Metabolic epoxidation is a critical step for the development of benzbromarone-induced hepatotoxicity

Article abstract of DOI:10.1124/dmd.117.077818

Benzbromarone (BBR) is effective in the treatment of gout; however, clinical findings have shown it can also cause fatal hepatic failure. Our early studies demonstrated that CYP3A catalyzed the biotransformation of BBR to epoxide intermediate(s) that reacted with sulfur nucleophiles of protein to form protein covalent binding both in vitro and in vivo. The present study attempted to define the correlation between metabolic epoxidation and hepatotoxicity of BBR by manipulating the structure of BBR. We rationally designed and synthesized three halogenated BBR derivatives, fluorinated BBR (6-F-BBR), chlorinated BBR (6-Cl-BBR), and brominated BBR (6-Br-BBR), to decrease the potential for cytochrome P450-mediated metabolic activation. Both in vitro and in vivo uricosuric activity assays showed that 6-F-BBR achieved favorable uricosuric effect, while 6-Cl-BBR and 6-Br-BBR showed weak uricosuric efficacy. Additionally, 6-F-BBR elicited much lower hepatotoxicity in mice. Fluorination of BBR offered advantage to metabolic stability in liver microsomes, almost completely blocked the formation of epoxide metabolite(s) and protein covalent binding, and attenuated hepatic and plasma glutathione depletion. Moreover, the structural manipulation did not alter the efficacy of BBR. This work provided solid evidence that the formation of the epoxide(s) is a key step in the development of BBR-induced hepatotoxicity.

Full text of DOI:10.1124/dmd.117.077818

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Title Page
Metabolic epoxidation is a critical step for the development of
benzbromarone-induced hepatotoxicity
Hui Wang, Ying Peng, Tingjian Zhang, Qunsheng Lan, Huimin Zhao, Wenbao Wang,
Yufei Zhao, Xu Wang, Jianxin Pang, Shaojie Wang, Jiang Zheng
Wuya College of Innovation (H. W., Y. P., H. Z., Y. Z., X. W., J. Z.), Shenyang
Pharmaceutical University, Shenyang, Liaoning, 11016, P. R. China; School of
Pharmacy (T. Z.), China Medical University, Shenyang, Liaoning, 110122, P. R.
China; Guangdong Provincial Key Laboratory of Drug Screening (Q. L., J. P.),
School of Pharmaceutical Sciences, Southern Medical University, Guangzhou,
Guangdong, 510515, P. R. China; Key Laboratory of Structure-Based Drug Design &
Discovery (Ministry of Education) (W. W., S. W.), School of Pharmaceutical
Engineering, Shenyang Pharmaceutical University, Shenyang, 110016, P. R. China;
State Key Laboratory of Functions and Applications of Medicinal Plants (J. Z.), Key
Laboratory of Pharmaceutics of Guizhou Province (J. Z.), Guizhou Medical
University, Guiyang, Guizhou, 550004, P. R. China.
1

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Running Title Page
Running title: Epoxidation and hepatotoxicity of benzbromarone
Corresponding Authors:
Jianxin Pang
School of Pharmaceutical Sciences, Southern Medical University, 1838 North
Guangzhou Ave, Guangzhou, Guangdong, 510515, China.
E-mail: pjx@smu.edu.cn
Tel: +86-20-61648671
Fax: +86-20-61648671
Shaojie Wang
School of Pharmaceutical Engineering, Shenyang Pharmaceutical University, 103
Wenhua Rode, Shenyang, Liaoning, 110016, P. R. China.
E-mail: sjwang_99@163.com
Tel: +86-24-43520230
Fax: +86-24-43520230
Jiang Zheng
Wuya College of Innovation, Shenyang Pharmaceutical University, 103 Wenhua
Rode, Shenyang, Liaoning, 110016, P. R. China; State Key Laboratory of Functions
and Applications of Medicinal Plants, Key Laboratory of Pharmaceutics of Guizhou
Province, Guizhou Medical University, Guiyang, Guizhou, 550004, P. R. China.
E-mail: zhengneu@yahoo.com
Tel: +86-24-23986361
Fax: +86-24-23986510
The number of text pages: 30
The number of tables: 3
The number of figures: 10
The number of schemes: 1
The number of references: 43
The number of words in the Abstract: 194
The number of words in the Introduction: 637
The number of words in the Discussion: 1389
2

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Abbreviations
ALT, alanine transaminase; AST, aspartate transaminase; BBR, benzbromarone; CE,
collision energy; CLint, intrinsic clearance; CUR, curtain gas; CXP, cell exit potential;
DMEM, Dulbecco’s modified Eagle’s medium; DP, declustering potential; DPBS,
dulbecco's phosphate buffered saline; ESI, electrospray ionization; GS1, ion source
gas 1; GS2, ion source gas 2; GSH, glutathione; hURAT1, human urate transporter 1;
HLMs, human liver microsomes; HRESIMS, high resolution electrospray ionization
mass spectroscopy; LC-MS/MS, liquid chromatography coupled to tandem mass
spectrometry; mBrB, monobromobimane; MDCK, Madin-Darby Canine Kidney;
MLMs, mouse liver microsomes; MRM, multiple-reaction monitoring; NAC,
N-acetylcysteine; NADPH, β-nicotinamide adenine dinucleotide 2′-phosphate reduced
tetrasodium salt; P450, cytochrome P450; PO, potassium oxonate; RLMs, rat liver
microsomes; SUA, serum uric acid; t1/2, half-life; TLC, thin-layer chromatography.
3

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Abstract
Benzbromarone (BBR) is effective in the treatment of gout but can also cause
fatal hepatic failure in clinic. Our early studies demonstrated that CYP3A catalyzed
the biotransformation of BBR to epoxide intermediate(s) which reacted with sulfur
nucleophiles of protein to form protein covalent binding both in vitro and in vivo.
The present study attempted to define the correlation between metabolic epoxidation
and hepatotoxicity of BBR by manipulating the structure of BBR. We rationally
designed and synthesized three halogenated BBR derivatives (6-F-, Cl-, or Br-BBR)
to decrease the potential for P450-mediated metabolic activation. Both in vitro and
in vivo uricosuric activity assays showed that the fluorinated BBR (6-F-BBR)
achieved favorable uricosuric effect, while the chlorinated BBR (6-Cl-BBR) and
brominated BBR (6-Br-BBR) showed weak uricosuric efficacy. Additionally,
6-F-BBR elicited much lower hepatotoxicity in mice. Fluorination of BBR offered
advantage to metabolic stability in liver microsomes, almost completely blocked the
formation of epoxide metabolite(s) and protein covalent binding, and attenuated
hepatic and plasma GSH depletion. Moreover, the structural manipulation did not
alter the efficacy of BBR. This work provided solid evidence that the formation of
the epoxide(s) is a key step for the development of BBR-induced hepatotoxicity.
4

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Introduction
Gout is the most prevalent inflammatory arthritis and occurs in approximately
0.2% to 0.35% of the general population (Fernandes et al., 2017). Specifically, gout
is an inflammatory arthritis characterized by deposition of urate in articular cartilage
or periarticular tissues (Perez-Ruiz et al., 2015; Keenan and Schlesinger, 2016).
Overproduction of uric acid and deficiency in renal excretion contribute to elevated
serum urate which is known as hyperuricemia. The incidence of gout grows is
considered to partially result from the decline of renal function (Smith et al., 2010).
Genetic and dietary factors also influence uric blood levels (Doherty, 2009). Recent
studies suggest that hyperuricemia plays essential roles in some important diseases
such as cardiovascular disease, diabetes, hypertension, and chronic renal disease
(Baker and Schumacher, 2010; Ruoff and Edwards, 2016). Gout has been a threat to
human health, and the needs for effective drugs that lower serum urate levels are
therapeutically urgent.
Insufficient renal excretion was reported to account for 70% of urate load (Choi
et al., 2005; Abhishek et al., 2017). Consequently, uricosurics that act directly on the
renal tubule and increase urate renal excretion by inhibiting uric acid reabsorption
emerge. Among commercially available uricosurics, benzbromarone (BBR, Figure 1)
is extremely potent and well tolerated, with most people meeting treatment goals,
compared with probenecid and sulfinpyrazone (Abhishek et al., 2016). As a
uricusuric, BBR shows potent inhibition property to human urate transporter 1
(hURAT1), a known transporter responsible for urate reabsorption (Wempe et al.,
2011; Ahn et al., 2016). Though effective, BBR is not licensed for use in the United
5

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
States, due to several reported idiosyncratic hepatotoxic events (Arai et al., 2002; Lee
et al., 2008; Kydd et al., 2014).
It is widely accepted that cytochromes P450 (P450)-mediated metabolic
activation is most frequently responsible for drug induced idiosyncratic hepatotoxicity
and drug withdrawal from the market (Leung et al., 2012; Dekker et al., 2016; Xuan
et al., 2016). Our early metabolic studies illustrated that CYP3A metabolized BBR
to epoxide intermediate(s) which reacted with sulfur nucleophiles of protein to form
protein covalent binding both in vitro and in vivo (Wang et al., 2016a; Wang et al.,
2016b). In an attempt to define the association between BBR bioactivation and its
hepatotoxicity, we modified the structure of BBR by decreasing the potential for the
P450-mediated metabolic activation and then evaluated the toxicological and
pharmacological responses to the structural manipulation.
Incorporation of halogens in drugs is a commonly employed strategy when
decrease in P450 metabolic degradation is needed (Wilcken et al., 2013; Ford and Ho,
2016; Obach et al., 2016). In the lead optimization of Ezetimib, a cholesterol
inhibitor, benzylic oxidation was blocked by introduction of fluorine at the phenyl
ring to mitigate P450-mediated oxidation (Rosenblum et al., 1998; Purser et al., 2008).
Bromination or iodination proximal to the phenolic hydroxyl group was carried out in
structural optimization of 13α-estrone to prevent the generation of reactive quinone
intermediates that might participate in estrogen-induced carcinogenesis (Patel and
Bhat, 2004; Bacsa et al., 2015). It is convinced that by inserting halogens adjacent to
or substitution of metabolically labile sites, the potential for metabolic clearance can
6

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
be attenuated by lowing metabolic rate or blocking metabolic pathways. Besides,
halogen substituents are frequently utilized to improve membrane permeability and,
therefore, to enhance oral absorption (Lu et al., 2010; Ford and Ho, 2016).
The objective of this study was to define the correlation of BBR bioactivation
with its hepatotoxicity by manipulating the structure of BBR. The correlation study
was achieved by 1) structural modification of BBR through introducing halogen
atoms to benzofuran ring at C6 position which is known to be a site of CYP3A
metabolism; 2) evaluation of metabolic activation and toxicity of the test compounds;
3) determination of metabolic stability and pharmacokinetic properties; and 4)
assessment of the efficacies of the new compounds in vitro and in vivo.
7

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Materials and Methods
1. Reagents
Benzbromarone (BBR, >98%) and potassium oxonate (PO) were purchased from
Aladdin Industrial Technology Co., Ltd. (Shanghai, China). 6-F-, Cl-, or Br-BBR
(Figure 1) were synthesized in our laboratory (see Supplemental Methods for
synthesis details).
Buthionine sulfoximine (BSO, >99%), N-acetyl cysteine
(NAC, >98%) and reduced nicotinamide adenine dinucleotide phosphate (NADPH)
were acquired from Sigma-Aldrich (St. Louis, MO). [8-14C]-Urate was obtained
from American Radiolabled Chemicals (St. Louis, MO). Liquid scintillation was
acquired from American Perkin Elmer Corporation. All organic solvents and
reagents were of either analytical grade obtained from Fisher Scientific (Springfield,
NJ) or HPLC grade from commercial sources.
2. Assay to measure the uricosuric effect in MDCK cells over-expressing
hURAT1
Cell culture. Madin-Darby Canine Kidney (MDCK) cells with over-expressed
hURAT1 were pre-prepared in our laboratory as described in our published work [32].
Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10%
fetal bovine serum, streptomycin (100 mg/mL), and penicillin (100 units/mL) in an
atmosphere of 5% CO2 and 95% air at 37 °C.
[14C]-Urate uptake experiment in MDCK-hURAT1 cells. Uptake assay was
performed as described previously with several modifications (Wu et al., 2017).
Briefly, cells were seeded in 24-well plates at a density of 1×105 cells/well. After 48
8

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
h, the culture medium was removed, and the resulting cells were washed three times
with uptake buffer (125 mM sodium gluconate, 4.8 mM potassium gluconate, 1.3 mM
calcium gluconate, 1.2 mM KH2PO4, 1.2 mM MgSO4, 5.6 mM glucose, and 25 mM
HEPES, pH=7.4). The cells were then incubated in another 0.5 mL of uptake buffer
for 10 min at 37 °C. Then the uptake buffer was removed, followed by addition of
200 μL of the medium containing [14C]-Urate (25 M) with or without the test
compounds to initiate the uptake process. The test compounds were dissolved in
DMSO and diluted with the culture medium to the final concentration of DMSO at <
1‰. The uptake step was set for 2 min and terminated by mixing with 0.6 mL of
ice-cold dulbecco's phosphate buffered saline (DPBS) in wells. The resultant cells
were washed with 1 mL of DPBS two times quickly, and then solubilized in 100 L of
0.1M NaOH. The radioactivity of the cell lysates was determined by a liquid
scintillation counter (Micro Beta2, PerkinElmer, Waltham, MA).
3. Animals
Male Sprague Dawley rats (200 ± 20 g) and male Kunming mice (20 ± 2 g)
were purchased from the Animal Center of Shenyang Pharmaceutical University
(Shenyang, China). Animals were housed in a controlled environment (temperature
25 °C and 12-h dark/light cycle) with free access to food and water. Animals were
fasted overnight and tap water ad libitum prior to the experiments. The protocol of
the animal experiments was approved by the Ethics Review Committee for Animal
Experimentation of Shenyang Pharmaceutical University.
9

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
4. Assay to determine the uricosuric effect in hyperuricemic rats
Hyperuricemia was induced in rats using uricase inhibitor potassium oxonate
(PO), according to previous reports (Haidari et al., 2012; Kou et al., 2016; Wang et al.,
2016). Briefly, rats were intraperitoneally (i.p.) treated with potassium oxonate 1 h
before the final administration of the test compounds. Animals were randomly
allocated into six groups (n = 8). Groups I and II served as a healthy control and
hyperuricemic control, respectively. Group III was hyperuricemic rats treated with
BBR, which served as the positive control.
Groups IV-VI consisted of
hyperuricemic rats dosed with 6-F-, Cl-, or Br-BBR. The animals were administered
(p.o.) with BBR or the other test compounds suspended in 0.9% CMC-Na at 10
mg/kg once daily for 8 days. On the eighth day, PO suspended in 0.9% saline
solution was injected i.p. at 250 mg/kg 1 h in the resulting animals before the final
administration of the test compounds. Blood samples were collected from the
retro-orbital sinus 2 h after PO administration for the assessment of serum uric acid
(SUA) levels. The blood samples were allowed to clot for 1 h at room temperature
and then were centrifuged (8,000g, 4 °C) for 10 min to obtain sera. The
concentrations of SUA were determined spectrophotometrically using a commercially
available standard diagnostic kit, according to manufacturers’ instruction (Nanjing
Jiancheng Technology Co., LTD, Nanjing, China).
5. Assessment of hepatotoxicities of BBR and 6-F-BBR
Male Kunming mice were randomly assigned into six groups (n = 4). Two
groups of mice were p.o. treated with BBR or 6-F-BBR dissolved in 0.9% CMC-Na at
10

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
dose of 50 mg/kg for 7 consecutive days. Another two groups of mice were i.p.
treated with BSO 1.5 h before the final administration of BBR or 6-F-BBR. Control
groups treated with 0.9% CMC-Na or BSO were included. Blood samples were
collected by cardiac puncture 24 h after the final administration. After clotting at
room temperature for 3 h, the blood samples were centrifuged at 8,000 g for 10 min at
4 °C. The resulting sera were collected for the determination of alanine
transaminase (ALT) and aspartate transaminase (AST) activities on a VITROS 5600
Integrated System (Ortho-Clinical Diagnostics, Rochester, NY).
6. Metabolic stability test
Mouse liver microsomes (MLMs) and rat liver microsomes (RLMs) were
prepared in our laboratory by following the protocol we previously published (Lin et
al., 2007). Human liver microsomes (HLMs) were acquired from BD Gentest
(Woburn, MA).
BBR or 6-F-BBR at a concentration of 1 M or 10 M was incubated with
MLMs, RLMs, or HLMs (1 mg/mL), and MgCl2 (3.2 mM) in a final volume of 0.5
mL of potassium phosphate buffer (0.1 M, pH 7.4). The mixture was preincubated at
37 °C for 3 min and then the reaction was commenced by addition of 1.0 mM
NADPH. Aliquots of 50 L were withdrawn at time points of 0, 5, 10, 15, 20, 30,
and 60 min, and the reaction was immediately quenched by adding equal volume of
ice-cold acetonitrile containing S-hexylglutathione (1.28 M) as an internal standard.
The samples were then vortex-mixed and spun at 16,000 g for 10 min at 4 °C. The
resulting supernatants (5 L) were subjected to liquid chromatography-tandem mass
11

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
spectrometry (LC-MS/MS) for the analysis of the depletion of the compound. Each
stability analysis was performed in triplicate. Natural log of the area ratio was
plotted against the incubation time using GraphPad Prism 5 (San Diego, CA, USA),
and the elimination rate constant (−k, the slope of the linear regression) was used to
determine the in vitro half-life (t1/2). For compounds with t1/2 < 60 min, the in vitro
intrinsic clearance (CLint) was calculated as:
 k
[E]
CLint =
(1)
where [E] is the concentration of microsomal protein used in the incubation.
7. Metabolic activation studies of BBR and 6-F-BBR in liver microsomes
The incubation mixtures contained 1.0 mg of protein/mL MLMs, 3.2 mM MgCl2,
75 M BBR or 6-F-BBR, and 40 mM NAC in a final volume of 0.5 mL of phosphate
buffer (pH 7.4). The reactions were initiated by addition of 1.0 mM NADPH and
quenched by adding 0.5 mL of ice-cold acetonitrile after 60 min incubation at 37 °C.
Control incubations containing no NADPH were included. The reaction mixtures
were vortex-mixed for 2 min, followed by centrifugation at 16,000 g for 10 min to
remove precipitated protein. The resulting supernatants were concentrated by N2
flushing at 40 °C and then redissolved in 100 L of 50% acetonitrile in water. A 5
L aliquot of the supernatants was injected into the LC-MS/MS system for analysis.
8. Determination of in vivo protein adductions derived from BBR and 6-F-BBR
Mice were treated i.p. with BBR or 6-F-BBR at 50 mg/kg. Liver tissues (0.2 g)
harvested 30 min post administration were homogenized in 2.0 mL of phosphate
12

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
buffer (pH 7.4). The resulting homogenates were denatured by heating in a water
bath at 60 °C for 30 min and then digested overnight with a mixture of chymotrypsin
(1.0 mg/mL) and Pronase E (2.0 mg/mL), according to our early study (Wang et al.,
2016b). The supernatants (5 L) were subjected to the LC-MS/MS system for
analysis.
9. Assessment of GSH level
Male Kunming mice were randomly divided into two groups (n = 4), and
exposed i.p. to 50 mg/kg BBR or 6-F-BBR dissolved in corn oil. Plasma and liver
samples were harvested post the administration at time of 0, 10, 30 min and 1, 2, 4, 8,
12 h. The liver tissues (0.2 g) were homogenized in 2.0 mL of phosphate buffer (0.1
M, pH 7.4). The assessment of plasma and hepatic GSH was achieved by
monobromobimane (mBrB) derivatization as described in our published method (Yu
et al., 2015), followed by the analysis of mBrB-GSH conjugate by LC-MS/MS.
10. Pharmacokinetic studies
Rats were randomly allocated into two groups (n = 6) and administered orally
with the test compounds (BBR or 6-F-BBR) suspended in 0.9% CMC-Na by gavage
at 10 mg/kg. Blood samples were collected from the retro-orbital sinus at 0 and 30
min as well as 1, 1.5, 2, 3, 4, 5, 6, 7, 8, 10, 12, and 24 h after dosing. The collected
blood samples were centrifuged (8,000g, 4 °C) for 10 min to obtain the plasma.
Aliqouts (30 μL) of plasma were mixed with ice-cold acetonitrile (90 μL) containing
S-hexylglutathione (1.28 nM) as an internal standard. The resulting mixtures were
13

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
vortex-mixed and centrifuged (16,000g, 4 °C) for 10 min to remove the precipitated
protein. Aliquots (2 L) of the supernatants were subjected to the LC-MS/MS
system for quantification of plasma BBR or 6-F-BBR, using validated specific and
sensitive LC–MS/MS methods. Both the LLOQs of LC-MS/MS methods for BBR
and 6-F-BBR were found to be 0.5 g/mL.
11. LC-MS/MS Method
An AB SCIEX Instruments 5500 triple quadrupole (Applied Biosystems, Foster
City, CA) equipped with an Agilent 1260 Series Rapid Resolution LC system (Agilent
Technologies, Santa Clara, CA) was employed for the LC-MS/MS analyses.
Samples were analyzed by multiple-reaction monitoring (MRM) scanning in
electrospray ionization positive ion mode (ESI+) using the following conditions: an
ion-spray voltage of 5500 V; source temperature, 650 °C; curtain gas (CUR), 35 psi;
ion source gas 1 (GS1), 50 psi; ion source gas 2 (GS2), 50 psi. The parameters of
ion pairs (declustering potential, DP; collision energy, CE; collision cell exit potential,
CXP) were m/z 425→279 (110, 35, 3) for BBR, m/z 443→279 (110, 35, 3) for
6-F-BBR, m/z 392.2→246.3 (86, 24, 5) for S-hexylglutathione, m/z 586→279 (110,
35, 15) for BBR-NAC conjugate derived from epoxide (A1), m/z 544→279 (110, 25,
10) for cysteine-based protein adduction by epoxide metabolite(s) of BBR (A2), and
m/z 498.2→192.2 (115, 50, 3) for mBrB-GSH conjugate. For the analysis of plasma
BBR and 6-F-BBR concentrations, A1, A2, and mBrB-GSH conjugate, the mobile
phase programs and the columns used were in accordance with our published studies
(Wang et al., 2016a; Wang et al., 2016b). Data were processed using Applied
14

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Biosystems/SCIEX Analyst software version 1.6.2.
12. Statistical analysis
Pharmacokinetic parameters were calculated from the mean plasma
concentration by non-compartmental model, using Drug and Statistic (DAS) 2.1.1
software (Mathematical Pharmacology Professional Committee of China, Shanghai,
China). Statistical analyses were performed using a Student’s t-test. *(p < 0.05),
**(p < 0.01) or ***(p < 0.001) were considered significant.
15

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Results
1. Characterization of 6-F-, Cl-, or Br-BBR
The general route for the synthesis of 6-F-, Cl-, or Br-BBR (Figure 1) is shown
in Supplemental Scheme 1. The synthesis started with anhydrous magnesium
chloride-catalyzed formylation of the corresponding 3-halogenated phenol with
paraformaldehyde to produce halogenated salicylaldehyde a. Compound a was then
cyclized with chloroacetone under the catalysis of KOH to give compound b,
followed by Wolff-Kishner-Huangminglong reduction, Friedel-Crafts acylation,
demethylation, and dibromination to give the final compounds. The C6-fluorinated,
-chlorinated, and -brominated BBR, named 6-F-BBR, 6-Cl-BBR, and 6-Br-BBR
1
respectively, were characterized by H NMR, 13C NMR (Figure 2 and 3) and
1
HRESIMS (Figure 4). The H NMR spectra depict the exact integration and the
13
position of each proton present in the individual compounds. The C NMR spectra
show the position of carbons pertaining to each compound. The HRESIMS spectra
display the deprotonated molecular ions [M-H]－ with the representative isotope
abundance patterns.
2. [14C]-Urate uptake in MDCK-hURAT1 cells
hURAT1-transgenic MDCK cells were used to measure the amounts of
[14C]-Urate taken up for BBR, 6-F-BBR, 6-Cl-BBR, and 6-Br-BBR. As shown in
Figure 5, BBR and 6-F-BBR at the concentration of 1 M exhibited 59.39% and
63.59% inhibition of the uptake, respectively, compared with that of control.
However, little inhibition was observed at the same concentration for 6-Cl-BBR and
16

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
6-Br-BBR. Even at the concentration of 15 M, 6-Cl-BBR and 6-Br-BBR only
showed weak inhibition of the uptake (21.72% and 13.33%, respectively). Clearly,
6-F-BBR, with similar efficacy as BBR, was a more potent hURAT1 inhibitor than
6-Cl-BBR and 6-Br-BBR.
3. Uricosuric efficacy in hyperuricemic rats
The effects of the test compounds on SUA levels were evaluated in potassium
oxonate-induced hyperuricemic model rats. As shown in Table 1, after 7 day
treatment with BBR or the test compounds, SUA levels in each group were
significantly reduced (p < 0.001, 0.001, 0.01, and 0.01 respectively), compared with
the normal control. There was no significant difference between BBR and 6-F-BBR
in lowering the SUA levels of healthy rats. As expected, intraperitoneally
administration of oxonate significantly (p < 0.001) elevated SUA levels compared
with that of the normal control. In addition, the groups treated with BBR, 6-F-, Cl-,
or Br-BBR all demonstrated significantly lower levels of SUA (p < 0.001, 0.001, 0.01,
and 0.01 respectively), compared with the hyperuricemic control. However, the
decreased levels of SUA achieved by 6-F-BBR were comparable to that of BBR,
which made us select 6-F-BBR as the lead uricosuric benzofuran derivative. As a
result, a thorough comparative study on metabolism and toxicities of BBR and
6-F-BBR was carried out as follows.
4. Hepatotoxicities of BBR and 6-F-BBR
The hepatotoxicities of BBR and 6-F-BBR were evaluated by the assessment of
17

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
serum ALT and AST activities in mice. Administration of BBR by oral gavage at 50
mg/kg for 7 consecutive days caused significant elevation of serum AST activity in
mice 24 h after the final administration, compared with that in vehicle-treated animals
(p < 0.01) and 6-F-BBR-treated animals (p < 0.05). A minor increase in serum ALT
activity was observed in animals treated with BBR but without statistical difference,
compared with that of vehicle-treated animals.
In a separate study, the
hepatotoxicities of BBR and 6-F-BBR were examined in mice pretreated with BSO
1.5 h before the final administration. Pretreatment with BSO was found to potentiate
6-F-BBR- and BBR-induced elevation of serum ALT and AST. However, the
elevations of ALT and AST induced by BBR were significantly higher than those
induced by 6-F-BBR (Figure 6). Both studies demonstrated that 6-F-BBR was less
hepatotoxic than BBR.
5. Metabolic stability
Metabolic stability of BBR and 6-F-BBR was examined in mouse, rat, and
human liver microsomes. The substrates all decayed in mono-exponential manners
(Supplemental Figure 1) in microsomes. As shown in Table 2, 50% of BBR at 1.0
M was metabolized in mouse, rat, and human liver microsomes in 5.6 4.2, and 8.6
min. However, the half life of 6-F-BBR in the microsomes was > 60, 8.4, and 13.6
min, respectively. At the concentration of 10 M, similar increased metabolic
stability for 6-F-BBR was observed in the microsomes. Taken together, 6-F-BBR
was much more stable in the microsome tests than BBR, and the introduction of a
6-fluoro substituent on the benzofuran ring (6-F-BBR vs BBR) offered resistance
18

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
against the metabolism.
6. In vitro metabolism (epoxidation) of BBR and 6-F-BBR
Our early study demonstrated that epoxide metabolite(s) was a primary reactive
metabolite of BBR which may be related to BBR hepatotoxicity (Wang et al., 2016a).
The potentials of metabolic activation (epoxidation) of BBR and 6-F-BBR were
evaluated in mouse liver microsomes supplemented with NAC as a trapping agent.
The representative chromatograms of NAC conjugate derived from epoxide
metabolite(s) of BBR and 6-F-BBR are shown in Figure 7. As expected, only minor
epoxide-derived NAC conjugate (A1, Scheme 1) was detected in
6-F-BBR-supplemented microsomal incubations. Apparently, an approximately
300-fold more A1 was produced in microsomal incubation of BBR. In addition, we
failed to detect the corresponding NAC conjugate derived from 6-F-BBR, indicating
that no epoxide metabolites were formed in the metabolism of 6-F-BBR and that the
introduction of fluorine atom to BBR deactivated the aromatic ring towards
epoxidation.
7. Epoxide-derived hepatic protein modification by BBR and 6-F-BBR
The detection of epoxide-derived liver protein covalent binding in 6-F-BBR- or
BBR-treated mice was achieved by proteolytic digestion of the protein adducts to the
corresponding cysteine conjugates. The representative chromatograms of the
cysteine adducts are shown in Figure 8. Epoxide-derived cysteine adduct A2
(Scheme 1) was detected in BBR-treated mice, while no such adduct was observed in
19

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
mice given 6-F-BBR.
8. Hepatic and plasma GSH depletion induced by BBR and 6-F-BBR in mice
Depletion of cellular GSH has been considered as a biomarker for the production
of reactive metabolites and early stage of cell injury. Time courses of hepatic and
plasma GSH levels were monitored in 6-F-BBR- and BBR-treated mice (Figure 9).
Hepatic and plasma GSH were dramatically depleted as much as 87 % and 92 %,
respectively, in mice after a hepatotoxic dose (50 mg/kg) of BBR, while 6-F-BBR
demonstrated much less potential to deplete hepatic and plasma GSH than BBR at the
same dose.
9. Pharmacokinetic behaviors
Plasma BBR and 6-F-BBR were monitored in rats given a single dose of BBR or
6-F-BBR at 10 mg/kg by oral gavage. The concentration-time (0-24 h) curves of the
compounds are shown in Figure 10, and the corresponding pharmacokinetic
parameters are presented in Table 3. Both BBR and 6-F-BBR revealed double-peak
property in their pharmacokinetic profiles. 6-F-BBR achieved a Cmax (54.59 g/mL)
and AUC (256.20 g/mL·h) as over twice as those of BBR-treated animals (26.80
g/mL and 123.66 g/mL·h, respectively). In addition, with the observed good
metabolic stability in RLMs, 6-F-BBR displayed a much lower in vivo clearance (0.04
L/h/kg) than that of BBR (0.09 L/h/kg). There was no significant difference in
half-life between BBR (3.86 h) and 6-F-BBR (2.59 h).
20

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Discussion
Though BBR shows great potential in the treatment of gout, it has not been
licensed for use in Europe, due to several reports of fatal idiosyncratic hepatic failure.
Idiosyncratic toxicities are considered to be associated with metabolic activation of
pro-toxicants. Our early study demonstrated that CYP3A-mediated epoxidation of
BBR at benzofuran ring was related to BBR-induced protein adduction and
hepatotoxicity (Wang et al., 2016a; Wang et al., 2016b). To avoid metabolic
activation-induced liver injury by BBR, we replaced the hydrogen at the metabolically
labile site with a halogen atom, which is a commonly used strategy to attenuate the
potential of P450-mediated metabolism (metabolic activation). Currently, 50% of
leading drugs on the pharmaceutical market are halogenated (Xu et al., 2014), and
there is a great potential to exploit halogens into drug design in medicinal chemistry.
In this study, three BBR derivatives substituted with a halogen (F, Cl, or Br) at
C6 of BBR, the site of epoxidation catalyzed by CYP3A, were designed and
synthesized to block the metabolic activation of BBR. As the initial step, we
evaluated the uricosuric efficacy of the resulting compounds (6-F-, Cl-, or Br-BBR) to
identify the most effective compound for further toxicity and metabolism
investigation. The efficacy evaluation started with assessment of inhibitory effect on
urate transport in hURAT1-transgenic cells, since uricosuric BBR is known to be a
potent inhibitor of hURAT1. 6-F-BBR was found to inhibit hURAT1 the most,
followed by BBR, 6-Cl-BBR, and 6-Br-BBR in order. In addition, we evaluated
their suppressive effects on SUA in rats with hyperuricemia (Table 1). No statistical
21

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
significance was observed in SUA levels in healthy rats treated with 6-F-, Cl-, or
Br-BBR, presumably due to the protective compensation reaction of healthy rats.
However, 6-F-BBR elicited more potency to decrease SUA level than 6-Cl-BBR (p <
0.001) and 6-Br-BBR (p < 0.05) in hyperuricemic rats. Both in vitro and in vivo
studies demonstrated that fluorinated BBR (6-F-BBR) was as effective as BBR and
more effective than the chlorinated and brominated BBR (6-Cl-BBR and 6-Br-BBR).
The observed favorable efficacy of 6-F-BBR was possibly attributed to the highly
electronegative nature of fluorine, the similarity in size of hydrogen (BBR) as that of
fluorine (fluorinated BBR), along with the low polarizability of fluorine (Hernandes et
al., 2010). These chemical characteristics of fluorine can productively influence
physicochemical and conformational properties of 6-F-BBR, which may contribute to
the improvement in oral absorption in rats (Table 3). As a result, this makes
6-F-BBR present dominated pharmacological properties.
As the core of the present study, hepatotoxicities of BBR and 6-F-BBR were
evaluated in vivo. We succeeded in development of a mouse model for the
assessment of hepatotoxicities of BBR and 6-F-BBR in mice pretreated with BSO.
BSO is known to have an ability to deplete GSH contents by inhibiting
γ-glutamylcysteine synthetase, a rate-limiting enzyme in the biosynthesis of GSH.
Significant elevations of serum ALT and AST were observed in BSO-pretreated mice
given BBR, while a minor or slight increase in serum ALT and AST was found in
animals treated with BBR alone (Figure 6). This made it possible to evaluate the
hepatotoxicity of BBR and hopefully its idiosyncratic liver injury, although the
22

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
mechanisms of the development of idiosyncratic toxicity remain unknown.
6-F-BBR was selected for further hepatotoxicity and metabolic activation
evaluation, since the fluorinated BBR was found to be the most effective uricosurics
among the compounds tested and be as effective as BBR (Table 1 and Figure 5).
The toxicity study was performed in mice pretreated with BSO. 6-F-BBR was found
to be much less hepatotoxic than BBR (Figure 6). The metabolism study
demonstrated that the metabolic activation (epoxidation) of 6-F-BBR was almost
abolished. Microsomal incubations of 6-F-BBR fortified with NAC revealed a
significant decrease in the formation of epoxide-derived NAC conjugate by nearly
99.7% compared with that of BBR, and no epoxide-derived protein adduction was
detected in mice given 6-F-BBR (Figures 7 and 8). It is likely that the introduced
fluorine blocked the site from metabolic epoxidation. Additionally, the electron
withdrawing effect of the fluorine decreased the electron density of the phenyl ring,
which disfavors the oxidation reaction. The observation of high volume of protein
adduction derived from BBR, combined with the finding of more potent
hepatotoxicity induced by BBR, led us to propose that epoxidation at C6 of BBR is a
key step for hepatotoxicity of BBR.
Our previous study demonstrated that the reaction of the epoxide metabolite(s) of
BBR with GSH contributes to the depletion of GSH (Wang et al., 2016b). The
present study found that treatment of BBR caused significant elevation in serum AST
activity in opposition to decrease in hepatic and plasma GSH content. On the
contrary, slight hepatotoxicity, along with attenuated GSH depletion, was observed in
23

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
mice given 6-F-BBR, indicating that the formation of the epoxide metabolite(s) was
slowed down by fluorine substitution. It should be noted that 6-F-BBR did not
protect mice completely from GSH depletion. A decrease of GSH level was also
observed during the first 30 min post administration of 6-F-BBR. This might result
from the formation of catechol (Scheme 1) and hydroquinone through
ipso-substitution of the dibromophenol ring (Kitagawara et al., 2015). We cannot
underestimate possible contribution of their oxidized forms (quinone metabolites, data
unpublished) which are known to be able to deplete GSH. The observed attenuated
GSH depletion at least partially resulted from the fluorine substitution. The
supporting evidence was that little metabolites derived from the epoxidation of
6-F-BBR were observed.
6-F-BBR was more stable than BBR in liver microsomes, particularly in mouse
liver microsomes (as much as 10-fold more stable, Table 2). This indicates that the
incorporation of fluorine kept the molecule from P450-mediated metabolism. It is
believed that the electron-withdrawing property of the fluorine substituent deactivated
the aromatic system towards metabolic degradation by P450 enzymes. Besides
epoxidation of BBR at benzofuran ring, BBR was suggested to be oxidized to a
quinone metabolite via sequential oxidation of 6-OH-BBR (McDonald and Rettie,
2007). Additionally, Naoki and coworkers demonstrated that BBR was metabolized
to a quinone methide derived from 1′,6-dihydroxy BBR (Cho et al., 2017). We
failed to detect the formation of such reactive metabolites directly from BBR in liver
microsomal systems, indicating that the two metabolic pathways are minor in
24

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
comparison with the epoxidation.
The observed improvement of 6-F-BBR
metabolic stability probably resulted mainly from retarding of the epoxidation of the
benzrofuran ring, and unlikely from that of the two minor pathways.
The metabolic stability studies demonstrated that the t1/2 of 6-F-BBR increased
twice as many as that of BBR in RLMs, and over two-fold increases in Cmax and AUC
values were observed in 6-F-BBR-treated rats relative to those of animals treated with
BBR at the same dose. This might result from the improvement of oral absorption,
presumably due to the increase of membrane permeability by fluorine substitution.
Permeability is one of the major considerations in the lead optimization of oral drugs.
Many drugs with aromatic fluorination are prone to penetrate cell membranes, thereby
increasing the intracellular concentration (Smart, 2001; Murphy and Sandford, 2015).
Furthermore, 6-F-BBR was much more stable than BBR in MLMs with t1/2 > 60 min.
It might be reasonable to assume that exposures of 6-F-BBR would be greater than
that of BBR in mice. Surprisingly, 6-F-BBR displayed much lower hepatotoxicity
than BBR (Figure 6). This suggests that the observed toxicity of BBR was induced
by its metabolite but not the parent compound itself. Interestingly, with the observed
good metabolic stability in RLMs, 6-F-BBR displayed a slightly increased elimination
rate than that of BBR in vivo (Figure 10). Explanation for the observed discrepancy
could arise from elevated phaseⅡmetabolism, such as glucuronidation or sulfation,
involved in biotransformation of 6-F-BBR. The resulting polar conjugates could be
eliminated quickly.
Conclusively, replacement of the hydrogen in the metabolically labile site by a
25

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
fluorine atom significantly attenuated the hepatotoxicity of BBR but did not alter its
uricosuric efficacy. The present work provided a solid evidence that the formation
of the epoxide(s) is a key step for the development of BBR-induced hepatotoxicity.
Additionally, the fluorinated BBR was identified as effective as BBR, and this study is
an excellent example of drug metabolism-guided rational drug design.
26

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Authorship Contributions:
Participated in research design: Zheng, S.J. Wang, and Pang.
Conducted experiments: H. Wang, Peng, Zhang, Lan, Y.F. Zhao, and X. Wang.
Performed data analysis: H. Wang, H.M. Zhao, W.B. Wang, and Peng.
Wrote or contributed to the writing of the manuscript: H. Wang and Zheng.
27

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
References
Abhishek A, Roddy E, and Doherty M (2017) Gout-a guide for the general and
acute physicians. Clin Med (Lond) 17: 54-59.
Ahn SO, Ohtomo S, Kiyokawa J, Nakagawa T, Yamane M, Lee KJ, Kim KH,
Kim BH, Tanaka J, Kawabe Y, and Horiba N (2016) Stronger uricosuric effects
of the novel selective URAT1 inhibitor UR-1102 lowered plasma urate in tufted
capuchin monkeys to a greater extent than benzbomarone. J Pharmacol Exp
Ther 357: 157-166.
Arai M, Yokosuka O, Fujiwara K, Kojima H, Kanda T, Hirasawa H, and Saisho H
(2002) Fulminant hepatic failure associated with benzbromarone treatment: a
case report. J Gastroenterol Hepatol 17: 625-626.
Bacsa I, Jójárt R, Schneider G, Wölfling J, Maróti P, Herman BE, Szécsi M, and
Mernyák E (2015) Synthesis of A-ring halogenated 13α-estrone derivatives as
potential 17β-HSD1 inhibitors. Steroids 104: 230-236.
Baker JF and Schumacher HR (2010) Update on gout and hyperuricemia. Int J
Clin Pract 64: 371-377.
Byun JH, Kim H, Kim Y, Mook-Jung I, Kim DJ, Lee WK, and Yoo KH (2008)
Aminostyrylbenzofuran derivatives as potent inhibitors for Abeta fibril
formation. Bioorganic & Medicinal Chemistry Letters 18: 5591-5593.
Cho N, Kobayashi K, Yoshida M, Kogure N, Takayama H, and Chiba K (2017)
Identification of novel glutathione adducts of benzbromarone in human liver
microsomes. Drug Metab Dispos 32: 46-52.
28

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Choi HK, Mount DB, and Reginato AM (2005) Pathogenesis of gout. Ann Intern
Med 143: 499-516.
Dekker SJ, Zhang Y, Vos JC, Vermeulen NP, and Commandeur JN (2016)
Different reactive metabolites of nevirapine require distinct glutathione
S-transferase isoforms for bioinactivation. Chem Res Toxicol 29: 2136-2144.
Doherty M (2009) New insights into the epidemiology of gout. Rheumatology
(Oxford) 48 Suppl 2: ii2-ii8.
Fernandes EA, Bergamaschi SB, Rodrigues TC, Dias GC, Malmann R, Ramos
GM, and Monteiro SS (2016) Relevant aspects of imaging in the diagnosis and
management of gout. Rev Bras Reumatol doi: 10.1016/j.rbr.2016.03.006.
Ford MC and Ho PS (2016) Computational tools to model halogen bonds in
medicinal chemistry. J Med Chem 59: 1655-1670.
Haidari F, Rashidi MR, and Mohammad-Shahi M (2012) Effects of orange juice
and hesperetin on serum paraoxonase activity and lipid profile in
hyperuricemic rats. Bioimpacts 2: 39-45.
Hernandes MZ, Cavalcanti SM, Moreira DR, de Azevedo Junior WF, and Leite
AC (2010) Halogen atoms in the modern medicinal chemistry: hints for the
drug design. Curr Drug Tragets 11: 303-314.
Hu K and Jeong JH (2006) A convergent synthetic study of biologically active
benzofuran derivatives. Arch Pharm Res 29: 476-478.
Keenan RT and Schlesinger N (2016) New and pipeline drugs for gout. Curr
Rheumatol Rep 18: 32.
29

DMD Fast Forward. Published on October 11, 2017 as DOI: 10.1124/dmd.117.077818
This article has not been copyedited and formatted. The final version may differ from this version.
DMD #
Kitagawara Y, Ohe T, Tachibana K, Takahashi K, Nakamura S, and Mashino T
(2015) Novel bioactivation pathway of benzbromarone mediated by
cytochrome P450. Drug Metab Dispos 43:1303-1306.
Kou Y, Li Y, Ma H, Li W, Li R, and Dang Z (2016) Uric acid lowering effect of
Tibetan Medicine RuPeng15 powder in animal models of hyperuricemia. J
Tradit Chin Med 36: 205-210.
Kydd AS, Seth R, Buchbinder R, Edwards CJ, and Bombardier C (2014)
Uricosuric medications for chronic gout. Cochrane Database Syst Rev
CD010457.
Lee MH, Graham GG, Williams KM, and Day RO (2008) A benefit-risk
assessment of benzbromarone in the treatment of gout. Was its withdrawal from
the market in the best interest of patients? Drug Saf 31: 643-665.
Leung L, Kalgutkar AS, and Obach RS (2012) Metabolic activation in
drug-induced liver injury. Drug Metab Rev 44: 18-33.
Lin G, Tang J, Liu XQ, Jiang Y, and Zheng J (2007) Deacetylclivorine: a
gender-selective metabolite of clivorine formed in female Sprague-Dawley rat
liver microsomes. Drug Metab Dispos 35: 607-613.
Lu Y, Wang Y, and Zhu W (2010) Nonbonding interactions of organic halogens in
biological systems: implications for drug discovery and biomolecular design.
Phys Chem Chem Phys 12: 4543-4551.
McDonald MG and Rettie AE (2007) Sequential metabolism and bioactivation of
the hepatotoxin benzbromarone: formation of glutathione adducts from a
30