Clementine juice has the potential for drug interactions – In vitro comparison
with grapefruit and mandarin juice

Dirk Theile, Nicolas Hohmann, Dominik Kiemel, Giuseppe Gattuso,

Davide Barreca, Gerd Mikus, Walter Emil Haefeli, Vedat Schwenger, Johanna
Weiss

PII: S0928-0987(16)30514-0
DOI: doi: 10.1016/j.ejps.2016.11.021
Reference: PHASCI 3807

To appear in:

Received date: 8 June 2016

Revised date: 7 November 2016
Accepted date: 19 November 2016

Please cite this article as: Theile, Dirk, Hohmann, Nicolas, Kiemel, Dominik, Gattuso,
Giuseppe, Barreca, Davide, Mikus, Gerd, Haefeli, Walter Emil, Schwenger, Vedat, Weiss,
Johanna, Clementine juice has the potential for drug interactions – In vitro comparison
with grapefruit and mandarin juice, (2016), doi: 10.1016/j.ejps.2016.11.021

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. 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.
 ACCEPTED MANUSCRIPT
1

Clementine juice has the potential for drug interactions – in vitro comparison with

grapefruit and mandarin juice

T
Dirk Theilea*, Nicolas Hohmanna*, Dominik Kiemela, Giuseppe Gattusob, Davide Barrecab,

IP
Gerd Mikusa, Walter Emil Haefelia, Vedat Schwengerb,c#, Johanna Weissa#

R
SC
a
Department of Clinical Pharmacology and Pharmacoepidemiology, University of Heidelberg,
Im Neuenheimer Feld 410, 69120 Heidelberg, Germany

NU
b
Dipartimento di Scienze Chimiche, Biologiche, Farmaceutiche ed Ambientali,
Università di Messina, Viale F. Stagno d'Alcontres 31, 98166 Messina, Italy
MA
c
Department of Endocrinology, Metabolism and Clinical Chemistry, Section of Nephrology,
University of Heidelberg, Im Neuenheimer Feld 162, 69120 Heidelberg, Germany
d
Clinical Centre Stuttgart, Clinic for Kidney and Hypertension Diseases, Kriegsbergstr. 60,
D

70174 Stuttgart, Germany

TE

* # These authors contributed equally.

P
CE
AC

Corresponding author:

Prof. Johanna Weiss, PhD

University Hospital Heidelberg

Department of Clinical Pharmacology and Pharmacoepidemiology

Im Neuenheimer Feld 410, 69120 Heidelberg, Germany

Tel: +49-6221/56-39402

Fax +49-6221/56-4642

Email address: [email protected]

 ACCEPTED MANUSCRIPT
2

Abstract

Adverse drug interactions due to grapefruit juice are well known prompting warnings even in

drug labels. Similar issues have not been reported for clementines and available data is scarce,

T
despite of genetic descent. We observed substantially increased tacrolimus trough

IP
concentrations in a renal transplant patient consuming high clementine amounts and, thus,

R
scrutinised the effects of clementine juice on drug metabolism and drug transporters in vitro

SC
and compared it to the effects of mandarin and grapefruit juice. All citrus juices profoundly

induced several drug transporters and drug metabolising enzymes, whereas the effects of

NU
grapefruit juice were most pronounced (e.g. 156-fold and 34-fold induction of cytochrome
MA
P450 (CYP) 3A4 mRNA by grapefruit juice and clementine juice, respectively). However, the

juices also inhibited e.g. CYP3A4, raising the question which effect prevails in vivo. Using an

enzymatic activity assay, we demonstrated that at least in vitro CYP3A4 inhibition prevails
D
TE

for both grapefruit and clementine juice, whereas for CYP1A2 induction appears to

predominate. Thus, inhibition of CYP3A4 is presumably the underlying reason for the
P

observed increase in the concentrations of the CYP3A4 substrate tacrolimus in the patient.
CE

Taken together, our data indicate that clementine juice as well as grapefruit juice and to a
AC

lesser extent also mandarin juice can induce several important drug metabolising enzymes and

drug transporters, but also inhibit some of these proteins. Our data indicate that clementine

juice similar to grapefruit juice bears the potential for profound interactions with drugs

potentially leading to adverse drug effects e.g. through over-exposure to CYP3A4 substrates.

Keywords:

Clementine / CYPs / drug transporters / grapefruit / mandarin / HPLC-DAD-ESI-MS-MS

 ACCEPTED MANUSCRIPT
3

1. Introduction

In November 2013, a 27-year old female with a functional living kidney allograft who was

well adjusted since July 2013 to an immunosuppressive therapeutic regime comprising total

T
daily oral doses of 4.5 mg tacrolimus, 1440 mg mycophenolate, and 4 mg methylprednisolone

IP
suddenly presented with a significantly increased tacrolimus whole blood trough level of 17.4

R
µg/ml during routine follow-up, which was ~2.5-fold increased compared to a previously

SC
measured concentration with the same dosing regimen. Recent medical history and physical

examination were unremarkable and a check for drug interactions was inconspicuous. Only

NU
one particular life style change, the consumption of large amounts (>1 kg/d) of clementines
MA
during autumn months, was detected. The patient was advised to stop clementine intake and

tacrolimus levels checked 14 days later had returned to a trough level (7.8 µg/ml) close to the

target concentration range of 3 – 7 ng/ml. Upon re-challenge with clementines, tacrolimus

D
TE

levels again increased to 19.6 µg/ml within another 14 days. With a positive de-challenge and

re-challenge, a definite causal relationship was established according to the WHO-UMC

P

causality scale and a probable relationship (6 points) was determined through the Naranjo
CE

algorithm. In consequence, the patient was advised to refrain from excessive clementine
AC

intake at home. Surprised by the observation, we investigated the possible underlying

pharmacokinetic mechanisms of this phenomenon in vitro.

The taxonomy of citrus fruits is complex and the nomenclature inconsistent. According to

Mabberley, there are only three edible species of origin: the citron (Citrus medica), the

pomelo (Citrus maxima), and the mandarin (Citrus reticulata) (Mabberley, 1997). Sweet

oranges (Citrus sinensis) represent crossbreeds of pomelo and mandarin, clementines (Citrus

clementina) are crossbreeds of sweet orange and mandarin, and grapefruits (Citrus paradisi)

are hybrids of pomelo and sweet orange (Nicolosi et al., 2000; Barrett et al., 1976). According

to genetic analyses, grapefruits share 70.4 % identity with pomelos and only 16.7 % with

clementines (Garcia-Lor et al., 2013).

 ACCEPTED MANUSCRIPT
4

To the best of our knowledge, there has been no interaction reported so far between

clementines or clementine juice with drugs. Only for mandarin juice there are a few reports

indicating a lack of interaction with CYP3A4 substrates such as cyclosporine or midazolam

T
(Sorkhi et al., 2007, 2008; Backman et al., 2000). In contrast, for grapefruit juice numerous

IP
interactions with drugs are known (Seden et al., 2010), including grapefruit juice-mediated

R
increases of tacrolimus bioavailability (Liu et al., 2009; Peynaud et al., 2007). Given the fact

SC
that clementines are genetically different from grapefruit and mandarins, they might exhibit a

fundamentally different interaction profile.

NU
Because tacrolimus pharmacokinetics is relevantly influenced by drug transporters such as P-
MA
glycoprotein (P-gp, ABCB1) (Iwasaki, 2007) and the cytochrome P450 (CYP) isozymes 3A4

and CYP3A5 and thus the respective inhibitors, inducers (Iwasaki, 2007), or genetic

polymorphisms (Provenzani et al., 2013), we hypothesised that inhibition of CYP3A4 and/or

D
TE

P-gp by clementines might be the underlying reason for the increased tacrolimus trough

levels. Consequently, we scrutinised the effect of clementine juice on the expression and
P

activity of CYP3A4 and P-gp and also included other important drug metabolising enzymes
CE

and drug transporters and compared it to the effects of mandarin juice and the partly well-
AC

known effects of grapefruit juice. The citrus juices used were also analysed for their flavonoid

and furanocoumarin content by liquid chromatography coupled to tandem mass-spectrometry

(LC-MS-MS).
 ACCEPTED MANUSCRIPT
5

2. Materials and methods

2.1 Materials

Cell culture media, phosphate buffered saline, (PBS), supplements, anti-β-actin (Clone AC-

T
74), aprotinin, and the GenElute™ Mammalian Total RNA Miniprep Kit were purchased

IP
from Sigma-Aldrich (Taufkirchen, Germany). Dimethyl sulfoxide (DMSO), crystal violet,

R
sodium dodecylsulphate (SDS), rifampicin, Tris-hydroxymethyl-aminomethan (TRIS),

SC
dithiothreitol (DTT), and Tween®20 were from AppliChem (Darmstadt, Germany). Foetal

calf serum (FCS) was obtained from Biochrom (Berlin, Germany). Rhodamine123 and the

NU
antibody against human P-glycoprotein (P-gp) clone C219 were from Calbiochem
MA
(Darmstadt, Germany), the antibodies against human CYP1A2 (clone D-3) and human

CYP3A4 (clone C-17) and the secondary donkey anti-goat antibody were from Santa Cruz

(Heidelberg, Germany), the secondary anti-mouse antibody was from GE Healthcare

D
TE

(Freiburg, Germany), and Rotiphorese® gel 30 was obtained from Carl Roth GmbH

(Karlsruhe, Germany). 2,3,7,8-tetrachlordibenzo-p-dioxin (TCDD, dioxin) was purchased

P

from LGC Standards GmbH (Wesel, Germany). Pheophorbide A (PhA) was from Frontier
CE

Scientific Europe (Carnforth, UK). Calcein acetoxymethylester (calcein-AM) was obtained

AC

from Invitrogen (Karlsruhe, Germany). 8-Fluorescein-cAMP (8-FcA) was purchased from

BIOLOG Life Science Institute (Bremen, Germany). Pefabloc was obtained from Serva

(Heidelberg, Germany), bromphenol blue, leupeptin, and pepstatin were from Biomol

(Hamburg, Germany). The RevertAid H Minus First Strand cDNA Synthesis Kit was

obtained from Fermentas (St.Leon-Rot, Germany). The Absolute QPCR SYBR Green Mix

was supplied by Abgene (Hamburg, Germany) and the QuantiTect ® Primer Assay by Qiagen

(Hilden, Germany). Primers were synthesised by Eurofins MWG Operon (Ebersberg,

Gemany). The SuperSignal®West Pico Chemiluminescent Substrate Kit and the BCA®

Protein Assay Kit and were purchased from Pierce (Rockford, USA). The nitrocellulose

membranes (Optitran BA-S 85) from Schleicher&Schuell BioScience (Dassel, Germany). The
 ACCEPTED MANUSCRIPT
6

P450-Glo CYP1A2 Induction/Inhibition Assay, the P450-Glo CYP1A2 Screening System

Dual-Glo, the Steady-Glo Luciferase Assay Systems, the P450-Glo CYP3A4 Assay with

Luciferin-IPA, the pGL4.21 vector, the pGL4.74 [hRluc/TK] renilla vector, and FuGENE®

T
HD Transfection Reagent were obtained from Promega Corporation (Madison, USA). The

IP
NR1I2 (NM_003889) Human cDNATrueClone® (pCMV6-XL4 vector containing the cDNA

R
of the PXR gene NR1I2) was obtained from OriGene (Rockville, USA). Gaze (pore size 150

SC
µm) was obtained from Neolab (Heidelberg, Germany). LY335979 was a kind gift of Eli Lilly

Company (Bad Homburg, Germany). HPLC-grade acetonitrile and water were supplied by

NU
Sigma-Aldrich (St. Louis, MO, USA), dimethylformamide (DMF) by Carlo Erba (Milano,
MA
Italy). Hesperidin, eriocitrin, narirutin, didymin, poncirin, sinensetin, nobiletin, tangeretin,

naringin, neohesperidin, bergamottin and epoxybergamottin were supplied from

D

Extrasynthèse (Genay, France). 6,8-di-C-Glu-apigenin and lucenin-2 4'-methyl ether were

TE

separated from Citrus limetta (Barreca et al., 2011), chrysoeriol 7-O-neohesperidoside from

Citrus bergamia (Barreca et al., 2016), and they were used as standards. The Iso-Disc P-34, 3
P

mm diameter PTFE membrane (0.45 m pore size) was obtained from Supelco (Bellefonte,
CE

PA, USA). Grapefruit juice (Citrus paradisi, not-from concentrate) was purchased from Aldi
AC

Süd GmbH & Co KG and mandarin juice (Citrus reticulata, not-from concentrate) was from

Völkel GmbH (Höhbeck, Germany).

2.2 Preparation of fresh clementine juice

Spanish clementines (Citrus clementina, var. clemenules) of the same lot the patient had

consumed were peeled and homogenised with a blender. The mixture was filtered with a sieve

and the filtrate centrifuged at 9000 x g for 5 min. The supernatant was then filtered with fine

gaze and centrifuged again at 9000 x g for 5 min and subsequently adjusted to pH 7.0 with 10
 ACCEPTED MANUSCRIPT
7

M NaOH, and frozen at -80 °C as aliquots. To minimise oxidative loss of ingredients, only

freshly thawed aliquots were used.

Grapefruit juice and mandarin juice were also adjusted to pH 7.0 and frozen in aliquots at -80

T
°C.

R IP
2.3 Identification and quantification of flavonoids and furanocoumarins

SC
2.3.1 Sample preparation

Prior to analysis, the juices (10.0 ml) were mixed with DMF (10.0 ml) and the mixture was

NU
centrifuged for 5 min at 2300 g. The supernatant liquid was then filtered through an Iso-Disc

P-34, 3 mm diameter PTFE membrane, 0.45 m pore size.

MA
2.3.2 LC-MS-MS analysis of flavonoids
D

LC-MS-MS analyses of juice samples were carried out with a ThermoQuest Model LCQ-Duo
TE

equipped with a diode array spectrophotometer and an ion trap mass spectrometer with an

electrospray ionisation source (ESI). Separation of each compound was performed on a 150
P

mm  4.6 mm i.d., Kinetex® 5 µm XB-C18, supplied by Phenomenex (Torrance, USA). The

CE

column was placed in a column oven set at 30 °C. The injection loop was 20 l, and the flow-
AC

rate was 0.5 ml/min. The mobile phase consisted of a linear gradient of acetonitrile in H2O as

follows: 5–20 % (0–15 min), 20–30 % (15–20 min), 30-100 % (20–35 min), 100 % (35–40

min), 100-5 % (40–45 min), and 5 % (45–55 min). UV spectra were recorded between 200

and 450 nm, and simultaneous detection by diode array was performed at 278 and 325 nm.

Operating parameters of the mass spectrometer were set as follows: capillary temperature 250

°C; spray needle voltage set at 4.50 kV; ES capillary voltage +3 and –47 V for positive and

negative polarity, respectively; tube lens offset 0 and –25 V for positive and negative polarity,

respectively. Nitrogen was used as a sheath gas with a flow of 50 arbitrary units. Mass

analysis was carried out in full-scan mode in the 80-900 amu range, both in positive and
 ACCEPTED MANUSCRIPT
8

negative mode. MS-MS spectra were obtained using an applied collision energy of 20–30 %

of instrument maximum. A source fragmentation of 20 V as a collision energy was used in

MS and MS-MS analysis when required. Each sample was tested three times and gave

T
superimposable chromatograms. Identification of the compoundswere performed by their

IP
retention time, UV spectra, MS and MS-MS data and by comparison with analytical standard

R
samples. The identified compounds were quantified by comparing the integrated peak areas

SC
with that standard compounds (0.1–600 mg/l). They have been analysed twice and used for

calibration of the peak areas. To confirm the linearity and reproducibility for the

NU
quantification of the analytes, the standards were analysed twice, and linear calibration curves
MA
were constructed from the averaged peak areas, thus obtaining the limit of detection (LOD,

0.012–0.027 g/ml) and the limit of quantification (LOQ, 0.036–0.081 g/ml) for all the
D

analytes. LOD was defined as a signal to noise ratio of 3. The LOQ was calculated as three
TE

times the LOD. Recovery of analytes was always >98%. Intra- and interday coefficients of

variation were below 7% for all analytes.

P
CE

2.4 Cell lines used

AC

2.4.1 P388 and P388/dx cells

The murine monocytic leukaemia cell line P388/dx cells over-expressing murine P-gp

(mdr1a/b) (Boesch et al., 1991) and the corresponding parental cell line P388 were used to

test the fruit juices for their potential to inhibit P-gp. Both cell lines were kindly provided by

Dr. D. Ballinari (Pharmacia & Upjohn, Milano, Italy). Cells were cultured under standard cell

culture conditions with RPMI 1640 medium supplemented with 10 % FCS, 2 mM glutamine,

500 mM -mercaptoethanol, 100 U/ml penicillin, and 100 µg/ml streptomycin sulphate. To

maintain P-gp expression, the culture medium for P388/dx contained 0.43 µM doxorubicin.

One day before the inhibition assay, both cell lines were fed with doxorubicin-free culture

medium.
 ACCEPTED MANUSCRIPT
9

2. 4.2 MDCKII-BCRP cells

MDCKII-BCRP cells over-expressing human breast cancer resistance protein (BCRP) (Pavek

T
et al., 2005) were used to quantify BCRP inhibition by citrus juices. The parental cell line

IP
MDCKII (available at ATCC) was used as a control. BCRP over-expressing cell line was

R
kindly provided by Dr. A. H. Schinkel (The Netherlands Cancer Institute, Amsterdam, The

SC
Netherlands). Cells were cultured under standard cell culture conditions in DMEM containing

10 % FCS, 2 mM glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin sulphate.

NU
MA
2.4.3 LS180 cells

The human colon adenocarcinoma cell line LS180 (available at ATCC, Manassas, VA, USA)

was used for induction experiments as a surrogate for the intestine being a major site of drug
D
TE

interactions and being ideal for investigating pregnane X receptor (PXR) and aryl

hydrocarbon receptor (AhR) mediated induction (Harmsen et al., 2008; Gupta et al., 2008;
P

Weiss et al., 2011; Brandin et al., 2007; Yamasaki et al., 2009; Li et al., 1998; Harper et al.,
CE

1991). Moreover, LS180 cells were used for the PXR reporter gene assay. Cells were cultured
AC

under standard cell culture conditions in DMEM supplemented with 10 % FCS, 2 mM

glutamine, 0.1 mM non-essential amino acids, 100 U/ml penicillin and 100 µg/ml

streptomycin sulphate.

2.4.4 HEK-OATP1B1 and HEK-OATP1B3 cells

The human embryonic kidney cell line HEK293 stably transfected with organic anion

transporter polypeptide 1B1 (OATP1B1) (HEK-OATP1B1), OATP1B3 (HEK-OATP1B3), or

the empty control vector (HEK293-VC G418) were used to assess inhibition of OATP1B1

and OATP1B3 (König et al., 2000a,b). Cell lines were kindly provided by Dr. D. Keppler

(German Cancer Research Centre, Heidelberg, Germany). Cells were cultured under standard
 ACCEPTED MANUSCRIPT
10

cell culture conditions with DMEM supplemented with 10 % FCS, 2 mM glutamine, 100

U/ml penicillin, 100 µg/ml streptomycin sulphate, and 800 µg/ml G418.

T
2.4.5 AZ-AhR cells

IP
AZ-AhR cells (human hepatoma HepG2 cells stably transfected with a construct containing

R
several AhR binding sites upstream of a luciferase reporter gene) (Novotna et al., 2011) were

SC
used for the AhR reporter gene assay. Cells were kindly provided by Dr. Zdenek Dvorak

(Olomouc, Czech Republic). Cells were cultured under standard cell culture conditions in

NU
DMEM supplemented with 10 % FCS, 2 mM glutamine, 100 U/ml penicillin, 100 µg/ml
MA
streptomycin sulphate, and 0.2 mg/ml hygromycin (every second passage).

2.5. P-gp inhibition assays (calcein uptake and rhodamine123 efflux assay)
D
TE

For investigating P-gp inhibition, a calcein assay was performed as published previously

(Fröhlich et al., 2004) using the P-gp over-expressing cell line P388/dx (Boesch et al., 1991)
P

and the parental cell line P388 as a control. Due to unspecific effects observed in the calcein
CE

assay we also used the rhodamine123 efflux assay as alternative as published previously
AC

(Storch et al., 2007). Each concentration (0.01 – 20 %) was tested in octuplet and each

experiment was performed in triplicate.

2.6 BCRP inhibition assay

Using MDCKII-BCRP cells (over-expressing human breast cancer resistance protein

(BCRP)) in comparison to the parental cell line MDCKII and pheophorbide A as specific

BCRP substrate, BCRP inhibition assays were conducted and measured by flow cytometry as

described and validated previously (Weiss et al., 2007). Each concentration (0.01 – 20 %) was

tested in octuplet and each experiment was performed in triplicate.

 ACCEPTED MANUSCRIPT
11

2.7 OATP inhibition assay

Inhibition of human organic anion transporting polypeptide 1B1 (OATP1B1) and OATP1B3

was quantified by flow cytometry assessing the uptake of 8-FcA into HEK293 cells over-

T
expressing the respective transporter (König et al., 2000a,b). The performance of the assay

IP
has been published previously (Weiss et al., 2013). Each concentration (0.1 – 20 %) was

R
tested in octuplet and each experiment was performed in triplicate

SC
2.8 Inhibition of CYP3A4 and CYP1A2

NU
Inhibition of CYP3A4 and CYP1A2 was assessed with the P450-Glo CYP3A4 Assay and
MA
the P450-Glo CYP1A2 Screening System according to the manufacturer’s instructions. The

kits contain a luminogenic substrate (luciferin-IPA for CYP3A4 and Luciferin-ME for
D

CYP1A2), which is converted by CYP3A4 or CYP1A2 into luciferin, generating light when
TE

incubated with the luciferin detection reagent of the kit. The citrus juices were tested for their

capacity to inhibit the production of the luminescent signal. Eight concentrations in triplicates
P

ranging from 0.2 to 50 % for CYP3A4 and from 0.0125 to 25 % for CYP1A2 were tested and
CE

each experiment was conducted thrice.

AC

2.9 Growth inhibition assay

Before induction experiments, antiproliferative effects of citrus juices were investigated in

LS180 cells to exclude an influence on cell proliferation. Proliferation was quantified by

crystal violet staining and the assays were conducted as described previously (Peters et al.,

2006). Each concentration was tested in octuplet and each experiment was performed in

quadruplicate. The IC20 values for proliferation inhibition were 13.4 ± 0.7 % (grapefruit

juice), 30.0 ± 12.7 % (clementine juice), and 18.0 ± 0.1 % (mandarin juice). Therefore, the

maximal concentration for the induction assay was set to 20 % for all juices to ensure that at

least 80 % of the cells survive and to use identical concentrations for all juices.
 ACCEPTED MANUSCRIPT
12

2.10 Induction assay

LS180 cells were seeded in culturing flasks and incubated for three days. Medium was then

T
changed to medium containing one of the juices (1 %, 2 %, 10 %, or 20 %) and cells were

IP
incubated for four consecutive days. Rifampicin (20 µM) served as a positive control for

R
PXR-driven genes and compound-free medium as a negative control. All media were adjusted

SC
to 0.03 % DMSO. After harvesting, cell samples were divided for RNA and protein

extraction. Each experiment was conducted in quadruplicate.

NU
MA
2.11 Quantification of mRNA expression by real-time RT-PCR

After four days of treatment with the respective juices, RNA was isolated using the GeneElute
D

Mammalian Total RNA Miniprep Kit and cDNA was synthesised with the RevertAid H
TE

Minus First Strand cDNA Synthesis Kit according to the manufacturers’ instructions. Gene

expression was quantified by real-time RT-PCR as described previously (Albermann et al.,

P

2005; Weiss et al., 2011) for the following genes: CYP1A1, CYP1A2, CYP3A4, CYP3A5,
CE

ABCB1, ABCC1 (multidrug resistance-associated protein 1 (MRP1), ABCC2 (MRP2), ABCC3

AC

(MRP3), ABCG2 (breast cancer resistance protein, BCRP), SLCO1B1 (solute carrier of

organic anion transporter 1B1, OATP1B1), UGT1A3 (UDP glucuronosyltransferase 1A3),

UGT1A9, and UGT2B7. Primer sequences and PCR conditions were published previously

(Albermann et al., 2005; König et al., 2010; Weiss et al., 2011; Dvorak et al., 2008 (CYP1A1;

Ayed-Boussema et al., 2011 (CYP1A2)).

The most stable housekeeping gene for normalisation was identified using geNorm (version

3.4, Center for Medical Genetics, Ghent, Belgium), which determines most stable reference

genes from a set of tested genes in a given cDNA sample panel (Vandesompele et al., 2002).

Among a panel of 7 housekeeping genes tested, glucuronidase β (GU) was the most stable

under the treatment with the citrus juices and was thus used for normalisation. Data were
 ACCEPTED MANUSCRIPT
13

evaluated as described previously (Albermann et al., 2005). All samples were amplified in

duplicate and the mean of the technical duplicate was used for further calculation.

T
2.12 Western blot analysis of P-gp, CYP1A1, and CYP3A4 expression

IP
To exemplarily verify gene induction observed with the citrus juices, protein expression of

R
CYP1A2, CYP3A4, and P-gp was analysed at least in triplicate by SDS-PAGE and western

SC
blotting according to standard protocols. Cell lysates containing 40 µg protein were mixed

with 5 x sample buffer (containing TRIS-HCl, SDS, DTT, bromphenol blue, and glycerol,

NU
aprotinin, pepstatin, and leupeptin) and then subjected to a 10 % SDS-PAGE and

electrotransferred to nitrocellulose nitrate membranes. After blocking, protein detection was

MA
carried out with a murine monoclonal antibody against human P-gp (clone C219, diluted

1:100 in TRIS-buffered saline containing 0.1 % Tween®20), or human CYP3A4 (clone C-17,
D

diluted 1:200), or human CYP1A2 (clone D-3, diluted 1:200), or -actin (Clone AC-74;
TE

diluted 1:40000). After extensive washing, blots were incubated with horseradish peroxidase-
P

linked secondary antibodies and bands were visualised by enhanced chemiluminescence using
CE

the SuperSignal®West Pico Chemiluminescent Substrate Kit and semi-quantified by

AC

FluorChem Q SA AlphaView Version 3.2.2, Cell Biosciences (Santa Clara, USA). Expression

was normalised to the loading control β-actin and the untreated medium control. Each blot

was conducted at least in triplicate.

2.13 PXR reporter gene assay

To investigate whether the citrus juices (0.5-20 %) can activate the nuclear receptor PXR,

reporter gene assays were conducted using the Dual-Glo Luciferase Assay System

according to the manufacturer’s instructions as published previously (Weiss et al., 2013).

Increases of PXR activity after 24 h incubation with the citrus juices were normalised to the
 ACCEPTED MANUSCRIPT
14

transfection efficiency control (renilla luminescence) and to the PXR activity of non-treated

controls set to 1 (= 100 %). Each juice was tested in triplicate.

T
2.14 AhR reporter gene assay

IP
To investigate whether the citrus juices (0.5-20 %) can activate the nuclear receptor AhR, a

R
reporter gene assay for AhR was conducted using the Steady-Glo Luciferase Assay System

SC
according to the manufacturer’s instructions as described previously (Weiss et al., 2014).

Citrus juice-induced increases of AhR activity were normalised to the activity of non-treated

controls. Each juice was tested in triplicate.

NU
MA
2.15 Quantification of CYP3A4 activity
D

To quantify CYP3A4 function in LS180 cells after 4 days of incubation with the citrus juices,
TE

CYP3A4 activity was measured with the P450-Glo CYP3A4 Assay (Luciferin-IPA)

according to manufacturer’s instructions as published previously (Weiss et al., 2012). Fold

P

changes were calculated by dividing treated values by the mean of the negative control values.
CE
AC

2.16 Quantification of CYP1A2 activity

To quantify CYP1A2 function in LS180 cells after 4 days of incubation with the citrus juices,

CYP1A2 activity was measured with the P450-Glo CYP1A2 Induction/Inhibition Assay

according to manufacturer’s instructions. In brief, cells on a 96 well plate were washed and

equilibrated with PBS containing 2 % FCS. Cells weren then incubated in 50 µl medium

containing 6 µM Luciferin-1A2 at 37 °C for 60 min. 25 µl of the solution of each well were

transferred to a 96-well luminometer plate and 25 µl Luciferin Detection Reagent were added.

After incubation for 20 min at room temperature, luminescence was measured in a Glomax 96

microplate luminometer (Promega Corporation (Madison, USA). Net signals were calculated
 ACCEPTED MANUSCRIPT
15

by subtracting background luminescence values and normalised to the cell count assessed by

crystal violet staining (Peters et al., 2006). Fold changes were calculated by dividing treated

values by the mean of the negative control values.

T
IP
2.17 Statistical analysis

R
Data were analysed using GraphPad Prism Version 6.02 and InStat Version 3.06 (GraphPad

SC
Software, San Diego, USA). IC50 values were calculated using the four-parameter fit

(sigmoidal dose-response curves with variable slope). Differences between mRNA expression

NU
and CYP1A1/CYP3A4 activity following incubation with the investigated compounds and the

respective vehicle controls were tested using ANOVA with Dunnett’s post hoc test. A p-value
MA
<0.05 was considered significant.
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
16

3. Results

3.1 Transporter inhibition by citrus juices

Possible P-gp inhibition by the citrus juices was investigated by means of the calcein assay.

T
Instead of increasing intracellular calcein fluorescence, citrus juices decreased it in P388/dx

IP
cells (over-expressing P-gp) as well as in parental cells P388. This suggested an unspecific

R
effect of the juices on the assay either by quenching the calcein fluorescence or by inhibiting

SC
intracellular esterases necessary to convert calcein-AM to calcein.

Thus, P-gp inhibition was assessed by an additional assay using rhodamine123 as P-gp

NU
substrate. In the rhodamine123 efflux assay, the juices at concentrations up to 20 % had no
MA
effect on intracellular fluorescence neither in the P-gp over-expressing cell line P388/dx nor

in the parental cell line P388, indicating absent P-gp inhibition. In contrast, the specific P-gp

inhibitor LY335979 (0.1 µM), which was used as a positive control, increased intracellular
D
TE

rhodamine123 fluorescence about 10-fold in P388/dx but not in P388 cells.

All citrus juices inhibited BCRP at low concentrations, with mandarin juice revealing the
P

most potent and clementine juice the weakest effect (Table 1).
CE

OATP1B1 and OATP1B3 were also inhibited by all juices at lower concentrations with
AC

grapefruit juice being the most potent inhibitor (Table 1).

3.2 CYP3A4 and CYP1A2 inhibition by citrus juices

All citrus juices inhibited CYP3A4 activity in the P450-Glo CYP3A4 assay, albeit with

different potency. Whereas grapefruit juice was a very potent CYP3A4 inhibitor, clementine

and mandarin juice were much weaker (Table 1). CYP1A2 was considerably inhibited by all

citrus juices with grapefruit juice being most potent (Table 1).

3.3 Induction of drug transporters and drug metabolising enzymes by citrus juices
 ACCEPTED MANUSCRIPT
17

All citrus juices concentration-dependently induced the mRNA expression of most of the

genes investigated in LS180 cells. Figure 1 depicts the effect for the highest concentration

tested (20 %). Grapefruit juice profoundly induced several PXR-driven genes (i.e. CYP3A4,

T
CYP3A5, UGT1A3, ABCB1, and ABCC2), mostly even exceeding the effect of the

IP
prototypical inducer rifampicin. AhR-driven genes were even more induced by grapefruit

R
juice (i.e. CYP1A1, CYP1A2, and ABCG2). Induction of CYP1A1 (12,829-fold) even

SC
exceeded that of the very strong AhR inducer and environmental pollutant dioxin (TCDD, 5

nM) (6,591-fold, data not shown). The induction observed for clementine and mandarin juice

NU
was generally lower than that for grapefruit juice (Figure 1). SLCO1B1 was only weakly
MA
induced by grapefruit juice and even repressed by clementine and mandarin juice. UGT2B7

mRNA was not changed by any of the citrus juiced tested (data not shown).

For CYP1A2, CYP3A4, and P-gp, induction by citrus juices was verified at the protein level
D
TE

(Figure 2).
P

3.4 Activation of PXR and AhR by citrus juices

CE

To mechanistically scrutinise the observed inductions, activation of responsible nuclear

AC

receptors (PXR and AhR) was investigated by luciferase-based reporter gene assays.

Matching the results obtained in the RT-PCR, grapefruit juice revealed to be a very strong

PXR and AhR activator, whereas mandarin and clementine juice also increased PXR and AhR

activity, but with much lower potency (Figure 3 and 4).

3.5 Net Effect of citrus juices on CYP3A4 and CYP1A2 activity

According to the data obtained, citrus juices appear to concurrently induce and also inhibit

CYP3A4 and CYP1A2. We therefore investigated which effect prevails (= net effect).

CYP1A2 and CYP3A4 activity in LS180 cells were measured after four days of treatment

with grapefruit, clementine, or mandarin juice. Rifampicin (20 µM) and TCDD (5 nM) again
 ACCEPTED MANUSCRIPT
18

served as positive controls for CYP3A4 or CYP1A2 induction, respectively. For rifampicin

and TCDD, which are inducers but no inhibitors of the respective enzymes, the induction was

clearly visible at the activity level (Figure 5 and 6). In contrast, treatment with citrus juices

T
demonstrated an overlap of induction and inhibition: Grapefruit juice significantly inhibited

IP
CYP3A4 activity even at the lowest concentration of 1 %. A net inhibition of CYP3A4 was

R
also observed for clementine juice, whereas for the mandarin juice the lower concentrations

SC
had no effect on CYP3A4 activity and the highest concentration induced CYP3A4 activity

about 1.5-fold. The net effect of grapefruit juice on CYP1A2 activity was null, for clementine

NU
juice induction prevailed at lower concentrations, and for mandarin juice induction prevailed
MA
up to 10 % (Figure 4 and 5).

3.6 Quantification of furanocoumarins and flavonoids in clementine, mandarin, and

D
TE

grapefruit juice

The three citrus juices were subjected to LC-MS-MS analysis to gain insight on their content
P

of flavonoids and furanocoumarins. In general, grapefruit juice was found to contain the
CE

highest amount of flavonoids, whereas mandarin juice the lowest (Table 2).
AC

The flavanone naringin was found to be the most abundant flavonoid present in grapefruit

juice (about 200 mg/l), followed closely by isomeric narirutin (about 100 mg/l).

Neohesperidin and hesperidin were quantified in the 12–14 mg/l range. Smaller amounts of

6,8-di-C-glucosyl-apigenin, eriocitrin, dydimin and poncirin were also found. In addition, the

polymethoxyflavones nobiletin and tangeretin were found (ca 1 mg/l each), along with the

expected furanocoumarins, i.e. bergamottin and epoxybergamottin (< 1 mg/l each).

Clementine juice showed a different profile altogether, with its chromatographic trace

dominated by the two rutinosides, hesperidin and narirutin (91 and 72 mg/l, respectively).

Dydimin (ca 2 mg/l) along with even smaller amounts (< 1 mg/ml) of 7-O-neohesperidosyl

chrysoeriol and of the di-C-glucosyl flavones (6,8-di-C-glucosyl-apigenin and 6,8-di-C-

 ACCEPTED MANUSCRIPT
19

glucosyl-diosmetin) were also found. Interestingly, this juice was found to be the richest in

terms of polymethoxyflavones content (ca 4.5 vs. 2 and 2.5 mg/l for grapefruit and mandarin,

respectively), with sinensetin appearing – together with nobiletin and tangeretin – as the main

T
component from this class of compounds. No furanocoumarines were detected in this juice.

IP
Mandarin juice is characterised, with respect to the two other juices analysed, by a significant

R
amount of the di-C-glucosyl flavones (6,8-di-C-glucosyl-apigenin: about 10 mg/l, 6,8-di-C-

SC
glucosyl-diosmetin: about 5 mg/l), even though the flavanone hesperidin is stil the major

component (about 50 mg/l), followed by the other rutinoside, narirutin in smaller amount

NU
(about 20 mg/l). As in clementine, sinensetin, nobiletin and tangeretin were the only
MA
polymethoxyflavones found, and again no furanocoumarines were detected.
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
20

4. Discussion

We report for the first time the effect of clementine and mandarin juices compared to

grapefruit juice on the expression and activity of drug transporters and drug metabolising

T
enzymes. Our data demonstrate pronounced inducing effects especially of grapefruit juice on

IP
PXR-driven genes like CYP3A4, ABCB1, and ABCG2, sometimes considerably exceeding the

R
effects mediated by the prototypical PXR ligand rifampicin (Figure 1). AhR-driven genes like

SC
CYP1A1 CYP1A2, and ABCG2 were also highly induced. The effects of the citrus juices again

partly exceeded the effects of the prototypical AhR ligand dioxin (Figure 1). Reporter gene

NU
assays confirmed that all juices contained PXR and AhR-activating ingredients with
MA
grapefruit juice generally exhibiting the highest potency (Figures 3 and 4).

However, the juices not only induced many drug transporters and drug metabolising enzymes,

some of the proteins were also potently inhibited. All juices inhibited BCRP at low
D
TE

concentrations (Table 1), matching data of another study, having demonstrated BCRP

inhibition by fruit juice components such as bergamottin, dihydroxybergamottin, and

P

tangeretin (Fleisher et al., 2015). In contrast, P-gp was not inhibited up to concentrations of
CE

20 %, although data of previous studies indicated that some components of grapefruit juice
AC

are P-gp inhibitors (Shirasaki et al., 2011; Dahan and Amidon 2009; de Castro et al., 2007;

El-Readi et al, 2010). Based on our in vitro data the increase in tacrolimus concentrations in

our patient cannot be attributed to P-gp inhibition by clementines.

Grapefruit juice mediated OATP1A2-inhibition is well characterised (Dresser et al., 2002;

Bailey et al., 2007). Some of its flavonoids like naringin also inhibit OATP1B1 and

OATP1B3 in vitro and high enough portal blood concentrations after extensive consumption

of grapefruit juice to inhibit these transporters in vivo are possible (Mandery et al., 2012).

This matches our data demonstrating that grapefruit juice is a potent inhibitor of OATP1B1

and OATP1B3. Moreover, we demonstrated for the first time that also mandarin and
 ACCEPTED MANUSCRIPT
21

clementine juices can inhibit OATP1B1 and OATP1B3, albeit with lower potency, which is

however most likely clinically irrelevant (Table 1).

CYP1A2 and CYP3A4 were profoundly induced by grapefruit juice and to a lesser extent by

T
mandarin and clementine juice. Concurrently, these enzymes can also be functionally

IP
inhibited. Clementine juice seems to be a weak CYP1A2 inhibitor (Table 1), whereas

R
grapefruit juice (Tasseneeyakul et al., 2000) and mandarin juice potently inhibit CYP1A2

SC
Thus, it is of crucial importance to assess the functional net effect resulting from these two

counteracting processes (induction vs. inhibition). We quantified the eventual net effect on

NU
protein activity after four days of exposure. The net effect differed between the juices. Gene
MA
induction of CYP1A2 activity by grapefruit juice was not functionally detectable at

concentrations between 1-20 %, for mandarin juice only a slight induction sustained, and for

clementine juice induction was outbalanced at least with high concentrations (Figure 6).
D
TE

Grapefruit juice contains mechanism-based inhibitors of CYP3A4 (Tasseneeyakul et al.,

2000; Egashira et al., 2004; Takanaga et al., 2000) that can apparently superimpose the strong
P

inducing effect since the net cellular CYP3A4 activity in LS180 cells was significantly
CE

reduced (Figure 5). The same effect (albeit with lower potency) was observed for clementine
AC

juice, suggesting that this juice also contains mechanism-based inhibitors. In contrast,

mandarin juice had hardly any inhibitory effects on purified CYP3A4 enzymes (Table 1) and

the net cellular effect was a slight induction at the highest concentrations tested after four

days of treatment (Figure 5).

We can only speculate which compounds in the different citrus juices elicited the effects

observed. The chemical composition and functional properties of citrus juices are profoundly

influenced e.g. by the genetic background (Goldenberg et al., 2014), cultivar (Milella et al.,

2011), maturity (Fujita et al., 2008) and by the extraction techniques (Alvarez et al., 2011).

Data on constituents of citrus fruits varies greatly. The most important components

considered responsible for drug-drug interactions are furanocoumarins (bergamottin,

 ACCEPTED MANUSCRIPT
22

dihydroxybergamottin) and flavonoids (hesperidin, naringin, tangeretin) (Seden et al., 2010;

Fujita et al., 2008; Guo et al., 2000; Dolton et al., 2013; Cho et al., 2009; Backman et al.,

2000). Among the juices under investigation, grapefruit juice is the only one to be known for

T
his relatively high furanocoumarins content. It contains up to 50 µM bergamottin, 85 µM

IP
dihydroxybergamottin (Xu et al., 2015), 19.5 mM naringin, and up to 570 µM hesperidin but

R
only small amounts of tangeretin (Nogata et al., 2006; Gattuso et al., 2007; Kawaii et al.,

SC
1999). Mandarins contain high levels of hesperidin (up to 22 mM (Nogata et al., 2006)) and

tangeretin (up to 142 µM (Nogata et al., 2006)), but no detectable concentrations of naringin,

NU
bergamottin, and dihydroxybergamottin (Peterson et al., 2006; Nogata et al., 2006; Fujita et
MA
al., 2008). Only one study investigated some of the flavonoids in the clementine cultivar we

used (Citrus clementina var. clemenules) (Sentandreu et al., 2007); the flavonoid with the

highest concentration in the juice was hesperidin (maximum 73 µM), whereas the
D
TE

polymethoxylated flavone tangeretin only reached concentrations of about 0.3 µM. Another

study investigating clementines of unknown breed found about 611 µM hesperidin and no
P

tangeretin or naringin in juice vesicles (Nogata et al., 2006). Investigations of Citrus

CE

clementina Hort. ex Tan. found even higher concentrations of about 1.23 mM for hesperidin
AC

in the juice (Dhuique-Mayer et al., 2005) and another study investigating this cultivar found

up to 272 µM hesperidin and only up to 4.7 µM naringin in the juice (Milella et al., 2011). So

far, no data on the possible occurrence or concentrations of the furoanocoumarins in

clementines was available.

The juices analysed herein showed a composition that does not significantly diverge from

data discussed above. In addition, we found that clementines, as well as mandarins, do not

contain detectable concentrations of the furanocoumarin derivatives, i.e. bergamottin,

dihydroxybergamottin, and epoxybergamottin, respectively; polymethoxyflavones, on the

other hand, are present as minor components in all samples investigated (Gattuso et al., 2007).
 ACCEPTED MANUSCRIPT
23

Thus, for grapefruit juice the effects observed on the expression and activity of CYP1A2 and

CYP3A4 as well as inhibition of BCRP and OATPs can most likely be attributed to the

combined action of furanocoumarins and naringin (Seden et al., 2010; Fujita et al., 2008; Guo

T
et al., 2000; Dolton et al., 2013; Cho et al., 2009; Backman et al., 2000; Mandery et al., 2012;

IP
Brill et al., 2009). In contrast, tangeretin and hesperidin might be the most important drug-

R
interactions provoking compounds in mandarins. For the di-C-glucosyl flavones no data on

SC
interactions with CYPs, OATPs or other drug transporters is available, but since their content

in the mandarin juice is much higher than in the clementine and grapefruit juice, they might

NU
also contribute to the effects observed for this citrus juice. For clementines the compounds
MA
causative for the effects observed also remain unclear. The main ingredient hesperidin is not

known to be a strong inducer (e.g. of CYP1A2, CYP3A4) or inhibitor (e.g. of CYP1A2,

CYP3A4, BCRP, or OATPs) and the second main ingredient narirutin seems to be an only
D
TE

very weak inhibitor of CYP3A4 (Ho and Saville, 2001). Together, this data points out that

either some of the ingredients more potently modulate drug metabolising enzymes than
P

expected before, or that some of the ingredients so far not tested for their effects also
CE

contribute the the phenomena observed, or that the particular compound composition in our
AC

clementines exhibit effects mechanistically unknown.

Concerning the suspected tacrolimus-clementine interaction, the effect of clementine

ingestion on tacrolimus whole blood concentrations observed in our patient can most

presumably be explained by CYP3A4 inhibition, which appears to exceed the inducing effects

observed in our in vitro experiments. These results match data from other studies

demonstrating that the inhibitory effect overweighs, when both induction and inhibition occur

simultaneously (Hafner et al., 2011) and also applies for grapefruit juice, where all in vivo

data indicate strong CYP3A4 inhibition (Seden et al., 2010), although we also found strong

induction at the mRNA and protein level.

 ACCEPTED MANUSCRIPT
24

Based on our findings, both a dual effect (inhibition and induction) from one compound as

well as single effects on the expression and activity of drug metabolising enzymes are

conceivable. If two or more compounds are responsible, our findings might only apply for the

T
clementine cultivar, harvest and batch we used for our study, because the occurence, percent

IP
distribution, and concentrations of constituents may differ markedly (de Castro et al., 2006).

R
For food-drug interactions, inter-batch variability is a known phenomenon. Testing two

SC
different kinds (labels) of grapefruit juice for their effect on tacrolimus pharmacokinetics, led

to a mean 2.1-fold increase of tacrolimus trough levels in one case while the other batch had

NU
no effect (statistically non-significant 20 % increase) (Liu et al., 2009).
MA
Hence, characterisation of effects of single compounds present in clementine juice are

objectives for further in vitro experiments. Moreover, the effects observed in our patient

warrants the conductance of a clinical trial to investigate induction and inhibition of CYP3A
D
TE

mediated metabolism through clementine juice.

P

5. Conclusions
CE

In conclusion, we have provided in vitro evidence and in vivo indication that clementines and
AC

clementine juice might lead to food-drug interactions similar to grapefruit juice. However, the

extent of interaction with clementines might greatly depend on the extent of intake, the

cultivar, and the harvest and is obviously different from the interaction potential of the closely

related mandarins. Interestingly, during the preparation of this manuscript another patient

formerly stable on tacrolimus presented himself with increased trough levels and admitted

extensive clementine consumption when asked directly. Hence, based on our in vitro findings

and recent clinical experiences, transplant patients on tacrolimus should refrain from

excessive consumption of clementines. To draw a definite conclusion, a controlled

pharmacokinetic trial evaluating tacrolimus and midazolam (protoptypical CYP3A4 substrate)

 ACCEPTED MANUSCRIPT
25

pharmacokinetics during intake of clementine juices of different batches, maturities etc seems

appropriate.

T
Acknowledgments

IP
We thank J. Kocher, S. Rosenzweig, C. Mueller, M. Wittnebel, and A. Fautsch for excellent

R
technical assistance.

SC
This research did not receive any specific grant from funding agencies in the public,

commercial, or not-for-profit sectors.

NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
26

References

Albermann, N., Schmitz-Winnenthal, F.H., Z'graggen, K., Volk, C., Hoffmann, M.M.,

Haefeli, W.E., Weiss, J., 2005. Expression of the drug transporters MDR1/ABCB1,

T
MRP1/ABCC1, MRP2/ABCC2, BCRP/ABCG2, and PXR in peripheral blood mononuclear

IP
cells and their relationship with the expression in intestine and liver. Biochem. Pharmacol. 70,

R
949-958.

SC
NU
Álvarez, R., Carvalho, C.P., Sierra, J., Lara, O., Cardona, D., Londoño-Londoño, J., 2012.

Citrus juice extraction systems: effect on chemical composition and antioxidant activity of
MA
clementine juice. J. Agric. Food Chem. 60, 774-781.
D

Ayed-Boussema, I., Pascussi, J.M., Maurel, P., Bacha, H., Hassen, W., 2011. Zearalenone
TE

activates pregnane X receptor, constitutive androstane receptor and aryl hydrocarbon receptor
P

and corresponding phase I target genes mRNA in primary cultures of human hepatocytes.
CE

Environ. Toxicol. Pharmacol. 31, 79-87.

AC

Backman, J.T., Mäenpää, J., Belle, D.J., Wrighton, S.A., Kivistö, K.T., Neuvonen, P.J., 2000.

Lack of correlation between in vitro and in vivo studies on the effects of tangeretin and

tangerine juice on midazolam hydroxylation. Clin. Pharmacol. Ther. 67, 382-390.

Bailey, D.G., Dresser, G.K., Leake, B.F., Kim, R.B., 2007. Naringin is a major and selective

clinical inhibitor of organic anion-transporting polypeptide 1A2 (OATP1A2) in grapefruit

juice. Clin. Pharmacol. Ther. 81, 495-502.

 ACCEPTED MANUSCRIPT
27

Barreca, D., Gattuso, G., Laganà, G., Leuzzi, U., Bellocco, E., 2016. C- and O-glycosyl

flavonoids in Sanguinello and Tarocco blood orange (Citrus sinensis (L.) Osbeck) juice:

identification and influence on antioxidant properties and acetylcholinesterase activity. Food

T
Chem. 196, 619–627.

R IP
Barreca, D., Bellocco, E., Caristi, C., Leuzzi, U., Gattuso, G., 2011. Flavonoid profile and

SC
radical-scavenging activity of Mediterranean sweet lemon (Citrus limetta Risso) juice. Food

Chem. 129, 417–422.

NU
MA
Barrett, H.C., Rhodes, A.M., 1976. A numerical taxonomic study of affinity relationships in

cultivated Citrus and its close relatives. Systematic Botany 1, 105–136.

D
TE

Boesch, D., Gaveriaux, C., Jachez, B., Pourtier-Manzanedo, A., 1991. In vivo circumvention

of P-glycoprotein-mediated multidrug resistance of tumor cells with SDZ PSC 833. Cancer
P

Res. 51, 4226-4233.

CE
AC

Brandin, H., Viitanen, E., Myrberg, O., Arvidsson, A.K., 2007. Effects of herbal medicinal

products and food supplements on induction of CYP1A2, CYP3A4 and MDR1 in the human

colon carcinoma cell line LS180. Phytother. Res. 21, 239-244.

Burhenne, J., Halama, B., Maurer, M., Riedel, K.D., Hohmann, N., Mikus, G., Haefeli, W.E.,

2012. Quantification of femtomolar concentrations of the CYP3A substrate midazolam and

its main metabolite 1'-hydroxymidazolam in human plasma using ultra performance liquid

chromatography coupled to tandem mass spectrometry. Anal. Bioanal. Chem. 402, 2439-

2450.
 ACCEPTED MANUSCRIPT
28

Brill, S., Zimmermann, C., Berger, K., Drewe, J., Gutmann,H., 2009. In vitro interactions

with repeated grapefruit juice administration--to peel or not to peel? Planta Med. 75, 332-335.

T
Cho, Y.A., Choi, D.H., Choi, J.S., 2009. Effect of hesperidin on the oral pharmacokinetics of

IP
diltiazem and its main metabolite, desacetyldiltiazem, in rats. J. Pharm. Pharmacol. 61, 825-

R
829.

SC
Dahan, A., Amidon, G.L., 2009. Grapefruit juice and its constituents augment colchicine

NU
intestinal absorption: potential hazardous interaction and the role of p-glycoprotein. Pharm.
MA
Res. 26, 883-892.

De Castro, W.V., Mertens-Talcott, S., Rubner, A., Butterweck, V., Derendorf, H., 2006.
D
TE

Variation of flavonoids and furanocoumarins in grapefruit juices: a potential source of

variability in grapefruit juice-drug interaction studies. J. Agric. Food Chem. 54, 249-255.
P
CE

de Castro, W.V., Mertens-Talcott, S., Derendorf, H., Butterweck, V., 2007. Grapefruit juice-
AC

drug interactions: Grapefruit juice and its components inhibit P-glycoprotein (ABCB1)

mediated transport of talinolol in Caco-2 cells. J. Pharm. Sci. 96, 2808-2817.

Dhuique-Mayer, C., Caris-Veyrat, C., Ollitrault, P., Curk, F., Amiot, M.J., 2005. Varietal and

interspecific influence on micronutrient contents in citrus from the Mediterranean area. J.

Agric. Food Chem. 53, 2140-2145.

Dolton, M.J., Roufogalis, B.D., McLachlan, A.J., 2012. Fruit juices as perpetrators of drug

interactions: the role of organic anion-transporting polypeptides. Clin. Pharmacol. Ther. 92,

622-630.
 ACCEPTED MANUSCRIPT
29

Dresser, G.K., Bailey, D.G., Leake, B.F., Schwarz, U.I., Dawson, P.A., Freeman, D.J., Kim,

R.B. 2002. Fruit juices inhibit organic anion transporting polypeptide-mediated drug uptake

T
to decrease the oral availability of fexofenadine. Clin. Pharmacol. Ther. 71, 11-20.

R IP
Dvorak, Z., Vrzal, R., Henklova, P., Jancova, P., 2008. JNK inhibitor SP600125 is a partial

SC
agonist of human aryl hydrocarbon receptor and induces CYP1A1 and CYP1A2 genes in

primary human hepatocytes. Biochem. Pharmacol. 75, 580-588.

NU
MA
Egashira, K., Ohtani, H., Itoh, S., Koyabu, N., Tsujimoto, M., Murakami, H., Sawada, Y.,

2004. Inhibitory effects of pomelo on the metabolism of tacrolimus and the activities of

CYP3A4 and P-glycoprotein. Drug Metab. Dispos. 32, 828-833.

D
TE

El-Readi, M.Z., Hamdan, D., Farrag, N., El-Shazly, A., Wink, M., 2010. Inhibition of P-
P

glycoprotein activity by limonin and other secondary metabolites from Citrus species in
CE

human colon and leukaemia cell lines. Eur. J. Pharmacol. 626, 139-145.
AC

Fleisher, B., Unum, J., Shao, J., An, G., 2015. Ingredients in fruit juices interact with

dasatinib through inhibition of BCRP: a new mechanism of beverage-drug interaction. J.

Pharm. Sci. 104, 266-275.

Fröhlich, M., Albermann, N., Sauer, A., Walter-Sack, I., Haefeli, W.E., Weiss, J., 2004. In

vitro and ex vivo evidence for modulation of P-glycoprotein activity by progestins. Biochem.

Pharmacol. 68, 2409-2416.

 ACCEPTED MANUSCRIPT
30

Fujita, T., Kawase, A., Niwa, T., Tomohiro, N., Masuda, M., Matsuda, H., Iwaki, M., 2008.

Comparative evaluation of 12 immature citrus fruit extracts for the inhibition of cytochrome

P450 isoform activities. Biol. Pharm. Bull. 31, 925-930.

T
IP
Garcia-Lor, A., Curk, F., Hager, S.T., Morillon, R., Ancillo, G., Navarro, L., Ollitraut, P.,

R
2013. A nuclear phylogenetic analysis: SNPs, indels and SSRs deliver new insights into the

SC
relationships in the ‘true citrus fruit trees’ group (Citrinae, Rutaceae) and the origin of

cultivated species. Ann. Bot. 111, 1-19.

NU
MA
Goldenberg, L., Yaniv, Y., Kaplunov, T., Doron-Faigenboim, A., Carmi, N., Porat, R., 2014.

Genetic diversity among mandarins in fruit-quality traits. J. Agric. Food Chem. 62, 4938-

4946.
D
TE

Guo, L.Q., Fukuda, K., Ohta, T., Yamazoe, Y., 2000. Role of furanocoumarin derivatives on
P

grapefruit juice-mediated inhibition of human CYP3A activity. Drug Metab. Dispos. 28, 766-
CE

771.
AC

Gupta, A., Mugundu, G.M., Desai, P.B., Thummel, K.E., Unadkat, J.D.. 2008. Intestinal

human colon adenocarcinoma cell line LS180 is an excellent model to study pregnane X

receptor, but not constitutive androstane receptor, mediated CYP3A4 and multidrug resistance

transporter 1 induction: studies with anti-human immunodeficiency virus protease inhibitors.

Drug Metab. Dispos. 36, 1172-1180.

Halama, B., Hohmann, N., Burhenne, J., Weiss, J., Mikus, G., Haefeli, W.E., 2013. A

nanogram dose of the CYP3A probe substrate midazolam to evaluate drug interactions. Clin.

Pharmacol. Ther. 93, 564-571.

 ACCEPTED MANUSCRIPT
31

Harmsen, S., Koster, A.S., Beijnen, J.H., Schellens, J.H., Meijerman, I., 2008. Comparison of

two immortalized human cell lines to study nuclear receptor-mediated CYP3A4 induction.

T
Drug Metab. Dispos. 36, 1166-1171.

R IP
Harper, P.A., Prokipcak, R.D., Bush, L.E., Golas, C.L., Okey, A.B., 1991. Detection and

SC
characterization of the Ah receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin in the human

colon adenocarcinoma cell line LS180. Arch. Biochem. Biophys. 290, 27-36.

NU
MA
Hohmann, N., Haefeli, W.E., Mikus, G., 2015. Use of microdose phenotyping to individualise

dosing of patients. Clin. Pharmacokinet. 54, 893-900.

D
TE

Ho, P.-C., Saville, D.J., 2001. Inhibition of human CYP3A4 activity by grapefruit flavonoids,

furanocoumarins and related compounds. J. Pharm. Pharmaceut. Sci. 4, 217-227.

P
CE

Iwasaki, K., 2007. Metabolism of tacrolimus (FK506) and recent topics in clinical
AC

pharmacokinetics. Drug Metab. Pharmacokinet. 22, 328-335.

Katzenmaier, S., Markert, C., Riedel, K.D., Burhenne, J, Haefeli, W.E., Mikus, G., 2011.

Determining the time course of CYP3A inhibition by potent reversible and irreversible

CYP3A inhibitors using A limited sampling strategy. Clin. Pharmacol. Ther. 90, 666-673.

Kawaii, S., Tomono, Y., Katase, E., Ogawa, K., Yano, M., 1999. Quantitation of flavonoid

constituents in Citrus fruits. J. Agric. Food Chem. 47, 3565-3571.

 ACCEPTED MANUSCRIPT
32

König, J., Cui, Y., Nies, A.T., Keppler, D., 2000. Localization and genomic organization of a

new hepatocellular organic anion transporting polypeptide. J. Biol. Chem. 275, 23161-23168.

T
König, J., Cui, Y., Nies, A.T., Keppler, D., 2000. A novel human organic anion transporting

IP
polypeptide localized to the basolateral hepatocyte membrane. Am. J. Physiol. Gastrointest.

R
Liver Physiol. 278, G156-G164.

SC
König, S.K., Herzog, M., Theile, D., Zembruski, N., Haefeli, W.E., Weiss, J., 2010. Impact of

NU
drug transporters on cellular resistance towards saquinavir and darunavir. J. Antimicrob.
MA
Chemother. 65, 2319-2328.

Li, W., Harper, P.A., Tang, B.K., Okey, A.B., 1998. Regulation of cytochrome P450 enzymes
D
TE

by aryl hydrocarbon receptor in human cells: CYP1A2 expression in the LS180 colon

carcinoma cell line after treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin or 3-

P

methylcholanthrene. Biochem. Pharmacol. 56, 599-612.

CE
AC

Liu, C., Shang, Y.F., Zhang, X.F., Zhang, X.G., Wang, B., Wu, Z., Liu, X.M., Yu, L., Ma, F.,

Lv, U., 2009. Co-administration of grapefruit juice increases bioavailability of tacrolimus in

liver transplant patients: a prospective study. Eur. J. Clin. Pharmacol. 65, 881-885.

Mabberley, D.J., 1997. A classification for edible Citrus (Rutaceae). Telopea 7, 167–172.

Mandery, K., Balk, B., Bujok, K., Schmidt, I., Fromm, M.F., Glaeser, H., 2012. Inhibition of

hepatic uptake transporters by flavonoids. Eur. J. Pharm. Sci. 46, 79-85.

 ACCEPTED MANUSCRIPT
33

Milella, L., Caruso, M., Galgano, F., Favati, F., Padula, M.C., Martelli, G., 2011. Role of the

cultivar in choosing Clementine fruits with a high level of health-promoting compounds. J.

Agric. Food Chem.59, 5293-5298.

T
IP
Nicolosi, E., Deng, Z.N., Gentile, A., Malfa, S., Continella, E., Tribulato, E., 2000. Citrus

R
phylogeny and genetic origin of important species as investigated by molecular markers.

SC
Theor. App. Genet. 100, 1155–1166.

NU
Nogata, Y., Sakamoto, K., Shiratsuchi, H., Ishii, T., Yano, M., Ohta, H., 2006. Flavonoid
MA
composition of fruit tissues of citrus species. Biosci. Biotechnol. Biochem. 70, 178-192.

Novotna, A., Pavek, P., Dvorak, Z., 2011. Novel stably transfected gene reporter human
D
TE

hepatoma cell line for assessment of aryl hydrocarbon receptor transcriptional activity:

construction and characterization. Environ. Sci. Technol. 45, 10133-10139.

P
CE

Gattuso, G., Barreca, D., Gargiulli, C., Leuzzi, U., Caristi, C., 2007. Flavonoid composition
AC

of Citrus juices. Molecules 12, 1641-1673.

Hafner, V., Jäger, M., Matthée, A.K., Ding, R., Burhenne, J., Haefeli, W.E., Mikus, G. 2010.

Effect of simultaneous induction and inhibition of CYP3A by St John's Wort and ritonavir on

CYP3A activity. Clin. Pharmacol. Ther. 87, 191-196.

 ACCEPTED MANUSCRIPT
34

Pavek, P., Merino, G., Wagenaar, E., Bolscher, E., Novotna, M., Jonker, J.W., Schinkel, A.H.,

2005. Human breast cancer resistance protein: interactions with steroid drugs, hormones, the

dietary carcinogen 2-amino-1-methyl-6-phenylimidazo(4,5-b)pyridine, and transport of

T
cimetidine. J. Pharmacol. Exp. Ther. 312, 144-152.

R IP
Peters, T., Lindenmaier, H., Haefeli, W.E., Weiss, J., 2006. Interaction of the mitotic kinesin

SC
Eg5 inhibitor monastrol with P-glycoprotein. Naunyn Schmiedebergs Arch. Pharmacol. 372,

291-299.

NU
MA
Peterson, J.J., Dwyer, J.T., Beecher, G.R., Bhagwat, S.A., Gebhardt, S.E., Haytowitz, D.B.,

Holden, J.M., 2006. Flavanones in oranges, tangerines (mandarins), tangors, and tangelos: a

compiliation and review of the data from the analytical literature. J. Food Compost. Anal. 19,
D
TE

S66-S73.
P

Peynaud, D., Charpiat, B., Vial, T., Gallavardin, M., Ducerf, C., 2007. Tacrolimus severe
CE

overdosage after intake of masked grapefruit in orange marmalade. Eur. J. Clin. Pharmacol.
AC

63., 721-722.

Provenzani, A., Santeusanio, A., Mathis, E., Notarbartolo, M., Labbozzetta, M., Poma, P.,

Provenzani, A., Polidori, C., Vizzini, G., Polidori, P., D'Alessandro, N., 2013.

Pharmacogenetic considerations for optimizing tacrolimus dosing in liver and kidney

transplant patients. World J. Gastroenterol. 19, 9156-9173.

Seden, K., Dickinson, L., Khoo, S., Back, D. 2010. Grapefruit-drug interactions. Drugs 70,

2373-2407.
 ACCEPTED MANUSCRIPT
35

Sentandreu, E., Izquierdo, L., Sendra, J.M., 2007. Differentiation of juices from clementine

(Citrus clementina), clementine-hybrids and satsuma (Citrus unshiu) cultivars by statistical

multivariate discriminant analysis of their flavanone-7-O-glycosides and fully methoxylated

T
flavones content as determined by liquid chromatography. Eur. Food Res. Technol. 224, 321-

IP
429.

R
SC
Shirasaka, Y., Suzuki, K., Nakanishi, T., Tamai, I., 2011. Differential effect of grapefruit

juice on intestinal absorption of statins due to inhibition of organic anion transporting

NU
polypeptide and/or P-glycoprotein. J. Pharm. Sci. 100, 3843-3853.
MA
Sorkhi, H., Oliaei, F., Moghadamnia, A.A., Pouramin, M., Firoozjahi, A.R., 2007 Effect of

orange and tangerine juice on cyclosporine levels in renal transplant recipients. Transplant.
D
TE

Proc. 39, 1228-1230.

P

Sorkhi, H., Moghadamnia, A.A., Oaliaee, F., PourAmir, M., Firoozjahi, A.R., Pasha, A.A.,
CE

Goodarzi, M.R., 2008. Effect of tangerine juice on cyclosporine levels in renal transplant
AC

children. Pediatr. Nephrol. 23, 499-501.

Storch, C.H., Theile, D., Lindenmaier, H., Haefeli, W.E., Weiss, J., 2007. Comparison of the

inhibitory activity of anti-HIV drugs on P-glycoprotein. Biochem. Pharmacol. 73, 1573-81.

Takanaga, H., Ohnishi, A., Murakami, H., Matsuo, H., , Higuchi, S., Urae, A., Irie, S., Furuie,

H., Matsukuma, K., Kimura, M., Kawano, K., Orii, Y., Tanaka, T., Sawada, Y., 2000.

Relationship between time after intake of grapefruit juice and the effect on pharmacokinetics

and pharmacodynamics of nisoldipine in healthy subjects. Clin. Pharmacol. Ther. 67, 201-

214.
 ACCEPTED MANUSCRIPT
36

Tassaneeyakul, W., Guo, L.Q., Fukuda, K., Ohta, T., Yamazoe, Y., 2000. Inhibition

selectivity of grapefruit juice components on human cytochromes P450. Arch. Biochem.

T
Biophys. 378, 356-363.

R IP
Weiss, J., Rose, J., Storch, C.H., Ketabi-Kiyanvash, N., Sauer, A., Haefeli, W.E., Efferth, T.,

SC
2007. Modulation of human BCRP (ABCG2) activity by anti-HIV drugs. J. Antimicrob.

Chemother. 59, 238-245.

NU
MA
Weiss, J., Herzog, M., Haefeli, W.E., 2011. Differential modulation of the expression of

important drug metabolising enzymes and transporters by endothelin-1 receptor antagonists

ambrisentan and bosentan in vitro. Eur. J. Pharmacol. 660, 298-304.

D
TE

Weiss, J., Theile, D., Haefeli, W.E., 2012. Rifampicin alters the expression of reference genes
P

used to normalize real-time quantitative RT-PCR data. Naunyn Schmiedebergs Arch.

CE

Pharmacol. 385, 1025-1034.

AC

Weiss, J., Theile, D., Spalwisz, A., Burhenne, J., Riedel, K.D., Haefeli, W.E., 2013. Influence

of sildenafil and tadalafil on the enzyme- and transporter-inducing effects of bosentan and

ambrisentan in LS180 cells. Biochem. Pharmacol. 85, 265-273.

Weiss, J., Theile, D., Dvorak, Z., Haefeli, W.E., 2014. Interaction potential of the

multitargeted receptor tyrosine kinase inhibitor dovitinib with drug transporters and drug

metabolising enzymes assessed in vitro. Pharmaceutics 6, 632-650.

 ACCEPTED MANUSCRIPT
37

Xu, J., Ma, L., Jiang, D., Zhu, S., Yan, F., Xie, Y., Xie, Z., Guo, W., Deng, X., 2015. Content

evaluation of 4 furanocoumarin monomers in various citrus germplasms. Food. Chem. 187,

75-81.

T
IP
Yamasaki, D., Nakamura, T., Okamura, N., Kokudai, M., 2009. Effects of acid and lactone

R
forms of 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors on the induction of

SC
MDR1 expression and function in LS180 cells. Eur. J. Pharm. Sci. 37, 126-132.

NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
38

Table 1: Inhibition of drug transporters, CYP1A2, and CYP3A4 by citrus juices

Protein inhibited Grapefruit IC50 [%] Clementine IC50 [%] Mandarin IC50 [%]

BCRP 3.0  1.5 9.8  1.4 1.8  0.8

T
P-gp None up to 20 % None up to 20 % None up to 20 %

IP
OATP1B1 1.0 ± 0.3 6.9  3.6 6.2 ± 2.0

R
OATP1B3 2.3 ± 0.7 10.4  4.3 8.8 ± 2.8

SC
CYP1A2 0.6  0.4 > 25 2.6  0.7

1.0  0.3 52.0  32.0 93.8  11.7

NU
CYP3A4 MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
39

Table 2: Flavonoid and furanocoumarin content in the citrus juices investigated (mg/l)

Compound Grapefruit Clementine Mandarin

juice juice juice

T
6,8-di-C-glucosyl-apigenin 1.30 ± 0.16 0.70 ± 0.09 11.4 ± 0.60

IP
6,8-di-C-glucosyl-diosmetin — 0.30 ± 0.02 4.80 ± 0.70

R
Eriocitrin 2.70 ± 0.35 — 1.70 ± 0.20

SC
Narirutin 97.0 ± 2.2 71.8 ± 3.70 17.0 ± 1.2

Chrysoeriol 7-O- — 0.70 ± 0.03 —

neohesperidoside
NU
MA
Naringin 205 ± 4.80 — —

Hesperidin 12.5 ± 0.70 91.0 ± 6.50 46.0 ± 3.15

D

Neohesperidin 14.7 ± 0.5 — —

TE

Dydimin 0.75 ± 0.05 2.25 ± 0.32 —

Poncirin 1.35 ± 0.10 — —

P

—
CE

Sinensetin 1.85 ± 0.11 1.03 ± 0.07

Nobiletin 1.04 ± 0.10 1.25 ± 0.17 0.84 ± 0.05

AC

Tangeretin 0.93 ± 0.07 1.32 ± 0.12 0.61 ± 0.08

Bergamottin 0.80 ± 0.05 — —

Epoxybergamottin 0.60 ± 0.07 — —

 ACCEPTED MANUSCRIPT
40

Figure legends

Figure 1: Effects of rifampicin (20 µM), grapefruit juice (20 %), clementine juice (20 %), and

mandarin juice (20 %) after four days of exposure on mRNA expression in LS180 cells

T
compared to untreated medium control. Expression data were normalised to the housekeeping

IP
gene GU. Data are expressed as mean  SEM for n = 4 biological replicates (juices) and n = 8

R
SC
biological replicates for medium and rifampicin. Data were analysed using ANOVA with

Dunnett’s post hoc test compared to the medium control. * p<0.05, ** p<0.01.

NU
Figure 2: Effects on protein expression of P-gp, CYP3A4, and CYP1A2 after four days of
MA
exposure of LS180 cells to rifampicin (20 µM), grapefruit juice (20 %), clementine juice (20

%), or mandarin juice (20 %). Effects are compared to untreated medium control and
D

normalised to β-actin used as a loading control. Depicted are the results of the semi-
TE

quantification of all blots (at least 3 for each compound).

P
CE

Figure 3: Concentration-dependent effect of citrus juices and the positive control rifampicin

on PXR activity in LS180 cells. Each curve depicts the results of three experiments with each
AC

concentration tested in triplicate. Data are expressed as mean  SEM.

Figure 4: Concentration-dependent effect of citrus juices and the positive control TCDD

(insert) on AhR activity in AZ-AhR cells. Each curve depicts the results of three experiments

with each concentration tested in triplicate. Data are expressed as mean  SEM.

Fig. 5: Changes in CYP3A4 activity in LS180 cells after four days of exposure to 1-20 %

citrus juices or 20 µM of rifampicin. Data were normalised to the medium control. Data are

expressed as mean  SEM for n = 20-27. Statistical significance was evaluated by ANOVA
 ACCEPTED MANUSCRIPT
41

with Dunnett’s multiple comparison test for post hoc pairwise comparison of the results with

the medium control. * P<0.05, ** P<0.01.

T
Fig. 6: Changes in CYP1A2 activity in LS180 cells after four days of exposure to 1-20 %

IP
citrus juices or 5 nM TCDD. Data were normalised to the medium control. Data are expressed

R
as mean  SEM for n = 16. Statistical significance was evaluated (1) by ANOVA with

SC
Dunnett’s multiple comparison test for post hoc pairwise comparison of the results with the

medium control including the exorbitant strong inducer TCDD (** P<0.01) and (2) by the

same test excluding TCDD ( ## P<0.01).

NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
42

T
IP
R
SC
NU
MA
D
PTE
CE
AC
 ACCEPTED MANUSCRIPT
43

T
IP
R
SC
NU
MA
D
PTE
CE
AC
 ACCEPTED MANUSCRIPT
44

T
IP
R
SC
NU
MA
D
PTE
CE
AC
 ACCEPTED MANUSCRIPT
45

T
IP
R
SC
NU
MA
D
PTE
CE
AC
 ACCEPTED MANUSCRIPT
46

T
IP
R
SC
NU
MA
D
PTE
CE
AC
 ACCEPTED MANUSCRIPT
47

T
IP
R
SC
NU
MA
D
PTE
CE
AC
 ACCEPTED MANUSCRIPT
48

Graphical abstract

T
IP
R
SC
NU
MA
D
P TE
CE
AC