Characterization of kinetics of human cytochrome P450s involved in bioactivation of flucloxacillin: inhibition of CYP3A-catalysed hydroxylation by sulfaphenazole

Article abstract of DOI:10.1111/bph.14548

Background and Purpose: The aim of this study was to characterize the human cytochrome P450s (CYPs) involved in oxidative bioactivation of flucloxacillin to 5-hydroxymethyl flucloxacillin, a metabolite with high cytotoxicity towards biliary epithelial cel

Full text of DOI:10.1111/bph.14548

Characterization of kinetics of human cytochrome P450s involved in
bioactivation of flucloxacillin: inhibition of CYP3A-catalyzed
hydroxylation by sulfaphenazole
Authors: Stefan J. Dekker, Floor Dohmen, Nico P.E. Vermeulen and Jan N.M.
Commandeur*
Affiliation:
Division of Molecular Toxicology, Amsterdam Institute for Molecules Medicine and Systems
(AIMMS), Vrije Universiteit, De Boelelaan 1108, 1081HZ, Amsterdam, the Netherlands.
Running head:
Bioactivation of flucloxacillin by human CYPs.
Correspondence: Dr. Jan N.M. Commandeur, Division of Molecular Toxicology, Amsterdam
Institute for Molecules Medicine and Systems (AIMMS), Vrije Universiteit, De Boelelaan
1108, 1081HZ, Amsterdam, the Netherlands. Phone: (31)205987595. E-mail:
j.n.m.commandeur@vu.nl
Number of figures: 8
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/bph.14548
This article is protected by copyright. All rights reserved.

BACKGROUND AND PURPOSE
The aim of this study was to characterize the human cytochrome P450s (CYPs) involved in
oxidative bioactivation of flucloxacillin to 5-hydroxymethylflucloxacillin, a metabolite with
high cytotoxicity toward biliary epithelial cells.
EXPERIMENTAL APPROACH
The CYPs involved in hydroxylation of flucloxacillin were characterized using recombinant
human CYPs, pooled liver microsomes in presences of CYP-specific inhibitors, and by
correlation analysis using a panel of liver microsomes from 16 donors.
KEY RESULTS
Recombinant CYPs showing the highest specific activity were CYP3A4, CYP3A7 and to
lower extent CYP2C9 and CYP2C8. Michaelis Menten enzyme kinetics were determined for
pooled HLM, recombinant CYP3A4, CYP3A7 and CYP2C9. Remarkably, sulfaphenazole
appeared a potent inhibitor of 5’-hydroxylation of FLX by both recombinant CYP3A4 and
CYP3A7.
CONCLUSIONS AND IMPLICATIONS
The combined results show that the 5’-hydroxylation of FLX is primarily catalysed by
CYP3A4, CYP3A7 and CYP2C9. The large variability of the hepatic expression of these
enzymes could affect the formation of 5’-hydroxymethylfucloxacillin, which may determine
the differences in susceptibility to flucloxacillin-induced liver injury. Additionally, the strong
inhibition in CYP3A-catalyzed flucloxacillin metabolism by sulfaphenazole suggest that
unanticipated drug-drug interactions could occur with coadministered drugs.
Abbreviations:
BEC, biliary epithelial cells; CYP, cytochrome P450; DILI, drug-induced liver injury; FLX,
flucloxacillin; GWAS, genome-wide association studies; HLM, human liver microsomes; 5'-
HM-FLX, 5'-hydroxymethyl flucloxacillin; NRS, NADPH-regenerating system.
This article is protected by copyright. All rights reserved.

What is known about this subject
 There are large inter-individual differences in plasma concentrations of flucloxacillin and
5’-hydroxymethyl flucloxacillin.
 5’-hydroxymethyl flucloxacillin is much more toxic to biliary epithelial cells than
flucloxacillin
 A previous study on four human CYPs showed highest activity of CYP3A4 in
flucloxacillin hydroxylation, however, only a single substrate concentration was used,
without quantification of specific activity due to lack of reference metabolite.
What this story adds
 Quantification of specific activities and enzyme kinetics of FLX-hydroxylation by HLM
and fourteen recombinant human CYPs.
 Next to CYP3A4, CYP3A7 and CYP2C9 may contribute significantly to flucloxacillin
hydroxylation.
 Sulfaphenazole, considered as a selective CYP2C9-inhibitor, strongly inhibits CYP3A-
mediated FLX-hydroxylation.
This article is protected by copyright. All rights reserved.

CYP3A4 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1337
CYP3A7 -
http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=263#1339
CYP2C9 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1326
CYP2C8 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1325
cyclosporine -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=1024
quinidine -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2342
warfarine -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=6853
NADP -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=3045
bufuralol -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=9834
chlorzoxazone -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2322
fluvoxamine -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=7189
ketoconazole -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2568
mephenytoin -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=7223
miconazole -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2449
quercetin -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=5346
quinidine -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2342
This article is protected by copyright. All rights reserved.

testosterone -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2858
ticlopidine -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=7307
Midazolam -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=3342
CYP1A2 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1319
CYP2B6 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1324
CYP2C8 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1325
CYP2C19 -
http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=262#1328
CYP2D6 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1329
CYP2E1 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1330
phenacetin -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=7402
CYP2A6 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1321
bupropion -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=7135
amodiaquine -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=10018
diclofenac -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2714
bufuralol -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=9834
CYP2J2 -
http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=262#1332
This article is protected by copyright. All rights reserved.

CYP3A5 -
http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=263#1338
CYP1A1 -
http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=1318
hydroxylapatinib -
http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=5692
Introduction
The isoxazolyl-penicillin flucloxacillin (FLX) is a narrow spectrum β-lactam
antibiotic which is active against gram-positive bacteria, such as methicillin-sensitive
Staphylococcus aureus (Sutherland et al., 1970; Leder et al., 1999). Although the antibiotic is
generally well tolerated at daily doses as high as 12 grams, it causes drug-induced liver injury
(DILI) of a cholesteric nature in susceptible patients (Fairley et al., 1993; Derby et al., 1993;
Eckstein et al., 1993; Koek et al., 1994; Russmann et al., 2005; De Abajo et al., 2004;
Andrews et al., 2009). Although the risk for FLX-induced DILI is in the region of 9 in
100.000 patients (Russmann et al., 2005), the fact that it is one of the most prescribed β-
lactam antibiotics makes FLX one of the drugs most often involved in DILI (De Abajo et al.,
2004). The mechanisms underlying FLX-induced cholestasis are still incompletely
understood. Association studies have identified several risk factors, such as prolonged
treatment with high daily doses and an age over 55 years (Fairley et al., 1993; Russmann et
al., 2005). Furthermore, a genome-wide association study (GWAS) identified the HLA-
B*57:01 allele as a strong risk factor for FLX-induced DILI, with an odds ratio of 80.6,
indicative for an important role of the immune system in the mechanism of liver toxicity
(Daly et al., 2009). Many forms of idiosyncratic DILI are believed to have an immunological
basis, in which covalent binding of drugs and/or metabolites to proteins may play an
important role by generation of a hapten and by causing cellular stress, leading to the release
of so-called danger signals which can stimulate the immune response (Uetrecht et al., 2007;
Tailor et al., 2015). Like many other penicillin-like antibiotics, the reactive β-lactam-rings of
FLX and its active metabolite 5’-hydroxymethyl FLX (5’-HM-FLX, Figure 1) have been
shown to react with lysine-residues of proteins thereby forming potential haptens (Jenkins et
al., 2009). Interestingly, cytotoxic (CD8+) T-cells were only activated by FLX-exposed
antigen-presenting cells in cells obtained from patients and volunteers with the HLA B*57:01
genotype, supporting the role of this genotype as risk factor for FLX-induced DILI (Monshi
This article is protected by copyright. All rights reserved.

et al., 2013; Wuillemin et al., 2013; Wuillemin et al., 2014). However, although the
association with HLA B*57:01 is one of the strongest ever reported, still only 1 in 1000
patients with the high-risk allele eventually develops FLX-induced DILI. Therefore, non-
genetic risk factors, which escape detection by GWAS-studies, may also determine the
sensitivity for FLX-induced cholestasis. Because 5’-HM-FLX was shown to be highly
cytotoxic to isolated biliary epithelial cells, whereas FLX was without effect, a high internal
exposure to 5’-HM-FLX may be considered as a potentially important risk factor (Lakehal et
al., 2001).
Many studies have investigated the pharmacokinetics of FLX in groups of healthy volunteers
or patients (Thijssen et al., 1982; Adam et al., 1983; Gath et al., 1995; Røder et al., 1995;
Landersdorfer et al., 2007; Dijkmans et al., 2012; Maier-Salamon et al., 2017). Dependent of
the dose regimen and route of administration (by i.v. infusion and/or orally) the maximal
plasma concentration (Cmax) of FLX can range 50 to 790 µM. Systemic clearance of FLX
appears to be mainly by renal clearance (65-71%) (Landersdorfer et al., 2007; Maier-
Salamon et al., 2017). Non-renal clearance of FLX occurs predominantly by cytochrome
P450-mediated hydroxylation of FLX to 5’-HM-FLX and partly by hydrolysis to the
corresponding penicilloic acid, Figure 1. In healthy volunteers plasma-concentrations of 5’-
HM-FLX are on average approximately 10% of those of FLX, whereas their half-lives are
comparable (Thijssen et al., 1982). Because of the high contribution of renal excretion of
FLX and 5’-HM-FLX, renal impairment was shown to lead to a significant increase in the
areas under the curve which may increase the risk for FLX-induced DILI (Gath et al., 1995).
The different pharmacokinetics studies have demonstrated that metabolism of FLX
shows significant interindividually variability, which may be a reflection of the large
interindividual variability of the enzymes involved in hydroxylation of FLX and/or de
mammalian dipeptidase involved in hydrolysis of FLX and 5’-HM-FLX to their
corresponding penicilloic acids. In the study of Lakehal et al., it was shown that CYP3A4
was able to catalyze the 5’-hydroxylation of FLX (Lakehal et al., 2001). However, the
involvement of only four CYP-isoforms were studied and FLX-hydroxylation was measured
at only one concentration. Another study showed that FLX induces CYP3A4 in cell-lines
exposed to 200-400 µM FLX (Huwyler et al., 2006). This may explain the decrease of
plasma concentrations of cyclosporine A, quinidine, voriconazole and warfarine when
coadministered together with FLX (Cynke et al., 1999; Comuth et al., 2012; Muilwijk et al.,
2017; Chaudhuri and Wade, 2018). Whether this induction resulted in increased formation of
5’-HM-FLX was not studied, however. Because a recent meta-analysis of Achour et al.
This article is protected by copyright. All rights reserved.

(Achour et al., 2014) showed that the hepatic expression levels of all drug metabolizing
human CYPs vary considerably, due to genetic polymorphism, enzyme induction and
epigenetic factors, it is critically important to fully characterize the CYPs involved in drug
metabolism in order to predict the interindividual variability in pharmacokinetics and the risk
to be victim of drug-drug interaction. Furthermore, a major involvement of CYP3A4 in drug
metabolism has been shown to complicate prediction of pharmacokinetics based on enzyme
kinetics because of the occurrence of atypical enzyme kinetics, such as substrate activation
(Houston et al., 2000; Ekins et al., 2003; Atkins et al., 2005; Houston et al., 2005).
Furthermore, inhibition of CYP3A4 by drugs appears substrate-dependent complicating
prediction of drug-drug interactions (Stresser et al., 2000).
The aim of the present study was therefore to characterize the bioactivation of FLX to
5’-HM-FLX in more detail by using a larger set of recombinant CYPs, by inhibition studies
using pooled human liver microsomes (HLM) and by correlation analysis using a panel of
HLM from 16 donors. Furthermore, for HLM and the most active CYPs the enzyme kinetics
of FLX-hydroxylation was determined.
Materials & methods
Chemicals
FLX (CAS: 1847-24-1) was a kind gift of Dr. Bayliss from the University of
Liverpool. β-Nicotinamide adenine dinucleotide phosphate disodium salt (NADP; CAS:
24292-60-2), glucose-6-phosphate (G-6-P; CAS: 54010-71-8), G-6-P dehydrogenase from
baker's yeast (Saccharomyces cerevisiae; CAS: 9001-40-5), (+)-N-3-benzyl-nirvanol (CAS:
790676-40-3), bufuralol (CAS: 60398-91-6), chlorzoxazone (CAS: 95-25-0),
diethyldithiocarbamate (CAS: 20624-25-3), fluvoxamine maleate (CAS: 22916-47-8),
ketoconazole (CAS: 65277-42-1), (S)-mephenytoin (CAS: 70989-04-7), miconazole (CAS:
22916-47-8), -naphtoflavone (CAS: 604-59-1), quercetin (CAS: 6151-25-3), quinidine
(CAS: 56-54-2), sulfaphenazole (CAS: 526-08-9), testosterone (CAS: 58-22-0), ticlopidine
(CAS: 53885-35-1), and troleandomycin (CAS: 2751-09-9) were purchased from Sigma-
Aldrich (St. Louis, MO). Midazolam (CAS: 59467-96-8) was from Duchefa Farma (Haarlem,
the Netherlands). All other chemicals were of analytical grade purity and obtained from
standard suppliers
Gentest Supersomes containing recombinant human CYPs were obtained from
Corning BV life sciences (Amsterdam, the Netherlands). The enzymes used were CYP1A1
This article is protected by copyright. All rights reserved.

(Lot No. 456ꢀ211), CYP1A2 (Lot No. 456ꢀ203), CYP2A6 (Lot No. 456ꢀ254), CYP2B6 (Lot
No. 456ꢀ255), CYP2C8 (Lot No. 456ꢀ252), CYP2C9*1(Arg144) (Lot No. 456ꢀ258),
CYP2C18 (Lot No. 456ꢀ222), CYP2C19 (Lot No. 456ꢀ259), CYP2D6*1 (Lot No. 456ꢀ217),
CYP2E1 (Lot No. 456ꢀ206), CYP2J2 (Lot No. 456ꢀ264), CYP3A4 (Lot No. 456ꢀ207),
CYP3A5 (Lot No. 456ꢀ256) and CYP3A7 (Lot No. 456ꢀ237).
Pooled HLM (50 donors, 20 mg/ml, lot No. 0710619) were obtained from Xenotech
(Lenexa, USA). A panel of HLM from single donors were prepared from livers kindly
provided by Kalycell (Plobsheim, France) (donors: S1399T, S1449T, S1352T, S1342T,
B1327T, S1405T, S1336T, S1446T, S1329T, S1339T, S1343T, S1442T, S1344T, S1441T,
S1334T and R1341T).
5’-HM-FLX was obtained biosynthetically by hydroxylation of FLX using a drug
metabolizing mutant of CYP102A1 from Bacillus Megaterium, M11 L437E, which contains
mutations R47L, E64G, F81I, F87V, E143G, L188Q, E267V, G415S and L437E compared
to wild-type CYP102A1. This mutant appeared to selectively convert FLX to 5’-HM-FLX
(Luirink et al., 2018). One mL of a solution of 100 mM FLX in DMSO was added to 49 mL
100 mM potassium phosphate buffer pH 7.4 and incubated in presence of 1 µM of M11
L437E and NADPH-regenerating system (NRS) for 20 hours at room temperature with
constant stirring. The NRS consisted of 100 µM NADPH, 10 mM glucose-6-phosphate, and
0.5 units/mL glucose-6-phosphate dehydrogenase (final concentrations). The reaction was
terminated by the addition of an equal volume ice-cold methanol containing 2% acetic acid.
The mixture was centrifuged at 4000 rpm for 2 hours at 4°C after which the supernatant was
filtered through a 0.2 µm Phenex RC membrane filter from Phenomenex (Utrecht, the
Netherlands). 5’-HM-FLX was purified by preparative HPLC using a Xbridge Prep C18-MS
(5 µm; 10 x 55 mm) column. Samples of 1.5 mL were injected and eluted with a flow rate of
3 mL/min. A gradient was constructed using eluent A (99% water/0.9% acetonitril/0.1%
formic acid) and eluent B (99% acetonitril/0.9% water/0.1% formic acid) and programmed
from 25% B to 50% B in 11 min after which it returned to 25% B in half a minute. After re-
equilibrating for 4.5 min the next sample was injected. A fraction collector was triggered to
collect peaks appearing at UV-detection at 272 nm. The fractions containing 5’-HM FLX
were combined and evaporated to dryness and stored at -80°C under N2. The identity of 5’-
1
19
HM FLX was confirmed by H NMR and F NMR using a Bruker Avance 250 (Fallanden,
Switzerland) equiped with a cryoprobe operating at 250,1 MHz. The metabolite had a purity
of 99% based on HPLC analysis.
This article is protected by copyright. All rights reserved.

Oxidative metabolism of FLX by HLM and recombinant human CYPs
FLX was initially incubated at concentrations of 10 and 100 µM in presence of 100
nM of fourteen Gentest Supersomes in order to identify the most active human CYPs.
Reactions were initiated in the presence of NRS (see above). Incubations were performed in
a 100 mM potassium phosphate buffer pH 7.4 supplemented with 5 mM MgCl2 and 2 mM
EDTA, and at a total volume of 50 µL. The incubations were started by the addition of NRS
and incubated for 10 min at 37 °C, during which metabolite production was linear (data not
shown). The reactions were terminated as described above. The samples were subsequently
centrifuged at 14,000 rpm (20,800g) for 20 min using a table top centrifuge. The supernatant
was filtered through 0.2 µm Phenex RC membrane filters from Phenomenex (Utrecht, the
Netherlands) and subsequently analysed by LC-MS, as described below. Since these
incubations were only performed to identify the most active enzymes duplicate experiments
were deemed sufficient since reproducibility of incubations with recombinant enzymes from
the same source was always within 5%,
For the most active Gentest Supersomes and pooled HLM, the enzyme kinetics of the
formation of 5’-HM-FLX were determined by incubating with eleven concentrations of FLX,
ranging from 0 to 500 µM. Each FLX concentration was incubated in duplicate for 10 min
with 0.5 mg/mL pooled HLM, 100 nM CYP3A4, CYP3A7 and CYP2C9 Supersomes in 100
mM potassium phosphate buffer, pH 7.4 supplemented with 5 mM MgCl2 and 2 mM EDTA.
The reactions were started, stopped and analysed as described above.
Effect of specific inhibitors of CYPs on oxidative metabolism of FLX by pooled HLM.
To study the involvement of individual human CYPs in the hydroxylation of FLX in
incubations with pooled HLM from 50 individuals, the effect of selective P450 inhibitors was
investigated. FLX was incubated at substrate concentrations of 10 and 100 µM with 0.5
mg/mL HLM, as described above, and in absence or presence of the following inhibitors: 10
µM α-naphtoflavone (αNF; CYP1A2), 3 µM ticlopidine (TIC; CYP2B6), 15 µM quercetin
(QCT; CYP2C8), 10 µM sulfaphenazole (SPZ; CYP2C9), 10 µM fluvoxamine (FVX;
CYP2C9/CYP1A2), 1 µM (+)-N-3-benzyl-nirvanol (BNV; CYP2C19), 2 µM quinidine
(QND; CYP2D6), 20 µM diethylthiocarbamate (DDC; CYP2E1), 3 µM troleandomycin
(TAO; CYP3A4); 2 µM miconazole (MCZ; CYP3A4/CYP2C9) and 0.1 µM (CYP3A4) and
1 µM (CYP3A) ketoconazole (KTZ) respectively. At these concentrations the inhibitors are
considered to be selective for the indicated P450 (Khojasteh et al., 2011; Kumar et al., 2006;
Niwa et al., 2009). The mechanism-based inhibitors TIC and TAO were pre-incubated with
This article is protected by copyright. All rights reserved.

HLM and NRS for 10 minutes before the addition of FLX. The inhibition experiment was
performed three times, with each condition incubated in duplicate.
Because the CYP2C9-specific inhibitor sulfaphenazole appeared to almost completely
block formation of 5’-HM-FLX by pooled HLM, it was also added at different concentrations
to incubations of CYP3A4 and CYP3A7 Supersomes with FLX (10 µM) and testosterone (50
µM).
Variability of oxidative metabolism of FLX by individual HLM and correlation analysis
To determine the variability of 5’-hydroxylation of FLX, FLX was incubated at 10
and 100 µM with a panel of individual HLM-fractions from 16 donors (0.5 mg/mL) which
were phenotyped with respect to activities of CYP1A2 (phenacetin O-deethylation), CYP2A6
(coumarine 7-hydroxylation), CYP2B6 (bupropion hydroxylation), CYP2C8 (amodiaquine
N-deethylation), CYP2C9 (diclofenac 4’-hydroxylation), CYP2C19 (mephenytoin 4-
hydroxylation), CYP2D6 (bufuralol 1-hydroxylation), CYP2E1 (chlorzoxazone 6-
hydroxylation) and CYP3A4 (midazolam 1’-hydroxylation and testosterone 6ß-
hydroxylation) using the concentrations described previously (den Braver-Sewradj et al.,
2017).
Analytical methods
FLX and its metabolites were measured using an Agilent TOF 6230 mass
spectrometer with electrospray ionization source coupled with an Agilent 1200 rapid
resolution LC system. The mass spectrometer was operated under the following conditions: a
capillary voltage of 3500V, 10 L/min nitrogen drying gas and 50 psig nitrogen nebulizing gas
at 350°C and operating in positive ion mode. Chromatographic separation of FLX and its
metabolites was achieved by reversed-phase liquid chromatography using a Phenomenex
Luna C18-column (5 µm; 4.6 x 150 mm). A gradient was constructed using 1% acetonitrile in
0.1% formic acid (A) and 99% acetonitrile in 0.1% formic acid (B) as eluents. The gradient
was programmed from 40% B to 99% B in 23.5 min after which the percentage went back to
40% B in 0.5 minute. Next the column was re-equilibrated in 40% B for 11 min. The flow
rate was 0.4 mL/min.
The chromatograms were analysed using Agilent Masshunter Qualitative Analysis
software (RRID:SCR_016657). Quantification was performed using a standard curve of 5’-
This article is protected by copyright. All rights reserved.

HM-FLX ranging from 1 nM to 6 µM. The lower limit of quantification of 5’-HM-FLX was
8 nM.
Data analysis
Enzyme kinetic parameters were determined using GraphPad Prism 5.00 for
Windows, GraphPad Software (San Diego, USA; RRID:SCR_002798) and fitted to the
Michaelis Menten equation. IC50's were calculated using the log(inhibitor) vs. response
function of Graphpad Prism 5.00 for Windows, Graphpad Software (San Diego, USA). The
Pearson correlation coefficient was calculated for the specific probe substrates and
flucloxacillin 5’-hydroxylation using GraphPad Prism 5.00 for Windows, GraphPad Software
(San Diego, USA). A two-tailed p-value was used with a confidence interval of 95%. The
data and statistical analysis in this manuscript comply with the recommendations on
experimental design and analysis in pharmacology (Curtis et al, 2015). However,
randomization of samples was deemed unneccessary since no systematic bias was observed
related to the sample position or in the order of sample analysis by LC-MS. Blinding was not
undertaken since prior knowledge of sample content was not expected to introduce bias of the
measurements. Also data acquisition and data analysis were performed by different
individuals.
Nomenclature of Targets and Ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries
in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS
Guide to PHARMACOLOGY (Harding et al., 2018), and are permanently archived in the
Concise Guide to PHARMACOLOGY 2017/18 (Alexander et al., 2017).
Results
Oxidative metabolism of FLX by HLM and recombinant CYPs.
As shown in Figure 2, initial incubations of 10 µM FLX for 60 min with pooled
HLM, in the presence or absence of NRS, resulted in the formation of only one NRS-
dependent metabolite eluting at 11.6 min and with m/z ([M+H]+) of 470.052, which
corresponded to the retention time and m/z-value of biosynthetical 5’-HM-FLX. The product
eluting at 9.3 min has a m/z of 472.078 and was also formed in absence of NRS. The mass
increase of 18 relative to FLX (m/z 454.059) indicates that this product most likely
corresponds to the penicilloic acid formed by hydrolysis of the beta-lactam ring of FLX
This article is protected by copyright. All rights reserved.

which is not dependent of CYP-activity. The penicilloic acid of 5'-HM-FLX , which was
identified in urine (Maier-Salamon et al., 2017), was not yet detectable under the incubation
conditions used. After checking the linearity of product formation with time, the enzyme
kinetics of 5’-HM-FLX formation by pooled HLM was studied in 10 minute incubations by
varying the concentration of FLX from 10 to 500 µM. As shown in Figure 3A, a hyperbolic
curve was obtained consistent with Michaelis-Menten kinetics. Enzyme kinetic parameters
obtained by non-linear regression were: Km 284 ± 38 µM and Vmax 30 ± 2 pmol/min/mg
protein.
To characterize which human CYP isoenzymes catalyze the formation of 5’-HM-
FLX, eleven different recombinant CYP Supersomes were incubated with 10 and 100 µM
FLX. These concentrations were chosen to discriminate between active isoforms at the
average Cmax in patients taking 1000 mg orally (100 µM) and high affinity enzymes active at
lower substrate concentration (10 µM). At a FLX-concentration of 10 µM, recombinant
CYP3A4, CYP3A7 and to a lesser extent CYP2C9 showed the highest activity in formation
of 5’-HM-FLX, Figure 4A. At 100 µM FLX, CYP3A7 was the most active CYP followed by
CYP3A4. Low activities of 5’-HM-FLX formation were observed, in decreasing order, in
incubations with CYP2C9, CYP2J2, CYP2C8, CYP2E1, CYP3A5, CYP1A1, CYP2D6 and
CYP1A2, Figure 4B. Since CYP3A4, CYP3A7 and CYP2C9 Supersomes showed highest
activity at both concentrations, the enzyme kinetics of 5’-HM-FLX formation was also
determined for these recombinant CYPs (Figures 3B, 3C and 3D). Similar to the incubations
with pooled HLM, Figure 3A, Michaelis-Menten kinetics was observed. Using the Michaelis
Menten model the following enzyme kinetic parameters were determined: CYP3A4, Km 124
± 15 µM, Vmax 197 ± 9 pmol/min/nmol P450; CYP3A7: Km 65 ± 8 µM, Vmax 193 ± 7
pmol/min/nmol P450; and CYP2C9: Km of 508 ± 82 µM, Vmax 67 ± 7 pmol/min/nmol P450.
Effect of isoenzyme-selective CYP-inhibitors on FLX metabolism by HLM.
Figure 5 shows the effects of isoenzyme-selective CYP-inhibitors on the oxidative
metabolism of FLX by pooled HLM. The three inhibitors of CYP3A-isoforms, KCZ, MCZ
and TAO, showed a relatively modest and differential inhibitory effect on 5’-HM-FLX
formation. Even at the highest concentration of KTZ, which inhibits the CYP3A-family, still
only 50% inhibition was observed at both concentrations of FLX. The other CYP3A4-
inhibitors showed only 20 to 40% inhibition at 10 µM FLX-concentration. Although
recombinant CYP2C9 showed much lower activity when compared to CYP3A4, Figures 3
and 4, an almost 90% inhibition of 5’-HM-FLX formation was observed in presence of
This article is protected by copyright. All rights reserved.

sulfaphenazole, which is considered to be a selective CYP2C9-inhibitor. However,
fluvoxamine, the other CYP2C9-inhibitor used did not show significant inhibition of 5’-HM-
FLX formation.
To study whether sulfaphenazole has the ability to inhibit CYP3A-mediated
hydroxylation of FLX, sulfaphenazole was added at different concentrations to incubations of
CYP3A4 and CYP3A7 with FLX and testosterone as substrates. As shown in Figure 6A and
6B, SPZ appeared to be a potent inhibitor of FLX hydroxylation by recombinant CYP3A4
and CYP3A7 with an IC50 of 4.1 ± 0.4 µM and 6.1 ± 0.5 µM respectively. At these SPZ-
concentrations, no inhibition of CYP3A4-catalyzed 6β-hydroxylation of testosterone was
observed. In HLM SPZ could inhibit both diclofenac 4’-hydroxylation and FLX
hydroxylation, but not testosterone hydroxylation, with an IC50 of 2.2 ± 1.1 µM for
diclofenac and an IC50 of 13 ± 1.3 µM for FLX, Figure 6C.
Variability of oxidative metabolism of FLX by individual HLM-fractions and correlation
analysis.
Figure 7 shows the variability of oxidative FLX metabolism which was observed in
incubations of FLX with HLM-fractions from sixteen donors. When incubating with 10 µM
FLX, a 6.5-fold difference was found between the lowest (S1342) and highest (S1441)
metabolising donor. When using 100 µM FLX as substrate concentration, the difference
between the lowest (S1342) and highest (S1343) metabolising donor was 3.3-fold. As shown
in Figure 8 and Table 1, the activities of 5’-HM-FLX formation showed very poor
correlations with activities of most CYP-specific reactions. At both concentrations of FLX,
the highest correlation coefficients were found with CYP2C9-catalyzed diclofenac 4’-
hydroxylation, with Pearson correlation coefficients of 0.78 and 0.58, respectively, and
CYP3A-catalyzed 6ß-hydroxylation, with correlation coefficients of 0.69 and 0.58,
respectively. Visual inspection of the data points of the correlation between CYP2C9 acitivity
and FLX hydroxylation shows a trend that the line does not cross the origin but at 280
pmol/min/mg protein at the x-axis. The fact that none of the correlation coefficiens
approaches unity may support the fact that multiple CYPs contribute to FLX hydroxylation.
Discussion
Previously it has been shown that 5’-HM-FLX appeared to be highly toxic to biliary
epithelial cells variability in the internal exposure to 5’-HM-FLX should be considered as an
important factor determining susceptibility to liver injury (Lakehal et al., 2001). Therefore,
This article is protected by copyright. All rights reserved.

the aim of present study was to further characterize the CYPs involved in hydroxylation of
FLX by using recombinant CYPs, specific inhibitors and correlation analysis. Furthermore,
the enzyme kinetic parameters of FLX hydroxylation were quantified using biosynthetical 5’-
HM-FLX as reference.
The results of the present study confirm that, upon incubation of FLX with HLM or
recombinant CYPs, 5’-HM-FLX is the only oxidative metabolite formed, Figure 2, consistent
with previous studies (Lakehal et al., 2001). By using recombinant CYPs, at both substrate
concentrations used CYP3A4 and CYP3A7 appeared to be the most active enzymes. At 10
µM FLX, CYP2C9 showed a 8-10 fold lower activity, Figure 4A. At 100 µM FLX, low
activities were also observed with CYP1A1, CYP1A2, CYP2C8, CYP2E1, CYP2J2 and
CYP3A5. The fact that Lakehal et al. (Lakehal et al., 2001) did not detect 5’-HM-FLX in
incubations with recombinant CYP1A2, CYP2C9 and CYP3A5 might be attributed to the
less sensitive analytical method used (HPLC-UV detection vs LC-MS) and/or the use of
yeast-expressed CYPs with lower specific activity than the Supersomes used in the present
study. Also, compared to the present study, incubations with recombinant CYPs were
performed at lower temperature (28 oC vs. 37 oC) and lower enzyme concentration (50 nM vs
100 nM CYP).
The present study confirms the previous study of Lakehal et al. (Lakehal et al., 2001).
that CYP3A4 is likely to play a major role in the formation of 5’-HM-FLX, since it is on
average the most abundant CYP in human liver (mean 93 pmol/mg protein) (Achour et al.,
2014). However, the meta-analysis of Achour et al. demonstrated that in the 713 human
livers analyzed CYP3A4 levels varies dramatically, ranging from 0 to 601 pmol/mg
protein(Achour et al., 2014). This large variability can contribute to the interindividual
variability in plasma-concentrations of 5’-HM-FLX (Maier-Salamon et al., 2017). As
mentioned, next to CYP3A4 also CYP3A7 showed high activity in FLX-hydroxylation.
CYP3A7 was originally considered a CYP only expressed in fetal tissues (Schuetz et al.,
1994). However, more recent studies have indicated that in approximately 10% of adult
livers, CYP3A7 is expressed to significant extents, up to 90 pmol/mg protein, which equals
the mean level of CYP3A4 (Sim et al., 2005; Ohtsuki et al., 2012). This strongly increased
CYP3A7 expression appears to be strongly associated with the CYP3A7*1C allele which has
arisen through the conversion of 60 bp of the promotor of CYP3A4 into the corresponding
region of CYP3A7 (Burk et al., 2002). To what extent the variability in CYP3A7-expression
contributes to the variability of FLX-hydroxylation remains to be established, and is
complicated by the present lack of specific inhibitors for the CYP3A-isoforms. The relatively
This article is protected by copyright. All rights reserved.

poor correlation between the rate of FLX-hydroxylation with the CYP-specific reactions,
Figure 8 and Table 1, may be explained by involvement of more than one enzymes.
Involvement of CYP3A-isoforms in drug metabolism is well known to result in very
diverse types of enzyme kinetics and poorly predictable drug-drug interactions. Dependent on
the substrate used, for all CYP3A-isoforms enzyme kinetics can appear as normal Michaelis-
Menten kinetics, or atypical kinetics such as autoactivation and substrate inhibition (Atkins et
al., 2005; Williams et al., 2002). As possible explanations for this complex behavior,
multiple binding sites in the active-site allowing accommodation of multiple substrate
molecules or the presence of an effector binding site has been proposed (Ekins et al., 2003;
Houston et al., 2005). Because of the multiple binding sites in CYP3A-enzymes, also the type
of drug-drug interactions appears complex and strongly substrate dependent (Ekins et al.,
2003; Stresser et al., 2000). Therefore to predict the ability to cause CYP3A4-inhibition the
use of multiple CYP3A4 substrates is recommended (Stresser et al., 2000). The inhibition
experiments in the present study seem to confirm the complicated inhibition pattern of
CYP3A-reactions. The established CYP3A4-inhibitors ketoconazole, troleandomycin and
miconazole showed relatively low degree of inhibition at the concentrations used.
Remarkably, sulfaphenazole, which is considered to be a highly selective inhibitor of
CYP2C9 (Khojasteh et al., 2011) showed almost strong inhibition of FLX-hydroxylation by
pooled HLM, CYP3A4 and CYP3A7, whereas the 6-beta-hydroxylation of testosterone was
not affected. These results suggest that FLX has a unique binding mode to the active site to
CYP3A4, explaining the relatively weak inhibition by 1 µM ketaconazole and an unexpected
strong inhibition by sulfaphenazole. A similar situation has been described recently in a study
where sulfaphenazole was able to significantly inhibit bioactivation of hydroxylapatinib by
HLM, although recombinant CYP2C9 showed only very low activity (Towles et al., 2016).
Therefore, CYP3A family in rare occurrences can be inhibited by sulfaphenazole in a
substrate dependent manner. The results of the present study contrast with the results of
Lakehal et al. which reported no inhibition of HLM-catalyzed hydroxylation of FLX by 25
µM sulfaphenazole (Lakehal et al., 2001). The reason for the discrepancy between these
results remains to be elucidated. A possible explanation might be the different sources of
enzymes used, since the ratio between CYP and reductase as well as the membrane
composition can affect specific activity of CYPs (Kim et al,. 2003).
In conclusion, the results of this present study confirm the involvement of CYP3A4 in
the hydroxylation of FLX, whereas also CYP3A7 and CYP2C9 may contribute significantly
in individuals expressing high hepatic levels. The large variability of both hepatic CYP3A
This article is protected by copyright. All rights reserved.

and CYP2C9 levels may contribute to the variability of plasma-levels of 5’-HM-FLX, which
is suspected to play a role in FLX-induced DILI by causing cytotoxicity to BEC. The strong
inhibition of CYP3A-catalyzed hydroxylation of FLX by the commonly used CYP2C9-
inhibitor SFZ suggest that unanticipated drug-drug interactions can occur with
coadministered drugs.
Acknowledgements
This work was supported by the European Community under the Innovative
Medicines Initiative (IMI) program through Grant Agreement number 115336.
Author contributions
S.J.D. and F.D. performed the experiments; S.J.D. and J.N.M.C. contributed to the writing;
S.J.D., N.P.E.V. and J.N.M.C. contributed to the study design.
Conflict of Interest.
All authors declare that they do not have conflicts of interest.
Declaration of transparency and scientific rigour
This Declaration acknowledges that this paper adheres to the principles for transparent
reporting and scientific rigour of preclinical research as stated in the BJP guidelines for
Design & Analysis, and as recommended by funding agencies, publishers and other
organisations engaged with supporting research.
This article is protected by copyright. All rights reserved.

References
Achour B, Barber J, Rostami-Hodjegan A (2014). Expression of hepatic drug-metabolizing
cytochrome P450 enzymes and their intercorrelations: A meta-analysis. Drug Metab
Dispos 42: 1349–1356.
Adam D, Koeppe P, Heilmann HD. (1983). Pharmacokinetics of amoxicillin and
flucloxacillin following the simultaneous intravenous administration of 4 g and 1 g,
respectively. Infection. 11: 150-154.
Andrews E, Daly AK (2008). Flucloxacillin-induced liver injury. Toxicology 254:158-163
Atkins WM (2005). Non-Michaelis-Menten kinetics in cytochrome P450-catalyzed reactions.
Annu Rev Pharmacol Toxicol. 45: 291-310.
Burk O, Tegude H, Koch I, Hustert E, Wolbold R, Glaeser H, et al. (2002). Molecular
mechanisms of polymorphic CYP3A7 expression in adult human liver and intestine. J
Biol Chem. 277: 24280-24288.
Comuth WJ, Comuth JA, Hauer RNW and Malingré MM (2012) Interaction of flucloxacillin
and quinidine. Eur J Clin Pharmacol 68: 891-893.
Curtis MJ, Bond RA, Spina D, Ahluwalia A, Alexander SP, Giembycz MA, et al.
Experimental design and analysis and their reporting: new guidance for publication in
BJP. Br J Pharmacol. 2015 ;172:3461-3471.
Cynke E, Binet I, Haefeli WE and Thiel G (1999) Flucloxacillin & cyclosporine A: an
unrecognized but relevant interaction in renal transplant recipients. Kidney Int. 55:
1156-1157.
Daly AK, Donaldson PT, Bhatnagar P, Shen Y, Pe’er I, Floratos A, et al. (2009). DILIGEN
Study; International SAE Consortium. HLA-B*5701 genotype is a major determinant
of drug-induced liver injury due to flucloxacillin. Nat Genet. 41: 816-819.
De Abajo FJ, Montero D, Madurga M, Garcia Rodriguez LA. (2004) Acute and clinically
relevant drug-induced liver injury: a population based case-control study. Br. J. Clin.
Pharmacol. 58: 71-80.
Den Braver-Sewradj SP, den Braver MW, van Dijk M, Zhang Y, Dekker SJ, Wijaya L.
(2018) Inter-individual variability in activity of the major drug metabolizing enzymes
in liver homogenates of 20 individuals. Curr. Drug Metab. (E-pub ahead of print)
Derby LE, Jick H, Henry DA, Dean AD (1993). Cholestatic hepatitis associated with
flucloxacillin. Med. J. Aust. 1581: 596-600.
This article is protected by copyright. All rights reserved.

Dijkmans AC, den Hartigh J, van Dissel JT, Burggraaf J (2012).A simplified oral
flucloxacillin absorption test for patients requiring long-term treatment. Therap Drug
Monit 34: 356–358
Eckstein RP, Dowsett JF, Lunzer MR (1993). Flucloxacillin induced liver disease:
histopathological findings at biopsy and autopsy. Pathology 25: 223–228
Ekins, S., Stresser, DM and Williams JA. In vitro and pharmacophore insights into CYP3A
enzymes (2003). TIPS 24: 161-166.
Fairley CK, McNeil JJ, Desmond P, Smallwood R, Young H, Forbes A, et al. (1993) Risk
factors for development of flucloxacillin associated jaundice. Br Med J. 306: 233–235
Gath J, Charles B, Sampson J, Smithurst B (1995). Pharmacokinetics and Bioavailability of
Flucloxacillin in Elderly Hospitalized Patients. J Clin Pharmacol 35: 31–36.
Harding SD, Sharman JL, Faccenda E, Southan C, Pawson AJ, Ireland S et al. (2018). The
IUPHAR/BPS Guide to PHARMACOLOGY in 2018: updates and expansion to
encompass the new guide to IMMUNOPHARMACOLOGY. Nucl Acids Res 46:
D1091-D1106.
Houston JB, Kenworthy KE (2000). In vitro-in vivo scaling of CYP kinetic data not
consistent with the classical Michaelis-Menten model. Drug Metab Dispos. 28: 246-54.
Houston JB, Galetin A (2005). Modelling atypical CYP3A4 kinetics: principles and
pragmatism. Arch Biochem Biophys. 433: 351-360.
Huwyler J, Wright M, Gutmann H, Drewe J (2006). Induction of Cytochrome P450 3A4 and
P-Glycoprotein by the isoxazolyl- penicillin antibiotic flucloxacillin. Curr Drug Metab
7: 119–126.
Jenkins RE, Meng X, Elliott VL, Kitteringham NR, Pirmohamed M, Park BK (2009).
Characterisation of flucloxacillin and 5-hydroxymethyl flucloxacillin haptenated HSA
in vitro and in vivo. Proteomics Clin Appl 3: 720–729.
Khojasteh SC, Prabhu S, Kenny JR, Halladay JS, Lu AYH (2011). Chemical inhibitors of
cytochrome P450 isoforms in human liver microsomes: A re-evaluation of P450
isoform selectivity. Eur J Drug Metab Pharmacokinet 36: 1–16.
Kim K, Ahn T, Yun C (2003) Membrane Properties Induced by Anionic Phospholipids and
Phosphatidylethanolamine Are Critical for the Membrane Binding and Catalytic
Activity of Human Cytochrome P450 3A4. Biochemistry 42: 15377-15387
Koek GH, Stricker BHC, Blok APR, Schalm SW, Desmet VJ (1994). Flucloxacillin-
associated hepatic injury. Liver 14: 225–229
This article is protected by copyright. All rights reserved.

Kumar V, Wahlstrom JL, Rock DA, Warren CJ, Gorman LA, Tracy TS (2006). CYP2C9
inhibition: impact of probe selection and pharmacogenetics on in vitro inhibition
profiles. Drug Metab Dispos 34: 1966-1975.
Lakehal F, Dansette PM, Becquemont L, Lasnier E, Delelo R, Balladur P, et al. (2001)
Indirect cytotoxicity of flucloxacillin toward human biliary epithelium via metabolite
formation in hepatocytes. Chem Res Toxicol. 14: 694-701.
Landersdorfer CB, Kirkpatrick CMJ, Kinzig-Schippers M, Bulitta JB, Holzgrabe U, Drusano,
GL et al. (2007), Population pharmacokinetics at two dose levels and
pharmacodynamic profiling of flucloxacillin. Antimicrob Agents and Chemother. 51:
3290-3297.
Leder K, Turnidge JD, Korman TM, Grayson ML (1999). The clinical efficacy of
continuous-infusion flucloxacillin in serious staphylococcal sepsis. J. Antimicrob.
Chemother. 43: 113-118.
Luirink RA, Dekker SJ, Capoferri L, Janssen LFH, Kuiper CL, Ari ME, Vermeulen NPE,
Vos JC, Commandeur JNM, Geerke DP (2018) A combined computational and
experimental study on selective flucloxacillin hydroxylation by cytochrome P450 BM3
variants. J. Inorg. Biochem. 184:115-122.
Maier-Salamon A, Elgendy SA, Meyer B, Vossen M, Thalhammer T, Thalhammer F, et al.
(2017). Pharmacokinetics of flucloxacillin and its metabolites in patients with renal
failure: impact on liver toxicity. Int J Clin Pharmacol Ther 55: 701-711.
Monshi MM, Faulkner L, Gibson A, Jenkins RE, Farrell J, Earnshaw CJ, et al. (2013).
Human leukocyte antigen (HLA)- B*57:01-restricted activation of drug-specic T cells
provides the immunological basis for flucloxacillin-induced liver injury. Hepatology
57: 727-739.
Muilwijk EW, Dekkers BGJ, Henriet SSV, Verweij PE, Witjes B, Lashof AMLO,
Groeneveld GH, van der Hoeven J, Alffenaar JWC, Russel FGM, van de Veerdonk F2,
Brüggemann RJM. (2017) Flucloxacillin results in suboptimal plasma voriconazole
concentrations. Antimicrob Agents Chemother. 6: e00915-17.
Niwa T, Shiraga T, Takagi A (2009). Effect of antifungal drugs on cytochrome P450 (CYP)
2C9, CYP2C19, and CYP3A4 activities in human liver microsomes. Biol Pharm Bull
28: 1805-1808.
Ohtsuki S, Schaefer O, Kawakami H, Inoue T, Liehner S, Saito A, et al (2012). Simultaneous
absolute protein quantification of transporters, cytochromes P450, and UDP-
glucuronosyltransferases as a novel approach for the characterization of individual
This article is protected by copyright. All rights reserved.

human liver: comparison with mRNA levels and activities. Drug Metab Dispos 40: 83-
92.
Røder BL, Frimodt-Møller N, Espersen F, Rasmussen SN (1995). Dicloxacillin and
flucloxacillin: pharmacokinetics, protein binding and serum bactericidal titers in
healthy subjects after oral administration. Infection 23: 107–112.
Russmann S, Kaye JA, Jick SS, Jick H (2005). Risk of cholestatic liver disease associated
with flucloxacillin and flucloxacillin prescribing habits in the UK: Cohort study using
data from the UK General Practice Research Database. Br J Clin Pharmacol 60: 76–82.
Schuetz JD, Beach DL, Guzelian PS (1994), Selective expression of cytochrome P450 3A
mRNAs in embryonic and adult human liver. Pharmacogenetics 4: 11-20.
Sim SC, Edwards RJ, Boobis AR, Ingelman-Sundberg M (2005). CYP3A7 protein
expression is high in a fraction of adult human livers and partially associated with the
CYP3A7*1C allele. Pharmacogenet Genomics 15: 625-631.
Stresser DM, Blanchard AP, Turner SD, Erve JCL, Dandeneau AA, Miller VP, et al (2000).
Substrate-dependent modeulation of CYP3A4 catalytic activity: analysis of 27 test
compounds with four fluorimetric substrates. Drug Metab Dispos 28: 1440–1448.
Sutherland R, Croydon EAP, Rolinson GN (1970). Flucloxacillin, a New Isoxazolyl
Penicillin, Compared with Oxacillin, Cloxacillin, and Dicloxacillin. Br Med J 4: 455–
460.
Tailor A, Faulkner L, Naisbitt DJ, Park BK (2015). The chemical, genetic and immunological
basis of idiosyncratic drug-induced liver injury. Hum Exp Toxicol 34: 1310-1317.
Thijssen HHW and Wolters J (1982). The metabolic disposition of flucloxacillin in patients
with impaired kidney function. Eur J Clin Pharmacol 22: 429–434
Towles JK, Clark RN, Wahlin MD, Uttamsingh V, Rettie AE, Jackson KD (2016).
Cytochrome P450 3A4 and CYP3A5-catalyzed bioactivation of lapatinib. Drug Metab
Dispos 44: 1584-1597.
Uetrecht J (2007). Idiosyncratic drug reactions: current understanding. Annu Rev Pharmacol
Toxicol 47: 513–539.
Williams JA, Ring BJ, Cantrell VE, Jones DR, Eckstein J, Ruterbories K, et al (2002).
Comparative metabolic capabilities of CYP3A4, CYP3A5, and CYP3A7. Drug Metab
Dispos 30: 883-891.
Wuillemin N, Adam J, Fontana S, Krähenbühl S, Pichler WJ, Yerly D (2013). HLA
haplotype determines hapten or p-i T cell reactivity to flucloxacillin. J Immunol. 190:
4956-4964.
This article is protected by copyright. All rights reserved.

Wuillemin N, Terracciano L, Beltraminelli H, Schlapbach C, Fontana S, Krahenbuhl S, et al.
(2014). T cells infiltrate the liver and kill hepatocytes in HLA-B(*)57:01-associated
oxacillin-induced liver injury. Am J Pathol 184: 1677-1682.
This article is protected by copyright. All rights reserved.

Figure 1: Oxidative metabolism of flucloxacillin (FLX) to 5’-hydroxymethyl fluclocaxillin
(5’-HM-FLX) and hydrolysis of FLX and 5’-HM-FLX to their corresponding penicilloic
acids.
This article is protected by copyright. All rights reserved.

Figure 2: LC-MS-analysis of an incubation of 100 µM FLX incubated for 60 minutes with
0.5 mg/mL pooled human liver microsomes with (solid line) or without (dotted line)
NADPH-regenerating system. (A) Extracted ion chromatogram of m/z 470.052
corresponding to the [M+H]+ of 5’-HM-FLX; (B) extracted ion chromatogram of m/z
472.078 corresponding to the [M+H]+ of the penicilloic acid of 5’-HM-FLX; (C) extracted
ion chromatogram of m/z 454.059 corresponding to the [M+H]+ of FLX.
This article is protected by copyright. All rights reserved.

Figure 3: Concentration dependency of the rate of FLX-hydroxylation as catalyzed by 0.5
mg/mL pooled HLM (A), 100 nM CYP3A4 Supersomes (B), 100 nM CYP3A7 Supersomes
(C) and 100 nM CYP2C9 supersomes (D). Each condition was incubated in duplicate and
each data point represent as mean whereas error bars represent the range of duplicate
measurements. Enzymes kinetic parameters obtained by non-linear regression according to
the Michaelis Menten-equation were: (A) HLM: Km 284 ± 38 µM and Vmax 30 ± 2
pmol/min/mg protein; (B) CYP3A4 Supersomes: Km 124 ± 15 µM and Vmax 197 ± 9
This article is protected by copyright. All rights reserved.

pmol/min/nmol P450, (C) CYP3A7 Supersomes: Km of 65 ± 8 µM and a Vmax 193 ± 7
pmol/min/nmol P450 and (D) CYP2C9 supersomes: Km of 508 ± 82 µM and a Vmax 67 ± 7
pmol/min/nmol P450.
This article is protected by copyright. All rights reserved.

Figure 4: Specific activities of formation of 5’-HM-FLX in incubations of FLX with CYP-
containing Supersomes (100 nM). (A) Incubations in presence of 10 µM FLX; (B)
incubations in presence of 100 µM FLX. Data represent mean of duplicate experiments; the
error bars represent the range of duplicate measurements.
This article is protected by copyright. All rights reserved.

Figure 5: The inhibition of 5'-HM-FLX formation by pooled HLM (0.5 mg/mL) by isoform-
specific inhibitors of CYPs. (A) Incubations performed in presence of 10 µM FLX; (B)
This article is protected by copyright. All rights reserved.

incubations performed in presence of 100 µM FLX. Activities are relative to the non-
inhibited reactions which showed specific activities of 0.52 pmol/min/mg protein (A) and 7.4
pmol/min/mg protein (B), respectively. 100% was defined as the mean of enzyme activities
in absence of inhibitor and 0% was defined as no enzyme activity. Data represent the
averages of a three independent experiments in with each condition was performed in
duplicate; error bars represent the standard deviation of the six individual measurements.
This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.