Predominant contributions of carboxylesterase 1 and 2 in hydrolysis of anordrin in humans

Article abstract of DOI:10.1080/00498254.2017.1333658

1.Anordrin (2α, 17α-diethynyl-A-nor-5α-androstane-2β, 17β-diol diproprionate) is post-coital contraceptive drug that is on the market in China for more than 30 years. This study aims to elucidate enzymes involved in anordrin hydrolysis, and to evaluate the significant role of carboxylesterases in anordrin hydrolysis in humans. 2.Human liver and intestinal microsomes, recombinant human carboxylesterase were selected as enzyme sources. In human liver microsomes, intrinsic clearance was 684 ± 83 μL/min/mg protein, which was considerably higher than the value of intestine microsomes (94.6 ± 13.3 μL/min/mg protein). Carboxylesterase (CES) 1 has more contribution than CES2 in human liver. 3.Inhibition studies were performed using representative esterase inhibitors to confirm esterase isoforms involved in anordrin hydrolysis. Simvastatin strongly inhibited hydrolytic process of anordrin in liver and intestine microsomes, with IC₅₀ values of 10.9 ± 0.1 and 6.94 ± 0.03 μM, respectively. 4.The present study investigated for the first time hydrolytic enzyme phenotypes of anordrin. Anordrin is predominantly catalyzed by CES1 and CES2 to generate the main active metabolite, anordiol. Moreover, anordrin and its metabolite anordiol can be altered by esterase inhibitors, such as simvastatin, upon exposure in vivo.

Full text of DOI:10.1080/00498254.2017.1333658

Xenobiotica
the fate of foreign compounds in biological systems
ISSN: 0049-8254 (Print) 1366-5928 (Online) Journal homepage: http://www.tandfonline.com/loi/ixen20
Predominant contributions of carboxylesterase 1
and 2 in hydrolysis of anordrin in humans
Jinfang Jiang, Xiaoyan Chen & Dafang Zhong
To cite this article: Jinfang Jiang, Xiaoyan Chen & Dafang Zhong (2017): Predominant
contributions of carboxylesterase 1 and 2 in hydrolysis of anordrin in humans, Xenobiotica, DOI:
10.1080/00498254.2017.1333658
To link to this article: http://dx.doi.org/10.1080/00498254.2017.1333658
Accepted author version posted online: 22
May 2017.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ixen20
Download by: [University of Florida]
Date: 24 May 2017, At: 07:54

Predominant contributions of carboxylesterase 1 and 2 in hydrolysis
of anordrin in humans
1
,2
1,2
1,2*
Jinfang Jiang , Xiaoyan Chen , Dafang Zhong
Abstract
. Anordrin (2α, 17α-diethynyl-A-nor-5α-androstane-2β, 17β-diol diproprionate) is
1
post-coital contraceptive drug that is on the market in China for more than 30 years.
This study aims to elucidate enzymes involved in anordrin hydrolysis, and to evaluate
the significant role of carboxylesterases in anordrin hydrolysis in humans.
2
. Human liver and intestinal microsomes, recombinant human carboxylesterase were
selected as enzyme sources. In human liver microsomes, intrinsic clearance was 684 ±
3 μL/min/mg protein, which was considerably higher than the value of intestine
8
microsomes (94.6 ± 13.3 μL/min/mg protein). Carboxylesterase (CES) 1 has more
contribution than CES2 in human liver.
3
. Inhibition studies were performed using representative esterase inhibitors to
confirm esterase isoforms involved in anordrin hydrolysis. Simvastatin strongly
inhibited hydrolytic process of anordrin in liver and intestine microsomes, with IC50
values of 10.9 ± 0.1 and 6.94 ± 0.03 μM, respectively.
4
. The present study investigated for the first time hydrolytic enzyme phenotypes of
anordrin. Anordrin is predominantly catalyzed by CES1 and CES2 to generate the
main active metabolite, anordiol. Moreover, anordrin and its metabolite anordiol can
be altered by esterase inhibitors, such as simvastatin, upon exposure in vivo.
Key words: anordrin; anordiol; carboxylesterase; hydrolysis in humans;
drug-drug interaction
1
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese
2
Academy of Sciences, 501 Haike Road, Shanghai 201203, China; University of Chinese
Academy of Sciences, No.19A Yuquan Road, Beijing 100049, China.
*
To whom correspondence should be addressed.
Tel/Fax: 86-21-50800738, E-mail: dfzhong@simm.ac.cn

Introduction
Anordrin (2α, 17α-diethynyl-A-nor-5α-androstane-2β, 17β-diol diproprionate) is
domestically synthetic, steroid-like estrogen that is used in China as a post-coital
contraceptive for more than 30 years (Chih-ping et al., 1976; Gu et al., 1984).
Anordrin is also marketed in combination with mifepristone (Sang et al., 1999).
Anordrin is a selective estrogen receptor modulator, with both weak estrogenic and
antiestrogenic properties (Mehta et al., 1981; Mehta et al., 1982; Chatterton et al.,
1
989). In cynomolgus monkeys, rapid hydrolysis of dipropionate esters is the main
metabolic pathway, resulting in forming of unesterified anordriol (Chatterton et al.,
994) (Figure 1). Anordriol shows 36-fold higher affinity than anordrin for cytosol
1
estrogen receptor of rat uterus in vitro, and is the main component responsible for
resistance to pregnancy (Gu et al., 1984), thus anordrin can be called as a prodrug for
the contraceptive indication. Moreover, α-anordrin shows inhibitory effect on multiple
kinds of tumors (Bin et al., 1989; Ma et al., 2000), indicating that anordrin can be
developed for variety of potential applications.
Esterases hydrolyze compounds containing ester, amide, and thioester bonds,
which contribute to metabolism of more than 10% of therapeutic drugs. In humans,
carboxylesterases (CES) are members of serine hydrolase superfamily. CES are
mainly categorized into five families: namely CES1, CES2, CES3, CES4A, and
CES5A (Holmes et al., 2010). Among these families, CES1 and CES2 reportedly
catalyze various drugs, such as imidapril (Takahashi et al., 2008), oseltamivir (Zhu
and Markowitz, 2013), and irinotecan (Khanna et al., 2000), and have different tissue
distribution, substrate specificity, immunological properties, and gene regulation.
CES1 is mainly expressed in liver, whereas CES2 is in intestine (Imai, 2006),
however, both share considerable individual differences in expression (Ross et al.,
2
012). Arylacetamide deacetylase (AADAC) is another serine hydrolase that is highly
expressed in livers; this enzyme is responsible for hydrolysis of flutamide (Watanabe
et al., 2009), phenacetin (Watanabe et al., 2010), rifamycin (Nakajima et al., 2011)

and others. CES1 generally prefers substrates containing large acyl and small
alcohol/amine moieties, whereas CES2 prefers substrates having large alcohol/amine
and small acyl moieties, AADAC and CES2 have similar substrate specificity
(Fukami and Yokoi, 2012; Fukami et al., 2015).
In previous study by Chatterton et al. on the pharmacokinetics of anordrin in
cynomolgus monkeys, dipropionate esters of anordrin were rapidly hydrolyzed to
dihydroxy metabolite anordriol after intravenous administration of 0.2 mg per kg
anordrin, half-life was about 3.5 min (Chatterton et al., 1994), monoester metabolite
was not detected and no other metabolites were found. Conversion also occurred in
sample storage, hydrolysis was hypothesized to be mediated by butyrylcholinesterase
in serum (Chatterton et al., 1994), however, this assumption was not confirmed, and
no more information can be accessed to understand metabolic profile in humans. As
shown in Figure 1, anordrin contains large alcohol group and small propionyl moiety,
this chemical structure is in accord with substrate specificity of CES2. Therefore, we
speculate that CES2 is involved in anordrin hydrolysis in humans. From our previous
results, hydrolytic process can be effectively inhibited by simvastatin, suggesting that
drug-drug interactions (DDI) may occur to cause efficacy reduction.
Although anordrin is used in China for more than 30 years, its metabolic enzyme
phenotype remains to be clarified. Hence, this study aims to identify enzymes
involved in anordrin hydrolysis and to evaluate role of CES in such process in
humans.
Materials and methods
Chemicals and reagents Anordrin (98% purity) was provided by Shanghai Aoqi
Pharmaceutical and Technology Co., Ltd. (Shanghai, China). Radioactive
3
[
H]-anordiol was provided by Shanghai Institute of Applied Physics. (Shanghai,
China). Levonorgestrel (LNG) was purchased from TCI (Shanghai, China).
Fenofibrate, phenacetin, procaine, loperamide, and bis-p-nitrophenyl phosphate
(BNPP) were purchased from Sigma-Aldrich (St. Louis, MO). Digitonin and

simvastatin were purchased from Dalian Meilun Biology and Technology Co., Ltd.
Dalian, China). Human liver microsomes (HLM, from Caucasian), human liver
(
cytosol (HLC), and human intestine microsomes (HIM) were purchased from BD
Gentest (Woburn, MA). Recombinant human CES1 and CES2 were purchased from
Cypex Ltd (10 mg/ml, Scotland, UK), and recombinant human AADAC was
purchased from CUSABIO Biotech Co., Ltd. (Wuhan, China). Deionized water (18.2
mΩ, TOC ≤ 50 ppb) was purified using Millipore Milli-Q Gradient Water Purification
System (Molsheim, France). All other chemicals were of analytical grade.
Anordrin hydrolase activity
Anordrin hydrolase activities in different enzyme sources were determined as
follows: incubation mixture (final volume of 0.1 mL) contained 100 mM potassium
phosphate buffer (containing 3.2 mM MgCl , pH 7.4) and enzyme sources (HLM: 0.1
2
mg/mL，HIM and HLC: 0.5 mg/mL; recombinant human CES1, CES2, and AADAC:
0
.1 mg/mL, and fresh human blood and plasma). Recombinant enzymes were diluted
with Tris-HCl buffer (50 mM Tris-HCl buffer, pH 7.4). After 5 min of preincubation
at 37°C, reactions were initiated by addition of anordrin (final concentration of 1 μM).
Final concentration of organic reagent was < 1% in incubation mixture. After 60 min
(10 min for HLM, incubation extended to 4 h in plasma and 6 h in blood) of
incubation, reactions were terminated by adding equal volume of ice-cold acetonitrile.
Control samples were incubated with thermally inactivated enzymes.
Substrate concentrations of phenacetin, fenofibrate, and procaine were all set to
2
00 µM. We confirmed linear hydrolytic rates with respect to protein concentrations
and incubation time. For phenacetin, enzyme source was 0.4 mg/mL, incubation time
was 30 min; for fenofibrate, enzyme sources was 0.05 mg/mL, incubation time was 5
min; for procaine, enzyme sources was 0.1 mg/mL, incubation time was 30 min.
Kinetics of anordrin hydrolysis
The parent depletion method was used to determine kinetics of anordrin
hydrolysis in different enzyme sources. In preliminary study, we confirmed linear

depletion rate of anordrin with respect to protein concentrations (<1.5 mg/mL HLM
and HIM, <1.0 mg/mL CES1 and CES2) and incubation time (<60 min in HLM and
HIM, <2 h in CES1 and CES2). For kinetic analysis, selected final conditions were as
follows: HLM : 0.1 mg/mL, 10 min；HIM: 0.2 mg/mL, 40 min；CES1 and CES2: 0.2
mg/mL, 30 min. Anordrin concentration ranged from 0.1 μM to 20 μM. Incubation
procedure was similar to method described for anordrin hydrolysis activity. Control
samples were incubated with thermally inactivated enzymes. Data are presented as
averages of triplicate experiments. Kinetic constants were obtained by fitting
Michaelis-Menten equation shown in Equation 1 to experimental data by using
nonlinear regression (Prism 5.0; GraphPad Software Inc.):
v = Vmax × S/(K + S)
(1)
m
where v is reaction velocity, Vmax is maximum velocity, K is Michaelis constant
m
(substrate concentration at 0.5Vmax), and S is substrate concentration. In vitro CLint
was calculated as Vmax/Km.
Studies on inhibition of anordrin hydrolase activity
To confirm esterase isoform involved in anordrin hydrolysis in humans, we
conducted inhibition studies on HIM, HLM, recombinant CES1 and CES2 using
representative esterase inhibitors, including digitonin, loperamide, simvastatin, and
BNPP. BNPP is nonselective inhibitor of esterases (Watanabe et al., 2009), and
digitonin and loperamide specifically inhibit CES1 and CES2 (Shimizu et al., 2014),
respectively. Simvastatin strongly inhibits CES1 and CES2 activities (Shimizu et al.,
2
014). Inhibitors (digitonin, loperamide, and simvastatin: 20 µM; BNPP: 1 mM) were
preincubated with different enzyme sources for 5 min at 37°C, and reactions were
initiated by adding anordrin solution (final 1 µM). After 30 min (10 min for HLM) of
incubation, reactions were terminated by adding equal volume of ice-cold acetonitrile.
Control samples were incubated with thermally inactivated enzymes.
Contribution of CES1 and CES2 to anordrin hydrolase activity in HLM
For relative contribution estimation, relative activity factor (RAF) method and

relative expression factor (REF) method are the two commonly used methods. For the
former one, RAF was applied as the ratio of activity values to evaluate contribution of
CES1 and CES2 to anordrin hydrolysis in HLM. As described previously by
Watanabe et al.(Watanabe et al., 2010), RAF values for CES1 (RAF
，
) were
CES1 HLM
determined as ratios of probe substrate fenofibrate hydrolase activities in HLM to
value of recombinant human CES1. Meanwhile, RAF values for CES2 (RAF
，
CES2 HLM
)
were determined as ratios of probe substrate procaine hydrolase activities in HLM to
value of recombinant human CES2. The following formlas were used to predict
anordrin hydrolase activities by CES1 (VCES1
VCES1，HLM = Vrec-CES1 × RAFCES1，HLM
VCES2，HLM = Vrec-CES2 × RAFCES2，HLM
，
，
HLM), and CES2 (VCES2 HLM) in HLM:
(2)
(3)
Vrec-CES1 and Vrec-CES2 are anordrin hydrolase activities by recombinant CES1 and
CES2, respectively. The following equations were used to calculate contributions of
CES1 and CES2 to anordrin hydrolase activities in HLM:
CES1 in HLM(%) = (VCES1，HLM /VHLM) × 100 (4)
CES2 in HLM(%) = (VCES2，HLM /VHLM) × 100 (5)
where VHLM value is observed anordrin hydrolase activities in HLM.
Another method for estimation of esterases to the anordrin hydrolysis in HLM is
REF. Instead of simple clearance values used in RAF method, enzyme contents, or
protein levels, was taken into consideration. This method has been applied to liver
uptake transporters (Hirano et al., 2006). The ratio of the expression level of each
esterase in HLM to that in recombinant enzyme system (pmol/mg protein) was
defined as R . The hydrolytic clearance by each esterase in HLM was separately
exp
calculated by multiplying the clearance of anordrin in recombinant enzyme (CLCES,test
)
by Rexp as described in the following equation:
CLHLM,CES1 = CLCES1,test × Rexp,CES1
CLHLM,CES2 = CLCES2,test × Rexp,CES2
(6)
(7)
The relative contribution (%) of each esterase to the hydrolysis in HLM was

defined by the ratio of CLHLM,CES for target esterase to that of the sum of CLHLM,CES
for CES1 and CES2. The expression level of each esterase in HLM was calculated
based on LC-MS/MS study described by Sato et al.(Sato et al., 2012).
LC-MS/MS bioanalytical method
Calibration curve concentration ranged from 0.500 nmol/L to 1000 nmol/L for
anordrin. LNG was used as internal standard. Calibration curve was fitted with linear
2
least-squares regression model (y =1/x ). Calibration samples were extracted
alongside study samples. Aliquots (25 μL) of either calibrant or study samples were
added to 1.5 mL polypropylene tube, a total of 25 μL of internal standard solution
(
LNG, 100 ng/mL) was then added, followed by 150 μL acetonitrile to precipitate
proteins. System was vortexed and centrifuged at 11,000 × g for 5 min. Finally, the
supernatant was injected into LC-MS/MS system.
LC-MS/MS data were acquired using LC-30AD liquid chromatographic system
(Shimadzu, Kyoto, Japan) coupled to Triple Quad 5500 mass spectrometer (AB
Sciex). Analyst V1.6.2 software (AB Sciex) was used for data processing.
Chromatographic separation was conducted on Phenomenex Luna 5u PFP (2) (50 ×
2
.0 mm I.D., 5 m), which was maintained at 40°C. Mobile phases used for gradient
elution were 40% (solvent A) 5 mM ammonium acetate–formic acid (100/0.1, v/v)
and 60% (solvent B) methanol. Initial mobile phase was 60%B for 0.7 min, followed
by linear increase to 95% over 0.1 min and then maintained for 1.2 min. Column was
equilibrated the with initial mobile phase. The total run time was 3.0 min, and flow
rate was 0.50 mL/min. Anordrin and LNG exhibited peak retention times of 2.5 and
2
.0 min (Figure 2), respectively. Mass spectrometer equipped with electrospray
ionization source was operated in positive multiple reaction monitoring mode. Ion
spray voltage and source temperature were set to 5,500 V and 500°C, respectively.
Nebulizer gas, heater gas, curtain gas, and collision-activated dissociation gas were
optimized at 50, 50, 30, and 9 psi, respectively. Samples were detected using multiple
reaction monitoring, and parent-to-product transitions were as follows: m/z

4
56.3→291.2 for anordrin, collision energy (CE) 15 eV; LNG m/z 313.2→245.2, CE
8 eV; fenofibrate hydrolysis metabolite m/z 319.1→233.0, CE 30 eV; procaine
1
hydrolysis metabolite m/z 138.1→120.0, CE 17 eV; phenacetin hydrolysis metabolite
p-phenetidine m/z 138.1→110.1, CE 21 eV.
Results
Anordrin hydrolase activity
Chatterton et al. have found that after intravenous administration of radioactive
14
[
3- C]anordrin to monkeys, the majority of the radioactivity in plasma after 15 min
14
was found as metabolite [ C]anordiol (Chatterton et al., 1994), i.e., the propionate
esters were cleaved. The monoesters were not present in detectable amounts
(Chatterton et al., 1994). After 4 h, the radioactivity in plasma was only formed as
14
[
C]anordiol (Chatterton et al., 1994) in radiochromatogram. It suggested that the
major metabolic pathway in vivo is hydrolysis to the main metabolite anordiol.
In addition, our laboratory has been testing a compound A (Figure 4), an
analogue of anordrin, which only differs in ester bonds: the side chain of ester of
3
compound A contains 4 carbons (Figure 4). After radioactive [ H]A was orally given
3
to ICR mice, the radioactivity in plasma was mainly formed as [ H]A and
3
[
H]anordiol，the sum of them accounted for >90% of total radioactivity in plasma,
3
while [ H]anordiol accounted for >75% of total radioactivity. Besides, in HLM
3
incubation system, [ H]anordiol was no longer metabolized to other substances. These
results suggested that the main metabolic pathway of the two compounds were
hydrolysis to anordiol, and the hydrolytic metabolite anordiol was difficult to convert
to other metabolites.
In the present study, anordiol was the only detected metabolite after incubation
with different enzyme sources under our experimental conditions. The
monopropionate ester of anordrin was not detectable under this experimental
condition, it is probably due to the rapid conversion to the dihydroxy compound,
anordiol. Figure 3 shows MS fragmentation profiles. m/z 456 and m/z 344 were parent

+
ions of anordrin and anordiol, respectively, adducted with NH , and m/z 291 was
4
typical fragment of steroidal structure. Meanwhile, owing to low ionization efficiency
of anordriol, detection sensitivity was not reached in LC-MS/MS (low limit of
detection ~ 300 nM). Because the depletion of anordrin represents formation of
metabolite anordiol, thus parent drug depletion method was used to determine
hydrolase activity of anordrin in various enzyme sources.
Probe substrates, including fenofibrate, procaine, and phenacetin were used as
markers to confirm hydrolase activities of these enzyme sources. Hydrolytic rates of
three substrates were 92.8, 3.34, and 0.0504 nmol/min/mg protein in recombinant
CES1, CES2, and AADAC, respectively, with values of 213, 4.01, and 0.102
nmol/min/mg protein in HLM, respectively (Table 2).
Anordrin is stable in human blood for 6 h and human plasma for 4 h (Figure 5).
Therefore, cholinesterase, paraoxonase, and albumin esterase were not assumed to be
involved in anordrin hydrolysis. Hydrolytic rate in HLM was approximately 500
pmol/min/mg protein, showing highest catalytic efficiency and considerably higher
values in HIM and HLC (Figure 5). CES1 and CES2 had comparable hydrolytic rates
(approximately 150 pmol/min/mg protein), whereas AADAC was proven not to be
responsible for anordrin hydrolysis.
Kinetics of anordrin hydrolysis
Hydrolase activity was detected through parent drug depletion, and data for
activities followed Michaelis-Menten equation (Figure 6). In HLM and HIM, the Km
values were 2.83 ± 0.52 and 2.24 ± 0.52 μM, respectively, indicating comparable
enzyme affinity. Calculated in vitro intrinsic clearances were 684 ± 83 and 94.6 ± 13.3
μL/min/mg protein in HLM and HIM, respectively. This study is the first to report on
kinetics of anordrin hydrolysis in HLM and HIM, and results showed that liver has
higher contribution to anordrin hydrolysis than intestine. CES1 and CES2 possessed
CL values of 117 ± 13 and 197 ± 17 μL/min/mg protein, respectively. Table 1
int
presents detailed Michaelis-Menten constants.

Inhibition studies on anordrin hydrolase activity
To evaluate contribution of each esterase to anordrin hydrolysis in human
intestine and liver, we investigated effects of various chemical inhibitors on anordrin
hydrolase activities (Figure 7). Nonspecific esterase inhibitor BNPP efficiently
inhibited anordrin hydrolysis in different enzyme sources. Digitonin and loperamide
are specific inhibitors of CES1 and CES2 (Shimizu et al., 2014), respectively,
effectively inhibiting anordrin hydrolysis in recombinant enzymes. However, these
drugs did not show inhibitory effects on HLM and HIM, it is probably because HLM
is a mixture of CES1 and CES2, and the weak inhibitory potency of loperamide
toward CES2 in HIM might be caused by the interference of some proteins.
Simvastatin is an effective esterase inhibitor according to several studies (Shimizu et
al., 2014; Wang et al., 2015). This finding is consistent with our study, which showed
that simvastatin strongly inhibited anordrin hydrolysis both in HLM and HIM at IC50
values were 10.9 ± 0.1 and 6.94 ± 0.03 μM (Figure 8), respectively. Although plasma
concentration of simvastatin was relatively low (Ahmed et al., 2013), liver
concentration was much higher; for example, liver-to-plasma concentration ratio was
approximately 10 to 20 in wild-type mice (Higgins et al., 2014), indicating that effect
of exposure of anordrin can be changed by esterase inhibitors, such as simvastatin.
Contribution of CES1 and CES2 to anordrin hydrolase activity in HLM
To estimate contribution of CES1 and CES2 to anordrin hydrolase activity in
HLM, RAF and REF values were calculated. For the RAF method, in HLM and CES1,
values of fenofibrate hydrolase activity were 213 and 92.8 nmol/min/mg protein,
respectively, and calculated RAFCES1，HLM was 2.30. In HLM and CES2, values of
procaine hydrolase activity were 4.01 and 3.34 nmol/min/mg protein, respectively,
and calculated RAF
，
was 1.20. Observed Vrec-CES1 and Vrec-CES2 were 123 and
CES2 HLM
1
37 pmol/min/mg protein, respectively. Thus, by using Equations (2) and (3),
calculated VCES1 and VCES2 were 283 and 164 pmol/min/mg protein,
，
，
HLM
HLM
respectively. Finally, the contribution ratios of CES1 and CES2 in HLM were

estimated according to Equations (4) and (5), resulting in CES1 and CES2 values of
5
9.4% and 34.6%, respectively (Table 2).
For the REF method, according to Sato’s research, in pooled HLMs, the overall
averages of quantified concentrations of CES1 and CES2 were 363 and 22.2 pmol/mg
protein (Sato et al., 2012). In this study, concentration of pooled HLMs was 0.1
mg/mL, while concentration of CES1 or CES2 in the samples was 0.2 mg/mL.
Molecular weights of CES1 and CES2 are about 60 kDa, 1 mg recombinant CES1 or
CES2 protein contains 16670 pmol moleculs. Rexp,CES1 and Rexp,CES2 were calculated to
-4
be 0.01089 and 6.66×10 , respectively. CL
and CLCES2,test was the observed
CES1,test
value of 117 and 197 μL/min/mg protein (Table 1), respectively. By the Equations (6)
and (7), the final relative contribution of CES1 and CES2 in HLM was estimated to be
9
0.7% and 9.3%, respectively.
For relative contribution estimation, REF method based on enzyme contents (or
protein levels) may be more accurate than the RAF method based on simple clearance
values.
Discussion
As increasing number of drugs are designed to avoid metabolism by CYP450, we
witnessed more in-depth research into esterases in humans, indicating their important
role in drug ADME/T evaluation. In present study, human blood and plasma, liver
microsomes, and intestine microsomes were selected as matrices, because they
contain different esterase subtypes. Butyrylcholinesterase, paraoxonase, and albumin
esterase are present in human plasma, whereas CES is absent (Li et al., 2005). Human
liver and intestine are main tissues responsible for drug metabolism and contain
abundant CES activities; human liver microsomes (HLM) is also standard matrix for
investigation of drug hydrolytic properties in humans (Watanabe et al., 2009;
Watanabe et al., 2010). In present study, after incubation with human primary
hepatocytes (data not shown), HLM, and HIM, anordiol was the only detected
metabolite (Figure 3). Anordrin was rapidly hydrolyzed in HLM (approximately 500

pmol/min/mg protein) but was stable in human blood and plasma (Figure 5).
Hydrolytic rate in HIM was about 20% of value for HLM. We further observed that
CES1 and CES2 were involved in anordrin hydrolysis, but not AADAC (Figure 5).
Chatterton et al.(Chatterton et al., 1994) and our experiments on the radioactive
compounds have proved that the main metabolic pathway of anordrin was hydrolysis
to anordiol. Thus the parent depletion method was used to determine kinetics of
anordrin hydrolysis in different enzyme sources. We performed kinetic analyses of
anordrin hydrolase activity by HLM, HIM, recombinant human CES1, and CES2
(Figure 6), showing that highest catalytic potency was detected in HLM (Table 1).
Obtained K value in HLM was similar to that in HIM (2.83 ± 0.52 versus 2.24 ± 0.52
m
μM), with CL values of 684 ± 83 and 94.6 ± 13.3 μL/min/mg protein, respectively.
int
These findings indicate that liver plays more important role than intestine in anordrin
hydrolysis. CES1 and CES2 kinetics show relatively similar enzyme affinity (13.6 ±
2
.8 versus 16.9 ± 3.0) and comparable intrinsic clearance (117 ± 13 versus 197 ± 17
μL/min/mg protein), suggesting that CES1 and CES2 play important role in anordrin
hydrolysis both in human liver and intestine. By using equations for RAF values of
each enzyme, contribution of CES1 and CES2 was 59.4% and 34.6%, respectively.
While by the REF method, the estimated contribution of CES1 and CES2 was 90.7%
and 9.3%, respectively. For relative contribution estimation, REF method based on
enzyme contents (or protein levels) may be more accurate than the RAF method based
on simple clearance values. These results suggest that CES1 and CES2 are the
predominant enzymes responsible for anordrin hydrolysis in humans, with CES1
having more contribution than CES2 in liver.
Aside from liver, the small intestine is also an important organ involved in the
first-pass metabolism after oral dosing. Although contribution of the liver to anordrin
hydrolysis was estimated to be higher than that of intestine, small intestine is logically
important for hydrolysis. Thus, direct conversion of anordrin to anordiol possibly
occur within intestine because of high expression of CES2 in human intestine

epithelial cells. Remaining parent compound is probably hydrolyzed by hepatic
esterases.
We performed inhibition studies by using several inhibitors to estimate
contribution of CES to anordrin hydrolysis (Figure 7). CES1 specific inhibitor
digitonin and CES2 specific inhibitor loperamide did not exert any inhibitory effect
on anordrin hydrolysis in HLM and HIM, which could be explained as the
compensation effect of CES1 and CES2 in HLM, while the weak inhibitory potency
of loperamide toward CES2 in HIM might be caused by the interference of some
proteins. The combination of specific inhibitors (digitonin and loperamide) in HLM
and HIM did not produce certain degrees of inhibition. In recombinant CES1 and
CES2, the two specific inhibitors effectively inhibited anordrin hydrolysis
(approximately 60% inhibition, Figure 7). Moreover, simvastatin strongly inhibited
anordrin hydrolysis both in HLM and HIM, with estimated IC values of 10.9 ± 0.1
5
0
and 6.94 ± 0.03 μM, respectively. Concentration in humans is relatively low (Ahmed
et al., 2013), but simvastatin is possibly concentrated in human liver (Higgins et al.,
2
014), this finding suggests that DDI may occur in vivo between anordrin and esterase
inhibitors, such as simvastatin.
The hydrolytic metabolite anordiol is the active compound for the termination of
early pregnancy, thus the parent drug anordrin can be called as a prodrug. Hydrolysis
is the bioactivation pathway in humans, and that this hydrolytic pathway is directly
related with in vivo exposure of active metabolite anordiol and pharmacological
efficacy. In this study, we found that some esterase inhibitors played some inhibition
effects on anordrin hydrolysis. Not limited to simvastatin, if anordrin is
coadministered with such inhibitors, the production of anordiol can be reduced, which
could lead to contraceptive failure. While side effects are also associated with the
varying exposures of parent drug and metabolite in vivo.
This study is the first to report on anordrin hydrolytic kinetics in HLM, HIM, and
recombinant enzymes. Enzyme phenotype was further investigated by using

recombinant enzymes and specific esterase inhibitors. In conclusion, we discovered
that anordrin is predominantly catalyzed by CES1 and CES2 in human liver and
intestine to generate the main metabolite, anordiol. Kinetic analyses of anordrin
hydrolase activity by HLM, HIM, recombinant human CES1 and CES2 were further
performed. We hypothesized that DDI toward anordrin can be induced in vivo by
esterase inhibitors, such as simvastatin. This finding should be considered in clinical
applications.
ABBREVIATIONS
CES1, carboxylesterase 1; CES2, carboxylesterase 2; AADAC, arylacetamide
deacetylase; HLM, human liver microsomes; HIM, human intestine microsomes;
HLC, human liver cytosol; LC-MS/MS, liquid chromatography–tandem mass
spectrometry; BNPP, bis-p-nitrophenyl phosphate；RAF, relative activity factor; LNG,
levonorgestrel; DDI, drug-drug interactions.
Acknowledgments
The study was supported by the National Natural Science Foundation of China (grant
number 81521005).
Author contribution
Jinfang Jiang and Dafang Zhong are responsible for the research design; Jinfang Jiang
conducted experiments; Dafang Zhong and Xiaoyan Chen contributed new reagents
or analytical tools; Jinfang Jiang and Dafang Zhong performed data analysis and
wrote this manuscript.
Disclosure
The authors report no conflicts of interest in this work.

Reference
Ahmed, T.A., Hayslip, J., Leggas, M. (2013). Pharmacokinetics of high-dose
simvastatin in refractory and relapsed chronic lymphocytic leukemia patients.
Cancer Chemother Pharmacol 72: 1369-74.
Bin, X., Peiqin, Z., Weijuan, Y. (1989). Antitumor action of Anordrin on experimental
tumors. Tumor 9: 197-9.
Chatterton, R.T., Jr., Berman, C., Walters, N.N. (1989). Anti-uterotrophic and
folliculostatic activities of anordiol (2 alpha,17 alpha-diethynyl-A-nor-5
alpha-androstane-2 beta,17 beta-diol). Contraception 39: 291-7.
Chatterton, R.T., Jr., Kowalski, W., Lu, Y.C., et al. (1994). Pharmacokinetics and
pharmacodynamics of anordrin (2 alpha, 17 alpha-diethynyl-A-nor-5
alpha-androstane-2 beta, 17 beta-diol diproprionate). Steroids 59: 217-23.
Chih-Ping, K., Ming-Kang, C., Hsiu-Chuan, C., et al. (1976). Pharmacological studies
of a contraceptive drug anordrin. Chin Med J (Engl) 2: 177-84.
Fukami, T., Kariya, M., Kurokawa, T., et al. (2015). Comparison of substrate
specificity among human arylacetamide deacetylase and carboxylesterases. Eur J
Pharm Sci 78: 47-53.
Fukami, T., Yokoi, T. (2012). The emerging role of human esterases. Drug Metab
Pharmacokinet 27: 466-77.
Gu, Z., Zhu, D., Qi, L., et al. (1984). Studies on the antifertility effects of anordrin
and its analogues. Acta Physiol Sinica 36: 611-3.
Higgins, J.W., Bao, J.Q., Ke, A.B., et al. (2014). Utility of Oatp1a/1b-knockout and
OATP1B1/3-humanized mice in the study of OATP-mediated pharmacokinetics
and tissue distribution: case studies with pravastatin, atorvastatin, simvastatin,
and carboxydichlorofluorescein. Drug Metab Dispos 42: 182-92.
Hirano, M., Maeda, K., Shitara, Y., et al. (2006). Drug-drug interaction between
pitavastatin and various drugs via OATP1B1. Drug Metab Dispos 34: 1229-36.
Holmes, R.S., Wright, M.W., Laulederkind, S.J., et al. (2010). Recommended
nomenclature for five mammalian carboxylesterase gene families: human, mouse,
and rat genes and proteins. Mamm Genome 21: 427-41.
Imai, T. (2006). Human carboxylesterase isozymes: catalytic properties and rational
drug design. Drug Metab Pharmacokinet 21: 173-85.
Khanna, R., Morton, C.L., Danks, M.K., et al. (2000). Proficient metabolism of
irinotecan by a human intestinal carboxylesterase. Cancer Res 60: 4725-8.
Li, B., Sedlacek, M., Manoharan, I., et al. (2005). Butyrylcholinesterase, paraoxonase,
and albumin esterase, but not carboxylesterase, are present in human plasma.
Biochem Pharmacol 70: 1673-84.
Ma, Z.C., Lou, L.G., Zhang, Z., et al. (2000). Antiangiogenic effect of alpha-anordrin
in vitro and in vivo. Acta Pharmacol Sin 21: 939-44.
Mehta, R.R., Jenco, J.M., Chatterton, R.T., Jr. (1981). Antiestrogenic and antifertility
actions of Anordrin (2 alpha, 17 alpha-diethynyl-A-nor-5 alpha-androstane-2
beta, 17 beta-diol 2,17-dipropionate). Steroids 38: 679-91.

Mehta, R.R., Jenco, J.M., Chatterton, R.T., et al. (1982). Antagonism of the Actions of
Estrogens, Androgens and Progesterone by Anordrin
(2-Alpha,17-Alpha-Diethynyl-a-nor-5-Alpha-Androstane-2-Beta,17-Beta-Diol
Dipropionate). Steroids 40: 65-80.
Nakajima, A., Fukami, T., Kobayashi, Y., et al. (2011). Human arylacetamide
deacetylase is responsible for deacetylation of rifamycins: Rifampicin, rifabutin,
and rifapentine. Biochem Pharmacol 82: 1747-56.
Ross, M.K., Borazjani, A., Wang, R., et al. (2012). Examination of the
carboxylesterase phenotype in human liver. Arch Biochem Biophys 522: 44-56.
Sang, G., Shao, Q., Zhang, L. (1999). A randomized multicentre clinical trial on
different doses of mifepristone alone and in combination with anordrin as
emergency contraception. Zhonghua fu chan ke za zhi 34: 331-4.
Sato, Y., Miyashita, A., Iwatsubo, T., et al. (2012). Simultaneous absolute protein
quantification of carboxylesterases 1 and 2 in human liver tissue fractions using
liquid chromatography-tandem mass spectrometry. Drug Metab Dispos 40:
1
389-96.
Shimizu, M., Fukami, T., Nakajima, M., et al. (2014). Screening of specific inhibitors
for human carboxylesterases or arylacetamide deacetylase. Drug Metab Dispos
4
2: 1103-9.
Takahashi, S., Katoh, M., Saitoh, T., et al. (2008). Allosteric kinetics of human
carboxylesterase 1: species differences and interindividual variability. J Pharm
Sci 97: 5434-45.
Wang, X.W., Zhu, H.J., Markowitz, J.S. (2015). Carboxylesterase 1-Mediated
Drug-Drug Interactions between Clopidogrel and Simvastatin. Biol Pharm Bull
3
8: 292-7.
Watanabe, A., Fukami, T., Nakajima, M., et al. (2009). Human Arylacetamide
Deacetylase Is a Principal Enzyme in Flutamide Hydrolysis. Drug Metab Dispos
3
7: 1513-20.
Watanabe, A., Fukami, T., Takahashi, S., et al. (2010). Arylacetamide Deacetylase Is a
Determinant Enzyme for the Difference in Hydrolase Activities of Phenacetin
and Acetaminophen. Drug Metab Dispos 38: 1532-7.
Zhu, H., Markowitz, J.S. (2013). Carboxylesterase 1 (CES1) genetic polymorphisms
and oseltamivir activation. Eur J Clin Pharmacol 69: 733-4.

Titles and legends to figures:
Figure 1. Chemical structures of anordrin and its hydrolysis metabolite anordiol.
Figure 2. Liquid chromatography–tandem mass spectrometry (LC-MS/MS) of
anordrin (upper trace) and internal standard LNG (lower trace) for the accurate
quantification of anordrin in different matrics. LNG: levonorgestrel.
Figure 3. Product ions of anordrin and its hydrolysis metabolite anordriol. m/z 456
+
and m/z 344 were parent ions adducted with NH of anordrin and anordriol,
4
respectively.
3
Figure 4. Chemical structure of compound A and [ H]anordiol.
Figure 5. Hydrolase activity of anordrin in different enzyme sources. Each column
represents the mean ± S.D. of triplicate determinations. HLM: human liver
microsomes; HIM: human intestine microsomes; HLC: human liver cytosol; CES:
carboxylesterase; AADAC: arylacetamide deacetylase.
Figure 6. Kinetic profiles for the substrate depletion of anordrin in HLM and HIM
(left), recombinant human CES1 and CES2 (right). The kinetic parameters were
estimated from the fitted curve using the computer program GraphPad designed for
nonlinear regression analysis. Each data point represents the mean ± S.D. of triplicate
determinations.
Figure 7. Inhibitory profile of chemical inhibitors against anordrin hydrolase activity
in different enzyme sources. Each column represents the mean ± S.D. of triplicate
determinations. **p < 0.01.
Figure 8. The inhibitory effect of simvastatin on anordrin hydrolysis catalyzed by
carboxylesterase in HLM and HIM. Data are presented as the mean ± SD of triplicate
determinations. The calculated IC were 10.9 ± 0.1 and 6.94 ± 0.03 μM in HLM and
5
0
HIM, respectively.

Table 1. Michaelis-Menten constants obtained in HLM, HIM, recombinant human
CES1 and CES2. Data are represented as the mean ± S.D. of triplicate determinations.
Km
Vmax
(pmol/min/mg protein)
1936 ± 149
CLint
(μL/min/mg protein)
684 ± 83
Enzyme Source
(
μM)
HLM
HIM
2.83 ± 0.52
2.24 ± 0.52
13.6 ± 2.8
16.9 ± 3.0
212 ± 21
94.6 ± 13.3
117 ± 13
CES1
CES2
1590 ± 163
3321 ± 312
197 ± 17

Table 2. RAF values calculated from the marker activity and the contributions of
CES1 and CES2 to anordrin hydrolysis in HLM.
Hydrolase Activity
Procaine Phenacetin
RAF
Contribution
Enzyme Source
Fenofibrate
Anordrin
CES1 CES2
CES1 CES2
nmol/min/mg nmol/min/mg pmol/min/mg pmol/min/mg
%
%
%
%
CES1
CES2
AADAC
HLM
92.8
123
137
3.34
4.01
50.4
102
0.721
476
213
2.30
1.20
59.4
34.6
HIM
54.5