Modulation of multidrug resistance-associated proteins function in erythrocytes in glycerol-induced acute renal failure rats

Article abstract of DOI:10.1111/jphp.12664

Objectives: Evaluation of the function of multidrug resistance-associated proteins (MRPs) expressed in erythrocytes and screening of endogenous MRPs modulator(s) in glycerol-induced acute renal failure (ARF) rats. Methods: Concentrations of 2,4-dinitrophenyl-S-glutathione (DNP-SG), a substrate for MRPs, in erythrocytes after administration of 1-chloro-2,4-dintrobenzene (CDNB), a precursor of DNP-SG, were determined in control and ARF rats. The screening of endogenous MRPs modulator(s) was performed using washed erythrocytes and inside-out erythrocyte membrane vesicles (IOVs) in vitro. Key findings: Accumulation of DNP-SG in erythrocytes was observed in ARF rats. Uraemic plasma components exhibited a greater inhibitory effect on DNP-SG uptake by IOVs than control plasma components and increased the DNP-SG accumulation significantly in washed erythrocytes. Several protein-bound uraemic toxins at clinically observed concentrations and bilirubin significantly inhibited DNP-SG uptake by IOVs. In washed erythrocytes, bilirubin (10 μm) and l-kynurenine (100 μm), a precursor of kynurenic acid being MRPs inhibitor, increased DNP-SG accumulation significantly. Conclusions: Glycerol-induced ARF rats contain various MRPs inhibitors in plasma, and membrane-permeable MRP substrates/inhibitors including their precursors inhibit the MRPs function in erythrocytes cooperatively.

Full text of DOI:10.1111/jphp.12664

Research Paper
Modulation of multidrug resistance-associated proteins
function in erythrocytes in glycerol-induced acute renal
failure rats
Aoi Matsushima, Keisuke Oda, Nobuhiro Mori and Teruo Murakami
Laboratory of Biopharmaceutics and Pharmacokinetics, Graduate School of Pharmaceutical Sciences, Hiroshima International University, Kure, Japan
Keywords
Abstract
acute renal failure; erythrocytes; inside-out
erythrocyte membrane vesicles; multidrug
resistance-associated proteins; uraemic toxin
Objectives Evaluation of the function of multidrug resistance-associated pro-
teins (MRPs) expressed in erythrocytes and screening of endogenous MRPs mod-
ulator(s) in glycerol-induced acute renal failure (ARF) rats.
Correspondence
Methods Concentrations of 2,4-dinitrophenyl-S-glutathione (DNP-SG), a sub-
strate for MRPs, in erythrocytes after administration of 1-chloro-2,4-dintroben-
zene (CDNB), a precursor of DNP-SG, were determined in control and ARF rats.
The screening of endogenous MRPs modulator(s) was performed using washed
erythrocytes and inside-out erythrocyte membrane vesicles (IOVs) in vitro.
Key findings Accumulation of DNP-SG in erythrocytes was observed in ARF
rats. Uraemic plasma components exhibited a greater inhibitory effect on DNP-
SG uptake by IOVs than control plasma components and increased the DNP-SG
accumulation significantly in washed erythrocytes. Several protein-bound
uraemic toxins at clinically observed concentrations and bilirubin significantly
inhibited DNP-SG uptake by IOVs. In washed erythrocytes, bilirubin (10 lM)
and L-kynurenine (100 lM), a precursor of kynurenic acid being MRPs inhibitor,
increased DNP-SG accumulation significantly.
Teruo Murakami, Laboratory of
Biopharmaceutics and Pharmacokinetics,
Graduate School of Pharmaceutical Sciences,
Hiroshima International University, 5-1-1
Hiro-koshingai, Kure, Hiroshima 737-0112,
Japan.
E-mail: t-muraka@ps.hirokoku-u.ac.jp
Received June 15, 2016
Accepted October 16, 2016
doi: 10.1111/jphp.12664
Conclusions Glycerol-induced ARF rats contain various MRPs inhibitors in
plasma, and membrane-permeable MRP substrates/inhibitors including their
precursors inhibit the MRPs function in erythrocytes cooperatively.
the expression of mRNA of multidrug resistance-associated
[
protein (MRP) 2 significantly in Caco-2 cells.
Introduction
5]
The elimination of renal disposition drugs is suppressed
under renal failure due to the suppression of glomerular fil-
tration rate or transporter-mediated renal secretion. The
elimination of non-renal disposition drugs is also sup-
pressed due to the alteration of transport and metabolism
Multidrug resistance-associated proteins family, one of
ATP-binding cassette (ABC) efflux transporter superfamily,
contains 13 members from MRP1 to MRP13 (ABCC1-13),
and widely expressed in normal tissues such as intestine,
[
9]
brain, lung, liver, kidney and skeletal muscle. ABC efflux
transporters including P-glycoprotein (ABCB1), MRPs
family and breast cancer resistance protein (BCRP, ABCG2)
have a role to prevent the influx of endogenous and exoge-
nous xenobiotics into cells and facilitate the efflux from
cells to prevent the intracellular accumulation of such
xenobiotics as a host defence-detoxification mechanism
[
1,2]
under renal failure.
Under renal failure, various endoge-
nous compounds including uraemic toxins are accumu-
lated in central circulation, and the alteration of transport
and metabolism is thought to be caused by uraemic tox-
[
3–5]
ins.
Uraemic toxins are classified into three groups, and
middle molecules and protein-bound solutes of uraemic
toxins are thought to exert various effects on the function
[
10,11]
together with various metabolic enzymes.
Among
[6–8]
and expression of transporters.
For example, depro-
them, MRPs transport relatively hydrophilic compounds
such as methotrexate, pravastatin and etoposide and
various conjugated compounds including glucuronide,
teinized uraemic serum from end-stage renal failure
patients increased pravastatin accumulation and decreased
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**
1

Modulated MRPs function in erythrocytes
Aoi Matsushima et al.
glutathione and sulphate conjugate of various endogenous
and exogenous compounds.
Synthesis of 2,4-dinitrophenyl-S-glutathione
[
5,9,11]
Previously, the MRP
function in various tissues was evaluated in rats treated
with bilirubin, an endogenous MRPs substrate, intra-
2,4-Dinitrophenyl-S-glutathione, a GSH-conjugated metabo-
lite of CDNB, was synthesized using FDNB according to
[
12]
[19]
venously.
Evaluation was made by measuring the tissue
Hinchman et al.
Briefly, 2.5 mmol FDNB dissolved in
2.5 ml methanol was slowly mixed with 3.75 mmol GSH dis-
concentrations of 2,4-dinitrophenyl-S-glutathione (DNP-
SG), a substrate of MRPs, after intravenous administration
of 1-chloro-2,4-dinitrobenzene (CDNB), a precursor of
DNP-SG. CDNB can penetrate biomembranes easily by
simple diffusion due to its high lipophilicity and is rapidly
metabolized to DNP-SG by glutathione S-transferase (GST)
in tissues. In that study, significantly higher DNP-SG con-
centrations were observed in the brain, liver, jejunum and
skeletal muscle in bilirubin-treated rats, suggesting that
bilirubin at a higher concentration suppresses MRP-
mediated efflux transport systemically.
solved in 1 M KHCO (12.5 ml) under stirring. After 15-min
3
incubation at room temperature, the solution was filtered
and acidified to approximately pH 2 with diluted HCl. The
precipitate was collected by vacuum filtration and was
washed with a sufficient amount of distilled water to remove
any extra GSH. DNP-SG synthesized was chromatographi-
[20,21]
cally pure, as well as those in our previous study.
Animal study
In this study, the effect of acute renal failure (ARF) on
MRPs function in erythrocytes was examined in rats, as
Male Sprague–Dawley (SD) rats aged 7–9 weeks old weigh-
ing 270–350 g were used. The protocol of the experiments
was reviewed and approved in advance, and experiments
with animals were performed in accordance with the Guide
for Animal Experimentation from the Committee of
Research Facilities for Laboratory Animal Sciences, Hir-
oshima International University, which is in accordance
with the Guidelines for Proper Conduct of Animal Experi-
ments from the Science Council of Japan. The licence num-
ber of this animal study was AE15-005. ARF was induced by
[13,14]
erythrocytes express MRP1, MRP4 and MRP5
and the
effect of ARF on MRP function in erythrocytes is not yet
evaluated. ARF was induced by intramuscular injection of
[
15]
5
0% glycerol after 24-h water deprivation,
where the
pathogenic mechanisms are due to the ischaemic injury
[
16]
and tubular nephrotoxicity including haemolysis.
In
haemolytic uraemic syndrome, the high concentration of
indirect bilirubin has also been observed in the concentra-
tion range from 0.98 to 8 mg/dl (16.8–136.8 lM) in
an injection of 50% glycerol (10 ml/kg) into the leg muscles
[15]
[
17]
patients.
all cases of renal dysfunction.
Haemolytic uraemic syndrome is involved in
The screening of endoge-
after a 24-h period of water deprivation in SD rats.
Con-
[18]
trol rats received the same volume of saline. The induction
of ARF states was confirmed by determining the concentra-
tions of blood urea nitrogen (BUN), indoxyl sulphate,
direct (conjugated) and indirect (unconjugated) bilirubin
in plasma 24 h after glycerol treatment.
nous modulating compounds against MRPs function was
performed using inside-out erythrocyte membrane vesicles
(
IOVs) and washed erythrocytes in vitro.
Materials and Methods
Evaluation of multidrug resistance-
associated protein function in erythrocytes
in rats in vivo
Materials
1
-Chloro-2,4-dinitrobenzene, reduced glutathione (GSH),
uric acid, uridine, urea, D(À)-mannitol, quinolic acid
and Blood Urea Nitrogen B-Test Wako were purchased
from Wako Pure Chemicals (Osaka, Japan). 1-Fluoro-
Multidrug resistance-associated protein function in ery-
throcytes was evaluated by determining the DNP-SG con-
centration in erythrocytes after administration of CDNB in
rats. Control and ARF rats were anaesthetized by an
intraperitoneal injection of pentobarbital (40 mg/kg) and
received CDNB (30 lmol/kg) intravenously from the jugu-
lar vein, or by a constant rate infusion (6 lmol/2 ml/h) via
the cannulae (polyethylene tubing PE50; Natsume Seisa-
kusho, Tokyo, Japan) inserted at the femoral vein. Blood
(0.4 ml each) was collected periodically from jugular vein.
A part of blood sample was centrifuged at 8000g for 15 min
to obtain plasma sample. A part of blood sample was filled
into haematocrit tubes (75 mm; Drummond Scientific
Company, Broomall, PA, USA) to measure haematocrit
value. Haematocrit tubes containing blood sample were
2
,4-dintrobenzene (FDNB) and hippuric acid sodium salt
were obtained from Tokyo Kasei (Tokyo, Japan). Biliru-
bin, indoxyl sulphate potassium salt, indole-3-acetic acid,
kynurenic acid, L-kynurenine, p-cresole, xantine, cytidine,
creatine, creatinine and inosine were purchased from
Sigma Chemical Co. Ltd. (St. Louis, MO, USA).
TM
QuantiChrom Bilirubin Assay Kit was purchased from
BioAssay Systems (Hayward, CA, USA). 3-Carboxy-4-
methyl-5-propyl-2-furanopropanoic acid (CMPF) was
purchased from Cayman Chemical (Ann Arbor, MI,
USA). All other chemicals used were of the highest pur-
ity available.
2
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**

Aoi Matsushima et al.
Modulated MRPs function in erythrocytes
centrifuged at 15 000 g using haematocrit rotor (Kubota
3220; Kubota, Tokyo, Japan). The concentration of DNP-
uraemic toxins were added to the incubation medium.
Concentrations of uraemic toxins used were the concentra-
[
23]
SG in whole blood was measured after haemolysis of blood
sample (50 ll) by adding 250 ll of distilled water. The con-
centration of DNP-SG in erythrocytes was calculated using
the following equation: C = [C – C 9 (1 À H )]/H ,
tions reported in various renal failure patients.
The
compounds examined were dissolved in incubation med-
ium containing a small amount of dimethyl sulfoxide
(DMSO), and the final concentration of DMSO in incuba-
tion medium was adjusted at 1%. Incubation was made for
30 min, and the ATP-dependent uptake of DNP-SG in
IOVs was calculated by subtracting the uptake in the pres-
ence of AMP from that in the presence of ATP.
e
w
p
t
t
where C , C , C and H represent the erythrocytes concen-
e
w
p
t
tration, whole blood concentration, plasma concentration
of DNP-SG and haematocrit, respectively.
Preparation of inside-out erythrocyte
membrane vesicles
Screening of multidrug resistance-
associated proteins modulator using
washed erythrocytes in vitro
Inside-out erythrocyte membrane vesicles were prepared
from rat erythrocytes by a spontaneous, one-step vesicula-
[
22]
tion method.
Briefly, rat erythrocytes were obtained by
Blood was obtained from untreated control rats and washed
three times with fivefold volume of ice-cold isotonic medium
(140 mM NaCl, 5 mM KCl, 20 mM Tris/HCl, 2 mM MgCl2,
0.1 mM EDTA, 5 mM glucose, pH 7.4) by repeating centrifu-
gation and dilution with isotonic medium. Haematocrit of
washed erythrocytes was adjusted to 0.45 by adding ice-cold
isotonic medium to packed erythrocytes and centrifuging it
using haematocrit rotor (Kubota 3220; Kubota) to confirm
the haematocrit value. The erythrocytes suspension was incu-
bated for 1 h on ice. After incubation, 100 ll of isotonic
medium containing endogenous compounds (uraemic toxin
or plasma components) was added to the washed erythro-
cytes suspension (100 ll), and the suspension was incubated
at 37 °C for 15 min. The accumulation study of DNP-SG in
erythrocytes was started by adding 100 ll of CDNB (30 lM)
dissolved in isotonic medium containing a small amount of
DMSO. The final concentrations of CDNB and DMSO in
the washed erythrocytes suspension were 10 lM and 1%,
respectively. After 15 min, the incubation was stopped by
adding threefold volume of ice-cold isotonic medium. The
suspension was centrifuged at 1500g and 4 °C for 10 min,
and the procedure was repeated twice. After haemolysis of
erythrocytes pellet by adding fivefold distilled water, protein
and DNP-SG concentrations were determined.
washing fresh whole blood with fivefold volume of isotonic
buffer (70 mM NaCl, 80 mM KCl, 10 mM HEPES, 0.1 mM
EGTA, pH 7.5) three times. The packed erythrocyte cells
were lysed by 60-fold volume of ice-cold hypotonic buffer
(
2 mM HEPES, 0.1 mM EGTA, pH 7.5), and pellet ghosts
were obtained by centrifugation (40 000g, 20 min, 4 °C).
The pellet ghosts were suspended with a half volume of
hypotonic buffer and incubated at 37 °C for 30 min, and
were dispersed in 10 ml of the suspension buffer (100 mM
mannitol, 10 mM HEPES/Tris, pH 7.4), The suspension
was centrifuged (40 000g, 20 min, 4 °C) to remove the
supernatant. The pellet including IOVs was re-suspended
in suspension buffer and was stored at À80 °C until use.
The fraction of IOVs in all vesicles was estimated to be
37.5% by measuring the acetylcholine esterase activity in
the absence and presence of 0.2% Triton-X.
[22]
Collection of plasma components
Plasma obtained from control and ARF rats was depro-
teinized by mixing with the same volume of acetonitrile.
After centrifugation at 1500g for 10 min, the supernatant
containing acetonitrile (50%) was collected and evaporated
to dryness with evaporator (EYELA TVE-1000; Tokyo
Rikakikai, Tokyo, Japan). The residue was stored in freezer
kept at À20 °C until use.
Sample analysis
Concentration of blood urea nitrogen (BUN) was deter-
mined using Blood Urea Nitrogen B-Test Wako. Concen-
trations of total and direct bilirubin were determined using
Screening of multidrug resistance-
associated proteins modulator using inside-
out erythrocyte membrane vesicles in vitro
TM
QuantiChrom Bilirubin Assay Kit. Concentration of pro-
tein was measured by Bradford method using bovine serum
[
24]
Effects of various endogenous compounds on MRP-
mediated DNP-SG uptake were examined using IOVs in
albumin as the standard.
After deproteinization of DNP-
SG samples by adding perchloric acid (final concentration
was more than 4%) and indoxyl sulphate samples by adding
twofold volume of methanol, the concentrations of DNP-
SG and indoxyl sulphate were determined by HPLC using a
column of J-Pak Wrapsil RP-18 (Jasco Engineering, Tokyo,
[22]
the same manner as reported.
Briefly, IOVs (5 mg pro-
tein/ml) were incubated with DNP-SG (20 lM) in the pres-
ence of 1 mM ATP or 1 mM AMP at pH 7.4 and 37 °C, and
various compounds such as plasma components and
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**
3

Modulated MRPs function in erythrocytes
Aoi Matsushima et al.
Japan). Mobile phases were a mixture of 1% acetic acid and
acetonitrile (85 : 15 (v/v)) for DNP-SG and a mixture of
pH 4.5, 0.2 M acetate buffer, and acetonitrile (93 : 7 (v/v))
for indoxyl sulphate, respectively. DNP-SG was detected at
UV 365 nm, and indoxyl sulphate was detected by fluorom-
eter at Ex. 280 nm and Em. 375 nm.
Statistical analysis
Data were presented as the mean Æ SE. Statistical analysis
was performed by Student’s t-test, or by one-way analysis
of variance followed by the Tukey test for multiple compar-
isons. A difference of *P < 0.05 was regarded as signifi-
cantly different from the corresponding value.
Results
Table 1 Biochemical data of control and acute renal failure (ARF)
rats
Biochemical data of control and acute renal
failure rats
Control
ARF
BUN (mg/dl)
13.7 Æ 0.70
3.63 Æ 0.61
1.76 Æ 0.09
1.01 Æ 0.12
124 Æ 6.94**
103 Æ 10.6**
29.1 Æ 7.96*
3.01 Æ 0.62*
To confirm the induction of ARF states, plasma con-
centrations of BUN, indoxyl sulphate and bilirubin were
determined in control and ARF rats (Table 1). In ARF
rats, concentrations of BUN and indoxyl sulphate were
approximately ninefold and 28-fold higher than those
in control rats, respectively. The plasma concentrations
of total and direct bilirubin, endogenous MRPs sub-
strate/inhibitor, in ARF rats were also significantly
higher than those in control rats.
Indoxyl sulphate (lM)
Total bilirubin (lM)
Direct bilirubin (lM)
BUN, blood urea nitrogen. ARF rats were induced by intramuscular
injection of 50% glycerol (10 ml/kg, v/v) after 24-h water deprivation
in rats. Data were obtained 24 h after the treatment with saline (Con-
trol) or 50% glycerol (ARF). Each value represents the mean Æ SE of
results from three to four rats. **P < 0.01 and *P < 0.05: significantly
different from the value for control.
Figure 1 Concentrations of 2,4-dinitrophenyl-S-glutathione in plasma (a), whole blood (b) and erythrocytes (c), and erythrocyte/plasma concen-
tration ratio (d) after intravenous administration of 1-chloro-2,4-dintrobenzene (30 lmol/kg) in control and acute renal failure rats. Rats received
saline (control) or 50% glycerol (acute renal failure) at a volume of 10 ml/kg and were used 24 h after treatment. Open and closed circles repre-
sent the results in control and acute renal failure rats, respectively. Each value represents the mean Æ SE of results from three rats. *P < 0.05,
*
*P < 0.01: significantly different from the value for control.
4
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**

Aoi Matsushima et al.
Modulated MRPs function in erythrocytes
(
Figure 3). Plasma components (corresponding to 10%
Accumulation of 2,4-dinitrophenyl-S-
glutathione in erythrocytes in acute renal
failure rats
deproteinized plasma) obtained from control rats exerted
no significant effect on DNP-SG accumulation in washed
erythrocytes. In contrast, plasma components (correspond-
ing to 10% deproteinized plasma) obtained from ARF rats
significantly increased DNP-SG accumulation in erythro-
cytes by approximately 3.5-fold of control, indicating that
plasma of ARF rats contains some MRP modulator(s).
1-Chloro-2,4-dinitrobenzene was administered intravenou-
sly or by a constant rate infusion, and concentrations of
DNP-SG in erythrocytes were determined in control and
ARF rats (Figures 1a and 2a). The difference in DNP-SG
plasma concentrations between control and ARF rats was
small. In contrast, ARF rats exhibited significantly higher
blood concentrations of DNP-SG compared to control rats,
indicating that DNP-SG is accumulated in erythrocytes in
ARF rats (Figures 1b, 1c, 2b). Although there was no rela-
tionship in plasma concentrations between DNP-SG and
indoxyl sulphate, an uraemic toxin, the higher indoxyl sul-
phate concentrations resulted in the higher DNP-SG accu-
mulation in erythrocytes in ARF rats (Figure 2d).
Screening of multidrug resistance-
associated proteins modulator using inside-
out erythrocyte membrane vesicles in vitro
MK-571 and probenecid, both are typical MRP inhibitors,
inhibited the DNP-SG uptake by IOVs significantly
(Figure 4). Bilirubin, an endogenous MRPs substrate but
not uraemic toxin, also inhibited DNP-SG uptake by IOVs
significantly. Plasma components obtained from control
and ARF rats significantly inhibited the DNP-SG uptake by
IOVs in a concentration-dependent manner, and greater
inhibitory effects were observed with ARF plasma compo-
nents (Figure 4). All small water-soluble solutes of uraemic
toxins examined in this study exerted no significant effect
on DNP-SG uptake by IOVs (Figure 5a). In contrast,
Effect of plasma components on 2,4-
dinitrophenyl-S-glutathione accumulation in
washed erythrocytes
2,4-Dinitrophenyl-S-glutathione concentrations in washed
erythrocytes after application of CDNB were determined
Figure 2 Concentrations of 2,4-dinitrophenyl-S-glutathione in plasma (a) and erythrocytes (b) during the constant rate infusion of 1-chloro-2,4-din-
trobenzene (6 lmol/2 ml/h), and the relationship of 2,4-dinitrophenyl-S-glutathione concentrations in plasma (c) and erythrocytes (d) with plasma
concentrations of indoxyl sulphate at 30 min in control and acute renal failure rats. Rats received saline (control) or 50% glycerol (acute renal failure)
at a volume of 10 ml/kg and were used 24 h after treatment. Open and closed circles represent the results in control and acute renal failure rats,
respectively. Each value represents the mean Æ SE of results from four rats. *P < 0.05: significantly different from the value for control.
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**
5

Modulated MRPs function in erythrocytes
Aoi Matsushima et al.
Figure 4 Effects of MK-571, bilirubin, probenecid and plasma com-
ponents obtained from control and acute renal failure rats on 2,4-dini-
trophenyl-S-glutathione uptake by inside-out erythrocyte membrane
vesicles. Control represents the results in the absence of any multidrug
resistance-associated protein substrate/inhibitor in the medium
(
37 °C). Plasma components were obtained from deproteinized
plasma of control and acute renal failure rats (24 h after glycerol
treatment). Inside-out erythrocyte membrane vesicles were incubated
with 2,4-dinitrophenyl-S-glutathione (20 lM) for 30 min at 37 °C.
Each value represents the mean Æ SE of results from three trials.
Figure 3 Effect of plasma components obtained from control and
acute renal failure rats on 2,4-dinitrophenyl-S-glutathione accumula-
tion after application of 1-chloro-2,4-dintrobenzene in washed
erythrocytes. Control value represents the concentration of 2,4-dini-
trophenyl-S-glutathione in washed erythrocytes in the absence of
plasma component in medium (37 °C). Plasma components were
obtained from deproteinized plasma of control and acute renal failure
rats (24 h after glycerol treatment). Concentrations of 2,4-dinitrophe-
nyl-S-glutathione in erythrocytes were determined 15 min after appli-
cation of 1-chloro-2,4-dintrobenzene (10 lM). Each value represents
the mean Æ SE of results from four trials. *P < 0.05: significantly dif-
*
*P < 0.01: significantly different from the value for control.
††
P < 0.01: significantly different from the value for control plasma.
increasing tendency of DNP-SG accumulation in washed
erythrocytes.
Different effects of kynurenic acid and L-
kynurenine on MRP function between IOVs
and washed erythrocytes
†
ferent from the value for control. P < 0.05: significantly different
from the value for control plasma.
Kynurenic acid inhibited DNP-SG uptake by IOVs in a
concentration-dependent manner, whereas L-kynurenine, a
precursor of kynurenic acid, did not show any inhibitory
effect on DNP-SG uptake even at a concentration of
100 lM (Figure 7a). In contrast, kynurenic acid did not
show any significant effect on DNP-SG accumulation in
washed erythrocytes, but L-kynurenine clearly increased the
DNP-SG concentration in erythrocytes in a concentration-
dependent manner (Figure 7b).
protein-bound solutes of uraemic toxins such as p-cresol,
hippuric acid, indoxyl sulphate, indole-3-acetic acid,
kynurenic acid and CMPF significantly inhibited DNP-SG
uptake by IOVs (Figure 5b).
Screening of multidrug resistance-
associated proteins modulator by washed
erythrocytes in vitro
Discussion
MK-571 and bilirubin significantly increased DNP-SG
accumulation in washed erythrocytes after application of
CDNB (Figure 6a). Protein-bound solutes of uraemic tox-
ins exerted no significant effect on DNP-SG accumulation
in washed erythrocytes (Figure 6b), different from the case
of IOVs (Figure 5b). Among them, however, L-kynurenine,
a precursor of kynurenic acid, and CMPF showed an
The effect of glycerol-induced ARF state on MRP func-
tion in erythrocytes was evaluated in rats. In control rats
in vivo, DNP-SG was not accumulated in erythrocytes
after administration of CDNB (Figures 1c and 2b),
possibly due to the rapid metabolism of CDNB to
DNP-SG by GST in erythrocytes followed by rapid
6
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**

Aoi Matsushima et al.
Modulated MRPs function in erythrocytes
plasma is eliminated, different from the case in in vivo
condition. In IOVs, plasma components obtained from
control and ARF rats significantly increased DNP-SG
accumulations, and ARF plasma components showed
greater DNP-SG accumulation, indicating that MRP inhi-
bitors are contained in plasma of control and ARF rats,
and the amount is greater in ARF rats than untreated
control rats (Figure 4). In the uptake study of DNP-SG
by IOVs, most of protein-bound uraemic toxins examined
inhibited DNP-SG uptake significantly, in addition to
endogenous MRP substrate bilirubin (Figure 5). In
washed erythrocytes, 10% plasma components obtained
from ARF rats exhibited great accumulation of DNP-SG
than that obtained from control rats (Figure 3). These
results would further suggest that membrane-permeable
MRP inhibitors are present in ARF plasma at a greater
amount than control rats. Among various test com-
pounds, bilirubin alone increased DNP-SG accumulation,
and uraemic toxins of protein-bound solutes examined
did not increase DNP-SG accumulation significantly
(
Figure 6). The discrepancy of inhibitory effects of pro-
tein-bound solutes between IOVs and washed
erythrocytes was considered to be derived from the mem-
brane permeation process of protein-bound solutes. In
inhibiting DNP-SG uptake by IOVs, the membrane per-
meation of uraemic toxin is not necessary, because MRP
transporter is located outside of erythrocyte vesicles. In
contrast, to inhibit MRP function in erythrocytes, inhibi-
tors must enter inside of erythrocytes across erythrocyte
membranes. In washed erythrocytes, L-kynurenine and
CMPF showed an increasing tendency of DNP-SG accu-
mulation, although the effect did not reach significant dif-
ference (Figure 6). To examine the possible membrane
permeation of L-kynurenine, the inhibitory effects of
kynurenic acid and L-kynurenine were compared in IOVs
and washed erythrocytes by increasing their concentra-
tions. Kynurenic acid significantly inhibited the DNP-SG
uptake by IOVs in a concentration-dependent manner,
but exerted no significant effect on DNP-SG accumula-
tion in washed erythrocytes. In contrast, L-kynurenine
exerted no significant effect on DNP-SG uptake by IOVs
but increased DNP-SG accumulation significantly in
Figure 5 Effects of small water-soluble solutes (a) and protein-
bound solutes (b) of uraemic toxins on 2,4-dinitrophenyl-S-glutathione
uptake by inside-out erythrocyte membrane vesicles (IOVs). Inside-out
erythrocyte membrane vesicles were incubated with 2,4-dinitrophenyl-
S-glutathione (20 lM) for 30 min at 37 °C. Each value represents the
mean Æ SE of results from three trials. *P < 0.05, **P < 0.01: signifi-
cantly different from the value for control.
MRPs-mediated efflux of DNP-SG from erythrocytes. In
contrast, in glycerol-induced ARF rats, DNP-SG was
greatly accumulated in erythrocytes (Figures 1c and 2b),
indicating that the MRPs-mediated efflux transport of
DNP-SG in erythrocytes was suppressed in glycerol-
induced ARF rats. The screening of modulator(s) for
MRPs function was performed using IOVs and washed
erythrocytes (without plasma protein) in vitro. IOVs of
erythrocytes would be used as a simple and rapid screen-
washed erythrocytes in
a
concentration-dependent
[
22,25]
ing tool of MRPs modulators against erythrocytes.
manner. Based on these results, L-kynurenine is consid-
ered to be taken up by erythrocytes and metabolized to
kynurenic acid in erythrocytes. In contrast, kynurenic acid
being MRPs inhibitor cannot penetrate erythrocyte mem-
brane. L-Kynurenine is known as a substrate for L-amino
acid transporter, and L-amino acid transporter is
However, IOVs technique does not involve the membrane
permeability process of testing modulating compounds.
To evaluate both of the membrane permeability and MRP
inhibitory effect of test compounds, washed erythrocyte
technique is preferable, in which the maximal membrane
permeability and therefore the maximal MRP modulating
effect can be obtained, because of the absence of plasma
protein binding, different from the case of in vivo condi-
tion. In addition, the metabolism of CDNB to DNP-SG in
[
26]
expressed in erythrocyte membranes.
aminotransferase, a metabolic enzyme converting L-kynur-
Also, kynurenine
[
27,28]
enine to kynurenic acid, exists in erythrocytes.
addition, CMPF also tended to increase DNP-SG
In
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**
7

Modulated MRPs function in erythrocytes
Aoi Matsushima et al.
Figure 6 Effects of MK-571 and bilirubin (a) and various protein-bound solutes of uraemic toxins (b) on 2,4-dinitrophenyl-S-glutathione accumu-
lation after application of 1-chloro-2,4-dintrobenzene in washed erythrocytes. Control value represents the results in the absence of multidrug
resistance-associated protein substrate/inhibitor in the medium (37 °C). Concentrations of 2,4-dinitrophenyl-S-glutathione in erythrocytes were
determined 15 min after application of 1-chloro-2,4-dintrobenzene (10 lM) at 37 °C. Each value represents the mean Æ SE of results from three
trials. **P < 0.01: significantly different from the value for control.
accumulation in erythrocytes (Figure 6). The partition
coefficient (log P) of CMPF between hydrochloric acid
and octanol is 1.2, and the distribution coefficient (log D)
between octanol and phosphate buffer at pH 7.4 is À0.59,
suggesting the possibility of membrane permeation
by simple perfusion due to its slight hydrophobic
erythrocytes. In normal condition, however, bilirubin
binds to albumin extensively at the level of more than
[
32]
99.9%,
indicating that the unbound bilirubin concen-
tration in plasma is very low. However, it is also known
that the protein binding of bilirubin is decreased in
renal patients, and CMPF was suggested as one of the
substances which contribute to the decreased binding
[
29,30]
character.
[
36]
Among various endogenous substances examined in
the present study, bilirubin significantly suppressed
MRP-mediated transport in IOVs and washed erythro-
cytes. The plasma concentrations of total bilirubin and
direct bilirubin were significantly increased in glycerol-
induced ARF rats, compared to control rats (Table 1),
possibly due to the haemolysis caused by glycerol. Indi-
rect bilirubin (unconjugated bilirubin) is a compound
with high permeability through lipid bilayer mem-
capacity of bilirubin in uraemic serum.
In general,
bilirubin is excreted into the intestinal lumen by biliary
excretion, not into urine under normal healthy condi-
tion, and the concentration of bilirubin in plasma
increases only under hepatic failure state. However, as
described already in the introduction, the high concen-
tration of indirect bilirubin has also been observed in
[
17]
haemolytic uraemic syndrome patients,
molytic uraemic syndrome is involved in all cases of
and the hae-
[
31,32]
[18]
brane.
bilirubin from plasma to the liver was also reported.
Recently, the uphill transport of indirect
[
renal dysfunction.
In the present study, the plasma
32]
concentration of bilirubin, in addition to uraemic toxins
such as BUN and IS, was significantly increased in glyc-
erol-induced ARF rats, possibly by haemolytic action of
glycerol. It has been reported that the P-glycoprotein
function in the liver was significantly suppressed in
glycerol-induced rats, in which the biliary excretion of
rhodamine 123, a typical substrate for P-glycoprotein,
Indirect bilirubin is metabolized to glutathione-conju-
gated bilirubin, a MRP substrate, by glutathione S-trans-
[
33,34]
ferases (GST) in hepatic cytosol.
rat erythrocytes.
GST also exists in
Thus, it may be considered that a
[35]
part of bilirubin distributed into erythrocytes by passive
diffusion is conjugated by glutathione, and both indirect
and direct bilirubin may inhibit MRP function in
[
15,37]
was significantly decreased in ARF rats.
Thus, it
8
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**

Aoi Matsushima et al.
Modulated MRPs function in erythrocytes
on the bile canalicular membrane, to clarify the mecha-
nism of the increased plasma level of bilirubin in glyc-
erol-induced ARF rats. In addition, it will be important
to examine the effect of the bilirubin accumulation in
plasma on renal function such as glomerular filtration
rate (GFR) and tubular secretion.
The suppression of MRP function in erythrocytes
under ARF state may not be explained only by several
endogenous compounds including bilirubin, L-kynurenine
and CMPF, by considering their unbound concentrations
in plasma. However, various endogenous compounds are
accumulated in plasma under ARF states, and they
would increase their unbound fractions by displacing the
protein binding each other. Further study is necessary
to clarify the contribution of protein binding under
ARF state.
Conclusion
In the present study, DNP-SG, a typical MRP substrate,
was found to be greatly accumulated in erythrocytes
after application of CDNB, a precursor of DNP-SG, in
glycerol-induced ARF rats, possibly due to the suppres-
sion of MRPs-mediated efflux transport in erythrocytes.
Some membrane-permeable endogenous substances such
as bilirubin, L-kynurenine and CMPF were thought to
have suppressive effects on MRP function in erythro-
cytes cooperatively. These results would suggest that
renal failure state can cause accumulation of MRPs sub-
strates in erythrocytes.
Figure 7 Effects of kynurenic acid and L-kynurenine on 2,4-dinitro-
phenyl-S-glutathione uptake by inside-out erythrocyte membrane vesi-
cles (a) and on 2,4-dinitrophenyl-S-glutathione accumulation after
application of 1-chloro-2,4-dintrobenzene in washed erythrocytes (b).
Inside-out erythrocyte membrane vesicles were incubated with
2,4-dinitrophenyl-S-glutathione (20 lM) for 30 min at 37 °C. Concen-
trations of 2,4-dinitrophenyl-S-glutathione in erythrocytes were deter-
mined 15 min after application of 1-chloro-2,4-dintrobenzene (10 lM)
at 37 °C. Each value represents the mean Æ SE of results from three
trials. *P < 0.05, **P < 0.01: significantly different from the value for
control.
Declaration
Conflict of interest
The Author(s) declare(s) that they have no conflict of inter-
est to disclose.
may be necessary to examine the effect of glycerol-
induced ARF state on the function of MRP2 expressed
erythromycin. Drug Metab Dispos
6. Vanholder R et al. What is new in
uremic toxicity? Pediatr Nephrol 2008;
23: 1211–1221.
References
2
004; 32: 1239–1246.
1
. Sun H et al. Effects of renal failure on
drug transport and metabolism. Phar-
macol Ther 2006; 109: 1–11.
4. Deguchi T et al. Involvement of organic
anion transporters in the efflux of
uremic toxins across the blood-brain
barrier. J Neurochem 2006; 96: 1051–
1059.
7. Akiyama Y et al. Indoxyl sulfate down-
regulates SLCO4C1 transporter thr-
ough up-regulation of GATA3. PLoS
One 2013; 8: e66518.
2
. Dreisbach AW, Lertora JJ. The effect
of chronic renal failure on drug meta-
bolism and transport. Expert Opin
Drug Metab Toxicol 2008; 4: 1065–
5. Tsujimoto M et al. Influence of serum
in hemodialysis patients on the expres-
sion of intestinal and hepatic trans-
porters for the excretion of pravastatin.
Ther Apher Dial 2012; 16: 580–587.
8. Saito H et al. Hepatic sulfotransferase as
a nephropreventing target by suppres-
sion of the uremic toxin indoxyl sulfate
accumulation in ischemic acute kidney
injury. Toxicol Sci 2014; 141: 206–217.
1074.
3. Sun H et al. Effect of uremic toxins
on hepatic uptake and metabolism of
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**
9

Modulated MRPs function in erythrocytes
Aoi Matsushima et al.
9
. Sodani K et al. Multidrug resistance
associated proteins in multidrug resis-
tance. Chin J Cancer 2012; 31: 58–72.
its mercapturic acid. Metabolism of 29. Costigan MG et al. Synthesis and
1-chloro-2,4-dinitrobenzene in isolated
perfused rat and guinea pig livers. J Biol
Chem 1991; 266: 22179–22185.
physicochemical properties of the
furan dicarboxylic acid, 3-carboxy-4-
methyl-5-propyl-2-furanpropanoic acid,
an inhibitor of plasma protein binding
in uraemia. J Pharm Pharmacol 1996;
48: 635–640.
1
1
1
1
0. Takano M et al. Expression and func-
tion of efflux drug transporters in the 20. Yokooji
T
et al. Modulation of
intestine. Pharmacol Ther 2006; 109:
37–161.
intestinal transport of 2,4-dinitrophe-
nyl-S-glutathione, a multidrug resis-
1
1. Murakami T, Takano M. Intestinal
efflux transporters and drug absorp-
tion. Expert Opin Drug Metab Toxicol
tance-associated protein 2 substrate, 30. Vanholder R, De Smet R. Pathophysi-
by bilirubin treatment in rats. J Pharm
Pharmacol 2005; 57: 579–585.
21. Yokooji T et al. Function of mul-
ologic effects of uremic retention
solutes. J Am Soc Nephrol 1999; 10:
1815–1823.
2008; 4: 923–939.
2. Yokooji T et al. Modulated function
of tissue efflux transporters under
hyperbilirubinemia in rats. Eur J Phar-
macol 2010; 636: 166–172.
tidrug resistance-associated protein 2 31. Zucker SD et al. Unconjugated biliru-
in acute hepatic failure rats. Eur J
Pharmacol 2006; 546: 152–160.
bin exhibits spontaneous diffusion
through model lipid bilayers and
native hepatocyte membranes. J Biol
Chem 2003; 274: 10852–10862.
22. Ogawa K et al. Interaction of valproic
acid and carbapenem antibiotics with
3. Klokouzas A et al. cGMP and glu-
tathione-conjugate transport in human
erythrocytes. The roles of the mul-
tidrug resistance-associated proteins,
multidrug resistance-associated pro- 32. Levitt DG, Levitt MD. Quantitative
teins in rat erythrocyte membranes.
assessment of the multiple processes
responsible for bilirubin homeostasis
in health and disease. Clin Exp Gas-
troenterol 2014; 7: 307–328.
Epilepsy Res 2006; 71: 76–87.
MRP1, MRP4, and MRP5. Eur J Bio- 23. Uremic toxin-Data base from Euro-
chem 2003; 270: 3696–3708.
pean Uremic Toxin Work Group.
http://www.uremic-toxins.org/DataBa-
se.html (February 13, 2016).
14. K o€ ck K et al. Expression of adenosine
33. Walkoff AW et al. Hepatic accumula-
tion and intracellular binding of con-
jugated bilirubin. J Clin Invest 1978;
61: 142–149.
triphosphate-binding cassette (ABC)
drug transporters in peripheral blood 24. Bradford MM. A rapid and sensitive
cells: relevance for physiology and
method for the quantitation of micro-
pharmacotherapy. Clin Pharmacokinet
gram quantities of protein utilizing 34. Simons PC, Jagt DL. Bilirubin binding
2
007; 46: 449–470.
the principle of protein-dye binding.
to human liver ligandin (glutathione
S-transferase). J Biol Chem 1980; 255:
4740–4744.
1
1
1
1
1
5. Huang ZH et al. Expression and func- Anal Biochem 1976; 72: 248–254.
tion of P-glycoprotein in rats with 25. Taogoshi T et al. Transport of prosta-
glycerol-induced acute renal failure.
Eur J Pharmacol 2000; 406: 453–460.
6. Lochhead KM et al. Anesthetic effects
glandin E1 across rat erythrocyte 35. Chikezie PC, Uwakwe AA. Activities
membrane. Biol Pharm Bull 2008; 31:
1288–1291.
of three erythrocyte enzymes of
hyperglycemic rats (Rattus norvegicus)
treated with Allium sativa extract.
J Diabetes Metab Disord 2014; 13:
50.
on the glycerol model of rhabdomyol- 26. Yao SY et al. Reconstitution studies of
ysis-induced acute renal failure in rats.
J Am Soc Nephrol 1998; 9: 305–309.
7. Chen SY et al. Nonenteropathic hemo-
amino acid transport system L in rat
erythrocytes. Biochem
655–660.
J 1993; 292:
36. Tsutsumi Y et al. Decreased bilirubin-
binding capacity in uremic serum
caused by an accumulation of furan
dicarboxylic acid. Nephron 2000; 85:
60–64.
lytic uremic syndrome: the experience 27. Hartai Z et al. Kynurenine metabo-
of a medical center. Pediatr Neonatol
011; 52: 73–77.
8. Trachtman H et al. Renal and neuro- 28. Hara T et al. High-affinity uptake of
lism in multiple sclerosis. Acta Neurol
Scand 2005; 112: 93–96.
2
logical involvement in typical Shiga
kynurenine and nitric oxide-mediated 37. Murakami T et al. Factors affecting
toxin-associated HUS. Nat Rev Nephrol
inhibition of indoleamine 2,3-dioxy-
genase in bone marrow-derived mye-
loid dendritic cells. Immunol Lett
2008; 116: 95–102.
the expression and function of
P-glycoprotein in rats: drug treat-
ments and diseased states. Pharmazie
2002; 57: 102–107.
2012; 8: 658–669.
9. Hinchman CA et al. Intrahepatic con-
version of a glutathione conjugate to
1
0
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2016), pp. **–**