Metabolism of fentanyl and acetylfentanyl in human-induced pluripotent stem cell-derived hepatocytes

Article abstract of DOI:10.1248/bpb.b17-00709

To evaluate the capability of human-induced pluripotent stem cell-derived hepatocytes (h-iPS-HEP) in drug metabolism, the profiles of the metabolites of fentanyl, a powerful synthetic opioid, and acetylfentanyl, an N-acetyl analog of fentanyl, in the cells were determined and analyzed. Commercially available h-iPSHEP were incubated with fentanyl or acetylfentanyl for 24 or 48 h. After enzymatic hydrolysis, the medium was deproteinized with acetonitrile, then analyzed by LC/MS. Desphenethylated metabolites and some hydroxylated metabolites, including 4′-hydroxy-fentanyl and β-hydroxy-fentanyl, were detected as metabolites of fentanyl and acetylfentanyl in the medium. The main metabolite of fentanyl with h-iPS-HEP was the desphenethylated metabolite, which was in agreement with in vivo results. These results suggest that h-iPSHEP may be useful as a tool for investigating drug metabolism.

Full text of DOI:10.1248/bpb.b17-00709

106
Biol. Pharm. Bull. 41, 106–114 (2018)
Vol. 41, No. 1
Regular Article
Metabolism of Fentanyl and Acetylfentanyl in Human-Induced
Pluripotent Stem Cell-Derived Hepatocytes
Tatsuyuki Kanamori,* Yuko Togawa Iwata, Hiroki Segawa, Tadashi Yamamuro,
Kenji Kuwayama, Kenji Tsujikawa, and Hiroyuki Inoue
National Research Institute of Police Science; 6–3–1 Kashiwanoha, Kashiwa, Chiba 277–0882, Japan.
Received September 1, 2017; accepted October 6, 2017
To evaluate the capability of human-induced pluripotent stem cell-derived hepatocytes (h-iPS-HEP) in
drug metabolism, the profiles of the metabolites of fentanyl, a powerful synthetic opioid, and acetylfentanyl,
an N-acetyl analog of fentanyl, in the cells were determined and analyzed. Commercially available h-iPS-
HEP were incubated with fentanyl or acetylfentanyl for 24 or 48h. After enzymatic hydrolysis, the medium
was deproteinized with acetonitrile, then analyzed by LC/MS. Desphenethylated metabolites and some hy-
droxylated metabolites, including 4ꢀ-hydroxy-fentanyl and β-hydroxy-fentanyl, were detected as metabolites
of fentanyl and acetylfentanyl in the medium. The main metabolite of fentanyl with h-iPS-HEP was the
desphenethylated metabolite, which was in agreement with in vivo results. These results suggest that h-iPS-
HEP may be useful as a tool for investigating drug metabolism.
Key words induced pluripotent stem cell; hepatocyte; fentanyl; metabolism
Human primary hepatocytes (h-PRM-HEP) are widely used Germany). h-iPS-HEP (Cellartis™ enhanced h-iPS-HEP, dif-
for drug metabolism studies because they are highly active ferentiated from ChiPSC18 and ChiPSC22), Cellartis HEP
in phase I and phase II drug metabolism.1) h-PRM-HEP are Coat, and Cellartis HEP supplement were purchased from
regarded as one of the best tools for the investigation of the TaKaRa Bio Inc. (Kusatsu, Japan). h-iPS-HEP (ReproHepa-
metabolism of drugs; however, there are problems with their to™), ReproHepato culture medium, and ReproHepato assay
use, including maintaining a stable supply of cells of a con- medium were purchased from ReproCELL Inc. (Yokohama,
sistent quality and occasional significant drops in cell viability Japan). h-PRM-HEP and KLC-SuM medium were purchased
resulting from the fragility of the cells.2)
from Kurabo Industries (Osaka, Japan). InVitroGRO CP
Induced pluripotent stem (iPS) cells, first developed in medium and InVitroGRO HT medium were purchased from
2006, can differentiate into many different types of cells, in- BioreclamationIVT (New York, NY, U.S.A.). Y-27632 was
cluding neurons, cardiomyocytes, and hepatocytes.3) Human purchased from Wako Pure Chemical Industries, Ltd. (Osaka,
iPS cell-derived hepatocytes (h-iPS-HEP) reportedly have Japan). Matrigel™ was purchased from Corning (Corning, NY,
catalytic activity from various CYP enzymes, in addition to U.S.A.). Phosphate-buffered saline (PBS), Williams Medium
liver-specific functions, such as albumin secretion and ammo- E and Leibovitz’s L-15 medium were purchased from Thermo
nia metabolism.4) Furthermore, h-iPS-HEP are robust and can Fisher Scientific (Waltham, MA, U.S.A.). All other reagents
be supplied stably, making these cells a potentially useful tool used were of analytical grade.
for drug metabolism studies.
Synthesis of Authentic Standards of Drugs and Metabo-
In this study, we investigated the ability of h-iPS-HEP to lites All synthesized standards were confirmed by positive
1
metabolize drugs using fentanyl and acetylfentanyl (Fig. 1) as electrospray ionization (ESI) mass spectrometry and H-NMR.
model drugs. Fentanyl is a powerful synthetic opioid, which is ESI mass spectra were obtained from a LCQ FLEET ion trap
1
used as an analgesic in patients with cancer.5) Acetylfentanyl mass spectrometer (Thermo Fisher Scientific). H-NMR spec-
is an N-acetyl analog of fentanyl, which is used as a substitute tra were measured on a JNM-ECA600 NMR spectrometer
for recreational opioid drugs, such as heroin.6) Fentanyl is (JEOL, Akishima, Japan). Tetramethylsilane was used as an
known to be metabolized by desphenethylation, 4′-hydroxyl- internal standard.
ation, and hydroxylation of the N-propionyl group.7) Acetylfen-
N-Phenyl-N-[1-(2-phenylethyl)-4-piperidinyl]propan-
tanyl is metabolized in a similar manner to fentanyl.8) We amide (Fentanyl) Fentanyl was synthesized according to the
evaluated how accurately the in vivo patterns of fentanyl and method of Siegfried.10) Briefly, 1-(2-phenylethyl)-4-piperidone
acetylfentanyl metabolism can be reproduced using h-iPS- and aniline were condensed in the presence of 3Å molecular
HEP.
MATERIALS AND METHODS
Materials Authentic standards of fentanyl, acetylfentanyl,
and their metabolites were synthesized in our laboratory. cis-
3-Methylfentanyl was synthesized in our laboratory by the
method reported previously.9) β-Glucuronidase/aryl sulfatase
(from Helix pomatia; β-glucuronidase, 32units/mL; aryl sul-
fatase, 102units/mL) was purchased from Merck (Darmstadt,
Fig. 1. Chemical Structures of Fentanyl and Acetylfentanyl
*To whom correspondence should be addressed. e-mail: kanamori@nrips.go.jp
© 2018 The Pharmaceutical Society of Japan

Vol. 41, No. 1 (2018)
Biol. Pharm. Bull.
107
sieves. The product was reduced by sodium borohydride to
4′-Hydroxy-fentanyl (free base)—1H-NMR (CDCl3) δ:
give N-phenyl-1-(2-phenylethyl)-4-piperidinamine (despropio- 1.01 (3H, t, J=7.1Hz), 1.96 (2H, q, J=7.1Hz), 1.98–2.05 (2H,
nyl-fentanyl, 0.897g, 65% yield from 1g of 1-(2-phenylethyl)- m), 2.11–2.21 (2H, m), 2.74–2.84 (2H, m), 3.01–3.16 (4H, m),
4-piperidone), and this was then propionylated with propionyl 3.55–3.64 (2H, m), 4.74–4.84 (1H, m), 6.78 (2H, d, J=8.7Hz),
chloride to give fentanyl. The free base of fentanyl was 7.03 (2H, d, J=8.7Hz), 7.09 (2H, d, J=7.2Hz), 7.40–7.50 (3H,
converted to the hydrochloride salt using hydrochloric acid m).
(84mg, 63% yield from 100mg of despropionyl-fentanyl).
Acetylfentanyl (N-phenyl-N-[1-(2-phenylethyl)-4-piperidinyl]- 1.78 (3H, s), 1.97–2.06 (2H, m), 2.13–2.22 (2H, m), 2.74–2.84
4′-Hydroxy-acetylfentanyl (free base)—1H-NMR (CDCl3) δ:
acetamide) hydrochloride was synthesized by the same meth- (2H, m), 3.00–3.15 (4H, m), 3.55–3.63 (2H, m), 4.74–4.83 (1H,
od (90mg, 70% yield from 100mg of despropionyl-fentanyl).
m), 6.79 (2H, d, J=7.8Hz), 7.02 (2H, d, J=7.8Hz), 7.10 (2H, d,
Fentanyl (free base)—1H-NMR (CDCl3) δ: 1.02 (3H, t, J= J=6.6Hz), 7.40–7.49 (3H, m).
7.3Hz), 1.96 (2H, q, J=7.3Hz), 1.97–2.02 (2H, m), 2.18 (2H,
4′-Hydroxy-3′-methoxy-fentanyl (free base)—1H-NMR
qd, J=3.3Hz, 13.0Hz), 2.80 (2H, qd, J=2.2Hz, 11.6Hz), (CDCl3) δ: 1.02 (3H, t, J=7.5Hz), 1.96 (2H, q, J=7.5Hz),
3.08–3.13 (2H, m), 3.19–3.23 (2H, m), 3.60 (2H, brd, J= 1.98–2.05 (2H, m), 2.18 (2H, qd, J=3.7Hz, 13.0Hz), 2.75–2.85
10.8Hz), 4.79 (1H, tt, J=3.8Hz, 12.3Hz), 7.08–7.12 (2H, m), (2H, m), 3.03–3.17 (4H, m), 3.59 (2H, brd, J=12.0Hz),
7.20–7.32 (5H, m), 7.40–7.49 (3H, m).
3.87 (3H, s), 4.79 (1H, tt, J=3.5Hz, 12.2Hz), 6.66 (1H,
Acetylfentanyl (free base)—1H-NMR (CDCl3) δ: 1.78 (3H, dd, J=1.5Hz, 7.8Hz), 6.77 (1H, d, J=1.5Hz), 6.83 (1H, d,
s), 1.98–2.02 (2H, m), 2.20 (2H, qd, J=3.3Hz, 13.1Hz), 2.80 J=7.8Hz), 7.09–7.13 (2H, m), 7.40–7.50 (3H, m).
(2H, qd, J=2.1Hz, 11.7Hz), 3.07–3.13 (2H, m), 3.19–3.23 (2H,
m), 3.60 (2H, brd, J=10.8Hz), 4.78 (1H, tt, J=3.5Hz, 12.2Hz), NMR (CDCl3) δ: 1.78 (3H, s), 1.98–2.06 (2H, m), 2.19 (2H,
7.08–7.12 (2H, m), 7.20–7.33 (5H, m), 7.40–7.49 (3H, m). qd, J=3.5Hz, 13.1Hz), 2.75–2.84 (2H, m), 3.03–3.17 (4H,
4′-Hydroxy-3′-methoxy-acetylfentanyl (free base)—1H-
N-Phenyl-N-(4-piperidinyl)propanamide (Nor-fentanyl) m), 3.59 (2H, brd, J=10.8Hz), 3.87 (3H, s), 4.78 (1H, tt,
1-Benzyl-4-piperidone and aniline were condensed in the J=3.4Hz, 12.2Hz), 6.66 (1H, dd, J=1.8Hz, 8.1Hz), 6.77
presence of 3Å molecular sieves, followed by reduction with (1H, d, J=1.8Hz), 6.83 (1H, d, J=8.1Hz), 7.09–7.12 (2H, m),
sodium borohydride and propionylation with propionyl chlo- 7.40–7.50 (3H, m).
ride. Debenzylation of the product by catalytic hydrogena-
N-Phenyl-N-[1-(2-phenylethyl)-4-piperidinyl]-3-hydroxy-
tion under acidic conditions (hydrochloric acid in methanol) propanamide (ω-Hydroxy-fentanyl) To a solution of 96mg
using 5% palladium on carbon as a catalyst gave nor-fentanyl of 3-benzyloxypropionic acid in 1mL of toluene, 2 drops of
hydrochloride (352mg, 70% yield from 356mg of 1-benzyl- N,N-dimethylformamide and 129mg of thionyl chloride were
4-piperidone). Nor-acetylfentanyl (N-phenyl-N-(4-piperidinyl)- added and the mixture was heated at 90°C for 1.5h. The
acetamide) hydrochloride was synthesized by the same meth- reaction mixture was added dropwise to a solution of 30mg
od (286mg, 69% yield from 309mg of 1-benzyl-4-piperidone). of despropionyl-fentanyl in 1mL of pyridine. After the ad-
Nor-fentanyl hydrochloride—1H-NMR (CD3OD) δ: 0.99 dition, water was added to the reaction mixture, the mixture
(3H, t, J=7.2Hz), 1.59 (2H, qd, J=4.2Hz, 13.0Hz), 1.98 was acidified with 3M hydrochloric acid, then extracted with
(2H, q, J=7.2Hz), 2.08 (2H, brd, J=13.8Hz), 3.11 (2H, td, chloroform. The extract was purified by flash chromatography
J=2.6Hz, 13.4Hz), 3.36–3.42 (2H, m), 4.77 (1H, tt, J=3.6Hz, (column, silica gel; solvent, n-hexane/ethyl acetate). Deben-
12.6Hz), 7.23–7.27 (2H, m), 7.46–7.54 (3H, m).
zylation of the product by catalytic hydrogenation under acidic
Nor-acetylfentanyl hydrochloride—1H-NMR (CD3OD) δ: conditions using 5% palladium on carbon as a catalyst gave
1.60 (2H, qd, J=4.0Hz, 13.0Hz), 1.76 (3H, s), 2.09 (2H, 25mg of ω-hydroxy-fentanyl hydrochloride (60% yield from
brd, J=13.8Hz), 3.10 (2H, td, J=3.0Hz, 13.5Hz), 3.36–3.43 despropionyl-fentanyl).
(2H, m), 4.77 (1H, tt, J=3.6Hz, 12.6Hz), 7.25–7.30 (2H, m),
7.47–7.56 (3H, m).
ω-Hydroxy-fentanyl (free base)—1H-NMR (CDCl3) δ:
1.98–2.03 (2H, m), 2.18 (2H, t, J=5.4Hz), 2.23 (2H, qd,
N-Phenyl-N-[1-[2-(4-hydroxyphenyl)ethyl]-4-piperidin- J=3.5Hz, 13.3Hz), 2.75–2.84 (2H, m), 3.08–3.14 (2H, m),
yl]propanamide (4′-Hydroxy-fentanyl) 4-Benzyloxyphen- 3.19–3.24 (2H, m), 3.61 (2H, brd, J=11.4Hz), 3.74 (2H, t,
ylacetic acid was reduced by lithium aluminum hydride to J=5.4Hz), 4.78 (1H, tt, J=3.4Hz, 12.2Hz), 7.09–7.14 (2H, m),
give 4-benzyloxyphenylethyl alcohol. 4-Benzyloxyphenylethyl 7.20–7.33 (5H, m), 7.42–7.50 (3H, m).
alcohol was treated with methanesulfonyl chloride to convert
N-Phenyl-N-[1-(2-phenylethyl)-4-piperidinyl]-2-hydroxy-
the alcohol to the methansulfonic acid ester.11) Condensation of propanamide ((ω-1)-Hydroxy-fentanyl) To a solution of
the ester with nor-fentanyl,12) followed by debenzylation gave 50mg of despropionyl-fentanyl in 2mL of dichloromethane,
4′-hydroxy-fentanyl hydrochloride (34mg, 59% yield from 40µL of triethylamine, and 40.3mg of 2-acetoxypropionyl
40mg of nor-fentanyl hydrochloride). 4′-Hydroxy-acetylfentan- chloride were added and stirred for 2h at room temperature.
yl (N-phenyl-N-[1-[2-(4-hydroxyphenyl)ethyl]-4-piperidinyl]- To the reaction mixture, 20mL of 0.1M hydrochloric acid
acetamide) hydrochloride (24mg, 41% yield from 40mg was added, and then extracted with chloroform. The solvent
of nor-acetylfentanyl hydrochloride), 4′-hydroxy-3′-methoxy- was evaporated to dryness under vacuum and the residue
fentanyl
(N-phenyl-N-[1-[2-(4-hydroxy-3-methoxyphenyl)- was treated with a small amount (ca. 15 drops) of 1M sodium
ethyl]-4-piperidinyl]propanamide) hydrochloride (26mg, 42% hydroxide in methanol for 5min to remove the acetyl group.
yield from 40mg of nor-fentanyl hydrochloride), and The solution was neutralized with 1M acetic acid in metha-
4′-hydroxy-3′-methoxy-acetylfentanyl
(N-phenyl-N-[1-[2-(4- nol, then water was added and the solution was basified with
hydroxy-3-methoxyphenyl)ethyl]-4-piperidinyl]acetamide) hy- 28% aqueous ammonia, then extracted with chloroform. The
drochloride (28mg, 44% yield from 40mg of nor-acetylfentan- residue was recrystallized from methanol to give 41mg of
yl hydrochloride) were synthesized by the same method.
(ω-1)-hydroxy-fentanyl (64% yield from despropionyl-fentan-

108
Biol. Pharm. Bull.
Vol. 41, No. 1 (2018)
yl). Hydroxyacetyl-fentanyl (N-phenyl-N-[1-(2-phenylethyl)-4- 2.88–2.98 (2H, m), 2.98–3.04 (1H, m), 3.16–3.22 (1H, m),
piperidinyl]-2-hydroxyacetamide) (free base) was synthesized 3.79–3.86 (2H, br), 4.82 (1H, tt, J=3.5Hz, 12.3Hz), 5.35 (1H,
by the same method (16mg, 27% yield from 50mg of despro- dd, J=1.8Hz, 10.2Hz), 7.10–7.14 (2H, br), 7.28–7.33 (1H, m),
pionyl-fentanyl).
7.33–7.38 (4H, m), 7.41–7.50 (3H, m).
(ω-1)-Hydroxy-fentanyl (free base)—1H-NMR (CDCl3)
Incubation of Drugs with h-iPS-HEP (Cellartis™) Cel-
δ: 1.09 (3H, d, J=6.5Hz), 1.87–1.93 (1H, m), 2.09–2.14 (1H, lartis enhanced h-iPS-HEP (≥1.23×107 viable cells/vial, from
m), 2.18 (1H, qd, J=3.9Hz, 13.3Hz), 2.36 (1H, qd, J=4.2Hz, ChiPSC18 or ChiPSC22) were thawed in 40mL of warm
13.2Hz), 2.75–2.85 (2H, m), 3.08–3.14 (2H, m), 3.19–3.25 InVitroGRO HT medium that contained 5µM of Y-27632.
(2H, m), 3.57–3.66 (2H, m), 4.75 (1H, tt, J=3.7Hz, 12.2Hz), After centrifugation at 100×g for 2min, the supernatant was
7.11–7.33 (7H, m), 7.45–7.53 (3H, m).
removed, then the cells were resuspended in 15mL of warm
Hydroxyacetyl-fentanyl (free base)—1H-NMR (CDCl3) δ: InVitroGRO CP medium that contained 5µM of Y-27632. A
2.04 (2H, brd, J=13.8Hz), 2.28 (2H, qd, J=3.6Hz, 12.9Hz), 1mL portion of the cell suspension was dispensed into each
2.77–2.86 (2H, m), 3.09–3.14 (2H, m), 3.20–3.24 (2H, m), well of a 24-well microplate coated with Cellartis HEP Coat
3.60–3.64 (2H, m), 3.70 (2H, s), 4.76 (1H, tt, J=3.4Hz, and incubated at 37°C and 5% CO2. After 1, 3, and 5d from
12.3Hz), 7.10–7.14 (2H, m), 7.20–7.33 (5H, m), 7.47–7.52 (3H, the start of incubation, the medium was changed to 0.5mL
m).
of warm fresh Williams Medium E that contained 2% Cel-
N-Phenyl-N-[1-(2-hydroxy-2-phenylethyl)-4-piperidin- lartis HEP supplement solution and 0.5% dimethyl sulfoxide.
yl]propanamide (β-Hydroxy-fentanyl) A mixture of 40mg After changing the medium at day 5, the drug (fentanyl hy-
of nor-fentanyl hydrochloride, 28mg of phenacyl chloride, and drochloride or acetylfentanyl hydrochloride dissolved in PBS)
15mg of sodium bicarbonate in 1.2mL of water–2-butanone was added to each well of the plate at a final concentration
(1:5) was heated at 90°C for 6h. Water was added to the reac- of 10µM, and incubated for 24 or 48h. The medium was col-
tion mixture, and the mixture was basified with 28% aqueous lected and stored at −30°C until analysis.
ammonia, then extracted with ethyl acetate. The extract was
Incubation of Drugs with h-iPS-HEP (ReproHepato™)
purified by flash chromatography (column, silica gel; solvent, ReproHepato h-iPS-HEP (8.2×106 cells/vial) was thawed in
chloroform/ethyl acetate). The purified product, β-hydroxy- 49mL of warm Leibovitz’s L-15 medium. After centrifuga-
fentanyl was converted to the hydrochloride salt (18mg, 31% tion at 350×g for 5min, the supernatant was removed, then
yield from 40mg of nor-fentanyl hydrochloride). β-Hydroxy- the cells were resuspended in 11mL of warm ReproHepato
acetylfentanyl
(N-phenyl-N-[1-(2-hydroxy-2-phenylethyl)-4- culture medium. A 640µL portion of the cell suspension was
piperidinyl]acetamide) hydrochloride was synthesized by the dispensed into each well of a 24-well microplate coated with
same method (34mg, 58% yield from 40mg of nor-acetylfen- Matrigel, and incubated at 37°C and 5% CO2. After 1, 3, and
tanyl hydrochloride).
5d from the start of incubation, the medium was changed to
β-Hydroxy-fentanyl (free base)—1H-NMR (CDCl3) δ: 1.02 0.5mL of warm fresh ReproHepato culture medium. After
(3H, t, J=7.3Hz), 1.97 (2H, q, J=7.3Hz), 1.98–2.02 (1H, m), 6d from the start of incubation, the medium was changed to
2.04–2.10 (1H, m), 2.19 (1H, qd, J=3.7Hz, 13.0Hz), 2.27 (1H, 0.5mL of warm fresh ReproHepato assay medium, then the
qd, J=3.9Hz, 13.1Hz), 2.88–2.98 (2H, m), 2.98–3.04 (1H, m), drug (fentanyl hydrochloride or acetylfentanyl hydrochloride
3.16–3.23 (1H, m), 3.79–3.86 (2H, br), 4.83 (1H, tt, J=3.8Hz, dissolved in PBS) was added to each well of the plate at a
12.4Hz), 5.34 (1H, dd, J=2.4Hz, 10.2Hz), 7.02–7.13 (2H, br), final concentration of 10µM, and incubated for 24 or 48h. The
7.28–7.33 (1H, m), 7.33–7.38 (4H, m), 7.41–7.50 (3H, m).
medium was collected and stored at −30°C until analysis.
Incubation of Drugs with h-PRM-HEP h-PRM-HEP
β-Hydroxy-acetylfentanyl (free base)—1H-NMR (CDCl3)
δ: 1.79 (3H, s), 1.97–2.03 (1H, m), 2.05–2.10 (1H, m), 2.21 (2×106 viable cells/vial) was thawed in 20mL of warm Leibo-
(1H, qd, J=3.9Hz, 13.1Hz), 2.29 (1H, qd, J=3.7Hz, 13.3Hz), vitz’s L-15 medium. After centrifugation at 170×g for 2min,
Table 1. Monitoring Ion, Collision Energy and Calibration Curve for Analysis of Compounds by Liquid Chromatography-Tandem Mass Spectrometry
Monitoring ion (m/z)
Compound
Collision energy (eV)
Calibration curve (µM)
Precursor ([M+H]+)
Product
Fentanyl
337.1
233.1
353.1
353.1
353.1
353.1
383.1
323.1
219.1
339.1
339.1
339.1
369.1
351.1
188.2
84.2
21
17
21
20
32
19
30
20
18
20
31
16
28
24
0.021–11
0.030–15
0.021–5.1
0.023–5.7
0.021–5.1
0.021–5.1
0.019–4.8
0.022–11
0.031–16
0.024–5.9
0.021–5.3
0.021–5.3
0.020–4.9
—
Nor-fentanyl
ω-Hydroxy-fentanyl
188.2
188.2
121.1
204.2
151.2
188.2
84.2
(ω-1)-Hydroxy-fentanyl
4′-Hydroxy-fentanyl
β-Hydroxy-fentanyl
4′-Hydroxy-3′-methoxy-fentanyl
Acetylfentanyl
Nor-acetylfentanyl
Hydroxyacetyl-fentanyl
4′-Hydroxy-acetylfentanyl
β-Hydroxy-acetylfentanyl
4′-Hydroxy-3′-methoxy-acetylfentanyl
cis-3-Methylfentanyl (Internal standard)
188.2
121.1
321.3
151.2
202.2

Vol. 41, No. 1 (2018)
Biol. Pharm. Bull.
109
Fig. 2. Total Ion Current Chromatograms (TIC) and Extracted Ion Chromatograms (EIC) Obtained from the Medium of the Hepatocytes Cultured
with Fentanyl
(a), Human iPS cell-derived hepatocytes (Cellartis h-iPS-HEP, from ChiPSC22); (b), Human primary hepatocytes (h-PRM-HEP). F-1, fentanyl; F-2, nor-fentanyl; F-3,
ω-hydroxy-fentanyl; F-4, (ω-1)-hydroxy-fentanyl; F-5, 4′-hydroxy-fentanyl; F-6, β-hydroxy-fentanyl; F-7, 4′-hydroxy-3′-methoxy-fentanyl.
into each well of a 24-well microplate. The drug (fentanyl hy-
drochloride or acetylfentanyl hydrochloride dissolved in PBS)
was added to each well of the plate at a final concentration of
10 µM, and incubated for 3h with shaking (90rpm) at 37°C and
5% CO2. The medium was collected and stored at −30°C until
analysis.
Identification of the Metabolites To a 25µL sample of
the medium, 15µL of 0.25M acetate buffer (pH 5.0) containing
β-glucuronidase/aryl sulfatase (β-glucuronidase, 0.01 unit) was
added and then incubated at 60°C for 1.5h. To the reaction
mixture, 250µL of acetonitrile was added and the mixture was
vortexed for 5s. After centrifugation at 10000×g for 5min,
the supernatant was evaporated to dryness under a nitrogen
stream. The residue was reconstituted with 50µL of the initial
mobile phase and centrifuged at 10000×g for 5min, then the
supernatant (10µL) was analyzed by LC-ion trap MS under
scan and product ion analysis modes. The conditions of analy-
Fig. 3. Electrospray Ionization (ESI) Product Ion Spectra of the Peaks
Detected in Fig. 2
sis were as follows: apparatus, Accela LC system connected
to LCQ FLEET ion trap mass spectrometer (Thermo Fisher
The protonated molecule of each compound was selected as the precursor ion.
Scientific); column, CORTECS C18 (2.1×50mm, 2.7µm, Wa-
F-1, fentanyl; F-2, nor-fentanyl; F-3, ω-hydroxy-fentanyl; F-4, (ω-1)-hydroxy-
ters, Milford, MA, U.S.A.) maintained at 40°C; mobile phase
composition, 0.1% formic acid (A) and methanol (B); linear
gradient mode, 20% B for 1min, 20% to 80% B over 8min,
fentanyl; F-5, 4′-hydroxy-fentanyl; F-6, β-hydroxy-fentanyl; F-7, 4′-hydroxy-3′-
methoxy-fentanyl.
the supernatant was removed, then the cells were resuspended 80% B for 2min, and 80% to 20% B over 0.1min; flow rate,
in 2mL of warm KLC-SuM medium and the cell viability was 0.2mL/min; MS interface, positive ESI; analysis mode, scan
measured by the trypan blue dye exclusion method. The cell (m/z 100–500) and product ion analysis (normalized collision
suspension was diluted to 5×105 viable cells/mL with KLC- energy, 35%; precursor ions, protonated molecules of drugs
SuM medium and 0.5mL of the cell suspension was dispensed and putative metabolites).

110
Biol. Pharm. Bull.
Vol. 41, No. 1 (2018)
Fig. 4. Proposed Metabolic Pathways for Fentanyl
Fig. 5. TICs and EICs Obtained from the Medium of the Hepatocytes Cultured with Acetylfentanyl
(a), Cellartis h-iPS-HEP (from ChiPSC22); (b), h-PRM-HEP; AF-1, acetylfentanyl; AF-2, nor-acetylfentanyl; AF-3, 4′-hydroxy-acetylfentanyl; AF-4, hydroxyacetyl-
fentanyl; AF-5, β-hydroxy-acetylfentanyl; AF-6, 4′-hydroxy-3′-methoxy-acetylfentanyl.
Quantitation of the Metabolites To a 25µL sample of reaction mixture, 10µL of internal standard solution (cis-
the medium, 15µL of 0.25M acetate buffer (pH 5.0) contain- 3-methylfentanyl hydrochloride, 50ng/10µL water) was added,
ing β-glucuronidase/aryl sulfatase (β-glucuronidase, 0.01 then 250µL of acetonitrile was added and the mixture was
unit) was added and then incubated at 60°C for 1.5h. To the vortexed for 5s. After centrifugation at 10000×g for 5min,

Vol. 41, No. 1 (2018)
Biol. Pharm. Bull.
111
50µL of the supernatant was mixed with 200µL of 0.1% for- for fentanyl hydrochloride, Table 1) for fentanyl, acetylfen-
mic acid. This was centrifuged at 10000×g for 5min, then tanyl, nor-fentanyl, and nor-acetylfentanyl and 0.008–2µg/mL
the supernatant (10µL) was analyzed by LC-triple quadrupole for the other compounds, with a correlation coefficient of 0.99.
MS. The conditions of analysis were as follows: apparatus,
NANOSPACE SI-2 LC system (Shiseido, Tokyo, Japan) con-
nected to TSQ Quantum triple quadrupole mass spectrometer
(Thermo Fisher Scientific); column, mobile phase composi-
RESULTS AND DISCUSSION
Identification of the Metabolites of Fentanyl and
tion, flow rate, and MS interface were the same as for iden- Acetylfentanyl The culture media obtained from h-iPS-HEP
tification of the metabolites; analysis mode, selected reaction and h-PRM-HEP were treated with conjugate-hydrolyzing
monitoring (SRM). SRM parameters are listed in Table 1.
enzymes, then deproteinated with acetonitrile, and analyzed
Calibration Curve Authentic standards of fentanyl, by LC-ion trap MS to identify the metabolites. The total ion
acetylfentanyl, and their metabolites were added to the InVi- chromatograms (TIC) and extracted ion chromatograms (EIC)
troGRO CP medium and processed as described above to obtained from the deproteinized media of h-iPS-HEP (Cellar-
obtain the calibration curves. Excellent linearity was obtained tis, from ChiPSC22) and h-PRM-HEP incubated with fentanyl
over the concentration range 0.008–4µg/mL (e.g. 0.021–11µM are shown in Fig. 2. Peaks corresponding to the desphenethyl-
ated metabolite (nor-fentanyl, F-2), ω-hydroxy-fentanyl (F-3),
(ω-1)-hydroxy-fentanyl (F-4), 4′-hydroxy-fentanyl (F-5), and
β-hydroxy-fentanyl (F-6) were detected, as the metabolites of
fentanyl, in the chromatograms obtained from h-iPS-HEP (a)
and h-PRM-HEP (b). The peak from 4′-hydroxy-3′-methoxy-
fentanyl (F-7) was detected in the chromatogram from h-
PRM-HEP (b), but not in the chromatogram from h-iPS-HEP
(a). The product ion spectra of the peaks F-1–F-7 are shown
in Fig. 3. All metabolites were identified by comparison of
the retention times and mass spectra with authentic standards.
The proposed metabolic pathways for fentanyl are shown in
Fig. 4. β-Hydroxy-fentanyl and 4′-hydroxy-3′-methoxy-fentanyl
were the first to be identified as metabolites of fentanyl in
the present study. 4′-Hydroxy-3′-methoxy-fentanyl is thought
to be formed via methylation of the intermediate metabolite,
3′,4′-dihydroxy-fentanyl. 4′-Hydroxy-3′-methoxy-fentanyl was
not formed in any h-iPS-HEP, indicating that the activity of
the enzyme involved in the methylation (i.e., catechol-O-meth-
yltransferase, COMT) was negligible in h-iPS-HEP.
The TICs and EICs obtained from h-iPS-HEP (Cellartis,
from ChiPSC22) and h-PRM-HEP incubated with acetylfen-
Fig. 6. ESI Product Ion Spectra of the Peaks Detected in Fig. 5
The protonated molecule of each compound was selected as the precursor
ion. AF-1, acetylfentanyl; AF-2, nor-acetylfentanyl; AF-3, 4′-hydroxy-acetylfen-
tanyl; AF-4, hydroxyacetyl-fentanyl; AF-5, β-hydroxy-acetylfentanyl; AF-6,
4′-hydroxy-3′-methoxy-acetylfentanyl.
Fig. 7. Proposed Metabolic Pathways for Acetylfentanyl

112
Biol. Pharm. Bull.
Vol. 41, No. 1 (2018)
tanyl are shown in Fig. 5. The peaks corresponding to the moiety followed by further hydroxylation and methylation
desphenethylated metabolite (nor-acetylfentanyl, AF-2), 4′-hy- of one of the hydroxyl groups.8) However, this group did not
droxy-acetylfentanyl (AF-3), hydroxyacetyl-fentanyl (AF-4), confirm the exact chemical structure of the metabolites. In
and β-hydroxy-acetylfentanyl (AF-5) were detected in the this study, we confirm the exact structure of the metabolites of
chromatograms obtained from h-iPS-HEP (a) and h-PRM- acetylfentanyl for the first time. β-Hydroxy-acetylfentanyl and
HEP (b). As for fentanyl, the 4′-hydroxy-3′-methoxy metabo- hydroxyacetyl-fentanyl are novel metabolites of acetylfentanyl.
lite (4′-hydroxy-3′-methoxy-acetylfentanyl, AF-6) was detected In previous reports, the N-desacyl metabolite (despropionyl-
in the medium of the h-PRM-HEP, but not in the medium of fentanyl) was found to be formed from both fentanyl and
the h-iPS-HEP. The product ion spectra of the peaks AF-1– acetylfentanyl8,13); however, this desacyl metabolite was not
AF-6 are shown in Fig. 6. The proposed metabolic pathways detected in any cell culture systems in the present study.
for acetylfentanyl are shown in Fig. 7. Melent’ev et al. have
Metabolite Profiles of Fentanyl and Acetylfentanyl in
reported that acetylfentanyl was metabolized by despheneth- h-iPS-HEP and h-PRM-HEP The deproteinated culture
ylation, deacetylation, and hydroxylation of the phenylethyl media were analyzed by LC-triple quadrupole MS under SRM
mode to quantitate the metabolites. The time–courses of the
formation of the main metabolites of fentanyl in Cellartis h-
iPS-HEP (ChiPSC18) are shown in Fig. 8. The amounts of all
of the metabolites of fentanyl increased with the incubation
time. Similar results were obtained for the other h-iPS-HEPs.
The metabolite profiles of fentanyl in h-iPS-HEP (48h incuba-
tion) and h-PRM-HEP (3h incubation) are shown in Fig. 9. The
majority of fentanyl (60–85%) remained unmetabolized after
incubation with the hepatocytes. The main metabolite of fen-
tanyl formed by h-iPS-HEP was nor-fentanyl, and the amounts
ranged from 3.3–14.5% of the initial amount of fentanyl. This
metabolite was also the main metabolite formed by h-PRM-
HEP (13.8% of the initial amount of fentanyl). 4′-Hydroxy-fen-
tanyl (0.2–2.4%) and β-hydroxy-fentanyl (0.9–3.5%) were the
next largest amounts of metabolites formed in h-iPS-HEP and
h-PRM-HEP. ω-Hydroxy-fentanyl and (ω-1)-hydroxy-fentanyl
were minor metabolites. 4′-Hydroxy-3′-methoxy-fentanyl was
formed only in h-PRM-HEP; however, the amount was very
small (less than 0.2% of the initial amount of fentanyl). Ac-
cording to a previous report, the urinary excretion of nor-fen-
tanyl and 4′-hydroxy-fentanyl, 24h after continuous infusion
Fig. 8. Time–Course of the Metabolite Formation of Fentanyl in Cel-
lartis h-iPS-HEP (ChiPSC18)
Each value is the average of two determinations.
Fig. 9. Metabolite Formation of Fentanyl in h-iPS-HEP and h-PRM-HEP
Each value is the average of two determinations.

Vol. 41, No. 1 (2018)
Biol. Pharm. Bull.
113
of fentanyl to patients (2.1–3.0mg/kg body weight), accounted why β-hydroxy-fentanyl has never been identified in bio-
for 8–25 and 3–6% of the initial amount of fentanyl, re- logical specimens from fentanyl users when it was formed in
spectively.7) The in vivo metabolite profile of fentanyl in this relatively large amounts in h-iPS-HEP and h-PRM-HEP. The
report was in good accordance with the in vitro metabolite metabolite profiles of fentanyl varied widely between the h-
profiles obtained in our study. In both cases, the first and iPS-HEP products. This was because of differences in the ex-
second largest amounts of fentanyl metabolites formed were pression of the drug-metabolizing enzymes in each h-iPS-HEP
nor-fentanyl and 4′-hydroxy-fentanyl, respectively. It is unclear product. According to the data sheets from the manufacturer,
Cellartis h-iPS-HEP from ChiPSC18 has substantially higher
CYP3A4 activity (more than 10 times) than h-iPS-HEP from
ChiPSC22. As expected, more nor-fentanyl, which is produced
by CYP3A4 from fentanyl,14) was formed in Cellartis h-iPS-
HEP from ChiPSC18 than in those from ChiPSC22. Although
the data are limited, above results suggested that the activity
of the drug-metabolizing enzyme in h-iPS-HEP was well cor-
related with the production of the metabolite.
The time–courses of the formation of the main metabo-
lites of acetylfentanyl in Cellartis h-iPS-HEP (ChiPSC18) are
shown in Fig. 10. As for fentanyl, the amounts of the main
metabolites of acetylfentanyl increased with the incubation
time. The metabolite profiles of acetylfentanyl in h-iPS-HEP
(48h incubation) and h-PRM-HEP (3h incubation) are shown
in Fig. 11. The main metabolites of acetylfentanyl formed
by h-iPS-HEP and h-PRM-HEP were nor-acetylfentanyl and
4′-hydroxy-acetylfentanyl, and the amounts ranged from
2.7–10.1 and 0.5–8.5% of the initial acetylfentanyl, respec-
tively. As for fentanyl, more of the desphenethylated metabo-
lite (nor-acetylfentanyl) was formed in Cellartis h-iPS-HEP
Fig. 10. Time–Course of the Metabolite Formation of Acetylentanyl in
from ChiPSC18 than in those from ChiPSC22, indicating
that this metabolic reaction was also catalyzed by CYP3A4.
Cellartis h-iPS-HEP (ChiPSC18)
Each value is the average of two determinations.
Fig. 11. Metabolite Formation of Acetylfentanyl in h-iPS-HEP and h-PRM-HEP
Each value is the average of two determinations.

114
Biol. Pharm. Bull.
(2010).
Vol. 41, No. 1 (2018)
The 4′-hydroxy-3′-methoxy metabolite was formed only in
h-PRM-HEP. β-Hydroxy-acetylfentanyl was the metabolite
formed in the third largest amount in h-iPS-HEP and h-PRM-
HEP; however, this metabolite has never been detected in in
vivo samples. Careful analysis of samples from fentanyl and
acetylfentanyl users may result in the detection of β-hydroxy
metabolites of these drugs in vivo in the future. The quantita-
tive values of the metabolites of fentanyl and acetylfentanyl
were not affected by enzymatic hydrolysis, indicating that the
metabolites did not undergo any conjugation.
3) Inoue H, Nagata N, Kurokawa H, Yamanaka S. iPS cells: a game
changer for future medicine. EMBO J., 33, 409–417 (2014).
4) Takayama K, Morisaki Y, Kuno S, Nagamoto Y, Harada K, Furuka-
wa N, Ohtaka M, Nishimura K, Imagawa K, Sakurai F, Tachibana
M, Sumazaki R, Noguchi E, Nakanishi M, Hirata K, Kawabata K,
Mizuguchi H. Prediction of interindividual differences in hepatic
functions and drug sensitivity by using human iPS-derived hepato-
cytes. Proc. Natl. Acad. Sci. U.S.A., 111, 16772–16777 (2014).
5) Barceloux DG. Fentanyl analogues. Medical toxicology of drug
abuse: synthesized chemicals and psychoactive plants. Wiley,
Hoboken, pp. 539–545 (2012).
h-iPS-HEP had drug-metabolizing activity comparable to
h-PRM-HEP, and showed a similar pattern for fentanyl me-
tabolite formation as seen in vivo, indicating that these cells
may be potentially useful for the prediction of in vivo drug
metabolism. In addition, h-iPS-HEP have the advantage that
the cells are robust (cell viability after thawing was more than
90%) and can be supplied stably. However, some enzymatic
activities (O-methylation and N-deacylation) were not detected
in h-iPS-HEP. Further study is needed to clarify the ability of
h-iPS-HEP to metabolize drug molecules.
6) Drug Enforcement Administration (DEA), Department of Justice.
Schedules of controlled substances: temporary placement of acetyl
fentanyl into schedule I. Fed. Regist., 80, 42381–42385 (2015).
7) Goromaru T, Kawauchi T, Katashima M, Matsuura H, Yoshimura
N, Sameshima T, Miyawaki T. Urinary excretion of fentanyl and its
metabolites at high-dose infusion. Masui, 35, 35–39 (1986), Japa-
nese.
8) Melent’ev AB, Kataev SS, Dvorskaya ON. Identification and ana-
lytical properties of acetyl fentanyl metabolites. J. Anal. Chem., 70,
240–248 (2015).
9) Kanamori T, Iwata YT, Segawa H, Yamamuro T, Kuwayama K,
Tsujikawa K, Inoue H. Characterization and differentiation of
geometric isomers of 3-methylfentanyl analogs by gas chromatogra-
phy/mass spectrometry, liquid chromatography/mass spectrometry,
and nuclear magnetic resonance spectroscopy. J. Forensic Sci., 62,
1472–1478 (2017).
10) “Siegfried,Synthesisoffentanyl”:‹https://erowid.org/archive/rhodium/
chemistry/fentanyl.html›, cited 25 July, 2017.
11) Crossland RK, Servis KL. Facile synthesis of methanesulfonate
esters. J. Org. Chem., 35, 3195–3196 (1970).
Acknowledgments This work was supported by the Japan
Society for the Promotion of Science (JSPS) KAKENHI Grant
Number 15K08895. We thank Victoria Muir, Ph.D., from
Edanz Group (www.edanzediting.com/ac) for editing a draft
of this manuscript.
Conflict of Interest The authors declare no conflict of
interest.
12) Srimurugan S, Murugan K, Chen C. A facile method for prepara-
tion of [2H3]-sufentanil and its metabolites. Chem. Pharm. Bull., 57,
1421–1424 (2009).
REFERENCES
13) Van Rooy HH, Vermeulen MP, Bovill JG. The assay of fentanyl
and its metabolites in plasma of patients using gas chromatography
with alkali flame ionisation detection and gas chromatography-mass
spectrometry. J. Chromatogr., 223, 85–93 (1981).
14) Labroo RB, Paine MF, Thummel KE, Kharasch ED. Fentanyl me-
tabolism by human hepatic and intestinal cytochrome P450 3A4:
implications for interindividual variability in disposition, efficacy,
and drug interactions. Drug Metab. Dispos., 25, 1072–1080 (1997).
1) Tetsuka K, Ohbuchi M, Tabata K. Recent progress in hepatocyte
culture models and their application to the assessment of drug
metabolism, transport, and toxicity in drug discovery: the value of
tissue engineering for the successful development of a microphysi-
ological system. J. Pharm. Sci., 106, 2302–2311 (2017).
2) Terry C, Dhawan A, Mitry RR, Lehec SC, Hughes RD. Optimiza-
tion of the cryopreservation and thawing protocol for human hepa-
tocytes for use in cell transplantation. Liver Transpl., 16, 229–237