Synthesis and evaluation of haloperidol ester prodrugs metabolically activated by human carboxylesterase

Article abstract of DOI:10.1016/j.ejps.2019.03.009

Two types of haloperidol prodrugs in which a chemical modification was carried out on the hydroxyl group or carbonyl group were synthesized, and their metabolic activation abilities were evaluated in a human liver microsome (HLM) solution, a human small i

Full text of DOI:10.1016/j.ejps.2019.03.009

Accepted Manuscript
Synthesis and evaluation of haloperidol ester prodrugs
metabolically activated by human carboxylesterase
Masato Takahashi, Tomoki Uehara, Minori Nonaka, Yuka
Minagawa, Riona Yamazaki, Masami Haba, Masakiyo Hosokawa
PII:
S0928-0987(19)30112-5
https://doi.org/10.1016/j.ejps.2019.03.009
PHASCI 4875
DOI:
Reference:
To appear in:
European Journal of Pharmaceutical Sciences
Received date:
Revised date:
Accepted date:
10 December 2018
1 March 2019
11 March 2019
Please cite this article as: M. Takahashi, T. Uehara, M. Nonaka, et al., Synthesis and
evaluation of haloperidol ester prodrugs metabolically activated by human
carboxylesterase, European Journal of Pharmaceutical Sciences, https://doi.org/10.1016/
j.ejps.2019.03.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Synthesis and evaluation of haloperidol ester prodrugs metabolically activated by human carboxylesterase
Masato Takahashi*, Tomoki Uehara, Minori Nonaka, Yuka Minagawa, Riona Yamazaki, Masami Haba,
Masakiyo Hosokawa
Faculty of Pharmacy, Chiba Institute of Science, 15–8, Shiomi-cho, Choshi, Chiba 288–0025, Japan
*
Corresponding author. Tel.: +81-479-30-4512; fax: +81-479-30-4683; e-mail: matakahashi@cis.ac.jp

ACCEPTED MANUSCRIPT
Two types of haloperidol prodrugs in which a chemical modification was carried out on the
hydroxyl group or carbonyl group were synthesized, and their metabolic activation abilities were
evaluated in a human liver microsome (HLM) solution, a human small intestine microsome (HIM)
solution and solutions of human recombinant carboxylesterases (hCESs). The metabolic activation
rates of alcohol ester prodrugs in HLM solution were similar to those in hCES2 solution, and
haloperidol pentanoate and haloperidol hexanoate showed high metabolic activation rates in the
synthesized alcohol ester prodrugs. In addition, haloperidol acetate and haloperidol
2
-methylbutanoate were hydrolyzed as slowly as haloperidol decanoate. The results suggested that
haloperidol prodrugs with a small chain or a branched chain are useful as prodrugs for sustained
release. The metabolic activation rate of the enol ester prodrug in HLM solution was similar to that
in hCES1 solution, and the enol ester prodrug was found to behave differently from alcohol ester
prodrugs, which were metabolically activated by hCES2.

ACCEPTED MANUSCRIPT
1
. Introduction
Haloperidol (1) is an antipsychotic drug with a butyrophenone skeleton that was developed in
Belgium in 1958. Treatment of schizophrenia was advanced by the development of haloperidol (1)
Froemming et al., 1989); however, there was a problem of some patients forgetting to take the
(
medicine. To solve this problem, haloperidol decanoate in which the hydroxyl group of haloperidol is
replaced with a decanoate group was developed. Haloperidol decanoate, which is intramuscularly
administered, is slowly metabolized and haloperidol (1) continues to be produced in the body for 4
weeks. Because of the long administration interval, it is thought to be helpful for some noncompliant
patients (Beresford and Ward, 1987). In addition, many ester prodrugs for sustained release such as
nalmefene prodrug (Gaekens et al., 2016), entecavir prodrug (Ho et al., 2018) and ibuprofen prodrug
(
Liu et al., 2017) have been studied recently.
Carboxylesterases (CESs) are hydrolytic enzymes that play an important role in metabolic
activation of ester–type and amide–type prodrugs (Hosokawa, 2008). Haloperidol decanoate is also
known to be metabolically activated by CESs (Nambu et al., 1987). The two dominant forms CES1
and CES2 are present in various mammals (Hosokawa et al., 2007; Satoh and Hosokawa, 1998;
Satoh and Hosokawa, 2006), and CESs contribute to hydrolysis of drugs in many tissues including
the liver, small intestine, lung and kidney. CES1 and a small amount of CES2 are expressed in the
human liver, in which many drugs are metabolized (Hosokawa et al., 1995). Human CES1 (hCES1)
catalyzes a hydrolysis reaction of ester compounds containing a small alkoxy group, such as
temocapril (Takai et al., 1997), meperidine (Zhang et al., 1997), oseltamivir (Shi et al., 2006),
atorvastatin prodrugs (Mizoi et al., 2016), indomethacin prodrugs (Takahashi et al., 2018) and
methylphenidate (Sun et al., 2004). In contrast, CES2 catalyzes the hydrolysis of ester compounds
containing a small acyl group, such as CPT–11 (Satoh et al., 1994), cocaine (Kamendulis et al.,
1
996) and heroin (Brzezinski et al., 1997). Therefore, an ester compound is metabolized by CES1 or
CES2 depending on the structure of the ester. Furthermore, among the substrates hydrolyzed by
CES1, it has been found that the rate of hydrolysis varies greatly depending on the size of the ester
structure and electron density (Mizoi et al., 2016; Imai et al., 2006). Examination of the substrate
recognizing ability of these CES isozymes is therefore important for understanding the metabolism
of drugs and metabolic activation of prodrugs.
Studies on haloperidol prodrugs have been intensively conducted in recent years. Besides
haloperidol decanoate, studies have been conducted on a haloperidol prodrug incorporating PEG
(Heath et al., 2016) or linear acyl ester (Morris and Heard, 2008; Morris and Heard, 2009). There
have also been studies on a prodrug of reduced haloperidol that was obtained by reducing the
carbonyl group of haloperidol (1) (Olivieri et al., 2016; Sozio et al., 2015). Although haloperidol
prodrugs having a linear acyl group have been reported, there has been no report on the metabolic
ability of haloperidol prodrugs having a branched alkyl acyl group. In this study, we synthesized and
evaluated the metabolic activation abilities of a linear acyl group–containing prodrug and branched

ACCEPTED MANUSCRIPT
chain acyl group–containing prodrug. Furthermore, focusing on the carbonyl group of haloperidol,
enol ester–type prodrug 3 was newly synthesized and its metabolic activation abilities were
evaluated.

ACCEPTED MANUSCRIPT
2
. Materials and Methods
2
.1. Materials
The ester derivatives were synthesized using the haloperidol (1), acid anhydride or acid chloride.
Materials for the reaction, such as haloperidol (1), acid anhydride, dimethylaminopyridine (DMAP)
were purchased from Wako pure chemical industries (Japan). Human intestine microsomes (HIM),
hCES1 and hCES2 were purchased from Corning inc. (Woburn, MA, USA). Human plasma sample
(single donor, treated by EDTA–2Na) was purchased from KAC Co., Ltd. (Amagasaki, Hyogo,
Japan). Human liver microsomes (HLM) were prepared from human livers which were obtained
from the Human and Animal Bridging (HAB) Research Organization, which is in partnership with
the National Disease Research Interchange (NDRI). The livers were homogenized and centrifuged at
9
4
,000 × G for 20 min at 4 °C, and the supernatant was ultra–centrifuged at 105,000 × G for 60 min at
°C. The microsome fraction was suspended in sucrose–EDTA–Tris (SET) buffer (pH 7.4).
2
.2. Synthetic protocols
A mixture of haloperidol (1) (1.0 mmol), acid anhydride (or acid chloride) (2.0 mmol), DMAP (0.1
mmol) and Et N (5.0 mmol) in CH CN (10 mL) was stirred at 80 °C under argon atmosphere for 1
3
3
day. After addition of sat. NaHCO solution (5 mL), and the whole was extracted with AcOEt (50
3
mL). The organic solution was washed with H O (5 mL×2) and brine (5 mL×2), and dried over
2
MgSO and evaporated under reduced pressure. The residue was purified by column chromatography
4
on SiO (ethyl acetate/n–hexane) to give an ester derivative 2. The synthesized compounds were
2
1
13
confirmed by H NMR, C NMR and IR analyses.
Haloperidol acetate 2a
–
1 1
isolated yield 29%; pale yellow solids; mp 101–102 °C; IR (KBr) 1737, 1695 cm ; H NMR (400
MHz) δ 1.78–1.92 (2H, m), 1.94 (3H, s), 2.20–2.26 (4H, m), 2.23–2.33 (2H, m), 2.36 (2H, t, J = 7.2
Hz), 2.68–2.70 (2H, m), 2.89 (2H, t, J = 7.2 Hz), 7.00–7.05 (2H, m), 7.14–7.17 (2H, m), 7.18–7.19
13
(2H, m), 7.89–7.94 (2H, m); C NMR (100 MHz) δ 21.9, 22.0, 35.4, 36.2, 49.2, 57.6, 79.9, 115.6 (d,
J = 22.0 Hz), 126.0, 128.5, 130.6 (d, J = 9.0 Hz), 133.0, 133.7 (d, J = 3.0 Hz), 143.0, 165.6 (d, J =
2
53.0 Hz), 169.5, 198.3; HPLC retention time (method A) t = 21.5 min, (method B) t = 13.7 min.
R R
Haloperidol butanoate 2b
–
1 1
isolated yield 5.0%; pale yellow solids; mp 53–54 °C; IR (KBr) 1733, 1681 cm ; H NMR (400
MHz) δ 0.85 (3H, t, J = 7.6 Hz), 0.86 (3H, t, J = 7.6 Hz), 1.5–1.59 (2H, m), 1.87–1.90 (4H, m),
2
.18–2.25 (2H, m), 2.18–2.25 (4H, m), 2.21 (2H, t, J = 7.6 Hz), 2.38 (2H, t, J = 7.2 Hz), 2.70–2.73
(2H, m), 2.91 (2H, t, J = 6.8 Hz), 7.03–7.08 (2H, m), 7.20–7.22 (2H, m), 7.16–7.19 (2H, m), 7.91–

ACCEPTED MANUSCRIPT
1
3
7
2
.96 (2H, m); C NMR (100 MHz) δ 13.8, 18.4, 21.9, 35.4, 36.2, 37.2, 49.2, 57.6, 79.5, 115.6 (d, J =
1.0 Hz), 126.0, 128.5, 130.6 (d, J = 9.0 Hz), 133.0, 133.6 (d, J = 3.0 Hz), 143.1, 165.6 (d, J = 253.0
Hz), 172.0, 198.3; HPLC retention time (method B) t = 18.6 min.
R
Haloperidol hexanoate 2c
–
1 1
isolated yield 46%; pale yellow oil; IR (KBr) 1735, 1685 cm ; H NMR (400 MHz) 0.89 (3H, t, J
7.2 Hz), 1.23–1.36 (4H, m), 1.55–1.62 (2H, m), 1.88–2.04 (4H, m), 2.24–2.35 (4H, m), 2.40–2.47
4H, m), 2.77–2.80 (2H, m), 2.98 (2H, t, J = 6.8 Hz), 7.09–7.14 (2H, m), 7.29–7.31 (4H, m), 7.98–
=
(
13
8
.03 (2H, m); C NMR (100 MHz) δ 13.9, 21.8, 22.3, 24.6, 31.3, 35.2, 35.4, 36.2, 49.2, 57.6, 79.5,
115.6 (d, J = 22.0 Hz), 126.1, 128.5, 130.6 (d, J = 9.0 Hz), 133.0, 133.6 (d, J = 3.0 Hz), 143.1, 165.6
(
d, J = 253 Hz), 172.2, 198.3; HPLC retention time (method B) t = 24.3 min.
R
Haloperidol octanoate 2d
–
1 1
isolated yield 28%; yellow oil; IR(KBr) 1735, 1685 cm ; H NMR (400 MHz) δ 0.88 (3H, t, J = 6.8
Hz), 1.11–1.31 (8H, m), 1.56–1.61 (2H, m), 1.87–2.04 (4H, m), 2.27–2.34 (4H, m), 2.40–2.47 (4H,
m), 2.76–2.79 (2H, m), 2.98 (2H, t, J = 7.2 Hz), 7.08–7.14 (2H, m), 7.23–7.29 (4H, m), 7.98–8.03
1
3
(
2H, m); C NMR (100 MHz) δ 14.1, 21.8, 22.6, 25.0, 28.9, 29.2, 31.7, 35.3, 35.4, 36.2, 49.2, 57.6,
7
1
9.5, 115.6 (d, J = 22.0 Hz), 126.1, 128.5, 130.6 (d, J = 9.0 Hz), 133.0, 133.6 (d, J = 3.0 Hz), 143.1,
65.6 (d, J = 253.0 Hz), 172.2, 198.3; HPLC retention time (method B) t = 28.8 min.
R
Haloperidol pentanoate 2f
–
1
1
isolated yield 28%; pale yellow solids; mp 31–41 °C; IR (KBr) 1729, 1683 cm ; H NMR (400
MHz) 0.90 (3H, t, J = 7.6 Hz), 1.28–1.37 (2H, m), 2.28–2.47 (8H, m), 2.77–2.79 (2H, m), 2.98 (2H, t,
13
J = 6.8 Hz), 7.08–7.14 (2H, m), 7.23–7.29 (4H, m), 7.97–8.02 (2H, m); C NMR (100 MHz) δ 13.7,
2
9
1.8, 22.3, 27.0, 35.0, 35.4, 36.2, 49.2, 57.6, 79.5, 115.6 (d, J = 21.0 Hz), 126.0, 128.5, 130.6 (d, J =
.0 Hz), 133.0, 133.6 (d, J = 3.0 Hz), 143.1,165.6 (d, J = 253.0 Hz), 172.2, 198.3; HPLC retention
time (method B) t = 22.1 min.
R
Haloperidol 3–methylbutanoate 2g
–
1 1
isolated yield 24%; pale yellow solids; mp 74.5–77 °C; IR (KBr) 1733, 1683 cm ; H NMR (400
MHz) 0.93 (6H, d, J = 6.8 Hz), 1.88–2.11 (5H, m), 2.16 (2H, d, J = 6.8 Hz), 2.32 (2H, t, J = 11.6 Hz),
2
.40–2.47 (4H, m), 2.76–2.79 (2H, m), 2.98 (2H, t, J = 6.8 Hz), 7.08–7.12 (2H, m), 7.24–7.30 (4H,
1
3
m), 7.97–8.02 (2H, m); C NMR (100 MHz) δ 21.9, 22.5, 25.6, 35.4, 36.2, 44.4, 49.2, 57.6, 79.6,
15.6 (d, J = 21.0 Hz), 126.1, 128.5, 130.6 (d, J = 9.0 Hz), 133.0, 133.7 (d, J = 3.0 Hz), 143.1,165.6
d, J = 253.0 Hz), 171.5, 198.3. HPLC retention time (method B) t = 22.0 min.
1
(
R
Haloperidol 2–methylbutanoate 2h
–
1
1
isolated yield 22%; pale yellow solids; mp 61–70 °C; IR (KBr) 1733, 1685 cm ; H NMR (400
MHz) δ 0.89 (3H, t, J = 7.2 Hz), 1.30–1.40 (2H, m) 1.49–1.56 (2H, m), 2.37 (3H, s), 2.86 (2H, t, J =
7
2
2
1
.6 Hz), 3.84–3.85 (5H, m), 6.68 (1H, dd, J = 8.8, 2.4 Hz), 6.89 (1H, d, J = 8.8 Hz), 6.95 (1H, d, J =
1
3
.4 Hz), 7.48 (2H, d, J = 6.8 Hz), 7.66 (2H, d, J = 6.8 Hz); C NMR (100 MHz) δ 11.8, 16.7, 21.9,
6.6, 35.4, 36.2, 42.0, 49.2, 57.7, 79.2, 115.6 (d, J = 21.0 Hz), 126.1, 128.5, 130.6 (d, J = 9.0 Hz),
33.0, 133.7 (d, J = 4.0 Hz), 143.1, 165.6 (d, J = 254.0 Hz), 174.9, 198.3; HPLC retention time

ACCEPTED MANUSCRIPT
(
method B) t = 21.7 min.
R
Haloperidol pivalate 2i
isolated yield 3%; pale yellow oil; IR (KBr) 1731, 1685 cm ; H NMR (400 MHz) δ 1.20 (9H, s),
.88–2.00 (4H, m), 2.29–2.35 (2H, m), 2.41–2.47 (4H, m), 2.79–2.81 (2H, m), 2.98 (2H, t, J = 7.2
–
1 1
1
1
3
Hz), 7.10–7.14 (2H, m), 7.20–7.30 (4H, m), 7.98–8.02 (2H, m); C NMR (100 MHz) δ 21.8, 27.3,
3
1
5.3, 36.2, 39.3, 49.2, 57.7, 79.0, 115.6 (d, J = 22.0 Hz), 125.8, 128.5, 130.6 (d, J = 9.0 Hz), 133.0,
33.6 (d, J = 3.0 Hz), 143.1, 165.6 (d, J = 253.0 Hz), 176.4, 198.3; HPLC retention time (method B)
tR = 21.1 min.
Haloperidol enol pivalate 3
–
1 1
isolated yield 11%; pale yellow solid; mp 116–123 °C; IR (KBr) 1747, 1664 cm , H NMR (400
MHz) δ 1.37 (9H, s), 1.70–1.73 (2H, m), 2.07–2.15 (2H, m), 2.28–2.33 (2H, m), 2.44–2.56 (4H, m),
2
.79–2.81 (2H, m), 5.74 (1H, t, J = 7.2 Hz), 6.98–7.02 (2H, m), 7.26–7.32 (2H, m),7.34–7.37 (2H,
m),7.42–7.45 (2H, m); 23.7, 27.3, 38.4, 39.2, 49.2, 57.3, 71.1, 115.4 (d, J = 22.0 Hz), 126.1, 126.3 (d,
J = 8.0 Hz), 128.4, 131.7 (d, J = 3.0 Hz), 132.8, 146.1, 146.4, 162.6 (d, J = 246.0 Hz), 176.1; HPLC
retention time (method B) t = 17.8 min.
R
2
.3. Hydrolysis reaction (HLM, HIM)
Enzyme solutions were diluted to 5 mg/mL solutions with 200 mM SET buffer (pH7.4). 25 mM
haloperidol ester (2a–2i, 3) in dimethylsulfoxide (DMSO) solutions (1 μL, final concentration 0.25
mM) were warmed with 5 mg/mL enzyme solutions in SET buffer (5 μL, final concentration 0.25
mg/mL), 200 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer (50 μL, final
concentration 100 mM) and purified water (44 μL) at 37 °C for 24 h. After addition of 0.5 mM
biphenyl in isopropanol (100 μL, 50 nmol) immediately, the whole was cooled on an ice bath at 10
min. The mixture was centrifuged at 21,600 × G for 15 min at 4 °C to precipitate protein, and then
the supernatant was filtered and analyzed by high performance liquid chromatography (HPLC).
2
.4. Plasma stability
1
0 mM haloperidol ester (2a, 2c, 2e, 2h or 3) in DMSO solutions (3 μL, final concentration 0.10
mM) were warmed with 200 mM HEPES buffer (147 μL), human plasma sample (150 μL) at 37°C.
The reaction solutions (30 μL) were collected at various time points (0, 1, 3, 6, 12, 24, 48 h). After
addition of 0.5 mM biphenyl in isopropanol (100 μL) immediately, the whole was cooled on an ice
bath at 10 min. The mixture was centrifuged at 21,600 × G for 15 min at 4 °C, and then the
supernatant was filtered and analyzed by HPLC.
2
.5. Hydrolysis reaction (hCES1, hCES2)

ACCEPTED MANUSCRIPT
Enzyme solutions were diluted to 5 mg/mL solutions with 200 mM SET buffer (pH7.4). 0.3125–20
mM haloperidol ester (2a–2i, 3) in DMSO solutions (1 μL, final concentration 0.003125–0.10 mM)
were warmed with 5 mg/mL enzyme solutions in SET buffer (5 μL, final concentration 0.25 mg/mL),
2
2
00 mM HEPES buffer (50 μL, final concentration 100 mM) and purified water (44 μL) at 37 °C for
4 h. After addition of 0.5 mM biphenyl in isopropanol (100 μL, 50 nmol) immediately, the whole
was cooled on an ice bath at 10 min. The mixture was centrifuged at 21,600 × G for 15 min at 4 °C
to precipitate protein, and then the supernatant was filtered and analyzed by HPLC.
2
.6. HPLC Analysis
Method A for compound 2a
The HPLC system was performed on a column of Mightysil RP–18 GP 150-4.6 (5 μm) at 30 °C with
elution at 1.0 mL/min using the mixture of 0.5% ethylamine in MeOH/0.5% ethylamine in H O =
2
6
5/35 for 30 min. The haloperidol was monitored at 220 nm. Analytical HPLC was performed in
biphenyl as an internal standard (t = 18.7 min). The hydrolysis rate was calculated from the
R
detecting area ratio of haloperidol (t = 10.8 min) and biphenyl.
R
Method B for compound (2b–2i, 3)
The HPLC system was performed on a column of Mightysil RP–18 GP 150-4.6 (5 μm) at 30 °C with
elution at 1.0 mL/min using the gradient of 0.5% ethylamine in MeOH/0.5% ethylamine in H O =
2
6
5/35 being ramped up from 65/35 to 95/5 for 30 min and then returned to 65/35 in 10 min. The
haloperidol was monitored at 220 nm. Analytical HPLC was performed in biphenyl as an internal
standard (t = 13.7 min). The hydrolysis rate was calculated from the detecting area ratio of
R
haloperidol (t = 9.3 min) and biphenyl.
R

ACCEPTED MANUSCRIPT
3
. Results and discussion
3
.1. Synthesis of prodrugs (2a–2i, 3) and these chemical properties
Haloperidol alcohol ester prodrugs (2a–2d, 2f–2h) were synthesized using haloperidol (1) and acid
anhydrides (Scheme 1). Using acid chloride instead of acid anhydride, it was possible to synthesize
both alcohol ester prodrug 2i and enol ester prodrug 3 (Scheme 2). These compounds were purified
1
13
by column chromatography, and the structure was confirmed by H–NMR, IR and C–NMR. The
E/Z conformation of enol ester 3 was determined by NOESY. Haloperidol decanoate (2e) was
purchased commercially and used for experiments.
Next, in order to evaluate the lipophilicity of the haloperidol prodrugs (2a–2i, 3), calculation of
ClogP value and retention time of HPLC were measured (Table 1). Due to the increase in the carbon
number of the modifying group, the ClogP value and retention time of HPLC were increased, it was
confirmed that any prodrug was higher in lipophilicity than haloperidol (1).
3
.2. Hydrolysis reaction in HLM and HIM
It was examined whether hydrolysis of the prodrugs (2a–2i, 3) proceeds in a human liver
microsome (HLM) solution (Fig. 1). Haloperidol hexanoate (2c) showed a higher metabolic
activation rate than those of the prodrugs (2a–2e). Acetate 2a was hydrolyzed as slowly as
haloperidol decanoate (2e). A comparison of the C5 acyl ester prodrugs (2f–2i) showed that the
hydrolysis rates of branched acyl esters (2g–2i) were slower than those of the linear acyl ester 2f.
Although 3-methylbutanoate 2g and pivalate 2i were not hydrolyzed in 24 hours, 2-methylbutanoate
2
h was hydrolyzed. The results suggested that haloperidol prodrugs with a small chain or a branched
chain are useful as prodrugs for sustained release.
The prodrugs were subjected to hydrolysis reaction in HIM solution and the stability of the
prodrugs in HIM solution was investigated (Fig. 2). Although the data showed that the hydrolysis
rates of alcohol ester prodrugs (2a–2i) in HIM solution are similar to those in HLM solution, the
hydrolysis rate of 3 was significantly reduced compared with that in HLM solution. Therefore, it was
thought that the kinds of hCES that catalyze the hydrolysis of 2 and the hydrolysis of 3 are different.
hCES1 and small amount of hCES2 are also expressed in the liver and hCES2 is expressed in the
small intestine. Since alcohol ester prodrugs (2a–2i) have a bulky alkoxy group, they are expected to
be mainly catalyzed by CES2. On the other hand, the enoxy position of enol ester prodrug 3 is more
vacant than the alkoxy position of alcohol ester prodrugs (2a–2i). For these reasons, it is thought that
alcohol ester prodrugs are mainly catalyzed by hCES2 and enol ester prodrugs are mainly catalyzed
by hCES1. The alcohol ester prodrugs (2a–2i), which are CES2–hydrolyzed types, may be

ACCEPTED MANUSCRIPT
hydrolyzed in both the liver and small intestine and are therefore to be inconvenient for oral
administration. The enol ester prodrug 3, which is a CES1–hydrolyzed type, is considered to be a
more suitable prodrug for oral administration since it is hardly hydrolyzed in the small intestine. It is
thought to be almost impossible to confirm the hydrolysis of alcohol ester prodrugs (2a–2i) with
CES1 since the alkoxy group has a structurally very bulky carbon near the ester.
3
.3. Plasma stability
The plasma stabilities of hexanoate 2c and enol pivalate 3, which have high hydrolysis rate in HLM
solution and acetate 2a, decanoate 2e and 2–methylbutanoate 2h, which have relatively low
hydrolysis rate in HLM solution were evaluated. The amount of haloperidol prodrugs (2a, 2c, 2e, 2h
and 3) when reaction was carried out for 0, 1, 3, 6, 12, 24 and 48 hours was calculated and the
stabilities of these esters was evaluated (Fig. 3). About 5–10% disappearance of haloperidol prodrugs
(2a, 2c, 2e, 2h and 3) was observed, however, haloperidol (1) was not detected even when reacted in
the presence of human plasma for 48 hours. Hydrolytic enzymes such as butyrylcholinesterase,
paraoxonase and albumin are mainly present, but carboxylesterases are no contained in human
plasma (Fukami and Yokoi, 2012). It is thought that the synthesized haloperidol prodrugs (2a–2i, 3)
are unlikely to be recognized by hydrolases in human plasma and can stably exist in human plasma.
3
.4. Kinetics analysis of hydrolysis reaction catalyzed by hCES1 or hCES2
To confirm that the haloperidol prodrugs (2a–2i, 3) are metabolically activated by hCESs, these
prodrugs were subjected to hydrolysis reaction in hCES1 or hCES2 solution. The Vmax and K values
m
were calculated by the Michaelis-Menten equation using nonlinear regression analysis with
GraphPad Prism 7 software, and CL was calculated by the V and K values. In the reaction in
int
max
m
hCES2 solution (Table 2), Vmax values of CES2 for ester prodrugs (2a–2i, 3) showed patterns similar
to the hydrolysis rates in HIM solution: Vmax values of pentanoate 2f and hexanoate 2c were
relatively high, while those of acetate 2a, decanoate 2e, 2-methylbutanoate 2h and pivalate 3 were
low. The K values of hCES2 for the haloperidol prodrugs increased as the number of carbons of the
m
prodrug became smaller or larger than the appropriate number. The K value of acetate 2a, which
m
has a relatively small steric hindrance, was larger than the K values of the other esters. It is thought
m
that the affinity of the enzyme varies greatly depending on the structure of the ester, and hexanoate
2
c and its analogs were thought to be easily recognizable by hCES2. When designing a sustained–
release prodrug that is metabolically activated by hCES2, it is important not to decrease the rate of
metabolic activation by the construction of a bulky ester but to control the rate of metabolic
activation in consideration of the affinity between an ester and a metabolic enzyme. The largest CLint
value was also for hexanoic acid ester. The CLint values of acetate 2a, decanoate 2e and 3–
methylbutanoate 2g were relatively small among the alcohol esters (2a–2i). Therefore, the CLint

ACCEPTED MANUSCRIPT
values of esters that are structurally different from hexanoate 2c were found to be small. These
results suggest that hCES2 may have catalyzed the hydrolysis of haloperidol prodrugs (2a–2i, 3).
Next, hydrolysis in hCES1 solution was examined (Table 3). Interestingly, although no hydrolysis
reaction was observed for alcohol ester prodrugs (2a–2i), enol ester prodrug 3 was effectively
hydrolyzed in hCES1 solution. Therefore, it is thought that the hydrolysis of alcohol ester prodrugs
(2a–2i) is catalyzed by hCES2 rather than hCES1 and that the hydrolysis of enol ester prodrug 3 is
catalyzed by hCES1 rather than hCES2.
A possible reason for the difference in hydrolysis of alcohol ester prodrugs (2a–2i) and hydrolysis
of the enol ester prodrug 3 is the difference in bulkiness of the ester in these prodrugs. The alkoxy
group of alcohol ester prodrugs (2a–2i) has a tertiary carbon neighboring the ester and is very bulky,
whereas the acyl group of these prodrugs (2a–2i) has relatively small chains. These esters that have a
small acyl group and a large alkoxy group were recognized by hCES2. In contrast, the enoxy group
of the enol ester prodrug 3 has a carbon neighboring the ester, which takes a planar structure and is
not bulky, whereas the acyl group of prodrug 3 has a tertiary carbon neighboring the ester. This ester,
which has a large acyl group and a small alkoxy group, is recognized by hCES1. Thus, by changing
the modification site of the acyl group, it is thought that two kinds of prodrugs with different
metabolic properties can be synthesized.
3
.5. Hydrolysis mechanism
The hydrolysis mechanisms of the synthesized haloperidol prodrugs (2a–2i, 3) are shown in Fig. 3
and Fig. 4. The hydrolysis reaction by CES is a typical ping–pong–bi–bi reaction (Satoh and
Hosokawa, 1998). Nucleophilic attack to ester is occurred by the hydroxyl group of serine which the
nucleophilicity of which is enhanced by the hydrogen bond with glutamine and histidine in CES.
Next, an acyl–enzyme complex is formed by elimination of the haloperidol (1). Subsequently, a
nucleophilic addition–elimination reaction in the acyl–enzyme complex by water occurs, the acyl
moiety is eliminated from the acyl-enzyme complex to form CES. The eliminated acyl group is
converted to the corresponding carboxylic acid by nucleophilic reaction of water. These carboxylic
acids are highly safe as contained in foods and essential oils.
The first nucleophilic addition step affects the metabolic activation ability depending on the type of
ester. From the results of X–ray structural analysis, the active center of hCES1 enzyme is composed
of a rigid site and a flexible site and has a wide space (Bencharit et al., 2003). On the other hand, one
of the loops constituting the flexible site is deleted in the amino acid sequence in hCES2, and it is
thought that the flexibility of the active center of hCES2 is inferior to that of hCES1. The difference
in substrate specificity between hCES1 and hCES2 is thought to be due to the difference in the site
size and flexibility. It is thought that hCES2 does not catalyze the hydrolysis of a compound having a
bulky acyl group because the steric hindrance of the active center prevents binding between the acyl
groups which is the first nucleophilic addition step, it is thought that the only ester having an acyl

ACCEPTED MANUSCRIPT
groups of the appropriate size can be bonded.
. Conclusions
In this study, two types of haloperidol prodrugs were synthesized, and the metabolic activation
4
abilities were investigated. Based on the results, prodrug 2 of the alcohol ester type is thought to be
metabolically activated mainely by hCES2, and it was found that haloperidol pentanoate and
haloperidol hexanoate have relatively high metabolic activation rates. Also, besides the prodrug with
a long acyl group such as decanoate, the prodrug with a short acyl group or a branched acyl group
showed the possibility of being useful in prodrugs for sustained release. In addition, it was found that
prodrug 3 of the enol ester type is more likely than an alcohol ester–type prodrug to be metabolized
in the liver. The application of these findings is expected to further accelerate the study of prodrugs
for alcohol–type and ketone–type pharmaceuticals.

ACCEPTED MANUSCRIPT
Acknowledgements
We thank the Human and Animal Bridging (HAB) Research Organization (Japan) for providing
human livers. This work was supported by JSPS KAKENHI Grant Number JP17K15519.
References
Bencharit, S., Morton, C. L., Hyatt, J. L., Kuhn, P., Danks, M. K., Potter, P. M., Redinbo, M. R.,
2
003. Crystal structure of human carboxylesterase 1 complexed with the Alzheimer’s drug tacrine:
from binding promiscuity to selective inhibition. Chem. Biol. 10(4), 341–349.
Beresford, R., Ward, A., 1987. Haloperidol Decanoate: A Preliminary Review of Its
Pharmacodynamic and Pharmacokinetic Properties and Therapeutic Use in Psychosis. Drugs 33(1),
3
1–49.
Brzezinski, M. R., Spink, B. J., Dean, R. A., Berkman, C. E., Cashman, J. R., Bosron, W. F., 1997.
Human liver carboxylesterase hCE-1: binding specificity for cocaine, heroin, and their metabolites
and analogs. Drug Metab. Dispos. 25(9), 1089–1096.
Froemming, J. S., Lam, Y. W. F., Jann, M. W., & Davis, C. M., 1989. Pharmacokinetics of
Haloperidol. Clin. Pharmacokinet. 17(6), 396–423.
Fukami, T., Yokoi, T., 2012. Drug Metab. Pharmacokinet. 27(5), 466–477.
Gaekens, T., Guillaume, M., Borghys, H., De Zwart, L. L., De Vries, R., Embrechts, R. C. A.,
Vermeulen, A., Megens, A. A. H. P., Leysen, J. E., Herdewijn, P., Annaert, P. P. Atack, J. R. 2016.
Lipophilic nalmefene prodrugs to achieve a one-month sustained release. J. Controll. Release 232,
1
96–202.
Heath, F., Newman, A., Clementi, C., Pasut, G., Lin, H., Stephens, G. J., Whalley, B. J., Osborn, H.
M. I., Greco, F., 2016. A novel PEG–haloperidol conjugate with a non-degradable linker shows the
feasibility of using polymer–drug conjugates in a non-prodrug fashion. Polym. Chem. 7(47), 7204–
7
210.
Ho, M. J., Lee, D. R., Im, S. H., Yoon, J. A., Shin, C. Y., Kim, H. J., Jang, S. W., Choi, Y. W., Han,
Y. T., Kang, M. J. (2018). Microsuspension of fatty acid esters of entecavir for parenteral sustained

ACCEPTED MANUSCRIPT
delivery. Int. J. Pharm. 543(1–2), 52–59.
Hosokawa, M., Endo, T., Fujisawa, M., Hara, S., Iwata, N., Sato, Y., Satoh, T., 1995. Interindividual
variation in carboxylesterase levels in human liver microsomes. Drug Metab. Dispos. 23(10), 1022–
1
027.
Hosokawa, M. 2008. Structure and catalytic properties of carboxylesterase isozymes involved in
metabolic activation of prodrugs. Molecules 13(2), 412–431.
Hosokawa, M., Furihata, T., Yaginuma, Y., Yamamoto, N., Koyano, N., Fujii, A., Nagahara, Y.,
Satoh, T., Chiba, K., 2007. Genomic Structure and Transcriptional Regulation of the Rat, Mouse,
and Human Carboxylesterase Genes. Drug Metab. Rev. 39(1), 1–15.
Imai, T., Taketani, M., Shii, M., Hosokawa, M., Chiba, K., 2006. Substrate specificity of
carboxylesterase isozymes and their contribution to hydrolase activity in human liver and small
intestine. Drug Metab. Dispos. 34(10), 1734–1741.
Kamendulis, L. M., Brzezinski, M. R., Pindel, E. V, Bosron, W. F., Dean, R. A., 1996. Metabolism
of cocaine and heroin is catalyzed by the same human liver carboxylesterases. J. Pharmacol. Exp.
Ther. 279(2), 713–717.
Liu, S., Pan, G., Liu, G., Neves, J. das, Song, S., Chen, S., Cheng, B., Sun, Z., Sarmento, B., Cui, W.,
Fan, C., 2017. Electrospun fibrous membranes featuring sustained release of ibuprofen reduce
adhesion and improve neurological function following lumbar laminectomy. J. Controll. Release 264,
1
–13.
Mizoi, K., Takahashi, M., Haba, M., Hosokawa, M., 2016. Synthesis and evaluation of atorvastatin
esters as prodrugs metabolically activated by human carboxylesterases. Bioorg. Med. Chem. Lett.
2
6(3), 921–923.
Morris, A. P., Brain, K. R., Heard, C. M., 2009. Skin permeation and ex vivo skin metabolism of
O-acyl haloperidol ester prodrugs. Int. J. Pharm. 367(1–2), 44–50.
Morris, A. P., Brain, K. R., Heard, C. M., 2008. Synthesis of haloperidol prodrugs and their
hydrolysis by porcine liver esterase. Drug Metab. Lett. 2(4), 275–279.
Nambu, K., Miyazaki, H., Nakanishi, Y., Oh-e, Y., Matsunaga, Y., Hashimoto, M., 1987. Enzymatic
hydrolysis of haloperidol decanoate and its inhibition by proteins. Biochem. Pharmacol. 36(10),

ACCEPTED MANUSCRIPT
1
715–1722.
Olivieri, M., Amata, E., Vinciguerra, S., Fiorito, J., Giurdanella, G., Drago, F., Caporarello, N.,
Prezzavento, O., Arena, E., Salerno, L., Rescifina, A., Lupo, G., Anfuso, C. D., Marrazzo, A., 2016.
Antiangiogenic Effect of (±)-Haloperidol Metabolite II Valproate Ester [(±)-MRJF22] in Human
Microvascular Retinal Endothelial Cells. J. Med. Chem., 59(21), 9960–9966.
Satoh, T., Hosokawa, M., Atsumi, R., Suzuki, W., Hakusui, H., Nagai, E., 1994. Metabolic
activation of CPT-11, 7-ethyl-10-[4-(1-piperidino)-1- piperidino]carbonyloxycamptothecin, a novel
antitumor agent, by carboxylesterase. Biol. Pharm. Bull. 17(5), 662–664.
Satoh, T., Hosokawa, M., 1998. THE MAMMALIAN CARBOXYLESTERASES: From Molecules
to Functions. Ann. Rev. Pharmacol. Toxicol. 38(1), 257–288.
Satoh, T., Hosokawa, M., 2006. Structure, function and regulation of carboxylesterases. Chem. Biol.
Interact. 162(3), 195–211.
Shi, D., Yang, J., Yang, D., Lecluyse, E. L., Black, C., You, L., Akhlaghi, F., Yan, B., 2006.
Anti-Influenza Prodrug Oseltamivir Is Activated by Carboxylesterase Human Carboxylesterase 1,
and the Activation Is Inhibited by Antiplatelet Agent Clopidogrel, J. Pharmacol. Ther. 319(3), 1477–
1
484.
Sozio, P., Fiorito, J., Di Giacomo, V., Di Stefano, A., Marinelli, L., Cacciatore, I., Cataldi, A.,
Pacella, S., Turkez, H., Parenti, C., Rescifina, A., Marrazzo, A., 2015. Haloperidol metabolite II
prodrug: Asymmetric synthesis and biological evaluation on rat C6 glioma cells. Euro. J. Med. Chem.
9
0, 1–9.
Sun, Z., Murry, D. J., Sanghani, S. P., Davis, W. I., Kedishvili, N. Y., Zou, Q., Hurley, T. D., Bosron,
W. F. 2004. Methylphenidate Is Stereoselectively Hydrolyzed by Human, J. Pharmacol. Exp. Ther.
3
10(2), 469–476.
Takahashi, M., Ogawa, T., Kashiwagi, H., Fukushima, F., Yoshitsugu, M., Haba, M., Hosokawa, M.,
018. Chemical synthesis of an indomethacin ester prodrug and its metabolic activation by human
2
carboxylesterase 1. Bioorg. Med. Chem. Lett. 28(6), 997–1000.
Takai, S., Matsuda, A., Usami, Y., Adachi, T., Sugiyama, T., Katagiri, Y., Tatemathu, M., Hirano, K.,
1
997. Hydrolytic profile for ester- or amide-linkage by carboxylesterases pI 5.3 and 4.5 from human
liver. Biol. Pharm. Bull. 20(8), 869–873.

ACCEPTED MANUSCRIPT
Zhang, J., Burnell, J. O. E. C., Dumaual, N., Bosron, W. F., 1999. Binding and Hydrolysis of
Meperidine by Human Liver Carboxylesterase hCE-1, J. Pharmacol. Exp. Ther. 290(1), 314–318.

ACCEPTED MANUSCRIPT
a
Scheme 1. Synthesis of haloperidol alcohol ester prodrugs (2a–2d, 2f–2h). 2e was purchased from
b
Wako Chemical Ind. no product was obtained in this condition.
Scheme 2. Synthesis of pivalate prodrugs (2i, 3) by haloperidol (1) and pivaloyl chloride.

ACCEPTED MANUSCRIPT
Fig. 1. Hydrolysis rates of haloperidol ester prodrugs (2a–2i, 3) in HLM solution. Values are
means±S.D (n = 3). nd: not detected.
Fig. 2. Hydrolysis rates of haloperidol ester prodrugs (2a–2i, 3) in HIM solution. Values are
means±S.D (n = 3). nd: not detected.
Fig. 3. Plasma Stabilities of haloperidol esters (2a, 2c, 2e, 2h, 3) in human plasma solution. Values
are means±S.D. (n = 3).

ACCEPTED MANUSCRIPT
Scheme 3. Proposed mechanism for the hydrolysis of an alcohol ester prodrug by CES.
Scheme 4. Proposed mechanism for the hydrolysis of an enol ester prodrug by CES.

ACCEPTED MANUSCRIPT
Table 1
Chemical properties of haloperidol (1) and ester prodrugs (2f–2i, 3). ClogP was
calculated using Chem3D 15.0. HPLC retention time was measured under
method B.
Compound
ClogP
HPLC retention
time (min)
9.3
1
3.85
4.70
5.76
6.82
7.88
8.94
6.29
6.07
6.16
5.94
5.42
2
2
2
2
2
2
2
2
2
3
a
b
c
d
e
f
13.7
18.6
24.3
28.8
31.4
22.1
g
h
i
22.0
21.7
21.1
17.8
Table 2
Kinetic parameters of haloperidol ester prodrugs (2a–2i, 3) in hCES2 solutions.
Values are means ± S.E. (n = 3). nd: not detected.
Compound
Vmax (nmol/mg
proteim/min)
23.0±4.80
22.4±1.76
29.8±1.66
23.4±1.85
11.2±1.42
27.0±3.32
nd
Km (μM)
CLint (mL/mg
proteiin/min)
0.0133
0.0858
1.16
2
2
2
2
2
2
2
2
2
3
a
b
c
d
e
f
1730±979
261±103
25.6±3.68
53.4±8.58
52.6±13.0
36.4±10.3
nd
0.438
0.213
0.742
g
h
i
nd
2.27±0.276
nd
25.5±8.16
nd
0.0889
nd
2.97±0.557
4.21±3.59
0.705

ACCEPTED MANUSCRIPT
Table 3
Kinetic parameters of haloperidol ester prodrugs (2f–2i, 3) in hCES1 solutions.
Values are means ± S.E. (n = 3). nd: not detected.
Compound
Vmax (nmol/mg
Km (μM)
CLint (mL/mg
proteim/min)
proteiin/min)
2
2
2
2
3
f
nd
nd
nd
g
h
i
nd
nd
nd
nd
nd
nd
nd
nd
nd
16.8±0.679
5.87±0.960
10.9

ACCEPTED MANUSCRIPT
Graphical abstract

Figure 1

Figure 2

Figure 3