Transport Properties of a System y+L Neutral and Basic Amino Acid Transporter

The properties of system y+L-mediated transport were investigated on rat system y+L transporter, ry+LAT1, coexpressed with the heavy chain of cell surface antigen 4F2 in Xenopusoocytes. ry+LAT1-mediated transport of basic amino acids was Na+-independent, whereas that of neutral amino acids, although not completely, was dependent on Na+, as is typical of system y+ L-mediated transport. In the absence of Na+, lowering of pH increased leucine transport, without affecting lysine transport. Therefore, it is proposed that H+, besides Na+ and Li+, is capable of supporting neutral amino acid transport. Na+ and H+ augmented leucine transport by decreasing the apparent K m values, without affecting the V max values. We demonstrate that although ry+LAT1-mediated transport of [14C]l-leucine was accompanied by the cotransport of 22Na+, that of [14C]l-lysine was not. The Na+ to leucine coupling ratio was determined to be 1:1 in the presence of high concentrations of Na+. ry+LAT1-mediated leucine transport, but not lysine transport, induced intracellular acidification in Chinese hamster ovary cells coexpressing ry+LAT1 and 4F2 heavy chain in the absence of Na+, but not in the presence of physiological concentrations of Na+, indicating that cotransport of H+ with leucine occurred in the absence of Na+. Therefore, for the substrate recognition by ry+LAT1, the positive charge on basic amino acid side chains or that conferred by inorganic monovalent cations such as Na+ and H+, which are cotransported with neutral amino acids, is presumed to be required. We further demonstrate that ry+LAT1, due to its peculiar cation dependence, mediates a heteroexchange, wherein the influx of substrate amino acids is accompanied by the efflux of basic amino acids.

Molecular cloning approaches have led to the identification of Na ϩ -dependent and Na ϩ -independent amino acid transporters (1). Four amino acid transporter families (three Na ϩ -dependent and one Na ϩ -independent) have been identified so far (2)(3)(4)(5)(6)(7). The amino acid permease family SLC7 consists of Na ϩindependent amino acid transporters composed of two subfamilies: the cationic amino acid transporter (CAT) 1 family, which consists of classical system y ϩ basic amino acid transporters, and the recently identified L-type amino acid transporter (LAT) family, which includes transporters associated with type II membrane glycoproteins, such as 4F2 heavy chain (4F2hc) and rBAT (related to b 0,ϩ amino acid transporter) (6,7). The members of the LAT family exhibit a variety of substrate selectivity and represent the amino acid transport systems L, asc, y ϩ L, x Ϫ C , and b 0,ϩ (8 -17). Among the members of the LAT family, the system y ϩ L transporters, y ϩ LAT1 and y ϩ LAT2, are unique because of their peculiar Na ϩ dependence (13,14). System y ϩ L was originally identified as a transport system that transports both neutral and basic amino acids in the erythrocyte plasma membrane (18). The transport of basic amino acids by this system was shown to be Na ϩ -independent, whereas that of neutral amino acids, although not completely, was dependent on Na ϩ (18 -21). This characteristic Na ϩ dependence was confirmed for mouse and human y ϩ LAT1 coexpressed with 4F2hc in Xenopus oocytes (13,14). In the case of erythrocyte system y ϩ L-mediated transport, it was proposed that Na ϩ (and Li ϩ ) increases the affinity of neutral but not basic amino acids for the binding sites (19).
To understand the mechanisms of substrate recognition and transport by system y ϩ L transporters, we isolated a y ϩ LAT1 cDNA from the rat kidney and expressed it in Xenopus oocytes. We report here that not only Na ϩ and Li ϩ , but also H ϩ , are capable of supporting the neutral amino acid transport and that these monovalent cations are, in fact, cotransported with the neutral amino acids. Because of the peculiar cation dependence, y ϩ LAT1 mediates the exodus of basic amino acids through an obligatory exchange mechanism.
EXPERIMENTAL PROCEDURES cDNA Cloning of Rat y ϩ LAT1-The cDNA for an expressed sequence tag (GenBank™ EBI/DDBJ accession number R82979) corresponding to the human y ϩ LAT1 was obtained from the Integrated and Molecular Analysis of Genomes and their Expression (IMAGE). A 0.3-kilobase EcoRI fragment was excised from the cDNA (IMAGE clone number 187050) and used as the probe for screening a rat kidney cDNA library (11,22,23). cDNAs in positive ZipLox phages were rescued into plasmid pZL1 and sequenced as described previously. A cDNA clone was isolated, which contained an open reading frame encoding a 512amino acid protein, designated as ry ϩ LAT1 (rat y ϩ LAT1: GenBank™ EBI/DDBJ accession number AB020520) with the amino acid sequence identity to y ϩ LAT1 from humans (90%) and mice (94%) (13,14).
Xenopus Oocyte Expression-cRNAs were obtained by in vitro transcription using SP6 RNA polymerase for ry ϩ LAT1 (rat y ϩ LAT1) in pZL1 linearized with RsrII and T7 RNA polymerase for rat 4F2hc in pBluescript II SK linearized with XbaI as described elsewhere (24). The Xenopus oocyte expression studies and uptake measurements were performed as described previously (22,25). The uptake of 14 C-labeled amino acids was measured 3 days after injection of the cRNA. For the coexpression experiments, 12.9 ng of ry ϩ LAT1 cRNA and 12.1 ng of 4F2hc cRNA at the molar ratio of 1:1 were mixed and injected into each oocyte. To express ry ϩ LAT1 or 4F2hc singly in Xenopus oocytes, 12.9 and 12.1 ng of ry ϩ LAT1 or 4F2hc cRNA was injected, respectively.
Amino Acid Uptake Measurements-Groups of six to eight oocytes were incubated in 500 l of standard uptake solution (containing 100 mM NaCl, 2 mM KCl, 1 mM CaCl 2 , 1 mM MgCl 2 , 10 mM HEPES, 5 mM Tris; pH 7.4) or Na ϩ -free uptake solution, which was similar in composition to the standard uptake solution except that NaCl was replaced by choline chloride, containing 0.5-3.0 Ci of radiolabeled compounds (22). For the high K ϩ uptake solution, NaCl in the standard uptake solution was replaced by KCl. For the Li ϩ uptake solution, NaCl in the standard uptake solution was replaced by LiCl. To prepare uptake solutions at pH 5.5, MES was used instead of HEPES for the buffer system (11).
Preliminary experiments to determine the time course of [ 14 C]Lleucine (50 M) uptake by oocytes expressing ry ϩ LAT1 and 4F2hc indicated that the uptake was linearly dependent on the incubation time up to 60 min (data not shown); hence, in all the following experiments, the uptakes were measured for 30 min and the values were expressed as picomoles/oocyte/min.
The K m and V max values for amino acid substrates were determined using the Eadie-Hofstee equation based on the ry ϩ LAT1-mediated amino acid uptakes (11). The ry ϩ LAT1-mediated amino acid uptakes were calculated as differences between the mean uptakes by oocytes injected with ry ϩ LAT1 and 4F2hc cRNAs and those by water-injected control oocytes.
In the experiments to investigate whether the amino acid transports were trans-stimulated by the intracellular amino acids, the oocytes were loaded with amino acids by microinjecting them with 100 nl of 100 mM nonradiolabeled amino acids. The individual oocytes were incubated in the standard uptake solution for 30 min after the injection of the amino acids, to allow recovery from the injury caused by the microinjection, and then transferred to a standard uptake solution containing 100 M [ 14 C]L-leucine; the uptake was measured for 30 min. 22 Na ϩ Uptake Measurements and Determination of Na ϩ to Amino Acid Coupling Ratio-Groups of six to eight oocytes were preincubated for 30 min in the 22 Na ϩ uptake solution (containing 40 mM NaCl, 60 mM choline chloride, 2 mM KCl, 1 mM CaCl 2 , 1 mM MgCl 2 , 1 mM ouabain, 0.1 mM amiloride, 0.1 mM bumetanide, 10 mM HEPES, 5 mM Tris; pH 7.4) (25). 22 Na ϩ uptake was measured for 30 min in a 22 Na ϩ uptake solution containing 20 Ci of 22 Na ϩ with or without 200 M L-leucine or L-lysine. Concurrently, the uptakes of 200 M [ 14 C]L-leucine and [ 14 C]L-lysine by the same batch of oocytes coexpressing ry ϩ LAT1 and 4F2hc were measured under the same conditions. In separate experiments, to determine the Na ϩ :amino acid stoichiometry, the dependence of 22 Na ϩ uptake and [ 14 C]L-leucine uptake by the same batch of oocytes on the concentration of leucine in the 22 Na ϩ uptake solution was compared.
Efflux Measurements-100 nl (ϳ3 nCi) of [ 14 C]amino acids (100 M) were injected into the oocytes (12). After the injection, the individual oocytes were incubated for 30 min in the standard uptake solution and then transferred to a standard uptake solution with or without 100 M nonradiolabeled amino acids. The radioactivity in the medium and the remaining radioactivity in the oocytes were measured. The efflux of preloaded [ 14 C]L-lysine was linearly dependent on the incubation time up to 45 min, when 100 M L-leucine was applied extracellularly to the oocytes expressing ry ϩ LAT1 and 4F2hc; hence, in all the following experiments, the efflux was measured for 30 min. The values were expressed as percentage of radioactivity (radioactivity in the medium or oocytes/(radioactivity in the medium ϩ radioactivity in the oocytes) ϫ 100%).
For the measurements of the uptake and the efflux of the radiolabeled amino acids or 22 Na ϩ in the present study, six to eight oocytes were used for each data point. Each data point in the figures represents the mean Ϯ S.E. of the uptake by six to eight oocytes. To confirm the reproducibility of the results, experiments were performed in triplicate for each measurement using different batches of oocytes and cRNA transcribed in vitro. Representative results from the experiments are shown in the figures.
Intracellular pH Monitoring-ry ϩ LAT1 and rat 4F2hc cDNAs were subcloned into the mammalian expression vector pcDNA3.1 (Invitrogen) and cotransfected into Chinese hamster ovary (CHO) cells by electroporation (16). In brief, the cell suspension (800 l, 2.4 ϫ 106 cells/ml) was mixed with ry ϩ LAT1 plasmid and 4F2hc plasmid (10 g for each). The mixture was electroporated at 250 V, 960 F in a 0.4-cm cuvette in a Gene Pulsar (Bio-Rad). The cells were cultured in Ham's F-12 nutrient mixture containing 10% fetal bovine serum for 3 days. They were then examined using an Ultima-z confocal laser scanning system (Meridian, Okemos, MI) connected to a Zeiss Epifluorescence microscope. The cells were incubated at room temperature for 1 h with Na ϩ -containing medium (containing 137 mM NaCl, 5.4 mM KCl, 1.3 mM CaCl 2 , 0.5 mM MgCl 2 , 0.4 mM MgSO 4 , 0.44 mM KH 2 PO 4 , 0.34 mM K 2 HPO 4 , 4.2 mM NaHCO 3 , 5.5 mM glucose; pH 7.4) containing 5 M 3Ј-O-acetyl-2Ј,7Ј-bis(carboxyethyl)-4(or 5)-carboxyfluorescein, diacetoxymethyl ester (26,27). The dye-loaded cells were washed three times with Na ϩ -containing medium or Na ϩ -free medium, which was similar in composition to the Na ϩ -containing medium except that NaCl was replaced by choline chloride, and then further incubated in the Na ϩcontaining medium or Na ϩ -free medium for fluorescence measurements. To examine the behavior of 2Ј,7Ј-bis(carboxyethyl)-4(or 5)-carboxyfluorescein in the dual-excitation mode, two laser sources were mounted on the confocal laser scanning system (26,27). The excitation light at 488 nm (pH-dependent wavelength) was provided by an argon ion laser. A He-Cd laser was used to adapt the confocal laser scanning microscope for excitation at 442 nm (pH-independent wavelength). The intensity of the fluorescence signal was measured with a photomultiplier and visualized on a personal computer.

RESULTS
Transport Activity and the Effect of Na ϩ -As reported for human and mouse y ϩ LAT1 (13,14), ry ϩ LAT1 required 4F2hc for its functional expression in Xenopus oocytes (Fig. 1a). All the following experiments were therefore performed by coexpressing ry ϩ LAT1 and 4F2hc. The uptake of 50 M [ 14 C]Lleucine or [ 14 C]L-lysine was measured in the presence or absence of Na ϩ . As shown in Fig. 1b, leucine transport was dependent on the extracellular Na ϩ , whereas lysine transport was not affected by the removal of Na ϩ from the extracellular medium. It is noticeable that the low level of leucine transport still remained in the absence of Na ϩ (Fig. 1b).
The Effects of Monovalent Cations on Leucine Transport-To determine which monovalent cations can take the place of Na ϩ to support leucine transport, the uptake of [ 14 C]L-leucine (200 M) was measured in Li ϩ uptake solution, high K ϩ uptake solution, and low pH uptake solution (Na ϩ -free uptake solution with a pH of 5.5). The values were compared with that in Na ϩ -free uptake solution, pH 7.5. As shown in Fig. 2, Li ϩ was able to take the place of Na ϩ to support leucine transport. Leucine transport was increased by lowering the pH, whereas it was not affected by increased K ϩ concentration of the uptake solution (Fig. 2).
Dependence of Substrate Transport on Na ϩ and pH-To determine the role of Na ϩ in ry ϩ LAT1-mediated transport, the concentration dependence of L-leucine transport and L-lysine transport was determined by measuring it in standard uptake solution (100 mM Na ϩ ) and in Na ϩ -free uptake solution (0 mM Na ϩ ). Although the K m value for leucine transport into oocytes coexpressing ry ϩ LAT1 and 4F2hc was dramatically increased in Na ϩ -free uptake solution, the V max value was less affected (K m ϭ 44 M, V max ϭ 6.7 pmol/oocyte/min in standard uptake solution containing Na ϩ 100 mM, and K m ϭ 423 M, V max ϭ 4.8 pmol/oocyte/min in Na ϩ -free uptake solution, both determined in the same batch of oocytes) (Fig. 3a). On the other hand, neither the K m nor the V max value for lysine transport was affected by changing Na ϩ concentration (K m ϭ 68 M, V max ϭ 9.0 pmol/oocyte/min in standard uptake solution containing Na ϩ 100 mM, and K m ϭ 50 M, V max ϭ 7.8 pmol/oocyte/min in Na ϩ -free uptake solution, both determined in the same batch of oocytes) (data not shown).
The effect of pH on ry ϩ LAT1-mediated transport was further examined by comparing the uptakes of 14 C-labeled leucine and lysine at pH 7.5 and 5.5. Although the K m value for leucine transport was markedly altered with a change of pH of the uptake solution, the V max value was less affected (K m ϭ 383 M, V max ϭ 5.7 pmol/oocyte/min in Na ϩ -free uptake solution, pH 7.5, and K m ϭ 119 M, V max ϭ 6.5 pmol/oocyte/min in Na ϩ -free uptake solution, pH 5.5, both determined in the same batch of oocytes) (Fig. 3b). On the other hand, neither the K m nor the V max value for lysine transport was affected with a change of pH (K m ϭ 74 M, V max ϭ 7.6 pmol/oocyte/min in Na ϩ -free uptake solution, pH 7.5, and K m ϭ 81 M, V max ϭ 6.7 pmol/ oocyte/min in Na ϩ -free uptake solution, pH 5.5, both determined in the same batch of oocytes) (data not shown). The concentration dependence profiles of neither leucine nor lysine uptake measured in the standard uptake solution (100 mM Na ϩ ) were affected by the pH of the solution (data not shown).
Transport of Na ϩ and H ϩ Accompanies Amino Acid Transport-The uptake of 22 Na ϩ was measured in the presence or absence of amino acids (200 M L-leucine or L-lysine) in the presence of 40 mM Na ϩ in the uptake medium containing ouabain, amiloride, and bumetanide to inhibit endogenous Na ϩ transport in the oocytes. In the oocytes coexpressing ry ϩ LAT1 and 4F2hc, marked 22 Na ϩ uptake was detected in the presence of leucine in the medium, whereas no significant 22 Na ϩ uptake was detected in the absence of leucine (Fig. 4a). Lysine, however, did not induce 22 Na ϩ uptake (Fig. 4a). In the same batch of oocytes, the uptakes of 200 M [ 14 C]L-leucine and [ 14 C]Llysine were measured under the same conditions as for the 22 Na ϩ uptake measurements to verify that the uptakes of [ 14 C]L-leucine and [ 14 C]L-lysine were almost identical (Fig. 4b).
Leucine-induced 22 Na ϩ uptake was calculated as the difference between the 22 Na ϩ uptakes measured in the presence and absence of leucine, and was compared with ry ϩ LAT1-mediated [ 14 C]L-leucine uptake. The ratio of the fluxes of 22 Na ϩ and [ 14 C]L-leucine was 1:0.88 (13.9 Ϯ 2.6 pmol/oocyte/min for 22 Na ϩ flux versus 12.3 Ϯ 1.0 pmol/oocyte/min for [ 14 C]Lleucine flux, both determined in the same batch of oocytes, Fig. 4, a and b).
To confirm the Na ϩ :leucine stoichiometry, the dependence of leucine-induced 22 Na ϩ uptake on the leucine concentration was compared with the concentration dependence of [ 14 C]L-leucine uptake measured under the same conditions in the same batch of oocytes. 22  calculated to be 1:1.17. 22 Na ϩ uptake was further measured in a medium containing only trace levels of Na ϩ (0.28 M). Addition of leucine (200 mM) to the medium increased the 22 Na ϩ uptake, however, the level of uptake was still much lower than that of [ 14 C]L-leucine uptake measured under the same conditions in the same batch of oocytes under the Na ϩ to leucine coupling ratio of 0.00015:1 (data not shown).
Cotransport of H ϩ with amino acids was examined by measuring the intracellular pH changes induced by extracellularly applied amino acids in CHO cells cotransfected with ry ϩ LAT1 and 4F2hc. As shown in Fig. 5, extracellularly applied leucine (1 mM) induced intracellular acidification in the Na ϩ -free medium, whereas in a Na ϩ -containing medium (141 mM Na ϩ ) extracellularly applied leucine had no effect on the intracellular pH. In contrast, extracellularly applied lysine (1 mM) did not induce intracellular pH changes regardless of the presence or absence of Na ϩ in the medium. No pH change was detected in the control mock-transfected CHO cells with the addition of amino acids to the medium (data not shown).
Amino Acid Exchange via ry ϩ LAT1-Trans-stimulation of [ 14 C]L-leucine uptake by intracellularly loaded amino acids was examined by microinjecting high concentrations of nonlabeled amino acids into the oocytes. As shown in Fig. 6a, the uptake of [ 14 C]L-leucine measured in the standard uptake solution was stimulated by the intracellular injection of L-lysine, whereas no such stimulatory effects on the uptake were observed following intracellular injection of neutral amino acids such as leucine. L-Arginine, L-ornithine, and L-histidine also trans-stimulated [ 14 C]L-leucine uptake (data not shown).
The effect of extracellularly applied L-leucine was examined on oocytes loaded with [ 14 C]L-lysine. As shown in Fig. 6b, L-leucine applied extracellularly to the oocytes induced the marked efflux of preloaded [ 14 C]L-lysine from the oocytes coexpressing ry ϩ LAT1 and 4F2hc, indicating that ry ϩ LAT1 is an amino acid exchanger.
Na ϩ dependence of extracellularly applied amino acids to induce efflux of preloaded [ 14 C]L-lysine was examined. As shown in Fig. 7, the efflux induced by L-leucine was dependent on extracellular Na ϩ , whereas that induced by L-lysine was not affected by the removal of Na ϩ from the extracellular medium. In addition, the efflux induced by other neutral amino acids such as L-isoleucine, L-glutamine, and L-methionine was also reduced by the removal of Na ϩ (data not shown). The efflux induced by L-arginine and L-ornithine was not affected by the removal of Na ϩ (data not shown).
To determine which intracellular amino acids were effectively exchanged via ry ϩ LAT1, the efflux of preloaded [ 14 C]Lamino acids was measured in the standard uptake solution (Na ϩ 100 mM) containing 100 M L-leucine. A high level of efflux of intracellularly injected 14 C-labeled basic amino acids such as L-lysine, L-arginine, and L-ornithine was induced by extracellularly applied L-leucine (Fig. 8). A lower level induction of the efflux of 14 C-labeled L-leucine, L-methionine, and L-histidine was detected (Fig. 8).

Requirement of Inorganic Monovalent Cations for Substrate
Recognition-ry ϩ LAT1-mediated L-lysine transport was not Na ϩ dependent, whereas the L-leucine transport was, although not completely, dependent on Na ϩ (Fig. 1b), consistent with previous observations on human and mouse y ϩ LAT1s (13,14). As reported for the erythrocyte system y ϩ L (20,21,28), Na ϩ decreased the K m value for L-leucine uptake while not affecting that of L-lysine uptake (Fig. 3a), suggesting that Na ϩ increases the apparent affinity of neutral amino acid substrates for the substrate-binding site of ry ϩ LAT1.
Based on the observation that ry ϩ LAT1 can still accept neutral amino acid substrates in the absence of Na ϩ , it is suggested that some other monovalent cations may also support the binding of neutral amino acids in the absence of Na ϩ (Fig. 1b). As shown in Fig. 3b, the K m value for L-leucine transport at pH 5.5 was lower than that at pH 7.5. Apparently, this is not due to the conformational changes of the transporter protein with the change of pH of the uptake solution, because L-lysine transport was not altered by the change of pH. It is proposed that H ϩ takes the place of Na ϩ in the absence of Na ϩ to support leucine transport. Besides Na ϩ and Li ϩ , H ϩ is also capable of supporting the binding of neutral amino acid substrates to the substrate-binding site of ry ϩ LAT1 (Fig. 9b).
Transport of Inorganic Monovalent Cations with Amino Acids-ry ϩ LAT1 not only requires Na ϩ for the binding of neutral amino acids to its substrate-binding site, but it also translocates the Na ϩ along with neutral amino acids. As shown in Fig.  4, although L-leucine transport was accompanied by 22 Na ϩ uptake, the transport of L-lysine was not coupled with 22 Na ϩ transport, consistent with the observation that L-leucine transport was Na ϩ -dependent, whereas that of L-lysine was not dependent on Na ϩ (Fig. 1b). The Na ϩ :L-leucine coupling ratio was 1:1 when measured in the presence of 40 mM Na ϩ .
In the presence of trace concentrations of Na ϩ , however, the number of Na ϩ coupled with L-leucine was markedly reduced. Considering the pH dependence of [ 14 C]L-leucine uptake in the absence of Na ϩ , it is supposed that H ϩ instead of Na ϩ is cotransported with L-leucine in the presence of trace concentrations of Na ϩ . In fact, as shown in Fig. 5, L-leucine induced intracellular acidification in CHO cells cotransfected with ry ϩ LAT1 and 4F2hc cDNAs in the absence of extracellular Na ϩ , lending support to the assumption of H ϩ cotransport. H ϩ was cotransported with L-leucine in the absence or presence of a low concentration of Na ϩ but not with L-lysine. Considering the low Na ϩ :L-leucine coupling ratio (0.00015:1, see "Results") in the presence of 0.28 M Na ϩ at pH 7.4 (0.04 M H ϩ ), the affinity for H ϩ seems to be higher than that for Na ϩ , provided that most of the leucine is cotransported with H ϩ under this condition. In agreement with this observation, the affinity of the Na ϩ /glucose cotransporter, which also accepts H ϩ in the absence of Na ϩ , is also much higher for H ϩ than that for Na ϩ (29).
Proposed Mechanisms of Substrate Recognition-As indicated above, Na ϩ , H ϩ , or Li ϩ is required for neutral amino acids to be accepted by ry ϩ LAT1. Therefore, the positive charge on the side chains of basic amino acids or that conferred by inorganic monovalent cations is presumed to be required for the recognition of amino acid side chains by the substratebinding site of ry ϩ LAT1 (Fig. 9, a and b). It should be noted that ry ϩ LAT1 is structurally related to the system y ϩ basic amino acid transporters CAT1, CAT2, CAT3, and CAT4 (ϳ30% identity at the amino acid level) (1,6,21). The substratebinding sites of CATs bind specifically with basic amino acids. There is, however, an interesting exception: neutral amino acids, in particular L-homoserine, are accepted with low affinity by CAT1 in the presence of Na ϩ (30), indicating that CAT1 behaves like a system y ϩ L transporter, particulary for L-homoserine. It is, therefore, suggested that the binding sites of ry ϩ LAT1 and CAT1 recognize the positive charges on the side chains of substrate amino acids by a similar mechanism, consistent with the structural similarity between the two proteins.
Considering the requirement of positive charges for substrate binding, it is reasonable to assume that the binding sites FIG. 5. pH changes associated with amino acid transport. The fluorescence intensities in 2Ј7Ј-bis(carboxyethyl)-4(or 5)-carboxyfluorescein -preloaded CHO cells coexpressing ry ϩ LAT1 and 4F2hc were monitored, before (a, d), 1 min (b, e), and 2 min (c, f) after the addition of 1 mM L-leucine to the Na ϩ -free medium (a, b, c) or Na ϩ -containing medium (d, e, f) (see "Experimental Procedures"). While L-leucine (1 mM) induced a decrease in fluorescence intensity, indicating a decrease of intracellular pH in the Na ϩ -free medium (a, b, c), L-leucine did not alter the intracellular pH in the Na ϩ -containing medium (d, e, f). The fluorescence intensities were visualized by the gradation of color values: white, ϳ4000; red, ϳ3500; yellow, ϳ2700; green, ϳ2000; blue, ϳ1000; purple, ϳ250; each value indicates an arbitrary score of the fluorescence intensity. of ry ϩ LAT1 and possibly those of the other system y ϩ L transporters (KIAA0245/y ϩ LAT2 from humans (13) and SPRM-1 from Schistosoma mansoni (9)) contain negatively charged amino acid residues to accept the positive charges of the basic amino acid side chains or of the inorganic cations with neutral amino acid side chains (Fig. 9, a and b). The positive charge recognition site is probably located at a certain distance from the ␣-carbon recognition site. Therefore, for neutral amino acids to be the substrates of ry ϩ LAT1, the length of the side chains must be within a limited range, so that the side chains accompanied with inorganic cations can fit well into the binding site of ry ϩ LAT1. L-Leucine, L-isoleucine, L-methionine, and L-glutamine, which are accepted by ry ϩ LAT1, are predicted to meet this requirement.
The positive charge recognition site of ry ϩ LAT1 accepts inorganic cations with the following proposed order of affinity: H ϩ Ͼ Li ϩ Ͼ Na ϩ Ͼ Ͼ K ϩ . Some of the Na ϩ -binding sites of Na ϩ -dependent transporters accept not only Na ϩ but also other inorganic monovalent cations such as H ϩ and Li ϩ . As already mentioned, the Na ϩ /glucose cotransporter accepts H ϩ in the absence of Na ϩ (29). In glutamate transporters, it is proposed that one of the three Na ϩ -binding sites is specific for Na ϩ , whereas the other two are less specific and also accept Li ϩ (31). The broad cation selectivity of the positive charge recognition site of ry ϩ LAT1 is reminiscent of the less specific Na ϩ -binding sites of Na ϩ -dependent transporters. The positive charge recognition site of the system y ϩ L transporters may be a primitive form of Na ϩ -binding sites; hence, system y ϩ L transporters might be the evolutionary link between Na ϩ -independent transporters and Na ϩ -dependent transporters.
Amino Acid Exchange Property-Consistent with the Xenopus oocyte endogenous system y ϩ L activity stimulated by the injection of 4F2hc cRNA and the mouse y ϩ LAT1 coexpressed with 4F2hc in Xenopus oocytes (14,32), ry ϩ LAT1 mediated amino acid exchange (Fig. 6b). ry ϩ LAT1-mediated uptake was trans-stimulated by substrate amino acids (Fig. 6a). In addition, the extracellular substrate selectivity and Na ϩ dependence for the efflux of preloaded [ 14 C]L-lysine were basically identical to those of the uptakes of [ 14 C]amino acids (Fig. 7). These results indicate that the amino acid efflux is tightly coupled to amino acid uptake, suggesting that ry ϩ LAT1 mediates an obligatory exchange of substrate amino acids.
The efflux of intracellularly injected 14 C-labeled basic amino acids such as L-lysine, L-arginine, and L-ornithine was efficiently induced by extracellularly applied amino acid substrates (Fig. 8). In contrast, the efflux of 14 C-labeled neutral amino acids such as glutamine, leucine, isoleucine, and methionine was low, although they are also substrates of y ϩ LAT1 in the standard uptake solution. This is understandable when considering the low intracellular concentration of Na ϩ . It is assumed that a sufficient concentration of Na ϩ is required for neutral amino acid substrates to bind not only to the extracellular binding sites but also to the intracellular binding sites.
Proposed Functional Roles-It has been proposed that, in the renal proximal tubules and small intestine, basic amino acids are absorbed from the luminal fluid via system b 0,ϩ transporter situated on the apical membrane of the epithelial cells and pass through the basolateral membrane via system y ϩ L transporter into the extracellular fluid and blood stream (32). The results of our present investigation indicate that ry ϩ LAT1 can fulfill this task once it exists on the basolateral membrane of the epithelial cells along with 4F2hc, which has already been shown to be localized on the basolateral membrane (33). As shown in Fig. 8, intracellularly loaded basic amino acids, but not neutral amino acids, in fact, efficiently moved out of the cells via ry ϩ LAT1 through the amino acid exchange mechanism, probably because the intracellular substrate-binding site prefers basic amino acids to neutral amino acids due to the low intracellular concentrations of Na ϩ . Thus, ry ϩ LAT1 appears to be well suited as an exit path for basic amino acids. The proposed mechanisms of substrate recognition are schematically shown for ry ϩ LAT1 (system y ϩ L) (a, basic amino acids; b, neutral amino acids). The binding site is proposed to be composed of two sites: one for the binding of charged ␣-amino and ␣-carboxyl moieties (indicated by ϩ or Ϫ symbols near the ␣-carbon shown by C), and the other for the binding of the substrate amino acid side chains (indicated by the stub connected by a line). The side-chain-binding site of ry ϩ LAT1 is proposed to be equipped with the machinery to accept a positive charge. The charged amino acid residues indicated by ϩ or Ϫ symbols are proposed to be present at the substrate-binding sites.