Cytochrome P450 and flavin-containing monooxygenase enzymes are responsible for differential oxidation of the anti-thyroid-cancer drug vandetanib by human and rat hepatic microsomal systems

Journal Pre-proof Cytochrome P450 and flavin-containing monooxygenase enzymes are responsible for differential oxidation of the anti-thyroid-cancer drug vandetanib by human and rat hepatic microsomal systems ´ Kateˇrina Jaklov ´ ´ Radek Indra, Petr Pompach, Katar´ına Vavrova, a, ˇ Heger, Vojtech ˇ Adam, Tomaˇ ´ s Eckschlager, Kateˇrina Zbynek ´ Volker Manfred Arlt, Marie Stiborova´ Kopeˇckova,

PII:

S1382-6689(19)30185-1

DOI:

https://doi.org/10.1016/j.etap.2019.103310

Reference:

ENVTOX 103310

To appear in:

Environmental Toxicology and Pharmacology

Received Date:

2 August 2019

Revised Date:

24 November 2019

Accepted Date:

2 December 2019

´ Please cite this article as: Indra R, Pompach P, Vavrova´ K, Jaklov a´ K, Heger Z, Adam V, Eckschlager T, Kopeˇckova´ K, Arlt VM, Stiborova´ M, Cytochrome P450 and flavin-containing monooxygenase enzymes are responsible for differential oxidation of the anti-thyroid-cancer drug vandetanib by human and rat hepatic microsomal systems, Environmental Toxicology and Pharmacology (2019), doi: https://doi.org/10.1016/j.etap.2019.103310

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Cytochrome P450 and flavin-containing monooxygenase enzymes are responsible for differential oxidation of the anti-thyroid-cancer drug vandetanib by human and rat hepatic microsomal systems

Radek Indraa, Petr Pompacha, Katarína Vavrováa, Kateřina Jáklováa, Zbyněk Hegerb, Vojtěch Adamb, Tomáš Eckschlagerc, Kateřina Kopečkovád, Volker Manfred Arlte, Marie

a

ro of

Stiborováa*

Department of Biochemistry, Faculty of Science, Charles University, Albertov 2030, 128 40

Department of Chemistry and Biochemistry, Mendel University in Brno, Zemedelska 1, 61300

re

b

-p

Prague 2, Czech Republic

Brno, Czech Republic

Department of Pediatric Hematology and Oncology, 2nd Medical Faculty, Charles University and

lP

c

University Hospital Motol, 150 06 Prague, Czech Republic Department of Oncology, 2nd Faculty of Medicine, Charles University and University Hospital

na

d

Motol, 150 06 Prague 5, Czech Republic

Department of Analytical, Environmental and Forensic Sciences, MRC-PHE Centre for

ur

e

Jo

Environment and Health, King’s College London, 150 Stamford Street, London SE1 9NH, United Kingdom

*Corresponding author: Marie Stiborová, Department of Biochemistry, Faculty of Science, Charles University, Albertov 2030, 128 40 Prague 2, Czech Republic; Tel.: +420-221951285; Fax: +420-221951283; E-mail: [email protected]

1

Highlights The anti-thyroid-cancer drug vandetanib is oxidized by rat liver enzyme systems



N-desmethylvandetanib and vandetanib N-oxide are formed during the enzyme oxidation



NADPH and NADH mediate the CYP- and FMO-catalyzed vandetanib oxidative reactions



Rat liver CYP2C11>>3A1 and FMO1 form N-desmethylvandetanib and N-oxide, respectively



Mechanisms of different vandetanib oxidation by rat and human systems are explained

ro of



ABSTRACT

We studied the in vitro metabolism of the anti-thyroid-cancer drug vandetanib in a rat animal

-p

model and demonstrated that N-desmethylvandetanib and vandetanib N-oxide are formed by NADPH- or NADH-mediated reactions catalyzed by rat hepatic microsomes and pure

re

biotransformation enzymes. In addition to the structural characterization of vandetanib

lP

metabolites, individual rat enzymes [cytochrome P450 (CYP) and flavin-containing monooxygenase (FMO)] capable of oxidizing vandetanib were identified. Generation of N-

na

desmethylvandetanib, but not that of vandetanib N-oxide, was attenuated by CYP3A and 2C inhibitors while inhibition of FMO decreased formation of vandetanib N-oxide. These results

ur

indicate that liver microsomal CYP2C/3A and FMO1 are major enzymes participating in the formation of N-desmethylvandetanib and vandetanib N-oxide, respectively. Rat recombinant

Jo

CYP2C11>>3A1>3A2>1A1>1A2>2D1>2D2 were effective in catalyzing the formation of Ndesmethylvandetanib. Results of the present study explain differences between the CYP- and FMO-catalyzed vandetanib oxidation in rat and human liver reported previously and the enzymatic mechanisms underlying this phenomenon.

Abbreviations: -NF, -naphthoflavone; BaP, benzo[a]pyrene; bw, body weight; CYP, cytochrome P450; DDTC, diethyldithiocarbamate; DMSO, dimethyl sulfoxide; EGFR, epidermal

2

growth factor receptor; FMO, flavin-containing monooxygenase, HPLC, high performance liquid chromatography; KC, ketoconazole; LQTS, long QT syndrome; PB, phenobarbital; PCN, pregnenolone-16-carbonitrile; POR, NADPH:cytochrome P450 oxidoreductase; RET, rearranged during transfection; r.t., retention time; TIE2, tyrosine kinase with immunoglobulin and EGF domains-2; TK, tyrosine kinase; TKI, tyrosine kinase inhibitor; VEGFR-2, vascular endothelial growth factor receptor.

Cytochromes P450; Flavin-containing monoxygenases.

-p

1. Introduction

ro of

Keywords: Vandetanib; Anti-thyroid-cancer drug; Tyrosine kinase inhibitor; Metabolism;

Vandetanib (Caprelsa, N-(4-bromo-2-fluorophenyl)-6-methoxy-7-[(1-methylpiperidin-4-

re

yl)methoxy]quinazolin-4-amine; Fig. 1) is an anticancer drug acting as tyrosine kinase inhibitor

lP

(TKI). It inhibits vascular endothelial growth factor receptor 2 (VEGFR-2), epidermal growth factor receptor (EGFR) and rearranged during transfection (RET) tyrosine kinase activity (Ciardiello et al., 2003; Commander et al., 2011; Hennequin et al., 2002; Martin et al., 2011;

na

Wedge et al., 2002). Further, it also shows activity towards protein tyrosine kinase 6 (BRK), tyrosine kinase with immunoglobulin and EGF domains-2 (TIE2), members of the ephrin receptor

ur

kinase and SRC tyrosine kinase families (Vozniak et al., 2012). Therefore, vandetanib is

Jo

considered a multiple TKI. Its administration reduces tumor cell-induced angiogenesis, tumor vessel permeability, and inhibits tumor growth and metastasis in mouse models of cancer (Vozniak et al., 2012). These activities result from their major targets; VEGFR-2, EGFR and RET, because they all are involved in signaling pathways promoting angiogenesis and tumor growth (Ferrara and Davis-Smyth, 1997; Kowanetz and Ferrara, 2006; Martin et al., 2011; Wells, 1999). Vandetanib is used for the treatment of symptomatic and/or progressive medullary thyroid cancer, probably because the RET tyrosine kinase mutation occurs in this type of thyroid cancer

3

(Dvořáková et al., 2008; Hadoux et al., 2016, 2017; Václavíková et al., 2009; Vozniak et al., 2012). Treatment with vandetanib is usually well-tolerated by patients but is contraindicated in people with congenital long QT syndrome (LQTS). In addition, various side effects such as diarrhea, rash and/or nausea can occur (FDA, 2011; Martin et al., 2012). In cancer chemotherapy, serious clinical consequences may occur from small alterations in drug metabolism affecting drug pharmacokinetics (Fujita, 2006). Although in general there is only little known about the metabolism of TKIs, the metabolism of vandetanib has been investigated

ro of

previously mainly in studies carried out by AstraZeneca (Australian Public Assessment Report for Vandetanib, 2013; Bates, 2003; Martin et al., 2011, 2012; Vozniak et al., 2012) but recently also by others (Attwa et al., 2018; Indra et al., 2019). The metabolic products formed by vandetanib in

-p

humans have been shown to be the same as those observed in animal models such as rat, mouse and dog. N-desmethylvandetanib and vandetanib N-oxide (Fig. 1) were determined as major

re

metabolites in excreta of humans and experimental animals (Martin et al., 2011, 2012). However,

lP

a recent comparison using hepatic microsomal subcellular systems showed that levels of the major vandetanib metabolites formed in rat, rabbit and mouse animal models differ significantly from those of human (Indra et al., 2019).

na

In contrast to the situation in humans, where studies have determined enzymes participating in the formation major vandetanib metabolites (Australian Public Assessment Report for

ur

vandetanib, 2013; Indra et al., 2019; Martin et al., 2011), little information is available on enzymes

Jo

involved in vandetanib metabolism in experimental animals. In previous studies cytochrome P450 (CYP) and flavin-containing monooxygenase (FMO) enzymes have been demonstrated to be key enzymes in the metabolism of vandetanib (Australian Public Assessment Report for vandetanib, 2013; Indra et al., 2019; Martin et al., 2011). Whereas hepatic CYP3A4 is considered to be the major enzyme oxidizing vandetanib to N-desmethylvandetanib, FMOs expressed in kidney (FMO1) and liver (FMO3) (Fig. 1) are suggested to be responsible for the formation of vandetanib N-oxide. Indeed, we recently (Indra et al., 2019) fully confirmed these suggestions and further

4

elucidated the mechanism how human CYP3A4 catalyzes with high efficiency the Ndemethylation of vandetanib (i.e. N-desmethylvandetanib formation). However, although knowledge of human enzymes involved in vandetanib metabolism is important to guide novel treatment regimens that can increase pharmacological efficacy of vandetanib, the utilization of animal models is still required for metabolic studies that cannot be carried in human patients. The latter studies for example can include the investigation of combination effects of vandetanib with other drugs (anticancer drugs), which can induce or inhibit biotransformation enzymes known to

ro of

metabolize vandetanib and thereby influence the pharmacological (anti-cancer) efficacy of vandetanib. This can also include studies aiming to test novel application forms of vandetanib, such as nanotransporters with unknown nanotransporter-related toxicity in humans, which may

-p

help to increase its therapeutic efficiency. This is a valuable reason because oncology patients frequently take a wide range of other drugs in addition to their therapy for cancer, possibly

re

including compounds that can induce or inhibit enzymes metabolizing anticancer drugs.

lP

In this work we studied the in vitro metabolism of vandetanib in subcellular systems (microsomes) isolated from rat liver expressing a variety of enzymes metabolizing different drugs and other xenobiotics as well as by rat recombinant enzymes. Results obtained were compared to

na

those previously obtained in human systems. Further, we elucidated the mechanism responsible for differences in the vandetanib metabolite profiles observed between human and rat subcellular

Jo

ur

enzymatic systems.

2. Material and methods 2.1. Chemicals and material Vandetanib was from LC Laboratories (Woburn, MA, USA). Benzo[a]pyrene (BaP), NADPH, NADH, phenobarbital (PB), pregnenolone-16α-carbonitrile (PCN) and other chemicals were all purchased from Sigma-Aldrich (St. Louis, MO, USA). The purity of all chemicals met standards set by the American Chemical Society, unless noted otherwise. All animal experiments were

5

conducted as reported (Indra et al., 2019), in accordance with the Regulations for the Care and Use of Laboratory Animals (311/1997, Ministry of Agriculture, Czech Republic), which follows the Declaration of Helsinki. Briefly, male Wistar rats (~125–150 g; AnLab, Czech Republic), male C57BL/6 mice (5–8 g; CXR Bioscience Ltd., Dundee, UK), and male New Zealand rabbits (2.5–3 kg; AnLab, Czech Republic) were employed as animal models. Rats, rabbits and mice used to prepare microsomal fraction from untreated animals were sacrificed by cervical dislocation, livers snap-frozen in liquid nitrogen and stored at ‒80°C until further processing (see below). For

ro of

enzyme induction male Wistar rats (~125-150 g; five weeks old) were treated with BaP (mainly CYP1A), PB (mainly CYP2B), ethanol (mainly CYP2E1) and PCN (mainly CYP3A) as follows: 1) Wistar rats were treated by gastric gavages with a single dose of 150 mg/kg body weight (bw)

-p

BaP dissolved in sunflower oil (1 ml) as reported (Hodek et al., 2013), rats used as controls

received 1 ml of sunflower oil only. 2) Wistar rats were treated with PB (0.1% in drinking water

re

for 6 days) as reported (Hodek et al., 2005), rats used as controls received drinking water only. 3)

lP

Wistar rats were treated with ethanol (0.1% in drinking water for 6 days) as described previously (Yang et al., 1985), rats used as controls received drinking water only. 4) Wistar rats were injected i.p. with 50 mg/kg bw PCN dissolved in maize oil (1 ml) once daily for four consecutive days as

na

reported (Chirulli et al., 2005; Stiborová et al., 2006), rats used as controls received 1 ml of maize oil only. In all treatment groups rats were sacrificed 24 h after the last treatment by cervical

ur

dislocation. For all treatment groups and controls, livers of the animals were removed immediately

Jo

after sacrifice, divided into small pieces (~1 g), snap-frozen in liquid nitrogen, and stored at ‒80°C until isolation of microsomal fractions. Pooled microsomes were used for all in vitro experiments in the present study and prepared from 10 livers/group for all animal species following established protocols (Hodek et al., 2013; Stiborová et al., 2003). Activities of CYP marker substrates in control microsomes did not differ significantly between solvents used. Microsomal fractions were stored at 80°C until analysis. Protein concentrations in the microsomal fractions were assessed using the bicinchoninic acid protein assay with bovine serum albumin as a standard (Wiechelman

6

et al., 1988). Rat recombinant enzymes were used in the forms of Supersomes and obtained from Gentest Corp. (Woburn, MA, USA). In Supersomes microsomal fractions were isolated from insect cells that are transfected with baculovirus constructs containing cDNA of rat CYP enzymes (CYP1A1/2, 2A1/2, 2B1, 2C6/11/12/13, 2D1/2, 2E1, 3A1/2). These Supersomes also express NADPH:CYP oxidoreductase (POR). However, because they are microsomes (particles of broken endoplasmic reticulum), other enzymes (proteins) of the endoplasmic reticulum membrane like NADH:cytochrome b5 reductase and cytochrome b5 are also expressed at basal levels in these

ro of

Supersomes. In addition, we utilized Supersomes which over-expressed cytochrome b5, in a molar ratio of CYP to cytochrome b5 of 1 to 5. In Supersomes where cytochrome b5 was not over-expressed (see above) pure cytochrome b5 was added to reach a molar ratio of CYP to

-p

cytochrome b5 of 1 to 5. Reconstitution of Supersomes with purified cytochrome b5 was

re

performed as described previously (Kotrbová et al., 2011; Stiborová et al., 2012a, 2016; Šulc et al., 2016). Human systems were employed for comparison as described (Indra et al., 2019).

lP

Briefly, this included male human hepatic microsomes (pooled sample; LOT #3043885) and human recombinant FMOs (i.e. FMO1 and FMO3) in the forms of Supersomes from Gentest

na

Corp.

ur

2.2. Oxidation of vandetanib by hepatic microsomes and CYP enzymes Unless stated otherwise, incubation mixtures used to study vandetanib metabolism contained the

Jo

following in a final volume of 500 μl for incubations containing hepatic microsomes and 250 μl for incubations with Supersomes: 100 mM potassium phosphate buffer (pH 7.4), 1 mM NADPH or NADH or NADPH with NADH, human, rat, rabbit and mouse hepatic microsomes (0.25 mg protein) (500 μl), or rat recombinant CYPs in Supersomes (100 pmol) (250 μl) or human recombinant FMO1 and FMO3 in Supersomes (100 pmol) and 50 µM vandetanib dissolved in 5 µl dimethyl sulfoxide (DMSO). The reaction was initiated by adding vandetanib. In control incubations, either microsomes or CYP or FMO or NADPH or NADH or vandetanib were 7

omitted. Incubations were performed at 37C for 20 min in open plastic Eppendorf tubes; vandetanib oxidation was linear up to 30 min of incubation. After incubation, 5 µl of 1 mM phenacetine (dissolved in methanol) was added as internal standard. The reaction was stopped by extraction with dichlormethane (2 × 1 ml). Extracts were evaporated, dissolved in 25 µl methanol and high-performance liquid chromatography (HPLC) analysis was used to separate vandetanib and its metabolites. HPLC conditions were as follows: 5 mm Ultrasphere ODS Beckman, 4.6 × 250 mm preceded by a C-18 guard column; the eluent was 0,5% v/v triethylamin in water (pH 3)

ro of

containing 30% acetonitrile with a flow rate of 1 ml/min, detection was at 254 nm. Vandetanib metabolites separated by HPLC were characterized by mass spectroscopy (see further details

below). Up to two vandetanib metabolites were detected with retention times (r.t.) of 7.6 and 8.9

-p

min, corresponding to N-desmethylvandetanib and vandetanib N-oxide, respectively (Fig. 2).

re

Recoveries of vandetanib metabolites were approximately 95%.

lP

2.3. Identification of vandetanib metabolites by mass spectrometry Vandetanib metabolites were identified by direct infusion of the diluted sample in 50% acetonitrile in water plus 1% acetic acid into a 12T solariX XR FT-ICR mass spectrometer (Bruker Daltonics,

na

Bremen, Germany). The mass spectrometer was operated in positive ion mode. Calibration of the instrument was performed using 1% solution of sodium trifluoracetic resulting in accuracy below

ur

2 ppm. Data acquisition and data processing were performed by ftmsControl 2.1.0 and

Jo

DataAnalysis 4.2 (Bruker Daltonics).

2.4. Inhibition studies Inhibition studies in rat liver microsomes were essentially conducted as reported previously (Stiborová et al., 2012b). The inhibitors employed were as follows: -Naphthoflavone (-NF), which inhibits CYP1A1 and 1A2; diamantane, which inhibits CYP2B; sulfaphenazole, which inhibits CYP2C; quinidine, which inhibits CYP2D; diethyldithiocarbamate (DDTC), which

8

inhibits CYP2E1 and CYP2A; and ketoconazole (KC), which inhibits CYP3A and methimazole, which inhibits FMOs. Values of IC50 using 0.05–500 M concentrations of inhibitors were also determined by the procedure described previously (Stiborová et al., 2006). Inhibitors were dissolved in 5 l DMSO, except for quinidine and -naphthoflavone which were dissolved in methanol, and except of DDTC which was dissolved in water, yielding a final concentration of 50 M in the incubation mixtures (equimolar concentrations of individual inhibitors with those of 50 M vandetanib). Inhibitors were incubated at 37C for 10 min with 50 M vandetanib prior to the

ro of

addition of NADPH. Then mixtures were incubated for further 20 min at 37C. For inhibition of FMO enzymes, which have a greater temperature sensitivity than CYPs (Schlenk 1998; Schlenk et al., 1993), microsomes were heated to 45C for 10 min first and then added to the incubation

-p

mixture to study vandetanib oxidation. Formation of vandetanib metabolites was analyzed by

re

HPLC as described above.

lP

2.5. Contributions of CYP enzymes to N-demethylation of vandetanib to N-desmethylvandetanib in rat livers

na

In order to calculate the contribution of individual CYP enzymes to vandetanib oxidation (i.e. formation of N-desmethylvandetanib) in rat livers, we utilized the velocities of vandetanib

ur

oxidation to N-desmethylvandetanib by Supersomal rat CYP enzyme systems in combination with the expression levels of CYPs in rat livers as reported in the scientific literature (Nedelcheva and

Jo

Gut, 1994; Večeřa et al.. 2011; Yamazaki et al., 2006; Zachařová et al., 2012). Contributions of these enzymes were calculated by relative activity factor because the activities of CYPs in Supersomes™ also need to be considered in addition to the relative contents of CYPs in the livers. Therefore, the contribution of each rat CYP enzyme that oxidizes vandetanib in rat livers was calculated by dividing the relative activity of each CYP oxidizing vandetanib [r.a.cypi] (rate of vandetanib oxidation to N-desmethylvandetanib) multiplied by the amounts of this CYP in rat liver, by the total relative activities (∑[r.a.cypi]) of all CYPs oxidizing this substrate. CYPs of the

9

2C subfamily (CYP2C6, 2C11, 2C12 and 2C13) are major enzymes expressed in rat livers accounting for ~55% of the CYP complement (Nedelcheva and Gut, 1994), with CYP2C11 (~50%) and 2C6 (~20%) contributing most to the CYP2C complement (Večeřa et al.. 2011; Yamazaki et al., 2006; Zachařová et al., 2012). Of the other CYPs, CYP2E1 (~15%), 3A (~10%), 2D (~7%), 2A (~6%), 2B (~5%) and 1A (~2%) enzymes are also present in rat livers (Nedelcheva and Gut, 1994).

ro of

2.6. Statistical analysis Data are expressed as mean ± SD. Data was analyzed using GraphPad Prism 7 using ANOVA with post-hoc Tukey HSD Test. All P-values are two-tailed and considered significant at the 0.05

-p

level.

re

3. Results

lP

3.1. Oxidation of vandetanib by rat hepatic microsomes

The efficiencies of rat hepatic microsomal subcellular systems to oxidize vandetanib were analyzed. Vandetanib was oxidized to two metabolites that were separated by HPLC (Fig. 2a) and

na

identified by mass spectrometry as N-desmethylvandetanib and vandetanib N-oxide (Fig. 2b). As we have shown previously (Indra et al., 2019), the formation of vandetanib metabolites in hepatic

ur

microsomes was dependent on NADPH which serves as cofactor for both POR-mediated CYP

Jo

catalysis and FMO-mediated oxidative reactions (Gao et al., 2018; Guengerich, 2018; Itoh et al., 1993; Krueger and Williams, 2005; Poulsen and Ziegler, 1995; Ziegler, 2002). This demonstrated that CYPs and/or FMOs are responsible for the formation of these metabolites in the subcellular systems. However, the profile of vandetanib metabolites formed in rat hepatic microsomes (Fig. 3) differed to that in human microsomes (Indra et al., 2019). Besides NADPH, in the present study we also investigated the ability of another nicotinamide oxidoreductase cofactor, NADH, to mediate vandetanib oxidation reactions. In the presence of

10

NADH instead of NADPH, rat hepatic microsomes also oxidized vandetanib to Ndesmethylvandetanib and vandetanib N-oxide. NADH was also effective to mediate the oxidation of vandetanib in human liver microsomes (Fig. 3). These results are consistent with other studies showing that NADH is capable of mediating the oxidation of several substrates of microsomal CYP enzymes (Kotrbova et al., 2011; Reed et al., 2018, 2019; Stiborova et al., 2016, 2017; Yamazaki et al., 1996). However, formation of N-desmethylvandetanib by rat and human microsomal subcellular systems was up to 8-fold lower when using NADH than NADPH as

ro of

cofactor (Fig. 3). The presence of both nicotinamide cofactors (NADPH together with NADH) in the incubation mixture containing rat microsomes resulted in a small 1.2-fold increase in

vandetanib oxidation to N-desmethylvandetanib compared to incubations with NADPH alone.

-p

However, no such effect was found in human microsomes (Fig. 3).

Formation of vandetanib N-oxide by rat and human hepatic microsomes was up to 2.5-times

re

lower when using NADH than NADPH as cofactor (Fig. 3). Although mammalian microsomal

lP

FMOs are considered to be NADPH-dependent enzymes, it has previously been demonstrated that NADPH can be partially replaced by NADH (Poulsen and Ziegler, 1995; Ziegler, 2002; Ziegler and Pettit, 1966). For example, oxidation of N,N-dimethyaniline to dimethylaniline N-oxide is

na

mediated both by NADPH and NADH (Ziegler and Pettit, 1966). Therefore, our results confirmed the participation of FMOs in the NADH-mediated oxidation of vandetanib to its N-oxide in human

ur

and rat hepatic microsomes. Nevertheless, in order to confirm the formation of vandetanib N-oxide

Jo

by microsomes in the presence of NADH instead of NADPH further, we also tested hepatic microsomes from two other model animals, rabbits and mice, to catalyze this reaction. Results obtained with rabbit and mouse liver microsomes were similar to those found for the human and rat hepatic microsomes (Fig. 3). When both NADPH and NADH were present in the incubation mixture containing microsomes formation of vandetanib N-oxide decreased up to 4-fold compared to incubations with NADPH alone (Fig. 3). Without NADPH or NADH, essentially no oxidation of vandetanib was detected in all tested hepatic microsomes (see Fig. 2a for rat microsomes).

11

Although the function of human CYP and FMO biotransformation enzymes in vandetanib oxidation was identified previously (Indra et al., 2019), no such information is available for the corresponding rat enzymes. In order to identify individual rat hepatic CYP and/or FMO enzymes oxidizing vandetanib and to estimate their contribution to the oxidation process, three approaches were utilized: (i) use of selective CYP and FMO inhibitors in rat microsomes; (ii) use of specific inducers of individual CYPs in rats; and (iii) use of recombinant CYP and FMO enzymes. In addition, levels of N-desmethylvandetanib formation by individual rat recombinant CYPs were

ro of

related to CYP enzyme expression levels in rat livers in order to evaluate the CYP contributions to this reaction in this organ.

-p

3.2. The effects of CYP and FMO enzyme inhibitors on vandetanib oxidation in rat liver microsomes

re

The role of hepatic enzymes in vandetanib oxidation was investigated by modulating the enzyme-

lP

catalyzed reactions using selective inhibitors of individual CYPs and FMOs. Two approaches were utilized in the inhibition studies: (i) evaluation of the % inhibition at inhibitor concentrations equimolar to that of the substrate (vandetanib); and (ii) determination of the IC50 values for

na

individual inhibitors. Inhibitors α-NF, diamantane, sulfaphenazole, quinidine, DDTC, ketoconazole, and methimazole were utilized as inhibitors of CYP1A, 2B, 2C, 2D, 2E1, 3A, and

ur

FMOs, respectively (Table 1). Ketoconazole was the most efficient in inhibiting the oxidation of

Jo

vandetanib to N-desmethylvandetanib, decreasing its formation by 94% in rat microsomes at 50 µM (equimolar) concentration (Table 1). The IC50 values for ketoconazole equaled to 2 µM in rat microsomes (Table 1). Sulfaphenazole inhibited the formation N-desmethylvandetanib by 40% in rat microsomes when used at concentrations equimolar to vandetanib (Table 1). The IC50 values for sulfaphenazole could not be exactly determined because even when tested at concentrations of 500 µM, observed inhibition was less than 50%. Inhibitors of other CYPs (α-NF, diamantane, quinidine, and DDTC) showed no inhibitory effects on vandetanib oxidation in rat microsomes

12

(Table 1). None of the tested inhibitors of CYPs was capable of inhibiting the formation of vandetanib N-oxide. These results demonstrate that CYP enzymes do not catalyze the oxidation of vandetanib to its N-oxide and that CYP2C and 3A enzymes are efficient in catalyzing the oxidation of vandetanib to N-desmethylvandetanib in rat hepatic microsomes. To estimate the contribution of enzymes oxidizing vandetanib to its N-oxide that are not sensitive to the inhibition by CYP inhibitors, the FMO inhibitor methimazole (Ballard et al., 2007; Gao et al., 2018) was used. Formation of vandetanib N-oxide was inhibited by 77% by

ro of

methimazole in rat hepatic microsomes, when used at concentrations equimolar to vandetanib, producing the IC50 value of 6 µM (Table 1). Therefore, FMOs, predominantly FMO1 as it is

highly expressed in rat livers (Itoh et al., 1993; Krueger et al., 2005), seems to be responsible for

-p

the oxidation of vandetanib to vandetanib N-oxide in rat liver microsomes.

In order to further resolve the contribution of CYP and FMO enzymes to vandetanib

re

metabolism, another experiment was employed. Since FMOs exhibit greater temperature

lP

sensitivity than CYPs (Schlenk, 1998; Schlenk et al., 1993), we also used heated microsomes to study the formation of vandetanib metabolites by rat hepatic microsomes. Only the formation of vandetanib N-oxide, but not that of N-desmethylvandetanib, was attenuated (to 2% of control) by

na

this procedure. Similar results were found for human hepatic microsomes, but the decrease in the vandetanib N-oxide formation was less pronounced (to 35% of control). These results support the

ur

suggestion that in the hepatic microsomes tested, N-desmethylvandetanib and vandetanib N-oxide

Jo

are generated by CYPs and FMOs, respectively.

3.3. The effects of CYP enzyme inducers on vandetanib oxidation in rats We pretreated rats with inducers of several CYPs: BaP as an inducer of CYP1A1/2 and 1B1; PB as an inducer of CYP2B and 2C; ethanol as an inducer of CYP2E1; and PCN as an inducer of CYP3A) (Stiborová et al., 2006). All hepatic microsomes isolated from these rats were capable of oxidizing vandetanib to its metabolites N-desmethylvandetanib and vandetanib N-oxide, but their

13

amounts and profiles differed depending on the inducers used (Fig. 4). Under the experimental conditions used, rat microsomes oxidized vandetanib to N-desmethylvandetanib most efficiently when those were isolated from livers of rats pretreated with PCN (rich in CYP3A), followed by those isolated from livers of PB-pretreated rats (rich in CYP2B and 2C) and those isolated from livers of control (uninduced) rats in which CYP2C enzymes are highly expressed. Microsomes from BaP-pretreated rats (rich in CYP1A and 1B1) and ethanol-pretreated rats were also capable of oxidizing vandetanib, but to a lesser extent. The latter microsomes formed N-

ro of

desmethylvandetanib at amounts up to one order of magnitude lower than those isolated from PCN-pretreated rats (Fig. 4). These findings suggest that CYPs of the 3A subfamily, induced in PCN-microsomes, followed by those of the 2C subfamily, highly expressed in microsomes from

-p

uninduced rats, play an important role in the oxidation of vandetanib to N-desmethylvandetanib. In contrast to the formation of N-desmethylvandetanib, vandetanib N-oxide was mainly formed

re

by microsomes of control (uninduced) rats, at amounts that were almost 4-fold higher than in

lP

microsomes of PCN-pretreated rats. Negligible amounts of vandetanib N-oxide were formed by microsomes of PB-pretreated rats. Other rat hepatic microsomes tested (i.e. BaP- and ethanolpretreated rats) oxidized vandetanib into vandetanib N-oxide at lower amounts than those of

na

control rats (P<0.001) (Fig. 4). The results suggest that FMO1, highly expressed in rat liver (Baráčková, 2019; Itoh et al., 1993), seems to be responsible for its formation which is in line with

Jo

ur

previous results testing human FMO1 (Indra et al., 2019; Martin et al., 2011).

3.4. Oxidation of vandetanib by recombinant CYP and FMO enzymes The use of recombinant CYP enzymes expressed in Supersomes™ in combination with their reductases, POR and NADH:cytochrome b5 reductase, was another approach to examine the activity of individual rat CYP enzymes to oxidize vandetanib. Under the experimental conditions used, several rat CYPs tested were capable of oxidizing vandetanib to N-desmethylvandetanib (Fig. 5) but not to vandetanib N-oxide. In the presence of NADPH, seven of the rat CYP enzymes

14

tested oxidized vandetanib to N-desmethylvandetanib, namely, CYP2C11 >> 3A1 > 3A2 > 1A1 > 1A2 > 2D1 and > 2D2, while CYP2A1, 2A2, 2B1, 2C6, 2C12, 2C13 and 2E1 were ineffective (Fig. 5A). Rat CYP2C11 was the most efficient enzyme oxidizing vandetanib to Ndesmethylvandetanib, being up to 5-fold more efficient than other CYPs (i.e. CYP3A1, 3A2, 1A1, 1A2, and 2D1/2) (Fig. 5A). Addition of NADH instead of NADPH as cofactor to the incubation mixtures resulted in up to 10-fold lower levels of N-desmethylvandetanib generated by CYP1A1, 2C11, 2D1, 2D2, 3A1 and 3A2 in the CYP-mediated systems. When rat CYP1A2 was tested in the

ro of

presence of NADH and cytochrome b5, formation of N-desmethylvandetanib was undetectable (Fig. 5B). When NADH was used together with NADPH in the incubation mixture, CYP2C11 and 1A2 produced up to 1.2-times higher levels of N-desmethylvandetanib, whereas no such effect was

-p

seen by the other rat CYPs tested (Fig. 5B).

Formation of vandetanib N-oxide was further studied by utilizing human recombinant FMOs,

re

FMO1 and FMO3, because no pure rat FMOs were available. Using NADPH as a cofactor these

lP

two FMOs were previously shown to oxidize vandetanib to its N-oxide (Indra et al., 2019). In the present study we found that NADH is also capable of catalyzing this reaction (Fig. 5C). Whereas FMO3 was less efficient to form this metabolite in the presence of NADH instead of NADPH,

na

FMO1, the enzyme highly expressed in rat liver, was 1.25-fold more effective in this reaction (Fig.

ur

5C). FMOs did not oxidize vandetanib to N-desmethylvandetanib.

Jo

3.5. Contributions of individual rat CYPs to oxidation of vandetanib to N-desmethylvandetanib in rat livers

Based on the results showing the velocities of vandetanib oxidation to N-desmethylvandetanib in experimental systems containing recombinant rat CYP enzymes in Supersomes™ (Fig. 5) and the relative amounts of CYP enzymes expressed in rat livers (Nedelcheva and Gut, 1994; Večeřa et al.. 2011; Yamazaki et al., 2006; Zachařová et al., 2012), we estimated the contribution of individual CYPs to this reaction in rat livers. In the presence of NADPH, the highest contribution

15

to vandetanib oxidation to N-desmethylvandetanib in rat liver was attributed to CYP2C11 (~79%), followed by CYP3A (~20.3%), CYP2D and 1A (less than 0.3%) subfamilies (Fig. 6).

4. Discussion In the present study we characterized the in vitro metabolism of vandetanib in hepatic subcellular system (microsomes) of a rat animal model and showed that the metabolic profile generated in the rat and human systems differed. This is an important phenomenon where the underlying

ro of

mechanism should be explained. Although it was previously suggested that the same metabolism of vandetanib occurs in rat and human (Australian Public Assessment Report for Vandetanib, 2013; Martin et al., 2011), the CYP- and FMO-generated amounts of individual vandetanib

-p

metabolites differ significantly between the human and rat experimental systems.

re

N-desmethylvandetanib and vandetanib N-oxide were formed by both NADPH- and NADHmediated reactions in rat hepatic microsomes in vitro and CYP and FMO enzymes participating in

lP

their formation were identified. Several approaches were utilized in the present study to identify individual rat CYP and FMO enzymes oxidizing vandetanib. We first employed selective CYP

na

and FMO inhibitors in rat microsomes. Whereas ketoconazole, an inhibitor of CYP3A, was the strongest inhibitor of the oxidation of vandetanib to N-desmethylvandetanib, sulfaphenazole,

ur

which inhibits CYP2C, was less efficient. Inhibitors of other CYPs, α-NF, diamantane, quinidine, and DDTC, inhibiting CYP1A, 2B, 2D, and 2E1, respectively, had no effect. Although, none of

Jo

tested CYP inhibitors was capable of inhibiting the formation of vandetanib N-oxide, its formation was strongly decreased by the FMO inhibitor methimazol (Ballard et al., 2007; Gao et al., 2018), which was in turn not able to inhibit the formation of N-desmethylvandetanib. Since FMO1 is the major FMO enzyme expressed in rat liver (Itoh et al., 1993; Krueger et al., 2005), this enzyme seems to be responsible for formation of vandetanib N-oxide in rat hepatic microsomes. This result corresponds to previous findings showing that the human FMO1 enzyme was the most efficient FMO to oxidize vandetanib to its N-oxide (Indra et al., 2019). Participation of this FMO in the

16

formation of vandetanib N-oxide in rat hepatic microsomes was also supported by the attenuation of its formation when heated microsomes were used which results from the greater temperature sensitivity of FMOs, a typical property of FMOs (Schlenk 1998; Schlenk et al., 1993). The inhibition effect of vandetanib oxidation to its N-oxide at higher temperature was also seen in human hepatic microsomes. With regards to the inhibitors used in the present study, it is important to point out that data found with inhibitors are sometimes difficult to interpret because inhibitors can act more

ro of

efficiently with one enzyme substrate than another. We therefore employed an additional experimental approach by utilizing hepatic microsomes of rats that were pre-treated with inducers specific for individual CYP enzymes. Microsomes isolated from the livers of rats pretreated with

-p

PCN (an inducer of CYP3A) and microsomes of uninduced rats rich in the CYP2C subfamily were most efficient in catalyzing the oxidation of vandetanib to N-desmethylvandetanib, which

re

supported the results of the inhibition studies. Employing recombinant CYP and FMO enzymes to

lP

study the oxidation of vandetanib to N-desmethylvandetanib and vandetanib N-oxide confirmed the findings that CYP2C11 and FMO1 enzymes play a major role in the reactions producing Ndesmethylvandetanib and vandetanib N-oxide, respectively.

na

In this study we found that oxidation of vandetanib to its N-oxide is not only mediated by NADPH, which is considered as cofactor of FMO enzymes, but also by NADH which was even

ur

more efficient than NADPH to mediate the formation of vandetanib N-oxide by FMO1. This

Jo

finding was unexpected, because in previous studies where NADH was shown to be active to serve as cofactor for FMO-catalyzed reactions the efficacy to mediate oxidation reactions was generally lower than NADPH (Poulsen and Ziegler, 1995; Ziegler, 2002; Ziegler and Pettit, 1966;). The vandetanib metabolite profiles formed by microsomes isolated from rat livers varied significantly from those formed by human liver microsomes, especially in the generated amounts of vandetanib N-oxide. Whereas human hepatic microsomes were more efficient in the formation

17

of N-desmethylvandetanib as compared to rat hepatic microsomes, they generated much lower amounts of vandetanib N-oxide. The underlying mechanisms explaining these differences can be based on the distinctive expression levels of CYPs and FMOs in human and rat livers and/or on the different efficiencies of orthologue forms of human and rat biotransformation enzymes to oxidize vandetanib. The expression levels of the selective CYP and FMO enzymes are indeed significantly different in hepatic microsomes of the tested animal model (rats) and in humans. In rat liver, CYPs of the 2C subfamily are mainly expressed (~55% of the CYP rat liver complement)

ro of

instead of enzymes of the CYP3A (~15%). In contrast, in human livers, the CYP3A4 enzyme is predominantly expressed, being the major CYP enzyme of this organ (~30% of the CYP liver

complement) (Rendic and DiCarlo, 1997). CYPs of the 2C subfamily (~15% for CYP2C9) and

-p

CYP1A2 (~13%) are also highly expressed in this organ but are less abundant (Rendic and DiCarlo, 1997).

re

Of the FMO enzymes, FMO1 which is predominantly expressed in rat liver, efficiently

lP

oxidizes vandetanib to its N-oxide and is the major form of the FMOs generating the N-oxide metabolite in this rat organ. FMO3, which is known to be highly expressed in human liver (Itoh et al., 1993; Krueger et al., 2005), also oxidizes vandetanib to its N-oxide, but to a lesser extent.

na

Therefore, the distinct expression of individual liver CYP and FMO enzymes might be the major reason explaining the differences in vandetanib oxidation found between rat and human hepatic

ur

microsomes. Concerning the efficacy of rat CYP enzymes to oxidize vandetanib, some of them

Jo

significantly differ from those of their human orthologues. The activity of CYP2C11 to oxidize vandetanib is 25-times higher than human CYP2C8 (Indra et al., 2019; present study). Because CYP2C11 is highly expressed in rat liver (~28% of the rat CYP liver complement) (Nedelcheva and Gut, 1994; Večeřa et al. 2011; Yamazaki et al., 2006; Zachařová et al., 2012), the highest contribution to N-desmethylvandetanib formation in rat liver is attributed to this CYP (80%). A more than 30-fold lower efficiency was found for rat CYP3A1 and 3A2 as compared to human CYP3A4 and both these CYPs are also less abundant in rat liver. In contrast, rat CYP3A1/2 and

18

human CYP3A5 showed the same efficacy in the reaction (Indra et al., 2019; present study). CYP3A of rat liver contributed ~20% to the oxidation of vandetanib, whereas in contrast human hepatic CYP3A4 contributes 99% towards N-desmethylvandetanib formation (Indra et al., 2019). Further, whereas human and rat CYP1A1 orthologues oxidize vandetanib with similar efficiencies, human CYP1A2, in contrast to rat CYP1A2, does not oxidize vandetanib at all. Human CYP2D6 is more than 3-times more effective to oxidize vandetanib than rat CYP2D1/2 orthologues (Indra et al., 2019; present study).

ro of

In summary these results demonstrate that the expression levels of enzymes oxidizing vandetanib and different efficiencies of rat and human orthologues of biotransformation enzymes in vandetanib oxidation represent the mechanisms underlying the differences in vandetanib

-p

metabolite profiles generated in rat and human subcellular enzymatic systems. Further, these results suggest that the rat animal model seems to be not optimal in mimicking the enzyme-

re

mediated metabolic situation of vandetanib oxidation in human liver. Therefore, this should be

lP

taken into account when evaluating the suitability of individual animal models for studying the metabolism of vandetanib as well as the other drugs. Thus it is highly recommended to combine in

ur

5. Conclusions

na

vivo pharmacokinetical studies with in vitro experimental systems to evaluate drug metabolism.

NADPH- and NADH-mediated formation of N-desmethylvandetanib and vandetanib N-oxide was

Jo

identified to be the predominant in vitro metabolic pathway of vandetanib oxidation in rat and human liver systems. However, differences in the amounts of both metabolites formed in rat and human enzymatic systems can be explained by at least two mechanisms: (i) by the distinctive expression levels of CYP and FMO enzymes oxidizing vandetanib, and (ii) by the different efficiencies of orthologues of human and rat CYPs and FMOs to oxidize vandetanib. In rats, CYP2C11 >> 3A1 > 3A2 are mainly important in the formation of N-desmethylvandetanib, while FMO1 oxidizes vandetanib to vandetanib N-oxide. Investigation of structural aspects of the

19

enzyme active sites that may account for differences in vandetanib oxidation remain a challenge for future studies.

Conflict of Interest Statement. The authors declare that there are no conflicts of interest.

Acknowledgements. The work was supported by Grant Agency of the Czech Republic (grant 18-

Jo

ur

na

lP

re

-p

ro of

10251S).

20

REFERENCES Attwa MW, Kadi AA, Darwish HW, Amer SM, Al-Shakliah NS., 2018. Identification and characterization of in vivo, in vitro and reactive metabolites of vandetanib using LC-ESIMS/MS, Chem. Cent. J. 12, 99. Australian Public Assessment Report for Vandetanib, 2013. Caprelsa Vandetanib, AstraZeneca Pty Ltd PM-2011-03002-3-4, 1-73. Ballard, J.E., Prueksaritanont, T., Tang, C., 2007. Hepatic metabolism of MK-0457, a potent

ro of

aurora kinase inhibitor: Interspecies comparison and role of human cytochrome P450 and flavin-containing monooxygenase. Drug Metab. Dispos. 9, 1447-1451.

Baráčková P., 2019. [The effect of tyrosine kinase inhibitors vandetanib and lenvatinib and

Faculty of Science, Charles University, 2019.

-p

cytotoxic alkaloid ellipticine on biotransformation enzymes] (in Czech). Diploma Thesis,

re

Bates, D., 2013. ZD-6474. AstraZeneca, Curr. Opin. Investig. Drugs. 4, 1468-1472.

lP

Ciardiello, F., Valuto, R., Damiano, V., Valuto, R., Trojanu, T., Vitaglian, D., Carlomagno, F., Veneziani, B.M., Fontanini, G., Bianco, A.R., Tortura, G., 2003. Antitumor effects of ZD6474, a small molecule vascular endothelial growth factor receptor tyrosine kinase inhibitor, with

9, 1546-1556.

na

additional activity against epidermal growth factor receptor tyrosine kinase. Clin. Cancer Res.

ur

Chirulli, V., Longo, V., Marini, S., Mazzaccaro, A., Fiorio, R., Gervasi, P.G., 2005. CAR and

Jo

PXR expression and inducibility of CYP2B and CYP3A activities in rat and rabbit lungs. Life Sci. 76, 2535-2546.

Commander, H., Whiteside, G., Perry, C., 2011. Vandetanib: first global approval. Drugs 71, 1355-1365. Dvořáková, S., Václavíková, E., Sýkorová, V., Včelák, J., Novák, Z., Dušková, J., Ryska, A., Laco, J., Čáp, J., Kodetová, D., Kodet, R., Krsková, L., Vlček, P., Astl, J., Veselý, D.,

21

Bendlová, B., 2008. Somatic mutations in the RET proto-oncogene in sporadic medullary thyroid carcinomas. Mol. Cell. Endocrinol. 284, 21-27. FDA Center for Drug Evaluation and Research. Approval package for vandetanib, application number NDA 022405Orig1s000; Medical Review. [cited May 19, 2011]; from http://www.accessdata.fda.gov/drugsatfda_docs/nda/2011/022405Orig1s000MedR.pdf. Ferrara, N., Davis-Smyth, T., 1997. The biology of vascular endothelial growth factor. Endocr.. Rev. 18, 4-25.

ro of

Fujita, K., 2006. Cytochrome P450 and anticancer drugs. Curr. Drug Metab. 7, 23-37. Gao, C., Catucci, G., Gilardi, G., Sadeghi, S.J., 2018. Binding of methimazole and NADP(H) to human FMO3: In vitro and in silico studies. Int. J. Biol. Macromol. 118, 460-468.

-p

Guengerich, F.P., 2018. Mechanisms of cytochrome P450-catalyzed oxidations. ACS Catal. 8, 10964-10976.

re

Hadoux, J., Pacini, F., Tuttle, R.M., Schlumberger, M., 2016. Management of advanced medullary

lP

thyroid cancer. Lancet Diabetes Endocrinol. 4, 64-71.

Hadoux, J., Schlumberger, M., 2017. Chemotherapy and tyrosine-kinase inhibitors for medullary thyroid cancer. Best Pract. Res. Clin. Endocrinol. Metab. 31, 335-347.

na

Hennequin, L.F., Stokem, E.S., Thomas, A.P., Johnstone, C., Plé, P.A., Ogilvie, D.J., Dukes, M., Wedge, S.R., Kendrew, J., Curwen, J.O., 2002. Novel 4-anilinoquinazolines with C-7 basic side

ur

chains: design and structure activity relationship of a series of potent, orally active, VEGF

Jo

receptor tyrosine kinase inhibitors. J. Med. Chem. 45, 1300-1312. Hodek, P., Koblihová, J., Kizek, R., Frei, E., Arlt, V.M., Stiborová, M., 2013. The relationship between DNA adduct formation by benzo[a]pyrene and expression of its activation enzyme cytochrome P450 1A1 in rats. Environ. Toxicol. Pharmacol. 36, 989-996. Hodek, P., Bořek-Dohalská, L., Sopko, B., Šulc, M., Smrček, S., Hudeček, J., Janků J., Stiborová, M., 2005. Structural requirements for inhibitors of cytochromes P450 2B: assessment of the enzyme interaction with diamondoids. J. Enzyme Inhib. Med. Chem. 20, 25-33.

22

Indra, R., Pompach, P., Martínek, V., Takácsová, P., Vavrová, K., Heger, Z., Adam, V., Eckschlager, T., Kopečková, K., Arlt, V.M., Stiborová, M., 2019. Identification of human enzymes oxidizing the anti-thyroid-cancer drug vandetanib and explanation of the high efficiency of cytochrome P450 3A4 in its oxidation. Int. J. Mol. Sci. 20, pii: E3392. Itoh, K., Kimura, T., Yokoi, T., Itoh, S., Kamataki, T., 1993. Rat liver flavin-containing monooxygenase (FMO): cDNA cloning and expression in yeast. Biochim. Biophys. Acta. 1173, 165-171.

ro of

Kotrbová, V., Mrázová, B., Moserová, M., Martínek, V., Hodek, P., Hudeček, J., Frei, E., Stiborová, M., 2011 Cytochrome b5 shifts oxidation of the anticancer drug ellipticine by

cytochromes P450 1A1 and 1A2 from its detoxication to activation, thereby modulating its

-p

pharmacological efficacy. Biochem. Pharmacol. 82, 669-680.

Kowanetz, M., Ferrara, N., 2006. Vascular endothelial growth factor signaling pathways:

re

therapeutic perspective. Clin. Cancer Res. 12, 5018-5022.

lP

Krueger, S.K., Williams, D.E., 2005. Mammalian flavin-containing monooxygenases: structure/function, genetic polymorphisms and role in drug metabolism. Pharmacol. Ther. 106, 357-387.

na

Martin, P., Oliver, S., Kennedy, S.J., Partridge, E., Hutchison, M., Clarke, D., Gilda, P., 2012.

221-237.

ur

Pharmacokinetics of vandetanib: three phase I studies in healthy subjects. Clin. Ther. 34 (2012)

Jo

Martin, P., Oliver, S., Robertson, J., Kennedy, S.J., Read, J., Duvauchelle, T., 2011. Pharmacokinetic drug interactions with vandetanib during coadministration with rifampicin or itraconazole. Drugs R D. 11, 37-51.

Nedelcheva, V, Gut, I., 1994. P450 in the rat and man: methods of investigation, substrate specificities and relevance to cancer. Xenobiotica 24,1151-1175. Poulsen, L.L., Ziegler, D.M., 1995. Multisubstrate flavin-containing monooxygenases: applications of mechanism to specificity. Chem. Biol. Interact. 96, 57-73.

23

Reed, L., Indra, R., Mrizova, I., Moserova, M., Schmeiser, H.H., Wolf, C.R., Henderson, C.J., Stiborova, M., Phillips, D.H., Arlt, V.M., 2019. Application of hepatic cytochrome b5/P450 reductase null (HBRN) mice to study the role of cytochrome b5 in the cytochrome P450mediated bioactivation of the anticancer drug ellipticine. Toxicol. Appl. Pharmacol. 366, 6474. Reed, L., Mrizova, I., Barta, F., Indra, R., Moserova, M., Kopka, K., Schmeiser, H.H., Wolf, C.R., Henderson, C.J., Stiborova, M., Phillips, D.H., Arlt, V.M., 2018. Cytochrome b5 impacts on

ro of

cytochrome P450-mediated metabolism of benzo[a]pyrene and its DNA adduct formation: studies in hepatic cytochrome b5/P450 reductase null (HBRN) mice. Arch. Toxicol. 92, 1625– 1638.

-p

Rendic, S., DiCarlo, F.J., 1997. Human cytochrome P450 enzymes: A status report summarizing their reactions, substrates, inducers, and inhibitors. Drug Metab. Rev. 29, 413-480.

re

Schlenk, D., 1998. Occurrence of flavin-containing monooxygenases in non-mammalian

lP

eukaryotic organisms. Comp. Biochem. Physiol. C Pharmacol. Toxicol. Endocrinol. 121, 185-195.

Schlenk, D., Ronis, M.J., Miranda, C.L, Buhler, D.R., 1993. Channel catfish liver

na

monooxygenases. Immunological characterization of constitutive cytochromes P450 and the absence of active flavin-containing monooxygenases. Biochem. Pharmacol. 45, 217-221.

ur

Stiborová, M., Stiborová-Rupertová, M., Bořek-Dohalská, L., Wiessler, M., Frei, E. 2003. Rat

Jo

microsomes activating the anticancer drug ellipticine to species covalently binding to deoxyguanosine in DNA are a suitable model mimicking ellipticine bioactivation in humans. Chem. Res. Toxicol. 16, 38-47.

Stiborová, M., Bořek-Dohalská, L., Aimová, D., Kotrbová, V., Kukačková, K., Janouchová, K., Rupertová, M., Ryšlaáa, H., Hudeček, J., Frei, E., 2006. Oxidation pattern of the anticancer drug ellipticine by hepatic microsomes – similarity between human and rat systems. Gen. Physiol. Biophys. 25, 245-261.

24

Stiborová, M., Indra, R., Moserová, M., Černá, V., Rupertová, M., Martínek, V., Eckschlager, T., Kizek, R., Frei, E. 2012a. Cytochrome b5 increases cytochrome P450 3A4-mediated activation of anticancer drug ellipticine to 13-hydroxyellipticine whose covalent binding to DNA is elevated by sulfotransferases and N,O-acetyltransferases. Chem. Res. Toxicol. 25, 1075-1085. Stiborova, M., Levova, K., Barta, F., Shi, Z., Frei, E., Schmeiser, H.H., Nebert, D.W., Phillips, D.H., Arlt, V.M., 2012b. Bioactivation versus detoxication of the urothelial carcinogen aristolochic acid I by human cytochrome P450 1A1 and 1A2. Toxicol. Sci. 125, 345–358.

ro of

Stiborová, M., Indra, R., Moserová, M., Frei, E., Schmeiser, H.H., Kopka, K., Phillips, D.H., Arlt, V.M., 2016. NADH:Cytochrome b5 reductase and cytochrome b5 can act as sole electron

donors to human cytochrome P450 1A1-mediated oxidation and DNA adduct formation by

-p

benzo[a]pyrene. Chem. Res. Toxicol. 29, 1325-1334.

Stiborová, M., Indra, R., Frei, E., Kopečková, K., Schmeiser, H.H., Eckschlager, T., Adam, V.,

re

Heger, Z., Arlt, V.M., Martínek,V., 2017. Cytochrome b5 plays a dual role in the reaction

Chem. 148, 1983-1991.

lP

cycle of cytochrome P450 3A4 during oxidation of the anticancer drug ellipticine. Monatsh.

Šulc, M., Indra, R., Moserová, M., Schmeiser, H.H., Frei, E., Arlt, V.M., Stiborová, M., 2016. The

na

impact of individual cytochrome P450 enzymes on oxidative metabolism of benzo[a]pyrene in human livers. Environ. Mol. Mutagen. 57, 229-235.

ur

Václavíková, E., Dvořáková, S., Sýkorová, V., Bílek, R., Dvořáková, K., Vlček, P., Skaba, R.,

Jo

Zelinka, T., Bendlová, B., 2009. RET mutation Tyr791Phe: the genetic cause of different diseases derived from neural crest. Endocrine 36, 419-424.

Večeřa, R., Zachařová, A., Orolin, J., Strojil, J., Skottová, N., Anzenbacher, P., 2011. Fenofibrateinduced decrease of expression of CYP2C11 and CYP2C6 in rat. Biopharm. Drug Dispos. 32, 482-487. Vozniak, J.M., Jacobs, J.M., 2012. Vandetanib. J. Adv. Pract. Oncol. 3,112-116.

25

Wiechelman, K.J., Braun, R.D., Fitzpatrick, J.D., 1988. Investigation of the bicinchoninic acid protein assay: identified cation of the groups responsible for color formation. Anal. Biochem. 175, 231-237. Wedge, S.R., Ogilvie, D.J., Dukes, M., Kendrew, J., Chester, R., Jackson, J.A., Boffey, S.J., Valentine, P.J., Curwen, J.O., Musgrove, H.L., Graham, G.A., Hughes, G.D., Thomas, A.P., Stokem, E.S., Curry, B., Richmond, G.H., Wadsworth, P.F., Bigley, A.L., Hennequin, L.F. 2002. ZD6474 inhibits vascular endothelial growth factor signaling, angiogenesis, and tumor

ro of

growth following oral administration. Cancer Res. 62, 4645-4655. Wells, A., 1999. EGF receptor. Int. J. Biochem. Cell. Biol. 31, 637-643.

Yamazaki, H., Nakano, M., Omak, Y., Ueng, Y.F,, Guengerich, F.P., Shimada, T., 1996. Roles of

-p

cytochrome b5 in the oxidation of testosterone and nifedipine by recombinant cytochrome P450 3A4 and by human liver microsomes. Arch. Biochem. Biophys. 325, 174-182.

re

Yamazaki, H., Shimizu, M., Nagashima, T., Minoshima, M., Murayama, N., 2006. Rat

lP

cytochrome P450 2C11 in liver microsomes involved in oxidation of anesthetic agent propofol and deactivated by prior treatment with propofol. Drug Metab. Dispos. 34, 1803-1805. Yang, C.S., Tu, Y.Y., Koop, D.R., Coon, M.J., 1985. Metabolism of nitrosamines by purified

na

rabbit liver cytochrome P-450 isozymes. Cancer Res. 45, 1140-1145. Zachařová, A., Siller, M., Špičáková, A., Anzenbacherová, E., Skottová, N., Anzenbacher, P.,

ur

Večeřa, R., 2012. Rosuvastatin suppresses the liver microsomal CYP2C11 and CYP2C6

Jo

expression in male Wistar rats. Xenobiotica 42, 731-736. Ziegler, D.M., 2002. An overview of the mechanism, substrate specificities, and structure of FMOs. Drug Metab Rev. 2002 Aug;34(3):503-11.

Ziegler, D.M., Petit, F.H., 1966. Microsomal oxidases. I. The isolation and dialkylarylamine oxygenase activity of pork liver microsomes. Biochemistry 5, 2932-2938.

26

Legends to Figures

re

-p

ro of

Fig. 1. Vandetanib oxidation by rat CYP and FMO enzymes. Cytochrome b5 (cyt b5).

lP

Figure 1

na

Fig. 2. Separation of vandetanib and its metabolites formed by rat hepatic microsomes in the presence of NADPH (black line), NADH (blue line) and without NADPH/NADH (control) (red

ur

line) using HPLC analysis. V – vandetanib; M1 – N-desmethylvandetanib; M2 – vandetanib Noxide (a). MS spectrum of vandetanib metabolites measured by 12T ESI-FTICR (b). Vandetanib

Jo

was observed as singly and doubly charged ion at m/z 475.1144 (1+) and m/z 238.0607 (2+) (Δppm 1.0). Insert A in (b) represents the detail MS spectrum of vandetanib product Ndesmethylvandetanib at m/z 461.0990 (Δppm 1.5). Insert B in (b) represents the detail MS spectrum of vandetanib product N-oxide of vandetanib at m/z 491.1098 (Δppm 1.8).

27

ro of

a

Jo

Figure 2

ur

na

lP

re

-p

b

Fig. 3. Oxidation of vandetanib by human, rat, rabbit and mouse hepatic microsomes to Ndesmethylvandetanib and vandetanib-N-oxide in the presence of NADPH, NADH and NADPH plus NADH. Values represent mean ± SD of three independent in vitro incubations (n = 3). ***P < 0.001, **P<0.01, significant differences between formation of N-desmethylvandetanib and vandetanib N-oxide in the presence of NADPH and NADH. ΔΔΔP<0.001, ΔΔP<0.01, significant

28

differences between formation of N-desmethylvandetanib and vandetanib N-oxide in the presence

ur

Figure 3

na

lP

re

-p

ro of

of NADPH and NADPH plus NADH (ANOVA with post-hoc Tukey HSD Test).

Jo

Fig. 4. Oxidation of vandetanib by hepatic microsomes of control (untreated) rats and rats pretreated with several CYP inducers (BaP ‒ benzo[a]pyrene; PB ‒ phenobarbital; EtOH ‒ ethanol; pregnenolone-16α-carbonitrile ‒ PCN). Values present mean ± SD of three independent in vitro incubations (n = 3). ***P<0.001, **P<0.01, significant differences between formation of N-desmethylvandetanib in individual microsomes; ΔΔΔP<0.001, significant difference between formation of vandetanib N-oxide in individual microsomes (ANOVA with post-hoc Tukey HSD Test). 29

ro of -p re

Figure 4

lP

Fig. 5. Oxidation of vandetanib to N-desmethylvandetanib by rat recombinant CYPs in the presence of NADPH (A), NADPH, NADH and NADPH plus NADH (B) and oxidation of

na

vandetanib by human FMO1 and FMO3 to vandetanib N-oxide (C). Values represent mean ± SD of three independent in vitro incubations (n = 3). Panel (B) ***P<0.001, **P<0.01, *P<0.05 significant differences between formation of N-desmethylvandetanib by CYP enzymes in the

ur

presence of NADPH and NADH. ΔΔΔP<0.001, ΔΔP<0.01 significant differences between

Jo

formation of vandetanib N-oxide in the presence of NADPH and NADPH plus NADH. Panel (C) ***P<0.001, *P<0.05 significant differences between formation of vandetanib N-oxide in the presence of NADPH and NADH, ΔΔΔP<0.001 significant differences between formation of vandetanib N-oxide in the presence of NADPH and NADPH plus NADH (ANOVA with post-hoc Tukey HSD Test). ND - not detected.

30

Figure 5

31

ro of

-p

re

lP

na

ur

Jo

-p

ro of

Fig. 6. Contributions of rat hepatic CYP enzymes to the formation of N-desmethylvandetanib.

Jo

ur

na

lP

re

Figure 6

32

Table 1: Effects of CYP and FMO inhibitors on oxidation of vandetanib by rat liver microsomes % of Inhibition IC50 (μM) N-Desmethylvandetanib Formation α-Napththoflavone (CYP1A) 0 NAb Diamantane (CYP2B) 0 NA *** Sulfaphenazole (CYP2C) NA 40  4 CYPs Quinidine (CYP2D) 0 NA DDTC (CYP2A, CYP2E1) 0 NA Ketoconazole (CYP3A) 94 ± 3*** 2 ± 0.2 Vandetanib N-oxide Formation FMOs Methimazol 77 ± 4*** 6 ± 0.5 a CYPs for which compounds acts as their specific inhibitors are listed in brackets.

ro of

Enzymes Inhibitora

Equimolar concentrations of individual inhibitors and vandetanib (50 μM), 0.1 nmol of microsomal CYP for CYP inhibition and 0.25 mg of microsomal protein for FMO

-p

inhibition were in incubation mixtures. Values represent mean ± SD of three independent

in vitro incubations (n = 3). bNA, not applicable. *** P<0.001, significant difference from

Jo

ur

na

lP

re

data of controls, without inhibitors (ANOVA with post-hoc Tukey HSD Test).

33