Synthesis and evaluation of Vitamin D receptor-mediated activities of cholesterol and Vitamin D metabolites

Article abstract of DOI:10.1016/j.ejmech.2016.01.002

A systematic study with phase 1 and phase 2 metabolites of cholesterol and Vitamin D was conducted to determine whether their biological activity is mediated by the Vitamin D receptor (VDR). The investigation necessitated the development of novel synthetic routes for lithocholic acid (LCA) glucuronides (Gluc). Biochemical and cell-based assays were used to demonstrate that hydroxylated LCA analogs were not able to bind VDR. This excludes VDR from mediating their biological and pharmacological activities. Among the synthesized LCA conjugates a novel VDR agonist was identified. LCA Gluc II increased the expression of CYP24A1 in DU145 cancer cells especially in the presence of the endogenous VDR ligand 1,25(OH)₂D_3.Furthermore, the methyl ester of LCA was identified as novel VDR antagonist. For the first time, we showed that calcitroic acid, the assumed inactive final metabolite of Vitamin D, was able to activate VDR-mediated transcription to a higher magnitude than bile acid LCA. Due to a higher metabolic stability in comparison to Vitamin D, a very low toxicity, and high concentration in bile and intestine, calcitroic acid is likely to be an important mediator of the protective Vitamin D properties against colon cancer.

Full text of DOI:10.1016/j.ejmech.2016.01.002

Accepted Manuscript
Synthesis and evaluation of vitamin D receptor-mediated activities of cholesterol and
vitamin D metabolites
Kelly A. Teske, Jonathon W. Bogart, Luis M. Sanchez, Olivia B. Yu, Joshua V.
Preston, James M. Cook, Nicholas R. Silvaggi, Daniel D. Bikle, Leggy A. Arnold
PII:
S0223-5234(16)30002-2
10.1016/j.ejmech.2016.01.002
EJMECH 8292
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 25 November 2015
Revised Date: 3 January 2016
Accepted Date: 4 January 2016
Please cite this article as: K.A. Teske, J.W. Bogart, L.M. Sanchez, O.B. Yu, J.V. Preston, J.M. Cook,
N.R. Silvaggi, D.D. Bikle, L.A. Arnold, Synthesis and evaluation of vitamin D receptor-mediated activities
of cholesterol and vitamin D metabolites, European Journal of Medicinal Chemistry (2016), doi: 10.1016/
j.ejmech.2016.01.002.
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.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Synthesis and Evaluation of Vitamin D Receptor-Mediated Activities of
Cholesterol and Vitamin D Metabolites
Kelly A. Teske, Jonathon W. Bogart, Luis M. Sanchez, Olivia B. Yu, Joshua V. Preston, James M.
Cook, Nicholas R. Silvaggi, Daniel D. Bikle, and Leggy A. Arnold*
Department of Chemistry and Biochemistry and Milwaukee Institute for Drug Discovery,
University of Wisconsin, Milwaukee, Milwaukee, Wisconsin 53211, United States
* Corresponding author: arnold2@uwm.edu
ABSTRACT
A systematic study with phase 1 and phase 2 metabolites of cholesterol and vitamin D was
conducted to determine whether their biological activity is mediated by the vitamin D receptor
(VDR). The investigation necessitated the development of novel synthetic routes for lithocholic
acid (LCA) glucuronides (Gluc). Biochemical and cell-based assays were used to demonstrate
that hydroxylated LCA analogs were not able to bind VDR. This excludes VDR from mediating
their biological and pharmacological activities. Among the synthesized LCA conjugates a novel
VDR agonist was identified. LCA Gluc II increased the expression of CYP24A1 in DU145 cancer
cells especially in the presence of the endogenous VDR ligand 1,25(OH)2D3. Furthermore, the
methyl ester of LCA was identified as novel VDR antagonist. For the first time, we showed that
calcitroic acid, the assumed inactive final metabolite of vitamin D, was able to activate VDR-
mediated transcription to a higher magnitude than bile acid LCA. Due to a higher metabolic
stability in comparison to vitamin D, a very low toxicity, and high concentration in bile and
1

ACCEPTED MANUSCRIPT
intestine, calcitroic acid is likely to be an important mediator of the protective vitamin D
properties against colon cancer.
KEYWORDS: Vitamin D Receptor; 1,25(OH)2D3; metabolites; lithocholic acid; calcitroic acid;
CYP24A1; DU145
ABBREVIATIONS: VDR, Vitamin D Receptor; 1,25(OH)2D3, calcitriol; NR, nuclear receptors; LCA,
lithocholic acid; LCME, lithocholic acid methyl ester; UDCA, ursodeoxycholic acid; CA, cholic
acid; DCA, deoxycholic acid; CDCA, chenodeoxycholic acid; HDCA, hyodeoxycholic acid; GLCA,
glycolithocholic acid; TLCA, taurolithocholic acid; SLCA, sulfonolithocholic acid; LCA Gluc I,
lithocholic acid glucuronide I; LCA Gluc II, lithocholic acid glucuronide II; Cyp24A1, 24-
hydroxylase gene.
Introduction
Many biological activities of vitamin D, once converted to calcitriol (1,25(OH)2D3), are
mediated by the vitamin D receptor (VDR, NR1I1), a ligand inducible transcription factor
belonging to the superfamily of nuclear receptors (NRs)[1]. When 1,25(OH)2D3 binds VDR, it
undergoes a conformational change and initiates a cascade of events including the association
of its heterodimeric partner, retinoid X receptor (RXR), followed by the recruitment of specific
coactivators needed for transcription [2]. Ligand activated VDR has been shown to regulate the
expression of VDR target genes involved in calcium homeostasis, immunity, and cellular growth
and differentiation. In addition to its role in vitamin D physiology, VDR has been shown to bind
bile acids and act as sensor for elevated concentrations of bile acids in the intestine [3]. Bile
2

ACCEPTED MANUSCRIPT
acids are end products of hepatic cholesterol catabolism and act as emulsifiers for ingestion
and intestinal absorption of hydrophobic nutrients like cholesterol, fatty acids and lipid-soluble
vitamins like vitamin D. Primary bile acids are produced from cholesterol in the liver through a
series of enzymatic reactions catalyzed by P450 enzymes followed by their secretion in the bile
as glycine or taurine conjugates, where they assist in lipid digestion and absorption [4]. Most
bile acids are reabsorbed in the intestine and recirculate to the liver, while others are converted
to secondary bile acids, such as lithocholic acid (LCA) [5]. LCA is poorly reabsorbed into the
enterohepatic circulation and can pass into the colon. At higher concentrations, LCA is toxic and
has been linked to the development of liver disease and colorectal cancer [6, 7]. VDR plays an
important role in maintaining bile homeostasis by inducing a LCA detoxification mechanism in
the liver and intestine through the upregulation of CYP3A4 [8]. The crystallographic structure of
VDR with LCA has been solved for both rVDR[9] and zVDR [10]. The zVDR crystallographic study
revealed the necessity of a second LCA, which anchors to VDR by forming direct H-bonds at a
considerably lower affinity compared to the usual ligand binding site but importantly stabilizes
the active protein conformation. These studies suggest that other LCA metabolites might have
stronger interactions with VDR than LCA, thus many LCA-derived ligands have been synthesized
and investigated [11, 12]. The results showed that esterification of the LCA side chain carboxyl
group with methyl, ethyl, and benzyl moieties resulted in very weak VDR agonists. However,
esterification of the C3 hydroxyl group (LCA acetate) exhibited a 30-fold improved agonistic
VDR-mediated transcriptional activity in comparison to LCA [11]. In colon cancer cells and
mouse intestine, LCA acetate upregulated CYP3A significantly more than LCA, suggesting that
LCA analogs could be developed into chemoprevention against colorectal cancer.
3

ACCEPTED MANUSCRIPT
Similar to the negative feedback loop responsible for the catabolic breakdown of LCA by
CYP3A4, 1,25(OH)2D3 metabolism is mediated by the VDR target gene, CYP24A1 [13]. Major
catabolism of 1,25(OH)2D3 occurs at the secosteroid aliphatic chain either forming C24-
oxidation pathway products or C23-hydroxylation pathway products [14]. The final product of
the 23-hydroxylation pathway is 1,25(OH)2D-26,23-lactone that exhibit antagonistic VDR
activities in vivo [15]. The final metabolite formed by the C24-oxidation pathway is the bile
excretory product calcitroic acid. CYP24A1 is induced greatly by 1,25(OH)2D3 in many different
cancer cells including prostate cancer cells DU145 [16]. The resulting fast metabolism of
1,25(OH)2D3 reduces its ability to induce cancer cell growth inhibition [17]. However, many
aspects of this pathway are still unknown, especially the role of vitamin D metabolites. Herein,
we report the systematic evaluation cholesterol and vitamin D metabolites in respect to VDR
binding, modulation of VDR-mediated transcription, and toxicity.
Results and Discussion
The following compounds depicted in Figure 1 were investigated in this study.
4

ACCEPTED MANUSCRIPT
Figure 1. Summary of the potential natural VDR ligands.
These include identified VDR agonists such as LCA, LCA acetate, and LCA methyl ester (LCME)
[11]. Secondly, we investigated vitamin D metabolite calcitroic acid [13], and cholesterol phase
1 metabolites ursodeoxycholic acid (UDCA), an FDA approved drug against primary biliary
cirrhosis [18], cholic acid (CA) a treatment for bile acid synthesis disorders due to enzymatic
effects [19], deoxycholic acid (DCA) used for the reduction of fat under the chin [20],
chenodeoxycholic acid (CDCA) a treatment against gallstones [21], and hyodeoxycholic acid
(HDCA), which is not marketed but has a similar activities to that of CDCA. In addition, phase 2
metabolites were studied such as glycolithocholic acid (GLCA), taurolithocholic acid (TLCA),
sulfonolithocholic acid (SLCA), lithocholic acid glucuronides I and II (LCA Gluc I and II).
5

ACCEPTED MANUSCRIPT
Chemistry
LCA derivatives and metabolites, which are not commercially available were synthesized
from LCA or LCME. LCA was the starting material used to synthesize LCA acetate and SLCA
(Scheme 1). A base-catalyzed esterification reaction using 4-dimethylaminopyridine and acetyl
chloride produced LCA acetate in a 95% yield. To obtain SLCA, sulfuric acid and acetic
anhydride in pyridine were used, as opposed to the reported method using pyridine sulfur
trioxide to make sulfonate steroids [22, 23]. The final product was converted into the
corresponding ammonium salt using 25% ammonia acetate at 0oC to obtain the final product in
a 97% yield.
Scheme 1. Conversion of LCA to LCA acetate and SLCA. i) Pyridine, acetyl chloride, cat. 4-
DMAP, rt, 1 h; ii) sulfuric acid, acetic anhydride, pyridine, 50-55oC, 30 min. iii) 25% ammonia in
water, 0oC, 15 min.
For the synthesis of LCA Gluc I and II (Scheme 2), LCA methyl ester (LCME) was used as
starting material. A Koenigs-Knorr condensation reaction of LCME with acetobromo-α-D-
glucuronic acid methyl ester in the presence of CdCO3 in dry benzene gave LCME Gluc I [24].
6

ACCEPTED MANUSCRIPT
1
The β-glycosidic linkage with LCA methyl ester at the C-3 position was confirmed by HNMR.
Hydrolysis with sodium hydroxide afforded the final product, LCA Gluc I, in an overall yield of
61%. LCA Gluc II was synthesized in four steps starting with the protection of LCME with t-
butyldimethylsilylether (TBS). Hydrolysis of LCME-TBS afforded the corresponding carboxylic
acid, LCA-TBS. Next, glucuronic acid was converted into a tetrabutylammonium salt (1)
(Scheme 3), which not only protected the free carboxyl group but also rendered it soluble in
organic solvents [25].
Scheme 2. Synthetic scheme for converting LCME to LCA O-glucuronide I and II. i) Dry
benzene, CdCO3, acetobromo-α-D-glucuronic acid methyl ester, reflux, 5 h; ii) MeOH, 1M NaOH,
rt, 4 h followed by 5% HCl; iii) TBDMSCl, DMF, imidazole, rt; iv) THF, 2M NaOH, 75oC, 24 h
7

ACCEPTED MANUSCRIPT
followed by 2M HCl; v) pyridine, CDI, reflux, 3 h; vi) a) NaH, tetrabutylammonium glucuronate,
50oC, 4-5 h; b) acetic acid; vii) THF, tetrabutylammonium fluoride, 40oC, H2O.
Scheme 3. Tetrabutylammonium salt (1) formation of glucuronic acid. i) MeOH, 40oC, 1 h.
In a one pot reaction, LCA-TBS was activated with 1,1’-carbonyldiimidazole followed by
the addition of tetrabutylammonium glucoronate (1). The reaction was quenched with acetic
acid to produce LCA-TBS Gluc II. To avoid accidental cleavage of the coupled sugar ring,
deprotection of the TBS group was accomplished under neutral conditions with
tetrabutylammonium fluoride as oppose to acidic conditions. This reaction was quenched with
water to yield the final product, LCA Gluc II, which was further purified by silica gel column
chromatography to remove any unreacted LCA followed by recrystallization in EtOH to yield
about 40 mg of product in an overall yield of 10%. Although this multistep synthesis produced
little product, it was beneficial to have the sugar as the donor and LCA as the acceptor molecule
since it omitted unnecessary protection and deprotection of the sugar hydroxyl groups.
Biological Evaluation
Each compound was initially evaluated with a VDR–coactivator binding assay using a
fluorescently labeled SRC2-3 coactivator peptide and recombinantly expressed VDR-LBD [26].
The results are summarized in Table 1.
8

ACCEPTED MANUSCRIPT
Table 1. Summary of VDR-mediated activities of cholesterol and vitamin D metabolites
Com-
VDR-SRC2-3
Activation EC50
(µM)a
VDR-SRC2-3
Inhibition IC50
(µM)a
VDR 2-Hybrid
EC50 (µM)b
VDR 2-Hybrid
IC50 (µM)c
Toxicity LD50
(µM)d
pound
LCA
>50
13.6 ± 4.6
>150
2.14 ± 1.03 (39%)
>150
46.7 ± 8.4
15.1 ± 5.1
>50
55.9 ± 6.5
50.4 ± 12.0
>50
LCME
>150
>150
>100
LCA Acetate
Calcitroic Acid
11.0 ± 3.2
2.29 ± 0.43
1.49 ± 0.65 (94%)
0.87 ± 0.29 (55%)
>100
>100
UDCA
CA
>150
>150
>150
>150
>150
>150
>150
>150
>50
>150
>150
>150
>150
>150
>150
>150
>50
>150
>50
>50
>50
>50
>50
>50
>150
>150
>150
>150
>150
>150
>150
>150
>150
>150
DCA
>150
CDCA
HDCA
TLCA
>150
>150
14.5 ± 4.3
40.1 ± 8.0
23.3 ± 5.5
31.5 ± 7.2
>150
GLCA
SLCA
>150
>150
LCA Gluc I
20.5 ± 13.1 (49%)
12.3 ± 3.9 (55%)e
6.52 ± 2.61 (29%)
21.3 ± 8.0 (70%)e
LCA Gluc II
>150
2.86 ± 0.55
>150
>150
Table 1 is divided into three sections by bolds lines: known VDR ligands, phase 1 metabolites and phase
a
2 metabolites. Fluorescence polarization at 635/685 nm with VDR-LBD (0.1 μM), Alexa Fluor-SRC2-3 (7
nM), with and without VDR-agonist LG190178 (0.75 µM). EC50 and IC50 values were obtained by non-
linear regression (Y = Bottom + (Top − Bottom)/(1 + 10(logIC50−X)(HillSlope)) using two independent
9

ACCEPTED MANUSCRIPT
b
experiments in quadruplet. The maximum concentration used was 450 µM. Two-hydrid assay in
HEK293T cells using VP16-VDR-LBD, SRC1-GAL4, and luciferase reporter plasmid vector with a 4xUAS
repeat with or without 10 nM 1,25(OH)2D3. Efficacy, shown in parenthesis was calculated in respect to
full activation with 10 nM 1,25(OH)2D3. cTwo-hydrid assay in HEK293T cells using VP16-VDR-LBD, SRC1-
GAL4, and luciferase reporter plasmid vector with a 4xUAS repeat in the presence of 10 nM 1,25(OH)2D3.
d
e
Treated HEK293 evaluated with Cell-TiterGlo (Promega); Transcription assay using a hVDR-Gal4
plasmid and a luciferase reporter plasmid containing a 4xUAS repeat.
LCME, LCA acetate, and calcitroic acid were not able to support the VDR-LBD recruitment of
the SRC2-3 labeled coactivator peptide in this assay and even LCA, with a reported cellular
activity of approximately 10 µM, exhibited a limited agonistic activity of greater than 50 μM [3].
The presence of two VDR binding sites for LCA might be responsible for the weak agonistic
binding observed in this assay [10]. However, the VDR-SRC2-3 binding was inhibited by LCA with
an IC50= 13.6 ± 4.6 μM, while the introduction of a methyl ester (LCME) rendered the molecule
inactive. LCA acetate had a similar inhibitor potency than LCA, whereas calcitroic acid was
approximately six times more potent with an IC50= 2.29 ± 0.43 μM. Obviously, the recruitment
of SRC2-3 by VDR-LBD is sensitive to any three-dimensional protein changes induced by ligand
binding. For instance, crystal structures of VDR with 1,25(OH)2D3 or LCA showed different
conformations of coactivator peptide MED1 [9]. The specificity of coregulator recruitment
induced by VDR ligands using this and other binding assays has been reported [27]. Thus, LCA
acetate and calcitroic acid and to a lesser extent LCA induce a three dimensional VDR
conformation that did not enable SRC2 to bind VDR.
10

ACCEPTED MANUSCRIPT
In addition, two cell-based assays were conducted in HEK-293T kidney cells. For the VDR
transcription assay, a CMV-VDR-Gal4 fusion plasmid and a luciferase reporter plasmid with a
4xUAS promoter were over expressed. The 2-hybrid assay utilized three plasmids: a fusion
plasmid of coactivator SRC1 and a GAL4 DNA binding domain, a fusion plasmid of VDR LBD and
VP16, and a luciferase reporter plasmid with a 4xUAS promoter repeat. The assay enables the
evaluation of ligand-dependent VDR recruitment of SRC1 in cells. The transcription and VDR–
SRC1 binding efficacy induced by VDR ligands was normalized to the response of 10 nM
1,25(OH)2D3. The compound cytotoxicity was determined in HEK293T cells by cellular ATP
quantification.
In contrast to the VDR–SRC2 interaction, VDR agonist LCA promoted the cellular interaction
between VDR and SRC1 with an EC50 value of 2.14 ± 1.03 μM. In comparison to the binding
efficacy of 1,25(OH)2D3, LCA only partially agonized VDR with an efficacy of 39%. When
competing with 1,25(OH)2D3, LCA showed weak inhibition in the 34-47 µM range, which was
likely caused by increased cell death at higher concentration due to its moderate toxicity.
Analog LCME did not activate transcription and exhibited similar toxicity as LCA. However,
LCME inhibited VDR–SRC1 recruitment at significantly lower concentrations with an IC50 value
of 15.1 ± 5.12 μM. Thus LCME behaves like a VDR antagonist supporting earlier results with VDR
antagonist GW0742 that VDR ligand carboxylic esters antagonize transcription more strongly
than their corresponding carboxylic acids [28]. The reported VDR agonist LCA acetate exhibited
an EC50 value of 1.49 ± 0.65 μM. No inhibition of VDR–SRC1 binding in the presence of
1,25(OH)2D3 was observed in cells. The efficacy of LCA acetate was 94% with respect to
1,25(OH)2D3. Importantly, calcitroic acid exhibited the highest agonistic potency among all
11

ACCEPTED MANUSCRIPT
metabolites investigated with an EC50 value of 0.87 ± 0.29 μM. The efficacy, however, was
lower than LCA acetate but higher than LCA at 55% with respect to 1,25(OH)2D3.
LCA phase 1 metabolites with hydroxyl groups attached on the B and C rings did not interact
with VDR at concentrations comparable to their concentration in the intestine (250 µM in
upper ileum to 1 mM in lower ileum) [29]. Although not toxic, these metabolites did not
promote or inhibit VDR-coactivator interactions or mediate VDR transcription. The results could
be attributed to the additional hydroxyl groups in what is typically a hydrophobic region of VDR
ligands. The VDR-LBD pocket is mostly hydrophobic with hydrogen bonding occurring at the
outer ends of the pocket. Therefore, hydroxyl groups in that hydrophobic region can promote
unfavorable binding.
The investigation of LCA conjugates demonstrated their inability to promote the interaction
between VDR-LBD and coregulator peptide SRC2-3 in contrast to 1,25(OH)2D3. However, LCA
Gluc II was able to bind VDR but induced a significant conformational change that excluded
SRC2-3 binding to VDR with an IC50 of 2.86 ± 0.55 µM. Other metabolites moderately inhibited
the VDR–SRC2-3 interaction with IC50 values ranging between 14.5-40 µM. In cells, LCA
conjugates LCA Gluc I and LCA Gluc II promoted the interaction between VDR and SRC1. The
observed EC50 values were 20.5 ± 13.1 µM and 6.52 ± 2.60 µM, respectively. These
concentrations are in the range of physiological concentrations of bile acids, which vary
between 2 mM to 10 mM in the small intestine [29]. In the presence of 1,25(OH)2D3, no
inhibition of VDR–SRC1 binding was observed. In addition, a transcription assay using a hVDR-
Gal4 and luciferase reporter plasmid confirmed the agonistic activity of LCA Gluc I and LCA Gluc
II. The efficacy for LCA Gluc II in this assay was significantly higher than LCA Gluc I with (70%)
12

ACCEPTED MANUSCRIPT
and (55%), respectively. Thus the interaction between VDR and SRC1 is biologically relevant in
contrast to VDR–SRC2 binding. As a control experiment, HEK293 cells were transfected with a
PPARα-Gal4 plasmid in combination with a luciferase reporter plasmid and treated with LCA
Gluc I or LCA Gluc II. Quantification of luminescence revealed no activation of transcription for
the PPARα-Gal4 system suggesting that LCA Gluc I and Gluc II do not interfere with the Gal4-Luc
detection system in general and that both compounds selectively agonize VDR in comparison to
nuclear receptors like PPARα.
Transcriptional modulation mediated by VDR in the presence of cholesterol and vitamin D
metabolites was confirmed in non-transfected DU145 cells. In comparison to 1,25(OH)2D3, the
ability to induce VDR target gene CYP24A1 was quantified by qRT-PCR. Therefore, selected VDR
ligands at a concentration of 7.5 µM were incubated in the absence and presence of 20 nM
1,25(OH)2D3 for 18 hours. The results are illustrated in Figure 2.
Figure 2. A) CYP24A1 gene regulation by VDR ligands (7.5 μM) in DU145 cells after 18 hours. Stars
represent P < 0.001 (***) compared to vehicle DMSO. B) CYP24A1 gene regulation by VDR ligands (7.5
13

ACCEPTED MANUSCRIPT
μM) in DU145 in the presence of 20 nM 1,25(OH)2D3 after 18 hours. Stars represent P < 0.001 (***)
compared to 1,25(OH)2D3. All values represent standard errors of mean calculated from three biological
independent experiments performed in triplicate. All significance values were determined using ANOVA
(Dunnett’s comparison test).
As reported for colon cancer cell line SW480, a strong induction of CYP24A1 in DU145 cells
was observed in the presence of 1,25(OH)2D3, whereas the transcriptional activation in
relationship to 1,25(OH)2D3 was less pronounced for LCA acetate and LCA with 24% and 6%,
respectively [12]. For the first time, we could show that calcitroic acid, which is thought to be
the inactive final metabolic product of 1,25(OH)2D3, was able to induce VDR target gene
CYP24A1. Found in high concentration in the liver and secreted into the duodenum via the
gallbladder, calcitroic acid and lithocholic acid are likely to have similar functions [30]. Thus,
calcitroic acid because of its high concentration in the intestine and superior ability to induce
transcription in comparison to LCA, might play a very important role in the detoxification of LCA
mediated by VDR target gene CYP3A [3]. The agonistic effect of LCA Gluc I, observed in the
transcription assays, could not be recapitulated as gene up-regulation of CYP24A1 in DU145
cells, however, LCA Gluc II induced a weak induction of CYP24A1, which was more pronounced
in the presence 1,25(OH)2D3. Interestingly, none of the other VDR agonists activated VDR in a
synergistic way, thus the interaction between VDR and LCA Gluc II might in addition to VDR
binding promote RxR heterodimerization. With respect to intestinal detoxification, LCA Gluc II
in addition to LCA might signal high concentrations of LCA conjugates via VDR to induce the
transcription of enzymes that initiated their catabolism.
Conclusion
14

ACCEPTED MANUSCRIPT
The LCA-based phase 1 metabolites investigated herein bearing hydroxyl functions are not
hydrophobic enough to interact with VDR. Thus their biological and pharmacological activity is
not expected to be mediated by VDR. Furthermore, we confirmed and quantified the reported
cellular agonistic activity of LCA and LCA acetate and identified LCME as a novel VDR antagonist.
Calcitroic acid, an assumed inactivated metabolite of 1,25(OH)2D3, was shown to have potent
agonistic activity with respect to VDR. Based on its ability to promote the catabolism of
1,25(OH)2D3 in prostate cancer cells via upregulation of CYP24A1, calcitroic acid might be able
to upregulate CYP3A in the colon thus playing a major role in intestinal detoxification and
parallel mediate the vitamin D protective effect against colon cancer. Further investigations in
this area are currently being carried out and include a comprehensive evaluation of biological
functions and distribution of calcitroic acid in the gut. Finally, LCA glucuronide (LCA Gluc II) was
identified as novel endogenous VDR agonist that enables signaling of high concentrations of bile
acid conjugates. This is especially important based on the fact more than 85% of bile acids are
conjugated in the gut. The observed synergistic activation of VDR in the presence of
1,25(OH)2D3 might enable LCA Gluc II to enhance LCA signaling in the intestine and accelerate
the production of metabolic enzymes for tissue protection.
Experimental Section
Chemical Compounds and Synthesis
Reagents including lithocholic acid, lithocholic methyl ester and taurolithocholic acid were
purchase from Sigma-Aldrich while glycolithocholic acid was purchased from Santa Cruz
Biotechnology. Calcitroic acid was purchased from Toronto Research Chemicals. Synthesized
15

ACCEPTED MANUSCRIPT
compounds were either purified via recrystallization or normal phase flash chromatography (SPI
Biotage, silica gel 230-400 mesh). Molecular masses were determined with a Shimadzu 2020
LC-MS (single quadrupole) instrument using direct injection. NMR spectra were recorded on a
Bruker 500MHz of 300MHz instrument with samples diluted in either CDCl3 or DMSO-D6.
Chemical shifts are reported in δ (parts per million: ppm) by reference to the hydrogenated
residues of the deuterated solvent as an internal standard (CDCl3: δ= 7.24 ppm (1H-NMR) and
δ= 77 ppm (13C-NMR) and DMSO-D6: δ= 2.50 ppm (1H-NMR) and δ= 39.5 ppm (13C-NMR).
LCA Acetate: Lithocholic acid (0.5 g, 1.3 mmol), was dissolved in dry pyridine (10mL) under N2
gas and cooled to 0oC. 4-dimethylaminopyridine (0.02 g, 0.13 mmol) and acetyl chloride (1.1
mL, 16 mmol) were added to the solution. The reaction was stirred at room temperature and
monitored by TLC using EtOAc-Hexanes-MeOH (4:1:1, v/v/v) and cerium molybdate as a
developing stain. After 1 hour, water (3 mL) was added and the solution was evaporated under
reduced pressure. Dichloromethane (15 mL) was used to dissolve the residue, which was
subsequently washed with saturated aqueous NaCl (20 mL) and water (15 mL). The organic
phase was dried over Na2SO4, then filtered and evaporated to dryness. The crude product was
co-evaporated with toluene (10 mL x2), ethanol (10 mL x2), acetonitrile (10 mL x 2) and DCM
(10 mL x 2) in order to remove traces of pyridine and purified by silica gel chromatography with
CH3OH/CH2Cl2 mixture (0%-5% strong). Yield: 95%; 1H-NMR (300 MHz) (CDCl3) δ 4.75 (m, 1H, H-
13
3), 2.05 (s, 3H, -CH3, Ac), 0.95 (m, 6H, H-18/19, 21), 0.67 (s, 3H, H-18/19); C NMR δ 178.57,
170.72, 74.43, 56.51, 55.99, 42.75, 41.89, 40.42, 40.16, 35.79, 35.35, 35.03, 34.58, 32.25, 31.06,
29.69, 28.18, 26.32, 23.33, 20.83, 18.55, 18.28, 12.05; MS DUIS (-ve) calcd. m/z for C26H42O4
[(M)] 418.3, found [(M-1)-] 417.4.
16

ACCEPTED MANUSCRIPT
LCA Sulfonate: Sulfuric acid (0.16 mL, 3 mmol) and acetic anhydride (0.28 mL, 3 mmol) were
mixed with dry pyridine (5 mL) and after 5 minutes of stirring at 50-55oC a solution of lithocholic
acid in 2 mL of pyridine was added. The mixture was stirred for 30 minutes at the same
conditions, cooled to 0oC, and 25% ammonia water (0.74 mL) was added. After 15 minutes of
stirring, the precipitate was filtered and the filtrate was placed on a rotary evaporator for
1
concentrating. The product isolated was not purified further. Yield: 97%; H-NMR (300 MHz)
13
(DMSO-D6) δ 3.96 (m, 1H, H-3), 0.88 (m, 6H, H-18/19, 21), 0.61 (s, 3H, H-18/19); C NMR δ
175.18, 75.98, 56.41, 55.93, 42.66, 35.77, 35.47, 35.24, 24.53, 33.77, 31.21, 31.08, 28.16, 27.27,
13
26.52, 24.31, 23.65, 20.85, 18.59, 12.34; CDEPT-135 δ Negative (-) CH2: 76.05, 56.42, 55.94,
42.06, 40.33, 35.82, 35.51, 24.31, 18.59, 12.34 Positive (+) CH and CH3: 40.09, 35.29, 33.80,
31.21, 31.13, 28.16, 27.27, 26.53, 23.64, 20.85; MS DUIS (-ve) calcd. m/z for C24H40O6S [(M)]
456.3, found [(M-1)-] 455.4.
LCME O-glucuronide I: To a solution of lithocolic methyl ester (400 mg) in anhydrous benzene
(16 mL) was added cadmium carbonate (400 mg), acetobromo-α-D-glucuronic acid methyl ester
(400 mg) and a quantity of molecular sieves (400 mg). The mixture was stirred at reflux. After
1 hour and 3 hours, additional quantities of acetobromo-α-D-glucuronic acid methyl ester (200
mg) and cadmium carbonate (200 mg) were added and the mixture stirred for 7 hours and was
monitored by TLC using hexane-EtOAc-AcOH (50:50:1, v/v/v) and cerium molybdate as the
developing stain. The precipitate was removed by filtration and washed with EtOAc. The
filtrate and washings were combine and evaporated to dryness under reduced pressure and the
oily residue was recrystallized in MeOH (5mL) to make white crystals. Yield: 43%; 1H-NMR (300
MHz) (CDCl3) δ 5.28-5.24 (m,2H), 5.01-4.95 (t, 1H, J= 9Hz), 4.69-4.66 (d, 1H, J= 9Hz, anomeric),
17

ACCEPTED MANUSCRIPT
4.06-4.03 (d, 1H, J= 9Hz), 3.78 (s, 3H, Ac), 3.69 (s, 3H, Ac), 3.62 (m, 1H, H-3), 2.07, 2.04 (s, 9H,
13
COCH3), 0.92 (m, 6H, H-18/19, 21), 0.65 (s, 3H, H-18/19); C NMR δ 174.76, 170.21, 169.35,
169.28, 167.32, 99.57, 80.56, 72.61, 72.20, 71.57, 69.50, 56.28, 55.89, 52.82, 51.46, 42.70,
42.17, 40.30, 40.09, 35.81, 35.35, 35.09, 34.63, 33.99, 31.05, 30.99, 28.16, 27.09, 27.01, 26.23,
24.16, 23.35, 20.83, 20.71, 20.63, 20.51, 18.25, 12.01; MS DUIS (+ve) calcd. m/z for C38H58O12
[(M)] 706.4, found [(M+ 18 (NH4))+] 724.8.
LCA O-glucuronide I: To a solution of LCME O-glucuronide I (70 mg) in MeOH (8 mL) an aqueous
solution of 2M NaOH was added dropwise until the mixture was basic (pH 14). The reaction
was stirred at room temperature overnight and monitored by TLC using hexanes-EtOAc-AcOH
(50:50:1, v/v/v) and cerium molybdate as the developing stain. After most of the solvent was
removed by rotary evaporation, the reaction product was diluted with water, neutralized with
3M HCl and then evaporated to dryness. The residue was re-suspended in anhydrous EtOH (10
mL) and the insoluble material was filtered off and washed with EtOH. The combined filtrate
was evaporated and the residue was recrystallized from MeOH resulting in white, flakey solid.
1
Yield: 100%: H-NMR (300 MHz) (DMSO-D6) δ 4.34-4.31 (d, 1H, J= 9Hz), 3.61-3.58 (d, 1H, J=
9Hz), 3.33-3.27 (2H, t, J= 9Hz), 3.20-3.14 (1H, t, J= 9Hz), 2.93 (m, 1H, H-3), 0.88 (m, 6H, H-18/19,
21), 0.61 (s, 3H, H-18/19); 13C NMR δ 175.30, 170.86, 101.33, 77.75, 76.55, 76.00, 73.65, 72.01,
56.45, 55.97, 51.65, 42.72, 41.86, 35.80, 35.29, 34.26, 31.13, 30.83, 28.17, 27.20, 26.85, 26.55,
24.31, 26.55, 24.31, 23.49, 20.86, 18.59, 12.32 ; MS DUIS (-ve) calcd. m/z for C30H48O9 [(M)] 552,
found [(M-1)-] 551.
LCME-TBS: To a solution of LCME (3 g, 0.0077 mol) and imidazole (3.74 g, 0.0231 mol, 3 equiv.)
in dry DMF (28 mL) tert-butyldimethylsilyl chloride (3.48 g, 0.0231 mols, 3 equiv.) was added
18

ACCEPTED MANUSCRIPT
dropwise. The mixture was stirred overnight at room temperature and monitored by TLC using
EtOAc-DCM- AcOH (5:95:1, v/v/v) and cerium molybdate as the developing stain. Upon
completion, the reaction was diluted with water and extracted with DCM (25 mL x 3) and then
dried using rotary evaporation. The resulting crude product was purified using silica gel
chromatography using EtOAc-Hexanes with 1% AcOH (1% -20% strong). A white solid was
1
isolated. Yield: 84%; H-NMR (300 MHz) (CDCl3) δ 3.67 (s, 3H, Ac) 3.59 (m, 1H, H-3), 0.92 (m,
13
6H, H-18/19, 21), 0.90 (s, 9H, t-butyl-Si), 0.63 (s, 3H, H-18/19), 0.06 (s, 6H, 2CH3-Si); C NMR δ
174.90, 135.15, 121.89, 72.85, 56.39, 55.93, 51.49, 42.70, 42.27, 40.20, 40.12, 36.90, 36.49,
35.84, 35.56, 35.36, 34.57, 31.57, 31.06, 31.00, 28.18, 27.28, 26.39, 25.97, 25.74, 24.20, 23.38,
20.79, 18.33, 18.24, 12.00; MS DUIS (+ve) calcd. m/z for C31H56O3Si [(M)] 504, found [(M-
TBDMS)]+ 373 and [(M+1+imidazole)+] 574.
LCA-TBS: LCME-TBS (3.25 g, 6.45 mmol) was dissolved in THF (20 mL) and to it 2M NaOH was
added until it reached pH=10. The reaction was stirred at 75oC for 24 hours and monitored by
TLC using EtOAc-Hexanes-AcOH (3:2:1, v/v/v) and cerium molybdate as the developing stain.
Once complete, most of the solvent was removed by rotary evaporation. The reaction product
was diluted with water and acidified with 3M HCl to pH=3. The solid formed was collected and
purified by silica gel chromatography using EtOAc-Hexanes with 1% AcOH (1-20% strong). A
1
white solid with low solubility in MeOH, EtOH and CHCl3 was obtained. Yield: 64%; H-NMR
(500 MHz) (DMSO-D6) δ 8.45 (s, 1H, COOH), 3.58 (m, 1H, H-3), 0.88 (m, 6H, H-18/19, 21), 0.85
13
(s, 9H, t-butyl-Si), 0.61 (s, 3H, H-18/19), 0.02 (s, 6H, 2CH3-Si); C NMR δ 175.14, 72.56, 56.45,
56.09, 42.80, 42.05, 37.11, 35.90, 35.49, 35.24, 34.62, 31.34, 31.24, 28.11, 27.26, 26.53, 26.28,
19

ACCEPTED MANUSCRIPT
24.28, 23.62, 20.92, 18.65, 18.23, 12.36; MS DUIS (-ve) calcd. m/z for C31H54O3Si [(M)] 490.3,
found [(M-1)-] 489.4.
Tetrabutylammonium Glucuronate (1): Glucuronic acid (2.9 g) was suspended in methanol (25
mL) to which tetrabutylammonium hydroxide 30-hydrate (12 g) was added. The mixture was
stirred at room temperature for 1 hour until a clear solution resulted. The solvent was
removed in a rotary evaporator at 40oC, thereby yielding a syrup. Addition of acetone (100 mL)
resulted in precipitation of the tetrabutylammonium glucoronate. That salt was separated by
filtration and washed with acetone. The filtrate and washings, on further concentration in a
rotary evaporator, resulted in precipitation of more salt. A flakey, white solid was obtained.
1
Yield: >95%; H-NMR (500MHz) (DMSO-D6) δ 5.75- 5.72 (d, 1H, J= 9Hz), 4.50-4.58 (d, 1H, J=
6Hz), 3.86-3.58 (m, 4H,OHs), 3.19-3.14 (m, 8H), 1.57 (m, 8H), 1.37-1.25 (m, 8H), 0.96-0.91 (t,
12H, J= 7.5Hz); MS DUIS (-ve) calcd. m/z for C22H45NO7 [(M)] 435.2, found [(M-
tetrabutylammonium)-] 193.
LCA-TBS-O-glucuronide II: LCA-TBS (1.0 g, 2.04 mmol) and 1,1’-carbonyldiimidazole (0.66 g, 4.08
mmol, 2 equiv.) were dissolved in dry pyridine (40 mL). The mixture was stirred at reflux
overnight and was monitored by both TLC using EtOAc-Hexanes-AcOH (4:1:1, v/v/v) and cerium
molybdate as a developing stain and mass spectrometry. The development of a peak at 542
m/z in the positive mode indicated the formation of LCA-TBS-imidazole coupled product. To
the same reaction mixrure, tetrabutylammonium glucuronate (2.22 g, 5.1 mmol, 2.5 equiv.), dry
pyridine (10 mL), and sodium hydride (15 mg, 0.04 mmol, 0.02 equiv.) were added and the
reaction was stirred at 50oC for 5 hours. The reaction was monitored by TLC (same conditions
as above) and mass spectrometry. The development of a peak at 666 m/z in the negative
20

ACCEPTED MANUSCRIPT
mode indicated the formation of LCA-TBS-O-glucuronide II product. The reaction was stopped
by careful addition of water. After the solution was made acidic with acetic acid, the product
was extracted with EtOAc (25mL x 3). The EtOAc layer was dried over Na2SO4 and then
evaporated to dryness with a rotary evaporator. The final yellow oil was not purified and was
1
directly used for the next reaction. Yield: >95% (crude); Crude sample: H-NMR (300 MHz)
(DMSO-D6) δ 4.69-4.67 (d, 2H, J= 6Hz anomeric), 3.47 (m, 1H, H-3), 0.89 (m, 6H, H-18/19, 21),
0.83 (s, 9H, t-butyl-Si), 0.63 (s, 3H, H-18/19), 0.07 (s, 6H, 2CH3-Si); 13C NMR δ 178.36 (glucuronic
acid COOH) and 175.73 (LCA COOR) MS DUIS (-ve) calcd. m/z for C36H62O9Si [(M)] 667, found
[(M-1)-] 666.
LCA O-glucuronide II: LCA-TBS-O-glucuronide II (1.3 g, 2.0 mmol) was dissolved in a 1.0M THF
solution of tetrabutylammonium fluoride (1.66 mL, 3 equiv.). The reaction was stirred at 40oC
for 2 days and was monitored by mass spectrometry with the disappearance of the starting
material peak at 666 m/z in the negative mode. When complete, the reaction was quenched
with addition of water and washed with EtOAc (3x). The EtOAc layer was dried over MgSO4 and
dried by rotary evaporation. The crude yellow oil re-suspended in EtOH and the insoluble solid
was filtered and washed with EtOH. The filtrate was collected and dried. The residue was then
purified using silica gel chromatography with MeOH-EtOAc with 1%AcOH (0-60% strong) to
remove the LCA impurity. Fractions containing the product were dried and then recrystallized
1
with EtOH to produce a cream colored solid. Yield: 10%; H-NMR (300 MHz) (DMSO-D6) δ 5.23-
3.77 (m, 5H, sugar-ring protons), 3.46 (m, 1H, H-3), 2.28-2.22 (m, 0.88 (m, 6H, H-18/19, 21),
13
0.62 (s, 3H, H-18/19); DEPT-135 δ Negative (-) CH2: 56.48, 40.25, 36.78, 35.64, 31.11, 30.86,
28.14, 27.38, 26.64, 20.89, Positive (+) CH and CH3: 93.07, 89.76, 70.77, 70.34, 56.52, 42.01,
21

ACCEPTED MANUSCRIPT
40.81, 40.53, 40.43, 40.25, 39.98, 39.70, 35.86, 35.21, 23.74, 19.02, 18.62, 12.34. MS DUIS (-
ve) calcd. m/z for C30H48O9 [(M)] 552, found [(M-1)-] 551.
Biological Assay Protocols
Reagents. LG190178 (a VDR agonist) was synthesized following a published procedure and used
as a positive control. 3-dibutylamino-1-(4-hexyl-phenyl)-propan-1-one (a VDR-coactivator
inhibitor) was used as a second positive control and was synthesized using a previously
published method. 1,25(OH)2D3 was purchased from Endotherm. The VDR-LBDmt DNA was
provided by D. Moras[31] and cloned into pMAL-z2X vector (New England Biolabs). For a
details on the expression and purification of VDR-LBD, see ref[27]. Alexa Fluor 647 Labeled
SRC2-3 was synthesized by labeling SRC2-3 (CKKKENALLRYLLDKDDTKD) with cysteine-reactive
Alexa Fluor 647 probe followed by reverse phase HPLC using a C18 column. VDR-CMV and
CYP24A1-luciferase reporter plasmids were provide by D.D. Bikle, VP16-VDR-LBD and SRC1-
GAL4 plasmid were provided by D.J. Mangelsdorf,[3] and hVDR-Gal4 was provided by S.D.
Ayers.
VDR Binding Determined by a Fluorescence Polarization (FP) Assay. Was conducted in 384-well
black polystyrene microplates (Corning, #3573). The assay solution contained buffer (25mM
PIPES, 50mM NaCl (Fisher), and 0.01% NP-40, at pH 6.75), 0.1 μM VDR-LBD and 7.5 nM Alexa
Fluor 647-labeled SRC2-3. For competitive inhibition, 0.75 μM LG190178 was also added to the
solution. 30 mM stock solutions of synthesized compounds made in DMSO were serially diluted
(1:3), and four 14 μL aliquots of each compound concentration was transferred to opaque 384-
22

ACCEPTED MANUSCRIPT
well polypropylene plates for storage. Small volume transfers were performed by a Tecan
Freedom EVO liquid handling system using a 50H hydrophobic coated pin tool that carried 100
nL (V&P Scientific). To obtain a maximum concentration of 150 μM, 300 nl of each compound
concentration was transferred into the 20 µl assay solution. Agonistic binding was determined
in the absence of LG190178 while competitive inhibition binding was determine in the presence
of LG190187.
Fluorescence polarization was detected after 30 minutes at an
emission/excitation wavelength of 635/685 nm (Alexa Fluor 647). LG190178 and DMSO were
used as positive and negative controls in the agonistic binding assay, respectively. In the
competitive inhibition binding assay 3-dibutylamino-1-(4-hexyl-phenyl)-propan-1-one was the
positive control while DMSO was used as a negative control. Controls were measured within
each plate to determine the z’ factor, which assed the quality of the assay, and enable data
normalization. Three independent experiments were carried out in quadruplicate, and data was
analyzed using nonlinear regression with a variable slope (GraphPrism).
Transcription Assays. Human embryonic kidney (HEK) 293T cells were purchased (ATCC) and
cultured in 75 cm2 flasks (CellStar) coated in matrigel (BD Bioscience, #354234). Cells were
grown in DMEM/High Glucose (Hyclone, #SH3024301) media to which non-essential amino
acids (Hyclone, #SH30238.01), 10 mM HEPES (Hyclone, #SH302237.01), 5 x 106 units of
penicillin and streptomycin (Hyclone, #SV30010), and 10% of heat inactivated fetal bovine
serum (Gibco, #10082147) were added. For the assay, cells at 70-80% confluency were
transfected by lipid-based methods where 2 mL of untreated DMEM/High Glucose media
(without additives) containing: 1) 2-Hybrid assay: 5.0 μg of VP16-VDR-LBD plasmid, 4.0 μg of
23

ACCEPTED MANUSCRIPT
SRC1-GAL4 plasmid, 16 μg of a luciferase reporter plasmid or 2) transcription assay: 1.5 µg of
hVDR-Gal4 plasmid and 16 µg of a 4xUAS luciferase reporter plasmid, LipofectamineTM LTX (75
μl, Life Technologies, #15338020), and PLUSTM reagent (25 μl) was added to the flask. After 16
hours of incubation at 37oC with 5% CO2, the cells were harvested with 3mL of 0.05% Trypsin
(Hyclone, #SH3023601), added to 10mL of the assay buffer, DMEM/High Modified buffer
without phenol red (Hyclone, #SH30284.01) containing all the above mentioned additives plus
10 mM sodium pyruvate and 2% percent charcoal treated FBS (Invitrogen, #12676-011) instead
of HI FBS, and spun down for 2 minutes at 1000 rpm. The media was removed and cells were
resuspended in the assay buffer, DMEM/High Modified buffer without phenol red. Prior to
adding cells to sterile white, optical bottom 384-well plates (NUNC, #142762), plates were
treated with 20 μL per well of a 0.25% matrigel solution. To each well, 20 μL of cells were
added to yield a final concentration of 15,000 cells per well. After 4 hours, plated cells were
treated with 100 nL of small molecules (1:3 serial dilution, with maximum final concentration at
150 µM) and controls (DMSO and 10 nM 1,25(OH)2D3 ) which were added using a Tecan
Freedom EVO liquid handling system with a 50H hydrophobic coated pin tool. In the
competitive inhibition assay, 1,25(OH)2D3 (10 nM) was also added to the assay wells containing
small molecule. After 16 hours of incubation at 37oC with 5% CO2, 20μL of Bright-Glo™
Luciferase Assay Kit (Promega, Madison, WI) was added to each well and luminescence was
read. Controls were measured within each plate to determine the z’ factor and to enable data
normalization. Three independent experiments were performed in quadruplicate and data was
analyzed using nonlinear regression with variable slope (GraphPrism).
24

ACCEPTED MANUSCRIPT
Cytotoxicity Assay. Hek293T cells were plates according to the transcription assay protocol and
100 nL of serially diluted (1:3 in DMSO) small molecules (final maximum concentration at 150
µM) were transferred using a Tecan Freedom EVO liquid handling system with a 50H
hydrophobic coated pin tool. The controls for the cytotoxicity assay used were 3-dibutylamino-
1-(4-hexyl-phenyl)-propan-1-one (150 μM in DMSO, positive) and DMSO (negative). After 18
hours, assay plates were evaluated by adding 20 µL of Cell Titer-Glo™ Luminescence Assay Kit
(Promega, Madison, WI) to each well and reading luminescence on a Tecan Infinite M1000
plate reader. Controls were measured within each plate to determine the z’ factor and to
enable data normalization. Three independent experiments were performed in quadruplicate
and data was analyzed using nonlinear regression with variable slope (GraphPrism).
Semi-quantitative Real Time-PCR. Reagents and Instrumentation: The prostate cancer cell line,
DU145, was purchased (ATCC) and in cultured in 75 cm2 flasks (CellStar). Cells were grown in
DMEM/High Glucose media to which non-essential amino acids, 10 mM HEPES, 5 x 106 units of
penicillin and streptomycin, and 10% of heat inactivated fetal bovine serum were added. Cells
at 80-90% confluency were harvested using 0.25% Trypsin, spun down in assay buffer, DMEM
media without phenol red containing the above mentioned additives (section 4.2.5), at 1000
rpm for 2 minutes. The media was removed and the cells were resuspended in the assay buffer
and transferred to 6 well plates coated in matrigel. Plates were incubated at 37oC overnight to
allow the cells to settle and adhere to the plate. The next day, DMSO (0.03%), test compound
(7.5 μM, LCA O-glucuronide I, LCA O-glucuronide II, Calcitroic acid, LCA) and 1,25(OH)2D3 (20
nM) were added to their own well in the 6-well plate and incubated at 37°C for 18 hours. Small
25

ACCEPTED MANUSCRIPT
molecules were either in the presence or absence of 1,25(OH)2D3. Cells were harvested
according to literature using a QIAshredder (Qiagen) to lyse the cells and a RNAeasy kit (Qiagen)
to isolate total RNA for each compound. RNA concentration was determine by UV at 260nm
using the Tecan Infinite M1000 plate reader. A QuantiFast SYBR Green RT-PCR Kit (Qiagen) was
used for the real time PCR following manufacturer’s recommendations. Primers used in these
studies are as follows: GAPDH FP 5’-ACCACAGTCCATGCCATCAC-3’, GAPDH RP 5’-
TCCACCACCCTGTTGCTGTA-3’; CYP24A1 FP 5’-CTTTGCTTCCTTTTCCCAGAAT-3’; CYP24A1 RP 5’-
CGCCGTAGATGTCACCAGTC-3’; Real-time rt-PCR was carried out on a Mastercycler (Eppendorf).
The ∆∆Ct method was used to measure the fold change in gene expression of target gene,
CYP24A1. Standard errors of mean were calculated from two biological independent
experiments performed in triplicates. Significance was calculated using Dunnett ANOVA in
GraphPrism.
Acknowledgement
This work was supported by the University of Wisconsin-Milwaukee, the University of
Wisconsin Milwaukee Research Growth Initiative, the National Institutes of Health [R03
DA031090], the University of Wisconsin Milwaukee Research Foundation (Catalyst grant), the
Lynde and Harry Bradley Foundation, the Richard and Ethel Herzfeld Foundation, and National
Institutes of Health [RO1 AR050023].
References
[1] D.J. Mangelsdorf, C. Thummel, M. Beato, P. Herrlich, G. Schutz, K. Umesono, B. Blumberg, P. Kastner,
M. Mark, P. Chambon, R.M. Evans, The Nuclear Receptor Superfamily: The Second Decade, Cell, 83
(1995) 835-839.
[2] C.K. Glass, M.G. Rosenfeld, The coregulator exchange in transcriptional functions of nuclear
receptors, Genes Dev, 14 (2000) 121-141.
26

ACCEPTED MANUSCRIPT
[3] M. Makishima, T.T. Lu, W. Xie, G.K. Whitfield, H. Domoto, R.M. Evans, e. al, Vitamin D receptor as an
intestinal bile acid sensor, Science, 296 (2002) 1313-1316.
[4] D.W. Russell, The enzymes, regulation and genetics of bile acid synthesis, Annual Review of
Biochemistry, 72 (2003) 137-174.
[5] J.M. Ridlon, D.-J. Kang, P.B. Hylemon, Bile salt biotransformations by human intestinal bacteria, J.
Lipid Res., 47 (2006) 241-259.
[6] I. Kim, K. Morimura, Y. Shah, Q. Yang, J.M. Ward, F.J. Gonzalez, Spontaneous hepatocarcinogensis in
farnesoid X receptor-null mice, Carcinogenesis, 28 (2007) 940-946.
[7] F.M. Nagengast, M.J. Grubben, I.P.V. Munster, Role of biles acids in colorectal carcinogenesis, Eur. J.
Cancer, 7-8 (1995) 1067-1070.
[8] J. Cheng, Z.-Z. Fang, J.-H. Kim, K.W. Krausz, N. Tanaka, J.Y.L. Chiang, F.J. Gonzalez, Intestinal CYP3A4
protects against lithocholic acid-induced hepatoxicity in instestine specific-deficient mice, Journal of
Lipid Research, 55 (2014) 455-465.
[9] H. Masuno, T. Ikura, D. Morizono, I. Orita, S. Yamada, M. Shimizu, N. Ito, Crystal structures of
complexes of vitamin D receptor ligand-bindig domain with lithocholic acid derivatives, J. Lipid Res., 54
(2013) 2206-2213.
[10] A.Y. Belorusova, J. Eberhardt, N. Potier, R.H. Stote, A. Dejaegere, N. Rochel, Structural insights into
the molecular mechanisms of vitamin D receptor activation by lithocholic acid involving a new mode of
ligand recognition, J. Med. Chem., 57 (2014) 4710-4719.
[11] R. Adachi, Y. Homna, H. Masuno, K. Kawana, I. Shimomura, S. Yamada, M. Makishima, Selective
activation of vitamin D receptor by lithocholic acid acetate, a bile acid derivative, J. Lipid Res., 46 (2005)
46-57.
[12] M. Ishizawa, M. Matsunawa, R. Adachi, S. Uno, K. Keda, H. Masuno, M. Shimizu, K.I. Iwasaki, S.
Yamada, M. Makishima, Lithocholic acid derivatives act as selective vitamin D receptor modulators
without inducing hypercalcemia, Journal of Lipid Research, 49 (2008) 763-772.
[13] Y. Tanaka, L. Castillo, H.F. Deluca, The 24-hydroxylation of 1,25-dihydroxyvitamin D3, J. Biol. Chem,
252 (1977) 1421-1424.
[14] T. Sakaki, N. Sawada, K. Komai, S. Shiozawa, S. Yamada, K. Yamamoto, Y. Ohyama, K. Inouye, Dual
metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24, Eur. J. Biochem., 267 (2000)
6158-6165.
[15] S. Ishizuka, S. Ishimoto, A.W. Norman, Biological activity assessment of 1 alpha,25-dihydroxyvitamin
D3-26,23-lactone in the rat, J Steroid Biochem, 20 (1984) 611-615.
[16] B.Y. Bao, H.J. Ting, J.W. Hsu, Y.F. Lee, Protective role of 1 alpha, 25-dihydroxyvitamin D3 against
oxidative stress in nonmalignant human prostate epithelial cells, Int J Cancer, 122 (2008) 2699-2706.
[17] G.J. Miller, G.E. Stapleton, T.E. Hedlund, K.A. Moffat, Vitamin D receptor expression, 24-hydroxylase
activity, and inhibition of growth by 1alpha,25-dihydroxyvitamin D3 in seven human prostatic carcinoma
cell lines, Clin Cancer Res, 1 (1995) 997-1003.
[18] H. Jackson, M. Solaymani-Dodaran, T.R. Card, G.P. Aithal, R. Logan, J. West, Influence of
ursodeoxycholic acid on the mortality and malignancy associated with primary biliary cirrhosis: a
population-based cohort study, Hepatology, 46 (2007) 1131-1137.
[19] K. Berendse, M. Engelen, S. Ferdinandusse, C.B. Majoie, H.R. Waterham, F.M. Vaz, J.H. Koelman,
P.G. Barth, R.J. Wanders, B.T. Poll-The, Zellweger spectrum disorders: clinical manifestations in patients
surviving into adulthood, J Inherit Metab Dis, (2015).
[20] B. Ascher, K. Hoffmann, P. Walker, S. Lippert, U. Wollina, B. Havlickova, Efficacy, patient-reported
outcomes and safety profile of ATX-101 (deoxycholic acid), an injectable drug for the reduction of
unwanted submental fat: results from a phase III, randomized, placebo-controlled study, J Eur Acad
Dermatol Venereol, 28 (2014) 1707-1715.
27

ACCEPTED MANUSCRIPT
[21] J.H. Iser, H. Dowling, H.Y. Mok, G.D. Bell, Chenodeoxycholic acid treatment of gallstones. A follow-
up report and analysis of factors influencing response to therapy, N Engl J Med, 293 (1975) 378-383.
[22] S. Bureeva, J. Andia-Pravdivy, A. Symon, A. Bichucher, V. Moskaleva, V. Popenko, A. Shpak, V.
Shvets, L. Kozlov, A. Kaplun, Selective inhibition of the interaction of C1q with immunoglobulins and the
classical pathway of complement activation by steroid and triterpenoids sulfates, Bioorg Med Chem, 15
(2007) 2489-2498.
[23] A.E. Sobel, P.E. Spoerri, Steryl Sulfates. I. Preparation and Properties, J. Am. Chem. Soc., 63 (1941)
1259-1261.
[24] T. Iida, S. Tazawa, Y. Ohshima, T. Niwa, J. Goto, T. Nambara, Analysis of conjugated bile acids in
human biological fluids. Synthesis of hyodeoxycholic acid 3- and 6-glycosides and related compounds,
Chem. Pharm. Bull, 42 (1994) 1479-1484.
[25] B. Becker, A.B. Barua, J.A. Olsen, All-trans-retinoyl B-glucuronide: new procedure for chemical
synthesis and its metabolism in vitamin A-deficient rats, Biochem J., 314 (1996) 249-252.
[26] P. Nandhikonda, W.Z. Lynt, M.M. McCallum, T. Ara, A.M. Baranowski, N.Y. Yuan, D. Pearson, D.D.
Bikle, R.K. Guy, L.A. Arnold, Discovery of the first irreversible small molecule inhibitors of the interaction
between the vitamin D receptor and coactivators, J Med Chem, 55 (2012) 4640-4651.
[27] A. Teichert, L.A. Arnold, S. Otieno, Y. Oda, I. Augustinaite, T.R. Geistlinger, R.W. Kriwacki, R.K. Guy,
D.D. Bikle, Quantification of the vitamin D receptor-coregulator interaction, Biochemistry, 48 (2009)
1454-1461.
[28] K. Teske, P. Nandhikonda, J.W. Bogart, B. Feleke, P. Sidhu, N. Yuan, J. Preston, R. Goy, L.A. Arnold,
Modulation of Transcription mediated by the Vitamin D Receptor and the Peroxisome Proliferator-
Activated Receptor delta in the presence of GW0742 analogs, J Biomol Res Ther, 3 (2014).
[29] T.C. Northfield, I. McColl, Postprandial concentrations of free and conjugated bile acids down the
length of the normal human small intestine, Gut, 14 (1973) 513-518.
[30] R.P. Esvelt, H.K. Schnoes, H.F. DeLuca, Isolation and characterization of 1 alpha-hydroxy-23-
carboxytetranorvitamin D: a major metabolite of 1,25-dihydroxyvitamin D3, Biochemistry, 18 (1979)
3977-3983.
[31] N. Rochel, J.M. Wurtz, A. Mitschler, B. Klaholz, D. Moras, The crystal structure of the nuclear
receptor for vitamin D bound to its natural ligand, Mol Cell, 5 (2000) 173-179.
28