Influence of stereochemistry on the bioactivation and glucuronidation of 4-ipomeanol

Article abstract of DOI:10.1124/jpet.118.249771

A potential CYP4B1 suicide gene application in engineered T-cell treatment of blood cancers has revived interest in the use of 4-ipomeanol (IPO) in gene-directed enzyme prodrug therapy, in which disposition of the administered compound may be critical. IPO contains one chiral center at the carbon bearing a secondary alcohol group; it was of interest to determine the effect of stereochemistry on 1) CYP4B1-mediated bioactivation and 2) (UGT)–mediated glucuronidation. First, (R)-IPO and (S)-IPO were synthesized and used to assess cytotoxicity in HepG2 cells expressing rabbit CYP4B1 and re-engineered human CYP4B1, where the enantiomers were found to be equipotent. Next, a sensitive UPLC-MS/ MS assay was developed to measure the IPO-glucuronide diastereomers and product stereoselectivity in human tissue microsomes. Human liver and kidney microsomes generated (R)- and (S)-IPO-glucuronide diastereomers in ratios of 57:43 and 79: 21, respectively. In a panel of 13 recombinantly expressed UGTs, UGT1A9 and UGT2B7 were the major isoforms responsible for IPO glucuronidation. (R)-IPO-glucuronide diastereoselectivity was apparent with each recombinant UGT, except UGT2B15 and UGT2B17, which favored the formation of (S)-IPO-glucuronide. Incubations with IPO and the UGT1A9-specific chemical inhibitor niflumic acid significantly decreased glucuronidation in human kidney, but only marginally in human liver microsomes, consistent with known tissue expression patterns of UGTs. We conclude that IPO glucuronidation in human kidney is mediated by UGT1A9 and UGT2B7. In human liver, it is mediated primarily by UGT2B7 and, to a lesser extent, UGT1A9 and UGT2B15. Overall, the lack of pronounced stereoselectivity for IPO’s bioactivation in CYP4B1-transfected HepG2 cells, or for hepatic glucuronidation, suggests the racemate is an appropriate choice for use in suicide gene therapies.

Full text of DOI:10.1124/jpet.118.249771

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
TITLE PAGE
Influence of Stereochemistry on the Bioactivation and Glucuronidation of
4-Ipomeanol
Aaron M. Teitelbaum, Matthew G. McDonald, John P. Kowalski, Oliver T. Parkinson,
Michele Scian, Dale Whittington, Katharina Roellecke, Helmut Hanenberg, Constanze
Wiek, and Allan E. Rettie
Department of Medicinal Chemistry, School of Pharmacy, University of Washington,
Seattle, WA 98105, USA (A.M.T., M.G.M., J.P.K., O.T.P., M.S., D.W., A.E.R.);
Department of Otorhinolaryngology and Head/Neck Surgery, Heinrich-Heine University,
40225 Düsseldorf, Germany (K.R., H.H., C.W.); Department of Pediatrics III, University
Children’s Hospital Essen, University of Duisburg-Essen, 45122 Essen, Germany (H.H.)
1

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
RUNNING TITLE PAGE
Running title: 4-Ipomeanol enantiomer metabolism
Corresponding author: Aaron M. Teitelbaum
Address: Drug Metabolism and Pharmacokinetics, Boehringer Ingelheim
Pharmaceuticals, Inc., 900 Ridgebury Road, Ridgefield, CT 06877-0368, USA
Tel: 203-791-6723
Fax: 203-791-6003
Email: aaron.teitelbaum@boehringer-ingelheim.com
Number of text pages: 29
Number of tables: 2
Number of figures: 6
Number of references: 35
Number of words in the abstract: 250
Number of words in the introduction: 611
Number of words in the discussion: 788
Nonstandard abbreviations:
CYP4B1, cytochrome P450 4B1; DLPC,1,2-Didodecanoyl-sn-glycero-3-phosphocholine;
DMEM, dulbecco’s modified eagle medium; ESI, electrospray ionization; EtOH, ethanol;
FBS, fetal bovine serum; GDEPT, gene-directed enzyme prodrug therapy; HKM, human
kidney microsomes; HLM, human liver microsomes; HPLC, high-performance liquid
chromatography, HRMS, high-resolution mass spectrometry; HSQC, heteronuclear
single quantum coherence; IPO, 4-Ipomeanol; KPi, potassium phosphate buffer; 4-
MUG, 4-methylumbelliferone glucuronide; MgCl2, magnesium chloride; NAC, N-Acetyl-L-
cysteine; NAL, Nα-Acetyl-L-lysine; Pen\Strep, Penicillin-Streptomycin; UDPGA, Uridine
5’-diphosphoglucuronic acid; UGT, Uridine 5’-diphosphoglucuronosyl transferase;
UPLC-MS/MS, ultra-performance liquid chromatography tandem-mass spectrometry;
ZnSO4, zinc sulfate
2

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Abstract: A potential CYP4B1 suicide gene application in engineered T-cell treatment
of blood cancers has revived interest in the use of 4-Ipomeanol (IPO) in gene-directed
enzyme prodrug therapy, where disposition of the administered compound may be
critical. IPO contains one chiral center at the carbon bearing a secondary alcohol
group; it was of interest to determine the effect of stereochemistry on (i) CYP4B1-
mediated bioactivation and (ii) UGT-mediated glucuronidation. (R)-IPO and (S)-IPO
were synthesized and utilized, firstly, to assess cytotoxicity in HepG2 cells expressing
rabbit CYP4B1 and re-engineered human CYP4B1, where the enantiomers were found
to be equipotent. Next, a sensitive UPLC-MS/MS assay was developed to measure the
IPO glucuronide diastereomers and product stereoselectivity in human tissue
microsomes. Human liver and kidney microsomes generated (R)- and (S)-IPO-
glucuronide diastereomers in ratios of 57:43 and 79:21, respectively. In a panel of
thirteen recombinantly expressed UGTs, UGT1A9 and UGT2B7 were the major
isoforms responsible for IPO glucuronidation. (R)-IPO-glucuronide diastereoselectivity
was apparent with each rUGT, except UGT2B15 and UGT2B17, which favored the
formation of (S)-IPO-glucuronide. Incubations with IPO and the UGT1A9 specific
chemical inhibitor niflumic acid significantly decreased glucuronidation in human kidney,
but only marginally in human liver microsomes, consistent with known tissue expression
patterns of UGTs. We conclude that IPO glucuronidation in human kidney is mediated
by UGT1A9 and UGT2B7, and in human liver, primarily by UGT2B7 and to a lesser
extent UGT1A9 and UGT2B15. Overall, the lack of pronounced stereoselectivity for
3

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
IPO’s bioactivation in CYP4B1-transfected HepG2 cells, or for hepatic glucuronidation,
suggests the racemate is an appropriate choice for use in suicide gene therapies.
Introduction
4-Ipomeanol (IPO) is a pulmonary pro-toxin that is produced biogenically in
sweet potatoes infected with the common mold Fusarium solani. Consumption of moldy
sweet potatoes by livestock results in selective pulmonary toxicity (mainly interstitial
pneumonia and edema), which is attributed to IPO metabolic bioactivation by
extrahepatic cytochrome P450 (Wilson and Burka, 1979; Devereux et al., 1982;
Verschoyle et al., 1993). IPO is initially oxidized on the furan ring and subsequently
undergoes a sigmatropic ring rearrangement to form an electrophilic enedial that
covalently binds to nucleophilic macromolecules in the lung (Baer et al., 2005). Other
species including rabbits, rats, guinea pigs, hamsters, and dogs also exhibit pulmonary
toxicity after IPO administration (Dutcher and Boyd, 1979; Smith et al., 1987; Gram,
1997).
Due to the selective lung toxicity observed in animal species, metabolic activation
of IPO was investigated over 30 years ago in both human non-small cell carcinoma
(NCI-H322 and NCI-H358) and small cell carcinoma cell lines (NCI-H128 and NCI-H69)
(Falzon et al., 1986). Bioactivation only occurred in the non-small cell carcinoma
cultures, which prompted studies in an orthotopic xenograft mouse model of non-small
cell lung cancer (McLemore et al., 1988). In addition to providing evidence that IPO
was able to inhibit the growth of tumors in the xenograft model, biopsied primary human
lung carcinomas were shown to bioactivate IPO. However, during phase 1 clinical trials
involving patients with non-small cell lung cancer, (Rowinsky et al., 1993; Kasturi et al.,
4

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
1998) no significant reduction in pulmonary tumors was observed and reversible
hepatotoxicity was dose limiting. In retrospect, it seems likely that IPO was ineffective in
this clinical situation because humans express an unstable pulmonary cytochrome P450
4B1 enzyme (CYP4B1), which is responsible for the initial oxidation of IPO on its furan
ring (Wiek et al., 2015).
In the late 1990’s, interest in IPO as a potential anticancer therapeutic resurfaced
because of developments in pharmacological gene therapy. This approach to cancer
treatment is based on the transfection of genes for bioactivating enzymes into tumor
cells, rendering them selectively susceptible to prodrugs that would otherwise be
ineffective. Studies performed by Rainov and co-workers, (Rainov et al., 1998b)
provided the first evidence that transfection of rabbit CYP4B1 into human and rat glioma
cells caused them to be highly sensitive to IPO treatment. Several other groups have
also shown IPO effectiveness following rabbit CYP4B1 transfection in vitro (Smith et al.,
1995; Mohr et al., 2000; Steffens et al., 2000; Frank et al., 2002; Hsu et al., 2003; Jang
et al., 2010; Moon et al., 2013). Additionally, tumors from CYP4B1-expressing glioma
cells implanted into nude mice exhibited abrogated growth following intraperitoneal
injection of IPO (200 µg/day for 7 days) (Rainov et al., 1998a). The most recent work
on IPO in gene-directed enzyme prodrug therapy (GDEPT) concerns re-engineering of
the ‘native’ human CYP4B1 protein for maximal IPO bioactivation activity (Wiek et al.,
2015; Roellecke et al., 2016). Briefly, the catalytically optimized human CYP4B1
enzyme was engineered by substituting a proline residue for the native serine residue in
the meandering region at position 427. Additionally, the engineered human CYP4B1
has 12 single amino acid substitutions which are coincident with human CYP4 enzymes
5

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
and ultimately aid in protein stability. When introduced into human liver cells and
primary T-cells, the catalytically optimized human CYP4B1 was as active as the native
rabbit CYP4B1 enzyme for inducing IPO-dependent cell death.
IPO is a chiral molecule and with the aforementioned new developments in
GDEPT, it is important to understand how stereochemistry impacts IPO bioactivation
and metabolism. IPO has been shown to undergo extensive glucuronidation in the rat
(Statham et al., 1982) and previous work from our laboratory describes IPO
glucuronidation in human liver as a potential mechanism of clearance, albeit to much
lesser extent than in a variety of other animal species (Parkinson et al., 2016).
Therefore, here we report on the influence of stereochemistry on IPO-dependent
cytotoxicity in CYP4B1-transduced HepG2 cells, and on glucuronidation by recombinant
UGT enzymes and human tissue microsomes.
6

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Materials and Methods
Chemicals
IPO (97% purity by 1H NMR and HPLC-UV analysis) was a gift from the National
Cancer Institute (Bethesda, MD). β-glucuronidase from Helix pomatia, 4-
methylumbelliferone-β-D-glucuronide hydrate (4-MUG), alamethicin, ethyl-β-oxo-3-
furan-propionate, (R)-(+)-propylene oxide, (S)-(+)-propylene oxide, uridine 5’-
diphosphoglucuronic acid trisodium salt (UDPGA), 21 wt% sodium ethoxide in EtOH,
Nα-Acetyl-L-lysine (NAL), N-Acetyl-L-cysteine (NAC), catalase, superoxide dismutase,
atazanavir and niflumic acid were purchased from Sigma Aldrich (St. Louis, MO).
Hecogenin and fluconazole were purchased from Santa Cruz Biotechnology, Inc.
(Dallas, TX). 1,2-Didodecanoyl-sn-glycero-3-phosphocholine (DLPC) was purchased
from Avanti Polar Lipids Inc. (Alabaster, Alabama). Furafylline was a gift from Dr. Kent
Kunze (University of Washington, Seattle). Dulbecco's Modified Eagle Medium (DMEM),
Fetal Bovine Serum (FBS), Penicillin-Streptomycin (Pen/Strep), and HEPES were
purchased from Thermo Fisher Scientific (Waltham, MA). All solvents utilized for
chemical reactions were reagent-grade and liquid chromatography solvents were
HPLC-grade.
NMR Spectroscopy and High Resolution Mass Spectrometry
1H and 13C NMR spectra were obtained on an Agilent DD2 500 MHz NMR
spectrometer (Agilent Technologies, Santa Clara, CA) at 25°C. Proton chemical shifts
(δ) are reported in ppm and referenced to the residual proton signals of acetonitrile-d3
(δH = 1.94 ppm) or chloroform-d3 (δH = 7.27 ppm). Carbon chemical shifts are
referenced to the acetonitrile-d3 methyl carbon signal (δC = 1.32 ppm) or the chloroform-
7

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
d3 signal (δC = 77.16 ppm). Coupling constants (J) are reported in Hz and proton
chemical data and multiplicities are reported as follows: s = singlet, d = doublet, t =
triplet, q = quartet, m = multiplet High resolution mass spectrometry (HRMS) was
acquired on a Thermo Fisher LTQ Orbitrap equipped with an ESI probe (Thermo
Scientific, Waltham, MA).
Chemical Synthesis of (R)- and (S)-4-Ipomeanol
The synthetic procedure for the preparation of individual IPO enantiomers was
based on the previous work described for the synthesis of racemic IPO (Alvarez-Diez
and Zheng, 2004) (Supplemental Figure 1). Briefly, ethyl-β-oxo-3-furan-propionate
(100 mg, 0.55 mmol) was dissolved in anhydrous EtOH (5 mL), followed by the addition
of sodium ethoxide (41 mg, 0.6 mmol) and either (R)-(+)-propylene oxide or (S)-(+)-
propylene oxide (670 mg, 11.5 mmol). The mixtures were stirred for 24 hours and
subsequently quenched by adjusting the pH to 7 with dilute HCl. Extraction with
diethylether followed by flash chromatography yielded (R) and (S)- furyl-γ-lactone in
67% and 71% yields, respectively. In the second step, each respective lactone was
hydrolyzed in the presence of 20M sulfuric acid at 65°C for 24 hours to yield (R)- and
(S)-IPO in 43% and 33% yield, respectively. Separate fractions of the (R) and (S)-IPO
were further purified by HPLC, utilizing Shimadzu LC-10ADVP pumps coupled to a
Shimadzu SPD-M10AVP photodiode array detector set to 254 nm (Shimadzu,
Columbia, MD). Chromatographic separation was performed with a Keystone chiral β-
OH column (150 x 2.0 mm, 5 μm) and a 1.5 mL/min isocratic elution composed of 15%
acetonitrile/85% water for a total run time of 15 minutes. (S)- and (R)-IPO eluted at 8.0
8

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
and 7.4 minutes, respectively, under these conditions. Collected fractions were
extracted with chloroform and the structures of the recovered enantiomers were
1
13
confirmed by H NMR, C NMR, and high resolution mass spectrometry, all of which
matched previously reported spectra (Boyd et al., 1972).
Effects of IPO Stereochemistry on Cytotoxicity
The cytotoxicity of (S)-, (R)-, and racemic IPO was assessed using a cell
proliferation assay. HepG2 cells were modified by lentiviral transduction to express
rabbit CYP4B1, re-engineered human CYP4B1 (‘P+12’), or a control vector as
previously described (Schmidt et al., 2015; Wiek et al., 2015; Roellecke et al., 2016).
Stably transduced cells were grown in tissue culture media (DMEM, 10% FBS, 1%
Pen/Strep, and 10 mM HEPES) and selected with puromycin treatment (2 μg/mL) over
a 7 day period. Cells were then plated into 96 wells at 10,000 cells/well, grown for 24
hours, and the CYP4B1 line dosed with (S)-, (R)-, and racemic- IPO at final
concentrations of 100, 50, 10, 5, 1, 0.5, 0.1, 0.05, and 0.01 μM (DMSO did not exceed
0.1%). The control line was dosed with racemic IPO at the same concentrations. Cells
were incubated for 48 hours and the viability was determined with the CellTiter-Glo
Luminescent Assay (Promega, Madison, WI). Vehicle-treated wells were set to 100%
viability and Lethal Dose 50% (LD50) values were calculated from dose-response curves
using nonlinear regression curve fitting with GraphPad Prism 7.00.
9

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Effects of IPO Stereochemistry on CYP4B1 Bioactivation
Metabolic incubations contained 25 pmol purified rabbit CYP4B1, (Roellecke et
al., 2016) 50 pmol cytochrome P450 Reductase, 25 pmol cytochrome b5, 2.5 μg DLPC,
100 U catalase, 50 μg superoxide dismutase, 10 mM NAL, and 10 mM NAC in 50 mM
KPi (potassium phosphate buffer), pH 7.4, in a final volume of 100 μL. (S)-, (R)-, and
racemic IPO were added to achieve a final concentration of 100 μM. The reaction was
initiated by the addition of NADPH (1 mM final concentration) and allowed to progress
for 30 minutes at 37°C. The reaction was quenched with an equal volume of acetonitrile
containing 2 μM furafylline as the internal standard and centrifuged at 16,000 x g for 5
minutes. A supernatant aliquot (5 μL) was analyzed by UPLC-MS/MS on a Waters
Acquity UPLC connected to a Xevo TQ-S instrument (Milford, MA) in ESI+ mode, with
the following settings: capillary 3.50 kV, cone 40.0 V, source offset 60.0 V, source
temperature 150°C, desolvation temperature 350°C, cone gas flow 150 L/hr, desolvation
gas flow 800 L/hr, and collision gas flow 0.15 mL/min.
Analytes were
chromatographically separated using a Waters HSS T3 column (2.1 × 100 mm, 1.8 μm),
starting with 95% mobile phase A (0.1% formic acid in H2O) and 5% mobile phase B
(0.1% formic acid in acetonitrile) at a flow rate of 0.3 ml/min. After holding for one
minute, B was increased linearly from 5 to 50% between 1 and 7 minutes, then from 50
to 100% between 7 and 7.5 minutes, and held finally at 100% between 7.5 and 9
minutes; total run time was 11 minutes following a two minute equilibration period.
Under these conditions, IPO (all species), NAC/NAL adducts (all species), and
furafylline eluted at 5.32, 4.96, 5.46 minutes, respectively, and were detected by
multiple reaction monitoring (MRM) at m/z 151.07/95.04, 482.19/353.18, and
10

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
261.30/81.00, respectively. For IPO-NAC/NAL adduct quantification, an IPO-NAC/NAL
adduct standard that we previously generated and characterized (Parkinson et al.,
2016) was utilized. A 200 μM stock solution of this standard was prepared in 50 mM
KPi, pH 7.4, and stored at -20°C. In duplicate, calibration standards were prepared
using the stock solution diluted into blank incubation matrix to obtain ten concentrations
spanning 0.024 to 12.5 μM. Metabolite analysis was performed with MassLynx V4.1 and
quantitation achieved by comparison of peak heights for the NAC/NAL adducts formed
from the IPO enantiomers to the standard curve.
Purification and Characterization of 4-Ipomeanol Glucuronide Diastereomers
A racemic mixture of (R,S)-IPO glucuronide diastereomers was synthesized as
previously described (Parkinson et al., 2016). Semi-preparative purification of (R)- and
(S)-IPO-glucuronides was accomplished utilizing a Shimadzu HPLC system equipped
with a Beckman Coulter Ultrasphere ODS column (250 x 10mm, 5μm) and UV/Vis
detector. Baseline separation of the diastereomers was achieved by an isocratic elution
(2.5 mL/min) with mobile phase consisting of 82% 10 mM aqueous formic acid and 18%
acetonitrile. Under these conditions, (R)-IPO-glucuronide and (S)-IPO-glucuronide
eluted at 7.9 and 8.7 min, respectively, and were detected at 254 nm. Each
diastereomeric peak was collected in a round bottom flask and the organic solvent was
evaporated. The aqueous components were frozen, lyophilized, and the respective
13
white powders characterized by NMR analysis (1H, C, and HSQC) in acetonitrile-d3
and by HRMS (Supplemental Figures 2 – 9).
11

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
1
(R)-IPO-glucuronide: H NMR (500 MHz, CD3CN): δ (ppm) 1.17 (d, J = 6.4 Hz, 3H),
1.73 – 1.90 (m, 2H), 2.82 – 2.98 (m, 2H), 3.08 (t, J = 8.5 Hz, 1H), 3.33 (t, J =9.0 Hz,
1H), 3.45 (t, J = 9.3 Hz, 1H) 3.78 (d, J = 9.6 Hz, 1H), 3.81 – 3.87 (m, 1H), 4.36 (d, J =
13
7.7 Hz, 1H), 6.74 (s, 1H), 7.54 (s, 1H), 8.22 (s, 1H); C NMR (125 MHz, CD3CN): δ
(ppm) 20.32, 32.04, 37.11, 72.62, 74.18 (2), 75.38, 76.91, 102.01, 109.21, 128.52,
145.52, 149.22, 170.72, 196.30; HRMS (ESI-) calcd for C15H19O9 [M-H]- 343.1024,
found: 343.1025 (error 0.44 ppm). HSQC
1
(S)-IPO-glucuronide: H NMR (500 MHz, CD3CN): δ (ppm) 1.22 (d, J = 6.2 Hz, 3H),
1.74 – 1.90 (m, 2H), 2.92 (td, J = 7.4, 6.9, 2.7 Hz, 2H), 3.13 (t, J = 8.5 Hz, 1H), 3.32 (t,
J = 9.1 Hz, 1H), 3.45 (t, J = 9.4 Hz, 1H) 3.77 (d, J = 9.8 Hz, 1H), 3.78 – 3.85 (m, 1H),
13
4.38 (d, J = 7.8 Hz, 1H), 6.75 (s, 1H), 7.55 (s, 1H), 8.23 (s, 1H); C NMR (125 MHz,
CD3CN): δ (ppm) 22.02, 31.68, 36.68, 72.51, 74.48, 75.18, 76.91, 77.06, 103.86,
109.24, 128.52, 145.64, 149.23, 170.72, 196.30; HRMS (ESI-) calcd for C15H19O9 [M-
H]- 343.1023, found: 343.1024 (error 0.17ppm).
UPLC-MS/MS Analysis
The chromatographic separation and quantification of IPO glucuronide
diastereomers was achieved using an Agilent 1290 Infinity liquid chromatograph (Agilent
Technologies, Santa Clara, CA) coupled to an AB Sciex ABI 4000 triple quadrupole
mass spectrometer (AB Sciex, Framingham, MA) equipped with an ESI probe or with
the aforementioned Waters Acquity UPLC and Xevo TQ-S mass spectrometer An
Acquity UPLC HSS T3 column (100 x 2.1 mm, 1.8 μm) (Waters, Milford, MA) equipped
with a pre-column C18 guard cartridge (Phenomenex, Torrence, CA) was employed for
12

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
the separation. The mobile phase consisted of (A) aqueous 10 mM formic acid and (B)
10 mM formic acid in acetonitrile at a flow rate of 0.350 mL/min. The total run time was
8.6 minutes with the following gradient elution: 5% B for 1 minute, linear gradient to
15% B between 1.0 and 1.5 minutes, held at 15% B until 3.0 minutes, linear increase to
40% B between 3.0 and 6.5 minutes, then re-equilibrated to initial conditions from 6.6-
8.6 minutes. Under these conditions, (R)- and (S)-IPO-glucuronides eluted at 4.9 and
5.1 minutes, respectively, and were detected by multiple reaction monitoring (MRM) at
m/z 345/151. 4-Methylumbelliferone glucuronide (4-MUG), which was used as the
internal standard, was detected by MRM at m/z 353/177, and eluted at 4.8 minutes. The
AB Sciex 4000 mass spectrometer was operated in positive-ion mode at 450°C with an
ion spray voltage set at 4200 V, entrance potential of 10V, and exit potential of 12 V.
The curtain gas was set at 25, ion source gas 1 and 2 at 35.
Stock solutions of each synthesized diastereomer were prepared in methanol at
a 5 mM concentration and stored at -20°C. Standard calibration stock solutions
containing each diastereomer were prepared in 100 mM Tris, pH 7.5 to achieve a
concentration range from 0.003 to 3.4 μM of (R)- and (S)-IPO-glucuronide. In duplicate,
calibration standards were prepared by spiking 30 μL of the calibration stock solutions
into 170 μL of a blank incubation matrix to achieve standard concentrations of 0.51 –
510 nM. Additionally, low (4.5 nM), medium (75 nM), and high (450 nM) quality control
samples were prepared and run with each standard curve. The limit of quantitation was
0.51 nM, the lowest point on the standard curve.
13

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Hydrolysis of Conjugates with β-glucuronidase
A 1 μM solution containing both (R)- and (S)-IPO-glucuronide and a 1 μM
solution of 4-MUG in sodium acetate, pH 5.0 (100 μL) were each incubated in the
presence of 1000 units/mL of β-glucuronidase from Helix pomatia at 37°C for 24 hr. The
reactions were quenched by the addition of 10 μL 15% ZnSO4 (aq) and the sample was
centrifuged at 15,000 x g for 5 minutes. The supernatant was subsequently analyzed
by the UPLC-MS/MS methods described above.
Recombinant UGT Screen
Thirteen recombinantly expressed UGT isoforms in baculovirus infected insect
cells (UGT1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B10, 2B15, and
2B17) were purchased from Corning Life Sciences (Tewksbury, MA). Incubations
consisted of recombinant UGT enzyme (0.25 mg total protein/mL), 100 mM Tris, pH 7.5,
IPO (100 μM), MgCl2 (5 mM), alamethicin (10 μg/mL), and UDPGA (5 mM) in a 200 μL
volume. Incubations were performed at 37°C for 60 min and the reactions were
quenched by the addition of 20 μL 15% ZnSO4 (aq.), centrifuged at 15,000 x g, and
injected onto the UPLC-MS/MS for analysis.
Microsomal UGT Activity
Human liver microsomes were prepared from 8 adult donors of mixed gender by
differential centrifugation according to previously published protocols (Lin et al., 2002).
Human kidney microsomes (Lot: 1310175, pool of 12, mixed gender) were purchased
from Xenotech (Kansas City, MO). Microsomal incubations contained human liver or
14

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
kidney microsomal protein (0.25 mg total protein/mL), 100 mM Tris, pH 7.5, IPO (100
μM), MgCl2 (5 mM), alamethicin (10 μg/mL) or CHAPS (0.5 mg/mL), UDPGA (5 mM),
and one of the following: atazanavir (20 μM), hecogenin (100 μM), niflumic acid (2.5
μM), or fluconazole (10 mM). All incubations contained 1% DMSO. Incubation mixtures
were stored on ice for 15 minutes, pre-incubated in a 37°C water bath for 2 minutes
followed by the addition of UDPGA (5 mM). Reactions were allowed to proceed for 60
minutes, at which time 20 μL 15% ZnSO4 (aq.) was added to quench the reactions. The
samples were centrifuged at 15,000 x g, and 20 μL of the supernatant was injected onto
the UPLC-MS/MS for analysis.
15

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Results
Effect of IPO Stereochemistry on CYP4B1-Dependent Cytotoxicity
The cytotoxicity of (S)-, (R)-, and racemic IPO was assessed using a HepG2 cell
proliferation (endpoint) assay, 48-hours after dosing (Figure 1A). No toxicity was
observed in vehicle treated wells as compared to wells with media (data not shown).
Control vector HepG2 cells showed no evidence of cytotoxicity from racemic IPO
treatment up to 100 μM, indicating the absence or low expression of any endogenous
IPO-bioactivating enzymes that could cause toxicity. Rabbit CYP4B1-expressing HepG2
cells exposed to (S)-, (R)-, and racemic IPO displayed nearly overlapping dose-
cytotoxicity response curves, which generated LD50 values of 5.8, 5.7, and 5.7 μM,
respectively. Almost identical results were obtained with the optimized human CYP4B1
‘P+12’ enzyme, with observed LD50 values of 5.3, 6.1, and 5.4 μM for (S)-, (R)-, and
racemic IPO respectively (Figure 1B). In the CYP4B1 expressing HepG2 cells, >99%
cell death was observed with all compounds at a dose of 50 μM.
Effect of IPO Stereochemistry on CYP4B1-Dependent Substrate Depletion and
NAC/NAL Adduct Formation
To determine the effect of IPO enantiomers on CYP4B1-mediated substrate
consumption and reactive intermediate formation, metabolism of the substrates in
reconstituted systems of purified rabbit CYP4B1 was investigated. IPO depletion, in 30
minute incubations with 25 pmol P450, showed no enantiomer preference (Figure 2A),
with averages of 18.8, 18.1, and 18.1% consumed substrate for (S)-, (R)-, and racemic
IPO respectively. In probing IPO furan oxidation by trapping the reactive intermediate
16

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
with NAC/NAL, no enantiomer preference was observed (Figure 2B). Quantification
yielded average adduct formation rates of 28.3, 28.4, and 32.2 pmol/pmol P450/30
minutes for (S)-, (R)-, and racemic IPO, respectively, thus demonstrating no
stereochemical differences in adduct formation by CYP4B1. Additionally, no IPO
enantioselectivity was observed in the reconstituted systems with regards to CYP4B1-
dependent substrate depletion or NAC/NAL adduct formation.
Identification and characterization of (R) and (S)-IPO glucuronide diastereomers
(R,S)-IPO-glucuronide diastereomers were synthesized as previously described
(Parkinson et al., 2016) and separated with baseline resolution using an UPLC-MS/MS
method (Figure 3A). The detection of 4-MUG (Figure 3B) was also achieved using the
same analytical method. A 1 µM stock solution of the glucuronides and 4-MUG were
each incubated in the presence of β-glucuronidase from Helix pomatia. In each case, all
glucuronides were successfully cleaved with >99% efficiency (Figures 3C and 3D). To
confirm the absolute configuration of each glucuronide diastereomer, (R)- and (S)-IPO
were initially synthesized by a modified procedure described previously (Alvarez-Diez
and Zheng, 2004). Each enantiomer was subsequently incubated with HLMs to form
their respective glucuronide (Figure 4B and 4C). When compared to an incubation with
racemic IPO (Figure 4A), (R)-IPO-glucuronide was identified as the first peak (4.86
min) and (S)-IPO-glucuronide as the second eluting peak (5.18 min). To further confirm
the identity of the glucuronide(s), 100 µM IPO was incubated with human liver
microsomes in the absence of UDGPA (Figure 4D).
17

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Optimization of microsomal incubation conditions: effects of buffer, pH,
alamethicin, and CHAPS on specific activity and diastereoselectivity of IPO
glucuronidation
Previously, we compared microsomal glucuronidation of racemic IPO across a
number of species including humans, where formation rates were low relative to
experimental animals (Parkinson et al., 2016). Therefore, for the current studies we
employed new lots of commercially available liver and kidney microsomes, used higher
microsomal protein and substrate concentrations, and performed glucuronide analysis
on more sensitive MS instruments. With these modifications in place we first evaluated
the specific activity and diastereoselectivity of glucuronidation in human liver
microsomal incubations containing either 100 mM KPi (pH 6.0 – 8.0) or 100 mM Tris
(pH 7.0 – 8.5 at 37°C), IPO (100 μM), 5 mM MgCl2, alamethicin (10 μg/mL), and
UDPGA (5 mM) (Figure 5). The MgCl2, alamethicin, and UDPGA concentrations were
mirrored based on the previous work conducted by Walksy et al (Walsky et al., 2012).
At pH 7.5, there was no dramatic difference in specific activity using either KPi or Tris,
although the diastereomeric ratio tended to favor (R)-IPO-glucuronide more using Tris
buffer. Specific activity increased as the pH increased from 6.0 – 8.0 in KPi, but with
little diastereoselectivity at the higher pHs. Additionally, we assessed the effect of either
the detergent CHAPS (Lett et al., 1992) or pore forming reagent alamethicin on IPO
glucuronidation in human liver microsomes (data not shown). These incubations
contained 100 mM Tris pH 7.5 at 37 °C, IPO (100 μM), 5 mM MgCl2, alamethicin (10
μg/mL) or CHAPS (0.5 mg/mL), and UDPGA (5 mM). Under these conditions there was
a pronounced difference between alamethicin and CHAPS. While diastereoselectivity
18

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
was retained in each case, the specific activity was dramatically increased when
alamethicin was used. Based on these optimization efforts, all further metabolic
incubations were carried out in Tris pH 7.5 buffer and with the inclusion of alamethicin.
Recombinant UGT Screen
To identify the specific isoform(s) responsible for IPO glucuronidation, a
commercially available panel of 13 UGT recombinant enzymes was investigated.
Screening with 0.25 mg/mL of each rUGT and substrate (100 μM) indicated that
UGT1A9, UGT2B7 and UGT2B15 were the predominant isoforms catalyzing
glucuronidation of IPO (Figure 6). UGT1A9 and UGT2B7 formed (R)-IPO-glucuronide
preferentially, whereas UGT2B15 was selective for formation of (S)-IPO-glucuronide.
The specific activities for formation of (R)- and (S)-IPO-glucuronide and the respective
diastereomeric ratios in the recombinant UGT enzymes and microsomes are provided in
Table 1.
Human Liver and Kidney Microsomal Incubations
To better evaluate which UGT isoforms dominate microsomal glucuronidation,
racemic IPO (100 μM) was incubated with human liver microsomes in the presence of
selective UGT inhibitors: 10 mM atazanavir (UGT1A1) (Zhang et al., 2005), 100 µM
hecogenin (1A4) (Uchaipichat et al., 2006a), 2.5 µM niflumic Acid (1A9) (Miners et al.,
2011), and 10 mM fluconazole (2B7) (Uchaipichat et al., 2006b) (Fang et al.,
2013)(Table 2). Glucuronidation was inhibited by fluconazole (60-65%), and marginally
inhibited by niflumic acid (16 - 28%), hecogenin (20 - 22%), and atazanavir (26%
inhibition of (R)-IPO-glucuronide). These data suggest that UGT2B7 dominates human
19

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
liver microsomal glucuronidation of IPO, although we cannot preclude contributions from
other UGT enzymes, such as UGT2B4, which is a major UGT isoform in human liver
microsomes. Because UGT2B7 and UGT2B15 exhibit opposite diastereoselectivities,
the net effect is that human liver microsomes display little diastereoselectivity for IPO
glucuronidation. In contrast, in human kidney microsomes, niflumic acid was a strong
inhibitor (68 - 76%) of human kidney microsomal glucuronidation, suggesting a
dominant role for UGT1A9 in this tissue. Renal glucuronidation of IPO was also
similarly inhibited by fluconaolze (36 - 58%) compared with hepatic glucuronidation,
providing evidence for a contribution by UGT2B7. (Table 2).
Discussion
IPO occupies an important position in the history of extrahepatic P450-dependent
toxicology (Gram, 1997). However, despite almost 50 years of research on this topic,
and interest in the compound for GDEPT, no prior data exist on the influence of
stereochemistry on the protoxin’s bioactivation or its inactivation via glucuronidation.
Given that the location of the chiral center in IPO is at C-4, one could hypothesize that
glucuronidation of the IPO enantiomers would be impacted more than bioactivation,
which occurs on the furan ring that is distant from C-4. In fact, for CYP4B1, regardless
of whether the source of the enzyme is the native rabbit sequence or the optimized
human (P+12) sequence, the cytotoxic effects of IPO enantiomers in transfected HepG2
cells is the same. Detailed analysis using purified reconstituted rabbit CYP4B1 further
confirmed that depletion of (R)-IPO and (S)-IPO in purified enzyme incubations as well
as the flux to the reactive ene-dial intermediate was enantiomer independent. In the
20

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
aggregate, it is clear that the IPO racemate is an appropriate choice for future work on
the development of a CYP4B1-dependent suicide gene system.
Enantioselective glucuronidation of drugs has been recognized since the early
1980s when the benzodiazepine, oxazepam, was shown to be preferentially
glucuronidated to S-oxazepam in humans (Seideman et al., 1981). Subsequently, many
additional drugs and hydroxylated metabolites were shown to undergo stereoselective
glucuronidation (Miners and Mackenzie, 1991) that reflected preferential metabolism by
certain UGT isoforms (Court, 2005; Rowland et al., 2013). To pursue questions related
to the stereochemistry of IPO glucuronidation we took the approach of quantifying the
diastereoselectivity of the enzymatic reactions. This was accessible once
chromatographic conditions were identified that resolved the disatereomers.
Independent chemical synthesis of the diastereomers and microsomal glucuronidation
reactions with the individual IPO enantiomers enabled unambiguous assignment of
product diastereoselectivity. Optimization of buffer composition, pH and inclusion of
alamethicin in metabolic reactions, together with other assay improvements facilitated
analyte detection and provided a robust method for determining the stereochemistry of
IPO formation and evaluating the specific isoforms that likely contribute to this potential
clearance pathway. The latter is important because if IPO glucuronidation is important
in humans in vivo, as it appears to be in rodents, (Statham et al., 1982) genetic
variability in UGT activity needs to be evaluated in terms of the risk for hepatotoxicity in
humans treated with IPO.
Peak plasma concentrations of IPO from previous clinical studies were about 100
µM (Rowinsky et al., 1993), and this concentration (total Cmax) was utilized in all in vitro
21

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
experiments described. Performing the rUGT, HLM, and HKM incubations at the
clinically reported IPO Cmax is optimal for generating data that has the likeliest chance of
translating to the clinic. Therefore, it was decided to only conduct all in vitro
experiments with the most physiologically relevant IPO concentration. Screening of a
panel of thirteen recombinantly expressed UGTs identified UGT1A9, UGT2B7, and UGT
2B15 as the dominant isoforms when incubations were performed with IPO. Notably,
however, incubations with human liver microsomes in the presence of selective UGT
inhibitors suggested that UGT1A9 and UGT2B7 are the more likely contributors in vivo.
Despite the high catalytic activity of recombinant UGT1A9, this isoform may not be a
prominent contributor to human liver IPO glucuronidation because it is a minor
constituent (6-9%) of the total UGT pool (Margaillan et al., 2015). Additionally, recent
evidence suggests the expression levels of UGT1A9 in HKMs in 3.1 fold higher that
HLMs and may account for the marginal inhibition by niflumic acid in HLMs (Nakamura
et al., 2016). Moreover, UGT1A9 exhibits pronounced selectivity for glucuronidation of
(R)-IPO over (S)-IPO, but this is not recapitulated in the diastereoselectivity found in
human liver microsomes. This observation further undermines a major contribution of
UGT1A9 to IPO glucuronidation in human liver. In contrast, UGT1A9 is clearly the
major contributor to human kidney glucuronidation, on the basis of both
diastereoselectivity and chemical inhibition studies. Human renal glucuronidation has
been reasonably well characterized, (Knights et al., 2013) and is important for drugs
such as propofol and mycophenolic acid (Gill et al., 2012; Knights et al., 2016).
Although polymorphisms are present in all UGT enzymes, high frequency deleterious
variants that impact the clearance of therapeutic drugs are limited largely to UGT1A1,
22

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
UGT2B15, UGT2B17, and to a lesser extent UGT1A9 (Stingl et al., 2014). While
UGT1A1 and UGT2B17 appear to have minimal roles in the glucuronidation of IPO,
UGT2B15 is probably a contributor based of the rUGT phenotypic screen. Since the
common mutation, D85Y, in UGT2B15 is a major determinant of oxazepam clearance in
vivo, (He et al., 2009) it would be interesting to determine if this mutation is a risk factor
for IPO-induced liver toxicity in humans. The opposite holds true for UGT1A9
polymorphisms, which are attributed to increased protein expression levels and
subsequently higher clearances of UGT1A9 substrates, such as mycophenolic acid
(Baldelli et al., 2007).
In summary, we have demonstrated that IPO stereochemistry does not affect
CYP4B1-mediated bioactivation of this pro-toxin, but does influence product
diasteroeselectivity generated from human liver and kidney UGTs. The major enzymes
capable of racemic IPO glucuronidation are UGT1A9, UGT2B7, and UGT2B15. Major
isoform-dependent differences in product diasteroeselectivity together with chemical
inhibition experiments help identify UGT1A9 and members of the UGT2B family as the
likely contributors to human hepatic glucuronidation of IPO. Therefore, beyond common
genetic polymorphisms in UGT2B15, we speculate that drug-drug interactions with
concomitantly administered UGT2B substrates might affect the overall clearance of
racemic IPO in pharmacological gene therapy. Future work in this area should also
focus on delineating the relative importance of other phase II detoxifying enzymes to
IPO metabolism in humans.
Acknowledgements
The authors would like to thank J. Scott Edgar for his assistance with high-resolution
23

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
mass spectrometry.
Authorship contributions
Participated in research design: Teitelbaum, McDonald, Kowalski, Parkinson,
Hanenberg, Rettie
Conducted experiments: Teitelbaum, McDonald Kowalski, Whittington, Roellecke,
Wiek, Scian
Performed data analysis: Teitelbaum, McDonald, Kowalski, Scian
Wrote or contributed to the writing of the manuscript: Teitelbaum, Kowalski, Rettie
24

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
References
Alvarez-Diez TM and Zheng J (2004) Mechanism-based inactivation of cytochrome P450 3A4
by 4-ipomeanol. Chem Res Toxicol 17:150-157.
Baer BR, Rettie AE and Henne KR (2005) Bioactivation of 4-ipomeanol by CYP4B1: adduct
characterization and evidence for an enedial intermediate. Chem Res Toxicol 18:855-864.
Baldelli S, Merlini S, Perico N, Nicastri A, Cortinovis M, Gotti E, Remuzzi G and Cattaneo D
(2007) C-440T/T-331C polymorphisms in the UGT1A9 gene affect the pharmacokinetics
of mycophenolic acid in kidney transplantation. Pharmacogenomics 8:1127-1141.
Boyd MR, Wilson BJ and Harris TM (1972) Confirmation by chemical synthesis of the structure
of 4-ipomeanol, a lung-toxic metabolite of the sweet potato, Ipomoea batatas. Nat New
Biol 236:158-159.
Court MH (2005) Isoform-selective probe substrates for in vitro studies of human UDP-
glucuronosyltransferases. Methods Enzymol 400:104-116.
Devereux TR, Jones KG, Bend JR, Fouts JR, Statham CN and Boyd MR (1982) In vitro
metabolic activation of the pulmonary toxin, 4-ipomeanol, in nonciliated bronchiolar
epithelial (Clara) and alveolar type II cells isolated from rabbit lung. J Pharmacol Exp
Ther 220:223-227.
Dutcher JS and Boyd MR (1979) Species and strain differences in target organ alkylation and
toxicity by 4-ipomeanol. Predictive value of covalent binding in studies of target organ
toxicities by reactive metabolites. Biochem Pharmacol 28:3367-3372.
Falzon M, McMahon JB, Schuller HM and Boyd MR (1986) Metabolic activation and
cytotoxicity of 4-ipomeanol in human non-small cell lung cancer lines. Cancer Res
46:3484-3489.
Fang ZZ, He RR, Cao YF, Tanaka N, Jiang C, Krausz KW, Qi Y, Dong PP, Ai CZ, Sun XY, Hong
M, Ge GB, Gonzalez FJ, Ma XC and Sun HZ (2013) A model of in vitro UDP-
glucuronosyltransferase inhibition by bile acids predicts possible metabolic disorders. J
Lipid Res 54:3334-3344.
Frank S, Steffens S, Fischer U, Tlolko A, Rainov NG and Kramm CM (2002) Differential
cytotoxicity and bystander effect of the rabbit cytochrome P450 4B1 enzyme gene by two
different prodrugs: implications for pharmacogene therapy. Cancer Gene Ther 9:178-188.
Gill KL, Houston JB and Galetin A (2012) Characterization of in vitro glucuronidation clearance
of a range of drugs in human kidney microsomes: comparison with liver and intestinal
glucuronidation and impact of albumin. Drug Metab Dispos 40:825-835.
Gram TE (1997) Chemically reactive intermediates and pulmonary xenobiotic toxicity.
Pharmacol Rev 49:297-341.
He X, Hesse LM, Hazarika S, Masse G, Harmatz JS, Greenblatt DJ and Court MH (2009)
Evidence for oxazepam as an in vivo probe of UGT2B15: oxazepam clearance is reduced
by UGT2B15 D85Y polymorphism but unaffected by UGT2B17 deletion. Br J Clin
Pharmacol 68:721-730.
Hsu H, Rainov NG, Quinones A, Eling DJ, Sakamoto KM and Spear MA (2003) Combined
radiation and cytochrome CYP4B1/4-ipomeanol gene therapy using the EGR1 promoter.
Anticancer Res 23:2723-2728.
25

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Jang SJ, Kang JH, Lee TS, Kim SJ, Kim KI, Lee YJ, Cheon GJ, Choi CW and Lim SM (2010)
Prodrug-activating Gene Therapy with Rabbit Cytochrome P450 4B1/4-Ipomeanol or 2-
Aminoanthracene System in Glioma Cells. Nucl Med Mol Imaging 44:193-198.
Kasturi VK, Dearing MP, Piscitelli SC, Russell EK, Sladek GG, O'Neil K, Turner GA, Morton
TL, Christian MC, Johnson BE and Kelley MJ (1998) Phase I study of a five-day dose
schedule of 4-Ipomeanol in patients with non-small cell lung cancer. Clin Cancer Res
4:2095-2102.
Knights KM, Rowland A and Miners JO (2013) Renal drug metabolism in humans: the potential
for drug-endobiotic interactions involving cytochrome P450 (CYP) and UDP-
glucuronosyltransferase (UGT). Br J Clin Pharmacol 76:587-602.
Knights KM, Spencer SM, Fallon JK, Chau N, Smith PC and Miners JO (2016) Scaling factors
for the in vitro-in vivo extrapolation (IV-IVE) of renal drug and xenobiotic
glucuronidation clearance. Br J Clin Pharmacol 81:1153-1164.
Lett E, Kriszt W, de Sandro V, Ducrotoy G and Richert L (1992) Optimal detergent activation of
rat liver microsomal UDP-glucuronosyl transferases toward morphine and 1-naphthol:
contribution to induction and latency studies. Biochem Pharmacol 43:1649-1653.
Lin YS, Dowling AL, Quigley SD, Farin FM, Zhang J, Lamba J, Schuetz EG and Thummel KE
(2002) Co-regulation of CYP3A4 and CYP3A5 and contribution to hepatic and intestinal
midazolam metabolism. Mol Pharmacol 62:162-172.
Margaillan G, Rouleau M, Klein K, Fallon JK, Caron P, Villeneuve L, Smith PC, Zanger UM and
Guillemette C (2015) Multiplexed Targeted Quantitative Proteomics Predicts Hepatic
Glucuronidation Potential. Drug Metab Dispos 43:1331-1335.
McLemore TL, Coudert BP and Adelberg S (1988) Metabolic activation of 4-Ipomeanol by
human pulmonary carcinoma cells propagated in vitro and intrabronchially in nude mice.
Clin Res 36:498A.
Miners JO, Bowalgaha K, Elliot DJ, Baranczewski P and Knights KM (2011) Characterization of
niflumic acid as a selective inhibitor of human liver microsomal UDP-
glucuronosyltransferase 1A9: application to the reaction phenotyping of acetaminophen
glucuronidation. Drug Metab Dispos 39:644-652.
Miners JO and Mackenzie PI (1991) Drug glucuronidation in humans. Pharmacol Ther 51:347-
369.
Mohr L, Rainov NG, Mohr UG and Wands JR (2000) Rabbit cytochrome P450 4B1: A novel
prodrug activating gene for pharmacogene therapy of hepatocellular carcinoma. Cancer
Gene Ther 7:1008-1014.
Moon BS, Jang SJ, Kim SJ, Lee TS, Chi DY, Lee BC, Kang JH and Kim SE (2013) Synthesis
and evaluation of a 18F-labeled 4-ipomeanol as an imaging agent for CYP4B1 gene
prodrug activation therapy. Cancer Biother Radiopharm 28:588-597.
Nakamura K, Hirayama-Kurogi M, Ito S, Kuno T, Yoneyama T, Obuchi W, Terasaki T and
Ohtsuki S (2016) Large-scale multiplex absolute protein quantification of drug-
metabolizing enzymes and transporters in human intestine, liver, and kidney microsomes
by SWATH-MS: Comparison with MRM/SRM and HR-MRM/PRM. Proteomics
16:2106-2117.
Parkinson OT, Teitelbaum AM, Whittington D, Kelly EJ and Rettie AE (2016) Species
Differences in Microsomal Oxidation and Glucuronidation of 4-Ipomeanol: Relationship
to Target Organ Toxicity. Drug Metab Dispos.
26

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Rainov NG, Dobberstein KU, Sena-Esteves M, Herrlinger U, Kramm CM, Philpot RM, Hilton J,
Chiocca EA and Breakefield XO (1998a) New prodrug activation gene therapy for cancer
using cytochrome P450 4B1 and 2-aminoanthracene/4-ipomeanol. Hum Gene Ther
9:1261-1273.
Rainov NG, Sena-Esteves M, Fraefel C, Dobberstein KU, Chiocca EA and Breakefield XO
(1998b) A chimeric fusion protein of cytochrome CYP4B1 and green fluorescent protein
for detection of pro-drug activating gene delivery and for gene therapy in malignant
glioma. Adv Exp Med Biol 451:393-403.
Roellecke K, Virts EL, Einholz R, Edson KZ, Altvater B, Rossig C, von Laer D, Scheckenbach
K, Wagenmann M, Reinhardt D, Kramm CM, Rettie AE, Wiek C and Hanenberg H
(2016) Optimized human CYP4B1 in combination with the alkylator prodrug 4-
ipomeanol serves as a novel suicide gene system for adoptive T-cell therapies. Gene Ther
23:615-626.
Rowinsky EK, Noe DA, Ettinger DS, Christian MC, Lubejko BG, Fishman EK, Sartorius SE,
Boyd MR and Donehower RC (1993) Phase I and pharmacological study of the
pulmonary cytotoxin 4-ipomeanol on a single dose schedule in lung cancer patients:
hepatotoxicity is dose limiting in humans. Cancer Res 53:1794-1801.
Rowland A, Miners JO and Mackenzie PI (2013) The UDP-glucuronosyltransferases: their role
in drug metabolism and detoxification. Int J Biochem Cell Biol 45:1121-1132.
Schmidt EM, Wiek C, Parkinson OT, Roellecke K, Freund M, Gombert M, Lottmann N, Steward
CA, Kramm CM, Yarov-Yarovoy V, Rettie AE and Hanenberg H (2015) Characterization
of an Additional Splice Acceptor Site Introduced into CYP4B1 in Hominoidae during
Evolution. PLoS One 10:e0137110.
Seideman P, Ericsson O, Groningsson K and von Bahr C (1981) Effect of pentobarbital on the
formation of diastereomeric oxazepam glucuronides in man: analysis by high
performance liquid chromatography. Acta Pharmacol Toxicol (Copenh) 49:200-204.
Smith AC, Barrett D, Stedham MA, el-Hawari M, Kastello MD, Grieshaber CK and Boyd MR
(1987) Preclinical toxicology studies of 4-ipomeanol: a novel candidate for clinical
evaluation in lung cancer. Cancer Treat Rep 71:1157-1164.
Smith PB, Tiano HF, Nesnow S, Boyd MR, Philpot RM and Langenbach R (1995) 4-Ipomeanol
and 2-aminoanthracene cytotoxicity in C3H/10T1/2 cells expressing rabbit cytochrome
P450 4B1. Biochem Pharmacol 50:1567-1575.
Statham CN, Dutcher JS, Kim SH and Boyd MR (1982) Ipomeanol 4-glucuronide, a major
urinary metabolite of 4-ipomeanol in the rat. Drug Metab Dispos 10:264-267.
Steffens S, Frank S, Fischer U, Heuser C, Meyer KL, Dobberstein KU, Rainov NG and Kramm
CM (2000) Enhanced green fluorescent protein fusion proteins of herpes simplex virus
type 1 thymidine kinase and cytochrome P450 4B1: applications for prodrug-activating
gene therapy. Cancer Gene Ther 7:806-812.
Stingl JC, Bartels H, Viviani R, Lehmann ML and Brockmoller J (2014) Relevance of UDP-
glucuronosyltransferase polymorphisms for drug dosing: A quantitative systematic
review. Pharmacol Ther 141:92-116.
Uchaipichat V, Mackenzie PI, Elliot DJ and Miners JO (2006a) Selectivity of substrate
(trifluoperazine) and inhibitor (amitriptyline, androsterone, canrenoic acid, hecogenin,
phenylbutazone, quinidine, quinine, and sulfinpyrazone) "probes" for human udp-
glucuronosyltransferases. Drug Metab Dispos 34:449-456.
27

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Uchaipichat V, Winner LK, Mackenzie PI, Elliot DJ, Williams JA and Miners JO (2006b)
Quantitative prediction of in vivo inhibitory interactions involving glucuronidated drugs
from in vitro data: the effect of fluconazole on zidovudine glucuronidation. Br J Clin
Pharmacol 61:427-439.
Verschoyle RD, Philpot RM, Wolf CR and Dinsdale D (1993) CYP4B1 activates 4-ipomeanol in
rat lung. Toxicol Appl Pharmacol 123:193-198.
Walsky RL, Bauman JN, Bourcier K, Giddens G, Lapham K, Negahban A, Ryder TF, Obach RS,
Hyland R and Goosen TC (2012) Optimized assays for human UDP-
glucuronosyltransferase (UGT) activities: altered alamethicin concentration and utility to
screen for UGT inhibitors. Drug Metab Dispos 40:1051-1065.
Wiek C, Schmidt EM, Roellecke K, Freund M, Nakano M, Kelly EJ, Kaisers W, Yarov-Yarovoy
V, Kramm CM, Rettie AE and Hanenberg H (2015) Identification of amino acid
determinants in CYP4B1 for optimal catalytic processing of 4-ipomeanol. Biochem J
465:103-114.
Wilson BJ and Burka LT (1979) Toxicity of novel sequiterpenoids from the stressed sweet potato
(Ipomoea batatas). Food Cosmet Toxicol 17:353-355.
Zhang D, Chando TJ, Everett DW, Patten CJ, Dehal SS and Humphreys WG (2005) In vitro
inhibition of UDP glucuronosyltransferases by atazanavir and other HIV protease
inhibitors and the relationship of this property to in vivo bilirubin glucuronidation. Drug
Metab Dispos 33:1729-1739.
28

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Footnotes
This study was supported in part by the National Institutes of Health [grant
R01GM49054] and by the University of Washington School of Pharmacy Brady Fund for
Natural Products. This work was also funded by the Strategische Forschungsverbund
of the Heinrich Heine University (to CW).
The authors declare no competing financial interest
29

JPET Fast Forward. Published on November 8, 2018 as DOI: 10.1124/jpet.118.249771
This article has not been copyedited and formatted. The final version may differ from this version.
JPET#249771
Figure Legends:
Figure 1: Cytotoxicity dose-response curves in HepG2 cells transfected with (A) rabbit
CYP4B1 and (B) human P+12 CYP4B1 from treatment with (S)-, (R)-, racemic IPO, and
control HepG2 cells treated with racemic IPO. Data shown are the mean, from triplicate
replicates, ± standard deviation.
Figure 2: Comparison of substrate depletion (A) and quantification of NAC/NAL adduct
formation (B) in reconstituted CYP4B1 incubations after 30 minutes, with 25 pmol P450
and 100 μM IPO substrates. Data shown are the mean, from triplicate replicates, ±
standard deviation.
Figure 3: Chromatographic separation of (R,S)-IPO-glucuronide (A) and 4-
methylumbelliferone glucuronide (B) by UPLC-MS/MS. Incubation of 1 µM (R,S)-IPO-
glucuronide (C) and 1 µM 4-methylumbelliferone glucuronide (D) in the presence of
Helix pomatia (1000 units/mL) in potassium acetate pH 5 for 24 hr at 37 °C.
Figure 4: Representative chromatographic traces of (R)- and (S)-IPO-glucuronides.
Incubation of 100 µM IPO in Tris buffer pH 7.5, alamethicin (10 µg/mL), MgCl2 (5 mM),
and UDPGA (5 mM) with 0.25 mg/mL human liver microsomes (A). Incubation of 100
µM (R)-IPO (B) and (S)-IPO (C) in human liver microsomes. Incubation of 100 µM
racemic IPO with human liver microsomes in the absence of UDPGA (D).
Figure 5:
Diastereoselective glucuronidation of racemic IPO in human liver
microsomes. Incubations were conducted with 100 µM IPO in potassium phosphate
buffer (pH 6.0 – 8.0) and Tris buffer (pH 7.0 – 8.5) with 0.25 mg/mL microsomal protein,
alamethicin (10 µg/mL), MgCl2 (5 mM), and UDPGA (5 mM) for 60 min at 37 °C. The
30