Pharmacokinetics, metabolism and off-target effects in the rat of 8-[(1H- benzotriazol-1-yl)amino]octanoic acid, a selective inhibitor of human cytochrome P450 4Z1: β-oxidation as a potential augmenting pathway for inhibition

XENOBIOTICA

021, VOL. 51, NO. 8, 901 –915

2

https://doi.org/10.1080/00498254.2021.1930281

RESEARCH ARTICLE

Pharmacokinetics, metabolism and off-target effects in the rat of 8-[(1H-

benzotriazol-1-yl)amino]octanoic acid, a selective inhibitor of human

cytochrome P450 4Z1: b-oxidation as a potential augmenting pathway

for inhibition

a,b

a

a,b

c

a

John P. Kowalski

, Robert D. Pelletier , Matthew G. McDonald , Edward J. Kelly and Allan E. Rettie

a

b

Department of Medicinal Chemistry, School of Pharmacy, University of Washington, Seattle, WA, USA; Department of Drug Metabolism

c

and Pharmacokinetics, Pfizer Boulder R&D, Boulder, CO, USA; Department of Pharmaceutics, School of Pharmacy, University of Washington,

Seattle, WA, USA

ABSTRACT

ARTICLE HISTORY

Received 26 February 2021

Revised 8 May 2021

Accepted 10 May 2021

1.

2.

3.

8-[(1H-1,2,3-benzotriazol-1-yl)amino]octanoic acid (8-BOA) was recently identified as a selective

and potent mechanism-based inactivator (MBI) of breast cancer-associated CYP4Z1 and exhibited

favourable inhibitory activity in vitro, thus meriting in vivo characterization.

The pharmacokinetics and metabolism of 8-BOA in rats was examined after a single IV bolus dose

of 10 mg/kg. A biphasic time-concentration profile resulted in relatively low clearance and a pro-

longed elimination half-life.

KEYWORDS

Cytochrome P450 4Z1;

mechanism-based inhibitor;

bioactivation; breast cancer;

rat; fatty acid; b-oxidation;

leaky hosepipe

The major circulating metabolites identified in plasma were products of b-oxidation; congeners

losing two and four methylene groups accounted for >50% of metabolites by peak area. The

–

(CH ) product was characterized previously as a CYP4Z1 MBI and so represents an active metab-

2 2

olite that may contribute to the desired pharmacological effect.

4

5

.

Ex vivo analysis of total CYP content in rat liver and kidney microsomes showed that off-target

CYP inactivation was minimal; liver microsomal probe substrate metabolism also demonstrated

low off-target inactivation. Standard clinical chemistries provided no indication of acute toxicity.

In silico simulations using the free concentration of 8-BOA in plasma suggested that the in vivo

dose used here may effectively inactivate CYP4Z1 in a xenografted tumour.

Introduction

pro-metastatic metabolite of arachidonic acid (McDonald et

al. 2017). While cell culture studies can be utilized to interro-

gate the role of CYP4Z1 in breast cancer, the lack of inter-

action with a tumour microenvironment limits the predictive

power of these experiments (Murayama and Gotoh 2019).

Therefore, to further the validation of CYP4Z1 as a drug tar-

get in breast cancer, a more complex cellular milieu, with its

attendant heterogeneity of signalling ligands, is desirable for

inhibitor testing.

The utility of the rat in pre-clinical rodent models of

human disease has become increasingly clear due to the

many physiological similarities between the species that are

absent when using mice. The benefits are especially pro-

nounced for breast cancer studies, as xenografted tumours

in rats better parallel those in humans in regard to oestrogen

dependency, hormone therapy sensitivity, development, duc-

During their lifetime, nearly 13% of US women will be diag-

nosed with invasive breast cancer, which is the most preva-

lent form and the second leading cause of mortality for this

group (DeSantis et al. 2019). The elucidation of new molecu-

lar drivers of this complex disease and new drugs to treat

metastases are urgently needed (Capper et al. 2016).

Cytochrome P450 4Z1 (CYP4Z1) has been identified as a

mammary tissue-restricted enzyme with an unknown physio-

logical role that is upregulated in breast cancer. Furthermore,

elevated transcript levels of CYP4Z1 are frequently associated

with increased tumour grade and aggressiveness (Rieger et

al. 2004, Radvanyi et al. 2005, Murray et al. 2010, Li et al.

2

017, Al-Esawi et al. 2020). Lastly, human CYP4Z1 is the only

currently known functional enzyme in this subfamily (Yang et

al. 2017).

Recently, we reported that 8-[(1H-1,2,3-benzotriazol-1-yl)a- tal origin, and simulation of phenotypic heterogeneity (Smits

mino]octanoic acid (8-BOA) is a selective and potent mech- et al. 2007, Iannaccone and Jacob 2009, Mollard et al. 2011).

anism-based inactivator (MBI) of CYP4Z1 that can effectively Additionally, recent advances in the generation of immuno-

inhibit this enzyme’s production of 14,15-epoxyeicosatrienoic deficient rats has furthered the use of this species in varied

acid (14,15-EET) (Kowalski et al. 2020), a pro-angiogenic and transplant settings, such as cell- or patient-derived xenograft

CONTACT John P. Kowalski

john.kowalski@pfizer.com

University of Washington Seattle Campus, Seattle 98195-0005, WA, USA

Supplemental data for this article can be accessed here .

ß 2021 Informa UK Limited, trading as Taylor & Francis Group

9

02

J. P. KOWALSKI ET AL.

1

(

CDX/PDX) models of tumour growth (Noto et al. 2018, He et S2(A) and the H NMR spectrum for 8-BOA has been pro-

al. 2019).

vided in Figure S2(B) for a comparison (reproduced from

Therefore, the aim of this pilot-scale study was to assess Kowalski et al. 2020 with the authors’ permission).

the in vivo characteristics of 8-BOA in the rat to facilitate the

design of future CDX/PDX model breast cancer studies with

this inhibitor. Specifically, our goals were to assess plasma

Depletion assays

exposure for feasibility of target enzyme inactivation, charac- To assess microsomal stability, the depletion of 8-BOA (and

terize circulating metabolites, and interrogate off-target CYP assay controls) was assessed after extended incubation in

inactivation for 8-BOA.

fortified microsomes. Reactions were set up on ice in tripli-

cate and contained 0.5 mg/mL of either RLM or RKM, in

1

00 mM potassium phosphate (KPi), pH 7.4, and 3.2 mM

Materials and methods

MgCl , in a final volume of 0.5 mL. 8-BOA was added to both

2

Reagents

RLM and RKM reactions, while 7-hydroxy coumarin and mida-

zolam were added only to RLM reactions; all substrates were

diluted 400-fold v/v from DMSO stocks, to achieve final con-

centrations of 1 lM (reactions contained 0.25% v/v DMSO).

The stocks of 8-BOA and 6-[(1H-1,2,3-benzotriazol-1-yl)ami-

no]hexanoic acid (6-BHA) used for all experiments were syn-

thesized for our previous study (Kowalski et al. 2020).

ꢀ

After a 5 min equilibration at 37 C, the reactions were

Benzotriazole-d was purchased from CDN Isotopes (Pointe-

0

4

initiated by the addition of a mixture of NADPH, uridine 5 -

Claire, Canada) and was used for inhouse synthesis of the

deuterated internal standard (see below). Pooled rat (male

Sprague Dawley) plasma and urine used as the matrix for

protein binding and standards for 8-BOA quantification was

purchased from BioIVT (Westbury, NY). Pooled rat (male

Sprague Dawley) liver and kidney microsomes (RLM, RKM),

derived from control and treated group rats for ex vivo

experiments, were generated according to standard proce-

dures for subcellular fractionation (Sadeque et al. 1992), and

quantified via BCA assay (Thermo Fisher Scientific, Waltham,

MA) according to the manufacturer’s protocol. RLM and RKM

used for in vitro experiments were purchased from Sekisui

Xenotech (Kansas City, KS). Chloroform-d was purchased

from Cambridge Isotope Laboratories, Inc (Andover, MA),

reduced b-nicotinamide adenine dinucleotide phosphate

diphosphoglucuronic acid (UDPGA), and alamethicin, a com-

monly added pore-forming reagent for in vitro investigation

0

of uridine 5 -diphospho-glucuronosyltransferase (UGT)-medi-

ated metabolism, to achieve final concentrations of 1.5 mM,

5

mM, and 10 lg/mL, respectively. Aliquots (50 lL) were

removed at times 0, 5, 15, 30, 60, and 90 min and added to

tubes containing identical volumes of ice-cold acetonitrile

containing 0.5 lM of the internal standard 8-BOA-d (for 8-

4

BOA stability studies) or 7-hydroxy coumarin-d₃(for 7-

hydroxy coumarin and midazolam stability studies) to

quench the reactions. All quenched reactions were stored at

ꢀ

ꢂ20 C until the end of the experiment. The samples were

ꢀ

centrifuged for 10 min at 10,000 x g and 4 C, and then the

supernatants were transferred to new tubes. For quantitation,

calibration curves were prepared by separately spiking

unlabelled standards for each test compound into an identi-

cal microsomal matrix spanning concentrations of

(

NADPH) was purchased from OYC Americas (Vista, CA), and

formulation reagents (DMSO, PBS, and Solutol) were pro-

vided by Alliance Pharma. All solvents (Optima grade) were

purchased from Thermo Fisher Scientific, and all other chem-

icals were purchased Sigma-Aldrich (St. Louis, MO).

3

0 nM–3 lM, in half-log dilutions, and underwent sample

workup identical to that described above. Aliquots for all

samples (5 lL) were then analysed according to the General

LC-MS/MS method for the quantitation of 8-BOA, 6BHA, and

assay controls described below.

Internal standard synthesis

2

8

-f[(4,5,6,7- H )-1H-1,2,3-benzotriazol-1-yl]aminogoctanoic

4

General LC-MS/MS method for the quantitation of

acid (8-BOA-d ) was synthesized using a modified protocol

4

8

-BOA, 6-BHA and assay controls

from our previous study and has been summarized in Figure

S1. The additional step, amination of benzotriazole-d , was

4

Aliquots were analysed by UPLC-MS/MS on a Waters Acquity

performed analogous to a previously reported method UPLC connected to a Waters Xevo TQ-S instrument in ESI^-

Campbell and Rees 1969). Accurate mass was determined mode, with the following settings: capillary 2.0 kV, source

(

ꢀ

via UPLC-MS on a Waters Acquity UPLC (Milford, MA) offset 60.0 V, source temperature 150 C, desolvation tem-

ꢀ

coupled to an AB Sciex TripleTOF 5600 mass spectrometer perature 350 C, cone gas flow 150 L/hr, desolvation gas flow

1

ꢀ

(

Framingham, MA). H NMR spectra was recorded at 25 C in 800 L/hr, collision gas flow 0.15 mL/min, cone voltage 30 V,

deuterated chloroform on a 499.73 MHz Agilent DD2 (Santa and collision energy 15 eV. Chromatographic separation of

1

Clara, CA) spectrometer. Purity as determined by H NMR analytes was achieved using a Waters BEH C₁₈column

1

spectroscopy was ꢁ95%. H NMR (500 MHz, Chloroform-d) d (2.1 ꢃ 50 mm, 1.7 lm), starting with 90% mobile phase A

ppm 3.42 (t, J ¼ 7.09 Hz, 2 H), 2.35 (t, J ¼ 7.34 Hz, 2 H), (0.1% formic acid in water) and 10% mobile phase B (0.1%

1

2

.59–1.69 (m, 2 H), 1.54 (quin, J ¼ 7.21 Hz, 2 H), 1.40–1.49 (m, formic acid in acetonitrile) at a flow rate of 0.3 mL/min. After

þ þ

H), 1.31–1.38 (m, 4 H). HRMS (ESI ) m/z [M þ H] calculated a 1 min hold, B was increased linearly from 10 to 95%

(

C H D N O ) 281.1910, observed 281.1909, d ppm 0.4.The between 1 and 3 min and held at 95% between 3.5 and

14 17 4 4 2

1

H NMR spectrum for 8-BOA-d has been provided in Figure 5.5 min; the total run time was 7 min following a 1.5 min

4

XENOBIOTICA

903

period for equilibration. The analytes and isotope-labeled between sample and reference cuvettes and CYP content

internal standards were detected by multiple reaction moni- was measured as described above.

-

toring (MRM) in electrospray negative mode (ESI ) with the

following transitions and retention times: 8-BOA m/z

Animal studies

2

3

2

75.15 > 154, 8-BOA-d₄279.16 > 154 (2.9 min), warfarin

07 > 161, warfarin-d₅312 > 161 (3.2 min), and 6-BHA Animal procedures were performed in full compliance with

47 > 118 (2.7 min). The following were detected by MRM in all applicable regulations by Alliance Pharma (Malvern, PA).

þ

ESI mode: 7-hydroxy coumarin m/z 163 > 119, 7-hydroxy Following a two day facility acclimation period, three male

coumarin-d₃166 > 122 (2.4 min), and midazolam 327 > 292 Sprague Dawley rats, weighing between 0.307–0.316 kg,

(

2.5 min). Analysis of the mass spectra was performed with received intravenous (IV) bolus administration of 8-BOA

Waters MassLynx V4.1.

(4 mg/mL in PBS, 67%; Solutol, 30%; DMSO, 3%) at 10 mg/kg

and a dose volume of 2.5 mL/kg. A control group of three

rats, weighing between 0.303–0.320 kg, received the same

formulation as above, minus 8-BOA. The study proceeded for

Plasma protein binding

In vitro analysis of plasma protein binding (PPB) was per- 24 hours and serial blood collections (tail bleeds) were taken

formed for 8-BOA, 6-BHA, and Warfarin as an assay control. at pre-dose, 5, 15, 30 min, 1, 2, 4, 8, and 24 hr, and plasma

Compounds were spiked separately into pooled male was separated. After euthanasia, necropsies were performed

Sprague Dawley rat plasma to achieve final concentrations of and liver and kidney were collected from each animal, and

3

lM. A dialysis method was followed using rapid equilibrium urine was collected from the bladders. Samples were then

dialysis devices according to the manufacturer’s protocol shipped on dry ice overnight for inhouse bioanalysis.

ThermoFisher Scientific). After the incubation period, ice- Standard clinical chemistries were performed by Phoenix Lab

cold acetonitrile, containing the internal standard 8-BOA-d₄(Mukilteo, WA) on plasma from both control and treated

for both 8-BOA and 6-BHA samples), or Warfarin-d , at groups for pre-dose and 24-hour timepoints.

(

5

0

.5 lM and 0.1% formic acid was added at a 3:1 ratio. The

ꢀ

samples were centrifuged for 10 min at 10,000 x g and 4 C,

and then the supernatants were transferred to new tubes.

Aliquots for all samples (5 lL) were then analysed according

Determination of 8-BOA and 6-BHA plasma

concentrations and pharmacokinetic analysis

to the General LC-MS/MS method for the quantitation of 8- Plasma samples were thawed on ice and 20 lL aliquots were

BOA, 6BHA, and assay controls.

transferred to new tubes. To these, 60 lL of ice-cold aceto-

nitrile, containing 5 lM 8-BOA-d₄and 0.1% formic acid,

was added. The samples were centrifuged for 10 min at

Determination of CYP content

ꢀ

1

0,000 x g and 4 C, and then the supernatants were trans-

In vitro inactivation

ferred to new tubes. A calibration curve was prepared for

Inactivation reactions were set up on ice and contained RLM quantification; 8-BOA or 6-BHA was spiked into pooled male

Sekisui Xenotech) at 1 mg/mL in 100 mM KPi, pH 7.4 and Sprague Dawley rat plasma diluted 200-fold v/v from DMSO

(

3

.2 mM MgCl , in a final volume of 1.2 mL. 8-BOA was diluted stocks to achieve concentrations ranging from 1 nM–1 mM, in

2

4

00-fold v/v from DSMO stocks to achieve final concentra- half-log dilutions (DMSO control was also prepared). The

tions of 100 and 500 lM, or an equal volume of DMSO was standards were subjected to the same sample workup as

added (reactions contained 0.25% v/v DMSO). After a 5 min described above and aliquots for all samples (5 lL) were

ꢀ

period for equilibration at 37 C, NADPH was added (1 mM then analysed according to the General LC-MS/MS method for

final concentration) to all conditions to initiate the reactions. the quantitation of 8-BOA, 6BHA, and assay controls.

Following

a

30 min incubation, 0.55 mL aliquots were

The quantitation data for both 6-BHA and 8-BOA were

removed from each condition and immediately placed into subjected to non-compartmental analysis (NCA) using

either a sample or reference cuvette. CYP content was then Certara WinNonlin (Princeton, NJ) with standard settings.

measured with an Olis Aminco DW-2 spectrophotometer Furthermore, after plotting the time – concentration profile,

(

Bogart, GA) following the standard procedure for carbon the 8-BOA data was also fit to the standard 2-compartment

monoxide (CO) binding difference spectra (Guengerich et al. model using Graphpad Prism 7.00 (San Diego, CA) according

009). Note: reactions were performed at separate times to to the following equation:

2

ensure incubation time consistency between conditions.

C ¼ Ae^ꢂ^a^tþ Be^ꢂ^b^t

(1)

Ex vivo CYP content

Mixtures were set up on ice in duplicate and contained

Pharmacokinetic parameters were subsequently calculated

either pooled RLM or RKM, that were derived from treated or using the resultant hybrid constants (for comparison to NCA)

control group rats, at concentrations of 2 mg/mL in 100 mM and this equation was used to calculate plasma 8-BOA con-

KPi, pH 7.4, with 0.5% sodium cholate w/v, in a final volume centrations for the CYP4Z1 inactivation simulations

of 1.2 mL. 0.55 mL aliquots for each condition were split (described later).

9

04

J. P. KOWALSKI ET AL.

Metabolite identification

acetonitrile) at a flow rate of 0.3 mL/min. After a 1 min hold,

B was increased linearly from 0 to 100% between 1 and

Sample preparation

1

4 min and held at 100% between 14 and 15.1 min; the total

Equal volumes of remaining plasma were separately pooled

from the time points spanning 5-min to 24-hr for treated

and control group rats. A 6-fold volume of ice-cold aceto-

nitrile was added, the samples vortexed briefly, centrifuged

run time was 20 min following a 4.9 min period for equilibra-

tion. Analysis of the mass spectra was performed with SCIEX

Metabolite Pilot V2.0 with the settings as described below.

ꢀ

for 10 min at 10,000 x g and 4 C, and then the supernatants

were transferred to new tubes. The samples were dried Metabolite Pilot settings

under a constant stream of N at room temperature, resus- 8-BOA and 6-BHA were each run as parent compounds (as 6-

2

pended in 150 lL of a 70/30 mixture of water/acetonitrile, BHA was identified later as a major metabolite that down-

ꢀ

and then stored at 4 C until use.

stream biotransformation processes could conceivably occur

Equal volumes of urine from the bladder collections at to), therefore reference MS/MS spectra for both was pro-

the 24-hr time point for the control group as well as for the vided. Control group samples for both urine and plasma

8

1

-BOA treated rats were combined and the internal standard matrices were run in conjunction with the treated group and

-BOA-d was spiked in to achieve a final concentration of acted as the background filtering control for metabolite

4

00 nM (to enable quantification of any remaining 8-BOA

peaks. The processing parameters for Metabolite Pilot in

both ESI ± modes consisted of the following TOF MS settings:

predicted metabolites, generic peak finding, application of

mass defect filter, and isotope pattern. The TOF MS/MS set-

tings consisted of the following: find at least 2 characteristic

product ions, find at least 1 characteristic neutral loss, and

consider internal neutral losses. The biotransformation set

contained 100 phase I/II and amino acid conjugation path-

ways. MS m/z tolerance was set to 20 ppm and the minimum

MS peak intensity was set to 200 CPS; MS/MS m/z tolerance

was set to 10 mDa and the minimum MS/MS peak intensity

was set to 100 cps. In addition to identifying protonated/de-

protonated metabolites, other common adducts were also

searched for. Results were manually filtered and metabolites

were excluded that either a) completely lacked MS/MS spec-

tra (and weren’t identified in the other ESI mode), b) showed

a < 10-fold difference in signal intensity between control and

sample, c) were readily interpreted as in-source fragmenta-

tion (unless they were notable and hence were included), or

d) displayed peak chromatography consistent with a slight

offset to a control peak and therefore was artifactually identi-

fied as a metabolite by the program.

parent). To create a callibration curve, pooled male Sprague

Dawley rat urine was spiked with 8-BOA diluted 200-fold v/v

from DMSO stocks to achieve final concentrations from

1^–¹⁰⁰⁰ⁿ^M^,ⁱⁿ^h^a^l^f^-^dⁱ^l^u^tⁱ^oⁿ^s^,^aⁿ^d⁸^-^B^O^A^-^d4 was added to

achieve a final concentration of 100 nM. The pH was adjusted

to ꢄ2 with acetic acid for all urine samples and they were

subsequently extracted three times with a 50/50 mixture of

diethyl ether/ethyl acetate. The combined organic fractions

were dried under a constant stream of N at room tempera-

2

ture, resuspended in 150 lL of a 70/30 mixture of water/

acetonitrile, and then stored at 4 ˚C until use. A portion was

assayed for 8-BOA concentration according to the General

LC-MS/MS method for the quantitation of 8-BOA, 6BHA, and

assay controls.

Qualitative LC-MS/MS profiling for 8-BOA rat metabolites

Aliquots (30 lL) were analysed by UPLC-MS/MS on a Waters

Acquity UPLC connected to an AB Sciex TripleTOF 5600 mass

spectrometer in both ESI ± modes (separately) with parent

and product ions acquired concurrently in the same period

using settings for experiment 1 and 2, respectively. For

experiment 1, TOF MS scanning was restricted to

0–1000 Da range with a cycle time of 0.55 sec and an accu-

mulation time of 0.1 sec. Information dependent acquisition

IDA) used switch criteria settings that required ions to be

100 m/z, exceed 150 cps, and to exclude former target ions

a

5

CYP activity assays

Reaction set-up

(

>

For CYP activity comparisons between the pooled RLM

derived from treated and control group rats, the probe sub-

strates and the biotransformations that were monitored (all

for 1 second. The nominal mass of 8-BOA was set to 276 Da,

a 700 Da formula width was used, and a mass defect of

1

58.6261 mDa was applied; mass tolerance was 50 mDa. done separately) were as follows: CYP1A/-2C phenacetin O-

Settings for DP, CE, and XA1 were 100, 5, and 136.62, deethylation, CYP2C/-2D dextromethorphan O-demethylation,

0

respectively. For experiment 2, TOF MS/MS scanning was and CYP3A/-CYP2C midazolam 1 -hydroxylation. Reactions

again restricted to a 50–1000 Da range with an accumulation were set up on ice in triplicate and contained microsomes at

time of 0.1 sec. For this product ion acquisition, settings for 0.25 mg/mL protein in 100 mM KPi pH 7.4, with 3.2 mM

DP, CE, CES, IRD, IRW, and XA1 were 100, 35, 20, 67, 25, and MgCl

0.16, respectively. For both experiment 1 and 2, settings for substrates were each added independently from DMSO

GS1, GS2, CUR, TEM, and ISVF were 60, 60, 35, 600, and stocks, diluted 400-fold v/v, to the reactions to achieve final

, in a final volume of 200 lL. The above listed probe

2

7

4

500, respectively.

concentrations of 10 lM (reactions contained 0.25% v/v

Chromatographic separation of analytes was achieved DMSO). After a 5 min period at 37 C for equilibration, the

ꢀ

using a Waters SunFire C₁₈column (2.1 ꢃ 150 mm, 3.5 lm), reactions were started by the addition of NADPH at a final

starting with 100% mobile phase A (0.1% formic acid in concentration 1 mM. The reactions were quenched after

water) and 0% mobile phase B (0.1% formic acid in 20 min with an equal volume of ice-cold acetonitrile

XENOBIOTICA

905

containing a mix of all deuterated internal standards at a

concentration of 200 nM each. The samples were centrifuged

for 10 min at 10,000 x g and 4 C, and then the supernatants

were transferred to new tubes. For quantitation, metabolite

standards were spiked all together into an identical matrix to

attain concentrations spanning 0.001–1 lM, in half-log dilu-

tions, and underwent sample workup identical to what is

described above.

ActiveCYP4Z1enzymeð%Þ

ꢀ

ꢁ

(4)

ꢂðkþkdegÞꢃDt

ꢀ

¼ e

þ ðkdeg ꢃ Dt ꢃ 100

The simulation was performed in Excel with the following

parameters:

fu ¼ 0:02

ꢂ1

k_i_n_a_c_t¼ 0:15min

ꢂ

ꢃ

lg

KI ¼ 2:2lM KI ¼ 0:61

mL

LC-MS/MS method for quantification of probe substrate

metabolites

À

Á

ꢂ1

kdeg ¼ 0:00048min

kdeg ¼ 0:00024min

kdeg ¼ 0:00012min ðt

t1=2, CYP ¼ 1day

À

Á

Aliquots (5 lL) were analysed by UPLC-MS/MS on a Waters

Acquity UPLC connected to a Waters Xevo TQ-S instrument

ꢂ1

t

¼ 2days

¼ 4daysÞ

1=2, CYP

ꢂ1

þ

1=2, CYP

in ESI mode, with the following settings: capillary 2.9 kV,

ꢀ

Dt ¼ 15min:

source offset 60.0 V, source temperature 150 C, desolvation

ꢀ

temperature 350 C, cone gas flow 150 L/hr, desolvation gas

Values for ½drugꢅ and k were determined for each time-

free

flow 800 L/hr, collision gas flow 0.15 mL/min, cone voltage

point of the time-step. Starting from time zero (100% active

CYP4Z1), all subsequent values calculated for active CYP4Z1

enzyme were performed on the previously calculated fraction

4

3

0 V, and collision energies of 20 eV for acetaminophen and

0 eV for the other analytes. Chromatographic separation uti-

lized a Waters HSS T3 column (2.1 ꢃ 100 mm, 1.8 lm), start-

ing with 95% mobile phase A (0.1% formic acid in water)

and 5% mobile phase B (0.1% formic acid in acetonitrile) at a

flow rate of 0.3 mL/min. After a 1 min hold, B was increased

linearly from 5 to 60% between 1 and 7 min, immediately

increased to 95%, and then held at 95% between 7 and

(

and converted to percent) such that inactivation was itera-

tive with time. Plasma drug_f_r_e_econcentrations were assumed

to equal intratumoral concentrations, the tumour was

assumed to be in the central compartment, and the micro-

somal protein binding of 8-BOA for the in vitro conditions

used to calculate kinact and KI in CYP4Z1 membranes was

assumed to be minimal. Although not determined experi-

mentally, if microsomal binding of 8-BOA to membranes

were substantial, this would result in a lowered effective KI

and increased potency of inactivation in this model.

8

.5 min; the total run time was 11 min following a 2.5 min

period for equilibration. The metabolites and isotope-labeled

internal standards were detected by MRM with the following

transitions and retention times: acetaminophen m/z

1

52 > 110, acetaminophen-d₃155 > 111 (3.3 min), dextro-

rphan 258 > 157, dextrorphan-d₃261 > 157 (4.5 min), 1-

hydroxymidazolam 342 > 203, and 1-hydroxymidazolam-d₄

Results

3

46 > 203 (5.4 min). Analysis of the mass spectra was per-

formed with Waters MassLynx V4.1 software.

Metabolic stability

In order to evaluate microsomal stability and the summative

0

(

potentially sequential) contributions of CYP- and uridine 5 -

Kinetics of CYP4Z1 inactivation from 6-BHA

diphospho-glucuronosyltransferase (UGT)- mediated metabol-

ism to clearance, 8-BOA at an initial concentration of 1 mM

was incubated for an extended period with RLM or RKM that

had been fortified with NADPH and UDPGA. After 90 min, 8-

BOA concentrations were reduced by ꢄ6 and ꢄ4% for RLM

and RKM, respectively (Figure 1(A)). Minimal depletion was

also evident when 3 mM of 8-BOA was used (data not

shown). Although calculation of an in vitro microsomal t₁_/₂

for 8-BOA was not practical due to the negligible decline

and margin of error for measurements, we estimate these

values to be >12 hr. In control RLM studies, midazolam and

Inhibition and probe substrate reactions were setup and run

analogous to the method in our previous study and LC-MS/

MS quantification was also similarly run (Kowalski et al.

2

020). Briefly, in reactions containing 5 pmol CYP4Z1, 6-BHA

was diluted 400-fold v/v from DMSO stocks to achieve final

concentrations of 0, 6, 10, 20, 40, 60, 80, and 120 lM (reac-

tions contained 0.25% v/v DMSO), and aliquots of the inhib-

ition reactions were taken at times of 0, 5, 10, and 15 min for

the residual activity reactions, using the probe substrate luci-

ferin-benzyl ether (Luc-BE).

7

-hydroxy coumarin depletion both displayed in vitro t₁_/₂val-

CYP inactivation simulations

ues of <5 min (data not shown).

The extrapolation of 8-BOA plasma concentrations to inacti-

vation of the target CYP4Z1, in a theoretical rat xenograft, Off-target CYP inactivation

was performed using the 2-compartment model Equation (1)

outlined above to calculate drug concentrations ([drug]), and

then using the following equations:

To assess degree of off-target microsomal CYP inactivation

caused by 8-BOA, carbon-monoxide (CO)-binding spectra

were recorded for NADPH-fortified RLM after incubation with

½

drugꢅ_f_r_e_e¼ fu ꢃ ½drugꢅ

(2)

(3)

8

-BOA or DMSO vehicle (Figure 1(B)). After 30-minute incuba-

k ¼ ð½drugꢅ ꢃ kinactÞ=ð½drugꢅ þ KIÞ

free

tions the CYP contents were 0.62, 0.57, and 0.34 nmol/mg

9

06

J. P. KOWALSKI ET AL.

Figure 2. Plasma time-concentration profile for 8-BOA, administered via IV

bolus at 10 mg/kg, and the metabolite 6-BHA, in male Sprague Dawley rats.

Non-compartmental analysis was utilized to calculate pharmacokinetic parame-

ters. Shown is the average ± SD (n ¼ 3).

Table 1. Pharmacokinetic parameters for 8-BOA and 6-BHA.

Parameter

max (hr)

8-BOA

6-BHA

T

–

0.14 ± 0.10

C

max (mg/mL // mM)

37 ± 10 // 130 ± 36

55 ± 6.5 // 200 ± 24

6.1 ± 1.0

1.5 ± 0.14 // 6.1 ± 0.10

C

0

(mg/mL // mM)

–

a

t1/2, elim. (hr)

1.8 ± 0.20

ꢆ

AUC0!1 (hr mg/mL)

CL (mL/min/kg)

V

14 ± 5.8

14 ± 7.3

7.9 ± 4.8

0.59 ± 0.50

0.84 ± 0.20

–

Figure 1. In vitro assessment of metabolic stability and off-target CYP inactiva-

tion. 8-BOA (1 mM) was incubated with pooled RLM or RKM (Xenotech) in the

presence of NADPH (1.5 mM), UDPGA (5 mM), and alamethicin (10 lg/mL).

Minimal depletion of 8-BOA was observed in either microsomal system over the

course of 90 minutes. Shown is the average and standard deviation from three

replicates (A). Pooled RLM was incubated in the presence of NADPH for

z

(L/kg)

ss (L/kg)

Shown are the averages ± SD (n ¼ 3)

a

Terminal phase likely not fully characterized

3

0 minutes with either DMSO (control) or 8-BOA, and standard CO-difference

spectra was recorded for each sample. Absorbance at 450 nm decreased with

increasing concentrations of 8-BOA, resulting in a ꢄ7 and ꢄ45% reduction of enabled the calculation of 8-BOA concentrations for the

total CYP content at 100 and 500 mM, respectively.

time-steps required for simulating CYP4Z1 inactivation. To

assess goodness of fit, a residual plot was analysed which

showed a random pattern, indicating that the application of

the 2-compartment model to the 8-BOA data was acceptable

protein for the DMSO control, 100, and 500 mM of 8-BOA,

respectively. This equates to dose-dependent losses of 7-45%

in liver microsomal CYP content.

(data not shown).

The time-concentration profile of the metabolite 6-BHA in

plasma is also depicted in Figure 2. NCA yielded values for

T_m_a_x, C_m_a_x, t₁_/₂, and AUC, which are summarized in Table 1.

Plasma protein binding (PPB)

As determined by rapid equilibrium dialysis with LC-MS/MS

quantitation, 8-BOA and 6-BHA (3 mM) were bound 98 ± 0.20

and 75 ± 4.0%, respectively, to rat plasma proteins. Warfarin,

used as an assay control, was bound 99 ± 0.10%, as expected.

A recognized limitation is that only single concentrations of

ligands were tested, as such, linearity was not assessed.

Metabolite identification

Plasma profiling

LC-MS/MS analysis was performed in both ESI and ESI

mode to ensure the best coverage of analytes, since ESI

would likely be required to detect some acid-containing

metabolites. Given the complexity of the matrices involved,

this ensured that mass-defect filtering with information-

dependent acquisition (IDA) would have the most opportuni-

þ

-

Pharmacokinetics of 8-BOA and 6-BHA

The time-concentration profile of 8-BOA in plasma after IV ties to trigger MS/MS fragmentation required for metabolite

bolus administration at 10 mg/kg is shown in Figure 2. Inter- identification.

animal variability was relatively low, and the profile exhibited

A complex profile of 25 total metabolites were detected

a multi-exponential pattern, indicative of both distribution by LC-MS/MS in the plasma of rats following IV administra-

and elimination phases being present. The pharmacokinetic tion of 8-BOA. The extracted ion chromatograms (XICs) for

þ

-

constants resulting from non-compartmental analysis (NCA) both ESI and ESI modes used for metabolite identification

are summarized in Table 1. Prolonged, albeit low, exposure are shown in Figure 3. While M3 – M10, M13, M15, and

to 8-BOA was observed, characterized by an extended t₁_/₂M16 were detected in both modes, M1, M2, M11, M12 and

þ

and low CL. The standard whole body 2-compartmental M14 were detected only in ESI , and M17 – M25 were

-

model was also fit to the data, which yielded the following detected only in ESI . Given that the plasma was pooled

hybrid values: A ¼ 43.45, a ¼ 3.32, B ¼ 0.068, b ¼ 0.12; this from timepoints spanning 5 min–24 hr, in order to increase

XENOBIOTICA

907

Figure 3. Metabolite identification in pooled rat plasma (5-min to 24-hr) via LC-MS/MS analysis. Overlaid extracted ion chromatograms (XIC) for detected metabo-

þ

lites in ESI mode. The metabolites M3 (loss of (CH

2

)

4

), M9 (loss of (CH

2

)

2

), M6 (hydroxylated), and M8 (hydroxylated), displayed the largest peak areas (A).

-

þ

Overlaid XIC for detected metabolites in ESI mode; metabolites with the largest peak areas matched that which were observed in ESI mode (B). M1, M2, M11,

M12 and M14 were detected only in ESI , M17 – M25 were detected only in ESI , and M3 – M10, M13, M15, M16 were detected in both modes. The Parent (8-

BOA, dotted peak) eluted at 9.3 min and has been plotted at 5% of intensity in both ESI and ESI modes for visual comparison; numerous metabolites show inten-

þ

-

þ

-

sities approximately ꢇ5% that of the Parent.

metabolite MS counts, the Parent 8-BOA constituted a dom- same four metabolites accounted for 48% of the total profile

ꢂ

inant signal. Therefore, the chromatograms showing metabo- in ESI mode.

lites in both modes have the Parent peak plotted at 5% of

þ

the detected intensity for ease of viewing. In ESI mode, M9,

M3, M6, and M8 were the major metabolites by peak area, ESI plasma metabolite assignments

þ

and accounted for 67% of the total profile, with M9 and M3 All metabolites were observed as the protonated ion

þ

themselves accounting for 28 and 20%, respectively. These [M þ H] , except M1 as [M] , and these were absent from

9

08

J. P. KOWALSKI ET AL.

Table 2. Plasma metabolites identified in ESIþ mode.

Peak ID

Biotransformation

Formula

m/z

ppm

R.T. (min)

Peak Area

Parent

M1

M2

M3

M4

M5

M6

M7

M8

N/A (8-BOA) [M þ H]þ

C

14

H

20

N

4

O

2

4

277.1662

364.198

1.2

0.2

1.4

9.26

5.04

5.96

7.04

7.14

7.27

7.61

7.69

8.01

8.03

8.64

8.69

9.02

9.10

9.79

10.13

10.51

7.83E þ 07

9.59E þ 04

1.80E þ 04

1.73E þ 06

5.79E þ 04

9.44E þ 04

8.74E þ 05

3.68E þ 05

7.53E þ 05

2.44E þ 06

5.69E þ 05

1.36E þ 05

1.40E þ 05

3.51E þ 05

1.49E þ 05

4.12E þ 05

4.59E þ 05

-(CH

2

)

4

þCarnitine [M]þ

17

26

5

-C

-(CH

8

H

2

14 2

)

O

[M þ H]þ

6

H N

4

135.0667

221.1034

293.1607

293.1609

291.1453

293.1607

249.1347

263.1502

275.1505

275.1506

305.1608

291.1813

275.1503

305.1976

4

[M þ H]þ

10

14

12

13

14

15

14

16

H

12

20

18

20

16

18

20

22

18

24

N

4

O

4

O

4

O

4

O

4

O

4

O

4

O

4

O

4

O

4

O

4

O

4

O

4

O

4

O

2

3

2

3

2

0.3

Oxidation [M þ H]þ

Ketone Formation [M þ H]þ

Oxidation [M þ H]þ

H

ꢂ0.3

0.4

0.5

ꢂ0.3

M9

-(CH

2

)

2

[M þ H]þ

0.4

M10

M11

M12

M13

M14

M15

M16

2

þMethylation [M þ H]þ

ꢂ0.2

0.8

1.4

Desaturation [M þ H]þ

Ketone þ Methylation [M þ H]þ

Methylation [M þ H]þ

ꢂ0.1

0.3

Desaturation [M þ H]þ

2x Methylation [M þ H]þ

0.1

1.3

pooled plasma derived from the control group (Table 2). cleavage of this bond is diagnostic for such regiospecificity.

LC-MS/MS analysis of the reference standard for the Parent The concomitant aldehyde that is formed from this fragmen-

8

-BOA (Figure S3) was identical to that observed in plasma tation (oxidation of the former b-hydroxyl group) also

and provided diagnostic fragments, e.g., m/z 135, 119, and resulted in the expected ion of m/z 233. Similarly, M8 was

0, which aided in metabolite assignments. Characteristic assigned as the a-hydroxy metabolite due to the strong sig-

9

dehydration of the carboxylic acid (minus 18 Da) and dehy- nal for the diagnostic ion of m/z 247, representing cleavage

dration plus loss of N₂(minus 46 Da) were also commonly between the a-hydroxy and carboxylic acid moiety. For M4

observed. For ease of viewing, the annotated structure frag- and M5, the fragment ions m/z 160 and 132 were consistent

ment ions are listed without defect values. These are shown with oxidation occurring at the c- through e-positions of the

in the MS/MS spectra where differential defects from equiva- aliphatic tail, and furthermore, no strong evidence of hydrox-

lent nominal masses can be discerned between structures.

ylation on the ABT moiety was observed. However, due to

the limited MS/MS spectra and lack of reference standards,

absolute regiospecificity could not be defined.

The largest metabolite by peak area, M9 (m/z 249.13

þ

[

M þ H] ), was assigned as having the formula C H N O

1

2 16 4 2

þ

that corresponded to the Parent 8-BOA undergoing one

round of b-oxidation (loss of two methylene units). M9 dis-

played ions m/z 231, corresponding to dehydration, and m/z

The metabolite M7 (m/z 291.14 [M þ H] ) exhibited frag-

mentation consistent with ketone formation at the b-carbon,

while fragmentation of the desaturation metabolites M11

þ

2

⁰³^,^c^o^r^r^e^s^p^oⁿ^dⁱⁿ^g^t^o^d^e^h^y^d^r^a^tⁱ^oⁿ^p^l^u^s^l^o^s^s^o^f^N2 (Figure

and M12 (m/z 275.15 [M þ H] ) placed the olefins on the

4

(A)). The MS/MS and retention time of M9 was identical to

alkyl-acid chain but could not define exact positions. M15,

another apparent desaturation product from the observed

m/z, eluted later than the Parent and had a matching reten-

tion time and accurate mass to an intermediate imine com-

pound (positioned adjacent to ABT) generated during the

synthesis of 8-BOA, hence it was assigned as such (data not

shown). Various other metabolites were consistent with par-

ent methylations of varying regiospecificity (M13, M16), loss

of the side chain -C H O to generate M2 (i.e., ABT, also

that of the reference standard for the chain-shortened ana-

log 6-[(1H-1,2,3-benzotriazol-1-yl)amino]hexanoic acid (6-BHA,

Figure S3 ), thus confirming the structure for this

major metabolite.

The metabolite with the next greatest peak area, M3 (m/z

þ

2

21.10 [M þ H] ), was assigned the formula C H N O ,

1

0 12 4 2

which is consistent with the Parent 8-BOA undergoing two

rounds of b-oxidation (loss of four methylene units). MS/MS

of M3 identified an ion at m/z 175 corresponding to dehy-

8

14 2

confirmed through a reference standard) and M1, the carni-

dration plus loss of N (Figure 4(B)). Additionally, the reten-

þ

2

tine conjugate of M3. All 16 metabolites identified in ESI

tion time of M3 matched the relative elution progression

observed for the Parent 8-BOA and the metabolite 6-BHA

mode are summarized in Table 2, and the MS/MS spectra

that aided in the structural assignments for all metabolites

have been provided in the Supporting Information .

(M9). Therefore, although lacking a reference standard, there

is strong evidence for assigning M3 as 4-[(1H-1,2,3-benzotria-

zol-1-yl)amino]butanoic acid (4-BBA).

-

The presence of four mono-hydroxylated metabolites in ESI plasma metabolite assignments

the profile was indicated by their shorter elution times and All metabolites were observed as the de-protonated ion [M-

-

gain of 16 Da, relative to the Parent, resulting in molecular H] (although other minor adducts were also present) and

þ

ions of m/z 293.16 [M þ H] , with M6 and M8 dominating. were absent from pooled plasma derived from the control

-

The MS/MS spectra for M6 is shown in Figure 4(C), where group (Table 3). In ESI mode, the major metabolites M9,

sequential dehydration generated fragments of m/z 275 and M6, and M8 all possessed fragmentation consistent with the

þ

2

57, followed by removal of CO to generate m/z 231. This previously assigned structures derived from their ESI frag-

-

fragment ion of m/z 231 was unique to M6 and definitively mentation profile, although the ESI MS/MS spectra proved

allows for assignment as the b-hydroxy metabolite since less diagnostic in general. While the molecular ion for M3

XENOBIOTICA

909

þ

Figure 4. LC-MS/MS analysis, in ESI mode, of the top three metabolites of 8-BOA (by peak area) in pooled rat plasma. M9, the metabolite with matching RT and

MS/MS as the chain-shortened analog 6-BHA was confirmed as the result of a loss of (CH from the Parent (A). M3 shows MS/MS spectra corresponding to a fur-

ther chain-shortening of 6-BHA, resulting in a loss of (CH from the Parent, and has been tentatively identified as 4-[(1H-1,2,3-benzotriazol-1-yl)amino]butanoic

2 2

)

2 4

)

acid (4-BBA) (B). M6 (Parent plus 16 Da), shows fragmentation consistent with hydroxylation at the b-carbon of the 8-BOA aliphatic tail (C).

-

(

m/z 219.09 [M-H] ) still generated a strong signal, no MS/MS presence of the ABT moiety. Although only the 1-b anomer

spectral fragments were triggered for this metabolite.

is depicted in the annotated scheme, M20 may in fact be

-

Notably, ESI analysis revealed M20 – M23 as glucuronide the 1-b anomer while the M22 cluster likely represents a mix

conjugates, which in total accounted for ꢄ13% of the total of the 2-, 3- and 4-a/b acyl-glucuronide anomers.

peak area for metabolites observed in this mode (Figure

Metabolites M21 and M23 also showed similar fragmenta-

-

S4(A)). M20 and M22 had identical mass fragmentation pro- tion profiles (m/z 319.14 [M-H] ) and were assigned the for-

-

files (m/z 451.18 [M-H] ) (Figure S4(B)), and were assigned the mula C H O , which corresponded to acyl-glucuronide

14 24 8

formula C H N O which is consistent with glucuronidation structures of the Parent minus the ABT moiety.

2

0 28 4 8,

of the Parent. Fragmentation produced the aglycone ion at Fragmentation of M21 (Figure S4(C)) produced the aglycone

m/z 275, with the corresponding dehydrated glucuronate at ion at m/z 143, the expected dehydrated glucuronate at m/z

m/z 175, and the further dehydrated glucuronate minus CO₂175, and the further dehydrated glucuronate minus CO at

2

at m/z 113. Direct fragmentation evidence for the acyl-glu- m/z 113. M21 and M23 had nearly overlapping retention

curonide is provided by the m/z 193 ion, which represents times with the aforementioned acyl-glucuronide cluster and

the full glucuronate that would be absent for the N-glucuro- therefore may be artifactually generated via in-source frag-

nide. Furthermore, the ion at m/z 133 confirmed the mentation of M22 or sample work-up.

9

10

J. P. KOWALSKI ET AL.

Table 3. Plasma metabolites identified in ESI- mode.

Peak ID

Biotransformation

N/A (8-BOA) [M-H]-

Formula

m/z

ppm

R.T. (min)

Peak Area

Parent

M17

M18

M19

M3

M4

M5

M6

M7

M20

M8

M9

M21

M22

M23

M10

M24

M13

M15

M25

M16

C

14

20

H N

4

O

2

275.1524

292.1399

548.1005

263.1156

219.0896

291.1463

291.1457

291.1468

289.1310

451.1824

291.1465

247.1207

319.1389

451.1822

319.1390

261.1358

319.1777

303.1466

273.1368

377.1726

303.1823

3.7

N/A

2.5

4.1

0.1

9.27

1.71

4.78

6.66

7.05

7.17

7.27

7.61

7.67

7.80

8.02

8.04

8.27

2.02E þ 07

2.28E þ 05

4.02E þ 04

6.94E þ 04

7.51E þ 05

3.61E þ 04

5.45E þ 04

5.85E þ 05

1.41E þ 05

8.33E þ 04

7.20E þ 05

6.07E þ 05

1.80E þ 05

3.43E þ 05

8.28E þ 04

1.12E þ 05

1.92E þ 05

2.23E þ 05

3.53E þ 04

5.70E þ 05

4.49E þ 05

Unassigned þ16.9885

Unassigned

Unassigned þ272.9481

-(CH

2

)

2

þOxidation [M-H]-

C

12

10

14

20

14

12

14

20

14

13

16

15

14

H

16

H

12

H

20

H

20

H

20

H

18

H

28

H

20

H

16

H

24

H

28

H

24

H

18

H

24

H

20

H

18

N

O

N

O

N

4

8

4

8

4

O

3

2

3

8

3

2

4

[M-H]-

Oxidation [M-H]-

Ketone Formation [M-H]-

Glucuronidation [M-H]-

Oxidation [M-H]-

ꢂ1.9

1.8

1.3

ꢂ2.3

0.9

-(CH

-C

2

H

)

2

[M-H]-

þGlucuronidation [M-H]-

2.7

6

4

N

4

ꢂ2.8

ꢂ2.7

ꢂ2.8

0.3

1.1

Glucuronidation [M-H]-

-C

þGlucuronidation [M-H]-

-(CH

O

8

8.31–8.64

8.56

6 4 4

H N

2

)

2

þMethylation [M-H]-

O

2

3

2

8.67

8.74

9.12

10.12

10.44

10.52

2x Methylation þ Oxidation [M-H]-

Ketone þ Methylation [M-H]-

Desaturation [M-H]-

4.0

Unassigned þ102.0212

Unassigned

C H N O

16 24 4 2

N/A

ꢂ1.0

2x Methylation [M-H]-

M17, M18, and M25, all exhibiting gains in mass over the Ex vivo analysis

Parent, could not be assigned as MS/MS fragmentation was

limited. All 20 metabolites identified in ESI mode and their

Off-target CYP inactivation

-

To assess the degree of off-target CYP inactivation in vivo

from 8-BOA treatment, liver and kidney microsomes prepared

from both control and treated groups were used for ex vivo

CO-binding experiments to quantify the total pool of CYP

content. Pooled liver microsomes yielded values of 0.71 and

assigned structures have been summarized in Table 3, with

the corresponding MS/MS spectra provided in the

Supporting Information.

0

.60 nmol/mg for the control and treated rats, respectively,

equating to a ꢄ15% decrease due to treatment with 8-BOA.

No appreciable difference in kidney microsomal CYP content

was evident, with concentrations equalling 0.13 and

Major urinary metabolites

Analysis of urine was performed on a limited bladder collec-

tion volume from the 24-hr necropsies, therefore, only the

major metabolites are reported. A small amount of the

Parent 8-BOA was excreted unchanged, with an average

concentration of ꢄ9 nM in the 24-hr urine. M3, the tetra-

0

.14 nmol/mg for control and treated samples, respectively

(

Figure 5(A)).

þ

demethylenation product, was the major metabolite in ESI

Off-target CYP activity

mode, with matching retention time and fragmentation to As another measure of off-target inhibition, ex vivo analysis

0

that seen in plasma (Figure S5(A,C)). Using 8-BOA calibration of midazolam 1 -hydroxylation (M1OH), dextromethorphan O-

standards, the average M3 concentration was estimated to demethylation (DOD), and phenacetin O-deethylation (POD)

be ꢄ400 nM (although due to a number of unknowns, this activities was performed to assess CYP3A, ꢂ2 C, and ꢂ1 A

should not be regarded as a semi-quantitative concentra- activity, respectively, in liver microsomes derived from both

tion). M21, the acyl-glucuronide minus ABT, appeared to be control and treated groups. Rates of M1OH, DOD, and POD

-

the major metabolite in ESI mode, with matching retention for control microsomes were 42.5 ± 1.0, 113 ± 0.7, and

time and fragmentation to that seen in plasma (Figure S5(B, 31.0 ± 1.0 pmol/mg protein/min, respectively (average ± SD).

Rates of M1OH, DOD, and POD for treated microsomes were

D)). Again, M21 may not be a unique metabolite in itself,

4

3.7 ± 2.0, 118 ± 6.0, and 26.8 ± 2.0 pmol/mg protein/min,

but rather an artefact of in-source fragmentation.

respectively (Figure 5(B)). While no decreases in M1OH or

DOD activity were evident, a minimal 8.6% decrease in POD

activity (CYP1A) was observed.

Characterization of mechanism-based inactivation of

CYP4Z1 by 6-BHA

Clinical chemistries

Time- and concentration-dependent loss of CYP4Z1-mediated Standard markers of liver and kidney function were assessed

Luc-BE O-debenzylation activity was observed from treatment in plasma for the vehicle control group and the 8-BOA

with 6-BHA (Figure S6 ), typical of MBIs. This resulted in treated group for both pre-infusion and 24 hr post-infusion

inactivation kinetic parameters of K ¼ 29 mM and k

timepoints. Most analytes changed very little between condi-

tions, with the exceptions of glucose (GLU) and aspartate

I

inact

ꢂ1

¼

0.19 min

.

XENOBIOTICA

911

and 4 days, respectively. Values at 24 hours post-treatment

were 55, 37, and 23% for t₁_/₂_,_C_Y_Pof 1, 2, and 4 days, respect-

ively. After 96 hr, values were 95, 78, and 49% for t₁_/₂_,_C_Y_Pof

1

, 2, and 4 days, respectively (data not shown).

Discussion

We recently designed and characterized 8-BOA as a selective,

and mechanism-based inhibitor of mammary-restricted

human CYP4Z1 (Kowalski et al. 2020). Through inhibition of

this target enzyme, we could expect to lower production of

its PUFA-derived pro-angiogenic metabolites (such as 14,15-

EET), thereby potentially reducing the growth of metastatic

breast cancer. However, detailed assessment of the inhibitor’s

Figure 5. Ex vivo assessment of off-target CYP inactivation in rats treated with

-BOA. Liver and kidney were harvested after 24 hours and microsomes were

8

prepared from rats receiving either 8-BOA (treated, grey bars) or a vehicle infu- in vivo pharmacokinetic characteristics is necessary to guide

sion (control, black bars). CYP content from pooled microsomes was analysed

the design of CDX/PDX-rodent model experiments and to

evaluate whether a prodrug approach or bioisosteric modifi-

via CO-binding difference spectra, where a ꢄ15% decrease in liver microsomal

CYP content was observed; no appreciable decrease in kidney microsomal CYP

content was evident. Shown is the average from two replicates (A). Pooled liver cations would further lead compound optimization. The com-

microsomes derived from either control or treated animals were incubated with

plexities surrounding pre-clinical models of breast cancer are

separate probe substrates in the presence of NADPH and rates of midazolam

well modelled in rats, and this species better mimics human

1’-hydroxylation (M1OH), dextromethorphan O-demethylation (DOD), and phen-

acetin O-deethylation (POD) were determined. A ꢄ9% decrease in POD activity physiology in comparison to mice (Smits et al. 2007,

was observed in the microsomes derived from treated compared to control ani-

Iannaccone and Jacob 2009, Mollard et al. 2011). Therefore,

mals, while no decreases in M1OH or DOD activity was evident. Shown is the

the rat was used to capitalize on these aspects and, addition-

ally, the practical considerations for this pilot study of ena-

bling serial blood draws with a minimal number of animals.

Preliminary in vitro experiments assaying for metabolic

stability of 8-BOA in liver and kidney microsomes (supple-

mented with co-factors to predict summative CYP- and UGT-

mediated clearance) indicated the test article is a highly sta-

ble compound, as negligible depletion of 8-BOA was

observed. We reasoned that this lack of depletion might be

due to potent inactivation of rat CYP isoforms by this ABT

analog. However, as evidenced by minimal inactivation of

the general pool of rat liver microsomal CYPs following treat-

ment with 100 mM of 8-BOA, this was clearly not the case in

average and standard deviation from three replicates (B).

Table 4. Kinetic parameters of CYP4Z1 inactivation for 8-BOA and 6-BHA.

Parameter

(mM)

8-BOA

6-BHA (M9)

a

K

I

2.2

0.15

29

-

1

a

kinact (min )

0.19

6.5E-3

0.25

116

-

1

-1

k

inact /K

I

(min mM )

6.8E-2

fu

0.02

K

I, fu adjusted (mM)

110

1.4E-3

-

1

-1

kinact /KI, fu adjusted (min mM )

1.6E-3

Shown are the averages from technical replicates (n ¼ 3)

a

Determined from a previous study (Kowalski et al., 2020)

aminotransferase (AST). Control and treated groups both saw vitro. Therefore, we hypothesized that 8-BOA may either

ꢄ2.7-fold decreases in GLU, and 2.0- and 1.7-fold increases, experience a prolonged half-life in vivo or be excreted sub-

respectively, in AST levels from pre-infusion to 24-hr. stantially unchanged.

Therefore, no indication of acute toxicity arising from treat-

A single low dose of 10 mg/kg was chosen to avoid

ment with the test article is evident, as the increased AST potential organ toxicities that might arise, as this was the

levels are likely from the vehicle formulation that was used. first in vivo study for this new chemical entity. IV bolus

The results for all analytes, groups, and timepoints are sum- administration was utilized to enable the highest potential

marized in Table S1.

plasma concentration as we were not concerned with oral

bioavailability at this stage of interrogation. This produced a

plasma C_m_a_xof 130 mM and a bi-phasic profile displaying a

prolonged elimination t₁_/₂of 6.1 hr. Although we cannot say

Simulated in vivo CYP4Z1 inhibition

The 8-BOA time – plasma concentration profile from the cur- at this time that 8-BOA would infiltrate into a tumour micro-

rent study was used to simulate the effect this inhibitor environment directly, the high calculated V provides evi-

z

would have on CYP4Z1, if it were expressed in a highly per- dence that 8-BOA may penetrate tissues to a substantial

fused xenografted tumour in the rat. Using constants extent. If future studies demonstrate that an increase in lipo-

obtained in vitro to characterize the inactivation kinetics of philicity is required to raise tissue concentrations, esterifica-

8

-BOA towards CYP4Z1 (i.e., K and k_i_n_a_c_t), the percent of tion of the carboxylic acid may be an effective strategy.

I

active CYP4Z1 enzyme remaining after exposure to this Lastly, a promisingly minimal CL value was observed; assum-

inactivator (over time) was modelled from time 0 (pre-dose) ing that hepatic clearance dominates, the extraction ratio for

to 96-hr post dose. From a starting point of 100% at time 0, 8-BOA in the rat is ꢄ20%, binning into the ‘low’ category

the lowest values of percent active CYP4Z1 remaining were (Davies and Morris 1993, Rowland and Tozer 2011).

modelled to occur at 1.25, 1.5, and 2.25 hr and were 18, 16,

Scaling in vitro data for the prediction of drug-drug inter-

and 15% for the scenarios corresponding to a t₁_/₂_,_C_Y_Pof 1, 2, actions due to mechanism-based inactivation of CYPs for

9

12

J. P. KOWALSKI ET AL.

lacked both a radiolabeled Parent and a majority of metab-

olite standards. We were particularly interested in assessing

the circulating metabolites present in plasma for this pilot

study to elucidate if any could contribute to either off- or

on-target activity (Loi et al. 2013). Therefore, exploratory

metabolite identification was performed with plasma pooled

from the 5-min through 24-hr timepoints. Although a

method such as described by Hamilton et al. using time-

point aliquots for an AUC-representative pool was considered

Figure 6. Simulation of active CYP4Z1 enzyme remaining (%) after treatment

with the inhibitor 8-BOA, in a theoretical breast cancer xenograft model. (Hamilton et al. 1981), low available sample volumes necessi-

Plasma concentrations of 8-BOA were calculated using the equation generated

from the time-concentration profile of this study and the enzyme remaining

tated the use of all available timepoints (particularly to

acquire adequate signal intensity for diagnostic MS/MS

after inactivation calculated from the unbound fraction of 8-BOA (2%) at these

ꢂ1

time points and the parameters K

I

¼ 2.2 mM and kinact ¼ 0.15 min . The results fragmentation).

from using different CYP turnover half-lives (t1/2, CYP), ranging from 1–4 days,

show that CYP4Z1 would remain 45–77% inactivated 24-hours post dose. [8-

BOA] corresponds to the left axis, and the active CYP4Z1 enzyme remaining (%)

for t1/2, CYP ¼ 1, 2, and 4 days corresponds to the right axis.

þ

Analysis of pooled plasma in ESI mode revealed 16

metabolites. The principal metabolite by peak area was M9,

which showed a loss of (CH ) from the Parent. This metab-

2

olite, 6-BHA, was confirmed through matching retention

times and MS/MS fragmentation to a chemical standard. By

inspection of diagnostic fragmentation ions, the next largest

metabolite M3, was assigned as the further chain-shortened

drug candidates has become common practice (Obach et al.

2

007). We previously characterized the MBI parameters in

vitro for the inactivation of CYP4Z1 by 8-BOA, resulting in a

4

-BBA, a total loss of (CH ) from the Parent. The presence

2 4

ꢂ1

K ¼ 2.2 mM and a k

I

¼ 0.15 min (Kowalski et al. 2020).

inact

of M3 and M9 provide strong evidence that b-oxidation of

-BOA is a major pathway of metabolism in the current study

These values, taken together with the pharmacokinetic pro-

file, allowed for an estimation of the potential CYP4Z1 inacti-

vation that would occur in a xenograft model where the

tumour expresses this enzyme. Contrary to an in vitro setting

where the target enzyme can become virtually all depleted,

de novo synthesis occurring in vivo will regenerate the

enzyme pool concurrently. The turnover half-lives that have

been determined for CYPs vary widely due to both isoform

differences as well as the technically challenging, and indir-

ect methods of measurement that are employed. As this

8

and may largely explain the discrepancy between micro-

somal in vitro and in vivo half-lives. It also seems likely that

the desaturation, mono-oxidation, and ketone formation

metabolites that were also identified may be intermediate

products from b-oxidation according to the ‘leaky hosepipe’

theory for this pathway (Stewart et al. 1973, Stanley and

Tubbs 1975). The presence of M19 as a hydroxylated variant

of M9, and M1 as the carnitine-conjugate of M3, are add-

itional markers for b-oxidation (Figure 7); both mitochondria

and peroxisomes could conceivably be the sites of this bio-

transformation (Suga 2003). It should be noted that the

multiplicity of isomers for desaturation, mono-oxidation, and

ketone formation metabolites certainly suggests contribu-

tions from other sources of metabolism as well.

Retrospective analyses of metabolites formed from in vitro

microsomal incubations, compared to plasma metabolites,

may be helpful in delineating between potential intermedi-

ate b-oxidation versus CYP products in future studies. The

use of additional in vitro matrices for metabolite identifica-

tion and comparison, such as hepatocytes and S9 fractions,

would also be helpful experiments to perform for further

interrogation.

ⁱⁿ^f^o^r^m^a^tⁱ^oⁿⁱ^s^c^l^e^a^r^l^y^l^a^c^kⁱⁿ^g^f^o^r^C^Y^P⁴^Z¹^,^w^e^u^s^e^d^a^t1

/2, CYP

range of 1–4 days to encompass the spectrum that has been

reported for this constant (Zhang and Wong 2005, Yang et

al. 2008). Furthermore, for this simple exercise, intratumoral

exposure to the unbound fraction of 8-BOA (2%) was mod-

elled as occurring in the central compartment, ignoring both

the lag-time to reach a peripheral tumour, and the potential

for the concentration to build inside the tumour.

The simulation in Figure 6 shows a rapid decline of active

CYP4Z1 enzyme after exposure to inhibitor. This resulted in a

nadir of 15–18%, suggesting CYP4Z1 activity would be

decreased by 82–85%. After 24 hr, values for active CYP4Z1

remained <60%, indicating that all three t₁_/₂_,_C_Y_Pscenarios

result in prolonged CYP4Z1 inactivation from a single admin-

-

Analysis of pooled plasma in ESI mode revealed many of

þ

istration of 8-BOA. Therefore, given the relatively rapid and the same metabolites as seen in ESI and again, M3 and M9

potent inactivation from 8-BOA and the time required for were majorly represented. As expected, the metabolite pro-

enzyme re-synthesis, maintaining constant high concentra- file was increased and totalled 25 between the two modes

tions for reversible inhibition in vivo is not expected to be of ionization, with glucuronide conjugates being newly

required for efficacy against the hypothesized pro-metastatic revealed. M20 at 7.8 min, and M22, denoting the cluster of

activity of CYP4Z1 (i.e., as a 14,15-EET synthase).

peaks from 8.3 to 8.6 min, showed fragmentation indicative

þ

-

Use of both ESI and ESI modes in LC-MS/MS proved to of acyl-glucuronides. These likely represent a mixture of the

be an effective strategy for metabolite identification as sev- 1-b anomer as well as the 2-, 3- and 4-a/b acyl-glucuronide

eral metabolites were unique in their mode of ionization and anomers akin to that seen in mouse metabolism of diclofe-

fragmentation profile. While clearly there are ionization dif- nac (Sarda et al. 2012). Furthermore, M21 and M23 were

ferences between metabolites, peak areas were used as a identified as ions with fragmentation indicating that they

metric for qualitative comparisons of relative amounts as we were acyl-glucuronide structures that had lost their ABT

XENOBIOTICA

913

2 2

Figure 7. Structures of proposed metabolites in pooled plasma that provide evidence for the b-oxidation pathway. Successive removal of (CH ) generates M9 (6-

BHA, confirmed through comparison to the chemical standard) and then M3 (tentatively assigned as 4-BBA) from the Parent 8-BOA, typical metabolites expected

from b-oxidation. The desaturations, hydroxylation, and ketone formation compounds, M11/M12, M6, and M7, respectively, may be products due to ‘intermediate

leakage’ from the b-oxidation pathway. The presence of M19 as a b-hydroxylated variant of M9, and M1 as the carnitine-conjugate of M3, are additional markers

for this route of metabolism. The presence of multiple isomers of several of the metabolites is evidence that they are also derived from other (most likely CYP) bio-

transformation pathways.

moiety. These metabolites had overlapping retention times

A notable drug example of b-oxidation metabolism occurs

with the presumed acyl-glucuronide cluster and therefore with the class of 3-Hydroxy-3-methylglutaryl-CoA reductase

may be artifactually generated via in-source fragmentation of inhibitors, referred to as the ‘statins’ (Li et al. 2006). At least

M22 or sample work-up. Regardless, acyl-glucuronidation in the case of lovastatin, the chain-shortened pentanoic acid

was identified as an additional biotransformation pathway derivative does not maintain any inhibitory activity towards

for 8-BOA.

its target enzyme (Vyas et al. 1990). Serendipitously, 6-BHA,

For ABT itself, the N-acetylated and N-glucuronidated the confirmed identity of the metabolite M9, had already

products are the major metabolites that have been identified undergone in-depth characterization as it was part of our

in rats (Ortiz de Montellano 2018). We were unable to find congeneric series of analogs and showed promise as a

these metabolites, although M24, corresponding to the CYP4Z1 MBI itself. 6-BHA exhibited a robust level of time-

product of bis-methylation plus oxidation, could conceivably dependent inhibition while maintaining a similar degree of

be due to acetylation on the exocyclic nitrogen followed by low off-target inhibition of human CYPs (Kowalski et al.

reduction, but MS/MS fragmentation was inconclusive. 2020). Therefore, the plasma concentrations of 6-BHA were

Seemingly, the flux towards biotransformation pathways tar- also determined and pharmacokinetic parameters for this

geted to the alkyl-carboxylic acid dominate over pathways metabolite were derived (Table 1). A rapid T_m_a_xof 0.14 hr

targeted to the ABT moiety. Along with the products from indicates 8-BOA undergoes quick biotransformation to 6-BHA

CYP-mediated oxidation, metabolites that were identified in and the elimination t₁_/₂for 6-BHA shows rapid flux to sec-

both modes of ionization included numerous methylated ondary metabolites such as 4-BBA. While it may be true that

species and unassigned adducts that possessed both more the rate constant for 6-BHA metabolism is more rapid than

or less hydrophobicity than the Parent (as discerned through that of its formation from 8-BOA, according to metabolite

retention times).

kinetics logic, the terminal t₁_/₂of the metabolite cannot be

Analysis of urine was performed only on samples faster than the parent (Houston 1981). Therefore, it is likely

obtained during the terminal 24-hour necropsy, and there- that the terminal elimination phase was not fully character-

fore, conclusions drawn from its analysis are limited. ized. Regardless, it is evident that in this scenario 6-BHA

Nevertheless, the major urinary metabolites by peak area in elimination should be considered as formation rate limited.

þ

-

ESI /ESI provided further evidence for two of the main path-

Although exhibiting a weaker affinity for CYP4Z1 (K > 10-

I

ways of metabolism observed from plasma analysis. M3, fold higher), the k_i_n_a_c_tdetermined for 6-BHA shows a simi-

assigned as 4-BBA, was the most dominant metabolite identi- larly quick rate of CYP4Z1 inactivation to that of 8-BOA, 0.19

þ

ꢂ1

fied in ESI mode with an estimated concentration of and 0.15 min , respectively. Additionally, the lowered PPB

ꢄ400 nM. As a comparison, the concentration of unchanged exhibited by 6-BHA, 75% versus 98% for 8-BOA, suggests

8

-BOA in the urine at the 24-hr collection was <10 nM. No that free concentrations of this metabolite are likely to be

metabolite pertaining to a full 8-BOA structure conjugated to more elevated in vivo. Taken together, a comparison of the

ꢂ1

a glucuronide (such as M20 or M22) was detected in urine 8-BOA and 6-BHA k_i_n_a_c_t/K (min mM ) ratios adjusted for

I,

-

in ESI mode. However, the major metabolite identified was PPB indicates comparable mechanism-based inactivation of

M21 (exhibiting the same retention time and MS/MS frag- CYP4Z1 could occur in vivo (outlined in Table 4). Therefore, it

mentation as that found in plasma), but again this metabol- is tempting to speculate that the b-oxidation of 8-BOA may

ite may be artifactual.

serve to augment the desired pharmacodynamic effect of

9

14

J. P. KOWALSKI ET AL.

CYP4Z1 inactivation through the generation of an active (AST) levels from pre-infusion to the 24-hr timepoint; this

metabolite in vivo. Future studies may characterize the was likely due to the infusion formulation that was utilized.

inhibitory potency and PPB of 4-BBA and also dose 6-BHA

itself to assess whether higher plasma exposure can be the rat was defined by low CL and extended plasma expos-

ure suggestive of pharmacodynamic efficacy in an appropri-

The control arm group of the study enabled ex vivo ate pre-clinical model. Metabolite identification revealed

In summary, the IV pharmacokinetic profile of 8-BOA in

obtained with this analog instead.

b-oxidation as a major biotransformation pathway, which

produced a circulating metabolite (6-BHA) that displays rapid

mechanism-based inhibition of CYP4Z1. Ex vivo analysis

showed only minor inactivation of off-target rat CYPs, match-

ing that seen previously with human CYP isozymes. No indi-

cation of acute toxicity was observed. Therefore, this in vivo

characterization of the inhibitor 8-BOA in rats suggests that

we could appropriately interrogate the role of CYP4Z1 in

breast cancer in a xenograft tumour model.

assessment of liver and kidney CYP content/activity as well

as clinical chemistries to evaluate acute toxicity. Given that

8

-BOA showed heightened CYP4Z1-selectivity for time-

dependent inhibition, we decided to limit our assessment to

the proclivity for off-target CYP inactivation to occur in the

rat. The observed ꢄ15% decrease in the CYP content in liver

microsomes derived from the treated animals, where a C

max

of 130 mM was determined, is in good agreement with our in

vitro analysis that showed 7% off-target CYP inactivation

from treatment with 100 mM of 8-BOA. Taken together, these

data demonstrate a relatively low propensity for 8-BOA to

target hepatic CYP isozymes in the rat.

Acknowledgements

The authors thank Dale Whittington and Dr. J. Scott Edgar for mass

spectrometry assistance and Dr. Brian Baer for useful drug metabolism

discussions. The authors recognize Dr. Ellen Joy Plein, who throughout a

Although no truly isozyme-specific probe substrates have

been described for rat liver CYPs, multiple studies have used

semi-selective metabolism of probe substrates (the minor long career contributed greatly to the University of Washington School

of Pharmacy. The Plein endowed funds that were established have

enabled many research opportunities, including the study pre-

sented here.

metabolizing isozymes are noted in parenthesis): M1OH to

assess CYP3A1 and -3A2 (-C11, -C13), DOD to assess CYP2C6

and -2C11 (-2D2), and POD to assess CYP1A2 (-2C6)

(

Kobayashi et al. 2002, Chovan et al. 2007). Therefore, M1OH

was used to probe residual CYP3A activity, DOD to probe Disclosure statement

CYP2C activity, and POD to probe CYP1A activity ex vivo in

The authors report no conflict of interest.

rat liver microsomes, A ꢄ9% decrease in POD activity was

observed in the microsomes derived from treated compared

to control animals, while no decreases in M1OH or DOD

activity was evident. Our previous characterization of 8-BOA

inactivation of off-target human CYP isoforms showed that

CYP1A2 experienced the largest degree of time-dependent

inhibition, characterized by IC₅₀and shifted-IC₅₀values of

ORCID

John P. Kowalski

http://orcid.org/0000-0002-0343-8425

References

6

14 and 55 mM, respectively, equating to a shift ratio of 11. Al-Esawi N.S.E., et al., 2020. Expression of CYT4Z1 in breast carcinoma;

Since rat liver POD activity is driven mainly by CYP1A2, the

minor ꢄ9% decrease in treated versus control microsomes

agrees with human CYP isozyme data. This indicates that

while overall a minor process, CYP1A family enzymes may

Correlating with clinicopathological parameters. Biomedical and

pharmacology journal, 13:1.

Campbell, C.D., and Rees, C.W., 1969. Reactive intermediates. Part I.

Synthesis and oxidation of 1- and 2-aminobenzotriazole. Journal of

the chemical society, (5):742–747.

experience the largest off-target CYP inactivation liability Capper, C.P., et al., 2016. The Metabolism, analysis, and targeting of ster-

oid hormones in breast and prostate cancer. Hormones and cancer, 7:

from 8-BOA in both the human and rat.

Lastly, the results of clinical chemistries indicated that 8-

BOA treatment causes no acute liver or kidney toxicity. Given

149–164.

Chovan, J.P., et al., 2007. Cytochrome P450 probe substrate metabolism

kinetics in Sprague Dawley rats. Xenobiotica. 37:459–473.

that ABT, the bioactivatable pharmacophore of 8-BOA, has Darnell, M., and Weidolf, L., 2013. Metabolism of xenobiotic carboxylic

acids: focus on coenzyme A conjugation, reactivity, and interference

with lipid metabolism. Chemical research in toxicology, 26:1139–1155.

Davies, B., and Morris, T., 1993. Physiological parameters in laboratory

been itself dosed to numerous species and causes little tox-

icity (Ortiz de Montellano 2018), this is not a surprising result.

While the potential for acyl-glucuronides to covalently mod-

animals and humans. Pharmaceutical research, 10:1093–1095.

ify proteins is well known (Faed 1984, Regan et al. 2010), DeSantis, C.E., et al., 2019. Breast cancer statistics, 2019. CA: A cancer

xenobiotic fatty acids also can interfere with lipid metabol-

ism. This can lead to their incorporation into triglycerides,

the reduction of carnitine levels, and inhibition of b-oxida-

tion, which can all trigger toxicological effects (Darnell and

Weidolf 2013). At least in the short time scale of this study

journal for clinicians, 69(6):438–451.

Faed, E.M. 1984. Properties of acyl glucuronides: implications for studies

of the pharmacokinetics and metabolism of acidic drugs. Drug metab-

olism reviews, 15:1213–1249.

Guengerich, F.P., et al., 2009. Measurement of cytochrome P450 and

NADPH-cytochrome P450 reductase. Nature protocols, 4:1245–1251.

Hamilton, R.A., et al., 1981. Determination of mean valproic acid serum

level by assay of a single pooled sample. Clinical pharmacology &

therapeutics, 3:408–413.

(

and low dose), no indication of interference with these sys-

tems was apparent based on the results of the liver and kid-

ney function tests. Of note, both control and treated groups

showed a significant increase in aspartate aminotransferase

He, D., et al., 2019. A novel immunodeficient rat model supports human

lung cancer xenografts. The FASEB journal, 33:140–150.

XENOBIOTICA

915

Houston, J.B., 1981. Drug metabolite kinetics. Pharmacology & therapeu- Regan, S.L., et al., 2010. Acyl glucuronides: the good, the bad and the

tics, 15(3):521–552.

ugly. Biopharmaceutics & drug disposition . 31:367–395.

Iannaccone, P.M., and Jacob, H.J., 2009. Rats! Disease models & mecha- Rieger, M.A., et al., 2004. Identification of a novel mammary-restricted

nisms, 2(5-6):206–210.

Kobayashi, K., et al., 2002. Substrate specificity for rat cytochrome P450

cytochrome P450, CYP4Z1, with overexpression in breast carcinoma.

Cancer research, 64:2357–2364.

Rowland, M., and Tozer, T., 2011. Clinical pharmacokinetics and pharma-

codynamics: Concepts and Applications. 4th ed. Philadelphia: Wolters

Kluwer Health/Lippincott Williams & Wilkins.

Sadeque, A.J., et al., 1992. Stereoselective sulfoxidation by human flavin-

containing monooxygenase. Evidence for catalytic diversity between

hepatic, renal, and fetal forms. Drug metabolism and disposition, 20:

(

CYP) isoforms: screening with cDNA-expressed systems of the rat.

Biochemical pharmacology, 63:889–896.

Kowalski, J.P., et al., 2020. Design and characterization of the first select-

ive and potent mechanism-based inhibitor of cytochrome P450 4Z1.

Journal of medicinal chemistry, 63(9):4824–4836.

Li, C., et al., 2006. Acyl-coenzyme a formation of simvastatin in mouse

liver preparations. Drug metabolism and disposition, 34:102–110.

Li, Y., et al., 2017. Tumoral expression of drug and xenobiotic metaboliz-

ing enzymes in breast cancer patients of different ethnicities with

implications to personalized medicine. Scientific reports, 7:4747.

Loi, C.M., et al., 2013. Which metabolites circulate? Drug metabolism and

disposition, 41:933–951.

832–839.

Sarda, S., et al., 2012. Diclofenac metabolism in the mouse: novel in vivo

metabolites identified by high performance liquid chromatography

coupled to linear ion trap mass spectrometry. Xenobiotica. 42:

179–194.

McDonald, M.G., et al., 2017. Expression and functional characterization Smits, B.M., et al., 2007. Genetically engineered rat models for breast

of breast cancer-associated cytochrome P450 4Z1 in Saccharomyces

cerevisiae. Drug metabolism and disposition, 45:1364–1371.

Mollard, S., et al., 2011. How can grafted breast cancer models be opti-

mized? Cancer biology & therapy, 12:855–864.

Murayama, T., and Gotoh, N., 2019. Patient-derived xenograft models of

breast cancer and their application. Cells, 8(6):621.

Murray, G.I., et al., 2010. Profiling the expression of cytochrome P450 in

breast cancer. Histopathology, 57:202–211.

Noto, F.K., et al., 2018. Sprague dawley Rag2-null rats created from engi-

neered spermatogonial stem cells are immunodeficient and permis-

sive to human xenografts. Molecular cancer therapeutics, 17:

cancer. Breast disease, 28:53–61.

Stanley, K.K., and Tubbs, P.K., 1975. The role of intermediates in mito-

chondrial fatty acid oxidation. Biochemical journal, 150:77–88.

Stewart, H. B., et al., 1973. Intermediates in fatty acid oxidation.

Biochemical journal, 132:61–76.

Suga, T., 2003. Drug metabolism in peroxisomes: involvement of peroxi-

somal beta-oxidation system in the oxidative chain-shortening of

xenobiotic acyl compounds. Drug metabolism and pharmacokinetics,

18:155–162.

Vyas, K.P., et al., 1990. Biotransformation of lovastatin. I. Structure eluci-

dation of in vitro and in vivo metabolites in the rat and mouse. Drug

metabolism and disposition, 18:203–211.

Yang, J., et al., 2008. Cytochrome p450 turnover: regulation of synthesis

and degradation, methods for determining rates, and implications for

the prediction of drug interactions. Current drug metabolism, 9:

2

481–2489.

Obach, R.S., et al., 2007. Mechanism-based inactivation of human cyto-

chrome p450 enzymes and the prediction of drug-drug interactions.

Drug metabolism and disposition, 35:246–255.

Ortiz de Montellano, P.R., 2018. 1-aminobenzotriazole: a mechanism-

based cytochrome P450 inhibitor and probe of cytochrome P450 biol-

ogy. Medicinal chemistry, 8(3):038.

384–394.

Yang, X., et al., 2017. CYP4Z1 - a human cytochrome P450 enzyme that

might hold the key to curing breast cancer. Current pharmaceutical

design, 23:2060–2064.

Radvanyi, L., et al., 2005. The gene associated with trichorhinophalangeal

syndrome in humans is overexpressed in breast cancer. Proceedings of

the national academy of sciences of the United States of America, 102: Zhang ZY and Wong YN. 2005. Enzyme kinetics for clinically relevant

1005–11010. CYP inhibition. Current drug metabolism, 6:241–257.

1

Article Doi

Pharmacokinetics, metabolism and off-target effects in the rat of 8-[(1H- benzotriazol-1-yl)amino]octanoic acid, a selective inhibitor of human cytochrome P450 4Z1: β-oxidation as a potential augmenting pathway for inhibition

DOI: 10.1080/00498254.2021.1930281

Source and publish data:

Authors:

Article abstract of DOI:10.1080/00498254.2021.1930281

Full text of DOI:10.1080/00498254.2021.1930281

Products guided by the article

R&D Labs maybe for 929-48-6

Relevant to this article

Hot Product