Cloning and Functional Expression of a Human Na+and Cl−-dependent Neutral and Cationic Amino Acid Transporter B0+ *

A Na+-dependent neutral and cationic amino acid transport system (B0+) plays an important role in many cells and tissues; however, the molecular basis for this transport system is still unknown. To identify new transporters, the expressed sequence tag database was queried, and cDNA fragments with sequence similarity to the Na+/Cl−-dependent neurotransmitter transporter family were identified. Based on these sequences, rapid amplification of cDNA ends of human mammary gland cDNA was used to obtain a cDNA of 4.5 kilobases (kb). The open reading frame encodes a 642-amino acid protein named amino acid transporter B0+. Human ATB0+ (hATB0+) is a novel member of the Na+/Cl−-dependent neurotransmitter transporter family with the highest sequence similarity to the glycine and proline transporters. Northern blot analysis identified transcripts of ∼4.5 kb and ∼2 kb in the lung. Another tissue survey suggests expression in the trachea, salivary gland, mammary gland, stomach, and pituitary gland. Electrophysiology and radiolabeled amino acid uptake measurements were used to functionally characterize the transporter expressed in Xenopus oocytes. hATB0+ was found to transport both neutral and cationic amino acids, with the highest affinity for hydrophobic amino acids and the lowest affinity for proline. Amino acid transport was Na+ and Cl−-dependent and was attenuated in the presence of 2-aminobicyclo-[2.2.1]-heptane-2-carboxylic acid, a system B0+ inhibitor. These characteristics are consistent with system B0+ amino acid transport. Thus, hATB0+ is the first cloned B0+ amino acid transporter.

A Na ؉ -dependent neutral and cationic amino acid transport system (B 0؉ ) plays an important role in many cells and tissues; however, the molecular basis for this transport system is still unknown. To identify new transporters, the expressed sequence tag database was queried, and cDNA fragments with sequence similarity to the Na ؉ /Cl ؊ -dependent neurotransmitter transporter family were identified. Based on these sequences, rapid amplification of cDNA ends of human mammary gland cDNA was used to obtain a cDNA of 4.5 kilobases (kb). The open reading frame encodes a 642-amino acid protein named amino acid transporter B 0؉ . Human ATB 0؉ (hATB 0؉ ) is a novel member of the Na ؉ /Cl ؊ -dependent neurotransmitter transporter family with the highest sequence similarity to the glycine and proline transporters. Northern blot analysis identified transcripts of ϳ4.5 kb and ϳ2 kb in the lung. Another tissue survey suggests expression in the trachea, salivary gland, mammary gland, stomach, and pituitary gland. Electrophysiology and radiolabeled amino acid uptake measurements were used to functionally characterize the transporter expressed in Xenopus oocytes. hATB 0؉ was found to transport both neutral and cationic amino acids, with the highest affinity for hydrophobic amino acids and the lowest affinity for proline. Amino acid transport was Na ؉ and Cl ؊ -dependent and was attenuated in the presence of 2-aminobicyclo-[2.2.1]-heptane-2-carboxylic acid, a system B 0؉ inhibitor. These characteristics are consistent with system B 0؉ amino acid transport. Thus, hATB 0؉ is the first cloned B 0؉ amino acid transporter.
Amino acids are involved in biosynthetic pathways, act as neurotransmitters, and are essential for metabolic processes. Amino acids do not permeate cell membranes and therefore require specialized transport proteins in order to cross the plasma membrane (1). Transporters are classified based on sequence similarity, amino acid substrate specificity, and ion dependence. Ion-independent transporters carry amino acids according to their electrochemical gradient, whereas ion-coupled transporters use ion motive force to concentrate amino acids inside the cell (2).
Mammalian plasma membrane amino acid transporters have been functionally classified into two groups based on their Na ϩ -independent or Na ϩ -dependent mechanism of action (3,4). Two gene families encode transporters that mediate Na ϩindependent amino acid transport. One such family, the cationic amino acid transporters, CAT1-CAT4, carries lysine, arginine, and histidine and possesses much lower affinity for other amino acids (5). Na ϩ -independent amino acid transport is also induced by another family of proteins, which include 4f2hc (6, 7) and rBAT (8,9). These proteins are not transporters themselves but rather have recently been shown to form heteromultimers with other proteins, yϩLAT (10,11), LAT1 (12), or xCT (13), which show homology to the CAT family. The amino acid substrate specificity of these complexes depends on the specific subunit composition (10 -13).
Na ϩ -dependent transporters utilize the electrochemical gradients of Na ϩ and other ions to actively transport amino acids. There are two gene families that encode Na ϩ -dependent amino acid transporters. One Na ϩ -dependent transporter family includes the excitatory amino acid transporters that transport glutamate and aspartate, EAAT1-5, the transporters for alanine, serine, and cysteine, ASCT1 and ASCT2, and the neutral amino acid transporter hATB 0 (14,15). In addition to cotransport of amino acids and Na ϩ , members of this family have been reported to cotransport H ϩ and countertransport K ϩ (16,17). An additional amino acid transporter family utilizes Cl Ϫ along with Na ϩ to transport amino acids and other organic substrates into the cell (18,19). The Na ϩ /Cl Ϫ -dependent transporter family includes transporters for ␥-aminobutyric acidlike substrates (e.g. betaine and taurine), monoamines (e.g. serotonin and dopamine), and amino acids (e.g. glycine and proline) (20).
The recent cloning of transporter genes enables correlation between individual transport proteins and transport systems described in specific cell types or tissues (21). Transport systems for amino acids have been classically characterized based on amino acid specificity, ion dependence, and pharmacological properties (1). One such transport system, designated B 0ϩ , is defined by Na ϩ -dependent transport of both neutral and cationic amino acids (22). System B 0ϩ transport has been reported in mouse blastocysts (22), Xenopus oocytes (23-26), a human intestinal cell line (27), rabbit small intestine (28), rabbit conjunctiva (29,30), rat pituitary gland (31), bullfrog lung (32), and human lung (33). The molecular basis of this transport system is yet unknown.
In this study, we report the cloning and functional expression of a novel human amino acid transporter, hATB 0ϩ . 1 hATB 0ϩ is a member of the Na ϩ /Cl Ϫ -dependent neurotransmitter transporter family and transports both neutral and cationic amino acids in a Na ϩ -and Cl Ϫ -dependent manner. Substrate specificity and pharmacology indicate that hATB 0ϩ is the first molecularly characterized system B 0ϩ amino acid transporter.

EXPERIMENTAL PROCEDURES
Molecular Cloning-Gene-specific primers were paired with adaptor ligated sequence-specific primers, AP1 and AP2, for rapid amplification of cDNA ends (RACE) (34) using Advantage Polymerase TM (CLON-TECH). Primers were originally designed based on GenBank accession number AA526963 and were subsequently designed based on 5Ј and 3Ј RACE clones and correspond to nucleotides 534 -558 and 577-596 for 5Ј RACE and nucleotides 430 -455 and 1969 -1994 for 3Ј RACE (Life Technologies, Inc.). For primary 5Ј and 3Ј RACE, a gene-specific primer was paired with AP1 to amplify mammary Marathon Ready TM cDNA (CLONTECH). This PCR product was diluted 1:500 for secondary RACE reactions using a nested gene-specific primer and AP2. All 5Ј and 3Ј RACE products were subcloned into pCR2.1-TOPO and transformed into TOP10 Escherichia coli using the TOPO-TA cloning kit (Invitrogen, Carlsbad, CA). Individual colonies were screened for insert size with PCR using M13 forward and reverse primers. DNA was sequenced at the University's Automated DNA Sequencing Facility on a Model 373A DNA Sequencer using the Taq DyeDeoxy TM Terminator Cycle Sequencing Kit (Applied Biosystems, Foster City, CA). Lasergene (DNA Star, Madison, WI) and Basic Local Alignment Search Tool (BLAST) were utilized for sequence analysis (35). To determine genomic structure, the BLAST 2 Sequences program was used to align genomic sequences (GenBank accession numbers AL034411 and Z96810) with the hATB 0ϩ cDNA sequence. Protein motifs were identified using ScanProsite (36), and membrane topology was predicted by TMpred (37) and Kyte-Doolittle hydrophobicity analysis.
Northern and Master Blots-The probe was synthesized by PCR amplification of nucleotides 710 -1896 of the hATB 0ϩ cDNA and labeled by random priming with [ 32 P]dCTP (Amersham Pharmacia Biotech) to a specific activity of approximately 1 ϫ 10 7 cpm/ml using the Random Prime Labeling Kit (Roche Molecular Biochemicals). After prehybridization for 30 min at 68°C in ExpressHyb TM solution (CLONTECH), a human Multiple Tissue Northern blot and a human Master blot (CLONTECH) were hybridized with the cDNA probe in ExpressHyb TM for 1 and 6 h, respectively. The Northern blot was washed three times with 2ϫ SSC and 0.05% SDS at room temperature and twice with 0.1ϫ SSC and 0.1% SDS at 50°C. The Master blot was washed five times in 2ϫ SSC and 1% SDS at 65°C and twice in 0.1ϫ SSC and 0.5% SDS at 55°C. Both blots were subsequently exposed to a PhosphorImager screen (Molecular Dynamics) for 24 h or to autoradiography film for 48 h at Ϫ70°C.
Expression in Xenopus Oocytes-To obtain the open reading frame for functional analysis, two primers corresponding to nucleotides 74 -105 and 2001-2027 of the hATB 0ϩ cDNA sequence were designed. PCR amplification of mammary gland Marathon Ready TM cDNA with these primers yielded a single DNA band of ϳ1.9 kb that was ligated into pCR2.1-TOPO. For efficient expression in Xenopus oocytes, the hATB 0ϩ coding sequence was transferred into pKSPA, a modified pBluescript KSϩ plasmid (Stratagene, La Jolla, CA) containing (A) 30 , using XbaI and HindIII sites. One clone was subsequently used for all functional studies. The template for cRNA synthesis was prepared by NotI digestion of hATB 0ϩ -pKSPA or by PCR of hATB 0ϩ -pKSPA using M13 forward and reverse primers. cRNA was synthesized in vitro with T7 RNA polymerase (mMessage mMachine; Ambion, Austin, TX). Xenopus oocytes were surgically removed, treated, and selected as described previously (38,39). In brief, Xenopus laevis (Nasco, Fort Atkinson, WI) were anesthetized with 0.2% Tricaine. Oocytes were removed and treated with 4 mg/ml type 1A collagenase (Sigma) in a Ca 2 -free solution (82.5 mM NaCl, 2 mM KCl, 1 mM MgCl 2 , and 5 mM HEPES (pH 7.5) with 50 g/ml gentamycin) for 1 h at room temperature. Oocytes were rinsed with the Ca 2ϩ -free solution and then rinsed with modified Ringer's solution (96 mM NaCl, 2 mM KCl, 1 mM MgCl 2 , 1.8 mM CaCl 2 , and 5 mM HEPES, pH 7.5) supplemented with 0.5 mM pyruvate and 50 g/ml gentamycin. Stage IV oocytes were selected and injected the following day with 5-10 ng of cRNA or water in a total volume of 50 nl. Oocytes were maintained at 19°C in Ringer's solution supplemented with 0.5 mM pyruvate and 50 g/ml gentamycin. Three to seven days after injection, oocytes were used for electrophysiology or uptake experiments. All experiments were performed at room temperature (21°C).
Electrophysiology-Two electrode voltage clamp experiments were conducted using a GeneClamp 500 amplifier (Axon Instruments, Foster City, CA). Current was measured upon application of increasing concentrations of amino acids ranging from 1 M to 10 mM for ϳ15 s and washed for a period of ϳ30 s in Ringer's solution. In ion dependence experiments, the NaCl concentration of the Ringer's solution was 100 mM, and Na ϩ and Cl Ϫ were substituted with equimolar concentrations of N-methyl-D-glucamine and gluconate, respectively. All experiments were conducted at a holding potential of Ϫ80 mV. Clampex was used to acquire data at 62.5 Hz (pClamp 6; Axon Instruments). After digital filtering at 1 Hz, data were analyzed by Clampfit (pClamp 6; Axon Instruments).

RESULTS AND DISCUSSION
hATB 0ϩ Cloning Strategy-Sequence homology is commonly used to identify novel genes of emerging gene families. The expressed sequence tag (EST) database was queried to identify new members of the Na ϩ /Cl Ϫ -dependent transporter family. After the identification of an EST (GenBank accession number AA526963) from mammary gland cDNA with homology to this family, RACE of human mammary gland Marathon Ready TM cDNA was performed to clone the full-length gene. hATB 0ϩ cDNA is 4.5 kb in length and has been submitted to GenBank (GenBank accession number AF151978). hATB 0ϩ possesses absolute identity with human ESTs (GenBank accession numbers AA526963, AA552658, and AA541466) and is highly similar (ϳ90%) to mouse ESTs (GenBank accession numbers AI006618, AI006510, AA592728, AI1429024, and AI605513).
Primary Structure-hATB 0ϩ is a member of the Na ϩ /Cl Ϫ -dependent neurotransmitter transporter family and shows the highest similarity (ϳ60%) to the glycine transporters GLYT1 (40 -42) and GLYT2 (43) and the proline transporter PROT (44). The isolated hATB 0ϩ cDNA contains an open reading frame of 1926 base pairs, which predicts a protein of 642 amino acids (Fig. 1A). Hydrophobicity prediction (37,45) of the primary amino acid sequence suggests 12 putative membranespanning domains, similar to other Na ϩ /Cl Ϫ -dependent transporters (Fig. 1A). Analysis of the amino acid sequence by ScanProsite (36) reveals several consensus sites for post-translational modification (Fig. 1A). There are seven possible glycosylation sites on the second putative extracellular loop and one on the third putative extracellular loop. Two consensus sites for protein kinase C phosphorylation are located at Ser-40 and Ser-261. The Ser-40 site is also present in the amino acid transporters hGLYT2 and hPROT, and the protein kinase C consensus site located at Ser-261 is highly conserved among the Na ϩ /Cl Ϫ -dependent neurotransmitter transporter family (20). Phorbol esters, protein kinase C activators, have been reported to regulate Na ϩ /Cl Ϫ -dependent transporter activity (46 -50) and membrane localization (51,52). However, regulation may or may not be mediated by direct phosphorylation by protein kinase C (47,53). hATB 0ϩ also shows a consensus site for phosphorylation by casein kinase II in the fourth putative intracellular loop at Thr-434.
Genomic Structure-Periodic query of GenBank revealed genomic sequences that were identical to hATB 0ϩ (GenBank accession number AL034411 and Z96810), and alignment with hATB 0ϩ cDNA predicts the gene structure. The coding sequence possesses 14 exons, each of which is ϳ100 -200 base pairs in length (Fig. 1B). Genomic organization is conserved among members of the Na ϩ /Cl Ϫ -dependent transporter family, and the coding sequence of the transmembrane domains is not interrupted by introns (54). The genomic sequences (GenBank accession numbers AL034411 and Z96810) were assigned to chromosome X at positions Xq24 and Xq22.1-23, respectively. Interestingly, several forms of nonspecific mental retardation and other central nervous system disorders have been mapped to this region (55)(56)(57)(58)(59)(60)(61)(62).
Tissue Distribution-A human Master blot was probed to determine the tissue distribution of hATB 0ϩ mRNA. The highest expression was detected in the lung, fetal lung, trachea, and salivary gland, and lower levels of expression were detected in the mammary gland, stomach, and pituitary gland. Hybridization in the colon, uterus, prostate, and testis was very low ( Fig.  2A). ESTs from human mammary gland, colon, and prostate and from mouse colon are in agreement with the Master blot tissue distribution data. A multiple tissue Northern blot showed no expression in the heart, brain, placenta, liver, skeletal muscle, kidney; or pancreas; however, transcripts of ϳ4.5 and ϳ2 kb were detected in the lung (Fig. 2B). The predominant transcript of ϳ4.5 kb corresponds to the length of the hATB 0ϩ isolated cDNA, 4.5 kb.
Functional Studies-The application of substrate generated ionic current for all members of the Na ϩ /Cl Ϫ -dependent transporter family studied electrophysiologically (63). We therefore utilized a two-electrode voltage clamp to functionally characterize hATB 0ϩ in the Xenopus oocyte expression system. Oocytes injected with hATB 0ϩ cRNA generated inward current in response to the application of neutral and cationic amino acids. The negatively charged amino acids, glutamate and aspartate, evoked no current. We never observed a current greater than 2 nA in uninjected or water-injected oocytes in response to the application of 1 mM of each amino acid. Despite some seasonal and batch-to-batch variation in expression levels, more than 20 batches of oocytes injected with hATB 0ϩ cRNA responded to neutral and cationic amino acids. Amino acid-induced inward current was observed at all voltages from Ϫ140 to ϩ40 mV, and the current was increased at more negative potentials (data not shown). Fig. 3A illustrates the typical current evoked by increasing concentrations of amino acid (1 M to 1 mM of phenylalanine). Dose-response data for all amino acids that generated current were saturable. Data were initially fit to the Hill equation, and Hill coefficients were determined to be approximately 1. Subsequently, data were fit to a curve assuming a Hill coefficient equal to 1. Table I presents the EC 50 values for all amino acids evoking transport current. hATB 0ϩ preferred hydrophobic amino acids but also had significant affinity for other neutral and cationic amino acids. The apparent affinity for nonpolar amino acids seems to increase with R group size, and the apparent affinity for polar amino acids seems to decrease with R group size. The affinity for proline was very low (EC 50 Ͼ 5 mM) and probably would not be physiologically relevant.
In addition to the 20 naturally occurring amino acids, related compounds with modified side chain or "core" amino acid structure were tested. D-Tyrosine evoked transport current with an EC 50 Ͼ 1 mM compared with L-tyrosine with an EC 50 of 92 M, indicating that hATB 0ϩ recognizes amino acids stereospecifically. In addition, ␤-alanine and 3,4-dihydroxyphenylalanine (but not ␥-aminobutyric acid, choline, taurine, and thyroxine) evoked inward current at concentrations of 1 mM (data not shown). hATB 0ϩ had broad substrate specificity compared with its most similar family members, GLYT1 and GLYT2, which only transport glycine and glycine derivatives (40,43), and PROT, which transports proline with the highest affinity but also transports phenylalanine, histidine, and cysteine (44). Interestingly, an insect K ϩ -coupled amino acid transporter, KAAT1, with sequence similarity to this family was also found to transport a broad range of amino acids (64).
The current generated by amino acid application is believed to reflect transport across the plasma membrane, but in order to verify the physical translocation of the amino acid into the cell, 3 H-amino acid uptake experiments were conducted (Fig. 3,  B-D). Consistent with electrophysiological data, oocytes injected with hATB 0ϩ cRNA showed higher uptake rates for L-[

H]leucine and L-[ H]arginine but not for L-[ H]glutamate
when compared with uninjected oocytes (Fig. 3, B and C). The difference in uptake rate between leucine and arginine (Fig. 3, B and C) is in agreement with the differences in EC 50 values between leucine-and arginine-induced transport current (Table I). L-[ 3 H]Leucine uptake was significantly attenuated in the presence of 1 mM leucine (data not shown), 1 mM L-arginine, and 1 mM L-glutamine, but not in the presence of 1 mM Lglutamate (Fig. 3B). L-[ 3 H]Arginine transport was inhibited by 1 mM L-leucine (Fig. 3C). Because arginine inhibited uptake of  (Table I). Uptake experiments demonstrated that hATB 0ϩ transports neutral and cationic amino acids (e.g. arginine and leucine). The combination of electrophysiology and uptake experiments indicates that the current measured rep-resents the transport process. The transport current measurements can therefore be used to assess hATB 0ϩ substrate specificity and affinity for all amino acids tested.
Pharmacological studies are an important tool for amino acid transport system classification. High concentrations (5-10 mM) of 2-aminobicyclo-[2.2.1]-heptane-2-carboxylic acid (BCH), a cyclic amino acid, have been shown to inhibit system B 0ϩ amino acid transport (21)(22)(23)(24)26). In Fig. 3D, 10 2. Tissue distribution of hATB 0؉ mRNA. A, hybridization of a 32 P-labeled cDNA probe to a human Master blot (CLONTECH) was used to determine the tissue distribution of hATB 0ϩ mRNA. amygd, amygdala; caud, caudate nucleus; cereb, cerebellum; frlb, frontal lobe; hippo, hippocampus; md-ob, medulla oblongata; occip, occipital lobe; putam, putamen; sub-n, substantia nigra; temp, temporal lobe; thal, thalamus; stn, subthalamic nuclei; sp-cd, spinal cord; sk-mu, skeletal muscle; bladr, bladder; uter, uterus; prost, prostate; stom, stomach; pancr, pancreas; pitu, pituitary; adren, adrenal gland; thyr, thyroid gland; saliv, salivary gland; mam, mammary gland; kidny, kidney; sm-int, small intestine; thym, thymus; leuk, peripheral leukocyte; bone, bone marrow; appen, appendix; trach, trachea; plac, placenta. Bottom row samples are negative controls: yeast total RNA, yeast tRNA, E. coli rRNA, E. coli DNA, Poly r(A), and human C o t1 DNA. Positive controls include 100 and 500 ng of human DNA. The highest level of expression was detected in the lung, fetal lung, trachea, and salivary gland, and lower levels of expression were detected in the mammary gland, stomach, and pituitary gland. Hybridization in the colon, uterus, prostate, and testis was weak but detectable. B, hybridization with the same probe to a Multiple Tissue Northern blot (CLONTECH) revealed two transcripts of ϳ4.5 and ϳ2 kb in the lung. Members of the Na ϩ /Cl Ϫ -dependent transporter family require both Na ϩ and Cl Ϫ for transport to occur (65). Fig. 4A illustrates that hATB 0ϩ L-[ 3 H]leucine transport was strongly dependent on Na ϩ and Cl Ϫ ions. The hATB 0ϩ -related component of L-[ 3 H]leucine uptake was found to decrease by Ͼ99% in Na ϩ -free and Cl Ϫ -free solutions. In agreement with uptake data, no transport current was measured in Na ϩ -free solution (Fig. 4B, inset). On the other hand, a small but significant current was generated by 100 M leucine in Cl Ϫ -free solution. This current was approximately 6% of the total current evoked in the presence of 108 mM Cl Ϫ (Fig. 4C, inset). A similar small Cl Ϫ -independent current was also reported for the ␥-aminobutyric acid transporter, GAT1, expressed in Xenopus oocytes (39), indicating that external Cl Ϫ is not absolutely required for some transport to occur. Fig. 4, B and C, describes the effect of increasing concentrations of Na ϩ or Cl Ϫ on L-leucine (100 M)induced transport current. Dose-response curves were fitted to the Hill equation. For Na ϩ , a Hill coefficient of 2.3 Ϯ 0.13, an EC 50 of 7.4 Ϯ 0.24 mM, and an I max of 31 Ϯ 1.6 nA were determined (mean Ϯ S.E.; n ϭ 5). The Cl Ϫ data yielded a Hill coefficient of 0.92 Ϯ 0.07, an EC 50 of 0.61 Ϯ 0.03 mM, and an I max of 32 Ϯ 2.5 nA (mean Ϯ S.E.; n ϭ 5). EC 50 values of 7.4 and 0.61 mM for Na ϩ and Cl Ϫ , respectively, indicate that under physiological conditions, these ions are not rate-limiting for amino acid transport. A Hill slope of Ͼ2 for Na ϩ suggests that the transport cycle involves the binding of at least two Na ϩ ions, and Hill slopes for Cl Ϫ and amino acids close to 1 suggest the binding of one Cl Ϫ ion and one amino acid. Therefore, we propose a transport stoichiometry of 2 or 3 Na ϩ , 1 Cl Ϫ , and 1 amino acid.
Possible Physiological Significance of hATB 0ϩ -Amino acid transport through hATB 0ϩ can be summarized by the following characteristics: 1) inward current associated with neutral and cationic amino acid application; 2) uptake of and competitive inhibition by neutral and cationic amino acids but not anionic amino acids; 3) low affinity for proline; 4) uptake inhibition by the competitive substrate BCH (5-10 mM); and 5) Na ϩ and Cl Ϫ dependence. The properties of hATB 0ϩ are similar to the properties of a transport system originally described in mouse blastocysts, system B 0ϩ (22,66,67). System B 0ϩ is defined by Na ϩcoupled transport of neutral and cationic amino acids. This transport system also shows sensitivity to BCH at high concentrations (5-10 mM) (22)(23)(24)26). The similarity of amino acid specificity, ion dependence, and BCH sensitivity suggest that hATB 0ϩ is the first transporter to possess all system B 0ϩ characteristics.
The pituitary gland is of special interest because amino acids (e.g. arginine and leucine) are known to act as secretagogues for anterior pituitary hormones (68). Amino acid-induced hormone secretion was found to be induced by an intracellular rise in Ca 2ϩ and dependent on extracellular Na ϩ (31). Based on the amino acid specificity that caused an increase in intracellular Ca 2ϩ , Villalobos et al. (31) hypothesized that the amino acid influx is through a Na ϩ -dependent transporter. Similar substrate specificity and the possible expression of hATB 0ϩ in the pituitary gland suggest that hATB 0ϩ may play a role in amino acid-induced pituitary secretion. A transporter could regulate hormone secretion as a direct result of Na ϩ , Cl Ϫ , or amino acid influx or due to depolarization of the cell membrane. These hypotheses are currently being investigated in our laboratory.
Transport measurements in lung epithelial cells provide the strongest evidence for hATB 0ϩ -mediated system B 0ϩ amino acid transport. Galietta et al. (33) reported a potential B 0ϩ transport system in cultured human bronchial epithelial cells using uptake and short circuit current measurements. They observed Na ϩ -dependent transport current in response to the application of L-arginine, L-lysine, and L-alanine with EC 50 values of 80, 66, and 26 M, respectively, and a much lower affinity for proline. Also, L-aspartate and taurine did not produce short circuit current in these cells. The reported amino acid transport in human bronchial epithelial cells may be mediated by hATB 0ϩ due to its high expression in the lung and corresponding substrate specificity and affinity. Therefore, hATB 0ϩ could play a significant role in the removal of amino acids, Na ϩ , and Cl Ϫ from the airway surface liquid. Localization of mRNA and protein and more extensive functional measurements will determine whether hATB 0ϩ underlies the amino acid transport in the lung, pituitary gland, and other tissues in which system B 0ϩ -like amino acid transport has been described.