaporator. Elemental analyses were performed by Columbia Analytical Services,
Inc. Tucson, Arizona.
4.1.1. 3-Hydroxy-7-keto-5-cholan-24-oic acid (1). To a suspension of
chenodeoxycholic acid (CDCA, 1.0 g, 2.5 mmol, Sigma-Aldrich) and silica gel (4 g, 200-
400 mesh, Aldrich) in anhydrous CHCl3 (2 mL) was added, portion-wise, pyridinium
chlorochromate (PCC, 0.81 g, 38 mmol) in 25 mL of CH2Cl2, and the reaction mixture
was stirred at room temperature for 15-20 min. The mixture was filtered and the filtrate
was washed with water (20 mL) and brine (20 mL). The organic layer was dried over
Na2SO4 and concentrated. The resulting crude oil was purified by flash column
chromatography (CH2Cl2: MeOH 95:5) to afford 1 as a solid (0.76 g, 78% yield), mp
201.2 oC (lit.22 mp 201.1 oC). 1H NMR (500 MHz, CD3OD)  3.51 (m, 1H), 2.94 (m, 1H),
2.52 (t, 1H), 2.30 (m, 2H), 2.19 (m, 6H), 1.70 (m, 2H), 1.43 (m, 4H), 1.31 (m, 6H), 1.19
(s, 3H), 1.12 (m, 4H), 0.92 (d, 3H), 0.66 (s, 3H). 13C NMR (125 MHz, CD3OD)  212.7,
176.8, 70.1, 54.8, 49.2, 48.9, 47.7, 46.0, 44.9, 43.0, 42.4, 38.9, 36.8, 35.1, 34.9, 33.7,
31.0, 30.6, 29.2, 27.8, 24.3, 22.0, 21.4, 17.3, 10.7. Anal. Calcd for C24H38O4 : C, 73.81;
H, 9.81. Found: C, 73.50; H, 9.63.
4.1.2. 3-Tetrahydropyranyloxy-7-keto-5-cholan-24-oic acid (2). To a solution of 1
(0.50 g, 1.3 mmol) in 16 mL of CHCl3:Cl2CH2:Et2O (1:1:2) were added p-toluensulfonic
acid (0.06 g, 0.3 mmol) and 3,4-dihydro-2H-pyrane (0.41 g, 4.9 mmol). The reaction
18

mixture was stirred at room temperature for 60 min. Water (20 mL) was added, and the
mixture was extracted with EtOAc (3 x 30 ml); the combined organic layers were washed
with brine, dried over Na2SO4, and concentrated. The resulting crude oil was purified by
flash column chromatography (CH2Cl2:Et2O 1:2) to afford 2 as a white solid (0.47 g, 76%
yield), mp 160.8 oC (lit.6 mp 157-159 oC). Selected 1H NMR (600 MHz, CDCl3)  4.73
(d, 1H), 3.86 (m, 1H), 3.59 (m, 1H), 3.46 (m, 1H), 2.82 (m, 1H), 1.17 (s, 3H), 0.92 (d,
3H), 0.63 (s, 3H). 13C NMR (125 MHz, CDCl3)  212.3, 179.8, 96.4, 62.8, 62.1, 19.8,
18.1, 11.4. Anal. Calcd for C29H46O5 : C, 73.38; H, 9.77. Found: C, 73.30; H, 9.76.
4.1.3. 3-Hydroxy-6-ethyl-7-keto-5-cholan-24-oic acid (3). To a solution of 2 (0.30
g, 0.63 mmol) in dry THF (20 mL) at -78oC were added dropwise n-butyllithium (1.0
mL, 1.6 M solution in hexane, 1.6 mmol), HMPA (0.7 g, 4 mmoL) and LDA (2.0 mL, 1.8
M in THF/heptane/ethylbenzene, 3.6 mmol). The reaction mixture was stirred for 30 min.
Iodoethane (2.0 g, 13 mmol) was slowly added and the reaction mixture was allowed to
warm overnight to room temperature. After rotary evaporation, water and ether were
added and the aqueous layer was acidified with 10% HCl and extracted with EtOAc (5 x
20 mL). The organic layers were washed with brine, dried over Na2SO4, and concentrated
to give a yellow oil. After a short column (CH2Cl2:Et2O 1:2), the crude semi-solid was
dissolved in ethanol (5 mL) and pyridinium p-toluenesulfonate (15 mg, 0.06 mmol) was
added. The reaction mixture was stirred at 55 oC for 4 h. After rotary evaporation, the
product was passed through a short column (CH2Cl2:Et2O 1:2) to obtain 3 as a semi-solid
(0.08 g, 30%) that was used in the next step without further purification.
4.1.4. General Method for Meerwein-Ponndorf-Verley (MPV) Type Reduction
19

To a solution of aluminum foil (30 eq.) in anhydrous isopropanol (80 eq.) was added
mercuric chloride (0.15 eq.). The reaction mixture was warmed up to reflux for 9-12
hours under an atmosphere of nitrogen until the aluminum foil completely dissolved into
the isopropanol. After the isopropoxide aluminum formed, it was used directly for the
reduction of 3-hydroxy-7-keto-5-cholan-24-oic acid (1) and 3-hydroxy-6-ethyl-7-
keto-5-cholan-24-oic acid (3). To this solution, 7-keto analog (1 eq.) was added and the
reaction mixture was heated to reflux for 3-5 hours under N2 and checked by TLC. After
cooling, the reaction mixture was dissolved in EtOAc (25 mL) and water was added (30
mL), the solution was acidified to pH 2 with 2N hydrochloric acid, and then extracted
with EtOAc (3 x 25 mL). The combined organic layers were washed with water and
brine, dried (Na2SO4), and evaporated under vacuum.
4.1.5. Usodeoxycholic acid (UDCA): yield: 82%.
Crystallization by EtOAc: pet ether 1:1 gave a white crystal, mp 202.5 oC (203-204 °C
(lit. Sigma)). Selected 1H NMR (600 MHz, CD3OD)  3.49 (brs, 2H), 2.37 (m, 1H), 2.22
(m, 1H), 1.93 (m, 2H), 1.87 (m, 3H), 0.97 (s, 3H), 0.96 (d, 3H), 0.71 (s, 3H). Selected 13C
NMR (125 MHz, CD3OD)  176.7, 70.7, 70.5, 56.1, 55.1, 20.9, 17.5, 11.0. Anal. Calcd
for C24H40O4 : C, 73.43; H, 10.27. Found: C, 73.29; H, 9.94.
4.1.6. 3, 7β-Dihydroxy-6-ethyl-5-cholan-24-oic acid (6EUDCA):
Yield: 80%. Solidification by MeOH/CH2Cl2/pet ether gave a beige powder, mp 109.5
oC. Selected 1H NMR (600 MHz, CDCl3):  3.50 (m, 1H), 3.36 (m, 1H, J = 8.18), 2.25
(m, 1H), 2.15 (m, 1H), 0.91 (d, 3H), 0.88 (s, 3H), 0.77 (t, 3H), 0.64 (s, 3H). 13C NMR
(125 MHz, CDCl3):  178.4, 75.2, 71.9, 55.8, 54.8, 43.7, 42.8, 41.2, 40.2, 39.6, 35.5,
20

34.4, 30.8, 30.7, 30.5, 28.3, 27.0, 23.5, 21.4, 20.8, 18.3, 12.0, 11.8. Anal. Calcd for
C26H44O4: C, 74.24; H, 10.54. Found: C, 74.00; H, 10.01.
4.2. Biology
4.2.1. Antibodies
Antibodies for pAKT S473, pGSK3beta, total AKT and total GSK3beta used for western
blot analysis were obtained from Cell Signaling (USA).
4.2.2. Cell culture
HepG2 cells were purchased from American Type Culture Collection. The cells were
cultured in Minimal Essential Medium (MEM) (Cellgro, Manassas, VA) supplemented
with 100 U/ml penicillin/streptomycin sulfate (Mediatech, Herndon, VA) and 10% (v/v)
heat-inactivated fetal bovine serum (Irvine Scientific, Santa Ana, CA). Primary
hepatocytes were isolated from mouse liver tissues and cultured in Dulbecco’s modified
Eagle medium and F-12 (Cellgro, Manassas, VA) supplemented with 100U/ml penicillin
G/streptomycin and 10% (v/v) heat-inactivated fetal bovine serum. Cells were cultured
in insulin or serum free media for 16 hr before performing all experiments unless
otherwise mentioned. Primary mouse hepatocytes were isolated as described from
C57BL6/J mice by using collagenase type IV; 5 × 105 cells per well were plated in six-
well collagen-coated plates and cultured in DMEM/Ham's F12 media.35
4.2.3. TR-FRET assay
To assess the binding of UDCA and 6EUDCA, we performed TR-FRET assay using
DY246 as an hFXR fluorescent probe as described earlier.30 For this assay, 30 nl of 10
mM testing chemical was transferred with a pintool into 15 μL of a mixture of GST–
21

hFXR and Tb-anti-GST in a 384-well black plate, and then 5 μL of 40 nM DY246 was
dispensed to give a final volume of 20 μL/well with 10 nM GST–hFXR, 1.5 nM Tb-anti-
GST, UDCA (1000 μM, 300 μM, or 100 μM) or 6EUDCA (100 μM, 30 μM, or 10μM)
and 10 nM DY246. In addition, selected wells containing 5 μM GW4064 or DMSO were
used as positive and negative controls, respectively. The final DMSO concentration was
0.23% in all wells. The plates were then spun down after a brief shake and incubated for
either 15 or 25 min at room temperature. The TR-FRET signal was then collected for
each well with an Envison plate reader using an excitation wavelength of 340 nm and
emission wavelengths of 520 nm and 490 nm.
4.2.4. RNA isolation and Quantitative Real-time PCR
RNA was isolation, reverse transcription reactions, and real time PCR were performed as
described previously.36 SHP primers forward 5’-GCC CTG CAC TCT CGC TTT CT-3’;
reverse 5’-CAA CTG GGC ACC GAG GCA ACA GTT G-3’were used to amplify using
Sybr green master mix purchased from applied biosystems (Foster City, CA).
Amplification of 36b4 with specific primers forward 5’-TGG AGT CTT TCT GGA GCC
TT-3’; reverse 5’-TCC TGT TGC AGG TGT GCG AT-3’ was used as internal
control. Absolute mRNA expression was quantified using standard curve method. All
primers used were Sybr green-based synthesized by IDT. Sybr green was purchased from
Applied Biosystems (Foster City, CA). Amplification of 36b4 was used as an internal
control. Absolute mRNA expression was quantified using the standard curve method.
4.2.5. Western blot analysis
Western blot analysis was performed according to methods described earlier.37 The
antibodies used are listed above (4.2.1). Briefly, serum-starved HepG2 cells or mouse
22

primary hepatocytes from flox/flox or fxr-/- mice cultured in collagen-coated 6-well
culture dishes were used to make total cell lysates. Proteins were resolved on SDS-
PAGE gels and transferred onto nitrocellulose membrane and blotted with antibodies
against phosphor AKT (S473) and pGSK3beta. The signal with secondary antibody was
detected with either an Odyssey 3.0 infrared scanner or ECL chemiluminiscence.
ACKNOWLEDGMENTS
We thank Drs. Arthur D. Riggs, Jyoti Kusari, and Keely Walker for helpful discussion.
We thank Ramani Ravirala for technical assistance. We also thank the Ruth B. and
Robert K. Lanman Chair Endowment for support and encouragement.
REFERENCES and notes
(1) Shaffer, E. A. Epidemiology and risk factors for gallstone disease: has the
paradigm changed in the 21st century? Curr Gastroenterol Rep 2005, 7, 132-140.
(2) Wang, H. H.; Portincasa, P.; de Bari, O.; Liu, K. J.; Garruti, G.; Neuschwander-
Tetri, B. A.; Wang, D. Q. Prevention of cholesterol gallstones by inhibiting hepatic
biosynthesis and intestinal absorption of cholesterol. Eur J Clin Invest. 2013, 43,
413-26.
(3) Salen, G.; Colalilo, A.; Verga, D.; Bagan, E.; Tint G. S.; Shefer S. Effect of high
and low doses of ursodeoxycholic acid on gallstone dissolution in humans
Gastroenterology, 1980, 78, 1412-1418.
(4) D’Amore, C.; Di Leva, F. S.; Sepe, V.; Renga, B.; Del Gaudio, C.; D’Auria, M. V.;
Zampella, A.; Fiorucci, S.; Limongelli, V. Design, synthesis, and biological
23

evaluation of potent dual agonists of nuclear and membrane bile acid receptors. J.
Med. Chem. 2014, 57, 937−954.
(5) Moschetta, A.; Bookout A. L.; Mangelsdorf, D. J. Prevention of cholesterol
gallstone disease by FXR agonists in a mouse model. Nat Med 2004, 10, 1352-
1358.
(6) 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. Identification of a nuclear receptor that is activated by farnesol
metabolites, Cell 1995, 81, 687-693.
(7) Watanabe, M.; Houten, S. M.; Mataki, C.; Christoffolete, M. A.; Kim, B. W.; Sato,
H.; Messaddeq, N.; Harney, J. W.; Ezaki, O. T.; Kodama, T.; Schoonjans, K.;
Bianco, A. C.; Auwerx, J. Bile acids induce energy expenditure by promoting
intracellular thyroid hormone activation. Nature 2006, 439, 484–489.
(8) Wang, H.; Chen, J.; Hollister, K.; Sower, L. C.; Forman, B. M. Endogenous bile
acids are ligands for the nuclear receptor FXR/BAR, Mol. Cell, 1999, 3, 543-553.
(9) Heathcote, E. J. Management of primary biliary cirrhosis. The American
Association for the Study of Liver Diseases practice guidelines. Hepatology 2000,
31, 1005–1013.
(10) Yu, D. D.; Sousa, K. M.; Mattern, D. L.; Vaidehi, N.; Forman, B. M.; Huang, W.
et al. Stereoselective Synthesis, Biological Evaluation, and Modeling of Novel
Bile Acid-Derived G-Protein Coupled Bile Acid Receptor 1 (GP-BAR1, TGR5)
Agonists, Bioorganic & Medicinal Chemistry 2015, 23, 1613-1628.
24

(11) Sharma R, Prichard D, Majer F, Byrne AM, Kelleher D, Long A, Gilmer JF.
Ursodeoxycholic acid amides as novel glucocorticoid receptor modulators. J Med
Chem. 2011, 54, 122-30.
(12) Campana, G.; Pasini, P.; Roda, A.; Spampinato, S. Regulation of ileal bile acid-
binding protein expression in Caco-2 cells by ursodeoxycholic acid: Role of the
farnesoid X receptor, Biochemical Pharmacology, 2005, 61, 1755-1763.
(13) Wittenburg, H.; Lyons, M. A.; Li, R.; Churchill, G. A.; Carey, M. C.; Paigen, B.
“FXR and ABCGs/ABCG8 as determinants of cholesterol gallstone formation
fromquantitative trait locus mapping in mice,” Gastroenterology, 2003, 125, 868–
881.
(14) Pellicciari, R.; Fiuorucci, S.; Camaioni, E.; Clerici, C.; Costantino, G.; Maloney,
P. R.; Morelli, A.; Parks, D. J.; Willson, T. M. 6α-Ethyl-Chenodeoxycholic Acid
(6-ECDCA), a Potent and Selective FXR Agonist Endowed with Anticholestatic
Activity. J. Med. Chem. 2002, 45, 3569-3572.
(15) Yu, D.; Mattern, D. L.; Forman, B. M. An Improved Synthesis of 6α-
Ethylchenodeoxycholic Acid (6ECDCA), a Potent and Selective Agonist for the
Farnesoid X Receptor (FXR), Steroids 2012, 77, 1335.
(16) US 0039435 Al, 2011.
(17) Liu, Z. Advance in methods for preparation of ursodeoxycholic acid, Yaoxue
Tongbao. 1988, 23, 583–86, Chem Abstr. 1989, 110, 115167s.
(18) Kanajawa, T. Proc. Japan Acad 1954, 30, 391.
(19) Hofmann, A. F. Pharmacology of ursodeoxycholic acid, an enterohepatic drug.
Scand J. Gastroenterol, 1994, 29, 1-15.
25

(20) Fieser, L.F.; Rajagopalan, S. Oxidation of steroid. III. Selective oxidation and
acylations in bile acid series. J Am Chem Soc 1950, 72, 5530–6.
(21) Sammuelson, B. Preparation of ursodeoxycholic acid and 3,7β,12-
trihydroxycholanic acid. Acta Chem Scand 1960, 14, 17–20.
(22) ES 489,661, 1980.
(23) Iida, T.; Chang, F. C. Potential Bile Acid metabolites. 6. Stereoisomeric 3,7-
Dihydroxy-5β-Cholanic Acids. J.Org. Chem. 1982, 47, 2969.
(24) ES 499,525, 1981.
(25) Piatkowski, W.; Mazurkiewicz, W. Pol. J. Appl. Chem. The oxidation
technology of hydroxycholanic acids to the corresponding oxo- derivatives.
1999, 43, 85-93.
(26) Hsia, S. L.; Matschiner, J. T.; Mahowald, T. A.; Eliott, W. H.; Doisy, E. A.;
Thayer, S.A.; Doisy, E. A. Bile acids. V. Chemical studies on new bile acids from
the rat and the hog. J. Biol. Chem. 1957, 811-823.
(27) (a) Meerwein, H.; Schmidt, R. Halogenated alcohols. Justus Liebigs Ann. Chem.
1925, 444, 221. (b) Ponndorf, W. The reversible exchange of oxygen between
aldehydes or ketones on the one hand and primary or secondary alcohols on the
other hand. Angew. Chem. 1926, 39, 138. (c) Verley, M. Bull. Soc. Chim. Fr. 1925,
37, 871.
(28) Miyashita, N; Yoshkoshi, A; Grieco, P. A. Pyridinium p-toluenesulfonate. A
mild and efficient catalyst for the tetrahydropyranylation of alcohols. J. Org.
Chem. 1977, 42, 3772-3774.
26

(29) Gioiello, A.; Macchiarulo, A.; Carotti, A.; Filipponi, P.; Costantino, G.; Rizzo, G.;
Adorini, L.; Pellicciari, R. Extending SAR of bile acids as FXR ligands: discovery
of 23-N-(carbocinnamyloxy)-3α,7α-dihydroxy-6α-ethyl-24-nor-5β-cholan-23-
amine. Bioorg. Med. Chem. 2011, 19, 2650−2658.
(30) Bhattacharyya, P. K.; Bankawala, Y. G. Determination of chenodeoxycholic acid
and ursodeoxycholic acid by nuclear magnetic resonance spectrometry. Analytical
Chem., 1978, 50, 1462-1465.
(31) Yu, D. D.; Lin, W.; Chen, T.; Forman, B. M. Development of Time Resolved
Fluorescence Resonance Energy Transfer-based Assay for FXR Antagonist
Discovery, Bioorganic & Medicinal Chemistry, 2013, 21, 4266-4278.
(32) 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. Identification of a chemical tool for the orphan nuclear receptor
FXR. J. Med. Chem. 2000, 43, 2971–2974.
(33) Nam, S.; Wen, W.; Schroeder, A.; Herrmann, A.; Yu, H.; Cheng, X.; Merz, K. H.;
Eisenbrand, G.; Li, H.; Yuan, Y. C.; Jove, R. Dual inhibition of Janus and Src
family kinases by novel indirubin derivative blocks constitutively-activated Stat3
signaling associated with apoptosis of human pancreatic cancer cells. Mol Oncol.
2013, 7, 369-78.
(34) 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. Conformationally
27

constrained farnesoid X receptor (FXR) agonists: Naphthoic acid-based analogs of
GW 4064. Bioorg. Med. Chem. Lett. 2008, 18, 4339-4343.
(35) Bissell, D. M; Guzelian, P. S. Phenotypic stability of adult rat hepatocytes in
primary monolayer culture. Ann N Y Acad Sci. 1980, 349, 85-98.
(36) Li, T; Kong, X; Owsley, E; Ellis, E; Storm, S; Chiang, J.Y. Insulin regulation of
cholesterol 7alpha-hydroxylase expression in human hepatocytes: roles of
forkhead box O1 and sterol regulatory element-binding protein 1c. J. Biol. Chem.
2006, 281, 28745-18754.
(37) Andrali, S. S; März, P; Ozcan, S. Ataxin-10 interacts with O-GlcNAc transferase
OGT in pancreatic beta cells. Biochem Biophys Res Commun. 2005, 337, 149-153.
Graphical Abstract
28