Báo cáo khoa học: Regulation of the expression and subcellular localization of the taurine transporter TauT in mouse NIH3T3 fibroblasts doc

Regulation of the expression and subcellular localization
of the taurine transporter TauT in mouse NIH3T3 fibroblasts
Jesper W. Voss, Stine F. Pedersen, Søren T. Christensen and Ian H. Lambert
The August Krogh Institute, Biochemical Department, Universitetsparken 13, Copenhagen, Denmark
The cellular level of the organic osmolyte taurine is a balance
between active uptake and passive leak via a volume sensitive
pathway. Here, we d emonstrate t hat NIH3T3 mouse fibro-
blasts express a saturable, high affinity taurine transporter
(TauT, K
m
¼ 18 l
M
), and that taurine uptake via TauT is
aNa
+
-andCl
–
-dependent process with an apparent
2.5 : 1 : 1 Na
+
/Cl
–
/taurine stoichiometry. Transport activ-
ity is reduced following acute administration of H
2
O
2
or
activators of p rotein kinases A or C. TauT transport a ctivity,
expression and nuclear localization are significantly
increased upon serum starvation (24 h), exposure to tumour

necrosis factor alpha (TNFa; 16 h), or hyperosmotic med-
ium (24 h); conditions that are a lso a ssociated with increased
localization of TauT to the cytosolic network of micro-
tubules. Conversely, transport activity, expression and
nuclear localization of TauT are reduced in a reversible
manner following long-term exposure (24 h ) to high extra-
cellular taurine con centration. In contrast to active taurine
uptake, swelling-induced taurine release is significantly
reduced fo llowing preincub ation w ith T NFa (16 h) but
unaffected by high extracellular taurine concentration
(24 h). Thus, in NIH3T3 cells, (a) active taurine uptake
reflects T auT expression; (b) TauT activity is modulated by
multiple stimuli, both acutely, and at the level of TauT
expression; (c) the subcellular localization of T auT is regu-
lated; and (d) volume -sensitive taurine release is not medi-
ated by TauT.
Keywords:TNFa; creatine; microtubules; reactive oxygen
species; v olume-sensitive t aurine leak pathway.
Taurine, amino ethane s ulfonic a cid, plays an essential role
not only as an organic os molyte and substrate for the
formation of bile salt, but also in the mo dulation of the
cellular, free Ca
2+
concentration and regulation of neuro-
transmission through interaction with GABA- and glycine-
gated Cl
–
channels [1–4]. More recently it has been shown
that taurine, via its r eaction with cellular hypochlorous acid
(HOCl), produces the less toxic taurine chloramine (TauCl)

and thus serves a tissue protective role against oxidative
injury [5,6]. He nce, changes i n the net cellular content of
taurine may have a dramatic impact on cell f unction.
The cellular taurine content is a balance between
synthesis from methionine/cysteine, active uptake via the
saturable, taurine transporter TauT, and release via a
volume-sensitive taurine leak pathway [7]. TauT has a high
affinity and s electivity towards taurine but a low transport
capacity, and active uptake of one molecule of taurine has
been demonstrated to require two to three Na
+
ions and
one Cl
–
ion [7]. TauT i s a member of the neurotransmitter
transporter family that includes the transporters for
serotonin (SEROT), c-amino butyric acid ( GAT1-3) as
well as the creatine transporter (CREAT) [8]. All members
of this family span the membrane 1 2 times, with t he N- and
C-terminal ends exposed to the c ytosolic compartment. The
cytosolic domains contain several serines, threo nines, and
tyrosines positioned in motifs highly conserved for phos-
phorylation. It was previously shown that NIH3T3 cells
release taurine via a volume-sensitive osmolyte transport
pathway. This pathway differs pharmacologically and
functionally from the volume-sensitive Cl
–
channel by its
sensitivity towards anion channel blockers and kinase
inhibitors, time course for activation and inactivation

following hypotonic exposure, as well as sensitivity towards
expression of constitutively active RhoA [7,9]. However,
little is known about the molecular identity o f the taurine
transporter responsible for the swelling-induced taurine
release.
The promoter region of rat and human taurine
transporter genes, TauT, contain consensus binding sites
for transcription factors p53 and NF-jB and for tonicity
response element binding protein (Ton-EBP) [10–12]. The
activity of p53 is regulated by a series of protein
phosphorylations, acetylations and glycosylations of par-
ticular regulatory domains of p53 [13], and p53 is
up-regulated by moderate hypertonicity that protects re nal
inner medullary co llecting duct cells from apoptosis [14].
TauT expression is down-regulated after activation of p53
in renal cells [15] but up-regulated by p53 in MCF-7
human breast cancer cells [16]. The activity of NF- jBis
typically modulated by interleukines (IL) a nd the cellular
Correspondence to I. H. Lambert, The August Krogh Institute,
Biochemical Department, Universitetsparken 13, DK-2100,
Copenhagen Ø, Denmark. E-mail:
Abbreviations: TauCl, taurine chloramine; HOCl, hypochlorous acid;
SEROT, transporters for serotonin; CREAT, creatine transporter;
Ton-EBP, tonicity respo n se elem en t b inding protein; IL, interleukines;
PKA/C, protein kinase A or C; ROS, reactive oxygen species.
(Received 2 9 July 2 004, revised 30 September 2004,
accepted 6 October 2004)
Eur. J. Biochem. 271, 4646–4658 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04420.x
antioxidant TauCl [5]. The tumour necrosis factor-a
(TNFa) is reported to increase the mRNA level of TauT

in rat brain capillary endothelial cells (TR-BBB13) [12]
and human intestinal epithelial CaCo-2 cells [17] as well as
the taurine uptake i n rat astrocytes [18] and CaCo-2 cells
[17]. Ton-EBP is phosphorylated under hyperton ic condi-
tions, whereupon it translocates to the nucleus, binds
to the tonicity response element (Ton-E) and initiates
transcription of hypertonicity-induced proteins such as
TauT [12].
The present work was initiated to characterize activity,
expression and subcellular localization o f TauT i n N IH3T3
fibroblasts. It i s demonstrated that: ( a) the saturable taurine
transporter TauT is present in NIH3T3 cells and taurine
uptake i s a Na
+
-andCl
–
-dependent process that correlates
with Taut expression; (b) it is possible to modulate the
expression of TauT and taurine transport activity by
multiple stimuli both acutely and at t he TauT expression
level; and (c) TauT does not promote volume-sensitive
taurine release.
Materials and methods
Chemicals
Bovine serum albumin (BSA), H
2
O
2
(1.03
M

), mouse
TNFa,(5lgÆmL
)1
in 1 m gÆmL
)1
BSA), N-methyl-
D
-
glutamine (NMDG), theophylline (40 m
M
, ddH
2
0), forsk-
olin (50 m
M
, 96% ethanol), 4b-phorbol 12-myristate
13-acetate (PMA, 40 l
M
, 96% ethanol), Ortho-phtaldehyde
(OPA), creatine, taurine, b-alanine, sucrose, mouse anti-
acetylated a-tubulin, gentamycin, amphotericin were from
Sigma Chemical (St. Louis, MO, USA). Rabbit anti-rat
TauT against the C-terminal TauT sequence was from
Alpha Diagnostics I nc (San Antonio, TX, USA). Rabbit
anti-human TauT against the N-terminal sequence was
donated by J. Mollerup (Institute of Molecular Biology,
University of Copenhagen). Goat anti-rabbit lactate dehy-
drogenase was from Abcam, Ltd (Cambridge, UK). Alka-
line phosphatase conjugated goat anti-rabbit IgG and
donkey anti-goat IgG were from Jackson Laboratories

(Bar Harbor, ME, USA). Alexa Fluor Ò 488 donkey anti-
rabbit I gG, Alexa Fluor Ò 488 mouse anti-chicken IgG and
Alexa F luorÒ 568 g oat anti-rabbit IgG were from Molecu-
lar Probes (Leiden, the Netherlands). [
14
C]Taurine was from
NEN Life Science Products (Boston, MA, USA). Penicillin,
streptomycin, Dulbecco’s modified Eagle’s medium
(DMEM), foetal bovine serum and trypsin were from
Invitrogen (Taastrup, Denmark).
Media
The phosphate buffered saline ( NaCl/P
i
) contained 137 m
M
NaCl, 2.6 m
M
KCl, 6.5 m
M
Na
2
HPO
4
,and1.5m
M
KH
2
PO
4
. Iso-osmotic NaCl medium contained 143 m

M
NaCl, 5 m
M
KCl, 1 m
M
Na
2
HPO
4
,1m
M
CaCl
2
,0.1m
M
MgSO
4
,5m
M
glucose and 10 m
M
N-2-hydroxyethyl pip-
erazine-N¢-2-ethanesulfonic acid ( Hepes). Iso-osmotic
NMDGCl solution was similar to NaCl with NMDG
being s ubstituted for sodium. Iso-osmotic KCl m edium
contained 150 m
M
KCl, 1.3 m
M
CaCl

2
,0.5m
M
MgCl
2
,and
10 m
M
Hepes. Hypo-osmotic NaCl/KCl solution were
prepared by reduction of the NaCl/KCl concentration in
the iso-osmotic solutions to 95 m
M
, with reducing t he other
components. Hypertonic NaCl medium w as prepared from
isotonic NaCl medium by supplementation with 100 m
M
sucrose. In all solutions, pH was adjusted at 7.40. Unless
otherwise noted, experiments were carried out room tem-
perature (typically 18–22 °C).
Cell culture
Measurements were performed on Swiss NIH3T3 mouse
fibroblasts (clone 7 obtained from B. M. Willumsen,
Institute of Molecular Biology, University of Copenhagen,
Denmark) and mouse myoblast (C2C12, American Type
Culture Collection, Manassas, VA, USA). F ibroblasts were
cultured in DMEM supplemented with 10% (v/v) heat
inactivated fetal bovine serum and 1% (w/v) penicillin/
streptomycin. C2C12 were cultured in DMEM with 10%
foetal bovine serum supplemented with 2 0 lgÆmL
)1

genta-
mycin, 3 lgÆmL
)1
amphotericin B, 100 lgÆmL
)1
strepto-
mycin sulfate and 100 IUÆmL
)1
penicillin. Cells were grown
at 37 °C, 5% CO
2and
95% humidity. Cell cultures were
passaged every 3–4 days by trypsinatio n (0.5%), and only
passages 10–30 were used for experiments.
Estimation of the taurine influx
NIH3T3 cells were grown to 80% confluence in six-well
polyethylene dishes (9.6 cm
2
per well). The c ells were
washed three times by gentle aspiration/addition of 1 mL
experimental solution. Following the final w ash, the cells
in five of the wells were exposed to isotonic NaCl
medium containing [
14
C]taurine ( 2.54 · 10
11
c.p.m.Æmmol
)1
,
1.4 l

M
). The sixth well had added isotope-free NaCl
medium and was used for estimation of the average protein
content (g protein per well) by the Lowry method [19] using
BSA as standard. At g iven time points (4–20 min) taurine
uptake was terminated by removal of the extracellular
medium, r apid add ition/aspiration of 1 mL i ce-cold M gCl
2
(100 m
M
), followed by cell lysis with 500 lL 96% (v/v)
ethanol. The ethanol was blown off and the cellular
[
14
C]taurine activity extracted by addition of 1 ml ddH
2
O
(1 h ), which was transferre d to a scintillation vial for
estimation of
14
C activity (b-scintillation counting, Ultima
Gold
TM
). The wells were washed twice with ddH
2
O. The
total [
14
C]taurine (c.p.m.) taken up by the cells in well one to
five w as in each case estimated a s the sum o f

14
C a ctivity in
the cell extract and water washouts. The taurine uptake
(nmolÆgprotein
)1
) at a given time point is calculated from
the ce llular [
14
C]taurine activity, the extra ce llular specific
activity and the protein content. The taurine influx (nmolÆ
gprotein
)1
Æmin
)1
) w as estimated as the slop e o f the cellular
taurine uptake plotted vs. time.
Estimation of the taurine efflux
Taurine efflux measurements were performed at room
temperature as described previously [20]. Briefly, NIH3T3
cells were grown to 80% confluence in six-well polyethy-
lene dishes (9.6 cm
2
per well) and loaded for 2 h in 1 ml
DMEM containing [
14
C]taurine (80 nCiÆmL
)1
). The pre-
incubation solution was aspirated and the cells washed
Ó FEBS 2004 TauT expression, activity and subcellular localization (Eur. J. Biochem. 271) 4647

five times with 1 ml isosmotic solution in order to
remove exce ss extracellular [
14
C]taurine a nd cellular
debris. The efflux was initiated after the final wash by
addition of one ml of experimental solution. The cells
were left for 2 min and then the medium was transferred
to scintillation vial and rapidly substituted by 1 mL fresh
medium. This procedure was carried out for 20 min.
Cells were lysed at the end of the experiment by addition
of 1 ml NaOH (0.5 m
M
). The total
14
C activity in the
cell system was estimated a s the sum of
14
C activity
(b-scintillation counting, Ultima Gold
TM
) in all the efflux
samples, the NaOH lysate plus the two final w ash outs
with ddH
2
O. The natural logarithm to the fraction of
14
C activity remaining in the fibroblast was plotted vs.
time, and the rate constant for the taurine e fflux (min
)1
)

at each time point was estimated as the negative slope of
the graph between the time point and the proceeding
time point. The taurine efflux at a g iven time point can
be estimated as t he product o f the rate constant and the
cellular taurine pool. I n the case where T auT activity w as
down-regulated and preloading of the cells was imposs-
ible, th e activity of the swelling-induced taurine transport
was followed as influx in isotonic or hypotonic N a
+
-free,
KCl media over 20 min.
Estimation of the cellular amino acid content
The amino acid content was estimated by OPA deriva-
tization followed by reversed-phase HPLC separation
(Gilson: 322-Pump, 2 34-Autoinjector, 155-UV/VIS detec-
tion, BioLab, Aarhus, Denmark). Cells, grown at 80%
confluence (75 cm
2
flasks) were washed three times with
NaCl/P
i
medium, the medium was a spirated and the cells
lyzed/deproteinized subsequently by addition of 1.2 mL
4% (v/v) sulfosalisylic acid. The cell homogenate was
transferred to eppendorf tubes and sonicated (2· 10 s) on
ice. Some suspension (200 lL) was d enaturated overnight
after dilution with 200 lLNaOH(2
M
) and used for
estimation of the protein content using the Lowry

procedure [19] and BSA as p rotein standard. The residual
1000 lL of the suspension was centrifuged (20 000 g,
10 min) and the supernatant filtered (Milex-GV, 0.45 lm).
The amino acids in the filtered sample were separated on
a Nucleosil column (Macherey-Nagel, D
€
uren, Germany,
C18, 250/4, 5 l
M
) using a 1 ml per min flow rate and
increasing the acetonitrile fraction in a 1 2.5 m
M
phosphate
buffer (pH 7.2) from 0% to 25% within the initial 24 min
and from 25% to 50% within the subsequent five min.
UV absorption (330 nm) of the samples and taurine
standard (0.1 m
M
) were used for estimation of the taurine
content in the samples. The cellular amino acid content
(lmolÆgprotein
)1
) was estimated from t he amino acid and
protein content in the flask.
Fractionation, SDS/PAGE and Western blotting
NIH3T3 cells grown to 80% confluence i n Petri dishes were
washed quickly in ice-cold NaCl/P
i
,treatedwith100lL
lysis buffer [150 m

M
NaCl, 20 m
M
Hepes, 10% (v/v)
glycerine, 1% (v/v) Triton X-100,1 m
M
EDTA, 1 m
M
NaF and 1 m
M
Na
3
VO
4
] and 1% (w/v) SDS, scraped off
with a rubber policeman and processed 10 times through a
27 gauge needle. For cell fractionation, fibroblasts were
lysed in lysis buffer containing 0.5% (v/v) T riton X-100 and
no SDS. The cell lysate was then centrifuged at 600 g fo r
10 min (4 °C) to give the nuclear fraction (pellet) and the
supernatant was centrifuged at 40 000 g for 1 h (4 °C) to
give the membrane fraction (pellet) and the cytosolic
fraction (supernatant). The two p ellets were washed once
in lysis buffer containing 0.5% (v/v) Triton X-100 and
resuspended in 60 lL lysis buffer containing 1% (v/v)
Triton X-100 and SDS. The protein concentrations were
estimated using a BCA protein k it (Pierce, BB Gruppen,
Denmark). Proteins were resolved by gel electrophoresis
under denaturing and reducing conditions and electropho-
retically transferred to a nitrocellulose membrane (Invitro-

gen) as previously described [21]. The membranes were
incubated with either rabbit anti-rat TauT (1 : 250), rabbit
anti-human TauT (1 : 250) or goat anti-rabbit lactate
dehydrogenase (1 : 800) Igs at room temperature for 2 h
or overnight a t 4 °C followed b y identification wi th species-
specific alkaline phosphatase-coupled secondary antibodies
[1 : 1200 (TauT); 1 : 900 (lactate dehydrogenase)] and
development with BCIP/NBT (Kirkegaard and Perry
Laboratories, Gaithersburg, MA, USA).
Immunocytochemistry
NIH 3T3 cells grown on glass coverslips in six-well test
plates (Nunc, Rosklide, Denmark) were fixed in 4% (v/v)
paraformaldehyde, permeabilized in 0.2% (v/v) Triton X-
100, quenched in NaCl/P
i
with 2% (w/v) BSA and
incubated with primary antibodies for 2 h at room temper-
ature or overnight at 4 °C: anti-acetylated a-tubulin
(1 : 400), rabbit anti-rat TauT (1 : 100) or rabbit anti-
human TauT (1 : 100) Igs. Cells were washed in NaCl/P
i
and incubated with 4,6-diamidino-2-phenylindole (DAPI)
(1 : 100), Alexa FluorÒ 488 donkey anti-rabbit IgG
(1 : 200), Alexa FluorÒ 488 mouse anti-chicken IgG
(1 : 600), and/or Alexa FluorÒ 568 goat anti-rabbit
IgG (1 : 600). For epifluorescence studies, fluorescence
was visualized on either a Microphot-FXA microscope
with EPI-FL3 filter (Nikon, DFA A/s, Copenhagen,
Denmark). Confocal microcopy was performed using a
Leica DM IRB/E microscope coupled to a Leica TSC

NT confocal laser scanning unit (Leica Lasertechnik
GmbH, Heidelberg, Germany). Excitation of DAPI and
Alexa F luorÒ 488 was carried out using the 364 nm UV
laser-line and the 488 nm argon/krypton laser-line,
respectively. Emission wavelengths, Photo Multiplyer
Tube (PMT) and laser intensity settings were optimized
to minimize bleed-through, and to set fluorescence
detected from preparations labelled with secondary
antibody only to zero. Images were taken using a 40·/
1.25 NA planapochromat objective, a 0.75 Airy disc
pinhole, and an optical slice thickness of 0.25 lm. Images
(512
2
pixels) were frame averaged and presented in
pseudocolour. Digital images were enhanced by Adobe
PHOTOSHOP
Ò
6.0.
Data and statistical analysis
The neural network algorithm, NetPhos 2.0 [22], manual
homology searches as well a s searches in the ELM database
4648 J. W. Voss et al. (Eur. J. Biochem. 271) Ó FEBS 2004
( were used to predict serines, threonines
and tyrosines in the intracellular domains of TauT suitable
for phosphorylation. Data are presented either as individual
experiments, representative of at least three independent
experiments, or as mean values ± SEM, n indicates the
number of i ndependent experiments. Statis tical s ignificance
was estimated by the Student’s t-test. For all statistical
evaluations, P-values < 0.05 were taken to indicate a

significant difference.
Results
TauT in NIH3T3 fibroblasts – affinity, substrate specificity
and ion dependency
TauT in most cell systems has a high affinity towards
taurine but a l ow transport capacity [7]. T he traces in Fig. 1
demonstrate that active taurine uptake in NIH3T3 mouse
fibroblasts is linear within the initial 20 min following the
addition of
14
C-labelled taurine. Using taurine uptake
within the initial 20 min in NaCl medium containing
extracellular taurine (0–50 l
M
) and fitting the uptake data
to a M ichaelis–Me nten equation revealed that the K
m
value
in NIH3T3 cells, i.e. t he extracellular taurine concentration
required for half maximal taurine uptake, is 18 ± 1 l
M
(n ¼ 3). It is also recognized that active taurine uptake via
TauT is Na
+
-andCl
–
-dependent in various cell systems and
that only close analogues t o taurine (such a s b-alanine) are
potential inhibitors of the active uptake [7]. The data shown
in Fig. 1 and summarized in Table 1 indicate that the i nitial

taurine uptake is reduced in the presence of b-alanine
(Fig. 1A) and following substitution of extracellular K
+
for
Na
+
or extracellular N O
3
–
for Cl
–
(Fig. 1B). From Fig. 1A
and T able 1 it i s also seen t hat t aurine uptake is r educed to
about 75% in the presence of 5 m
M
creatine. Creatine
(a-methylguanido acetic acid) is accumulated by muscle
cells via the active creatine transporter CreaT, which has a
structure similar to that o f TauT and exhibits the same
requirement for Na
+
and Cl
–
for initiation of active uptake
of creatine [23]. However, Western blotting using a
polyclonal antibody raised against the human creatine
transporter (Research Diagnostics Inc, Flanders, NJ USA)
and C2C12 myoblasts as a positive c ontrol, indicated that
CreaT is apparently ab sent in NIH3T3 cells (n ¼ 3, data
not shown). C reatine resembles GABA (c-amino butyric

acid), which reduces the active taurine uptake in, e.g.
Ehrlich ascites tumour cells [24], and it is possible that
creatine binds to TauT with low affinity and competitively
reduces active taurine uptake. This was not investigated
further. From Fig. 2 it is seen that active taurine uptake in
NIH3T3 cells is a sigmoidal function of the extracellular
Na
+
concentration ([Na
+
]
o
) and a hyperbolic function of
the extracellular Cl
–
concentration ([Cl
–
]
o
). Fitting these
uptake data to Hill type equations it is estimated that 2.5
Na
+
and 1 Cl
–
ions are required for initiation of the uptake
of one taurine molecule (Fig. 2). From the data presented in
Figs 1–2 and Table 1 it is suggested that taurine uptake in
NIH3T3 cells is mediated by a system that exhibits
characteristics typical for TauT in other cell types, i.e.

NIH3T3 TauT has a high specificity and a ffinity towards
taurine a nd 2.5 Na
+
and 1 Cl
–
are i nvolved in the uptake
of one taurine.
Acute regulation of TauT transport activity
by PKC, PKA and H
2
O
2
TauT possesses several cytosolic serine and threonine
residues that are potential targets for protein kinase A and
C (PKA, PKC) mediated phosphorylation [7,25]. From
Fig. 3 it is seen that acute exposure e ither to PMA to
stimulate PKC, or t o forskolin plus theophylline t o increase
cellular cAMP level and thus presumably stimulate PKA,
results in a reduced taurine uptake. Exposing the NIH3T3
cells to H
2
O
2
also reduces the taurine uptake (Fig. 3).
Reactive oxygen species (ROS) exhibits a variety of
physiological effects [7] and H
2
O
2
, which is highly cell

permeable, is reported to elevate the phosphorylation of
tyrosine residues, most probably via inhibition of a p rotein
Fig. 1. Substrate specificity and ion requirement of TauT. NIH3T3 cells, grown at 80% confluence, were exposed to isotonic N aCl medium
containing [
14
C]taurine (2.54 · 10
11
c.p.m.Æmmo l
)1
,1.4l
M
). At time points 4–20 min taurine uptake was terminated by removal of the extra-
cellular medium and the cellular [
14
C]taurine was extracted. The cellular taurine uptake (nmolÆgprotein
)1
)atagiventimepointwascalculatedfrom
the cellular [
14
C]taurine activity, the extra cellular specific activity and the protein content. (A) Taurine uptake was followed in the abse nce (control
cells) and in t he presence of 5 m
M
creatine or 5 m
M
b-alanine. ( B) Taurine uptake w as fo llowed i n control cells e xposed to NaCl medium and in cells
incubated with N a
+
-free KCl m edium or Cl
–
free NaNO

3
medium. T he cu rve s in A and B are all r e presentative o f three inde penden t s ets o f
experiments.
Ó FEBS 2004 TauT expression, activity and subcellular localization (Eur. J. Biochem. 271) 4649
tyrosine phosphatase [26]. Thus, t hese data could indicate
that an increased phosphorylation of TauT or a putative
regulator of TauT could be involved in the H
2
O
2
-induced
reduction in the active taurine uptake in NIH3T3 cells.
Expression and subcellular localization of TauT
The p romoter r egion of the TauT gene in human [10] and
rat [11] contains consensus binding sites for the transcription
factors p 53 and NF-jB. The cellular expression of p 53 is
reported to be increased following serum-starvation [27] or
exposure to hypertonic conditions [14], whereas NF-jB
activity is typically modulated by IL (TNFa) . Active taurine
uptake in NIH3T3 cells is significantly increased by 11%
and 17% following exposure to TNFa (2 0 ngÆmL
)1
)for
16 h or serum starvation for 24 h, respectively (Fig. 4).
Epifluorescence microscopy analysis shows that a poly-
clonal TauT antibody, raised against the C-terminal
sequence of the rat TauT, localizes to a region of NIH3T3
control cells, which appears to be within the nucleus, as well
as in the cytosol and at the plasma membrane (Fig. 5A, row
1; TauT red colour, microtubules green colour). Following

TNFa treatment, TauT immunolocalization is augmented,
particularly in the nucleus and the perinuclear area
(Fig. 5A, row 2). Thus, the data in Figs 4 and 5 indicate a
correlation between taurine transport and total cellular
TauT expression. To verify the apparent cytoplasmic and
nuclear localization of TauT, we employed confocal laser
scanning microscopy of TauT and DAPI-stained NIH3T3
cells. From t he confocal visualization s tudies in Fig. 5B it is
seen that TauT (green colour) appears throughout the
nuclear compartment. It is emphasized that the p inhole size
was kept s mall enough to exclude that staining observed in
these c ompartments is caused by out-of-focus fluorescence
from transporters localized, e.g. to the p lasma m embrane.
As controls, a similar pattern of TauT localization was
observed with an antibody raised against the N-terminal
part of TauT, and immunolocalization of the antibody
raised against the C-terminal part of TauT was abolished by
a specific blocking peptide to t his antibody (data not
shown).
Besides being regulated by p53 and NF-jB, TauT
expression and TauT activity are also sensitive to exogenous
taurine [28]. From Fig. 6A it can be seen that polyclonal
antibody raised against the C-terminal sequence of the rat
TauT, recognizes a m ajor protein band at 67 kDa in lysates
of whole NIH3T3 cells, corresponding to the molecular
mass of TauT. The expression of this protein is reduced
following 24 h exposure to DMEM supplemented with
1m
M
or 100 m

M
exogenous taurine, and i ncreased follow-
Fig. 2. N a
+
/Cl
–
/taurine stoichiometry for taurine uptake v ia TauT . N IH3T3 cells, grown at 80% co nfluen ce, were exposed to isotonic mediu m
containing [
14
C]taurine ( 2.54 · 10
11
c.p.m.Æmmol
)1
,1.4l
M
) f or 20 min. Taurine uptake w as terminated and the cellular [
14
C]taurine (c.p.m. per
20 min) estimated. (A) The extracellular Na
+
concentration was varied between 0 and 150 m
M
adjusting the concentration to 150 m
M
with
NMDG. The influx w as plotted vs. the extracellular Na
+
concentration ([Na
+
]) and fitted to the Hill type equation: Y ¼ (V

max
[Na
+
]
n
)/
((K
Na
)
n
+[Na
+
]
n
), where V
max
is the maximal uptake, K
Na
is the Na
+
concentration required f or half maximal u ptake a nd n is the number of ions
required for initiation of uptake of one taurine. (B) The extracellular Cl
–
concentration was varied between 0 and 150 m
M
adjusting the con-
centration to 150 m
M
with NaNO
3

. The influx was plotted vs. the extracellular Cl
–
concentration ([Cl
–
]) and fitted to the Hill type equation: Y ¼
Y
o
+(V
max
[Cl
–
]
n
)/((K
Cl
)
n
+[Cl
–
]
n
), where Y
o
is taurine uptake in the absence of Cl
–
,andK
Cl
is the Cl
–
concentration required for half maximal

uptake. The stoichiometry values i ndic ated on the fi gure were estimated for N a
+
and Cl
–
in five a nd three sets of experiments, respectively.
Table 1. TauT substrate specificity a nd ion d ependency. Taurine u ptake
was estimat ed a s in dicated i n F ig. 1 in the absence or presence of 5 m
M
creatine/b-alanin e and in Na
+
-free NMDG-medium, KCl medium or
Cl
–
free NaNO
3
medium. The absolute tau rine influx (nmolÆgpro-
tein
)1
Æmin
)1
) w as estimated by linear regression, u sing values in the
time frame 4–20 min . Taurine uptake from three sets of paired
experiments is given relative to control values ± SEM. P indicates the
level of significance in a paired Student’s t-test against the control
value.
Taurine influx P
Substrate specificity
Control 1
Creatine 0.730 ± 0.027 0.005
b-alanine 0.007 ± 0.001 <0.001

Ion dependency
Control, NaCl 1
NaNO
3
0.116 ± 0.018 0.01
KCl 0.003 ± 0.001 <0.001
NMDG 0.001 ± 0.001 <0.001
4650 J. W. Voss et al. (Eur. J. Biochem. 271) Ó FEBS 2004
ing exposure to DMEM supplemented with 100 m
M
sucrose. Similar data where obtained using the antibody
raised against the N-terminal sequence of T auT, confirming
antibody specificity (data not shown). Pre-exposure to
exogenous taurine for 24 h is accompanied by a reduction in
the initial taurine uptake, whereas the taurine uptake is
significantly increased following pre-exposure to sucrose
(Fig. 6B). It is noted that the reduced taurine uptake
following exposure to 100 m
M
taurine is not secondary to
cell shrinkage induced by the hypertonic conditions, as
evidenced by the effec t of addition of 100 m
M
sucrose to
increase extracellular osmolarity to the same extent. The
taurine-induced down regulation of the taurine uptake in
NIH3T3 cells is reversible. This is seen from Fig. 6C where
it is shown that in NIH3T3 cells, pre-exposed to 100 m
M
taurine for 24 h normal transport capacity is g radually

regained and has returned to control values within 24 h. It is
noted that TauT activity in cells exposed to DMEM
supplemented with 100 m
M
sucrose for 48 h is slightly
increased compared to that in control cells exposed to
DMEM, indicating that TauT activity in NIH3T3 cells is
not affected by long-term hypertonic exposure. Figure 7A
shows that the immunofluorescence intensity of a ntibody
against the C-terminal sequence of the rat TauT is
dramatically reduced following long-term exposure to high
concentrations of taurine (compare 1st and 2nd row) and
increased following long-term exposure to high s ucrose (3rd
row), indicating that the variations in the TauT transport
activity in Fig. 6B reflect TauT expression. The fluorescence
intensity of the cells shown in Fig. 7 is enhanced by
electronic manipulation (to the same extent for all p anels,
i.e. maintaining the same relative intensity) in comparison to
those in Fig. 5. This was done in order to facilitate the
visualization of the inhibitory effect of the taurine, which
was so marked, that there was no or very little visible
labelling in the taurine-treated cells in the nonenhanced
images. The fluorescence images presented in Fig. 7B
(frames 0 –12 h ) show the t ime-dependent effect of sucrose
on the level of TauT expression in cells pre-exposed to h igh
concentrations of taurine for 24 h. It is seen that the level of
TauT increases at about 4 h after sucrose was su bstituted
for taurine in the incubation medium. Thus, in accordance
with the data in Figs 4 and 5 the active taurine uptake
correlates with TauT expression.

In order to confirm changes in subcellular level of TauT
expression and TauT localization upon supplementation of
taurine and sucrose, we performed Western blotting analysis
of TauT expression in cytosolic, nuclear and membrane
fractions of NIH3T3 cells. The cytosolic protein lactate
dehydrogenase w as used as a c ontrol to ensure that nuclear
and membrane fractions were not contaminated with
cytosolic TauT. The cell fractionation experiment presen ted
in Fig. 8 confirms the presence of a major 67 kDa TauT
protein in whole NIH3T3 fibroblasts, a nd that this pro tein
is mainly localized to the cytosolic fraction. In contrast, we
find that TauT in the nuclear and membrane fractions
mainly appears as a 90 kDa protein. The intensity of both
protein bands is reduced in NIH3T3 cells exposed to
100 m
M
extracellular taurine compared to cells exposed to
100 m
M
sucrose. In particular, the protein level of the
nuclear and membrane-associated 90 kDa forms of TauT is
down-regulated at least eight and three times, respectively,
in the taurine supplemented cells. Thus, the variation of
nuclear TauT expression, observed by immunofluorescence
microscopy analysis upon taurine and sucrose sup plemen-
tation (Fig. 7), probably reflects variation in the level of the
90 kDa TauT protein. Three TauT protein bands in the
range 50–70 kDa have been previously demonstrated by
immune blotting in Ehrlich ascites tumour cells and it has
been suggested that the band in the Ehrlich cells with the

highest apparent molecular mass represents a phosphoryl-
ated form of TauT [29]. Whether the variance in the
Fig. 4. Augmentation of the taurine uptake by long-term exposure to
TNFa or serum-free conditions. NIH3T3cellsweregrowninDMEM
(control), DMEM containing TNFa (20 ng ÆmL
)1
, 16 h) or serum-free
DMEM (24 h, serum-starved) before initiation o f the influx e xperi-
ment. Taurine up take was followed with time using [
14
C]taurine as
indicated in Fig. 1. The influx, estimated from the slope of the uptake
curves, was estimated and in ea ch case giv en r elative to t he influx i n
control cells ± SEM (n ¼ 3). # , Significantly different f rom t he control
(P <0.05).
Fig. 3. Modulation of taurine uptake by phosphorylation. NIH3T3 cells
were grown to 80% confluence. The cellular taurine uptake was fol-
lowed with t ime in isotonic N aCl m edium using [
14
C]taurine as indi-
cated in Fig. 1. PM A (50 n
M
), forskolin (10 l
M
) plus theoph ylline
(0.5 m
M
), an d H
2
O

2
(2 m
M
) w ere i nclude d i n t he experimental medium
from the time of the initiation of the influx experiment. The influx was
estimated f rom the slop e of t he uptake cu rves and i n each case given
relative to the infl ux in c ontrol cells ± SEM (n ¼ 3). #, Significantly
different from the c ontrol (P < 0.05).
Ó FEBS 2004 TauT expression, activity and subcellular localization (Eur. J. Biochem. 271) 4651
molecular mass of TauT from the nuclear/membrane and
the cytosolic fractions in NIH3T3 fibroblasts reflects
differences in post-transcriptional modifications such as
glycosylation and/or phosphorylation or expression of
different TauT isoforms was not investigated further.
We fu rther observed that TauT in sucrose-treated cells
strongly localizes in a punctuate pattern along the c ytosolic
network o f acetylated microtubules (Fig. 9), indicating that
localization and functional targeting of TauT to subcellular
domains, such a s the plasma membrane, may be coupled to
microtubules-associated carrier vesicles.
Effect of long-term exposure to TNFa and exogenous
taurine on the volume-sensitive taurine leak pathway
The experiments in Fig. 10 were performed in order to
evaluate whether up/down r egulation of T auT transport
activity is paralleled by a similar up/down regulation of t he
activity of the volume s ensitive taurine efflux pathway. It is
seen that the rate constant for taurine release is increased
transiently following hypotonic exposure and that the
maximal rate constant is reduced following pre-exposure
to TNFa (Fig. 10A). It is estimated that pre-exposure to

20 ng TNFaÆmL
)1
for 16 and 48 h reduces the maximal rate
constant for the swelling-induced taurine release from
NIH3T3 cells to 75% and 60% of the control value,
respectively (Fig. 10B). A 16 h exposure to TNFa
also reduces the rate constant for taurine release from
NIH3T3 cells under isotonic conditions by 30%, i.e. the
rate constant in three sets of experiments was e stimated
at 0.0020 ± 0.00005 min
)1
(control cells) and 0.0014 ±
0.00005 min
)1
(20 ng TNFaÆmL
)1
,16h;P ¼ 0.005). T he
cellular taurine content was, in three sets of separate
experiments, estimated at 0.031 ± 0.003 and 0.025 ±
0.003 lmolÆgprotein
)1
in control cells and cells treated
with TNFa, respectively. Pasantes-Morales and coworkers
[30] have similarly estimated the cellular taurine content in
NIH3T3 cells under control conditions at 0.052 lmolÆg
protein
)1
. As the taurine efflux at a given time point is equal
to the r ate constant times the cellular pool it is estimated
Fig. 5. Modulation of subcellular localization and expression of TauT. (A) NIH3T3 cells, grown on cover slips at 60–70% confluence, were grown in

DMEM (control cells, 1st row) or DMEM plus TNFa (20 n gÆmL
)1
, 16 h , 2nd ro w). Cells were fi xed in paraformaldehyde (4%, 15 min) and
permeabilized with Triton X-100 ( 0.2%, 10 min ). T he nucleus (blue) was visualized wit h DAPI. T he microtubules (green) were marked with a
primary antibody against acetylated a-tubuline and visualized with Alexa 488. TauT (red) was marked with a primary antibody raised against the
C-terminus of TauT and visualized with Alexa 568. The merged column (3rd column) is an overlay of the 1st column (nucleus, microtubules) and
the 2nd column (TauT). Each image is representative of at least t hree images from separate experiments. (B) Cells were incubated in isotonic
medium und er co ntrol c o nditio ns, fixed , a nd TauT and nuclei w ere l abelled a s described in Materials and methods. Cells were vi ewed using a
40·/1.25 NA planapochromat objective on a Leica DM IRB/E microscope coupled to a Leica TSC NT confocal laser scanning unit. Visualization
of TauT (green) was carried out by excitation of DAPI and Alexa FluorÒ 488 using t he 364 nm UV laser-line and the 4 88 nm ar gon/krypton laser-
line, respectively. Emission wavelengths, PMT and laser intensity se ttings were optim ized to minimize bleed-through, and to set fluore scence
detected from preparations labelled with secondary antibody only to zero. Images were taken at a 0.75 Airy disc pinhole, and an o ptical slic e
thickness of 0.25 lm, frame averaged and prese nte d i n p seudocolo ur. Im ages shown w ith a re o ptical s lices taken at 0 .75 lm intervals, moving
towards the bottom of t he cell from left to right. Th e experiment shown is r epresen tative of three independent experiments.
4652 J. W. Voss et al. (Eur. J. Biochem. 271) Ó FEBS 2004
that 16 h e xposure to TNFa reduces the maximal taurine
efflux under hypotonic conditions by 40%. Thus, TNFa
exposure has opposing effects on the activity of TauT and
the v olume-sensitive taurine leak pathway i n N IH3T3 cells.
Serum starvation increased the maximal rate constant for
the s welling-induced taurine influx in hypoton ic media (200
mOsm) by 20 ± 7% (n ¼ 4).
Figure 10C shows that a n increase in taurine transport
via the volume-sensitive taurine leak pathway following
reduction in the extracellular tonicity can be demonstrated
as an increase in taurine uptake in Na
+
-free hypotonic KCl
medium. This technique was used as down regulation of
TauT prevents the [

14
C]taurine preloading of the cells
required for the standard efflux procedure. Exposing
NIH3T3 cells, preincubated for 24 h with DMEM medium
containing 100 m
M
sucrose or 100 m
M
taurine to hypotonic
KCl medium, resulted in an increase in the taurine uptake
which is similar o r slightly larger than the influx seen in
NIH3T3 cells pre-exposed to DMEM alone (Fig. 10D).
The l atter is most probably a consequence of the fact that
cells pre-exposed to hypertonic conditions experience a
more dramatic reduction in the extracellular osmolarity
when exposed to the hypotonic conditions. Thus, the
volume-sensitive taurine r elease seems to not be affected b y
long-term hypertonic e xposure o r t o l ong-term exposure t o
high extracellular taurine concentrations.
Discussion
The organic osmolyte, taurine, is present in high concen-
trations in heart- and skeletal m uscles, brain, kidney and
retina. The cellular level of taurine is a balance between
active uptake via TauT and passive leak via a volume
sensitive pathway. Within recent years it has become evident
that taurine is a multifunctional molecule th at r egulates cell
volume, cellular free Ca
2+
concentration, cellular oxidative
status and interferes w ith cell survival [7,31,32]. The data

presented in Figs 1–3 and 5,6 and Table 1 indicate that
TauT is present in the NIH3T3 mouse fibroblasts and
exhibits typical characteristics of mammalian TauT, i.e.
TauT has a high affinity/specificity for taurine, the uptake of
one molecule of taurine via TauT involves 2 Na
+
and 1 Cl
–
,
and the TauT transport activity is reduced following acute
stimulation of PKC and PKA. Stimulation of PKC and
PKA has n o detectable effect on the e xpression of TauT in
NIH3T3 fibroblasts (data not shown) and most probably
does not involve modulation of the transcriptional rate of
TauT. Taurine release from NIH3T3 cells is increased
Fig. 6. Reversibility of substrate-induced down regulation of TauT activity. (A) NIH3T3 fibroblasts were grown in DMEM (control), DMEM plus
1m
M
/100 m
M
taurine or DMEM plus 100 m
M
sucrose for 24 h. The cells were lyzed, sonicated and proteins separated by SDS/PAGE (10%) and
visualized with Western blotting using a primary antibody raised against th e C-terminus of TauT and a secondary, alkaline, phosphatase
conjugated antibody. The bands represent a 67 kDa protein. The blot is representative three se ts of paired experiments. (B) Cells were pretreated
with taurine and sucrose as indicated in panel A. Cells were washed five times prior to the initiation of the influx experiments, in order to remove
excess unlabelled taurine, and t aurine uptake was f ollowed w ith t ime using [
14
C]taurine as outlined in Fig. 1. In o rder to p reserve the tonicity during
the washing procedure and the influx experiment we used isotonic standard NaCl medium for control cells and cells pretreated with 1 m

M
taurine,
and standard N aCl medium supplemented with 100 m
M
sucrose for cells treated with 1 00 m
M
taurine or 100 m
M
sucrose. The i n flux, estimated
from the slope of the uptake curves, is in e ach case given relative to the influx in control cells ± SEM (n ¼ 3). (C) Cells, grown in DMEM
supplemented with 100 m
M
taurine for 24 h, were washed and incubated for a nother 2, 4, 8, 12, 2 4 h in DMEM supplemented with 100 m
M
sucrose. Cells were grown 48 h in DMEM (control cells) and DMEM supplemented with 100 m
M
sucrose, respective ly. Taurine uptake was
followed with time (4.20 min) in isoton ic NaCl me dium (control) or s tandard NaCl m edium s upplemented with 100 m
M
sucrose (Tau/Suc, Sucrose)
and the influx estimated by linear regression as indicated above. Values for 24 h taurine plus 24 h suc rose and for 48 h sucrose treatment are given
relative to the isotonic c ontrol ± S EM, represe nting five and t hree in depende nt sets of e x perimen ts. Valu es for c ells incub ated f or 24 h with taurine
followed by 2 to 12 h incubation in NaCl medium supplemented with sucrose are given relative to the isotonic control and represents the mean of
two sets o f experiments. #, S ignific antly different from c ontrol (P <0.05).
Ó FEBS 2004 TauT expression, activity and subcellular localization (Eur. J. Biochem. 271) 4653
following osmotic cell swelling (Fig. 10) and it has previ-
ously been demonstrated that the swelling-induced taurine
efflux is via a volume-sensitive taurine leak pathway which is
sensitive to various anion ch annel blockers but different
from the swelling-induced Cl

–
efflux pathway [7,25].
Role of reactive species in the regulation
of taurine transport
From the data in Figs 3 and 4 it is seen that the active
taurine t ransport in NIH3T3 i s reduced following exposure
to H
2
O
2
and increased following preincubation with TNFa.
In the case of TNFa we also observed an increased
expression of TauT in NIH3T3 cells (Fig. 5). It has been
demonstrated recently that that e xposure to H
2
O
2
increases
taurine release from NIH3T3 cells following hypotonic
incubation and it w as suggested that this effect reflected an
inhibition of a p rotein tyrosine phosphatase [20]. Several of
the serines, t hreonines and tyrosin es in the intracellular
domains of TauT are situated in motifs highly suitable as
targets for protein kinases. It is possible that a shift of TauT
to a m ore tyrosine phosphorylated state i n congruence with
Fig. 7. Reversibility of substrate-induced down regulation of TauT expression. (A) NIH3T3 cells, grown on cover slips at 60–70% confluence, were
exposed to DM EM (control cells, 1st ro w), DMEM p lus 100 m
M
taurine (24 h, 2nd row) or DM EM s upplemen ted w ith 100 m
M

sucrose (24 h, 3rd
row) as indicat ed i n F ig. 6 A,B. C ells were fixed in paraformaldehyde (4%, 15 min) and permeabilized with Triton X-100 ( 0.2%, 10 mi n). T he
nucleus (blue), the microtubu les (green) and TauT (red) were visualized as indicated in the legend to Fig. 5 . The 3rd column is the merge of the 1st
column (nucleus, microtubules)andthe2ndcolumn(TauT).Eachimageisrepresentative of at least 5 images from separate experimental setups.
(B) Cells, grown in DMEM plus 100 m
M
taurine for 24 h, were washed and exposed to DMEM plus 100 m
M
sucroseforthetimeperiodindicated
(Frame 2… 12 h). Cells were fixed and TauT ( red ) detected as indic ated above. Images are re presentative of two s e ts of experiments.
4654 J. W. Voss et al. (Eur. J. Biochem. 271) Ó FEBS 2004
increased serine/threonine phosphorylation of TauT redu-
ces the active taurine uptake following exposure to H
2
O
2
.
Kang and coworkers [12] found that exposing the blood-
barrier TR)BBB13 cells to TNFa (2 0 ngÆmL
)1
)resultedin
an increased mRNA level of the TauT and a concomitant
1.7-fold increase in taurine uptake. The promoter to TauT
contains a b inding site for the transcription factor NF-jB
andactivationofNF-jB involves phosphorylation of the
inhibitory complex IjB, dissociation of the heterodimer
NF-jB(p50,RelA)fromIjB, translocation of NF-jBto
the nucleus and subsequent initiation of gene expression.
HOCl which is a highly reactive oxidant species generated
from H

2
O
2
and Cl
–
, is converted to the less reactive TauCl
by combination with t aurine. I t has been proposed recently
by M iyamoto and coworkers [5] that TauCl is involved in
oxidation of IjB which prevents phosphorylation of IjB
and a ctivation o f NF-jB. Thus, the effect of longterm
exposure to TNFa, and most probably t o H
2
O
2
,ontaurine
uptake could in volve modulation o f the NF- jBmediated
regulation of the transcription of TauT.TNFa is a well-
known inducer of apoptosis. However, Lang and coworkers
[33] have demonstrated previously that taurine uptake in
Jurkat lymphocytes is increased following stimulation with
the apoptosis inducing ligand Fas (CD95) which presum-
ably relieves the apoptotic process. Whether an increased
taurine content in TNFa-treated cells actually counteracts
apopthosis in NIH3T3 cells is under investigation.
High extracellular taurine down-regulates TauT
expression and transport activity
From Figs 6 and 7 i t is seen that exposure to growth
medium supplemented with 100 m
M
taurine reduces the

expression as well as transport a ctivity o f TauT in NIH3T3
Fig. 8. N uc lear localizatio n o f TauT. Cells, g rown in DMEM supplemented with 100 m
M
taurine or 100 m
M
sucrose for 24 h were l yzed , so nicated,
fractionated and proteins f rom the different fractions were separated by SDS/PAGE (10%). TauT was v isualized with a primary antibody raised
against the C-terminus of TauT and a secondary, alkaline, phosphatase conjugated antibody. The cytosolic protein lactate dehydrogenase, used to
exclude c ytosolic contamination of nuclear and membrane fractions, was visualized with goat anti-rabbit lactate dehydrogenase and alkaline
phosphatase conjugated donkey anti-goat. C, whole cell homogenate; Nu, nuclear fraction; Cyt, cytosolic fraction and Mem, membrane fraction.
The gel is representative of three set of experiments.
Fig. 9. M icro tubule association of TauT. (A) NIH 3T3 cells, grown on cover slips at 60–70% con fluence, were wash ed, fixed, pe rmeabiliz ed and
preceded for d etection of TauT and c ytosolic m icrtubules a s i ndicated i n F ig. 7 . TauT w as visualized using a C -terminal antibody raised against the
rat TauT. Images are representative of at least three sets of experiments. Frames A–D represent tubuline system (A: green), TauT (B: red), merged
image ( C), a nd enlargement of the framed section from C (D) with the microtubules and TauT images slightly shifted in o rder to facilitate
visualization of t he TauT/microtubules colocalization.
Ó FEBS 2004 TauT expression, activity and subcellular localization (Eur. J. Biochem. 271) 4655
cells, whereas exposure to growth m edium supplemented
with 100 m
M
sucrose increases the expression as well as
transport activity of TauT. Kang and coworkes [12]
demonstrated accordingly that hypertonic treatment of
TR-BBB13 cells increased the level of TauT mRNA as well
as active taurine uptake, whereas excess taurine conditions
resulted in a reduced level of TauT mRNA and a
concomitant reduction in taurine uptake. Furthermore,
Bitoun and Tappaz [ 28] have demonstrated that increasing
the cellular t aurine content p revented osmolarity-induced
up-regulation of TauT mRNA expression in astrocyte

primary cultures and they suggested that TauT expression is
controlled by a taurine-induced down-regulation and an
osmolarity-induced up-regulation. In this context i t is noted
that Ton-EBP , which transactivates osmo-protective g enes,
is expressed and up-regulated following hypertonic expo-
sure [34,35] and that the cloned promoter region o f TauT ,
has revealed a consensus binding sites f or Ton-EBP [12].
Thus, the osmolarity-induced up-regulation in NIH3T3
cells most probably involves Ton-EBP. More recently it has
been demonstrated that exposure of NIH3T3 cells to
hypertonic sucrose media is accompanied b y an increased
level of p53 in its phosphorylated, active state in the nucleus
(M. B . F riis, C . F riborg, L. Schneider, M B. Nielsen, I. H.
Lambert, S. T. Christensen and E. K. Hoffmann, unpub-
lished observations). Serum-s tarvation is reported t o induce
a similar i ncrease in p53 and phosphorylated p53 in variou s
cells [27] as well as an increase in transport activity o f TauT
in NIH3T3 cells (Fig. 4). Our findings are thus consistent
with, a lthough by no means unique to, the interpretation
that p53 contribu tes to t he hypertonicity-induced effect on
TauT expression and activity. The effect o f high e xtracel-
lular taurine concentration has been reported to involve
aCa
2+
-sensitive interaction of taurine with at least one
unidentified cis-ele ment in the 5¢-flankin g region of the
TauT promoter region [36,37]. An alternative explanation to
the effect of high extracellular taurine on TauT expression
and activity could b e a concomitant i ncreased cellular level
of taurine and TauCl which, as stipulated above, would lead

to oxidation of IjB and consequently impairment of the
NF-jB-mediated regulation of the transcription of TauT.
From Fig. 6C it is seen that the taurine-induced down-
regulation of TauT is reversible and that TauT in NIH3T3
cells regain 50% o f their transport c apacity within 12 h and
full capacity 24 h after removal of extracellular taurine. This
reversibility is reflected in a concomitant increased expres-
sion of TauT (Fig. 7B) and most probably as a result of
restoration of normal cellular taurin level.
Fig. 10 . Effect of long-term exposure to TNFa a nd high e xtracellular taurine on the swelling-induced taurine transport. (A) Cells, grown to 80%
confluence, were incubated for 16 h in DMEM in the absence (control) or presence of 20 ng ÆmL
)1
TNFa.[
14
C]Taurine was included in the
preincubation medium during the last 2 h. The cells were washed and the release of [
14
C]taurine followed with time in NaCl medium with a shift in
osmolarity from 300 mOsm to 200 mOsm at a time 6 m in after the initiat ion of the e fflux experiment. The rate co nstant is plotted vs. time. The flux
curves are representative of three identical sets of experiments. (B) The maximal rate constant for the swelling-induced taurine efflux was estimated
in three sets of experiments for c ontro l cells, cells pre-exposed to T NF a (20 ngÆmL
)1
, 1 6 h) o r cells serum-starved for 2 4 h. Values a re given ±
SEM. (C) NIH3T3 cells, grown at 80% confluence, were exposed to isotonic (300 mOsm) or hypotonic (200 mOsm) , Na
+
-free KCl medium
containing [
14
C]taurine (2.54 · 10
11

c.p.m.Æmmol
)1
,1.4l
M
), and the uptake followed with time. The curves are r epresentative of three sets of
experiments. ( D) Taurine uptake was estimated for cells grown in: (a) DMEM and subsequently exposed to isotonic or hypotonic Na
+
-free KCl
medium; ( b) DMEM supplemente d with 100 m
M
sucrose fo r 24 h and then exposed to hypotonic Na
+
-free KCl medium; and (c) DMEM
supplemented with 100 m
M
taurine for 24 h and then exposed to hypotonic Na
+
-free KCl medium. The taurine uptake was in each case estimated
as the increase in cellular activity (c.p.m.Æmg protein
)1
) in the time frame 4–12 min following the shift to the Na
+
-free KCl medium and g iven as
mean values of three sets of experiments ± SEM.
4656 J. W. Voss et al. (Eur. J. Biochem. 271) Ó FEBS 2004
Subcellular distribution of TauT
Interestingly, TauT seems to localize to both the
cytosolic, plasma membrane and nuclear compartment
of NIH3T3 cells as judged from the immunolocalization
and fractionation studies presented in Figs 5,7 and 8. It

is generally assumed that activity and function of TauT
strongly correlates with its localization to the plasma
membrane in most cell types. However, it is well
recognized that multiple membrane transport proteins,
including the closely related Na
+
-andCl
–
-coupled
neurotransmitter transporters, are regulated by membrane
insertion and retrieval [38]. In fact, in neurons in rat
prefrontal cortex, the norepinephrine transporter was
recently found to localize predominantly to the cytoplasm
with no apparent association to organelles [39]. Thus, the
cytoplasmic localization of TauT noted in the present
study most probably (at least in part) reflects the
presence of an intracellular pool of transporters that
can be mobilized to the plasma membrane upon specific
stimuli. Indeed, we find that TauT, upon reversal of
substrate-induced down regulation of TauT expression
with sucrose, strongly localizes in a punctuated pattern
along the cytosolic network of microtubules (Fig. 9),
indicating that translocation and functional targeting of
TauT to the plasma membrane is coupled to carrier
vesicles transported along the microtub ular network. A
fraction of the TauT proteins in the cytoplasm may also
represent immature TauT present in the ER/Golgi
compartments. The nuclear localization of TauT, the
high molecular mass of nuclear TauT, and its apparent
increase upon TNF-a treatment, is puzzling, and more

studies are needed to determine the phosphorylation state
and the potential role of TauT in the nucleus. However,
it is interesting to note that another ion transport
protein, the organellular chloride channel CLIC4, was
recently shown to translocate to the nucleus upon TNFa
treatment, and proposed to play a role in the cellular
stress response following apoptotic stimuli [40].
Is the swelling-induced taurine efflux via TauT?
Release of t aurine is a characteristic feature of ischemia
and i t h as been suggested th at the efflux of taurine i nvolves
both osmotic stress and a Na
+
-dependent reversal of TauT
[41]. T he opposing effect of TNFa and high extrac ellular
taurine concentration on the active taurine uptake via
TauT and the passive taurine release via the volume-
sensitive leak pathway (Fig. 4 vs. 10A,B; Fig. 6 B v s. 10D)
clearly indicate that TauT is not involved in the s welling-
induced taurine release in NIH3T3. T he molecular identity
of the v olume-sensitive taurine leak pathway is unresolved.
It has been demonstrated recently that phospholipase A
2
(iPLA
2
) and 5-lipoxygenase activity is required for the
swelling-induced ac tivation of the volume-sensitive taurine
leak pathway in N IH3T3 cells and t hat ROS, p roduced b y
a NAD(P)H oxidase complex, could inhibit protein
tyrosine phosphase activity, reduce Src kinase-activity
and thereby potentiate the swelling-induced taurine release

[20,42]. Liu and McHowat [43] have demonstrated that
TNFa interferes with two types of PLA
2
with different
substrate specificity in rat ventricular myocytes, i.e.
TNFa stimulates cytosolic iPLA
2
activity, that prefer-
ably acts on plasmenyl choline and alkylacyl glycerol-
phosphorylcholine, and decreases iPLA
2
activity, that acts
preferentially on phosphatidylcholine. As activation of
iPLA
2
is a permissive, initial upstream event in the
swelling-induced activation of taurine release from
NIH3T3 cells it is conceivable that T NFa might interfere
with the activation and/or expression of iPLA
2
in NIH3T3.
Acknowledgements
This work was supported by the Danish Natio nal Research Council
and ÔFonden af 1870Õ (I.H.L.), The Danish National Research Council
grant 2 1-02-0120 (S.T.C.), and the Carlsbe rg Foundation (S.F.P.). Dr
Nanna K. J ørgensen, D epartment o f Me dical Physiology, U niversity o f
Copenhagen is acknowle dged for t echnical advice i n conn ection with
the fluorescence data. Dorthe Nielsen is gratefully acknowledged for
technical assistance.
References

1. Barakat, L., Wang, D. & Bordey, A. (2002) Carrier-mediated
uptake and release of taurine from Bergmann glia in rat cerebellar
slices. J . Physiol. 541, 7 53–767.
2. Huxtable, R.J. (1992) Physiological actions of taurine. Physiol.
Rev. 72 , 101–163.
3. Huxtable, R.J. (2000) Expanding the circle 1975–99: sulfur bio-
chemistry and insight s on the biological functions of taurine. Adv.
Exp. Med. 483, 1–25.
4. Schaffer, S., Takahashi, K. & Azuma, J. (2000) Role of osmo-
regulation in t he actions of taurine. Amino Acids 19 , 527–546.
5. Miyamoto, Y., Kanayama, A., Inoue, J.I., Konishi, Y.S. &
Shimizu, M. (2003) Taurine is involved in oxidation of I kappa B
alpha at Met45 – N-halogenated taurine and anti-inflammatory
action. In Taurine in the 21st Century (eds Lombardini, J.B.,
Schaffer, S.W. & Azuma, J.) pp. 373–380. Kluwer Academic/
Plenum Publishers, NY, US A.
6. Schuller-Levis, G.B. & Park, E. (2003) Taurine: New implications
foranoldaminoacid.FEMS Microbiol. Lett. 226, 195–202.
7. Lambert, I.H. (2004) Regulation of the cellular content of the
organic osmolyte taurine in mammalian cells. Neurochem. Res. 29,
27–63.
8. Lill, H. & Nelson, N. (1998) Homologies and family relationships
among Na+ /Cl– n eurotransmitter transporters. Methods Enzy-
mol. 296, 4 25–436.
9. Pedersen, S.F., Beisner, K.H., Hougaard, C., Willumsen, B.M.,
Lambert, I.H. & Hoffmann , E.K. (2002) R ho fam ily GTP b inding
proteins are involved in the regulatory Volume decrease process in
NIH3T3 mouse fi bro blasts. J. Ph ysiol. (Lond), 541 , 779–796.
10. An, H.S., Han, H.C., P ark, T., Park, K .K. & Kim, H .W. (2003)
Cloning of human intestin al tau rine t ransporter and production of

polyclonal antibody. Taurine 5: Beginning 21St Century 526,
131–138.
11. Han, X., Budreau, A.M. & Chesney, R.W. (2000) Cloning and
characterization of th e promote r region o f the rat taurine trans-
porter (TauT) gene. Adv. Exp. Med. 483, 97–108.
12. Kang, Y.S., Ohtsuki, S., Takanaga, H., Tomi, M., H osoya, K. &
Terasaki, T . (2002) Regulation of t aurine transport a t the blood–
brain barrier by tumor necrosis factor-alpha taurine and hyper-
tonicity. J. Neurochem. 83, 1188–1195.
13. Oren, M. (1999) Regulation of the p53 tumor suppressor protein.
J. Bi ol. Chem. 274, 36031–36034.
14. Dmitrieva, N., M ichea, L. & B urg, M . (20 01) p53 Protects r enal
inner medullary cells from hypertonic stress by restricting DNA
replication. Am. J . Physiol Renal Ph ysiol. 281, F 522–F530.
Ó FEBS 2004 TauT expression, activity and subcellular localization (Eur. J. Biochem. 271) 4657
15. Han, X.B., Patters, A.B. & Chesney, R.W. (2002) Transcriptional
repression of taurine transporte r gene (TauT ) by p53 in renal ce lls.
J. Bi ol. Chem. 277, 39266–39273.
16. Han, X.B., Patters, A.B. & Chesney, R.W. (2003) Transactivation
of Ta uT by p53 in M CF-7 cells – The role o f estrogen r eceptors.
Taurine 5 : Beginning 21 st Century 526 , 139–147.
17. Mochizuki, T., Satsu, H. & Shimizu, M. (2002) Tumor necrosis
factor alpha stimulates taurine uptake and transporter gene
expression in human intestinal C aco-2 cells. FEBS Lett. 517, 9 2–96.
18. Chang, R.C.C., S tadlin, A. & Ts ang, D . (200 1) Effects of tumor
necrosis factor alpha on taurine uptake in cultured rat astrocytes.
Neurochem . Int. 38, 2 49–254.
19. Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall, R.J.
(1951) Protein measurement with the folin phenol reagent. J. Biol.
Chem. 193, 265–275.

20. Lambert, I.H. (2003) Reactive o xygen species regulate swelling-
induced taurine efflux in NIH3T3 mouse fibroblasts. J. Membr.
Biol. 192, 19–32.
21. Christensen, S.T., Guerra, C., Wada, Y., Valentin, T., Angeletti,
R.H., Satir, P. & Hamasaki, T. (2001) A regulatory light chain of
ciliary outer arm dynein in Tetrahymena thermophila. J. Biol.
Chem. 276, 20048–20054.
22. Blom, N., Gammeltoft, S. & Brunak, S. (1999) Sequence and
structure-based prediction o f eukaryotic protein ph osphorylatio n
sites. J. Mol. Bi ol. 294, 1351–1362.
23. Nash, S.R., Giros, B., Kingsmore, S.F., Rochelle, J.M., Suter,
S.T., Gregor, P., Seldin, M.F. & Caron, M.G. (1994) Cloning,
pharmacological characterization, and genomic localization of the
human creatine transporter. Receptors C hannels 2, 165–174.
24. Lambert, I.H. (1985) Taurine t ransport in Ehrlich ascites tumour
cells Specificty and chloride de pendenc e. Mol. Physiol. 7, 323–332.
25. Lambert, I.H. (2003) Regulation of the volume-sensitive taurine
efflux pathway in NIH3T3 mouse fibroblasts. I n Taurine in the
21st Ce ntury (eds Lombardini, J.B., Sc haffer, S.W. & Azuma, J .)
pp. 115 –122, Kluwer A cademic/Plenum Publishers, NY, USA.
26. Meng, T.C., Fukada, T. & Tonks, N.K. (2002) Reversible oxi-
dation and inactivation of protein tyrosine phosphatases i n vivo.
Mol. Cell 9, 387–399.
27. Liu, J.D., Wang, Y.J., Chen, C.H., Yu, C.F., Chen, L.C., Lin,
J.K., Liang, Y.C., Lin, S.Y. & Ho, Y.S. (2003) Mole cular
mechanisms of G (0) /G (1) cell-cycle arrest and ap optosis induce d
by terfenadine in h uman cancer cells. Mol Carcinogenesis 37,
39–50.
28. Bitoun, M. & Tappaz, M. (2000) Taurine down-regulates basal
and osmolarity-induced gene expression of its transporter, but not

the gene e xpression of its biosynthetic enzym es, in astrocyte pri-
mary cu ltures. J. Neurochem. 75, 919–924.
29. Poulsen, K .A., L itm an, T., Eriksen, J., Mollerup, J. & Lambert,
I.H. (2002) Downregulation of taurine uptake in multidrug
resistant Ehrlich ascites tumor cells. Amino Acids 22 , 333–350.
30. Pasantes-Morales, H., Sanchez, O.R., Miranda, D. & Moran, J .
(1997) Volume regulation in NIH/3T3 cells not expressing
P-glycoprotein. I. Regulatory volume decrease. Am.J.Physiol.
272, C 1798–C1803.
31. Lang, F ., Madlung, J., Siemen, D ., Ellory, C., Lepple-Wienhues,
A. & Gulbins, E. (2000) The involvement of caspases in the CD95
(Fas/Apo-1) but not swelling-induced cellular taurine release from
Jurkat T-lymphocytes. Pflugers Arch. 440 , 93–99.
32. Takahashi, K., Ohyabu, Y., Takahashi, K., Solodushko, V.,
Takatani,T.,Itoh,T.,Schaffer,S.W.&Azuma,J.(2003)Taurine
renders the cell re sistant to ischemia-induced injury i n cultured
neonatal rat cardiom yocytes. J. Cardiovascular Pharmacol. 41,
726–733.
33. Lang, F., Madlung, J., Uhlemann, A.C., Risler, T. & Gulbins, E.
(1998) Cellular taurine release triggered by stimulation of the
Fas(CD95) receptor in Jurkat lymphocy tes. Pflugers Arch. 436,
377–383.
34. Loyher, M.L., M utin, M., Woo,S.K.,Kwon,H.M.&Tappaz,
M.L. (2004) Transcription factor tonicity-resp onsive enhancer-
binding protein (TonEBP) which transactivates osmoprotective
genes is expressed and upregulated following acute systemic
hypertonicity in neurons in brain. Neuroscience 124, 89–104.
35. Miyakawa,H.,Woo,S.K.,Dahl,S.C.,Handler,J.S.&Kwon,
H.M. (1999) Tonic ity-respon sive enhancer binding protein, a rel-
like protein that stimulates transcription in response to hyper-

tonicity. Proc. Natl Acad. Sci. USA 96, 253 8–2542.
36. Han,X.,Budreau,A.M.&Chesney, R.W. (2000) Identification of
promoter elements involved in adaptive re gulation of the taurine
transporter gene: role of cytosolic Ca2+ signaling. Adv. Exp.
Med. 483, 535–544.
37. Tappaz, M.L. (2004) Taurine biosynthetic enzymes and taurine
transporter: Molecular identification and regulations. Neurochem.
Res. 29, 83–96.
38. Beckman, M.L. & Quick, M.W. (1998) Neurotransmitter trans-
porters: Regulators of function and functional regulation.
J. Memb r. Biol. 164 , 1–10.
39. Miner, L.H., Schroeter, S., Blakely, R.D. & Sesack, S.R. (2003)
Ultrastructural localization of the norepinephrine transporter in
superficial and d eep laye rs of t he rat p relimbic p refron tal c ortex
and its spatial relationship to probable dopamine terminals.
J. Comp . Neurol. 466 , 478–494.
40. Suh, K.S., Mutoh, M ., Nagashima, K., Fernandez-Salas, E.,
Edwards, L.E., Hayes, D.D., Crutchley, J.M., Marin, K.G.,
Dumont,R.A.,Levy,J.M.,Cheng,C.,Garfield,S.&Yuspa,S.H.
(2004) The organellular chloride channel protein CLIC4/mtCLIC
translocates to the nucleus in response to cellular stress and
accelerates apoptosis. J . Biol. Chem. 279, 4632–4641.
41. Schaffer, S.W., Pastuk h, V., Solodushko, V., Kram er, J. &
Azuma, J. (2002) Effect of ischemia, calcium depletion and
repletion, acidosis and hypoxia on cellular taurine content. Amino
Acids 23, 3 95–400.
42. Lambert, I.H. (2004) Modulation of volume-sensitive taurine re-
lease from NIH3T3 mouse fibroblasts by reactive oxygen species.
In Cell Volume and Signal Transduction (P. Lauf, ed), pp. 000–000.
Kluwer A cademic/Plenum P ublishers, NY, U SA.

43. Liu, S.J. & McHowat, J. (1998) Stimulation o f different phos-
pholipase A (2) isoforms by TNF-alpha and IL-1 beta in adult rat
ventricul ar myocytes. Am.J.Physiol.275, H1462–H1472.
4658 J. W. Voss et al. (Eur. J. Biochem. 271) Ó FEBS 2004