Báo cáo khoa học: ABCG transporters and disease pdf

MINIREVIEW
ABCG transporters and disease
Owen M. Woodward
1
, Anna Ko
¨
ttgen
2,3
and Michael Ko
¨
ttgen
2,4
1 Department of Physiology, Johns Hopkins University, School of Medicine, Baltimore, MD, USA
2 Renal Division, University Medical Centre Freiburg, Freiburg, Germany
3 Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
4 Department of Nephrology, Johns Hopkins University, School of Medicine, Baltimore, MD, USA
ABCG family
Members of the ABCG family are half transporters
with one ABC cassette in the amino terminus followed
by six putative transmembrane domains (see also
reviews on other ABC transporters in the minireview
series in this issue [1–3]). Full transporters contain two
ABC cassettes and 12 transmembrane domains. Half
transporters assemble to homodimeric and heterodi-
meric complexes to form functional transporters. Fig-
ure 1 provides an overview of the human ABC
transporter superfamily and lists the members of the
ABCG or White family, which is most closely related
to the ABCA family. Currently, five members of the
ABCG subfamily are known to exist in humans:
ABCG1, ABCG2, ABCG4, ABCG5 and ABCG8.

The ABCG1 gene is located on chromosome 21q22.3
[4]. Its product ABCG1 is found in multiple tissues
and has a role in macrophage lipid transport [5].
ABCG2, mapped to chromosome 4q22, was initially
identified in placenta tissue [6] and as a xenobiotic
transporter from a human breast cancer cell line [7]. It
was therefore also termed ‘breast cancer resistance pro-
tein’ (BCRP). The ABCG4 gene is located on chromo-
some 11q23.3 [8,9]. The gene product ABCG4 shows
highest homology to ABCG1, and a role in macro-
phage lipid metabolism has also been proposed [9].
The human ABCG5 and ABCG8 genes, located adja-
cent to each other on chromosome 2p21, were both
identified in the search for genetic causes of a rare
autosomal-recessive lipid metabolism disorder, sitoster-
olemia [10].
ABCG transporters and disease
Members of the ABCG family are known to play a role
in lipid transport across membranes. Loss-of-function
mutations in ABCG5 or ABCG8 cause sitosterolemia,
Keywords
ABCG2; gout; GWAS; hyperuricemia; urate
Correspondence
M. Ko
¨
ttgen, Renal Division, University
Medical Centre Freiburg, Freiburg, Germany
Fax: +49 (0)761 27063240
Tel: +49 (0)761 27032990
E-mail: michael.koettgen@uniklinik-

freiburg.de
(Received 17 December 2010, revised 18
February 2011, accepted 6 May 2011)
doi:10.1111/j.1742-4658.2011.08171.x
ATP-binding cassette (ABC) transporters form a large family of transmem-
brane proteins that facilitate the transport of specific substrates across
membranes in an ATP-dependent manner. Transported substrates include
lipids, lipopolysaccharides, amino acids, peptides, proteins, inorganic ions,
sugars and xenobiotics. Despite this broad array of substrates, the physio-
logical substrate of many ABC transporters has remained elusive. ABC
transporters are divided into seven subfamilies, A–G, based on sequence
similarity and domain organization. Here we review the role of members of
the ABCG subfamily in human disease and how the identification of dis-
ease genes helped to determine physiological substrates for specific ABC
transporters. We focus on the recent discovery of mutations in ABCG2
causing hyperuricemia and gout, which has led to the identification of urate
as a physiological substrate for ABCG2.
Abbreviations
ABC, ATP-binding cassette; SNP, single nucleotide polymorphism.
FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS 3215
a disorder characterized by the accumulation of plant
and fish sterols including cholesterol [10–12]. Clinical
characteristics of sitosterolemia are xanthomatosis and
premature atherosclerosis, resulting in early onset of
cardiovascular disease and lethal myocardial infarction
[13]. Mutations in ABCG5 or ABCG8 cause increased
intestinal absorption and decreased biliary elimination
of plant sterols and cholesterol, leading to a 50- to
200-fold increase in plasma plant sterol concentrations
[13,14]. The encoded proteins ABCG5 and ABCG8

form obligate heterodimers that are expressed in the
apical membrane of enterocytes and in the canicular
membrane of hepatocytes [15]. They limit the absorp-
tion of plant sterols and cholesterol by secreting these
sterols from enterocytes back into the intestinal
lumen, and by excretion of sterols from hepatocytes
into bile. Disruption of ABCG5 and ABCG8 in mice
results in a 3-fold increase in the fractional absorption
of plant sterols, a 30% increase in plasma sitosterol
levels, and a reduction in biliary cholesterol levels
[16]. Thus these mice display many characteristics seen
in patients with sitosterolemia. In accordance with the
phenotypes observed upon disrupted function of
ABCG5 and ABCG8 in humans or mice, it was
recently shown that sterols are the direct substrates of
ABCG5 and ABCG8. Inside-out membrane vesicles
prepared from Sf9 insect cells overexpressing ABCG5
and ABCG8 or from liver membranes showed ATP-
dependent transfer of both cholesterol and sitosterol
[17,18].
To date no functional mutations in ABCG1 and
ABCG4 have been linked to any monogenic human
disease, although ABCG1 has been implicated in car-
diovascular disease, obesity and diabetes (reviewed in
[19]). Abcg1
) ⁄ )
mice on a high-cholesterol diet display
an attenuated endothelium-dependent arterial vasore-
laxation as well as reduced activity of endothelial nitric
oxide synthase, consistent with a role of ABCG1 in

maintaining endothelial cell function by promoting
efflux of cholesterol and 7-oxysterols [20]. In contrast,
ABCG4 is highly expressed in the central nervous sys-
tem. Detailed studies of the brains of Abcg4
) ⁄ )
mice
(< 1 year old) did not identify any pathological
changes, however [19]. Both proteins have been shown
to transport lipids including cholesterol, but their pre-
cise role in vivo remains to be elucidated. It is of great
interest whether future studies will establish a role for
these transporters in inherited human disorders.
Discovery of ABCG2 variants in
association studies of human disease
ABCG2 was first identified as a multidrug resistance
protein (Fig. 2) [7]. It has been shown to transport a
wide range of structurally and functionally diverse sub-
strates such as chemotherapeutics, antibiotics and
HMG-CoA reductase inhibitors. Yet, physiological
substrates and the roles of ABCG2 in vivo had
remained elusive until very recently. As was the case
for ABCG5 and ABCG8, an important physiological
function of ABCG2 was uncovered through genetic
studies of human disease. In a series of genetic and
physiological studies over the past 3 years, it was
established that ABCG2 functions as a novel urate
transporter that promotes urate excretion in the
human kidney.
A genome-wide association study among more than
11 000 individuals of European ancestry, including rep-

lication in an additional 11 000 European ancestry and
3800 African American study participants, identified
common alleles in ABCG2 as associated with serum
urate levels and risk of gout [21]. Gout is a common
form of arthritis with a prevalence of about 1–3% in
western countries [22,23]. Patients with gout experience
very painful attacks caused by the precipitation of
monosodium urate crystals in joints, which triggers
subsequent inflammation. Elevated serum urate levels
are therefore a key risk factor for gout. Earlier studies
showed that serum urate levels are highly heritable
[24]. In fact, the majority of inter-individual variation
of urate levels in a population can be explained by
additive genetic effects. A genome-wide association
study was initiated among individuals participating
in three large, population-based prospective studies
(Atherosclerosis Risk in Communities Study, Framing-
ham Heart Study, Rotterdam Study) in an effort to
discover genes that might explain the genetic effects on
serum urate levels. Each study participant had serum
urate levels measured and genotyping performed either
ABCD
ABCB
ABCC(I)
ABCC(II)
ABCG
ABCA
ABCE
ABCF
ABCG2

ABCG1
ABCG4
ABCG5
ABCG8
Human ABC family members Human ABCGs Disease phenotype
Gout, hyperuricemia
Sitosterolemia
Sitosterolemia
?
?
Fig. 1. Phylogenetic tree of all human ABC genes and specifically
the ABCG subgroup of genes (after [19,66]). Disease phenotypes
reported include only human diseases associated with specific
ABCG mutations, not information from model organisms.
ABCG transporters and disease O. M. Woodward et al.
3216 FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS
as part of a high-throughput single nucleotide poly-
morphism (SNP) chip or as targeted replication geno-
typing. Gout status was ascertained by self-report or
based on the intake of gout-specific medication [21].
Of more than 500 000 SNPs surveyed, the ABCG2 var-
iant with the strongest effect on serum urate concen-
trations was the SNP rs2231142: each additional copy
of the minor T allele was associated with mean serum
urate concentrations approximately 0.25 standard devi-
ations higher among individuals of European ancestry
(P =3· 10
)60
), corresponding to approximately
0.30 mgÆdL

)1
higher mean serum urate per copy of the
T allele (Table 1). The odds of gout were increased by
74% with each copy of the T allele (odds ratio 1.74,
95% confidence interval 1.51–1.99, P =4· 10
)15
).
The association between the risk allele and serum urate
and gout was significantly stronger in men than in
women [21,25].
Since this first study, the effect of the rs2231142 T
allele on mean serum urate levels and the risk of gout
has been replicated in many diverse study populations
and is consistently observed with comparable effect
sizes (Table 1). Replication of a finding in study popu-
lations of different ancestry, where risk allele frequency
and correlation patterns between nearby genomic vari-
ants may differ, is an important feature of a functional
genetic variant. Interestingly, the allele frequency of
the T risk allele in a Japanese study population was
reported as 31% [26], which is approximately three
times more common than the T allele frequency
observed in individuals of European ancestry. While
the prevalence of gout in Japan is lower than in coun-
tries where a western diet is consumed, the prevalence
of gout among US individuals of Asian ancestry has
been reported as three times higher than that of US
individuals of European ancestry [27].
R
L

L
A
A
M
A
T
T
T
R
V
S
G
G
G
F
I
T
Q
R
R
V
K
K
S
G
E
A
D
R
R

V
V
K
K
L
L
G
E
E
E
I
IN
N
N
D
H
Q
Q
R
V
V
V
V
V
L
L
S
G
F
E

N
M
T
T
QD
D
S
K
R
V
K
L
L
G
F
P
C
Y
R
K
S
G
F
P
P
C
N
N
A
V

L
L
S
G
G
G
I
N
A
D
R
K
P
P
S
G
G
G
R
V
V
K
K
K
K
L
L
L
L
L

L
S
S
S
GG
G
PP
E
E
I
I
II
N
NN
M
A
A
AT
DD
Y
N
E
A
I
P
E
S
I
D
L

L
F
T
L
S
G
EIMT
D
I
I
P
F
C
LR
I
H
A
N
T
T
T
T
T
G
L
D
S
S
K
K

K
L
L
L
S
G
G
G
F
F
F
F
Q
P
P
I
M
M
A
A
A
A
D
H
G
G
L
S
S
S

V
L
L
L
L
L
R
R
R
Q
Q
I
I
Y
Y
Y
S
S
H
E
E
A
T
V
V
V
V
L
Q
I

S
F
I
I
I
I
A
A
L
G
G
Y
K
F
R
S
S
E
E
I
I
L
G
Y
Y
Y
Y
V
V
K

H
S
P
C
M
M
D
R
T
I
I
I
L
L
L
F
F
Y
V
S
S
P
F
N
T
I
A
Q
Q
L

L
L
G
F
Y
Y
H
S
S
P
R
W
C
N
M
I
I
A
A
A
L
L
G
F
V
V
K
H
W
T

L
I
F
F
C
C
C
D
D
D
A
A
A
Q
Q
Q
Q
Q
G
G
G
G
G
G
G
G
G
G
F
F

FF
F
F
FF
Y
Y
Y
Y
Y
V
V
V
V
V
V
V
V
K
K
K
K
K
K
K
K
E
E
E
E
P

P
P
P
R
W
W
T
T
T
T
T
TT
T
T
NN
N
N
N
N
N
N
N
M
M
M
M
L
L
L
L

L
L
L
L
L
L
L
L
I
I
I
I
I
I
A
A
A
A
A
A
A
S
S
S
S
S
S
S
S
S

S
L
L
L
L
L
L
L
L
L
L
V
V
F
G
G
CC
T
Q
Q
Q
Q
Y
Y
Y
K
K
KK
K
H

H
E
E
E
EEE
E
EE
E
P
PP
P
R
R
R
W
N
N
N
I
I
I
II
I
I
I
A
AAA
A
A
A

S
S
S
SS S
S
LL
L
L
L
L
L
V
V
V
V
F
F
F
F
FF
F
G
G
G
G
C
TT
T
T
T

T
K
K
K
K
K
KKK
N
N
N
LL
D
D
D
D
DS
S
395
469
565
644
414
450
495
505
584
625
Signature
Walker A
Walker B

Q
E
P
M
I
A
V
V
V
FF
G
G
T
NN
N
S
S
SS
P
F
H
E
V
F
G
C
T
T
K
N

N
L
L
D
S
S
A
A
A
I
V12M
N-terminus
C-terminus
M
M
M
M
M
T
A
A
A
A
L
F
F
Y
V
V
S

S
S
F
524
476
Y
Q126X
G268R
S441N
F506fs
Q141K
44
288
P
P
A
A
D
D
Fig. 2. Topographical representation of the ABCG2 monomer in the plasma membrane. Transmembrane domains experimentally determined
by Wang et al. (2008) [67]; nucleotide binding domain (NBD) begins at Y44 and ends at residue N288 [68]. The Walker A and B and ABC sig-
nature motif of the nucleotide binding domain are identified, as are the six human polymorphisms associated with hyperuricemia and gout
(in red) [21,41]. Amino acid residues: pink, aromatic; green, + charged; light blue, ) charged; white, nonpolar; yellow, polar residues.
O. M. Woodward et al. ABCG transporters and disease
FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS 3217
Physiological function of ABCG2
A connection between ABCG2 and urate metabolism
or gout had not been described until this first genome-
wide association study. It was known, however, that
human ABCG2 is expressed in the apical membrane of

human proximal tubule cells [28], the main site of
urate handling in the human kidney. We therefore
investigated whether urate is a physiological substrate
of ABCG2, and whether the Q141K variant, encoded
by rs2231142, leads to altered urate transport and as a
consequence to elevated serum urate levels and
increased risk of gout.
In order to test whether ABCG2 was a yet unknown
urate transporter, ABCG2 was expressed in Xenopus
oocytes [29]. Accumulation of radiolabeled urate in oo-
cytes expressing ABCG2 was decreased by 75% com-
pared with water-injected control oocytes (Fig. 3A).
The reduced urate accumulation was caused by
ABCG2-mediated urate efflux from cells rather than
by the inhibition of urate uptake, as shown in experi-
ments monitoring the decrease of intracellular urate
over time in oocytes preloaded with radiolabeled urate
(Fig. 3B). Although it was known that the major site
of urate excretion in humans is the proximal tubule in
the kidney, the molecular identity of the transporters
mediating urate secretion at the apical membrane of
proximal tubular cells had only been poorly under-
stood. To study ABCG2 function at this location,
urate accumulation and localization of ABCG2 was
studied in native LLC-PK1 cells, a porcine proximal
tubule cell line. These experiments revealed that
ABCG2 mediates the apical secretion of urate in proxi-
mal tubule cells (Fig. 3D). A similar function and
localization has been shown for MRP4 [30,31], but
polymorphisms in MRP4 have not been linked to

hyperuricemia and gout in humans.
Given the vast literature on ABCG2 with dozens of
structurally diverse substrates it appears surprising at
first glance that urate was not found to be a physiolog-
ical substrate earlier. Notably, ABCG2 knockout mice
do not develop gout. One of the reasons for urate stay-
ing under the radar of ABCG2 research may be that
gout is a complex genetic disease with multiple contrib-
uting genetic and environmental factors. More impor-
tantly though, there are striking species differences in
purine metabolism within the animal kingdom. Urate
is the end product of purine metabolism in humans.
Humans and higher primates have much higher serum
urate levels than other mammals because they lack the
enzyme uricase, which converts urate into allantoin
[32]. Therefore genetic factors that predispose to
hyperuricemia and gout cannot be easily studied in
rodent models.
Q141K is a functional variant in ABCG2
Several lines of evidence in the initial genome-wide
association study by Dehghan et al. [21] suggested
that the rs2231142 variant may be functional. First,
Table 1. Effect sizes of the ABCG2 rs2231142 (Q141K) variant on risk of gout and mean urate levels in study populations of different
ancestry.
Study sample
ethnicity
Sample
size
Risk allele
frequency (T)

Odds ratio for
gout per T allele,
95% CI
Effect on mean
serum urate per T allele Ref.
European ancestry 22 871 0.11 1.74 0.24 standard deviation changes [21]
European ancestry 28 141 0.11 NA 0.17 z-score units [44]
European ancestry 28 283 0.11
a
1.86 0.30 mgÆdL
)1
[45]
European ancestry 4492 0.11–0.12 NA 0.34 mgÆdL
)1
[62]
European ancestry 2246 0.1 (controls),
0.14 (cases)
1.37 NA [63]
African American 3843 0.03 1.71 0.22 standard deviations [21]
Japanese 739 0.32 2.5 in a subset
of gout patients
 0.4 mgÆdL
)1
[41]
Japanese 3923 0.31 1.37 for genotype
TG, 4.37 for genotype TT
[26]
Japanese 5165 0.23–0.30 NA 0.1 mgÆdL
)1
per risk allele [64]

New Zealand
population
Cases ⁄ controls:
185 ⁄ 284 Maori,
173 ⁄ 129 Pacific
Islanders, 214 ⁄ 562
Caucasian
1.08 Maori, 2.80 Pacific
Islanders, 2.20 Caucasian
[65]
a
highly correlated SNP rs2199936 was studied (r2 = 0.92 in HapMap CEU r22).
ABCG transporters and disease O. M. Woodward et al.
3218 FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS
the variant is located in exon 5 of ABCG2 and leads
to a glutamine-to-lysine amino acid substitution
(Q141K) in ABCG2. This substitution is predicted to
have a possibly damaging effect by the functional
prediction program polyphen-2 [33]. Second, the glu-
tamine residue at position 141 is highly conserved
across species. No other common variants in the
ABCG2 gene region showed association with serum
urate levels after accounting for the effect of
rs2231142 [21,29].
However, while genome-wide association studies
have been extremely successful at establishing associa-
tions between common SNPs and a multitude of com-
plex diseases [34], these studies cannot establish
whether a disease-associated SNP is causally related to
the disease or merely a naturally occurring genetic

marker that is correlated with another, unknown func-
tional variant. To test whether the rs2231142 is such a
functional variant, the transport capacity of the
encoded Q141K mutation was compared with that of
wild-type ABCG2. Oocytes expressing ABCG2 Q141K
showed 54% reduced urate transport rates compared
with oocytes expressing wild-type ABCG2 (Fig. 3C).
This is consistent with previous studies showing
impaired transport of chemotherapeutic agents by
ABCG2 Q141K [35,36] (and reviewed in [37]). While it
is difficult to compare the results from different trans-
port assays and substrates, the reduction of transport
of the Q141K variant compared with wild-type
ABCG2 appears to be of similar magnitude. The Q141
residue is located in the nucleotide binding domain of
ABCG2 (Fig. 2), and Q141K ABCG2 expression is sig-
nificantly lower than wild-type when overexpressed in
mammalian cells [35,36,38,39]. Interestingly, the F508
mutation in CFTR, a related ABC transporter, is
located right next to this position in the nucleotide
binding domain and is commonly mutated in cystic
fibrosis patients [40]. And like the Q141K ABCG2
mutation, expression of the deleted F508 CFTR
mutant is significantly lower than wild-type suggesting
a common pathophysiology (Woodward, unpublished
observations).
0.0
0.5
1.0
1.5

Urate accumulation
pmol per oocyte·120 min
–1
Urate accumulation
pmol per oocyte·120 min
–1
H
2
O ABCG2
∗∗
A
0204060
0.4
0.6
0.8
1.0
Relative urate remaining
Time (min)
∗∗
∗∗
∗∗
∗∗
∗∗
B
Lumen
Blood
0.0
0.3
0.6
0.9

WT Q141K
∗∗
C
Others
3
Others
4
SLC2A9
URAT1
SLC2A9
Others
1
Others
2
U-
U-
U-
U- U-
U-
U-
U-
U-
U-
U-
U-
U-
U-
U-
U-
U-

U-
U-
D
ABCG2
Fig. 3. ABCG2 is a urate transporter. (A) C-14 urate accumulation from Xenopus oocytes injected with mRNA coding for either ABCG2 or
H
2
O controls. (B) Urate efflux in oocytes incubated overnight in 500 lM C-14 urate as relative efflux over time (blue, control; red, ABCG2).
(C) Urate accumulation in oocytes expressing either the wild-type ABCG2 or the mutant Q141K ABCG2. (**P < 0.01, ± SEM) (A–C originally
from [29]; ª 2009 by the National Academy of Sciences of the USA). (D) Model of urate transport in the proximal tubule of the human
kidney overlying fluorescent micrograph of LLCPK-1 proximal tubule cell with endogenous ABCG2 labeled in green and the nucleus in blue.
Proteins influencing urate absorption and secretion and with significance for human diseases are shown with the direction of urate transport
indicated [21,69,70]. Other transporters expressed in the human kidney and shown to transport urate in model systems:
1
OAT4;
2
OAT1,
OAT3;
3
MRP4;
4
OAT1, OAT3 [71,72].
O. M. Woodward et al. ABCG transporters and disease
FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS 3219
The role of ABCG2 as a urate transporter with
mutations leading to hyperuricemia and gout was
recently confirmed and further investigated by Matsuo
et al. [41]. The investigators of this study identified sev-
eral non-synonymous coding variants in ABCG2
through sequencing of the ABCG2 gene in 90 hyperuri-

cemia patients in a Japanese population. In addition to
Q141K, Q126X was identified as a novel loss-of-func-
tion variant. Q126X was assigned to a different haplo-
type than Q141K and shown to increase gout risk
(odds ratio 5.97) to an even greater extent than the
Q141K variant. In addition, 10% of the gout patients
studied had genotype combinations of the Q141K and
Q126X variants that resulted in more than a 75%
reduction of ABCG2 function compared with patients
that were homozygous for the non-risk allele at both
variants (odds ratio 25.8, 95% confidence interval
10.3–64.6).
Many additional SNPs and their role in ABCG2
function have been analyzed [37,42], but these studies
have not addressed the impact of other SNPs in urate
transport and gout. Future studies will have to test
whether additional functional SNPs also affect serum
urate concentrations in humans.
Urate transport is complex: in the kidney, urate
transport is bidirectional and involves multiple differ-
ent transport and regulatory proteins [32,43]. This is
reflected in the complex genetic architecture of serum
urate levels and risk of gout: two recent large gen-
ome-wide association studies identified variants in
multiple genes associated with serum urate concentra-
tions (SLC2A9, ABCG2, SLC17A1, SLC22A11,
SLC22A12, SLC16A9, GCKR, LRRC16A, PDZK1,
the R3HDM2–INHBC region and RREB1) [44,45].
The effect of the individual common risk alleles in
these genes on mean serum urate concentrations and

the risk of gout is modest. The range of the pheno-
typic variation in serum urate levels in the studied
populations that could be explained by the individual
genetic variants ranged from 0.1% to 3.5%. However,
the effect of urate-increasing alleles at different geno-
mic loci can add up: Yang et al. [45] estimated
from several large population-based studies that mean
urate levels increased from approximately 4.5 to
6.2 mgÆdL
)1
across a genetic score composed of the
risk alleles at eight different genomic loci. Similarly,
the prevalence of gout increased from 2% to more
than 20% at the upper extreme of the risk score.
Some of the genes identified in the two large studies
mentioned above encode for known urate transporters
(SLC2A9, ABCG2, SLC17A1, SLC22A11, SLC22A12)
or regulators thereof (PDZK1). For the remaining
genes, little is known about a possible connection of
the gene product to urate metabolism in humans and
therefore this constitutes a new area for future
research.
ABCG2 function in other tissues
ABCG2’s physiological function has been difficult to
identify because of the large number of known sub-
strates and varied tissue expression. Suggested physio-
logical roles include functioning as a xenobiotic
transporter, conferring xenobiotic protection in tissues
like the liver, intestine, placenta and CNS [37]; and as
a transporter of heme and other porphyrins, prevent-

ing their accumulation in erythrocytes and stem cells
[46,47]. As noted above ABCG2 plays a significant role
in urate transport in the human kidney, but does
ABCG2 expression in other tissues fit with this newly
postulated function? Here we would like to discuss the
putative physiological role of ABCG2-mediated urate
transport in other tissues. In addition to the kidney,
ABCG2 is expressed at high levels in the liver, at the
blood–brain barrier, in the placenta and in mammary
glands. An examination of ABCG2 at each of these
locations suggests that ABCG2 expression is consistent
with sites of urate transport. In human hepatocytes,
ABCG2 is expressed in the basolateral membrane [48]
oriented to mediate efflux into the biliary canaliculus.
Though ABCG2 is effectively situated to remove drugs
and toxins from the liver, it is also well situated to
export urate out of the liver via the biliary system, a
known urate excretion pathway [49]. ABCG2, in addi-
tion to the urate transporter MRP4 [31], are the only
identified urate transporters positioned to secrete urate
into the biliary system, and thus ABCG2 could be
playing a substantial role in the liver-mediated urate
excretion pathway. At the blood–brain barrier,
ABCG2 is expressed on the luminal membrane of
endothelial cells, seemingly well positioned to protect
the brain from accumulating xenotoxins [50]. However,
there is also ample evidence that misregulation of urate
at the blood–brain barrier has profound effects on
brain function and health. Cerebrospinal fluid urate
levels and serum urate levels are correlated [51,52] but

urate concentration in cerebrospinal fluid is only 7%
of that in serum [52], suggesting an important role for
urate secretion from the cerebrospinal fluid. Higher
serum urate levels are associated with cognitive dys-
function [53] but are also protective against developing
Parkinson’s disease [52]. Thus a tight regulation of
cerebrospinal fluid urate appears important. High
expression of ABCG2 at the blood–brain barrier may
help maintain appropriate urate concentrations in the
brain and the cerebrospinal fluid.
ABCG transporters and disease O. M. Woodward et al.
3220 FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS
Pregnancy has a profound effect on ABCG2
expression at two sites. First, ABCG2 is expressed
highly in the apical membrane of placental syncytio-
trophoblasts and is hypothesized to aid in the protec-
tion of the fetus from toxins or to regulate fetal
estrogen levels by transporting estrogen precursor
molecules [54]. However, ABCG2-mediated efflux of
urate from the placenta may be critical for normal
fetal development. It was recently reported that high
urate levels in amniotic fluid correlated with lower
birth weights, finding a 2 mgÆdL
)1
decrease in amni-
otic urate results in a 120 g increase in birth weight
[55]. Second, pregnancy and lactation increases
ABCG2 expression in mammary gland alveolar epi-
thelial cells. This can result in the concentrating of
xenotoxins, if present in the mother, into breast milk

[56], a seemingly undesirable outcome for a nursing
infant. This apparent contradiction prompted the pro-
posal that ABCG2 may be mostly transporting non-
toxic substitutes like riboflavin [57]. Yet ABCG2
knockout models show no reduction of this vitamin
in breast milk [58]. In contrast, there is some evidence
that human breast milk plays an important role in
delivering antioxidants, including urate, to infants
[59]. Interestingly, while human breast milk contains
urate, it does not contain orotic acid, which is found
in high concentrations in other mammalian milk
[60,61]. Orotic acid is a strong uricosuric compound,
and its disappearance from human milk is consistent
with the evolutionarily conserved loss of uricase
rs2231142
P = 4*10
–27
0
20
40
60
30
25
20
15
10
5
0
IBSP MEPE SPP1 PKD2
ABCG2

PPM1K
Disease
GWAS
Physiology
Treatment ?
0
2
46810
0.00
0.05
0.10
Efflux: (pmol per oocyte·min
–1
)
Internal oocyte concentration (µM)
Fig. 4. The cycle of translational research can begin with the description of a disease phenotype like the destruction of joints that occurs in
patients with gout from urate crystal deposition. Genome-wide association studies allow the identification of genes that associate with ele-
vated serum urate levels and gout. Subsequent in-depth physiological characterization of the gene and its protein product lays the foundation
for an improved understanding of physiology and pathophysiology and may reveal a therapeutic target. Finally, drug development can be
attempted in order to better treat hyperuricemia or gout (X-ray kindly provided by Janet Maynard).
O. M. Woodward et al. ABCG transporters and disease
FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS 3221
function to increase urate levels in humans. In sum-
mary, a role of ABCG2-mediated urate secretion in
several non-renal tissues is conceivable and needs to
be investigated in more detail.
Pharmacological modulation of ABCG2, both inhi-
bition and activation, has been proposed as therapeutic
strategies for numerous human diseases. For instance,
inhibition of ABCG2 has been tested to overcome

multidrug resistance in cancer therapy. However, based
on the function of ABCG2 in urate excretion, one pos-
sible side effect of ABCG2 inhibitors could be
increased serum urate concentrations and gout attacks.
Further studies on ABCG2 are needed to learn more
about its function in different tissues and the relevance
of additional physiological substrates. These studies
may help to predict therapeutic effects as well as side
effects of drugs targeting ABCG2.
Future perspectives and conclusion
In summary, mutations in members of the ABCG
family have led to the identification of physiological
substrates and functions of these transporters. We
anticipate that future studies will continue to
uncover additional novel physiological substrates and
functions for ABC transporters and define additional
roles in human disease. The powerful combination of
genetic and physiological approaches not only may
identify novel mechanisms but may also help to
identify novel therapeutic targets. ABCG2 represents
an attractive drug target since pharmacological acti-
vation of ABCG2 may help to promote urate excre-
tion from the body. The discovery of ABCG2 as a
novel urate transporter is a prime example for trans-
lational research. Hopefully, the fast translation from
bedside to bench will eventually lead back to the
bedside and benefit patients suffering from gout
(Fig. 4).
Acknowledgements
We acknowledge the work of many others whose work

we could not cite due to space constraints. O.M.W.
was supported by NIDDK: DK032753-25A1, A.K.
was supported by the Emmy Noether programme of
DFG and M.K. was supported by DFG KFO 201
and by Alfried Krupp von Bohlen und Halbach
Foundation.
References
1 Nagao K, Tomioka M & Ueda K (2011) Function and
regulation of ABCA1 – membrane meso-domain
organization and reorganization. FEBS J 278,
3190–3203.
2 Pollock NL & Callaghan R (2011) The lipid translo-
case, ABCA4: seeing is believing. FEBS J 278, 3204–
3214.
3 Chen Z-S (2011) Multidrug resistance proteins
(MRPs ⁄ ABCCs) in cancer chemotherapy and genetics
disease. FEBS J 278, 3226–3245.
4 Chen H, Rossier C, Lalioti MD, Lynn A, Chakravarti
A, Perrin G & Antonarakis SE (1996) Cloning of the
cDNA for a human homologue of the Drosophila white
gene and mapping to chromosome 21q22.3. Am J Hum
Genet 59, 66–75.
5 Klucken J, Buchler C, Orso E, Kaminski WE, Porsch-
Ozcurumez M, Liebisch G, Kapinsky M, Diederich W,
Drobnik W, Dean M et al. (2000) ABCG1 (ABC8), the
human homolog of the Drosophila white gene, is a reg-
ulator of macrophage cholesterol and phospholipid
transport. Proc Natl Acad Sci USA 97, 817–822.
6 Allikmets R, Schriml LM, Hutchinson A, Romano-
Spica V & Dean M (1998) A human placenta-specific

ATP-binding cassette gene (ABCP) on chromosome
4q22 that is involved in multidrug resistance. Cancer
Res 58, 5337–5339.
7 Doyle LA, Yang W, Abruzzo LV, Krogmann T, Gao
Y, Rishi AK & Ross DD (1998) A multidrug resistance
transporter from human MCF-7 breast cancer cells.
Proc Natl Acad Sci USA 95, 15665–15670.
8 Annilo T, Tammur J, Hutchinson A, Rzhetsky A, Dean
M & Allikmets R (2001) Human and mouse orthologs
of a new ATP-binding cassette gene, ABCG4. Cytogenet
Cell Genet 94, 196–201.
9 Engel T, Lorkowski S, Lueken A, Rust S, Schluter B,
Berger G, Cullen P & Assmann G (2001) The human
ABCG4 gene is regulated by oxysterols and retinoids in
monocyte-derived macrophages. Biochem Biophys Res
Commun 288, 483–488.
10 Berge KE, Tian H, Graf GA, Yu L, Grishin NV,
Schultz J, Kwiterovich P, Shan B, Barnes R & Hobbs
HH (2000) Accumulation of dietary cholesterol in
sitosterolemia caused by mutations in adjacent ABC
transporters. Science 290, 1771–1775.
11 Dean M (2005) The genetics of ATP-binding cassette
transporters. Methods Enzymol 400, 409–429.
12 Lee MH, Lu K, Hazard S, Yu H, Shulenin S, Hidaka
H, Kojima H, Allikmets R, Sakuma N, Pegoraro R
et al. (2001) Identification of a gene, ABCG5, important
in the regulation of dietary cholesterol absorption.
Nat Genet 27, 79–83.
13 Sudhop T & von Bergmann K (2004) Sitosterolemia –
a rare disease. Are elevated plant sterols an additional

risk factor? Z Kardiol 93, 921–928.
14 Berge KE (2003) Sitosterolemia: a gateway to new
knowledge about cholesterol metabolism. Ann Med 35,
502–511.
ABCG transporters and disease O. M. Woodward et al.
3222 FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS
15 Graf GA, Li WP, Gerard RD, Gelissen I, White A,
Cohen JC & Hobbs HH (2002) Coexpression of ATP-
binding cassette proteins ABCG5 and ABCG8 permits
their transport to the apical surface. J Clin Invest 110,
659–669.
16 Yu L, Hammer RE, Li-Hawkins J, Von Bergmann K,
Lutjohann D, Cohen JC & Hobbs HH (2002) Disrup-
tion of Abcg5 and Abcg8 in mice reveals their crucial
role in biliary cholesterol secretion. Proc Natl Acad Sci
USA 99, 16237–16242.
17 Wang J, Sun F, Zhang DW, Ma Y, Xu F, Belani JD,
Cohen JC, Hobbs HH & Xie XS (2006) Sterol transfer
by ABCG5 and ABCG8: in vitro assay and reconstitu-
tion. J Biol Chem 281, 27894–27904.
18 Wang J, Zhang DW, Lei Y, Xu F, Cohen JC, Hobbs
HH & Xie XS (2008) Purification and reconstitution of
sterol transfer by native mouse ABCG5 and ABCG8.
Biochemistry 47, 5194–5204.
19 Tarr PT, Tarling EJ, Bojanic DD, Edwards PA &
Baldan A (2009) Emerging new paradigms for ABCG
transporters. Biochim Biophys Acta 1791, 584–593.
20 Terasaka N, Yu S, Yvan-Charvet L, Wang N, Mzhavia
N, Langlois R, Pagler T, Li R, Welch CL, Goldberg IJ
et al. (2008) ABCG1 and HDL protect against endothe-

lial dysfunction in mice fed a high-cholesterol diet.
J Clin Invest 118, 3701–3713.
21 Dehghan A, Kottgen A, Yang Q, Hwang SJ, Kao WL,
Rivadeneira F, Boerwinkle E, Levy D, Hofman A,
Astor BC et al. (2008) Association of three genetic loci
with uric acid concentration and risk of gout: a gen-
ome-wide association study. Lancet 372, 1953–1961.
22 Choi HK, Mount DB & Reginato AM (2005) Patho-
genesis of gout. Ann Intern Med 143, 499–516.
23 Lawrence RC, Felson DT, Helmick CG, Arnold LM,
Choi H, Deyo RA, Gabriel S, Hirsch R, Hochberg
MC, Hunder GG et al. (2008) Estimates of the preva-
lence of arthritis and other rheumatic conditions in the
United States. Part II. Arthritis Rheum 58, 26–35.
24 Yang Q, Guo CY, Cupples LA, Levy D, Wilson PW &
Fox CS (2005) Genome-wide search for genes affecting
serum uric acid levels: the Framingham Heart Study.
Metabolism 54, 1435–1441.
25 Tanaka Y, Slitt AL, Leazer TM, Maher JM & Klaassen
CD (2005) Tissue distribution and hormonal regulation
of the breast cancer resistance protein (Bcrp ⁄ Abcg2)
in rats and mice. Biochem Biophys Res Commun 326,
181–187.
26 Yamagishi K, Tanigawa T, Kitamura A, Kottgen A,
Folsom AR & Iso H (2010) The rs2231142 variant of
the ABCG2 gene is associated with uric acid levels and
gout among Japanese people. Rheumatology (Oxford)
49, 1461–1465.
27 Krishnan E, Lienesch D & Kwoh CK (2008) Gout
in ambulatory care settings in the United States.

J Rheumatol 35, 498–501.
28 Huls M, Brown CD, Windass AS, Sayer R, van den
Heuvel JJ, Heemskerk S, Russel FG & Masereeuw R
(2008) The breast cancer resistance protein transporter
ABCG2 is expressed in the human kidney proximal
tubule apical membrane. Kidney Int 73, 220–225.
29 Woodward OM, Kottgen A, Coresh J, Boerwinkle E,
Guggino WB & Kottgen M (2009) Identification of a
urate transporter, ABCG2, with a common functional
polymorphism causing gout. Proc Natl Acad Sci USA
106, 10338–10342.
30 van Aubel RA, Smeets PH, Peters JG, Bindels RJ &
Russel FG (2002) The MRP4 ⁄ ABCC4 gene encodes a
novel apical organic anion transporter in human kidney
proximal tubules: putative efflux pump for urinary
cAMP and cGMP. J Am Soc Nephrol
13, 595–603.
31 Van Aubel RA, Smeets PH, van den Heuvel JJ &
Russel FG (2005) Human organic anion transporter
MRP4 (ABCC4) is an efflux pump for the purine end
metabolite urate with multiple allosteric substrate bind-
ing sites. Am J Physiol Renal Physiol 288, F327–333.
32 Anzai N, Kanai Y & Endou H (2007) New insights into
renal transport of urate. Curr Opin Rheumatol 19,
151–157.
33 Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE,
Gerasimova A, Bork P, Kondrashov AS & Sunyaev SR
(2010) A method and server for predicting damaging
missense mutations. Nat Methods 7, 248–249.
34 Manolio TA (2010) Genomewide association studies

and assessment of the risk of disease. N Engl J Med
363, 166–176.
35 Mizuarai S, Aozasa N & Kotani H (2004) Single nucle-
otide polymorphisms result in impaired membrane
localization and reduced atpase activity in multidrug
transporter ABCG2. Int J Cancer 109, 238–246.
36 Morisaki K, Robey RW, Ozvegy-Laczka C, Honjo Y,
Polgar O, Steadman K, Sarkadi B & Bates SE (2005)
Single nucleotide polymorphisms modify the transporter
activity of ABCG2. Cancer Chemother Pharmacol 56,
161–172.
37 Polgar O, Robey RW & Bates SE (2008) ABCG2:
structure, function and role in drug response. Expert
Opin Drug Metab Toxicol 4, 1–15.
38 Furukawa T, Wakabayashi K, Tamura A, Nakagawa
H, Morishima Y, Osawa Y & Ishikawa T (2009) Major
SNP (Q141K) variant of human ABC transporter
ABCG2 undergoes lysosomal and proteasomal degrada-
tions. Pharm Res 26, 469–479.
39 Imai Y, Nakane M, Kage K, Tsukahara S, Ishikawa E,
Tsuruo T, Miki Y & Sugimoto Y (2002) C421A poly-
morphism in the human breast cancer resistance protein
gene is associated with low expression of Q141K protein
and low-level drug resistance. Mol Cancer Ther 1,
611–616.
40 Fuller CM & Benos DJ (1992) Cftr!. Am J Physiol 263,
C267–286.
O. M. Woodward et al. ABCG transporters and disease
FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS 3223
41 Matsuo H, Takada T, Ichida K, Nakamura T,

Nakayama A, Ikebuchi Y, Ito K, Kusanagi Y, Chiba
T, Tadokoro S et al. (2009) Common defects of
ABCG2, a high-capacity urate exporter, cause gout: a
function-based genetic analysis in a Japanese popula-
tion. Sci Transl Med 1, 5ra11.
42 Koshiba S, An R, Saito H, Wakabayashi K, Tamura A
& Ishikawa T (2008) Human ABC transporters ABCG2
(BCRP) and ABCG4. Xenobiotica 38, 863–888.
43 So A & Thorens B (2010) Uric acid transport and
disease. J Clin Invest 120, 1791–1799.
44 Kolz M, Johnson T, Sanna S, Teumer A, Vitart V,
Perola M, Mangino M, Albrecht E, Wallace C, Farrall
M et al. (2009) Meta-analysis of 28,141 individuals
identifies common variants within five new loci that
influence uric acid concentrations. PLoS Genet 5,
e1000504.
45 Yang Q, Kottgen A, Dehghan A, Smith AV, Glazer
NL, Chen MH, Chasman DI, Aspelund T, Eiriksdottir
G, Harris TB et al. (2010) Multiple genetic loci influ-
ence serum urate and their relationship with gout and
cardiovascular disease risk factors. Circ Cardiovasc
Genet 3, 523–530.
46 Jonker JW, Buitelaar M, Wagenaar E, Van Der Valk
MA, Scheffer GL, Scheper RJ, Plosch T, Kuipers F,
Elferink RP, Rosing H et al. (2002) The breast cancer
resistance protein protects against a major chlorophyll-
derived dietary phototoxin and protoporphyria. Proc
Natl Acad Sci USA 99, 15649–15654.
47 Krishnamurthy P, Ross DD, Nakanishi T, Bailey-Dell
K, Zhou S, Mercer KE, Sarkadi B, Sorrentino BP &

Schuetz JD (2004) The stem cell marker Bcrp ⁄ ABCG2
enhances hypoxic cell survival through interactions with
heme. J Biol Chem 279, 24218–24225.
48 Ieiri I, Higuchi S & Sugiyama Y (2009) Genetic
polymorphisms of uptake (OATP1B1, 1B3) and efflux
(MRP2, BCRP) transporters: implications for inter-indi-
vidual differences in the pharmacokinetics and pharma-
codynamics of statins and other clinically relevant
drugs. Expert Opin Drug Metab Toxicol 5, 703–729.
49 Kountouras J, Magoula I, Tsapas G & Liatsis I (1996)
The effect of mannitol and secretin on the biliary
transport of urate in humans. Hepatology 23 , 229–233.
50 Scherrmann JM (2005) Expression and function of
multidrug resistance transporters at the blood–brain
barriers. Expert Opin Drug Metab Toxicol 1, 233–246.
51 Bowman GL, Shannon J, Frei B, Kaye JA & Quinn JF
(2010) Uric acid as a CNS antioxidant. J Alzheimers
Dis 19, 1331–1336.
52 Schlesinger I & Schlesinger N (2008) Uric acid in
Parkinson’s disease. Mov Disord 23, 1653–1657.
53 Vannorsdall TD, Jinnah HA, Gordon B, Kraut M &
Schretlen DJ (2008) Cerebral ischemia mediates the
effect of serum uric acid on cognitive function. Stroke
39, 3418–3420.
54 Mao Q (2008) BCRP ⁄ ABCG2 in the placenta: expres-
sion, function and regulation. Pharm Res 25, 1244–
1255.
55 Gao T, Zablith NR, Burns DH, Skinner CD & Koski
KG (2008) Second trimester amniotic fluid transferrin
and uric acid predict infant birth outcomes. Prenat

Diagn 28, 810–814.
56 Jonker JW, Merino G, Musters S, van Herwaarden AE,
Bolscher E, Wagenaar E, Mesman E, Dale TC &
Schinkel AH (2005) The breast cancer resistance protein
BCRP (ABCG2) concentrates drugs and carcinogenic
xenotoxins into milk. Nat Med 11, 127–129.
57 van Herwaarden AE & Schinkel AH (2006) The func-
tion of breast cancer resistance protein in epithelial bar-
riers, stem cells and milk secretion of drugs and
xenotoxins. Trends Pharmacol Sci 27, 10–16.
58 van Herwaarden AE, Wagenaar E, Merino G, Jonker
JW, Rosing H, Beijnen JH & Schinkel AH (2007) Mul-
tidrug transporter ABCG2 ⁄ breast cancer resistance pro-
tein secretes riboflavin (vitamin B2) into milk. Mol Cell
Biol 27, 1247–1253.
59 Aycicek A, Erel O, Kocyigit A, Selek S & Demirkol
MR (2006) Breast milk provides better antioxidant
power than does formula. Nutrition 22, 616–619.
60 Ferreira IM (2003) Quantification of non-protein nitro-
gen components of infant formulae and follow-up
milks: comparison with cows’ and human milk. Br J
Nutr 90, 127–133.
61 Indyk HE & Woollard DC (2004) Determination of
orotic acid, uric acid, and creatinine in milk by liquid
chromatography. J AOAC Int 87, 116–122.
62 Brandstatter A, Lamina C, Kiechl S, Hunt SC, Coassin
S, Paulweber B, Kramer F, Summerer M, Willeit J,
Kedenko L et al. (2010) Sex and age interaction with
genetic association of atherogenic uric acid concentra-
tions. Atherosclerosis 210, 474–478.

63 Stark K, Reinhard W, Grassl M, Erdmann J, Schunkert
H, Illig T & Hengstenberg C (2009) Common polymor-
phisms influencing serum uric acid levels contribute
to susceptibility to gout, but not to coronary artery
disease. PLoS ONE 4, e7729.
64 Tabara Y, Kohara K, Kawamoto R, Hiura Y, Nishim-
ura K, Morisaki T, Kokubo Y, Okamura T, Tomoike
H, Iwai N et al. (2010) Association of four genetic loci
with uric acid levels and reduced renal function: the J-
SHIPP Suita study. Am J Nephrol 32, 279–286.
65 Phipps-Green AJ, Hollis-Moffatt JE, Dalbeth N, Merri-
man ME, Topless R, Gow PJ, Harrison AA, Highton J,
Jones PB, Stamp LK et al. (2010) A strong role for the
ABCG2 gene in susceptibility to gout in New Zealand
Pacific Island and Caucasian, but not Maori, case and
control sample sets. Hum Mol Genet 19, 4813–4819.
66 Dean M (2002) The human ATP-binding cassette
(ABC) transporter superfamily [Internet]. National
Library of Medicine (US)(ID:NBK3), NCBI, Bethesda,
ABCG transporters and disease O. M. Woodward et al.
3224 FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS
MD. Available from: />books/NBK3/
67 Wang H, Lee EW, Cai X, Ni Z, Zhou L & Mao Q
(2008) Membrane topology of the human breast cancer
resistance protein (BCRP ⁄ ABCG2) determined by epi-
tope insertion and immunofluorescence. Biochemistry
47, 13778–13787.
68 Li YF, Polgar O, Okada M, Esser L, Bates SE & Xia
D (2007) Towards understanding the mechanism of
action of the multidrug resistance-linked half-ABC

transporter ABCG2: a molecular modeling study. J Mol
Graph Model 25, 837–851.
69 Enomoto A, Takeda M, Tojo A, Sekine T, Cha SH,
Khamdang S, Takayama F, Aoyama I, Nakamura S,
Endou H et al. (2002) Role of organic anion transport-
ers in the tubular transport of indoxyl sulfate and the
induction of its nephrotoxicity. J Am Soc Nephrol 13,
1711–1720.
70 Vitart V, Rudan I, Hayward C, Gray NK, Floyd J,
Palmer CN, Knott SA, Kolcic I, Polasek O, Graessler J
et al. (2008) SLC2A9 is a newly identified urate trans-
porter influencing serum urate concentration, urate
excretion and gout. Nat Genet 40, 437–442.
71 Enomoto A & Endou H (2005) Roles of organic anion
transporters (OATs) and a urate transporter (URAT1)
in the pathophysiology of human disease. Clin Exp
Nephrol 9, 195–205.
72 Wright AF, Rudan I, Hastie ND & Campbell H (2010)
A ‘complexity’ of urate transporters. Kidney Int 78,
446–452.
O. M. Woodward et al. ABCG transporters and disease
FEBS Journal 278 (2011) 3215–3225 ª 2011 The Authors Journal compilation ª 2011 FEBS 3225