Roles of human cytochrome P450 1A2 in coumarin 3,4-epoxidation mediated by untreated hepatocytes and by those metabolically inactivated with furafylline in previously transplanted chimeric mice

Abstract

Coumarin is a naturally occurring component of food products but is of clinical interest for its potential hepatotoxicity in humans. In the current study, the pharmacokinetics of coumarin in humanized-liver mice after oral and intravenous administrations (30 mg/kg) were investigated for its transformations to metabolically active coumarin 3,4-epoxide (as estimated by the levels of o-hydroxyphenylacetic acid) and to excretable 7-hydroxycoumarin. After oral administration, control mice metabolized coumarin to o-hydroxyphenylacetic acid at roughly the same rate as that to 7-hydroxycoumarin (total of unconjugated and conjugated forms). In contrast, the in vivo biotransformation of coumarin to o-hydroxyphenylacetic acid by humanized-liver mice was around two orders of magnitude less than that to conjugated and unconjugated 7-hydroxycoumarin. After intravenous administrations of coumarin, differences were observed in the plasma concentrations of o-hydroxyphenylacetic acid between humanized-liver mice treated with furafylline (daily oral doses of 13 mg/kg for 3 days) and untreated humanized-liver mice. The mean values of the areas under the plasma concentration versus time curves and the maximum concentrations for o-hydroxyphenylacetic acid were significantly lower in the group treated with furafylline (45% and 57% of the untreated values, respectively). These results suggested that the metabolic activation of coumarin in humans was mediated mainly by P450 1A2, which was suppressed by furafylline, and that humanized-liver mice orally treated with furafylline might constitute an in vivo model for metabolically inactivated P450 1A2 in human hepatocytes transplanted into chimeric mice.

INTRODUCTION

In vitro determinations of the major roles of cytochrome P450 (P450) enzymes in drug oxidation reactions using human liver preparations constitute a key method for estimating and understanding a variety of drug interactions (Kusama et al., 2009; Brown et al., 2005). This in vitro approach is carried out differently in different institutions. We previously employed the mechanism-based inactivation approach of selective P450 inhibition to evaluate the roles of different human P450 isoforms, especially P450 2C isoforms, in typical P450 probe substrate oxidations (Murayama et al., 2018). Furthermore, mechanism-based inactivation of P450 2C9 by treatment with tienilic acid (Hutzler et al., 2009) resulted in similar inhibition ratios in probe S-warfarin depletion rates and S-warfarin 7-hydroxylations in vitro in human liver microsomes (Murayama et al., 2018) and in vivo in humanized-liver mice pretreated with tienilic acid (Miura et al., 2020).

Coumarin occurs naturally and is found in various plants such as flowering cherry (Lake, 1999; Abraham et al., 2010). Although it reportedly causes hepatotoxicity in rats via the reactive metabolite coumarin 3,4-epoxide (Rietjens et al., 2010; Vassallo et al., 2003), coumarin is used as a medicine and a cosmetic fragrance ingredient in humans, its tolerable daily intake being 0.1 mg/kg body weight (Abraham et al., 2010). A risk assessment case study of coumarin has been recently reported (Baltazar et al., 2020; Moxon et al., 2020). The 7-hydroxylation of coumarin is slow in rats, and the activation of coumarin to its reactive 3,4-epoxide assumes relatively greater importance in rodents in general (Murayama and Yamazaki, 2021).

Coumarin elimination mediated by rat liver microsomes was quite slow and occurred mainly via 3,4-epoxidation, which ultimately led to the formation of o-hydroxyphenylacetic acid (Born et al., 2000; Tanaka et al., 2017). In in vitro experiments using liver microsomes from humans and rats, coumarin was eliminated via 7-hydroxylation mediated extensively by human liver microsomal P450 2A6 and quite slowly by rat microsomal P450 2A1 (Yamazaki et al., 1994). Contributions to 3,4-epoxidation in humans were predominantly by P450 1A2 (Murayama and Yamazaki, 2021) and minorly by human P450 2E1 (Born et al., 2002), whereas 3,4-epoxidation in rats was extensively mediated by P450 1A2 (Born et al., 2002). In in vitro experiments, it was found that mouse and rat liver cytosolic fractions mediated the detoxication of coumarin 3,4-epoxide produced by liver microsomes by converting it to o-hydroxyphenylacetic acid (Vassallo et al., 2004). Although cases of hepatotoxicity have been reported in patients treated with medicinal coumarin and in healthy subjects under dietary foodstuff consumption, there are limited in-depth reports on the in vivo metabolic profiles of coumarin in humans.

Furafylline, a 1,3,8-trisubstituted xanthine, results in the loss of human P450 1A2 activity (Racha et al., 1998). The aim of the present study is to discover the roles of the enzymes responsible for the metabolic activation of coumarin in humans in vivo. In the current study, we first precisely assessed the pharmacokinetics of coumarin in vivo using control and humanized-liver mice. We investigated coumarin oxidations via 3,4-epoxidation, ultimately leading to the formation o-hydroxyphenylacetic acid, and via 7-hydroxylation forming 7-hydroxycoumarin in humanized-liver mice treated with furafylline, the mechanism-based inhibitor of P450 1A2 (Racha et al., 1998). We herein report a strategy using suppressed liver P450 1A2 activities in humanized-liver mice to elucidate its role in the in vivo activation of coumarin to its primary reactive 3,4-epoxide.

MATERIALS AND METHODS

Chemicals and animals

Coumarin and 7-hydroxycoumarin were obtained from Fujifilm Wako Pure Chemicals (Osaka, Japan) and furafylline from Sigma-Aldrich (St. Louis, MO, USA). o-Hydroxyphenylacetic acid, 7-hydroxycoumarin sulfate, 7-hydroxycoumarin glucuronide, and 8-cyclopentyl-1,3-dimethylxanthine were obtained from Toronto Research Chemicals (Toronto, Canada). Valsartan was obtained from Tokyo Chemical Industry (Tokyo, Japan). Other chemicals were sourced as reported previously (Murayama and Yamazaki, 2021).

An improved TK-NOG mouse (NOG-TKm30, Central Institute for Experimental Animals, Kawasaki, Japan) was used in this study (Uehara et al., 2021) under the approval of the Animal Ethics Committee of the Central Institute for Experimental Animals (Permit Number: 20060A). Control mice and humanized-liver NOG-TKm30 mice (males and females weighing ~20–30 g) were treated with single oral administrations of coumarin (30 mg/kg). In separate experiments, humanized-liver mice were orally pretreated for 3 days with furafylline at a dose of 13 mg/kg. On the 4th day, the humanized-liver mice were intravenously administered a single dose of coumarin (30 mg/kg). The oral dose of furafylline (13 mg/kg) to humanized-liver mice necessary to achieve metabolic inactivation of P450 1A2 was established based on reported inhibitory doses of 90–125 mg (1.3–1.9 mg/kg) of furafylline in humans (Segura et al., 1986; Tarrus et al., 1987a) by applying a species factor of 10 and on an oral inhibitory dose of 10 mg/kg in rats (Tarrús et al., 1987b). Plasma samples were collected from control and humanized-liver mice at 0.25, 0.5, 1, 2, 4, 7, and 24 hr after coumarin administration. Accumulated urinary and fecal samples (0–24 hr) were also collected from control and humanized-liver mice. Plasma and urine samples (10 µL) were deproteinized by adding 10 and 90 µL, respectively, of acetonitrile and centrifuged at 20,000 × g for 10 min at 4°C. Fecal samples were homogenized in a fivefold (v/w) volume of methanol and sonicated for 5 min. The mixture was then centrifuged (4°C; 20,000 × g; 10 min) and the supernatant extract was collected.

In vivo metabolic studies of coumarin

Coumarin and its metabolites in the samples were determined using a liquid chromatography (LC) system. An API 5500 tandem mass spectrometer (AB Sciex, Framingham, MA, USA) directly coupled to a Nexera UHPLC System liquid chromatograph (Shimadzu, Kyoto, Japan) equipped with a C₁₈ column (YMC-Triart C18, particle size: 3 µm, 3.0 × 100 mm; YMC, Kyoto, Japan) was operated in electrospray positive ionization mode for coumarin; 7-hydroxycoumarin; and 8-cyclopentyl-1,3-dimethylxanthine (internal standard) or in electrospray negative ionization mode for 7-hydroxycoumarin glucuronide, 7-hydroxycoumarin sulfate, o-hydroxyphenylacetic acid, and valsartan (internal standard). The column temperature was set at 40°C and the flow rate was 0.30 mL/min. Gradient elution using 0.1% (v/v) formic acid in (A) water and (B) acetonitrile was conducted as follows: 20%–80% B (0–12 min), 100% B (12.01–15 min), and 20% B (15.01–18 min). The precursor-to-product mass transitions were m/z 147.1 to 91.1 for coumarin (retention time, 9.6 min), m/z 163.1 to 107.0 for 7-hydroxycoumarin (8.0 min), m/z 337.2 to 160.9 for 7-hydroxycoumarin glucuronide (4.5 min), m/z 241.1 to 160.9 for 7-hydroxycoumarin sulfate (4.7 min), m/z 249.1 to 192.2 for 8-cyclopentyl-1,3-dimethylxanthine (11.7 min), and m/z 434.1 to 179.0 for valsartan (13.9 min). Analyte samples (2.0 µL) were infused using an auto-sampler as described previously (Murayama and Yamazaki, 2021).

The plasma concentration–time data for individual humanized-liver mice were analyzed using noncompartmental analysis in Phoenix WinNonlin 8.2 (Pharsight, Mountain View, CA, USA) to determine individual pharmacokinetic parameters. Statistical analyses of these pharmacokinetic parameters in both humanized-liver mice treated with furafylline and in untreated mice were carried out using GraphPad Prism (GraphPad Prism Software, La Jolla, CA, USA).

RESULTS AND DISCUSSION

The in vivo pharmacokinetics of coumarin (30 mg/kg) were analyzed after oral administrations in control and humanized-liver mice (Fig. 1). Fifteen minutes after oral administration, 7-hydroxycoumarin (unconjugated and conjugated forms), o-hydroxyphenylacetic acid, and coumarin were detected in sampled plasma from control and humanized-liver mice. In control mice, the plasma concentrations of 7-hydroxycoumarin (unconjugated and conjugated forms) and o-hydroxyphenylacetic acid (generated as a result of coumarin 3,4-epoxidation) were roughly comparable and both were higher than those of the substrate coumarin under the present conditions (Fig. 1A). As expected, in the humanized-liver mice, coumarin was predominantly metabolized to 7-hydroxycoumarin glucuronide and moderately metabolized to 7-hydroxycoumarin sulfate as a result of the extensively primary 7-hydroxylation of coumarin (Fig. 1B). In control mice, the main fecal metabolite was o-hydroxyphenylacetic acid (Fig. 1D); in contrast, in humanized-liver mice, the major urinary metabolite was 7-hydroxycoumarin glucuronide (Fig. 1C). It was evident that the metabolic activation of coumarin to its active 3,4-epoxide by liver enzymes in vivo occurred in control mice and humanized-liver mice in a similar manner. The pharmacokinetics of coumarin in humanized-liver mice transplanted with human hepatocytes suggested a higher clearance of 7-hydroxycoumarin conjugates from plasma via the urinary pathway in mice with humanized livers than in control mice, presumably because of a different molecular sieve for hepatic binary excretion.

Fig. 1

Plasma concentrations (A, B) and urinary (C) and fecal (D) excretions of coumarin, o-hydroxyphenylacetic acid, 7-hydroxycoumarin, and its sulfate, and glucuronide in four control mice (A, C, D) and four humanized-liver mice (B, C, D) after a single oral dose of 30 mg coumarin/kg body weight. Data represent mean concentrations with standard deviations in samples taken from four individual control mice and four humanized-liver mice for coumarin (open circles and boxes), o-hydroxyphenylacetic acid (red triangles and boxes), 7-hydroxycoumarin (open gray squares and boxes), and its sulfate (light gray squares and boxed), and glucuronide (gray squares and boxes).

Intravenous administrations of coumarin (30 mg/kg) were performed in humanized-liver mice treated with daily oral doses of furafylline (13 mg/kg) for 3 days and in untreated humanized-liver mice. Plasma concentrations of coumarin and its metabolites are presented in Fig. 2. The plasma levels of coumarin (Fig. 2A) and 7-hydroxycoumarin (unconjugated and conjugated forms) (Fig. 2C) were roughly similar in treated and untreated mice. However, a significant difference was observed in the metabolic profiles of o-hydroxyphenylacetic acid (an indicator of 3,4-epoxidation) between the untreated and treated mice (Fig. 2B). After intravenous administrations of coumarin in humanized-liver mice treated with furafylline, the mean values of the areas under the plasma concentration versus time curves (AUC) and the maximum concentrations (C_max) of o-hydroxyphenylacetic acid were significantly lower (45% and 57%, respectively) than those in the untreated group (Table 1). There were no apparent differences in the profiles for coumarin or 7-hydroxycoumarin (unconjugated and conjugated forms) between the treated and untreated humanized-liver mice.

Fig. 2

Plasma concentrations of coumarin (A); o-hydroxyphenylacetic acid (B); and 7-hydroxycoumarin (open squares), its sulfate (light gray squares), and its glucuronide (gray squares, C) in humanized-liver mice after intravenous administration of coumarin (30 mg/kg body weight) in untreated humanized-liver mice (n = 4, in black, gray, and red) and in those pretreated for 3 days with furafylline (13 mg/kg, symbols in blue, n = 4). Mean values and SD bars are shown. *Significantly different from untreated mice, p < 0.05 (two-way analysis of variance).

Table 1. Pharmacokinetic parameters of coumarin and its metabolites after intravenous administration of 30 mg/kg body weight in both untreated humanized-liver mice and humanized-liver mice pretreated with daily oral doses of furafylline (13 mg/kg) for 3 days.

Analyte	Parameter	Untreated, n = 4			Treated, n = 4
Coumarin	C0, µg/mL	35.1	±	19.8 (100)	24.4	±	12.3 (70)
	AUC, µg·hr/mL	9.9	±	1.9 (100)	6.9	±	3.7 (70)
o-Hydroxyphenylacetic acid	Cmax, µg/mL	11	±	1 (100)	6.3	±	0.8 (57)**
	AUC, µg·hr/mL	20.9	±	11.0 (100)	9.4	±	2.9 (45)*
7-Hydroxycoumarin	Cmax, µg/mL	0.22	±	0.11 (100)	0.23	±	0.05 (105)
	AUC, µg·hr/mL	0.19	±	0.09 (100)	0.18	±	0.02 (95)
7-Hydroxycoumarin sulfate	Cmax, µg/mL	9.2	±	1.7 (100)	14.2	±	3.3 (154)
	AUC, µg·hr/mL	8.4	±	0.7 (100)	9.0	±	2.3 (107)
7-Hydroxycoumarin glucuronide	Cmax, µg/mL	133	±	66 (100)	223	±	40 (168)
	AUC, µg·hr/mL	203	±	113 (100)	256	±	61 (126)

The numbers in parentheses are percentage values compared with the untreated group. The plasma concentration–time data for individual humanized-liver mice were analyzed using noncompartmental analysis in Phoenix WinNonlin 8.2 (Pharsight, Mountain View, CA, USA) to determine individual pharmacokinetic parameters.

*Significantly different, *p < 0.05 and **p < 0.01.

Under the current conditions, in humanized-liver mice, the in vivo metabolic ratio of coumarin to o-hydroxyphenylacetic acid, generated via 3,4-epoxidation, was two orders of magnitude smaller than the metabolic ratio to unconjugated/conjugated 7-hydroxycoumarin. The reported mean fraction of coumarin metabolized to o-hydroxyphenylacetic acid mediated by P450 1A2, as discovered in vitro using human liver microsomes selectively inactivated by furafylline, was 0.65 (Murayama and Yamazaki, 2021). This value was similar to the calculated suppression of in vivo pharmacokinetic parameters obtained in vivo using the current humanized-liver model (0.43 for C_max and 0.55 for AUC) (Table 1). These results implied that the oral pretreatment of humanized-liver mice with furafylline had effectively suppressed P450 1A2 function and could therefore act as an in vivo model for metabolically inactivated human P450 1A2 hepatocytes. The half-life of furafylline (90–125 mg, 1.3–1.9 mg/kg) eliminated from human plasma is reportedly ~2 days (Segura et al., 1986; Tarrus et al., 1987a). The mechanism-based and competitive inhibitory effects of furafylline were deemed to be adequate for the current humanized-liver mice model under the current conditions.

Different ratios of urinary 7-hydroxycoumarin and o-hydroxyphenylacetic acid after oral administration of coumarin (2 mg) in a Jordanian population have been reported (Hadidi et al., 1998). In the current study, extensive urinary excretion of 7-hydroxycoumarin glucuronide was seen in the humanized-liver mice (Fig. 1C). Although coumarin predominantly underwent 7-hydroxylation mediated by P450 2A6, our findings suggested that, in humans, the metabolic activation of coumarin via 3,4-epoxidation was mediated mainly by P450 1A2, as demonstrated in both in vitro and in vivo experiments. Although human P450 1A2 was the dominant form mediating o-hydroxyphenylacetic acid formation, there is a potential minor role of P450 2E1 in the formation of o-hydroxyphenylacetic acid in human liver (Murayama and Yamazaki, 2021). In our preliminary experiments, humanized-liver mice had approximately twice levels of leaked human aspartate aminotransferase in plasma detected by an immunoassay kit (AlphaLISA Human AST kit, PerkinElmer, Waltham, MA, USA) 24 hr after oral doses of coumarin (300 mg/kg), without any hepatic pathological changes. Additional research using the current approach should elucidate the metabolic balance and clearance rates of natural food-derived coumarin in individual subjects. The metabolic oxidation of coumarin via 3,4-epoxidation to form o-hydroxyphenylacetic acid was dependent on inducible human P450 1A2 and could inform individual human risk assessments of dietary-derived coumarin, for which hepatotoxicity is especially evident in rodents.

In conclusion, in the current study, the metabolic oxidations of coumarin to 7-hydroxycoumarin and to its active 3,4-epoxide (as determined by the formation rates of o-hydroxyphenylacetic acid) were explored in vivo. Liver P450 1A2 mediated the in vivo activation of coumarin to its primary reactive 3,4-epoxide, and the rates of 3,4-epoxidation were similar in both control and humanized-liver mice (Fig. 1). In summary, humanized-liver mice orally pretreated with furafylline provided an in vivo model of metabolically inactivated P450 1A2 human hepatocytes. This model could be utilized to evaluate in vivo victim drugs in drug interactions dependent on human P450 1A2. The modified approach of using humanized-liver mice with selectively inactivated liver P450 1A2 (the current study), inactivated P450 2C9 (Miura et al., 2020), and other P450 enzymes inactivated in vivo may deliver further important information in future drug interaction studies. The current results indicated that the metabolic oxidation rates of coumarin via 3,4-epoxidation to form o-hydroxyphenylacetic acid, mediated mainly by includible P450 1A2 in human livers in vivo, could inform individual human risk assessment of dietary-derived coumarin.

ACKNOWLEDGMENTS

The authors thank Drs. Takashi Yamada, Fumiaki Shono, and Yusuke Kamiya for their assistance and David Smallbones for copyediting a draft of this article. This work was supported partly by the Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research 20K07164, the Food Safety Commission of Japan (JPCAFSC20202006), and the METI Artificial Intelligence-based Substance Hazard Integrated Prediction System Project, Japan.

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Abraham,  K.,  Wöhrlin,  F.,  Lindtner,  O.,  Heinemeyer,  G. and  Lampen,  A. (2010): Toxicology and risk assessment of coumarin: focus on human data. Mol. Nutr. Food Res., 54, 228-239.
•  Baltazar,  M.T.,  Cable,  S.,  Carmichael,  P.L.,  Cubberley,  R.,  Cull,  T.,  Delagrange,  M.,  Dent,  M.P.,  Hatherell,  S.,  Houghton,  J.,  Kukic,  P.,  Li,  H.,  Lee,  M.Y.,  Malcomber,  S.,  Middleton,  A.M.,  Moxon,  T.E.,  Nathanail,  A.V.,  Nicol,  B.,  Pendlington,  R.,  Reynolds,  G.,  Reynolds,  J.,  White,  A. and  Westmoreland,  C. (2020): A next-generation risk assessment case study for coumarin in cosmetic products. Toxicol. Sci., 176, 236-252.
•  Born,  S.L.,  Caudill,  D.,  Fliter,  K.L. and  Purdon,  M.P. (2002): Identification of the cytochromes P450 that catalyze coumarin 3,4-epoxidation and 3-hydroxylation. Drug Metab. Dispos., 30, 483-487.
•  Born,  S.L.,  Caudill,  D.,  Smith,  B.J. and  Lehman-McKeeman,  L.D. (2000): In vitro kinetics of coumarin 3,4-epoxidation: application to species differences in toxicity and carcinogenicity. Toxicol. Sci., 58, 23-31.
•  Brown,  H.S.,  Ito,  K.,  Galetin,  A. and  Houston,  J.B. (2005): Prediction of in vivo drug-drug interactions from in vitro data: impact of incorporating parallel pathways of drug elimination and inhibitor absorption rate constant. Br. J. Clin. Pharmacol., 60, 508-518.
•  Hadidi,  H.,  Irshaid,  Y.,  Vågbø,  C.B.,  Brunsvik,  A.,  Cholerton,  S.,  Zahlsen,  K. and  Idle,  J.R. (1998): Variability of coumarin 7- and 3-hydroxylation in a Jordanian population is suggestive of a functional polymorphism in cytochrome P450 CYP2A6. Eur. J. Clin. Pharmacol., 54, 437-441.
•  Hutzler,  J.M.,  Balogh,  L.M.,  Zientek,  M.,  Kumar,  V. and  Tracy,  T.S. (2009): Mechanism-based inactivation of cytochrome P450 2C9 by tienilic acid and (+/-)-suprofen: a comparison of kinetics and probe substrate selection. Drug Metab. Dispos., 37, 59-65.
•  Kusama,  M.,  Maeda,  K.,  Chiba,  K.,  Aoyama,  A. and  Sugiyama,  Y. (2009): Prediction of the effects of genetic polymorphism on the pharmacokinetics of CYP2C9 substrates from in vitro data. Pharm. Res., 26, 822-835.
•  Lake,  B.G. (1999): Coumarin metabolism, toxicity and carcinogenicity: relevance for human risk assessment. Food Chem. Toxicol., 37, 423-453.
•  Miura,  T.,  Uehara,  S.,  Shimizu,  M.,  Murayama,  N.,  Utoh,  M.,  Suemizu,  H. and  Yamazaki,  H. (2020): Different Roles of Human Cytochrome P450 2C9 and 3A Enzymes in Diclofenac 4′- and 5-Hydroxylations Mediated by Metabolically Inactivated Human Hepatocytes in Previously Transplanted Chimeric Mice. Chem. Res. Toxicol., 33, 634-639.
•  Moxon,  T.E.,  Li,  H.,  Lee,  M.Y.,  Piechota,  P.,  Nicol,  B.,  Pickles,  J.,  Pendlington,  R.,  Sorrell,  I. and  Baltazar,  M.T. (2020): Application of physiologically based kinetic (PBK) modelling in the next generation risk assessment of dermally applied consumer products. Toxicol. In Vitro, 63, 104746.
•  Murayama,  N.,  Yajima,  K.,  Hikawa,  M.,  Shimura,  K.,  Ishii,  Y.,  Takada,  M.,  Uno,  Y.,  Utoh,  M.,  Iwasaki,  K. and  Yamazaki,  H. (2018): Assessment of multiple cytochrome P450 activities in metabolically inactivated human liver microsomes and roles of P450 2C isoforms in reaction phenotyping studies. Biopharm. Drug Dispos., 39, 116-121.
•  Murayama,  N. and  Yamazaki,  H. (2021): Metabolic activation and deactivation of dietary-derived coumarin mediated by cytochrome P450 enzymes in rat and human liver preparations. J. Toxicol. Sci., 46, 371-378.
•  Racha,  J.K.,  Rettie,  A.E. and  Kunze,  K.L. (1998): Mechanism-based inactivation of human cytochrome P450 1A2 by furafylline: detection of a 1:1 adduct to protein and evidence for the formation of a novel imidazomethide intermediate. Biochemistry, 37, 7407-7419.
•  Rietjens,  I.M.,  Punt,  A.,  Schilter,  B.,  Scholz,  G.,  Delatour,  T. and  van Bladeren,  P.J. (2010): In silico methods for physiologically based biokinetic models describing bioactivation and detoxification of coumarin and estragole: implications for risk assessment. Mol. Nutr. Food Res., 54, 195-207.
•  Segura,  J.,  Garcia,  I. and  Tarrus,  E. (1986): Some pharmacokinetic characteristics of furafylline, a new 1,3,8-trisubstituted xanthine. J. Pharm. Pharmacol., 38, 615-618.
•  Tanaka,  Y.,  Fujii,  W.,  Hori,  H.,  Kitagawa,  Y. and  Ozaki,  K. (2017): Changes in coumarin kinetics and subcellular localization of CYP2E1 contribute to bile duct damage and reduce hepatocellular damage after repeated administration of coumarin in rats. Toxicol. Lett., 280, 99-105.
•  Tarrus,  E.,  Cami,  J.,  Roberts,  D.J.,  Spickett,  R.G.,  Celdran,  E. and  Segura,  J. (1987a): Accumulation of caffeine in healthy volunteers treated with furafylline. Br. J. Clin. Pharmacol., 23, 9-18.
•  Tarrús,  E.,  García,  I.,  Roberts,  D.J. and  Segura,  J. (1987b): An animal model for the detection of drug-induced inhibition of caffeine metabolism. Methods Find. Exp. Clin. Pharmacol., 9, 311-316.
•  Uehara,  S.,  Higuchi,  Y.,  Yoneda,  N.,  Kawai,  K.,  Yamamoto,  M.,  Kamimura,  H.,  Iida,  Y.,  Oshimura,  M.,  Kazuki,  Y.,  Yamazaki,  H.,  Hikita,  H.,  Takehara,  T. and  Suemizu,  H. (2021): An improved TK-NOG mouse as a novel platform for humanized liver that overcomes limitations in both male and female animals. Drug Metab. Pharmacokinet., in press.
•  Vassallo,  J.D.,  Hicks,  S.M.,  Daston,  G.P. and  Lehman-McKeeman,  L.D. (2004): Metabolic detoxification determines species differences in coumarin-induced hepatotoxicity. Toxicol. Sci., 80, 249-257.
•  Vassallo,  J.D.,  Morrall,  S.W.,  Fliter,  K.L.,  Curry,  S.M.,  Daston,  G.P. and  Lehman-McKeeman,  L.D. (2003): Liquid chromatographic determination of the glutathione conjugate and ring-opened metabolites formed from coumarin epoxidation. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 794, 257-271.
•  Yamazaki,  H.,  Mimura,  M.,  Sugahara,  C. and  Shimada,  T. (1994): Catalytic roles of rat and human cytochrome P450 2A enzymes in testosterone 7 alpha- and coumarin 7-hydroxylations. Biochem. Pharmacol., 48, 1524-1527.