Role of human organic anion transporter 4 in the

transport of ochratoxin A

Ellappan Babu a,b, Michio Takeda a, Shinichi Narikawa a, Yukari Kobayashi a,Atsushi Enomoto a, Akihiro Tojo c, Seok Ho Cha a, Takashi Sekine a,

Dhanapal Sakthisekaran b, Hitoshi Endou a,*

aDepartment of Pharmacology and Toxicology, Kyorin University School of Medicine, 6-20-2 Shinkawa, Mitaka, Tokyo 181-8611, JapanbDepartment of Medical Biochemistry, University of Madras, Chennai, India

cDivision of Nephrology, Department of Internal Medicine, University of Tokyo, Tokyo, Japan

Received 21 May 2001; received in revised form 21 January 2002; accepted 4 February 2002

Abstract

The purpose of this study was to investigate the characteristics of ochratoxin A (OTA) transport by multispecific human organic anion

transporter 4 (hOAT4) using mouse proximal tubule cells stably transfected with hOAT4 (S2 hOAT4). Immunohistochemical analysis

revealed that hOAT4 protein was localized to the apical side of the proximal tubule. S2 hOAT4 expressed hOAT4 protein in the apical side as

well as basolateral side and the cells were cultured on the plastic dish for experiments. S2 hOAT4 exhibited a time- and concentration-

dependent, and a saturable increase in OTA uptake, with an apparent Km value of 22.9F 2.44 AM. The OTA uptakes were inhibited by

several substrates for the OATs. Probenecid, piroxicam, octanoate and citrinin inhibited OTA uptake by hOAT4 in a competitive manner

(Ki = 44.4–336.4 AM), with the following order of potency: probenecid>piroxicam>octanoate>citrinin. The efflux of OTA by S2 hOAT4 was

higher than that by mock. Addition of OTA resulted in slight decrease in viability of S2 hOAT4 compared with mock. These results indicate

that hOAT4 mediates the high-affinity transport of OTA on the apical side of the proximal tubule, whereas the transport characteristics of

OTA are distinct from those by basolateral OATs. D 2002 Elsevier Science B.V. All rights reserved.

Keywords: Ochratoxin A; Organic anion transporter; Proximal tubule; Transport; Cell line

1. Introduction

Ochratoxin A (OTA) is a naturally occurring mycotoxin

produced as a secondary metabolite in different species of

Aspergillus and Penicillum [1]. Contamination of foods,

particularly cereals, nuts and coffee beans, with OTA has

been noted in Eastern and Central Europe, North Africa,

North America and Japan [2]. As shown in Fig. 1A, OTA is

a structural analogue of phenylalanine and contains a

chlorinated dihydroisocoumarin moiety [3]. Recently, OTA

has received considerable attention because of its deleteri-

ous effects on human and animal health. OTA accumulates

in several tissues in the body, with the kidney being its main

target where it exerts toxic and carcinogenic effects [4].

Three forms of human renal disease appear to be also caused,

at least in part, by enhanced exposure to OTA: Balkan

endemic nephropathy, chronic interstitial nephritis and kar-

yomegalic interstitial nephritis [5]. However, the pathophy-

siological role of OTA in human renal disease has not yet

been completely determined.

In the kidney, OTA mainly impairs proximal tubular

functions and causes glucosuria, enzymuria and a decrease

in para-aminohippuric acid (PAH) transport [6–8]. An in

vivo study revealed that probenecid, a typical inhibitor of

organic anion transporter (OAT), inhibited the renal clear-

ance of OTA [9]. From these results, it is thought that the

excretion of OTA into the urine is mainly by tubular

secretion, presumably via the OAT system, and this may

play an important role in OTA accumulation and develop-

ment of nephrotoxicity. On the other hand, the mechanism

for the reabsorption of OTA on the apical side of the

proximal tubule has been elucidated in membrane vesicles,

cultured cells and isolated proximal tubules [9–12]. Using

the rat nephron segments, it was shown that pH-independent

0167-4889/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.

PII: S0167 -4889 (02 )00187 -8

* Corresponding author. Tel.: +81-422-47-5511x3451;

fax: +81-422-79-1321.

E-mail address: endouh@kyorin-u.ac.jp (H. Endou).

www.bba-direct.com

Biochimica et Biophysica Acta 1590 (2002) 64–75

reabsorption of OTA in the proximal straight tubule is

mediated by OAT-K1, whereas pH-dependent reabsorption

in the proximal convoluted tubule is mediated by H + -

dipeptide co-transporter [8,11]. However, the molecular

mechanism for the apical OTA transport remains unknown.

The secretion or reabsoption of various anionic substan-

ces, drugs and xenobiotics is an important physiological

function of the renal proximal tubule [13]. Recently, cDNAs

encoding transporters mediating organic anion transport

have been successively cloned, including OAT1 [14,15],

OAT2 [16], OAT3 [17,18], OAT4 [19], OAT-K1 [20], OAT-

K2 [21], organic anion-transporting polypeptide (oatp1)

[22], oatp2 [23], oatp3 [24], multiple resistance-associated

protein 2 (MRP2) [25] and human-type I sodium-dependent

inorganic phosphate transporter (NPT1) [26].

We have recently characterized OTA transport by human

OAT1 (hOAT1) and human OAT3 (hOAT3) using proximal

tubule cells stably expressing hOAT1 and hOAT3 (S2hOAT1 and S2 hOAT3) [27]. In the current study, we

undertook to elucidate the role of human OAT4 (hOAT4)

in OTA transport on the apical side of the proximal tubule

using mouse proximal tubule cells stably transfected with

hOAT4.

2. Materials and methods

2.1. Materials

[3H]OTA (547.6 GBq/mmol) was purchased from Mor-

avek Biochemicals Inc. (Brea, CA, USA). [3H]Estrone

sulfate (ES) (1961 GBq/mmol) and [14C]mannitol (532.4

GBq/mmol) were purchased from Perkin Elmer Life Scien-

ces (Boston, MA, USA). OTA, PAH, citrinin, octanoate,

piroxicam, penicillin G, cimetidine, aspartame and indome-

thacin were obtained from Sigma Chemical Co. (St. Louis,

MO, USA). Fetal bovine serum (FBS), trypsin and geneticin

were from Gibco/BRL (Gaithersburg, MD, USA), recombi-

nant epidermal growth factor (EGF) were from Wakunaga

(Hiroshima, Japan), insulin was from Shimizu (Shizuoka,

Japan), RITC 80-7 culture medium was from Iwaki Co.

(Tokyo, Japan), and TfX-50 was from Promega (Madison,

WI, USA).

2.2. Immunohistochemical analysis of hOAT4 protein in

human kidney

For the generation of antibody against hOAT4, rabbits

were immunized with keyhole limpet hemocyanin-conju-

gated synthesized peptides, GNRQEAVTVESTSL, corre-

sponding to cysteine and the 14 amino acids of the COOH

terminus of hOAT4.

Light-microscopic analysis of the hOAT4 protein was

performed as previously described [28]. Briefly, waxed

sections (2 Am) were cut and stained by the labeled

streptavidin–biotin method. After dewaxing, the sections

were incubated with 3% H2O2 for 15 min and then with

blocking serum for 15 min. The sections were then incu-

bated with a polyclonal antibody against hOAT4 (1:500

dilution) for 2 h. The sections were rinsed with Tris-buffered

saline containing 0.1% Tween 20 (TBST) and incubated

with the biotinylated secondary antibody against rabbit

immunoglobulin (Dako, Glostrup, Denmark) for 1 h. After

rinsing with TBST, the sections were incubated for 30 min

with horseradish peroxidase (HRP)-conjugated streptavidin

solution. HRP labeling was detected using a peroxidase

substrate solution with diaminobenzidine (0.8 mM; Dojindo

Laboratories, Kumamoto, Japan). The sections were coun-

terstained with hematoxylin before examination under a

light microscope. For preabsorption experiment, the hOAT4

peptide (200 Ag/ml) was added to the hOAT4-specific

antibody solution and incubated overnight at 4 jC. Usingthis preabsorbed antibody, the immunohistochemistry was

performed as described above.

2.3. Cell culture and establishment of cells stably expressing

hOAT4

As described previously [29], the second portion of the

proximal tubule (S2), derived from transgenic mice harbor-

ing the temperature-sensitive simian virus 40 large T-antigen

Fig. 1. Chemical structure of ochratoxin A (A), aspartame (B) and piroxicam

(C).

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–75 65

gene was microdissected and S2 cells were established. The

full-length cDNA of hOAT4 [19] was subcloned into

pcDNA3.1 (Invitrogen, Carlsbad, CA, USA), a mammalian

expression vector. S2 cells stably expressing hOAT4 (S2hOAT4) were obtained by transfecting S2 cells with

pcDNA3.1-hOAT4 coupled with pSV2neo, a neomycin

Fig. 2. Immunohistochemical staining illustrating hOAT4 protein in human kidney and cells stained with polyclonal antibody against hOAT4. (A) human

kidney cortex (� 100), (B) human kidney cortex(� 400), (C) human kidney medulla (� 100), (D) human kidney cortex (� 400), (E) S2 hOAT4 (� 100) and

(F) mock (� 100). Immunoreactivity was completely abolished by pretreatment of antibody with hOAT4 oligopeptide (D). a: arteriole, C: cortical collecting

duct, D: distal tubule, O: outer medullary collecting duct and P: proximal tubule.

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–7566

resistance gene, using a liposome [30]. S2 cells transfected

with pcDNA3.1 lacking an insert and pSV2neo were des-

ignated as S2 pcDNA3.1 and used as the control (mock). S2hOAT4 and mock were grown in a humidified incubator at

33 jC and 5% CO2, using RITC 80-7 medium that con-

tained 5% FBS, 10 Ag/ml transferrin, 0.08 U/ml insulin, 10

ng/ml recombinant EGF and 400 Ag/ml geneticin [30]. The

cells were subcultured in a medium containing 0.05%

trypsin–EDTA solution (containing in mM: 137 NaCl, 5.4

KCl, 5.5 glucose, 4 NaHCO3, 0.5 EDTA and 5 HEPES; pH

Fig. 2 (continued ).

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–75 67

7.2) and used for 25–30 passages. Clonal cells were isolated

using a cloning cylinder, and screened by determining the

optimal substrate for hOAT4, [3H]ES [19]. S2 hOAT4

exhibited a dose- and time-dependent increase in ES uptake

activities. The S2 monolayer was determined to be leaky

based on the results of a study wherein we estimated

paracellular secretion from cells cultured on a permeable

support, using D-[14C]mannitol as an indicator. Light-micro-

scopic analysis of the hOAT4 protein in S2 hOAT4 was

performed as previously described [30]. The cells were

cultured on Transwell chambers (Coster, Cambridge, MA,

USA). After the cells were washed three times with 50 mM

NH4Cl in phosphate-buffered saline (PBS) (containing in

mM: 136 NaCl, 2.7 KCl, 8.2 NaHPO4�12H2O and 15

KHPO4, pH 7.4), they were fixed with paraformaldehyde

lysine periodate (PLP) solution (containing 0.01 M NaIO4,

0.075 M lysine, 0.0375 M phosphate buffer with 2%

paraformaldehyde; pH 6.2) for 2 h at 4 jC. The fixed cells

were stained with the antibody against hOAT4 and HRP as

described above. In addition, vertical sections of S2 hOAT4

stained with polyclonal antibody against hOAT4, showed

that the subcellular localization of hOAT4 protein was

mainly on the cell membrane (Fig. 2E), whereas no staining

was observed in mock (Fig. 2F). Both the basolateral and

apical portions of the membrane showed positive staining.

Therefore, the cells were cultured on a solid support for use

in the following experiments.

2.4. Uptake experiment

Uptake experiments were performed as previously

described [30]. The S2 cells were seeded in 24-well tissue

culture plates at a cell density of 1�105 cells/well. After 2

days of culture, the cells were washed three times with

Dulbecco’s modified PBS (D-PBS) solution (containing in

mM: 137 NaCl, 3 KCl, 8 NaHPO4, 1 KH2PO4, 1 CaCl2 and

0.5 MgCl2; pH 7.4) containing 5 mM D-glucose, and then

preincubated with the same solution for 10 min in a water

bath at 37 jC. The cells were then incubated in D-PBS

containing 5 mM D-glucose with [3H]OTA at 100 nM for

time course experiments, and 1 AM for inhibition experi-

ments at 37 jC. The uptake was stopped by the addition of

ice-cold D-PBS, and the cells were washed three times with

the same solution. The cells in each well were lysed with 0.5

ml of 0.1 N sodium hydroxide and 2.5 ml of aquasol-2, and

radioactivity was determined using a h-scintillation counter

(Aloka, LSC-3100).

2.5. Inhibition study

To evaluate the inhibitory effects of several substrates for

an OAT on hOAT4-mediated OTA uptake, S2 hOAT4 was

incubated for 1 min at 37 jC in D-PBS containing 1 AM[3H]OTA in the absence or presence of various inhibitors

(0.2 mM) including PAH, probenecid, penicillin G, aspar-

Fig. 2 (continued ).

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–7568

tame, octaonate, cimetidine, citrinin, indomethacin and

piroxicam. PAH, probenecid, penicillin G and cimetidine

were dissolved in D-PBS. Indomethacin was initially dis-

solved in ethanol and then diluted with D-PBS. Aspartame,

octanoate, citrinin and piroxicam were dissolved in dime-

thylsulfoxide and then diluted with the incubation medium.

The final concentrations of ethanol and dimethylsulfoxide

were adjusted to less than 0.1%, which did not affect the

hOAT4-mediated OTA uptake in our system.

2.6. Kinetic analysis

After preincubation as described above, S2 hOAT4 was

incubated for 1 min at 37 jC in D-PBS containing different

concentrations of [3H]OTA in the absence or presence of

various inhibitors. The concentrations of the inhibitors used

are listed in the Table 1. On the basis of [3H]OTA uptake

under each condition, double reciprocal plot analyses were

performed as previously described [31]. When the inhibition

was found to be competitive, Ki values were calculated

based on the following equation:

Ki ¼ concentration of inhibitor=½ðKm OTA with inhibitor

=Km OTA without inhibitorÞ � 1�

2.7. Efflux study

Efflux study was performed as previously described

[30]. S2 hOAT4 and mock were seeded in 24-well tissue

Table 1

The Ki values of the inhibitory effects of various substrates on OTA uptake

by hOAT4

Drug Ki

value

(AM)

Inhibitor

(AM)

Total plasma

concentration

(AM)

Unbound

fraction

(%)

Unbound

plasma

concentration

(AM)

Probenecid 44.4 500 170 [40] 11.0 [41] 18.7

Piroxicam 107.8 400 24.0 [42] 0.02 [42] 0.0048

Octanoate 235.5 1000

Citrinin 366.4 1000

S2 hOAT4 was incubated with solution containing various concentrations of

[3H]OTA in the absence or presence of various substrates at 37 jC. The Ki

values were estimated from a Lineweaver–Burk plot.

Fig. 3. Time course of OTA uptake by hOAT4. S2 hOAT4 and mock were

incubated in solution containing 100 nM [3H]OTA at 37 jC for 30 s to

5 min. Each value represents the meanF S.E. of four determinations.

Fig. 4. Specificity of OTA uptake by hOAT4. S2 hOAT4 and mock were

incubated in solution containing 100 nM [3H]OTA and 100 nM

[14C]mannitol at 37 jC for 5 min. (A) OTA, (B) mannitol. Each value

represents the meanF S.E. of four determinations. *P < 0.05 and * *P <

0.001 vs. mock.

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–75 69

culture plates at a cell density of 1�105 cells/well. After

the cells were cultured for 2 days, they were washed three

times with D-PBS and then preincubated in the same

solution for 10 min in a water bath at 37 jC. Thereafter,the monolayers were incubated with 300 nM [3H]OTA for

30 min at 37 jC, washed immediately once with 1500 AlD-PBS, and incubated at 37 jC with 500 Al of D-PBS.

Since the time course of OTA efflux was prompt and three

times washing of cells takes time, the significant amount of

OTA was effluxed during washing of cells with 500 Al D-PBS. Thus, in order to minimize the OTA efflux during

washing, the cells were washed once with 1500 Al D-PBS.After the incubation for the indicated periods of time, 50 Alof the supernatant was collected. Then, the medium was

aspirated immediately and the cell monolayers were washed

three times with the medium and solubilized in 0.5 ml of 0.1

N sodium hydroxide. The radioactivities within the super-

natant and cell lysates were measured by counting the

radioactivity. The rate of efflux at each time point was

presented using the following formula: (effluxed [3H]OTA

by S2 hOAT4� effluxed [3H]OTA by mock)/([3H] OTA by

S2 hOAT4 at time 0� [3H]OTA accumulated by mock at

time 0).

2.8. Cell viability assay

Cell viability was assessed using the MTT assay

[30,32,34]. As described above, after S2 hOAT4 and mock

were seeded at a cell density of 2� 105 cells/well in 24-

well plates and cultured for 2 days, cells were cultured in a

medium containing 5% FBS with or without OTA for 2

days at 33 jC. The number of S2 hOAT4 and mock 2 days

after inoculation was 6� 105 and 4� 105 cells/well,

respectively. After incubation, 50 Al of 5 mg/ml of MTT

(3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bro-

mide) was added to the medium (0.5 ml), and the cells

were further incubated for 4 h. After solubilizing the cells

with isopropanol/HCl solution, optical intensity was meas-

ured at 570 nm with 630 nm as the reference (Beckman,

Du640).

2.9. Statistical analysis

Data are expressed as means + S.E. Statistical differences

were determined using Student’s unpaired t-test. Differences

were considered significant at P < 0.05.

Fig. 6. Effects of various inhibitors on hOAT4-mediated OTA uptake. S2hOAT4 was incubated in solution containing 1 AM [3H]OTA in the absence

or presence of various inhibitors (0.2 mM) at 37 jC for 1 min. Each value

represents the meanF S.E. of four determinations. *P < 0.05, **P < 0.01

and ***P < 0.001 vs. control.

Fig. 5. Concentration dependence of OTA uptake by hOAT4. (A) Con-

centration dependence and (B) Eadie–Hofstee plot of the uptake of

[3H]OTA. S2 hOAT4 and mock were incubated in solution containing 2.5 to

40 AM [3H]OTA at 37 jC for 1 min. The values for the uptake by mock were

subtracted from those by S2 hOAT4. Each value represents meanF S.E. of

four determinations from one typical experiment.

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–7570

3. Results

3.1. Immunohistochemical analysis of hOAT4 protein in

human kidney

Light-microscopic analysis of 2-Am waxed sections

demonstrated that there was specific immunostaining for

hOAT4 in the proximal tubules (Fig. 2A–C). As shown in

Fig. 2A and B, in the cortex, there was staining for hOAT4

in the proximal convoluted tubules. In addition, in the outer

medulla, hOAT4 was located in the proximal straight

tubules (Fig. 2C). In contrast, no immunostaining for

hOAT4 was observed in glomeruli, the other nephron seg-

ments and arterioles. By preincubation of the antibody with

hOAT4 peptide, the immunostaining was completely dimin-

ished (Fig. 2D). The specificity of the antibody for hOAT4

was verified by these results.

3.2. OTA uptake by hOAT4

We evaluated the time-dependent uptake of OTA in S2hOAT4. As shown in Fig. 3, the amount of OTA uptake

Fig. 7. Kinetic analysis of the inhibitory effects of various inhibitors on OTA uptake by hOAT4. S2 hOAT4 was incubated in solution containing various

concentrations of [3H]OTA (2.5–80 AM) in the presence or absence of probenecid (A), piroxicam (B), octanoate (C) and citrinin (D) at 37 jC for 1 min, and

analysis of Lineweaver–Burk plots was performed from one typical experiment.

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–75 71

by S2 hOAT4 was significantly higher than that by mock.

In order to confirm the specificity of OTA uptake medi-

ated by hOAT4, S2 hOAT4 and mock were incubated in a

solution containing 100 nM [3H] OTA and 100 nM [14C]

mannitol for 5 min. As shown in Fig. 4, both OTA (A)

and mannitol (B) uptake values by S2 hOAT4 were higher

than that by mock (N = 4, *P < 0.05 and **P < 0.001 vs.

mock). However, OTA uptake by hOAT4 was approxi-

mately fourfold of that by mock, whereas mannitol uptake

by hOAT4 was approximately twofold of that by mock.

Thus, it was suggested that hOAT4 mediates the specific

uptake of OTA. The concentration dependence of hOAT4-

mediated OTA uptake was examined. The specific uptake

of OTA in S2 hOAT4 revealed saturable kinetics (Fig.

5A), and the Eadie–Hofstee plot gave a single straight

line (Fig. 5B). The estimated Km value for hOAT4-

mediated OTA uptake was 22.9F 2.44 AM (N = 4). The

results suggest that hOAT4 is responsible for the high-

affinity transport of OTA.

3.3. Inhibitory effects on OTA uptake by hOAT4

We examined the effects of various substrates on OTA

uptake by hOAT4. The substances used were as follows:

PAH, probenecid (an organic anion transport inhibitor),

penicillin G, aspartame (an artificial sweeter), octaonate (a

fatty acid), citrinin (mycotoxin), cimetidine (a H2 receptor

antagonist as well as an organic cation transport inhibitor),

indomethacin and piroxicam (an anti-inflammatory drugs).

These substances except cimetidine are organic anions.

Aspartame is the N-aspartl-phenylalanine methyl ester, thus

a structural analogue of phenylalanine and OTA [33] (Fig.

1A). In addition, as shown in Fig. 1C, piroxicam also

exhibited structural analogy to OTA. We have already

determined the inhibitory effects of these substances on

rat-OAT1-, hOAT1- and hOAT3-mediated OTA uptake

[27,34], and other researchers have also found the inhibitory

effects of some of the substances on OTA transport in

tubules or cells [33,35–39]. As shown in Fig. 6, probenecid,

penicillin G, aspartame, octanoate, citrinin, indomethacin

and piroxicam significantly inhibited hOAT4-mediated OTA

uptake (N = 4, *P < 0.05, **P < 0.01 and ***P < 0.001 vs.

control), whereas PAH and cimetidine did not (N = 4, N.S.).

3.4. Kinetic analysis of inhibitory effects of various drugs on

OTA uptake by hOAT4

To further elucidate the inhibitory effects of hOAT4-

mediated OTA uptake, inhibitory kinetics of probenecid,

piroxicam, octanoate and citrinin was analyzed. Typical

results are shown in Fig. 7. The Lineweaver–Burk plot

analysis of probenecid (A), piroxicam (B), octanoate (C)

and citrinin (D) inhibition on hOAT4-mediated OTA uptake

demonstrated that these compounds inhibited OTA uptake

by hOAT4 in a competitive manner. The Ki value for each

inhibitor tested is shown in the Table 1.

3.5. Efflux of OTA by hOAT4

In order to determine whether hOAT4 mediates the efflux

of OTA, we compared the efflux of OTA by S2 hOAT4 with

that by mock. As shown in Fig. 8, S2 hOAT4 exhibited

significant amount of OTA efflux, in which the amount of

OTA efflux by mock was subtracted.

Fig. 8. Efflux of OTA by hOAT4. After 30 min of incubation in a solution

containing 300 nM [3H]OTA at 37 jC, S2 hOAT4 and mock were incubated

in solution at 37 jC for 30 min. The amount of [3H]OTA within the

supernatant and the cell lysate was determined. The rate of efflux at each

time point was presented as the following formula: (effluxed [3H]OTA by

the S2 hOAT4� effluxed[3H]OTA by mock)/([3H]OTA accumulated by S2hOAT4 at time 0� [3H]OTA accumulated by mock at time 0). Each value

represents the mean + S.E. of three determinations.

Fig. 9. Effects of OTA on viability of S2 hOAT4. S2 hOAT4 and mock were

cultured in medium containing 5% FBS in the absence or presence of

various concentrations of OTA at 33 jC for 48 h. Each value represents the

meanF S.E. of four determinations.

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–7572

3.6. Effects of OTA on viability of S2 hOAT4

To elucidate the cytotoxic effects of OTA on S2 hOAT4,

S2 hOAT4 and mock were cultured for 48 h in the absence

or presence of various concentrations of OTA. As shown in

Fig. 9, OTA slightly inhibited viability in S2 hOAT4

compared with mock, and no significant difference was

observed (N = 4, N.S. vs. mock).

4. Discussion

HOAT4 was isolated from human placenta and also

shown to be localized to the kidney [19]. HOAT4 mediated

the uptake of various substrates including PAH, ES and

dehydroepiandrosterone sulfate; however, its substrate spe-

cificity was relatively narrow compared with that of hOAT1

and hOAT3. In the current study, hOAT4 was shown to be

localized to the apical membrane of the proximal tubule

within the kidney. Based on this, we characterized OTA

transport using S2 hOAT4. In S2 hOAT4, hOAT4 is sorted to

the apical side as well as the basolateral side (Fig. 2E). Thus,

in confluent monolayers of S2 hOAT4 on the plastic dish,

not only the apical side but also the lateral side of hOAT4

protein could be exposed to uptake solution containing

various substrates because tight junction is leaky. In this

regard, we have also isolated Madin–Darby canine kidney

(MDCK) cells stably expressing hOAT4 (MDCK hOAT4),

where the unidirectional transport of OTA from the baso-

lateral side to the apical side of the cells on the permeable

support was recognized and tight junction was well devel-

oped (data not shown). The Km value for hOAT4-mediated

OTA uptake in MDCK hOAT4 cultured on the plastic dish

was similar to that in S2 hOAT4, i.e., 56.2F 4.12 AM(unpublished observation).

In the current study, hOAT4-mediated OTA uptake was

found to increase in a dose- and time-dependent manner and

was saturable. The crucial parameter for the estimation of

the transport rate of a substrate is the concentration of its

free, unbound form. In this regard, although more than 99%

of OTA is bound to plasma proteins [43,44], only little

albumin is filtered by the glomerulus while free OTA is

filtered. Thus, free OTA concentration in the primary urine

is not necessarily negligible in the uptake of OTA into the

proximal tubule. In addition, hOAT4-mediated OTA uptake

was inhibited by various substrates for OATs. Furthermore,

hOAT4 mediated the efflux of OTA. Thus, it was suggested

that hOAT4 mediates the reabsorption and the efflux of

OTA on the apical side of the proximal tubule.

The intracellular content of OTA in S2 hOAT4 preincu-

bated in a solution containing 300 nM [3H]OTA for 30 min

was 2.11 pmol/mg protein. The intracellular volume of

cultured fibroblasts was reported to be 3.0 Al/106 cells using[14C]urea [45]. Assuming that the intracellular volume of S2hOAT4 is similar to that of fibroblasts, the intracellular

concentration of OTA in S2 hOAT4 preincubated for 30 min

is estimated as 1.17 AM. According to the results of dose-

dependent experiments (Fig. 5A), OTA uptake by S2 hOAT4

incubated in a solution containing 1.17 AM [3H]OTA for 2

min was 5.57 pmol/mg protein. On the other hand, during

the 2-min efflux after 30-min incubation, OTA efflux by S2hOAT4 was 0.99 pmol/mg protein. Based on these, OTA

efflux activity by S2 hOAT4 appears to be much lower than

OTA uptake activity.

To our knowledge, there are some reports concerning the

involvement of apical OATs in the transport of OTA. oatp-1

was shown to mediate the reabsorption of OTA (Km = 18.9

AM) [46]. In addition, as described above, OAT-K1 was

thought to mediate the pH-independent reabsorption of OTA

in the proximal straight tubule [8]. Furthermore, MRP2

mediated the efflux of OTA [25]. The relative contribution

of these OATs including oatp-1, OAT-K1, MRP2 and

hOAT4 to the reabsorption and the efflux of OTA on the

apical side of the proximal tubule should be elucidated in

future studies. In addition to these apical OATs, the role of

OAT-K2 [21], oatp-1 [22] and NPT1 [26] in the transport of

OTA should be further clarified.

The hOAT4-mediated OTA uptake was significantly

inhibited by the presence of probenecid, penicillin G,

aspartame, octaonate, citrinin, indomethacin and piroxicam.

These results are consistent with those of previous reports

that various substrates for OATs including PAH, probenecid,

piroxicam and octanoate inhibited peritubular OTA uptake

or OTA secretion in rabbit isolated proximal tubules [35–

39]. In addition, the results of the inhibition experiments

were in contrast to the previous results that these inhibitors,

except aspartame, exhibited significant inhibitory effects on

OTA uptake by basolateral OATs, hOAT1 and hOAT3 [27].

The rank order of Ki values of various inhibitors of hOAT4-

mediated OTA uptake was the same as that for hOAT3- but

not hOAT1-mediated OTA uptake [27]. The Ki values of

probenecid, piroxicam, octanoate and citrinin for hOAT4-

mediated OTA uptake were much higher than those for

hOAT1- and hOAT3-mediated OTA uptake except that of

citrinin for hOAT1-mediated OTA uptake (more than three-

fold) [27,47]. Thus, it was suggested that there exist some

differences in the affinity and the effects of various sub-

strates on OTA uptake between apical OAT (hOAT4) and

basolateral OATs (hOAT1 and hOAT3). As far as we know,

there is only one report describing the effects of these

substrates on the apical OTA transport, i.e., aspartame

exhibited no significant inhibitory effects on apical OTA

uptake in MDCK cells [37]. The discrepancy between the

result of this study and our result may be due to the OTA

uptake in MDCK cells mediated not only by OAT but also

by proton-dependent peptide transporter.

In humans, the steady-state maximum plasma concen-

tration of probenecid was reported to be 170 AM [40] and its

unbound fraction was 11.0% [41]. Thus, the steady-state

maximum concentration of unbound probenecid in the

plasma was estimated to be 18.7 AM. Since the therapeuti-

cally relevant concentration of a drug in the plasma are

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–75 73

thought to be within fivefold of the steady-state maximum

concentration of a drug in the plasma [48], the therapeuti-

cally relevant concentration of unbound probenecid in the

plasma is though to be within 93.5 AM. In addition, since an

unbound drug is filtered through glomerulus, the concen-

tration of unbound probenecid within the tubular fluid in the

proximal tubule may be within 93.5 AM. Thus, the Ki value

of probenecid for hOAT4-mediated OTA uptake (4.40 AM)

was much lower than the concentration of unbound probe-

necid within the tubular fluid in the proximal tubule (more

than threefold) [47]. Based on these, it was predicted that

probenecid within the tubular fluid could inhibit OTA

uptake in the apical side of the proximal tubule.

Piroxicam exhibits inhibitory effects on hOAT4-medi-

ated OTA reabsorption on the apical side of the proximal

tubule. However, piroxicam appears to have no clinical

significance. The steady-state plasma concentration of pir-

oxicam was reported to be 24.0 AM and its unbound fraction

was 0.02% [42]. The steady-state maximum concentration

of unbound piroxicam in the plasma is estimated to be

approximately 0.0048 AM. This concentration of piroxicam

is much lower than the Ki value for hOAT4-mediated OTA

uptake, 4.88 AM.

Since the addition of OTA to S2 hOAT4 culture resulted

in no significant effects on viability compared with mock, it

was suggested that hOAT4-mediated OTA uptake may not

significantly contribute to OTA-induced nephrotoxicity in

vivo. The results were in contrast to those in which addition

of OTA resulted in a significant decrease in the viability of

S2 cells stably expressing hOAT1 and hOAT3 (unpublished

observation). Previous studies have demonstrated that pir-

oxicam was shown to inhibit OTA-induced nephrotoxicity

[49]. In addition, piroxicam inhibited OTA uptake mediated

by hOAT1 and hOAT3 [27], and hOAT4. However, hOAT4

did not mediate OTA-induced nephrotoxicity. These results

suggest that piroxicam prevented OTA-induced nephrotox-

icity by inhibiting hOAT1 and hOAT3. On the other hand,

aspartame was also shown to inhibit OTA-induced neph-

rotoxicity [33,50]. In addition, aspartame inhibited OTA

uptake mediated by hOAT4, but not hOAT1 and hOAT3

[27]. Since hOAT4 did not mediate OTA-induced nephro-

toxicity, it was suggested that aspartame inhibited OTA

uptake by inhibiting some transporters other than hOAT1,

hOAT3 and hOAT4. However, there is another possibility

for the involvement of hOAT4 in the OTA-induced neph-

rotoxicity. Since the Km value of hOAT4 (22. 9 AM) is much

higher than those of hOAT1 (0.42 AM) and hOAT3 (0.75

AM) (more than threefold) [27,47], OTA taken up via

basolateral OATs does not move readily across the apical

membrane. This may lead to the buildup of a high concen-

tration of OTA, resulting in the selective damage of the

proximal tubular cells.

Citrinin is also a nephrotoxic mycotoxin, whose structure

is related to OTA. Citrinin has been shown to cause renal

dysfunction, such as glucosuria and proteinuria [51]. Since

citrinin inhibited hOAT4-mediated OTA uptake in a com-

petitive manner, it is possible that citrinin is taken up by

hOAT4 and accumulated OTA induces OTA-induced neph-

rotoxicity.

In conclusion, these results suggest that hOAT4 mediates

not only reabsorption but also the efflux of OTA in the

apical side of the proximal tubule. In addition, since hOAT4

was originally isolated from human placenta [19], it is

possible that hOAT4 in the human placenta plays some role

in the elimination of toxic substances including OTA from

the fetus.

Acknowledgements

This study was supported in part by Grants-in-Aid from

the Ministry of Education, Sports, Science and Technology

(nos. 11671048, 11694310 and 13671128), the Science

Research Promotion Fund of the Japan Private School

Promotion Foundation and Research on Health Sciences

Focusing on Drug Innovation from the Japan Health

Sciences Foundation.

References

[1] J. Harwig, T. Kuiper-Goodmann, P.M. Scott, M. Rechcigl, Handbook

of Foodborne Diseases of Biological Origin, CRC Press, Boca Raton,

FL (1983) 193–238.

[2] WHO, Environmental Health Criteria Geneva, 1990, pp. 105.

[3] I. Baudrimont, M. Murn, A.M. Betbeder, J. Guilcher, E.E. Creppy,

Toxicology 95 (1995) 147–154.

[4] T. Kuiper-Goodman, P.M. Scott, Biomed. Environ. Sci. 2 (1989)

179–248.

[5] P. Simon, J. Toxicol. 15 (1996) 239–249.

[6] M. Gekle, S. Silbernagl, J. Pharmacol. Exp. Ther. 276 (1993) 316–

321.

[7] M. Gekle, S. Silbernagl, Renal Physiol. Biochem. 17 (1994) 40–49.

[8] A. Dahlmann, W.H. Dantzler, S. Silbernagl, M. Gekle, J. Pharmacol.

Exp. Ther. 286 (1998) 157–162.

[9] P.P. Sokol, G. Ripich, P.D. Holohan, C.R. Ross, J. Pharmacol. Exp.

Ther. 246 (1988) 460–465.

[10] A.F. Stein, T.D. Phillips, L.F. Kubena, R.B. Harvey, J. Toxicol. En-

viron. Health 16 (1985) 593–605.

[11] M. Zingerle, S. Silbernagl, M. Gekle, J. Pharmacol. Exp. Ther. 280

(1997) 220–224.

[12] G. Schwerdt, M. Gekle, R. Freudinger, S. Mildenberger, S. Silbernagl,

Biochim. Biophys. Acta 1324 (1997) 191–199.

[13] J.B. Pritchard, D.S. Miller, Physiol. Rev. 73 (1993) 765–796.

[14] T. Sekine, N. Watanabe, M. Hosoyamada, Y. Kanai, H. Endou, J. Biol.

Chem. 272 (1997) 18526–18529.

[15] M. Hosoyamada, T. Sekine, Y. Kanai, H. Endou, Am. J. Physiol. 276

(1999) F122–F128.

[16] T. Sekine, S.H. Cha, M. Tsuda, N. Apiwattanakul, N. Nakajima, Y.

Endou, H. Endou, FEBS Lett. 429 (1998) 179–182.

[17] H. Kusuhara, T. Sekine, N. Utsunomiya-Tate, M. Tsuda, R. Kojima,

S.H. Cha, Y. Sugiyama, Y. Kanai, H. Endou, J. Biol. Chem. 274

(1999) 13675–13680.

[18] S.H. Cha, T. Sekine, J.I. Fukushima, Y. Kanai, Y. Kobayashi, T. Goya,

H. Endou, Mol. Pharmacol. 59 (2001) 1277–1286.

[19] S.H. Cha, T. Sekine, H. Kusuhara, E. Yu, Y.J. Kim, D.K. Kim,

Y. Sugiyama, Y. Kanai, H. Endou, J. Biol. Chem. 275 (2000)

4507–4512.

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–7574

[20] H. Saito, S. Masuda, K.I. Inui, J. Biol. Chem. 271 (1996) 20719–

20725.

[21] S. Masuda, K. Ibaramoto, A. Takeuchi, H. Saito, Y. Hashimoto, K.I.

Inui, Mol. Pharmacol. 55 (1999) 743–753.

[22] E. Jacquemin, B. Hagenbuch, B. Stieger, A.W. Wolkoff, P.J. Meier,

Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 133–137.

[23] B. Noe, B. Hagenbuch, B. Stieger, P.J. Meier, Proc. Natl. Acad. Sci.

U. S. A. 94 (1997) 10346–10350.

[24] T. Abe, M. Kakyo, B. Sakagami, T. Tokui, T. Nishio, M. Tanemoto,

H. Nomura, S.C. Hebert, S. Matsuno, H. Kondo, H. Yawo, J. Biol.

Chem. 273 (1998) 22395–22401.

[25] I. Leier, J. Hummel-Eisenbeiss, Y. Cui, D. Keppler, Kidney Int. 57

(2000) 1636–1642.

[26] H. Uchino, I. Tamai, K. Yamashita, Y. Minemoto, Y. Sai, H. Yabuuchi,

K. Miyamoto, E. Takeda, A. Tsuji, Biochem. Biophys. Res. Commun.

270 (2000) 254–259.

[27] K.Y. Jung, M. Takeda, D.K. Kim, A. Tojo, B.S. Yoo, S. Narikawa, M.

Hosoyamada, S.H. Cha, T. Sekine, H. Endou, Life Sci. 69 (2001)

2123–2135.

[28] A. Tojo, T. Sekine, N. Nakajima, M. Hosoyamada, Y. Kanai, K.

Kimura, H. Endou, J. Am. Soc. Nephrol. 10 (1999) 464–471.

[29] M. Hosoyamada, M. Obinata, M. Suzuki, H. Endou, Arch. Toxicol. 70

(1996) 284–292.

[30] M. Takeda, A. Tojo, T. Sekine, M. Hosoyamada, Y. Kanai, H. Endou,

Kidney Int. 56 (1999) 2128–2136.

[31] N. Apiwattanakul, T. Sekine, A. Chairoungdua, Y. Kanai, N. Nakaji-

ma, S. Sophasan, H. Endou, Mol. Pharmacol. 55 (1997) 847–854.

[32] T. Mossmann, J. Immunol. Methods 65 (1983) 55–63.

[33] E.E. Creppy, I. Baudrimont, A. Marie, J. Toxicol. Sci. 23 (1998) 165–

172.

[34] M. Tsuda, T. Sekine, M. Takeda, S.H. Cha, Y. Kanai, M. Kimura, H.

Endou, J. Pharmacol. Exp. Ther. 289 (1999) 1301–1305.

[35] C.E. Groves, M. Morales, S.H. Wright, J. Pharmacol. Exp. Ther. 284

(1998) 943–948.

[36] L.P. Sullivan, J.J. Grantham, J. Am. Soc. Nephrol. 2 (1992) 1192–

1200.

[37] G. Schwerdt, R. Freudinger, S. Silbernagl, M. Gekle, Toxicology 16

(1998) 193–202.

[38] J.R. Welborn, C.E. Groves, S.H. Wright, J. Am. Soc. Nephrol. 9

(1998) 1973–1982.

[39] C.E. Groves, G. Nowak, M. Morales, J. Am. Soc. Nephrol. 10 (1999)

13–20.

[40] D.W. Nierenberg, J. Pharmacol. Exp. Ther. 226 (1983) 1–6.

[41] P.G. Dayton, T.F. Yu, W. Chen, L. Berger, I.A. West, A.B. Gutman,

J. Pharmacol. Exp. Ther. 140 (1963) 278–286.

[42] Interview form, Pfizer Co., Ltd.

[43] F.S. Chu, Arch. Biochem. Biophys. 147 (1971) 359–366.

[44] S. Hagelberg, K. Hult, R. Fuchs, J. Appl. Toxicol. 9 (1989) 91–96.

[45] B. Vasquez, F. Ishibashi, B.V. Howard, In Vitro 18 (1982) 643–649.

[46] M. Kontaxi, U. Eckhardt, B. Hagerbuch, B. Stieger, P.J. Meier, E.

Petzinger, J. Pharmacol. Exp. Ther. 279 (1996) 1507–1513.

[47] L. Zhang, M.E. Schaner, K.M. Giacomini, J. Pharmacol. Exp. Ther.

286 (1998) 354–361.

[48] L. Zhang, W. Gorset, C.B. Washington, T.F. Blaschke, D.L. Kroetz,

K.M. Giacomini, Drug Metab. Dispos. 28 (2000) 329–334.

[49] E.E. Creppy, I. Baudrimont, A.M. Betbeder, Toxicol. Lett. 82 (1995)

869–877.

[50] I. Baudrimont, A.M. Betbeder, E.E. Creppy, Arch. Toxicol. 71 (1997)

290–298.

[51] R.D. Phillips, A.W. Hayes, W.O. Berndt, W.L. Williams, Toxicology

16 (1980) 123–137.

E. Babu et al. / Biochimica et Biophysica Acta 1590 (2002) 64–75 75