(Received for publication, September 17, 1996, and in revised form, December 18, 1996)
From the rCAT3 (at ationic
mino acid ransporter 3), a cDNA that
encodes a novel member of the murine CAT family was isolated. The protein encoded by rCAT3 contained 619 amino acids, 53-58% of which
were identical with those of the murine CAT family proteins previously
described (mouse CAT1, CAT2a, CAT2b, and rat CAT1). Transient
expression of rCAT3 and L-[14C]arginine
incorporation experiments in COS7 cells verified a high affinity system
y+ transporter activity of rCAT3. First, rCAT3-mediated
L-[14C]arginine incorporation was
time-dependent and saturable with half-saturation constant
(Km) values of 103 ± 12 µM
(mean ± S.E., n = 3). Second, the incorporation
was specific for cationic amino acids as evidenced from the inhibition
by L-arginine, L-lysine, and
L-ornithine. Third, neither sodium nor chloride ions in the extracellular medium were required for the activity. Fourth, the incorporation was inhibited by high potassium-induced membrane depolarization. On Northern blot using RNAs from various rat tissues, the expression of rCAT3 mRNA was restricted to the brain. These results indicated a role of rCAT3 in the system y+
transporter activity in the nervous tissue.
Nitric oxide, a highly diffusible molecule involved in signal
transduction in the brain, is formed from the terminal guanidino group
of L-arginine by an enzyme NO synthetase
(NOS)1 (1). Of the two classes of NOS
isoforms (constitutive and inducible), neuronal NOS belongs to the
constitutive type, and it is generally accepted that calcium/calmodulin
signaling is the primary mechanism that regulates its activity (1, 2). Several lines of evidence have suggested that besides the regulation of
NOS activity, NO synthesis is also regulated by availability of the
substrate L-arginine (3) and that L-arginine is
provided to NOS via transport from the extracellular space rather than from the intracellular metabolic pool. For example,
N-methyl-D-aspartate-induced NO production in
neuroblastoma cells depends on extracellular L-arginine
(4). NO synthesis by a lipopolysaccharide-activated macrophage was also
dependent on extracellular L-arginine and was inhibited by
L-lysine or L-ornithine (5). Inhibitory effects of these cationic amino acids on NO synthesis were also observed in
brain synaptosomes (6). Furthermore, exogenous L-arginine enhanced NO synthesis in brain slices (7, 8).
The amino acids are incorporated from the exterior medium to the
intracellular space via carrier proteins on the cell surface. Among the
several types of amino acid transport systems distinguished by
substrate specificity, cationic amino acids such as
L-arginine, L-lysine, and
L-ornithine share the same transporter with a
sodium-independent activity, designated system y+ (9).
Although the presence of system y+ activity in
neuronal/glial cells has been demonstrated by functional studies (10,
11), a molecular basis for the activity remains unclear.
At least two genes, mCAT1 (ouse ationic
mino acid ransporter 1) and mCAT2, have been
identified and verified to encode a high affinity system y+
activity (Km = ~100 µM) in murine
non-neuronal tissues. mCAT1, originally identified as a receptor for an
ecotropic retrovirus (12), was later found to encode a system
y+ activity by expression studies in Xenopus
oocytes (13, 14). mCAT1 is widely expressed in various tissues and cell
lines and is believed to be a ubiquitous form of the CAT gene family
(14). mCAT2 gives rise to two alternative splice forms, mCAT2a and
mCAT2b. mCAT2b, originally identified as a T-cell early activated gene (15), was also found to encode a system y+ activity with
properties similar to that encoded by mCAT1 (16). mCAT2b has been shown
to be expressed in some types of cells including T cells and vascular
smooth muscle cells (17). A comparative study on the distribution of
mCAT2b, however, is lacking, and it is not known whether neuronal/glial
cells express this transporter. mCAT2a does not encode system
y+ but a low affinity, high capacity transporter (18). The
expression of mCAT2a is restricted to liver (18).
By using an RNase protection assay, Stoll et al. (19)
demonstrated a marked enrichment of the rat CAT1 (rCAT1) mRNA in
brain capillary, suggesting a role of rCAT1 in amino acid transport by
endothelial cells at the blood-brain barrier. The relatively low
expression level of rCAT1 in the whole brain mRNA might imply the
presence of a brain-specific, but as yet unknown transporter distinct
from rCAT1. More recently, Wu et al. (20) examined the
tissue distribution of rCAT1 mRNA and found a ubiquitous
distribution of a 7.9-kilobase pair hybridization signal and a
brain-specific 3.4-kilobase pair hybridization signal, the latter of
which might represent a brain-specific homologue of rCAT1.
In light of the regulation of NO synthesis by L-arginine
transport, we wished to identify the molecular basis for the system y+ activity in brain. The current study employed homology
screening of a rat brain cDNA library and led to the isolation of a
novel member of the CAT gene family, designated here as rCAT3.
Total RNA
was extracted from rat cerebellum as described (21). cDNA was
synthesized from 1 µg of the total RNA by SuperScript II reverse
transcriptase (Life Technologies, Inc.) using random hexamers as
primers. A part of the reaction mixture was used directly as a template
for PCR. The sequences of the primers used for PCR were
5 A cDNA library from adult Wistar rat
brain poly(A)+ RNA was constructed in The
EcoRI fragment of The probe was prepared by labeling the
EcoRI fragment of pBSKrCAT3 with [32P]dCTP
(110 TBq/mmol, Amersham) exactly as described above. A rat multiple
tissue Northern blot membrane was obtained from Clontech (Palo Alto,
CA). Blotting procedures were as described (23). The blot was
visualized for radioactivity with a BAS2000 image analyzer (Fuji Photo
Film Co. Ltd., Tokyo, Japan).
Where necessary, statistical analysis
was done by analysis of variance.
cDNA
prepared by reverse transcription from rat cerebellum RNA served as a
template for PCR. An ~550-base pair product was obtained using a
primer set derived from the 12th and 14th TMs of mCAT1 (12). The
deduced amino acid sequence of this PCR product displayed ~70%
identity with that of the corresponding region of mCAT1 and served as a
probe for library screening. By hybridization screening of a rat brain
cDNA library under high stringency conditions, one positive clone
( The nucleotide and deduced amino acid sequences of the
The predicted amino acid sequence of rCAT3 had a 53-58% identity with
those of other system y+ transporters previously reported
(Fig. 1C, GeneWorksTM, IntelliGenetics). The
leucine zipper motif is completely conserved in all members of the CAT
family. Although individual sequences are variable, all of them have a
potential N-glycosylation site and a potential protein
kinase C phosphorylation site on the same extra- and intracellular
loops, respectively. mCAT2a and mCAT2b are believed to be alternative
splicing products of the same gene, and their sequences differ from
each other only in a 41-amino acid stretch encoding the loop between
the 8th and 9th TMs and a part of the 9th TM (amino acids 358-398 of
mCAT2b). This divergent sequence has been shown to be responsible for
the differential kinetics of L-arginine transport activity
encoded by mCAT2a and mCAT2b (16). Within this stretch, the sequence of
rCAT3 is homologous with those of the others in an order of rCAT1 > mCAT1 > mCAT2b > mCAT2a (GeneWorksTM,
IntelliGenetics).
L-Arginine transport activity of rCAT3 was
tested by transient expression of the cDNA and
L-[14C]arginine incorporation assays in COS7
cells. In every experiment, one-half of the plates were transfected
with an empty vector plasmid and assayed in parallel. Basal
incorporations exhibited by vector-transfected cells were 40-50% of
those exhibited by pMErCAT3-transfected cells. rCAT3-mediated
incorporations were calculated as the differences between the values
from pMErCAT3- and vector-transfected cells.
rCAT3-mediated incorporation of
L-[14C]arginine (100 nM) was
time-dependent and was close to linear in the first 10 min
(Fig. 2A). Therefore, the reaction time was
set to 10 min in the following experiments. The rCAT3-mediated
incorporation of L-[14C]arginine was
dose-dependent and was saturated over the concentration of
0.5 mM (Fig. 2B) as expected for a
carrier-mediated process. Eadie-Hofstee plot analysis (Fig.
2B, inset) of the saturation isotherms gave
half-saturation constant (Km) values of 103 ± 12 µM (mean ± S.E., n = 3). The
Km values obtained were comparable with those of the
L-arginine incorporation in Xenopus oocytes
mediated by mCAT1 (14) or mCAT2b (16), both of which were verified to
encode a high affinity system y+ transporter. They were
also comparable with those for the system y+ transport
activity determined in brain slices, cultured neurons, and glial cells
(10, 11). In addition, they were at least twice as low as those of the
L-arginine incorporation in Xenopus oocytes
mediated by mCAT2a (18), which encodes a liver-specific, low affinity
transporter distinct from system y+.
The substrate specificity was examined by testing the ability of
unlabeled amino acids to inhibit the rCAT3-mediated
L-[14C]arginine incorporation. Of the 22 naturally occurring L-amino acids, only the 3 basic amino
acids (L-arginine, L-lysine, and L-ornithine) competed the incorporation (Fig.
3). Both of the two structural analogues of
L-arginine tested (D-arginine and L-citrulline) also caused a significant inhibition of the
incorporation (Fig. 3). There was, however, a statistically significant
difference between the effects of L and D
isomers of arginine, suggesting a stereospecificity of the
incorporation. To demonstrate that the inhibitory effects of the
cationic amino acids observed were due to direct competition for rCAT3,
we conducted incorporation experiments of radioactive
L-lysine and L-ornithine (Fig.
4). In both cases, the rCAT3-mediated incorporations
were dose-dependent, and Eadie-Hofstee plot analysis (not
shown) gave Km values of 147 ± 22 and 219 ± 26 µM (mean ± S.E., n = 3) for
incorporations of L-[3H]lysine and
L-[14C]ornithine, respectively. The
specificity for cationic amino acids, together with the high affinity
for L-arginine, indicated a system y+ transport
activity of rCAT3.
Next, we examined the dependence of the rCAT3 activity on extracellular
monovalent ions and membrane potentials by manipulating the ionic
constitutions of the extracellular medium. Neither the isotonic
substitution of Na+ with Li+ nor that of
Cl
Currently available NOS inhibitors are structural
analogues of L-arginine, and some of them also inhibit
L-arginine transport (10, 32). Therefore, we examined the
ability of three drugs, L-NMMA, L-NIO and
L-NAME, to inhibit rCAT3-mediated
L-[14C]arginine incorporation in COS7 cells
(Fig. 6). When the transfected cells were exposed to
L-[14C]arginine (100 nM) in the
presence or absence of the NOS inhibitors (1 mM), both
L-NMMA and L-NIO caused a significant
inhibition of the rCAT3-mediated
L-[14C]arginine incorporation, while
L-NAME caused only a marginal effect. There was also a
statistically significant difference between the effects of
L-NMMA and L-NIO. Thus, the rank order of
potency of the three drugs was L-NMMA > L-NIO
On Northern blot
of poly(A)+ RNA from various rat tissues, a single band
with a size of 3.3 kilobase pairs was detected in the brain only with a
rCAT3 cDNA probe (Fig. 7), suggesting the brain-specific expression of rCAT3. Although direct evidence is lacking, it is very likely that the brain-specific 3.4-kilobase pair
hybridization signal of rCAT1 detected by Wu et al. (20) was
rCAT3.
We have isolated a cDNA clone (rCAT3) that
encodes a novel member of the murine CAT family. Expression and
functional characterization of the gene verified a high affinity,
system y+ transporter activity of rCAT3. The expression of
rCAT3 mRNA is restricted to the brain. The rCAT3 cDNA will be
an essential tool to further clarify the molecular basis of the system
y+ activity in the nervous tissues and that of the
regulation of NO synthesis by L-arginine transport.
Department of Pharmacology,
Probe Preparation by Reverse Transcription-PCR
-GGTCTTACGGTACCAGCCAG-3
(sense) and 5
-GGACGCTTCCTCACTG-3
(antisense). These sequences correspond to the nucleotide sequences at
positions 1461-1480 and 1977-1996 of mCAT1 (12). The PCR contained 1 µM dNTPs, 67 mM Tris-Cl (pH 8.8), 16.7 mM (NH4)2SO4, 10 mM 2-mercaptoethanol, 0.5 mM dimethylsulfonate,
2 mM MgCl2, 2.5 units of Taq DNA
polymerase (Takara, Ostu, Japan), 1 µM concentration of
each primer, and the template. Thirty-five cycles of the following temperature conditions were used for amplification: 94 °C for 1 min,
55 °C for 2 min, 72 °C for 2 min, and final extension at 72 °C
for 10 min. An amplified DNA was separated on an agarose gel,
extracted, subcloned into pCR1000 (Invitrogen, NV Leak, Netherlands), and then sequenced using a BcabestTM dideoxy sequencing kit
(Takara). The PCR product was labeled with [32P]dCTP (110 TBq/mmol, Amersham, Buckinghamshire, UK) using a Random Primers DNA
Labeling SystemTM (Life Technologies, Inc.) and was used as
a hybridization probe.
gt10 vector
(Stratagene, La Jolla, CA) (22), and hybridization screening of the
library (~106 clones) was performed as described (23). A
single positive clone (
rCAT3) was extracted from the plaque, and the
EcoRI fragment of the phage was subcloned into pBlueScript
SK+ (Stratagene) to give pBSKrCAT3. The nucleotide sequence of the
pBSKrCAT3 insert was determined as described above.
rCAT3 was subcloned into the mammalian
expression vector pME18sf
to give pMErCAT3. COS7 cells were routinely
maintained in Dulbecco's modified Eagle's medium supplemented with
10% fetal bovine serum in a CO2 incubator. The cells were seeded on six-well plates (105 cells/well) and transfected
with pMErCAT3 or with vector plasmid (for basal incorporations) using
LipofectAMINETM (Life Technologies, Inc.) according to the
manufacturer's instructions. Two days later, they were subjected to
L-[14C]arginine incorporation assay. The
cells were washed twice with HEPES-buffered saline (HBS) (150 mM NaCl, 10 mM HEPES, pH 7.5, 1 mM
CaCl2, 1 mM MgCl2, and 5 mM KCl) and further incubated in HBS at 37 °C for 10 min. The reaction was started by changing the media to HBS containing
L-[14C]arginine (11 GBq/mmol). The
concentrations of L-[14C]arginine that were
applied and the reaction time used are indicated in the figure legends.
The reaction was stopped by washing the cells three times with ice-cold
HBS. The cells were then lysed in 0.2 N NaOH, 1% sodium
dodecyl sulfate, and the radioactivity incorporated was determined by a
solid scintillation counter. Incorporation assays with
L-[3H]lysine (3.55 TBq/mmol) and
L-[14C]ornithine (9.69 GBq/mmol) were done in
exactly the same way. All of the radioactive amino acids were obtained
from Amersham. Protein concentration of the cell lysate was determined
using a MicroBCA kit (Pierce). All of the unlabeled amino acids and the
three NOS inhibitors
(NG-monomethyl-L-arginine
(L-NMMA)),
L-N5-(1-iminoethyl)ornithine
(L-NIO)),
NG-nitro-L-arginine methyl
ester (L-NAME)) used for the inhibition experiments were
obtained from Sigma.
rCAT3 Is a Novel Member of the Murine CAT Family
rCAT3) was isolated.
rCAT3 insert
(Fig. 1A) indicated that it encoded a
complete coding sequence of rCAT3. The 5
-proximal ATG triplet is
followed by a 1857-base pair open reading frame encoding 619 amino acid
residues with a calculated molecular mass of ~67 kilodaltons. The
sequence around this ATG fits the consensus sequence for eukaryotic
translation initiation sites (24). rCAT3 lacks a readable signal
sequence in the N terminus (25). Fig. 1B depicts a putative
membrane topology of rCAT3 predicted by a hydrophobicity analysis
(GeneWorksTM, IntelliGenetics, Mountain View, CA). rCAT3
contains 14 putative TMs. One potential N-linked
glycosylation site is located on an extracellular loop between the 5th
and 6th TMs, and one potential protein kinase C phosphorylation site is
located on an intracellular loop between the 10th and 11th TMs. Another
structural feature of rCAT3 is a leucine zipper consensus sequence in
the first TM (Fig. 1A). Originally identified as a
functional unit required for dimerization of DNA-binding proteins (26),
the leucine zipper motif has also been found in membrane proteins
including voltage-gated potassium channels (27), glucose transporters
(28), and some neurotransmitter transporters (29), where it may mediate
subunit oligomerization.
Fig. 1.
Primary structure of rCAT3. A,
nucleotide and deduced amino acid sequences of rCAT3 cDNA.
Nucleotides are numbered in the 5 to 3
direction beginning with the
first methionine codon. Triple asterisk, the stop codon
flanking the open reading frame; underline, putative transmembrane domains; asterisk, a potential
N-linked glycosylation site; double underline, a
putative protein kinase C phosphorylation site; plus, a
leucine zipper motif. B, schematic representation of the
putative membrane topology of rCAT3 predicted by a hydrophobicity
analysis. C, homology alignments of rCAT3 and other CAT
family proteins. Residues conserved in all members are
boxed.
[View Larger Versions of these Images (52 + 83K GIF file)]
Fig. 2.
Kinetics of rCAT3-mediated
L-[14C]arginine incorporation in COS7 cells.
A, time course. COS7 cells transfected with pMErCAT3 or
vector plasmid were incubated with
L-[14C]arginine (100 nM) for the
time indicated, and the radioactivity incorporated was determined as
described under "Experimental Procedures." The incorporation was
normalized for the protein recovered from each well, and the
rCAT3-mediated incorporation was defined as the difference between the
values from pMErCAT3- and vector-transfected cells. B,
saturation isotherms. COS7 cells transfected with pMErCAT3 or vector
plasmid were incubated for 10 min with increasing concentrations of
L-[14C]arginine. To obtain concentrations
higher than 100 nM,
L-[14C]arginine (100 nM) was
diluted with unlabeled L-arginine, and the radioactivities
recovered were normalized for the reduced specific activities.
rCAT3-mediated incorporation was determined as described above.
Inset is an Eadie-Hofstee plot of the same data. The
means ± S.E. of triplicate determinations obtained in a single
experiment are shown in both A and B. These
experiments were replicated three times with similar results.
[View Larger Version of this Image (13K GIF file)]
Fig. 3.
Inhibition of rCAT3-mediated
L-[14C]arginine incorporation in COS7 cells
by cationic amino acids. COS7 cells transfected with pMErCAT3 or
vector plasmid were incubated for 10 min with L-[14C]arginine (100 nM) in the
absence (None) or presence of the individual amino acid
indicated (1 mM). All of the amino acids applied were L isomers except for D-arginine. rCAT3-mediated
incorporation was determined as described in the legend to Fig. 2. The
data are presented as relative to the incorporation in the absence of
the unlabeled amino acid. Each bar represents the mean ± S.E. of three determinations, each done in triplicate. *,
p < 0.01, significantly different from the control
values (analysis of variance).
[View Larger Version of this Image (21K GIF file)]
Fig. 4.
Dose dependence of rCAT3-mediated
incorporations of L-[3H]lysine
(circles) or L-[14C]ornithine
(squares) in COS7 cells. The assays were performed exactly as described for L-[14C]arginine in
the legend to Fig. 2B. The reaction time was 10 min. Shown
are the means ± S.E. of triplicate determinations obtained in a
single experiment. These experiments were replicated three times with
similar results. The Km values obtained were given
in the text.
[View Larger Version of this Image (15K GIF file)]
with CH3COO
caused any
change in the rCAT3-mediated L-[14C]arginine
incorporation (data not shown). In contrast, substitution of
Na+ with K+ caused a dose-dependent
inhibition of the rCAT3 activity (Fig. 5). These results
are consistent with the functional properties of system y+
described previously, i.e. it is dependent neither on the
extracellular Na+ nor on Cl
(9) and is
inhibited by high K+-induced membrane depolarization (9,
30, 31).
Fig. 5.
Inhibition of rCAT3-mediated
L-[14C]arginine incorporation in COS7 cells
by high concentrations of K+. COS7 cells transfected
with pMErCAT3 or vector plasmid were incubated for 10 min with
L-[14C]arginine (100 nM) in HBS
containing increasing concentrations of K+. Normal HBS
contained 5 mM KCl and 150 mM NaCl. To increase
the concentrations of K+, NaCl was replaced with equimolar
KCl. rCAT3-mediated incorporation was determined as described in the
legend to Fig. 2. The data are presented as relative to the
incorporation in normal HBS. Shown are the means ± S.E. of three
determinations, each done in triplicate.
[View Larger Version of this Image (11K GIF file)]
L-NAME. Similar rank order of
potency was observed for these drugs to inhibit system y+
L-arginine transport in endothelial cells (32) and in
neuronal cells (10). These results gave further evidence for the system y+ transport activity of rCAT3 and support the notion that
the transport of L-NAME is mediated by a system different
from y+ (32).
Fig. 6.
Inhibition of rCAT3-mediated
L-[14C]arginine incorporation in COS7 cells
by NOS inhibitors. COS7 cells transfected with pMErCAT3 or vector
plasmid were incubated for 10 min with
L-[14C]arginine (100 nM) in the
absence or presence of the individual NOS inhibitor (1 mM).
rCAT3-mediated incorporation was determined as described in the legend
to Fig. 2. The data are presented as relative to the incorporation in
the absence of the drug. Each bar represents the mean ± S.E. of three determinations, each done in triplicate. *,
p < 0.01, significantly different from each other
(analysis of variance).
[View Larger Version of this Image (14K GIF file)]
Fig. 7.
Brain-specific expression of rCAT3
mRNA. Northern blotting was performed as described under
"Experimental Procedures." A, an autoradiograph showing
the hybridization signal of a rCAT3 cDNA probe on a rat multiple
tissue Northern blot membrane. Molecular sizes are given on the
right. B, internal control. The same blot was
probed with a human -actin cDNA.
[View Larger Version of this Image (40K GIF file)]
*
This work was supported in part by research grants from the
Ministry of Education, Science, and Culture of Japan.The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
¶
Research fellow of the Japan Society for the Promotion of
Science.
To whom correspondence should be addressed. Tel.:
81-75-753-4477; Fax: 81-75-753-4402; E-mail:
masaki{at}mfour.med.kyoto-u.ac.jp.
1
The abbreviations used are: NOS, nitric-oxide
synthetase; CAT, cationic amino acid transporter; PCR, polymerase chain
reaction; TM, transmembrane domain; HBS, HEPES-buffered saline;
L-NMMA, NG-monomethyl-L-arginine;
L-NIO,
L-N5-(1-iminoethyl)ornithine;
L-NAME,
NG-nitro-L-arginine methyl
ester; rCAT, rat cationic amino acid transporter; mCAT, mouse cationic
amino acid transporter.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.