Original paper 451

C3435T polymorphism in the MDR1 gene affects the

enterocyte expression level of CYP3A4 rather than Pgp in
recipients of living-donor liver transplantation
Maki Gotoa , Satohiro Masudaa , Hideyuki Saitoa , Shinji Uemotob ,
Tetsuya Kiuchib , Koichi Tanakab and Ken-ichi Inuia
The bioavailability of structurally unrelated drugs is limited
by active secretion via the multidrug resistance gene
(MDR1) product P-glycoprotein (Pgp) from enterocyte into
lumen as well as intestinal metabolism by cytochrome
P450 IIIA4 (CYP3A4). In the present study, we analyzed
whether genetic polymorphism of the MDR1 had some
influence on the intestinal expression levels of Pgp and
CYP3A4 and the tacrolimus concentration/dose ratio over
the first postoperative days in recipients of living-donor
liver transplantation (LDLT). Genotyping assays were
performed for the major 10 polymorphisms in the MDR1
gene by the polymerase chain reactionrestriction enzyme
length polymorphism method. The allele frequencies of
variations at five positions were almost comparable with
those in the former studies in Caucasians and Japanese,
but there was no variation at the other five positions.
Although no polymorphism correlated with the intestinal
expression of MDR1 mRNA or the tacrolimus
concentration/dose ratio in the LDLT recipients, the
C3435T polymorphism significantly affected the intestinal
expression level of CYP3A4 mRNA as follows; 3435C/
C>3435C/T (P < 0.05 vs. 3435C/C)>3435T/T (P < 0.01 vs.

Introduction
The multidrug resistance (MDR) gene product P-glycoprotein (Pgp) is expressed in the plasma membrane in
normal tissue as well as several types of tumor cells and
acts as an ATP-driven efflux pump [1]. Normally, Pgp
is expressed in the brain, liver, kidney and small
intestine and contributes to drug pharmacokinetics such
as absorption, distribution, and excretion [1,2]. Therefore, the level of protein expression and the activity of
Pgp in these tissues will affect the pharmacokinetics of
drugs, that is, the drug efficacy and adverse effects. In
the small intestine, Pgp is localized at the brush-border
membranes and mediates extrusion of structurally unrelated drugs including the immunosuppressants tacrolimus and cyclosporin [3], as a possible mechanism of
inter-individual variation in the bioavailability of these
drugs.
Tacrolimus and cyclosporin are widely used to suppress
the rejection of graft organs after transplantation. However, the bioavailability of these drugs varies widely
0960314X & 2002 Lippincott Williams & Wilkins

3435C/C). Therefore, the identified polymorphisms

including C3435T in the MDR1 gene were indicated to have
no influence on the intestinal expression level of Pgp or
the tacrolimus concentration/dose ratio in the recipients of
LDLT. On the other hand, the C3435T polymorphism of
MDR1 was suggested to correlate with the enterocyte
expression of CYP3A4 rather than Pgp linking unknown
genetic variation in CYP3A4 gene. Pharmacogenetics
12:451457 & 2002 Lippincott Williams & Wilkins

Pharmacogenetics 2002, 12:451457

Keywords: SNP, metabolic enzyme, tacrolimus, transporter, small intestine
a

Department of Pharmacy, Faculty of Medicine, Kyoto University Hospital, Kyoto,

Japan and b Department of Transplantation and Immunology, Graduate School of
Medicine, Kyoto University, Kyoto, Japan
Correspondence to Professor Ken-ichi Inui, PhD, Department of Pharmacy, Kyoto
University Hospital, Sakyo-ku, Kyoto 606-8507, Japan
Tel: +81 75-751-3577; fax: +81 75-751-4207;
e-mail: inui@kuhp.kyoto-u.ac.jp
Received 7 January 2002
Accepted 8 May 2002

among individuals, and the monitoring of the blood

concentration is necessary to avoid adverse effects such
as renal and neuronal toxicities [4]. Pgp and the metabolic enzyme cytochrome P450 3A (CYP3A4) expressed
in the small intestine and liver are considered to be
factors, which influence the bioavailability of tacrolimus, respectively [5]. We have reported that the
tacrolimus concentration/dose ratio was closely correlated to the mucosal expression level of MDR1, but not
CYP3A4, in the recipients of living-donor liver transplantation (LDLT) and in a recipient of living-donor
small bowel transplantation (LDSBT) [6,7]. Therefore,
the expression level of MDR1 in the intestine is
suggested to be a pharmacokinetic factor in the variation in bioavailability of orally administered tacrolimus.
Recently, a genotype (C3435T) in the MDR1 mRNA
was identified as a factor responsible for the intestinal
expression level of Pgp in the Caucasians [8]. The
intestinal expression level of Pgp was higher for the
3435C/C allele than 3435T/T allele, and the area under

452 Pharmacogenetics 2002, Vol 12 No 6

the plasma concentration curve (AUC) value of orally

administered digoxin was higher for the 3435T/T allele
than 3435C/C allele [8]. However, Kim et al. [9] reported that fexofenadine AUC values during the first
4 h after oral administration in European-American and
African-American subjects were significantly lower in
those with the 3435T/T allele than with the 3435C/C
allele. Therefore, the reason for the discrepancy in the
pharmacological effects of the C3435T single nucleotide polymorphism (SNP) in the MDR1 mRNA between these two studies should be clarified. In
addition, it should be determined whether C3435T
SNP and/or other genetic variations have a significant
effect on the intestinal expression and/or transport
activity of Pgp or another absorptive barrier such as
CYP3A4.
In the present study, we assessed 10 genotypes including C3435T of the MDR1 gene in recipients of LDLT
by the polymerase chain reactionrestriction enzyme
length polymorphism (PCRRFLP) method. Thereafter, the influences of each SNP were evaluated by
comparing with the intestinal expression levels of
MDR1 mRNA and CYP3A4 mRNA and the tacrolimus
concentration/dose ratio during the first seven postoperative days.

(Takara, Shiga, Japan), according to the following

profile; 94 8C for 30 s, 60 8C for 30 s, and 72 8C for 30 s,
35 cycles, followed by a single additional 10-min
extension at 72 8C. The PCR products were digested
with a restriction enzyme according to the condition of
manufacturers instructions and separated on 3.5%
agarose gel.
Phenotyping of MDR1 polymorphisms

The mRNA expression level of MDR1 or CYP3A4 in

the mucosal specimens was quantified by competitive
PCR as described previously [7]. In this study, phenotyping by comparing the SNPs in the MDR1 gene with
the expression level of MDR1 or CYP3A4 mRNA was
performed in all 69 patients. The relationship between
the MDR1 SNPs and the tacrolimus concentration/dose
ratio was examined in 46 patients who were not administered either tacrolimus intravenously or drugs that
might affect MDR1 or CYP3A4 expression or activity.
Statistical analysis

Logarithmic transformations of the mRNA expression

levels of MDR1 and CYP3A4 were performed to
improve normality before the statistical analysis. The
statistical differences between various parameters were
determined with ANOVA (Sheffes test).

Materials and methods

Results

Subjects

Frequency of MDR1 variants in LDLT patients

Sixty-nine recipients of LDLT at Kyoto University

Hospital between 6 November 1998 and 10 May 2000
were enrolled in this study after providing the written
informed consent. The genetic analysis was performed
in accordance with the Declaration of Helsinki and its
amendments, and was approved by the Ethics Committee of Kyoto University.

The genotype of MDR1 was assessed at 10 positions in

69 LDLT (28 male and 41 female) recipients. The age
of the recipients ranged from 0.6 to 59.6 years and the
primary diseases were biliary atresia (n 29), cirrhosis
(n 11), primary sclerosing cholangitis (n 4), primary
biliary cirrhosis (n 4), fulminant hepatic failure
(n 3), Alagille syndrome (n 2), and others (n 16).
Figure 1 shows distribution histograms of logarithmically transformed mucosal mRNA expression levels
of MDR1 and CYP3A4. The average mRNA expression
levels of MDR1 and CYP3A4 were 0.35  0.05 (0.02
5.75) and 1.65  0.60 (0.01182.5) amol/g total RNA,
respectively. Table 1 shows the frequency of SNPs in
the 69 LDLT patients. In the present study, the
variants in exon 2 G-1A, A61G (Asn21 Asp) in exon 2,
T307C (Phe103 Leu) in exon 5, G1199A (Ser400 Asn) in
exon 11 and C+44T in intron 12 were not observed.
Variants leading to an amino acid exchange were observed only at a mutation in exon 21 G2677T/A
(Ala893 Ser or Ala893 Thr). Allele frequencies of G2677T/
A variants were 43.5% for the T allele and 12.3% for
the A allele, respectively; only one carrier homozygous
for 2677A/A was observed. In the exonic variants which
have no influence on the amino acid sequence, C1236T
in exon 12 and C3435T in exon 26 were also observed
and the frequency of the T allele was 65.2% and
44.2%, respectively. The most frequent intronic variants were C/T in exon 6+139 (allelic frequency,

Genotyping of the MDR1 gene

Genomic DNA from a frozen mucosal homogenate in a

guanidinium thiocyanate solution was isolated with a
MagNAPure LC DNA isolation kit (Roche, Mannheim,
Germany), and the extraction of total RNA and reverse
transcription were performed as described previously
[6]. The genomic DNA and reverse-transcribed cDNA
were used for genotyping by PCRRFLP. The specific
primer sequences and restriction enzymes used were as
described by Cascorbi et al. [10]. For exon 2, A61G
mutation mismatch primers were created, which introduced a Bsp45 I site in the presence of a mutation. The
primer sequences were as follows; forward, 59AGGAGCAAAGAAGAAGAACTTTTTTAAAACTGT
A-39; reverse, 59-ACCTGGTCATGTCTTCCTCC-39.
The restriction enzyme-digested fragments were 35 bp
and 280 bp for the wild-type allele, and 315 bp for the
variant allele. The PCR conditions were as follow; after
denaturing at 94 8C for 3 min, the PCR was performed
with 0.25 M of each primer and Taq DNA polymerase

MDR1 C3435T and intestinal CYP3A4 level Goto et al. 453

63.2%) and T/A in exon 1776 (31.7%). A linkage was

observed between exon 6+139 C/T and exon 12
C1236T; 92% of samples homozygous for exon 12 SNP
(T/T) have the genotype T/T in exon 6+139. Other
incomplete linkage disequilibriums were observed between G2677T and C3435T.

Fig. 1

18 A

20 B

16

18

14

16
14
Frequency

Frequency

12
10
8
6

Effect of MDR1 SNPs on MDR1 and CYP3A4 expression

We analyzed the relationships between each SNP of

the MDR1 gene and the mRNA expression level of
MDR1 as well as CYP3A4. As shown in Table 2, there
was no significant correlation between the SNPs and
mRNA expression level of MDR1. However, the variation in exon 26 (C3435T) had a significant correlation
with the CYP3A4 mRNA expression level; the expression level of intestinal CYP3A4 mRNA was 8-fold lower
in those with the genotype 3435T/T than the genotype
3435C/C (P , 0.01), while the level was 4-fold lower in
the subjects with the heterozygous genotype 3435C/T
than with 3435C/C (P , 0.05). In addition, there was
no relationship between the primary diseases of recipi-

2
0
102

12
10

2
101
1
MDR1 mRNA
(amol/g total RNA)

10

0
2
3
103 102 101 1 10 10 10
CYP3A4 mRNA
(amol/g total RNA)

Distribution histograms of mRNA expression levels of MDR1 and

CYP3A4 in 69 recipients of LDLT. Distribution of intestinal mRNA
expression levels of MDR1 (A) and CYP3A4 (B), which were
logarithmically transformed to improve normality, in 69 recipients of
LDLT.

Table 1

MDR1 genetic variants in LDLT patients (n

Frequency
(%)

Caucasiansa

100.0
0.0

G/G
G/A
A/A

100.0
0.0
0.0

82.8
16.4
0.8

FokI

21 Asn
21 Asp

100.0
0.0

A/A
G/A
G/G

100.0
0.0
0.0

78.9
19.8
1.2

Bsr GI

T
C

103 Phe
103 Leu

100.0
0.0

T/T
C/T
C/C

100.0
0.0
0.0

100.0
0.0
0.0

BanII

C
T

36.8
63.2

C/C
C/T
T/T

10.3
52.9
36.8

69.4
46.7
13.8

SspI

G
A

400 Ser
400 Asn

100.0
0.0

G/G
G/A
A/A

100.0
0.0
0.0

89.2
10.5
0.3

Eco57 I

C
T

Wobble

34.8
65.2

C/C
C/T
T/T

10.1
49.3
40.6

34.8
48.4
16.8

HaeIII

C
T

100.0
0.0

C/C
C/T
T/T

100.0
0.0
0.0

90.5
9.3
0.2

Tsp45I

T
A

68.4
31.7

T/T
T/A
A/A

42.6
51.5
5.9

28.9
49.7
21.3

ApoI

G
T
A

893 Ala
893 Ser
893 Thr

45.4
43.5
12.3

G/G
G/A
G/T
T/A
T/T
A/A

22.9
5.8
39.1
15.9
15.9
1.4

32.0
2.1
47.0
1.6
17.3
0.0

C
T

Wobble

55.8
44.2

C/C
C/T
T/T

30.4
50.7
18.8

21.2
49.7
29.1

Position

Allele

Effect

Exon 2

Exon 21

G
A

Initiation of
translation?

A
G

Exon 5

Intron 6

Exon 11

Exon 12

Intron 12

Intron 16

Exon 21

Exon 26

cDNA 61

cDNA 307

Exon 6+139

cDNA 1199

cDNA 1236

Exon 12+44

Exon 1776

cDNA 2677

cDNA 3435

These values were from Cascorbi et al. [10].

Allele
frequency

68 for intron polymorphism)

Genotype

Location

Exon 2

69 for exon, n

Restriction
enzyme

Bsr I
BanI

MboI

454 Pharmacogenetics 2002, Vol 12 No 6

these patients was inversely related to the MDR1

mRNA expression level.

Effect of SNPs on MDR1 and CYP3A4 mRNA expression

level in the LDLT patients

Table 2

Location

SNP

Intron 6

C+139T

MDR1 (amol/g CYP3A4 (amol/

total RNA)
g total RNA)

Genotype

C/C
C/T
T/T

7
36
25

0.67  0.22
0.33  0.07
0.32  0.08

2.7  1.0
2.1  0.6
1.0  0.4

Exon 12

C1236T

C/C
C/T
T/T

7
34
28

0.67  0.11
0.32  0.07
0.35  0.09

2.7  1.2
1.9  0.5
1.1  0.4

Intron 16

T-76A

T/T
T/A
A/A

29
35
4

0.39  0.09
0.30  0.06
0.82  0.22

1.1  0.4
1.9  0.2
5.2  3.0

Exon 21

G2677T/A

G/G
G/A
G/T
T/A
T/T
A/A

15
4
27
11
11
1

0.47  0.12
0.98  0.75
0.25  0.07
0.31  0.11
0.48  0.15
0.23

3.1  1.0
10.6  8.0
1.5  0.5
2.0  1.4
0.4  0.2
0.8

Exon 26

C3435T

C/C
C/T
T/T

21
35
13

0.58  0.20
0.25  0.05
0.40  0.10

4.7  1.7
1.3  0.4
0.6  0.3

Each data represents the mean  SE of the indicated patients numbers (n).
 P , 0.05, CC vs. CT,  P , 0.01, CC vs. TT.

ents and the mRNA expression level of MDR1 or

MDR1 SNPs.
MDR1 SNPs and tacrolimus concentration/dose ratio

Next, to examine whether the MDR1 SNPs correlate

with the activity of Pgp as an absorptive barrier, we
analyzed the correlation between the MDR1 SNPs and
the tacrolimus concentration/dose ratio. As shown in
Table 3, no significant correlation was observed between the MDR1 SNPs and the tacrolimus concentration/dose ratio during the first seven postoperative days.
However, the tacrolimus concentration/dose ratio in

Three-point substitutions of MDR1 in the open reading

frame and MDR1 mRNA expression

We analyzed the correlation between the haplotype for

the three-point substitutions in the open reading frame,
the MDR1 and CYP3A4 mRNA expression levels, and
the tacrolimus concentration/dose ratio at positions
1236, 2677 and 3435. As shown in Table 4, the subjects
homozygous for the CC/GG/CC allele seemed to have
higher expression levels of MDR1. In contrast, haplotypes which have a homozygous mutation at all three
points (TT/TT/TT genotype) tended to have a reduced expression of MDR1. Because the heterozygous
genotype (CT/GT/CT) showed a lower expression
level than both homozygous (CC/GG/CC and TT/TT/
TT) genotypes, no correlation between the three-point
substitution linkages and MDR1 mRNA expression
levels were observed. In addition, one subject with the
A allele at 2677 and the wild-type allele at the other
two positions had a slightly lower expression level of
MDR1 and CYP3A4. The mRNA expression level of
MDR1 mRNA in the patients homozygous for the
wild-type allele at all three points (CC/GG/CC) tended
to be higher than that in the patients with some
variations elsewhere.

Discussion
Pharmacogenetic evaluations of several genes related to
drug pharmacokinetics should be efficient to personalize drug therapy by optimizing the dosage. For several
metabolic enzymes, a relationship between genetic
polymorphisms and clinical effects has been reported,
and studies of this kind help to improve drug efficacies

Effect of SNPs on MDR1 and CYP3A4 mRNA expression and tacrolimus

concentration/dose ratio in LDLT patients (n 46)

Table 3

Location

SNP

Intron 6

CYP3A4
MDR1 (amol/g (amol/g total
total RNA)
RNA)

Concentration/
dose ratio

Genotype

C+139T

C/C
C/T
T/T

5
23
18

0.76  0.22
0.35  0.08
0.34  0.11

2.5  1.5
1.6  0.4
1.0  0.3

119.2  35.2
163.6  17.2
171.2  23.5

Exon 12

C1236T

C/C
C/T
T/T

5
23
18

0.76  0.22
0.34  0.07
0.36  0.10

2.5  1.0
1.8  0.3
0.8  0.3

119.2  35.2
168.6  16.7
164.8  24.6

Intron 16

T-76A

T/T
T/A
A/A

19
24
3

0.39  0.12
0.34  0.07
0.76  0.40

1.0  0.4
1.6  0.4
3.8  2.0

166.1  22.9
160.5  16.6
144.3  57.7

Exon 21

G2677T/A

G/G
G/A
G/T
T/A
T/T

11
3
18
5
9

0.38  0.10
0.55  0.30
0.28  0.09
0.71  0.35
0.44  0.15

1.7  0.4
4.1  2.0
1.6  0.6
1.0  0.7
0.7  0.3

143.2  22.4
221.4  55.5
175.9  23.5
126.8  43.1
155.6  26.2

Exon 26

C3435T

C/C
C/T
T/T

15
22
9

0.48  0.15
0.30  0.09
0.44  0.18

3.0  0.9
1.1  0.4
0.7  0.2

152.7  22.4
170.4  20.6
155.6  26.2

Each data item represents the mean  SE of the indicated patients numbers (n).  P , 0.05, CC vs. TT.

MDR1 C3435T and intestinal CYP3A4 level Goto et al. 455

Three-point substitutions of MDR1 in the open reading frame and MDR1

expression

Table 4

C1236T
CT
TT
CC
CT
TT
CC
CT
CC
CT
CT
TT
CT
TT
TT
CT

G2677T/A

C3435T

Frequency
(%)

MDR1 (amol/g
total RNA)

CYP3A4 (amol/g
total RNA)

GT
GT
GG
GG
GG
GA
GA
AA
TA
GT
GT
GG
TT
TA
GT

CC
CC
CC
CC
CC
CC
CC
CC
CT
CT
CT
CT
TT
TT
TT

1
2
4
6
3
2
2
1
10
12
11
2
11
1
1

1.5
2.9
5.8
8.7
4.4
2.9
2.9
1.5
14.5
17.4
15.9
2.9
15.9
1.5
1.5

0.04
3.18
0.89  0.07
0.34  0.16
0.51  0.40
0.66
1.47
0.23
0.34  0.13
0.27  0.10
0.18  0.06
0.33
0.47  0.13
0.11
0.21

1.4
27.6
2.8  1.4
3.3  1.4
4.8  4.0
4.7
14.2
0.8
1.8  1.4
1.1  0.6
1.1  0.1
1.4
0.4  0.2
6.1
4.8

Each data item represents the mean  SE of the indicated patients numbers (n).

and prevent adverse effects [11]. It is now well

acknowledged that intestinal MDR1 as well as CYP3A4
acts as an absorptive barrier limiting the bioavailability
of various drugs, including tacrolimus [3,5]. Therefore,
genetic polymorphisms related to the intestinal expression level and/or functional characteristics of MDR1
and CYP3A4 would influence the individual dosage
regimen of tacrolimus in LDLT recipients. Recently, a
silent or wobble variation of C3435T in the MDR1
mRNA was suggested to be closely related with the
intestinal expression level of Pgp [8]. In the present
study, we analyzed whether 10 SNPs of MDR1 including the C3435T have some influence on the relationship between the intestinal MDR1 expression level
and the tacrolimus concentration/dose ratio in LDLT
patients.
In a genetic analysis in 461 Caucasians, the allele
frequencies of variations at four positions of exon 21,
cDNA 61, cDNA 1199 and exon12+44 were 9%, 11.2%,
5.5% and 4.9%, respectively, and the frequency of
variation at cDNA 307 was 0% [10]. In the 69 Japanese
LDLT recipients, no variation was found at all these
five positions including cDNA 307 (Table 1). Because
the size of the population in the present study was only
69 and the allele frequencies of these SNPs were low
even in a large population of 461 individuals, genetic
analysis in a larger number of subjects is needed to
compare the allele features in Japanese with those in
Caucasians. In the present study, we found a moderate
frequency of the A allele at cDNA 2677 in Japanese
(Table 1), and the allele frequency of G2677A was
comparable with that in the report of Tanabe et al. [12].
Although the Pgp expression levels in the placenta
tended to be higher in those with the G allele than
with the T or A allele [12], we could not find a
correlation like that in the small intestine. Therefore,
the influence of G2677T/A variation on the Pgp expres-

sion level would differ between the placenta and small

intestine. Considering that the G2677A allele frequency
was only 1.9% in 461 Caucasians, the Ala893 Thr variation derived from G2677A would be more frequent in
Japanese than in Caucasians. Although Kim et al. [9]
demonstrated that the Pgp-mediated extrusive activity
of digoxin in the MDR1-Ser893 (2677T allele) transfectant was 2-fold higher than that in the MDR1-Ala893
(2677G allele) transfectant, MDR1-Thr893 (2677A
allele)-mediated digoxin activity was not examined.
The functional significance of MDR1 2677A (MDR1Thr893 ) should be elucidated by examining the transport activities among the three types of MDR1 cDNA
2677 allele.
We reported that the intestinal expression levels of Pgp
rather than CYP3A4 correlated in an inverse proportion
to the tacrolimus trough levels in LDLT recipients and
in a LDSBT recipient [6,7]. Lown et al. [13] reported
that the hepatic CYP3A4 and intestinal Pgp were two
major factors that predict the oral clearance or peak
concentration of cyclosporin, whereas the intestinal
CYP3A4 did not predict any of the cyclosporin pharmacokinetics in 19 recipients of kidney transplantation.
Despite of some reports suggesting the significant
metabolic activity of intestinal CYP3A4 for drug bioavailability [3], these findings suggested that the intestinal absorption rates of tacrolimus and cyclosporin
could be predicted from the expression and/or transport
activity of intestinal Pgp rather than CYP3A4. Hoffmeyer et al. [8] reported that the intestinal expression
level of Pgp and the AUC for orally administered
digoxin in subjects with the 3435C/C allele were higher
and lower than those with the 3435C/T or 3435T/T
allele, respectively. In contrast, the AUC for orally
administered digoxin was higher in Japanese with the
3435C/C allele than 3435T/T allele [14]. On the other
hand, the MDR1 C3435T polymorphism did not influ-

456 Pharmacogenetics 2002, Vol 12 No 6

ence the cyclosporin oral dosage or trough levels in 123

stable Caucasian renal transplant recipients (over 6
months after transplantation) [15]. In addition, the 9%
increase in digoxin AUC at time zero to 4 h induced by
grapefruit juice ingestion in 12 healthy volunteers
(P 0.01) was not affected by the MDR1 C3435T
polymorphism [16]. In the present study, we could not
detect a significant effect of that polymorphism on the
intestinal expression level of Pgp and the tacrolimus
concentration/dose ratio in the LDLT recipient (Table
3). These findings suggested that the influences of
MDR1 C3435T polymorphism on the intestinal expression and/or functional characteristics of Pgp were
relatively weak, and/or relate to some unknown phenotype(s) different from the absorption of tacrolimus with
or without several genetic linkages.
Because tacrolimus and cyclosporin were metabolized
from the circulation via CYP3A4-mediated metabolism
in the liver, the variations of hepatic CYP3A4 would
correlate with the systemic clearance of tacrolimus
and cyclosporin. Min et al. [17] reported that the
systemic clearance and bioavailability of cyclosporin
were 4.3  0.9 ml/min/kg and 32.8  6.6% in AfricanAmericans and 3.7  0.5 ml/min/kg and 39.3  7.1% in
European-Americans, with a significant racial interaction (P 0.0001 and P 0.0489, respectively). In
addition, it was demonstrated in the recipients of renal
transplantation that black patients required about 2-fold
higher dose of tacrolimus in comparison with that in
white/Asian patients to achieve standard trough levels
of tacrolimus (1015 ng/ml) [18]. The racial interaction
for the hepatic clearance of cyclosporin and tacrolimus
might be explained by an effect of genetic background
on the hepatic and intestinal CYP3A4. Surprisingly,
we have found that the intestinal expression level
of CYP3A4 was correlated with the MDR1 C3435T
polymorphism, i.e., 3435C/C.3435C/T (P , 0.05 vs.
3435C/C).3435T/T (P , 0.01 vs. 3435C/C; Table 2).
It is of interest to consider what would happen to the
low expression level of enterocyte CYP3A4 in patients
with the 3435T/T variation. Schaeffeler et al. [19]
demonstrated a significantly higher frequency of the
3435C/C genotype in West Africans (83%) and AfricanAmericans (61%) than in European-Americans (26%)
and in Japanese (34%) (P , 0.0001). These findings led
us to hypothesize that the MDR1 C3435T polymorphism would be linked with an unknown genetic variation
of the CYP3A4 gene, and thereby, the higher systemic
clearance and lower bioavailability of cyclosporin in
Africans than European-Americans might be derived
from a difference of MDR1 3435C/C frequency, e.g. the
expression level of CYP3A4 in the liver and intestine.
In the present study, the hepatic expression level of
CYP3A4 could not be evaluated because the graft liver
was of a different genetic background from the patient.
If the expression level of CYP3A4 in the small intestine

correlates to that in the liver, the enterocyte expression

level of CYP3A4 obtained using a biopsy specimen
would be useful to predict the hepatic expression level
of CYP3A4. Therefore, the in vivo effect of MDR1
C3435T on the intestinal and hepatic CYP3A4 should
be clarified by evaluating the pharmacokinetics of
CYP3A4 substrate such as midazolam bioavailability
and erythromycin breath test, respectively. Hence, the
unknown polymorphism(s) affecting the intestinal and/
or hepatic expression levels of CYP3A4 may be found
by comparing the nucleotide sequence of the CYP3A4
gene in the subjects with the C and T allele of the
MDR1 C3435T polymorphism. To our knowledge, this
is the first report suggesting a genetic linkage between
the MDR1 and CYP3A4 genes.
In the present study, the genetic polymorphisms of
MDR1 were not suggested to be major determinants of
the inter-individual variation in the tacrolimus concentration/dose ratio in recipients of LDLT. Because we
only investigated LDLT recipients who had severe
hepatic disease at surgery, the possibility of selection
bias must be considered. However, allele frequencies
in these recipients for MDR1 were consistent with
published allele frequencies for healthy subjects, as
described above. Therefore, the present data for the
intestinal expression levels of MDR1 and CYP3A4 and
tacrolimus concentration/dose ratio could have varied
with liver function at surgery as well as with MDR1
polymorphisms. In the case of recipients of LDLT
who received tacrolimus therapy, other factors may
explain the relation between enterocyte Pgp expression and the MDR1 genetic polymorphism, such as
physiological status, concomitant disease and graft liver
mass. However, there was a large inter-individual
variation in the enterocyte MDR1 mRNA expression
in these patients, between 0.01 and 5.45 amol/g total
RNA (Fig. 1). Considering these findings, a strong
genetic background associated with the intestinal
expression level of MDR1 even in the LDLT recipients with hepatic dysfunction at the surgery should
exist.
In summary, we performed genotyping of the MDR1
gene in 69 LDLT recipients and analyses of the
relationship between MDR1 expression and the tacrolimus concentration/dose ratio. No genetic variant
showed a significant effect of MDR1 expression or
tacrolimus concentration/dose ratio. However, the exon
26 (C3435T) polymorphism of the MDR1 gene correlated with the intestinal expression level of CYP3A4.
Although the correlation between the MDR1 C3435T
polymorphism and the hepatic expression levels of
CYP3A4 should be examined in future, the C3435T
polymorphism assessment could be used for predicting
a potential CYP3A4 contents in the liver and a maintenance dosage regimen of tacrolimus.

MDR1 C3435T and intestinal CYP3A4 level Goto et al. 457

Acknowledgements
This work was supported in part by a grant-in-aid from
Japan Health Sciences Foundation, and by a grant-inaid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

References
1

10

11
12

13

14

15

16

17

18
19

Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, Gottesman

MM. Biochemical, cellular, and pharmacological aspects of the multidrug
transporter. Annu Rev Pharmacol Toxicol 1999; 39:361398.
Cordon-Cardo C, OBrien JP, Boccia J, Casals D, Bertino JR, Melamed
MR. Expression of multidrug resistance gene product (P-glycoprotein)
in human normal and tumor tissues. J Histochem Cytochem 1990;
38:12771287.
Zhang Y, Benet LZ. The gut as a barrier to drug absorption: combined
role of cytochrome P450 3A and P-glycoprotein. Clin Pharmacokinet
2001; 40:159168.
Venkataramanan R, Swaminathan A, Prasad T, Jain A, Zuckerman S,
Warty V, et al. Clinical pharmacokinetics of tacrolimus. Clin Pharmacokinet 1995; 29:404430.
Hebert MF. Contributions of hepatic and intestinal metabolism and Pglycoprotein to cyclosporine and tacrolimus oral drug delivery. Adv Drug
Deliv Rev 1997; 27:201214.
Hashida T, Masuda S, Uemoto S, Saito H, Tanaka K, Inui K. Pharmacokinetic and prognostic of intestinal MDR1 expression in recipients of livingdonor liver transplantation. Clin Pharmacol Ther 2001; 69:308316.
Masuda S, Uemoto S, Hashida T, Inomata Y, Tanaka K, Inui K. Effect of
intestinal P-glycoprotein on daily tacrolimus trough level in a living-donor
small bowel recipient. Clin Pharmacol Ther 2000; 68:98103.
Hoffmeyer S, Burk O, Richter OV, Arnold HP, Brockmoller J, Johne A,
Cascorb I, Gerloff T, Roots I, Eichebaum M, Brinkmann U. Functional
polymorphisms of the human multidrug-resistance gene: multiple
sequence variations and correlation of one allele with P-glycoprotein
expression and activity in vivo. Proc Natl Acad Sci USA 2000;
97:34733478.
Kim RB, Leake BF, Choo Ef, Dresser GK, Kubba SV, Schwartz UI, et al.
Identification of functionally variant MDR1 alleles among European Americans and African Americans. Clin Pharmacol Ther 2001; 70:189199.
Cascorbi I, Gerloff T, Johne A, Meisel C, Hoffmeyer S, Schwab M, et al.
Frequency of single nucleotide polymorphisms in the P-glycoprotein drug
transporter MDR1 gene in white subjects. Clin Pharmacol Ther 2001;
69:169174.
Evans WE, Relling MV. Pharmacogenomics: translating functional
genomics into rational therapeutics. Science 1999; 286:487491.
Tanabe M, Ieiri I, Nagata N, Inoue K, Ito S, Kanamori Y, et al. Expression
of P-glycoprotein in human placenta: Relation to genomic polymorphism
of the multidrug resistance (MDR-1) gene. J Exp Pharmacol Ther 2001;
297:11371143.
Lown KS, Mayo RR, Leichtman AB, Hsiao HL, Turgeon DK, Schmiedlin
RP, et al. Role of intestinal P-glycoprotein (mdr1) in interpatient variation
in the oral bioavailability of cyclosporine. Clin Pharmacol Ther 1997;
62:248260.
Sakaeda T, Nakamura T, Horinouchi M, Kakumoto M, Ohmoto N, Sakai
M, et al. MDR1 genotype-related pharmacokinetics of digoxin after single
oral administration in healthy Japanese subjects. Pharm Res 2001;
18:14001404.
von Ahsen N, Richter M, Grupp C, Ringe B, Oellerich M, Armstrong VW.
No influence of the MDR-1 C3435T polymorphism or a CYP3A4
promoter polymorphism (CYP3A4-V allele) on dose-adjusted cyclosporin
A trough concentrations or rejection incidence in stable renal transplant
recipients. Clin Chem 2001; 47:10481052.
Becquemont L, Verstuyft C, Kerb R, Brinkmann U, Lebot M, Jaillon P,
et al. Effect of grapefruit juice on digoxin pharmacokinetics in humans.
Clin Pharmacol Ther 2001; 70:311316.
Min DI, Lee M, Ku YM, Flanigan M. Gender-dependent racial difference in
disposition of cyclosporine among healthy African American and white
volunteers. Clin Pharmacol Ther 2000; 68:478486.
Andrews PA, Sen M, Chang RWS. Racial variation in dosage requirements of tacrolimus. Lancet 1996; 348:1446.
Schaeffeler E, Eichelbaum M, Brinkmann U, Penger A, Asante-Poku S,
Zanger UM, et al. Frequency of C3435T polymorphism of MDR1 gene
in African people. Lancet 2001; 358:383384.