Chiral Inhibition of Rivaroxaban Derivatives Towards UDP-Glucuronosyltransferase (UGT) Isoforms

Article abstract of DOI:10.1002/chir.22505

Rivaroxaban is an oral direct factor Xa (FXa) inhibitor clinically used to prevent and treat thromboembolic disorders. Drug-drug interaction (DDI) exist for rivaroxaban and the inhibitors of CYP3A4/5. This study aims to investigate the inhibition of rivaroxaban and its derivatives with a chiral center towards UDP-glucuronosyltransferases (UGTs). Chemical synthesis was performed to obtain rivaroxaban derivatives with different chiral centers. UGTs supersomes-catalyzed 4-methylumbelliferone (4-MU) glucuronidation was employed to evaluate the inhibition potential towards various UGT isoforms. A significant influence of rivaroxaban derivatives towards UGT1A3 was observed. Chiral centers produce different effects towards the effect of four pairs of rivaroxaban derivatives towards UGT1A3 activity, with stronger inhibition potential of S1 than R1, but stronger inhibition capability of R2, R3, R4 than S2, S3, and S4. Competitive inhibition of R3 and R4 towards UGT1A3 was demonstrated by Dixon and Lineweaver-Burk plots. In conclusion, the significant influence of rivaroxaban derivatives towards UGT1A3's activity was demonstrated in the present study. The chirality centers highly affected the inhibition behavior of rivaroxaban derivatives towards UGT1A3. Chirality 27:936-943, 2015.

Full text of DOI:10.1002/chir.22505

CHIRALITY 27:936–943 (2015)
Chiral Inhibition of Rivaroxaban Derivatives Towards UDP-
Glucuronosyltransferase (UGT) Isoforms
ZHUHUA YAO,1,† YONG-ZHE LIU, AI-LUN MA, SHU-FEN WANG, DAN LU, CUI-MIN HU, YAN-YAN ZHANG, HAINA WANG,
3,†
2
3
5
6
7
8
9
2,4
3
3
LINGYUN HU, JUN DENG, * KUN YANG, * AND ZHONG-ZE FANG *
1
Department of Cardiology, Tianjin Union Medicine Centre, 300121, Tianjin, People’s Republic of China
2
School of Pharmaceutical Science and Technology, Key Laboratory for Modern Drug Delivery & High-Efficiency, Tianjin University, Tianjin, People’s
Republic of China
3
Department of Toxicology, School of Public Health, Tianjin Medical University, Tianjin, People’s Republic of China
4
Collaborative Innovation Center of Chemical Science and Engineering, Tianjin University, Tianjin, People’s Republic of China
5
Department of Immunology, Tianjin Key Laboratory of Cellular and Molecular Immunology, Tianjin Medical University, Tianjin, People’s
Republic of China
6
7
Tianjin Life Science Research Center, Department of Microbiology, School of Basic Medical Sciences, Tianjin Medical University, Tianjin, People’s
Republic of China
Joint Center for Translational Medicine, Dalian Institute of Chemical Physics, Chinese Academy of Sciences and First Affiliated Hospital of Liaoning
Medical University, Dalian, People’s Republic of China
College of Pharmaceutical Sciences, Shandong University, Jinan, People’s Republic of China
8
9
Shandong Cancer Hospital and Institute, Shandong, People’s Republic of China
ABSTRACT
Rivaroxaban is an oral direct factor Xa (FXa) inhibitor clinically used to prevent
and treat thromboembolic disorders. Drug–drug interaction (DDI) exist for rivaroxaban and the
inhibitors of CYP3A4/5. This study aims to investigate the inhibition of rivaroxaban and its deriv-
atives with a chiral center towards UDP-glucuronosyltransferases (UGTs). Chemical synthesis
was performed to obtain rivaroxaban derivatives with different chiral centers. UGTs
supersomes-catalyzed 4-methylumbelliferone (4-MU) glucuronidation was employed to evaluate
the inhibition potential towards various UGT isoforms. A significant influence of rivaroxaban deriv-
atives towards UGT1A3 was observed. Chiral centers produce different effects towards the effect
of four pairs of rivaroxaban derivatives towards UGT1A3 activity, with stronger inhibition potential
of S1 than R1, but stronger inhibition capability of R2, R3, R4 than S2, S3, and S4. Competitive in-
hibition of R3 and R4 towards UGT1A3 was demonstrated by Dixon and Lineweaver-Burk plots. In
conclusion, the significant influence of rivaroxaban derivatives towards UGT1A3’s activity was
demonstrated in the present study. The chirality centers highly affected the inhibition behavior
of rivaroxaban derivatives towards UGT1A3. Chirality 27:936–943, 2015. © 2015 Wiley Periodicals, Inc.
KEY WORDS: rivaroxaban; UDP-glucuronosyltransferases (UGTs); drug-drug interaction;
chirality
Rivaroxaban is an oral direct factor Xa (FXa) inhibitor, and
has been clinically used to prevent and treat thromboembolic
disorders.1 Based on the structural basis of rivaroxaban,
many efficient derivatives were prepared to investigate the
structure–activity relationship for antithrombotic therapies.2
Rivaroxaban has one (S)-type chiral center, and different
types of chiral centers have been introduced into the struc-
tures of rivaroxaban derivatives. The (S)-enantiomer has
stronger therapeutic efficacy, and the (R)-enantiomer and its
derivatives have not served as drugs, but might be useful in
the new therapeutic utilization of rivaroxaban.
diarrhea through catalyzing the glucuronidation of SN-38,
which is the active substance of irinotecan. The inhibition
7
of the SN-38 glucuronidation process by xenobiotics can re-
sult in the elevation of the toxicity intensity of irinotecan.8
The inhibition towards the UGTs-catalyzed reaction can also
disrupt the metabolism of endogenous substances, such as
the influence of indinavir and sorafenib towards the metabolism
of bilirubin.11 The inhibition of rivaroxaban and its derivatives
towards UGTs’ activity was investigated in the present study.
Drugs with different chiral centers will exert varied phar-
macokinetic properties. For example, terazosin enantiomers
exhibited different pharmacokinetic behaviors in healthy
Chinese male subjects.12 A completely different metabolic
pathway was elucidated for praziquantel (PZQ), which is a
–10
Rivaroxaban can be rapidly absorbed after oral administra-
tion. Cytochrome P450 (CYP)-catalyzed oxidation and CYP-
independent hydrolytic reaction have been demonstrated to
3
be the major metabolic pathway of rivaroxaban. The major
involved CYP isoforms contain CYP3A4/5 and CYP2J2.4
*Correspondence to: Jun Deng, School of Pharmaceutical Science and Tech-
nology, Key Laboratory for Modern Drug Delivery & High-Efficiency, Tianjin
University, Tianjin, 300072 People’s Republic of China. E-mail: jdeng@tju.
edu.cn
Drug–drug interaction was highly speculated to be involved4
between rivaroxaban and the strong inhibitors of CYP3A4/5.
Human UDP-glucuronosyltransferases (UGTs) are impor-
tant phase II drug-metabolizing enzymes involved in the
Kun Yang or Zhong-Ze Fang, Department of Toxicology, School of Public
Health, Tianjin Medical University, Tianjin, China. E-mail: yangkun@tijmu.
edu.cn or fangzhongze@tijmu.edu.cn
5
†
glucuronidation of drugs or their phase I metabolites. The
These two authors equally contributed to this work.
UGTs-catalyzed glucuronidation reaction plays a key role in6
detoxification of endogenous and exogenous substances.
For example, intestinal UGT1A1 protects irinotecan-induced
Received for publication 17 April 2015; Accepted 19 July 2015
DOI: 10.1002/chir.22505
Published online 1 October 2015 in Wiley Online Library
(wileyonlinelibrary.com).
©
2015 Wiley Periodicals, Inc.

RIVAROXABAN’S INHIBITION TOWARDS UGTS
937
prevalent drug to treat schistosomiasis.10 Drug metabolism-
related drug–drug interaction can also be affected by chiral
properties. For example, the experiment performed by Sun
et al. showed the strong stereoselective interaction between
+
spectrum (ESI): m/e (% relative intensity) 436.8 (100) (M+H) ; HRMS
+
(ESI): m/z calcd for C19H18O5N3SCl (M+H) 436.0734.
N-({(5S)-2-oxo-3-[4-(3-oxomorpholin-4-yl)-phenyl]-1,3-oxazolidin-
-yl}methyl)-thiophene-2-carboxamide (S2). Compound S2 was
synthesized according to the procedure described for S1, yield: 48%;
mp: 200–203°C; Rf = 0.70 (MeOH:CH Cl = 1:10); [α]D23 = –44.62 [c
.1 MeOH]; 1H NMR (400 MHz, CDCl ) δ 3.70–3.77 (m, 3H), 3.83–3.89
(m, 2H), 4.02–4.10 (m, 3H), 4.33 (s, 2H), 4.85 (br, 1H), 6.77 (t, J = 5.98
Hz, 1H), 7.07 (t, J = 4.36 Hz, 1H), 7.31–7.33 (d, J = 8.88 Hz, 2H), 7.49–
7.51 (dd, J = 0.78, 4.96 Hz, 1H), 7.54–7.56 (d, J = 8.92 Hz); 13C NMR
5
tetrahydropalmatine enantiomers and cytochrome P450
3
(
(
CYP) isoforms.1 Our recent work also demonstrated that
2
2
S)-carprofen exhibited stronger inhibition than (R)-carprofen
0
3
towards the activity of phase II DME UDP-glucuronosyl-
1
4
transferase (UGT) 2B7.
The aim of this study was to investigate the inhibition of
rivaroxaban and its derivatives towards UGT isoforms. The
compounds with different chiral centers were synthesized
and the inhibition potential was determined using recombi-
nant UGTs-catalyzed 4-methylumbelliferone (4-MU) glucuro-
nidation reaction.
(
100 MHz, CDCl3) δ 42.3, 47.7, 49.6, 64.1, 68.5, 71.9, 119.1, 126.2, 127.8,
28.7, 130.8, 136.6, 137.2, 154.3, 162.6, 166.8; mass spectrum (ESI): m/e
% relative intensity) 423.8 (100) (M+Na) ; HRMS (ESI): m/z calcd for
1
(
+
C19H19O5N3S (M+Na)+ 424.0943.
5
-Bromo-N-({(5S)-2-oxo-3-[4-(3-oxomorpholin-4-yl)-phenyl]-1,3-
oxazolidin-5-yl}methyl)-thiophene-2-carboxamide (S3). Compound
S3 was synthesized according to the procedure described for S1, yield:
MATERIALS AND METHODS
2 2
37%; mp: 205–208°C; Rf = 0.70 (MeOH:CH Cl = 1:10); [α]D23 = –20.00 [c
Materials
0
.1 MeOH]; 1H NMR (400 MHz, CDCl ) δ 3.65–3.85 (m, 5H), 4.02–4.05
3
4
-Methylumbelliferone (4-MU), 4-methylumbelliferone-β-D-glucuro-
nide (4-MUG), Tris-HCl, 7-hydroxycoumarin, and uridine-5′-diphosphoglu-
curonic acid trisodium salt (UDPGA) were purchased from Sigma-Aldrich
St. Louis, MO). Recombinant human UGT isoforms (UGT1A1, UGT1A3,
UGT1A6, UGT1A7, UGT1A8, UGT1A9, UGT1A10, UGT2B7) expressed
in baculovirus-infected insect cells were obtained from BD Gentest
Woburn, MA). All other reagents were of high-performance liquid chro-
(m, 3H), 4.33 (s, 2H), 4.82 (br, 1H), 6.96 (t, J = 5.86 Hz, 1H), 7.00–7.01 (d,
J = 3.96 Hz, 1H), 7.30 (t, J = 6.88 Hz, 3H), 7.52–7.55 (d, J = 8.88 Hz, 2H),
3
7.54–7.56 (d, J = 8.92 Hz); 13C NMR (100 MHz, CDCl ) δ 42.4, 47.6, 49.7,
(
6
4.1, 68.5, 71.8, 119.1, 126.3, 128.7, 130.8, 136.6, 137.3, 154.3, 161.6, 166.9;
+
mass spectrum (ESI): m/e (% relative intensity) 503.4 (100) (M+Na) ;
HRMS (ESI): m/z calcd for C19H18O5N3SBr (M+Na) 502.0048.
+
(
matography (HPLC) grade or of the highest grade commercially available.
6
-Chloro-N-({(5S)-2-oxo-3-[4-(3-oxomorpholin-4-yl)phenyl]-1,3-
oxazolidin-5-yl}methyl)pyridine-2-carboxamide (S4). Compound
Chemical Synthesis of Rivaroxaban and Its Derivatives
S4 was synthesized according to the procedure described for S1, yield: 27%;
mp: 117–120°C; Rf = 0.70 (MeOH:CH
MeOH]; 1H NMR (400 MHz, CDCl ) δ 3.72–3.96 (m, 5H), 4.03 (t, J = 4.96
Hz, 2H), 4.12 (t, J = 8.9 Hz, 1H) 4.33 (s, 2H), 4.89 (br, 1H), 7.31–7.33 (d, J
2 2
Cl = 1:10); [α]D23 = –72.69 [c 0.1
All synthesis reactions were performed in flame-dried glassware under
a nitrogen atmosphere. Solvents were distilled prior to use. Reagents
were purchased from J&K Scientific (China), Beijing Ouhe (China),
Aldrich (Milwaukee, WI), Acros (Somerville, NJ), Alfa Aesar (Ward Hill,
MA), or TCI (Portland, OR) unless otherwise noted. Chromatographic
3
=
2
8.8 Hz, 2H), 7.47–7.49 (d, J = 7.92 Hz, 1H), 7.55–7.58 (d, J = 8.84 Hz,
H), 7.82 (t, J = 7.76 Hz, 1H), 8.08–8.10 (d, J = 7.52 Hz), 8.29 (t, J = 6.2
1
13
3
Hz); 13C NMR (100 MHz, CDCl ) δ 42.2, 47.8, 49.6, 53.4, 64.1, 68.5, 71.7,
separations were performed using Kangbino 48-75 Å SiO
NMR spectra were obtained on 400 MHz Bruker Avance (Billerica,
MA) spectrometers using CDCl with tetramethylsilane (TMS) or
residual solvent as standard unless otherwise noted. Melting points were
determined using Laboratory Devices MEL-TEMP and were
2
.
H and
C
1
1
19.0, 121.1, 126.1, 127.5, 136.7, 137.2, 140.0, 149.4, 150.3, 154.2, 163.9,
66.8; mass spectrum (ESI): m/e (% relative intensity) 452.7 (100)
3
+
+
(M+Na) ; HRMS (ESI): m/z calcd for C20H19O5N3SCl (M+Na) 453.0942.
a
5
-Chloro-N-({(5R)-2-oxo-3-[4-(3-oxomorpholin-4-yl)-phenyl]-1,3-o
uncorrected/calibrated. Thin-layer chromatography (TLC) analysis was
performed using Kangbino glass-backed plates (60 Å, 250 μm) and
visualized using UV and KMnO stains. Low-resolution mass spectra
4
were obtained using an Agilent (Palo Alto, CA) 1100 series LS/MSD.
High-resolution mass spectra were obtained using a Q-TOF micro
xazolidin-5-yl}methyl)-thiophene-2-carboxamide (R1). Compound
R1 was synthesized according to the procedure described for S1, yield:
2 2
34%; mp: 233–235°C; Rf = 0.71 (MeOH:CH Cl = 1:10); [α]D23 = 76.67 [c
0
.1 MeOH]; 1H NMR (400 MHz, CDCl ) δ 3.63–3.84 (m, 5H), 4.02–4.06
3
(
m, 3H), 4.33 (s, 2H), 4.81 (br, 1H), 6.85–6.86 (d, J = 4.0 Hz, 1H), 7.04 (t,
J = 5.96 Hz, 1H), 7.33 (t, J = 6.84 Hz, 3H), 7.52–7.54 (d, J = 8.92 Hz, 2H);
3C NMR (100 MHz, CDCl ) δ 42.4, 47.6, 49.7, 64.1, 68.5, 71.8, 119.1,
(
1
Bruker) spectrometer. Compounds 4-{4-[(5R)-5-(Aminomethyl)-2-oxo-
,3-oxazolidin-3-yl]phenyl}morpholin -3-one and 4-{4-[(5S)-5-(Aminome-
1
3
thyl)-2-oxo-1,3-oxazolidin-3-yl]phenyl}morpholin-3-one were synthesized
according to the literature procedure.2
126.3, 127.1, 127.9, 136.0, 136.5, 136.6, 137.2, 154.4, 161.7, 166.9; mass
spectrum (ESI): m/e (% relative intensity) 436.8 (100) (M+H)+; HRMS
(
ESI): m/z calcd for C19H18O5N3SCl (M+H)+ 436.0734.
5
-Chloro-N-({(5S)-2-oxo-3-[4-(3-oxomorpholin-4-yl)-phenyl]-1,3-
oxazolidin-5-yl}methyl)-thiophene-2-carboxamide (S1). To
solution of 5-chlorothiophene-2-carboxylic acid (13.3 mg, 0.082 mmol) in
CH Cl (2.0 mL) was added DCC (28 mg, 0.136 mmol) at room tempera-
ture. The resulting mixture was stirred for 2 h. After which 4-{4-[(5S)-5-
aminomethyl)-2-oxo-1,3-oxazolidin-3-yl]phenyl}morpholin-3-one (14.5 mg,
.05 mmol) was added. Then the mixture was stirred for 6 h at room tem-
perature. The mixture was then quenched with H O and extracted with
CH Cl (5 mL × 3). The combined organic phase was dried over Na SO
and concentrated under reduced pressure. The resulting crude product
was purified via flash column chromatography (SiO ; isocratic eluent: 3%
MeOH in CH Cl ) to provide S1 as a white solid in 36% yield (8.0 mg), yield:
6%; mp: 233–235°C; R = 0.71 (MeOH:CH Cl = 1:10); [α]D23 = –33.75 [c
.1 MeOH]; H NMR (400 MHz, CDCl ) δ 3.63–3.84 (m, 5H), 4.02–4.06
a
N-({(5R)-2-oxo-3-[4-(3-oxomorpholin-4-yl)-phenyl]-1,3-oxazolidin-
5-yl}methyl)-thiophene-2-carboxamide (R2). Compound R2 was
synthesized according to the procedure described for S1, yield: 57%;
mp: 200–203°C; Rf = 0.70 (MeOH:CH Cl = 1:10); [α]D23 = –44.62 [c
0.1 MeOH]; 1H NMR (400 MHz, CDCl3) δ 3.70–3.77 (m, 3H), 3.83–3.89
(m, 2H), 4.02–4.10 (m, 3H), 4.33 (s, 2H), 4.85 (br, 1H), 6.77 (t, J = 5.98
Hz, 1H), 7.07 (t, J = 4.36 Hz, 1H), 7.31–7.33 (d, J = 8.88 Hz, 2H), 7.49–
7.51 (dd, J = 0.78, 4.96 Hz, 1H), 7.54–7.56 (d, J = 8.92 Hz); 13C NMR
(100 MHz, CDCl3) δ 42.3, 47.7, 49.6, 64.1, 68.5, 71.9, 119.1, 126.2, 127.8,
128.7, 130.8, 136.6, 137.2, 154.3, 162.6, 166.8; mass spectrum (ESI): m/e
2
2
2
2
(
0
2
2
2
2
4
2
+
2
2
(% relative intensity) 423.8 (100) (M+Na) ; HRMS (ESI): m/z calcd for
C19H19O5N3S (M+Na) 424.0943.
+
3
0
f
2
2
1
3
(m, 3H), 4.33 (s, 2H), 4.81 (br, 1H), 6.85–6.86 (d, J = 4.0 Hz, 1H), 7.04 (t,
5-Bromo-N-({(5R)-2-oxo-3-[4-(3-oxomorpholin-4-yl)-phenyl]-1,3-
J = 5.96 Hz, 1H), 7.33 (t, J = 6.84 Hz, 3H), 7.52–7.54 (d, J = 8.92 Hz, 2H);
oxazolidin-5-yl}methyl)-thiophene-2-carboxamide
pound R3 was synthesized according to the procedure described for
S1,: yield: 42%; mp: 205–208°C; Rf = 0.70 (MeOH:CH Cl = 1:10);
Chirality DOI 10.1002/chir
(R3). Com-
1
3
3C NMR (100 MHz, CDCl ) δ 42.4, 47.6, 49.7, 64.1, 68.5, 71.8, 119.1,
1
26.3, 127.1, 127.9, 136.0, 136.5, 136.6, 137.2, 154.4, 161.7, 166.9; mass
2
2

9
38
MA ET AL.
) δ 3.65–3.85
[
(
1
α]D23 = –20.00 [c 0.1 MeOH]; 1H NMR (400 MHz, CDCl
m, 5H), 4.02–4.05 (m, 3H), 4.33 (s, 2H), 4.82 (br, 1H), 6.96 (t, J = 5.86 Hz,
H), 7.00–7.01 (d, J = 3.96 Hz, 1H), 7.30 (t, J = 6.88 Hz, 3H), 7.52–7.55 (d,
J = 8.88 Hz, 2H), 7.54–7.56 (d, J = 8.92 Hz); 13C NMR (100 MHz, CDCl
δ 42.4, 47.6, 49.7, 64.1, 68.5, 71.8, 119.1, 126.3, 128.7, 130.8, 136.6, 137.3,
54.3, 161.6, 166.9; mass spectrum (ESI): m/e (% relative intensity) 503.4
3
For better understanding the molecular interactions between ligand
and protein, the docking between flexible small molecule and rigid pro-
tein was performed using Autodock v. 4.2. Eight compounds (S1, R1,
S2, R2, S3, R3, S4, R4) were docked into the UGT1A3 enzyme. The non-
polar hydrogen atoms of the UGT1A3 enzyme were merged, and Kollman
charges were then added to the protein structure using the
AutoDockTool. Gasteiger partial charges were assigned to the eight
inhibitors. The grid points for Autogrid calculations were set to
3
)
1
+
+
(
100) (M+Na) ; HRMS (ESI): m/z calcd for C19H18O5N3SBr (M+Na)
5
02.0048.
1
00*100*100, which was large enough to cover the entire ligand-binding
6
-Chloro-N-({(5R)-2-oxo-3-[4-(3-oxomorpholin-4-yl)phenyl]-1,3-
site and accommodate ligands to move freely. The grid point spacing for
grid-box was 0.375 Å. The Lamarckian genetic algorithm (LGA) was used
for protein-fixed ligand-flexible docking calculations. Fifty search at-
tempts (ga_run parameter) were performed for each ligand. Other
docking parameters were set to the software’s default values. After the
docking search was completed, the best conformation was chosen with
the lowest docked energy. The interactions between inhibitors and the
UGT1A3 enzyme including hydrogen bonds and hydrophobic interac-
tions were analyzed.
oxazolidin-5-yl}methyl)pyridine-2-carboxamide (R4). Compound
R4 was synthesized according to the procedure described for S1, yield: 26%;
mp: 117–120°C; Rf = 0.70 (MeOH:CH2Cl2 = 1:10); [α]D23 = –72.69 [c 0.1
MeOH]; 1H NMR (400 MHz, CDCl3) δ 3.72–3.96 (m, 5H), 4.03 (t, J = 4.96
Hz, 2H), 4.12 (t, J = 8.9 Hz, 1H) 4.33 (s, 2H), 4.89 (br, 1H), 7.31–7.33 (d,
J = 8.8 Hz, 2H), 7.47–7.49 (d, J = 7.92 Hz, 1H), 7.55–7.58 (d, J = 8.84 Hz,
2
H), 7.82 (t, J = 7.76 Hz, 1H), 8.08–8.10 (d, J = 7.52 Hz), 8.29 (t, J = 6.2
Hz); 13C NMR (100 MHz, CDCl3) δ 42.2, 47.8, 49.6, 53.4, 64.1, 68.5, 71.7,
1
1
19.0, 121.1, 126.1, 127.5, 136.7, 137.2, 140.0, 149.4, 150.3, 154.2, 163.9,
66.8; mass spectrum (ESI): m/e (% relative intensity) 452.7 (100)
Inhibition Kinetic Determination
+
+
(M+Na) ; HRMS (ESI): m/z calcd for C20H19O5N3SCl (M+Na) 453.0942.
For some representative compounds R3 and R4, the inhibition kinetic
behaviors (including kinetic type and parameters) were determined.
The reaction velocity was determined at different concentrations of sub-
strates and inhibitors, and the incubation time is the same as that in the
initial screening study. The two most common plot methods (Dixon and
Lineweaver-Burk plots) to determine the inhibition type were employed
in the present study. The inhibition parameters (K ) were calculated
i
through correlating the slopes from the Lineweaver-Burk plots versus
the concentrations of inhibitors.
In Vitro Incubation Mixture to Determine the Inhibitory
Capability
The in vitro incubation system to evaluate the inhibition of UGTs’ activ-
1
4,15
ity by compounds has been described in the previous literature.
brief, the incubation system (total volume = 200 μL) contained recombi-
nant human UGT isoforms, 5 mM UDPGA, 5 mM MgCl , 50 mM Tris-
In
2
HCl (pH = 7.4), and 4-MU in the absence or presence of different concen-
trations of various rivaroxaban or its derivatives. The incubation time
used and protein concentration were previously determined to ensure
the reaction rate within the linear range. The incubation reaction was ini-
tiated through addition of UDPGA to the mixture after a 5-min
preincubation at 37°C. The reactions were quenched by adding 100 μL
acetonitrile with 7-hydroxycoumarin (100 μM) as internal standard. The
mixture was centrifuged at 20,000g for 10 min, and an aliquot of superna-
tant was transferred to an autoinjector vial for HPLC analysis. The HPLC
system (Shimadzu, Kyoto, Japan) contained an SCL-10A system control-
ler, two LC-10AT pumps, an SIL-10A autoinjector, and an SPD-10AVP
UV detector. Chromatographic separation was carried out using a C18
column (4.6 × 200 mm, 5 μm, Kromasil) at a flow rate of 1 mL/min and
UV detector at 316 nm. The mobile phase consisted of acetonitrile (A)
RESULTS
Inhibition Profiles of Rivaroxaban Derivatives Towards
Various UGT Isoforms
The inhibition potential of rivaroxaban derivatives towards
various UGT isoforms was investigated at 100 μM. As shown
in Figure 1, 100 μM of rivaroxaban derivatives weakly
inhibited the activity of UGT1A1, UGT1A6, UGT1A8,
UGT1A9, UGT1A10, and UGT2B7, with less than 75% inhibi-
tion towards all these UGT isoenzymes. For UGT1A3 inhibi-
tion by rivaroxaban derivatives, the activity of UGT1A3 was
inhibited 73.9%, 44.5%, 49.3%, 74.7%, 43.9%, 78.5%, 45.3%,
and H
condition was used: 0–15 min, 95–40% B; 15–20 min, 10% B; 20–30 min,
5% B. The calculation curve was generated by peak area ratio
4-MUG/internal standard) over the concentration range of 4-MUG
.1–100 mM. The curve was linear over this concentration range, with
2
O containing 0.5% (v/v) formic acid (B). The following gradient
8
3.9% by S1, R1, S2, R2, S3, R3, S4, and R4, respectively. S4
9
(
0
inhibited more than 75% activity of UGT1A7-catalyzed 4-MU
glucuronidation. Other rivaroxaban derivatives exhibited
weak inhibition capability towards UGT1A7-catalyzed 4-MU
glucuronidation, with the activity of UGT1A7 inhibited by less
than 75%. The chiral properties exhibited influence on the
inhibition capability towards UGT1A1 and UGT1A3-catalyzed
2
an r value >0.99. The limits of detection and quantification were deter-
mined at signal-to-noise ratios of 3 and 10, respectively. The accuracy
and precision for each concentration were more than 95%.
4
-MU glucuronidation. R2 and R4 exhibited stronger inhi-
Molecular Docking
bition potential towards the activity of UGT1A1 than S2 and
S4 (P < 0.05). S1 exerted stronger inhibition towards
UGT1A3 than R1 (P < 0.01). In contrast, R2, R3, and R4
showed stronger inhibition potential than S1, S2, and S3
The 3D structure of the UGT1A3 enzyme is still unknown. Thus, we de-
veloped a 3D structure for the UGT1A3 enzyme. The homology modeling
method has been applied to the building of this 3D model. The amino acid
sequence of the UGT1A3 enzyme in the FASTA format was retrieved
from the NCBI database (Accession number: NP061966), which was used
for homology modeling of the 3D structure of the UGT1A3 enzyme. The
templates for homology modeling of the UGT1A3 structure were in-
cluded the crystal structures of oleandomycin glycosyltransferase (PDB
code: 2iya), flavonoid 3-O glycosyltransferase (PDB code: 2c1x), and hy-
droquinone glucosyltransferase (PDB code: 2vce). The Modeller 9v14
program was used for predicting the 3D model of the UGT1A3 enzyme
according to the known crystal structures of homologous proteins. The
best model can be selected based on the lowest value of the Modeller
objective function and DOPE (discrete optimization protein energy) score
build by the Modeller program. PROCHECK was used to determine the
stereochemical quality of the prediction model, and check the rationality
of the predicted enzyme structure.
(
P < 0.01). Furthermore, concentration-dependent inhibition
behavior was determined (Fig. 2). The results showed that
S1 exhibited stronger inhibition potential than R1 towards
UGT1A3 at tested concentrations. In contrast, R2, R3, and
R4 exerted stronger inhibition ability than S2, S3, and S4 in
a concentration-dependent manner.
Molecular Docking to Understand the Interaction Between
Rivaroxaban Derivatives and UGT1A3
Eight inhibitors were docked into the cavity of UGT1A3 by
molecular docking method (Fig. 3A), which explored the
interactions between ligand and protein. It is a big task to
Chirality DOI 10.1002/chir

RIVAROXABAN’S INHIBITION TOWARDS UGTS
939
Fig. 1. Screening of inhibitory capability of rivaroxaban derivatives towards recombinant UGT isoforms-catalyzed 4-MU glucuronidation. The incubation system
(
2
total volume = 200 μL) contained recombinant human UGT isoforms, 5 mM UDPGA, 5 mM MgCl , 50 mM Tris-HCl (pH = 7.4), 4-MU, and 100 μM of rivaroxaban
derivatives. The residual activity (% control) = the activity of UGTs-catalyzed 4-MU glucuronidation at 100 μM of rivaroxaban derivatives / the activity of UGTs-cat-
alyzed 4-MU glucuronidation at 0 μM of rivaroxaban derivatives *100%.
recognize the ligand binding domain of UGT1A3, and the
molecular docking method was used to generate the active
pocket of UGT1A3. The active site of UGT1A3 enzyme for
binding with S1, R1, S2, R2, S3, R3, S4, and R4 is composed
of residues Ser39, His40, Leu42, Leu77, Arg112, Ser113,
Met116, Leu117, Met120, Val154, Arg174, Asn175, Cys224,
His225, Phe239, Arg258, Asn283, Arg284, Lys285, Ser307,
Leu308, Gly309, Ser310, Val312, Ser313, Arg337, Trp355,
Leu356, Pro357, Gln358, Asn359, His373, Ala374, Gly375,
Ser376, His377, Gly378, Gln381, Phe395, Gly396, Asp397,
Gln398, and Asn401, which is shown in Figure 3B. In the
active pocket of UGT1A3, eight inhibitors made hydrogen
bonds to UGT1A3 enzyme, which are shown in Figure 4.
Ligand S1 formed a hydrogen bond with residues Ser39,
Ser376, His377, Gly378, and Asp397 in the active pocket of
the UGT1A3 enzyme; however, ligand R1 just formed four
Chirality DOI 10.1002/chir

9
40
MA ET AL.
Fig. 2. Concentration-dependent inhibition of rivaroxaban derivatives towards recombinant UGT1A3-catalyzed 4-MU glucuronidation. Each data point represents
the mean value of duplicate experiments.
residue Ser310, Ser376, His377, and Asp397, and its chiral
compound R3 formed hydrogen bonds with residues
Asn175, Ser310, and Gln358 in the UGT1A3 enzyme. Ligand
S4 made hydrogen bonds to residues Asn175 and Glu381,
and its chiral compound R4 formed hydrogen bonds with res-
idues Asn175, Ser310, and Gln358 in UGT1A3. The hydropho-
bic contacts between the eight inhibitors and UGT1A3
appeared in the active pocket of enzyme (Fig. 5). Ligand S1
formed hydrophobic contacts with residues Ser39, Leu117,
His225, Gly309, Trp355, Gln358, His373, Gly375, Ser376,
His377, Gly378, Glu381, Phe395, and Gly396. Ligand R1
formed analogous hydrophobic contacts with residues
Ser39, Gly309, Trp355, Leu356, Gln358, His373, Gly375,
His377, Phe395, Gly396, and Gln398. The hydrophobic con-
tacts formed by UGT1A3-S2 contain residues Ser39, Leu42,
Leu117, Asn175, Gly309, Gln358, His373, Gly375, His377,
Gly378, Phe395, and Gly396. Ligand R2 formed hydrophobic
contacts with residues Ser39, Gly309, Trp355, Leu356,
Gln358, His373, Gly375, His377, Gly378, Gly396, and
Gln398. Ligand S3 made hydrophobic contacts to residues
Ser39, Leu42, Leu117, His225, Gly309, Ser310, Trp355,
His373, Gly375, Ser376, His377, Phe395, and Gly396. Ligand
Fig. 3. The binding of eight rivaroxaban derivatives towards the activity
cavity of UGT1A3. (A) Far view map of the binding of rivaroxaban derivatives
towards the activity cavity of UGT1A3; (B) The active pocket of UGT1A3 en-
zyme binding with inhibitors S1 (red), R1 (green), S2 (magenta), R2 (cyan),
S3 (orange), R3 (blue), S4 (gray), and R4 (wheat).
hydrogen bonds, involved in residues Asn175, Ser310,
Gln358, and Gln398 in UGT1A3. Ligand S2 made hydrogen
bonds with the residues Ser39, Ser376, His377, and Gly378
of UGT1A3, whereas its chiral compound R2 formed different
hydrogen bonds. Ligand R2 formed hydrogen bonds with
residues Asn175, Ser310, and Gln358 in the active pocket of
the UGT1A3 enzyme. Ligand S3 made hydrogen bonds to
Chirality DOI 10.1002/chir

RIVAROXABAN’S INHIBITION TOWARDS UGTS
941
Fig. 4. The UGT1A3 enzyme in complex with S1, R1, S2, R2, S3, R3, S4, and R4, and hydrogen bond is shown by line.
Fig. 5. The hydrophobic contacts between UGT1A3 and S1, R1, S2, R2, S3, R3, S4, and R4. The protein and inhibitors are analysis by LigPlot.
R3 formed hydrophobic contacts with residues Ser39, Gly309,
Ser310, Val312, Trp355, Leu356, Gln358, His373, Gly375,
His377, Phe395, Gly396, and Gln398. The hydrophobic con-
tacts formed by the interaction of UGT1A3-S4 were involved
in residues Ser39, His40, Leu42, Asn283, Arg284, Lys285,
Gly309, Pro357, Gln358, His373, Gly375, His377, and
Gly378. Ligand R4 made hydrophobic contacts to residues
Ser39, Gly309, Trp355, Gln358, His373, Gly375, His377,
Gly378, Glu381, Phe395, Gly396, and Gln398.
calculated the binding free energy values for the eight inhib-
itors binding into the UGT1A3 enzyme, the binding free
energy value for S1, R1, S2, R2, S3, R3, S4, R4 were –10.61,
–10.20, –9.62, –9.90, –9.65, –10.65, –9.28, and –10.74 (Table 1),
respectively, which is in accordance with the experimental
result that the rank of the inhibition activity for the eight
inhibitors were S1>R1, S2<R2, S3<R3, and S4<R4.
Inhibitory Kinetic Type and Parameters of Represent
Rivaroxaban Derivatives Towards UGT1A3
The top rank in the 50 conformations were obtained from
the molecular docking search for results analysis. The eight
inhibitors made analogical bind to the UGT1A3 enzyme. We
R3 and R4 were selected as representative compounds to
determine their inhibition kinetics towards UGT1A3-
Chirality DOI 10.1002/chir

9
42
MA ET AL.
TABLE 1. Docking results of S1, S2, S3, S4, R1, R2, R3, and R4 with UGT1A3 enzyme
Intermolecular energy
(kcal/mol)
Internal energy
(kcal/mol)
Torsional free energy
(kcal/mol)
Unbound system’s energy
Binding free energy
(kcal/mol)
Inhibitor
(kcal/mol)
S1
S2
S3
S4
R1
R2
R3
R4
À11.80
À10.81
À10.84
À11.07
À11.69
À11.09
À11.84
À11.94
À0.96
À0.84
À1.14
À0.92
À0.96
À1.08
À0.98
À0.85
1.19
1.19
1.19
1.19
1.49
1.19
1.19
1.19
À0.96
À0.84
À1.14
À0.92
À0.96
À1.08
À0.98
À0.85
À10.61
À9.62
À9.65
À9.87
À10.20
À9.90
À10.65
À10.74
Fig. 6. Inhibition kinetics of R3 towards the activity of UGT1A3. (A) Dixon
plot for the inhibition of R3 towards the activity of UGT1A3; (B) Lineweaver-
Burk plot for the inhibition of R3 towards the activity of UGT1A3; (C) Second
plot for the inhibition of R3 towards the activity of UGT1A3. The plot was
drawn using the slopes of the lines from Lineweaver-Burk plot towards the
concentration of R3.
Fig. 7. Inhibition kinetics of R4 towards the activity of UGT1A3. (A) Dixon
plot for the inhibition of R4 towards the activity of UGT1A3; (B) Lineweaver-
Burk plot for the inhibition of R4 towards the activity of UGT1A3; (C) Second
plot for the inhibition of R4 towards the activity of UGT1A3. The plot was
drawn using the slopes of the lines from Lineweaver-Burk plot towards the
concentration of R4.
catalyzed 4-MU glucuronidation. As shown in Figures 6A, 7A,
the intersection point was located in the second quadrant in
the Dixon plot. The intersection point in Lineweaver-Burk
plot was located in the vertical axis (Figs. 6B, 7B). The sec-
ond plot was drawn using the slopes from the Lineweaver-
Burk plot towards the concentrations of R3 or R4. As shown
in Figures 6C, 7C, the fitting equation was y=276.7x+50 and
y=80.3x+10 for the inhibition of R3 and R4 towards UGT1A3-
catalyzed 4-MU glucuronidation. Using the equations, the
towards UGT isoforms.17 Some drugs also exhibited inhibi-
tory behavior on the activity of UGT isoforms. For example,
the antiviral drug arbidol showed strong inhibition towards
the activity of UGT1A9 and UGT2B7.18 The antidiabetes drug
glimepiride exhibited strong inhibition towards UGT1A6.19
The herbal ingredient andrographolide exhibited highly
specific inhibition towards UGT2B7. The inhibition of
UGTs’ activity by all these compounds strongly results in
the disrupted metabolism of endogenous and exogenous
substances.
15
inhibition kinetic parameters (K ) were calculated to be 0.2
i
and 0.1 μM for the inhibition of R3 and R4 towards UGT1A3,
respectively.
In this study, the inhibition of UGTs’ activity by rivaroxaban
derivatives was investigated to indicate the possible drug–
drug interaction or disrupted metabolic disorders of
endogenous substances. Rivaroxaban derivatives showed
the most significant influence towards UGT1A3. UGT1A3
is an important UGT isoform involved in the metabolism
DISCUSSION
UGTs play an important biochemical function through cata-
lyzing the metabolism of a remarkable number of structurally
diverse, endogenous, and exogenous substrates, such as
2
0,21
of estrogen and bile acids.
Recently, an experiment
performed by Wang et al. showed that human UGT1A3
can conjugate 25-hydroxyvitamin D3, which is an important
1
6
bilirubin, steroids, fatty acids, and bile acids. Additionally,
these compounds can also inhibit the activity of UGTs. For
example, endogenous substances like bile acids have been
reported to exert strong inhibition potential towards several
UGT isoforms. Among them, taurolithocholic acid (TLCA)
2
2
metabolite of vitamin D3. Therefore, rivaroxaban deriva-
tives might disrupt the metabolism of these endogenous
substances. Additionally, UGT1A3 is also involved in the
metabolism of some clinical drugs (e.g., nonsteroidal
antiinflammatory drugs, etc.). Therefore, potential drug–
drug interactions might exist between rivaroxaban deriva-
tives and UGT1A3’s substrates.
15
exhibited the strongest inhibition towards UGT isoforms.
The lipid components phosphatidylcholine (PC) and
lysophosphatidylcholines (LPC) showed strong inhibition
Chirality DOI 10.1002/chir

RIVAROXABAN’S INHIBITION TOWARDS UGTS
943
5
. Fang ZZ, Cao YF, Hu CM, Hong M, Sun XY, Ge GB, Liu Y, Zhang YY,
Yang L, Sun HZ. Structure-inhibition relationship of ginsenosides towards
UDP-glucuronosyltransferases (UGTs). Toxicol Appl Pharmacol
Molecular docking was used to understand the interaction
between rivaroxaban derivatives and UGT1A3. We found that
the alteration of chirality significantly changed the inhibition
potential towards the activity cavities of UGT1A3, as indicated
by computational modeling. In silico, the alteration of chirality
for ligand S1 made the binding ability weaken according to
the binding free energy between inhibitor and enzyme, which
may be associated with the changes of hydrogen bonds and
hydrophobic interactions between inhibitor and UGT1A3
enzyme. The hydrogen bonds analysis show that ligand S1
made seven hydrogen bonds to the enzyme, which is signifi-
cantly higher than S2, S3, S4. Some of the hydrogen bonds in
UGT1A3-S1 were formed by the interaction between S2, S3,
S4, and UGT1A3; however, ligand S1 made more hydrogen
bonds to residue Ser376 and His377 in the enzyme, which
implies the two residues may influence the binding activity
of UGT1A3-S1. Ligands R1, R2, R3, and R4 formed analogous
hydrogen bonds with the UGT1A3 enzyme; all of them
formed hydrogen bonds with residues Ser310 and Gln358,
which may play an important role in binding of (R)-type
rivaroxaban derivatives into the active pocket of the UGT1A3
enzyme. For the hydrophobic interaction, no specific resi-
dues play an important role.
2013;267:149–154.
6
. Rowland A, Miners JO, Mackenzie PI. The UDP-glucuronosyltransferases:
their role in drug metabolism and detoxification. Int J Biochem Cell Biol
2013;45:1121–1132.
7. Chen S, Yueh MF, Bigo C, Barbier O, Wang K, Karin M, Nguyen N,
Tukey RH. Intestinal glucuronidation protects against chemotherapy-
induced toxicity by irinotecan (CPT-11). Proc Natl Acad Sci U S A
2013;110:19143–19148.
8
. Yu J, Han JC, Gao YJ. Biotransformation of glucoaurantio-obtusin to-
wards aurantio-obtusin increases the toxicity of irinotecan through in-
creased inhibition towards SN-38 glucuronidation. Phytother Res
2
014;28:1577–1580.
9
. Wei J, Ye X, Liu T. Toxicity magnification of irinotecan’s toxicity by
levothyroxine through inhibition of SN-38 glucuronidation. Lat Am J
Pharm 2014;33:1385–1388.
1
0. Wang H, Fang ZZ, Zheng Y, Zhou K, Hu C, Krausz KW, Sun D, Idle JR,
Gonzalez FJ. Metabolic profiling of praziquantel enantiomers. Biochem
Pharmacol 2014;90:166–178.
11. Zucker SD, Qin X, Rouster SD, Yu F, Green RM, Keshavan P, Feinberg J,
Sherman KE. Mechanism of indinavir-induced hyperbilirubinemia. Proc
Natl Acad Sci U S A 2001;98:12671–12676.
2. Liu M, Zhang D, Yang M, Zhao T, Wang X, Zhang Y, Han J, Liu H. Phar-
macokinetics of terazosin enantiomers in healthy Chinese male subjects.
Chirality 2012;24:1047–1050.
3. Sun SY, Wang YQ, Li LP, Wang L, Zeng S, Zhou H, Jiang HD.
Stereoselective interaction between tetrahydropalmatine enantiomers
and CYP enzymes in human liver microsomes. Chirality 2013;25:43–47.
1
However, the hydrophobic residues Leu117, Trp355, and
Phe395 make contacts to ligand S1, which may contribute to
the stronger binding activity of UGT1A3-S1.
1
In conclusion, the significant influence of rivaroxaban deri-
vatives towards UGT1A3’s activity was demonstrated in the
present study. The chirality centers highly affected the inhibi-
tion behavior of rivaroxaban derivatives towards UGT1A3.
14. Fang ZZ, Wang H, Cao YF, Sun DX, Wang LX, Hong M, Huang T, Chen
JX, Zeng J. Enantioselective inhibition of carprofen towards UDP- glucuro-
nosyltransferase (UGT) 2B7. Chirality 2015;27:189–193.
1
5. Fang ZZ, He RR, Cao YF, Tanaka N, Jiang C, Krausz KW, Qi Y, Dong PP,
Ai CZ, Sun XY, Hong M, Ge GB, Gonzalez FJ, Ma XC, Sun HZ. A model of
in vitro UDP-glucuronosyltransferase inhibition by bile acids predicts pos-
sible metabolic disorders. J Lipid Res 2013;54:3334–3344.
ACKNOWLEDGMENTS
This work was supported by the National Natural Science
Foundation of China (No. 81202586, 81202587, 81202588),
the Scientific Research Foundation for the Returned
Overseas Chinese Scholars, State Education Ministry, Tianjin
Project of Thousand Youth Talents, Tianjin city funded inter-
national projects to culture selected outstanding postdoctoral,
and Shandong Natural Science Foundation of China (No.
16. Radominska-Pandya A, Bratton SM, Redinbo MR, Miley MJ. The crystal
structure of human UDP-glucuronosyltransferase 2B7 C-terminal end is
the first mammalian UGT target to be revealed: the significance for
human UGTs from both the 1A and 2B families. Drug Metab Rev
2010;42:133–144.
1
7. Gao X, Qu H, Ai CZ, Cao YF, Huang T, Chen JX, Zeng J, Sun XY, Hong M,
Gonzalez FJ, Liu Z, Fang ZZ. Regulation profile of phosphatidylcholines
(PCs) and lysophosphatidylcholines (LPCs) components towards UDP-
glucuronosyltransferases (UGTs) isoforms. Xenobiotica 2015;45:197–206.
2
013BSE27023 and ZR2010HL023).
1
8. Liu X, Huang T, Chen JX, Zeng J, Fan XR, Xu-Zhu YZW, Sun XY,
Hong M, Sun HZ. Arbidol exhibits strong inhibition towards UDP-
glucuronosyltransferase (UGT) 1A9 and 2B7. Pharmazie 2013;68:945–950.
CONFLICT OF INTEREST
The authors declare no conflicts of interests.
1
9. Fu JF, Ren QY, Zhang NY, Gao B, Tu YY, Fu GQ, Li DH, Zhang YS. Inhi-
bition potential of glimepiride (gli) towards important UDP-
glucuronosyltransferase (UGT) isoforms in human liver. Pharmazie
2012;67:715–717.
LITERATURE CITED
1
. Depasse F, Busson J, Mnich J. Effect of BAY 59-7939—a novel, oral, direct
factor Xa inhibitor—on clot-bound factor Xa activity in vitro. J Thromb
Haemost 2005;3:P1104.
20. Lankisch TO, Gillman TC, Erichsen TJ, Ehmer U, Kalthoff S, Freiberg N,
Munzel PA, Manns MP, Strassburg CP. Aryl hydrocarbon receptor-
mediated regulation of the human estrogen and bile acid UDP-
glucuronosyltransferase 1A3 gene. Arch Toxicol 2008;82:573–582.
21. Erichsen TJ, Aehlen A, Ehmer U, Kalthoff S, Manns MP, Strassburg CP.
Regulation of the human bile acid UDP-glucuronosyltransferase 1A3 by
the farnesoid X receptor and bile acids. J Hepatol 2010;52:570–578.
22. Wang Z, Wong T, Hashizume T, Dickmann LZ, Scian M, Koszewski NJ,
Goff JP, Horst RL, Chaudhry AS, Schuetz EG, Thummel KE. Human
UGT1A4 and UGT1A3 conjugate 25-hydroxyvitamin D3: metabolite struc-
ture, kinetics, inducibility, and interindividual variability. Endocrinology
2014;155:2052–2063.
2
. Roehrig S, Straub A, Pohlmann J. Discovery of the novel antithrombotic
agent 5-chloro-N-({(5S)-2-oxo-3- [4-(3-oxomorpholin-4-yl)phenyl]-1,3-
oxazolidin-5-yl} methyl) thiophene- 2-carboxamide (BAY 59-7939): an oral,
direct factor Xa inhibitor. J Med Chem 2005;48:5900–5908.
3
. Weinz C, Schwarz T, Kubitza D. Metabolism and excretion of
rivaroxaban, an oral, direct factor Xa inhibitor, in rats, dogs, and humans.
Drug Metab Dispos 2009;37:1056–1064.
4
. Mueck W, Kubitza D, Becka M. Co-administration of rivaroxaban with
drugs that share its elimination pathways: pharmacokinetic effects in
healthy subjects. Br J Clin Pharmacol 2013;76:455–466.
Chirality DOI 10.1002/chir