key: cord-0008176-fr8spifc
authors: Khunweeraphong, Narakorn; Nagamori, Shushi; Wiriyasermkul, Pattama; Nishinaka, Yumiko; Wongthai, Printip; Ohgaki, Ryuichi; Tanaka, Hidekazu; Tominaga, Hideyuki; Sakurai, Hiroyuki; Kanai, Yoshikatsu
title: Establishment of Stable Cell Lines With High Expression of Heterodimers of Human 4F2hc and Human Amino Acid Transporter LAT1 or LAT2 and Delineation of Their Differential Interaction With α-Alkyl Moieties
date: 2019-06-11
journal: J Pharmacol Sci
DOI: 10.1254/jphs.12124fp
sha: 5d119a4cad45a112b332047e35127b78608e5a10
doc_id: 8176
cord_uid: fr8spifc

System L is a major transport system for cellular uptake of neutral amino acids. Among system L transporters, L-type amino acid transporter 1 (LAT1) is responsible for the nutrient uptake in cancer cells, whereas L-type amino acid transporter 2 (LAT2) is a transporter for non-cancer cells. In this study, we have established HEK293 cell lines stably expressing high levels of human LAT1 and LAT2 forming heterodimers with native human 4F2hc of the cells. We have found that l-[(14)C]alanine is an appropriate substrate to examine the function of LAT2, whereas l-[(14)C]leucine is used for LAT1. By using l-[(14)C]alanine on LAT2, we have for the first time directly evaluated the function of human LAT2 expressed in mammalian cells and obtained its reliable kinetics. Using α-alkyl amino acids including α-methyl-alanine and α-ethyl-l-alanine, we have demonstrated that α-alkyl groups interfere with the interaction with LAT2. These cell lines with higher practical advantages would be useful for screening and analyzing compounds to develop LAT1-specific drugs that can be used for cancer diagnosis and therapeutics. The strategy that we took to establish the cell lines would also be applicable to the other heterodimeric transporters with important therapeutic implications.

Amino acid transport across the plasma membrane is mediated via amino acid transporters situated on the plasma membrane. Amino acid transporters were originally described as amino acid transport systems (1). Among them, system L, a Na + -independent neutral amino acid transport agency, is a major route for living cells to take up neutral amino acids including branched-chain or aromatic amino acids (1) . By means of expression cloning, we isolated a cDNA encoding the first isoform of system L transporter named LAT1 (L-type amino acid transporter 1, SLC7A5) (2) . Following LAT1, we furthermore identified the second isoform of system L transporter named LAT2 (L-type amino acid transporter 2, SLC7A8) (3) . LAT1 or LAT2 forms heterodimers via a disulfide bond with the heavy chain of 4F2 antigen (4F2hc), which is essential for the proper targeting of LAT1 and LAT2 to the plasma membrane (4) .

In normal tissues, mRNA of LAT1 has been detected † These authors contributed equally to this work. # Research Fellow of the Japan Society for the Promotion of Science *Corresponding author. ykanai@pharma1.med.osaka-u.ac.jp Published online in J-STAGE on July 31, 2012 (in advance) doi: 10 .1254/jphs.12124FP

With α-Alkyl Moieties in brain, ovary, testis, and placenta, whereas its protein has only been reported in the blood-brain barrier and placenta barrier at low levels (2, 4 -6) . LAT1 is now regarded as a cancer cell-type transporter because LAT1 is highly upregulated in malignant tumors and its expression is related to the poor prognosis of patients (7 -15) . LAT1 in cancer cells is, thus, supposed to provide cancer cells with neutral amino acids including several essential amino acids for nutrition (14, 16, 17) . The LAT1-specific high-affinity inhibitor has been developed and its antitumor effect has in fact been demonstrated in experimental animals (18) . LAT1 can also be involved in the tissueand cellular delivery of amino acid-related drugs. Anti-Parkinson drug L-DOPA, anticonvulsant gabapentin, and antitumor phenylalanine mustard melphalan are the substrates of LAT1 (15, 19, 20) . L-DOPA and gabapentin are supposed to be delivered to the brain by LAT1 at the blood-brain barrier, whereas LAT1 contributes to transport of melphalan to cancer cells (15, 19 -21) . Recently it has been proposed that L-[3-18 F]fluoro-α-methyltyrosine, a radiotracer of positron emission tomography (PET) for cancer diagnosis, and borono-phenylalanine for boron neutron capture therapy are accumulated in cancer cells by LAT1 expressed in cancer cells (22 -24) , indicating further applications of LAT1 to cancer diagnosis and therapeutics. In contrast, LAT2 is a normal cell-type transporter widely expressed in an animal body including the small intestine, kidney, brain, placenta, ovary, testis, skeletal muscle, and pancreas (3, 4, 15, 25) . In the epithelia of small intestine and renal proximal tubules, LAT2 is localized at the basolateral membrane and responsible for the transepithelial absorption of amino acids from the luminal fluid to blood (25, 26) . LAT2 as well as LAT1 participates in the permeation of amino acids and amino acid-related drugs such as L-DOPA through the blood-brain barrier and placenta barrier (27 -30) . One of the notable differences in the substrate selectivity of LAT1 and LAT2 lies in the acceptance of α-methyl moieties. LAT1 accepts α-methyl-L-tyrosine, α-methyl-L-phenylalanine, and α-methyl-L-DOPA as substrates, whereas LAT2 shows less interaction with these aromatic α-methyl amino acids (21, 30, 31) .

The chemical compounds targeting LAT1 and avoiding the interaction with LAT2 are, thus, proposed to be drug candidates for cancer diagnosis and therapeutics. Recently, it has been shown that LAT2 is upregulated in some non-cancer lesions (32) . Thus, LAT2-specific inhibitors might also be considered therapeutically. To develop such drugs and screen chemical compounds, it is essential to establish in vitro simple assay systems to analyze the interaction with LAT1 and LAT2 separately. However, establishing cell lines stably expressing heterodimeric transporters such as LAT1 and LAT2 suffers from some difficulties. Firstly, most of the host cell lines originally express endogenous LAT1-4F2hc heterodimeric complex, so that the overexpressed exogenous LAT1 or LAT2 has to expel endogenous LAT1 to form heterodimers with 4F2hc to appear on the cell surface. Because the amount of endogenous 4F2hc limits the level of cell surface expression of exogenous LAT1 or LAT2, it would be difficult to obtain high level of functional activity even on overexpression of LAT1 or LAT2. Secondly, because host cells or mock cells exhibit high level of system L activity due to the endogenous LAT1-4F2hc heterodimer, it is difficult to distinguish exogenously introduced LAT1 or LAT2 from endogenous LAT1 activity.

We previously generated mouse S2 cell lines stably expressing human LAT1 (S2-hLAT1 cell) and human LAT2 (S2-hLAT2 cell) (30) . The S2 cell was chosen because it was an easy host cell to establish stably transfected cell lines (33) . To evaluate the expressed functions, we used L-[ 14 C]leucine as a substrate for both LAT1 and LAT2 because L-leucine is a common substrate for LAT1 and LAT2. L-[ 14 C]leucine uptakes in the S2-hLAT1 and S2-hLAT2 cells measured in Na + -free medium were supposed to be mediated by exogenously expressed human LAT1 and human LAT2, respectively (30) . Although the contribution of LAT1 to L-[ 14 C]leucine uptake of S2-hLAT1 cells has been confirmed by LAT1specific inhibitor KYT0353 (18) , that of LAT2 to L-[ 14 C]leucine uptake of S2-hLAT2 cells has not been proved directly because LAT2-specific inhibitors are not available for the moment. Additionally, S2 cell requires various growth factors and less common culture medium for maintaining (30) , so that it is not desirable for largescale screening of chemical compounds. Further concern on the S2-hLAT1 and S2-hLAT2 cells is that the exogenously introduced human LAT1 or human LAT2 forms heterodimers with endogenous mouse 4F2hc. In this study, to overcome such disadvantages of S2-hLAT1 cells and S2-hLAT2 cells, we have used conventional HEK293 (Human Embryonic Kidney 293) (34) as a host cell to establish new cell lines stably expressing high levels of human LAT1 and human LAT2 and searched for more appropriate substrates to evaluate LAT2 function compared with L-[ 14 C]leucine. Additionally, using newly established and validated cell lines and assays, we have examined the differential interaction of LAT1 and LAT2 with α-alkyl moieties by introducing wider spectrum of compounds.

All chemicals used in this study were of analytical Fig. 1 .

HEK293 cells were obtained from HSRRB (Osaka) and maintained in Minimum Essential Medium (MEM, Wako) containing 10% (v/v) heat-inactivated fetal bovine serum (Invitrogen, San Diego, CA, USA) supplemented with Non-Essential Amino Acid (Wako) under humidity with 5% CO 2 at 37°C. The full-length cDNAs for human LAT1 (GenBank/EMBL/DDBJ accession no. AB018009) and human LAT2 (GenBank/EMBL/DDBJ accession no. AB037669) were subcloned into pcDNA3.1(+), a mammalian expression vector (Invitrogen), to obtain pcDNA3.1(+)-hLAT1 and pcDNA3.1(+)-hLAT2, respectively. HEK293 cells were seeded onto a 60-mm dish at a density of 2 × 10 5 cells/dish and cultured for 36 h. At 60% -70% confluence, cells were transfected with 3 μg of pcDNA3.1(+) vector, pcDNA3.1(+)-hLAT1, or pcDNA3.1(+)-hLAT2 by using Lipofectamine TM 2000 (Invitrogen) according to the manufacturer's instructions. At 48 h after transfection, cells were subcultured into a 10-cm plate containing the culture medium supplemented with G418 (0.9 g/l). Cells were seeded at range between 0.5 -100 × 10 4 cells per plate to obtain individual colonies. The culture medium was changed every 2 days for 2 -3 weeks until the single colony was visualized. Once the colony was formed, a single colony was isolated by using a cloning cylinder. Cells in the isolated clone were trypsinized for 3 -5 min. Then, the cells were transferred into a 24-well plate containing 500 μl of the culture medium supplemented with G418. The cells were subcultured into a 6-well plate and subsequently 10-cm plates and maintained in the presence of G418. L-[ 14 C]Leucine uptake was measured for 30 clones of HEK293-hLAT1 cells stably expressing human LAT1 and 20 clones of HEK293-hLAT2 cells stably expressing human LAT2. Among the clones tested, a few clones that showed higher L-[ 14 C]leucine uptake compared with HEK293-mock cells stably transfected with vector plasmid were selected. The final selection relied on the stability of the uptake activities after 10 passages. HeLa S3 cells and S2 cells were cultured as described elsewhere (24, 30, 35) 

To isolate total RNA, cells were cultured as a monolayer in a 10-cm dish to confluence. Cells were washed with PBS and lysed in 1 ml of ISOGEN (NIPPON GENE, Tokyo). RNA isolation was performed following the manufacturer's instructions. The first-strand cDNA from the stable cell lines were generated by utilizing SuperScript TM III Reverse Transcriptase (Invitrogen). The specific primers designed to correspond to the sequences of LAT1, LAT2, LAT3 (GenBank/EMBL/ DDBJ accession no. NM_003627. 

For crude membrane preparation, cells cultured on 10-cm plates to confluence were collected and suspended in 500 μl buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 5 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM Na 3 VO 4 , and 1 mM NaF. Cells were then disrupted by ultra-sonication in ice-cold water. Cell debris was separated by centrifugation at 1,000 × g for 5 min at 4°C. The supernatants were ultracentrifuged at 45,000 × g for 30 min at 4°C. After the centrifugation, the crude membrane pellet was lysed with 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 1% NP-40. The protein concentrations were measured by BCA protein assay (Thermo Fisher Scientific, Waltham, MA, USA). The lysates were mixed with the Laemmli sample buffer with (reducing condition) or without (non-reducing condition) 100 mM dithiothreitol (DTT) and then subjected to SDS-PAGE.

Plasma membrane fractions were prepared as described previously (36) with minor modification. The cell pellet was suspended in homogenization buffer containing 10 mM Tris-HCl (pH 7.5), 250 mM sucrose, 100 mM NaCl, 1 mM EDTA, and Protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN, USA). The cells were homogenized and centrifuged at 1,000 × g for 5 min at 4°C. The supernatants were centrifuged at 430,000 × g in a Hitachi S100AT4 rotor (Hitachi Koki, Tokyo) for 15 min and the membrane pellet was resuspended in 0.75 ml of 30% iodixanol solution [20 mM Tris-HCl (pH 7.5), 1 mM EDTA, 30% (w/v) iodixanol (Axis-Shield PoC AS, Oslo, Norway), 125 mM sucrose]. The membrane suspension was overlayed by 3 ml each of 25%, 17.5%, 10%, and 2.5% iodixanol solutions sequentially and centrifuged at 100,000 × g in a Beckman SW41 Ti rotor (Beckman Coulter, Brea, CA, USA) for 16 h. One milliliter of each fraction was collected from the top. All fractions were analyzed by SDS-PAGE and western blotting. After confirmation by western blotting with plasma membrane markers or organelle markers, fractions from the 2.5% -10% iodixanol interface were pooled as the plasma membrane fraction. The protein concentrations were measured by BCA protein assay. The membrane fractions were dissolved in 1% Fos-Choline-12 ( Affymetrix, Santa Clara, CA, USA), mixed with the Laemmli sample buffer, and subjected to SDS-PAGE.

After SDS-PAGE, the separated proteins were transferred electrophoretically to a Hybond-P PVDF transfer membrane (GE Healthcare Life Sciences, Pittsburgh, PA, USA). The membrane was pre-blocked in blocking solution [5% (w/v) skim milk] at room temperature for 1 -2 h. The membrane was then incubated with the blocking solution containing 1:10,000 dilution of antibody, either affinity-purified rabbit anti-LAT1 C-terminus polyclonal antibody (19) , rabbit anti-LAT2 antiserum against amino acid residues from 505 to 516 (30), or anti-CD98 (H300) antibody against 4F2hc (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The membrane was treated with horseradish-peroxidase-conjugated antirabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) as the secondary antibody. The signal was developed by using Immobilon Western Chemiluminescent HRP Substrate (Millipore, Billerica, MA, USA) and visualized under the LAS-4000 mini Luminescent image analyzer (Fujifilm, Tokyo).

The cells were harvested from a 10-cm dish and lysed in the lysis buffer [50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, and Protease Inhibitor Cocktail (Roche Applied Science)]. The cell lysates were incubated overnight at 4°C with either control mouse IgG (Santa Cruz) or anti-4F2hc monoclonal antibody HBJ127 (37) provided by Dr. Yasuhiko Ito, Chubu University. The lysates were mixed with Dynabeads protein G (Invitrogen) and incubated for 1.5 h with rotation at 4°C. The beads were washed with the lysis buffer and resuspended in the Laemmli sample buffer. The samples were heated at 65°C for 10 min and then subjected to SDS-PAGE. The separated proteins were detected by western blotting as described above.

The functional clones of stable cell lines were analyzed in the uptake experiments as described previously (30, 38) . Briefly, HEK293-mock, HEK293-hLAT1, and HEK293-hLAT2 cells were seeded at the density of 1 × 10 5 cells/well onto collagen-coated 24-well plates. The cells were used for the uptake experiments two days after seeding when the cells reached to approximately 90% -100% confluence. After removal of the culture medium, the cells were carefully washed with prewarmed Na + -free HBSS (Hank's balance salt solution) containing 125 mM chlorine chloride, 4.8 mM KCl, 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , 1.3 mM CaCl 2 , 5.6 mM glucose, and 25 mM HEPES (pH 7.4) and then pre-incubated in 500 μl of pre-warmed Na + -free HBSS at 37°C for 10 min before adding substrates for the uptake experi-ment. The cells were then incubated at 37°C for 1 min in 250 μl of uptake medium: Na + -free HBSS containing L-[ 14 C]leucine (5.0 MBq/mmol) or L-[ 14 C]alanine (5.0 MBq/mmol). Subsequently, the cells were washed three times with ice-cold Na + -free HBSS. The cells were then lysed with 500 μl of 0.1 N NaOH and the lysate was mixed with 3.5 ml of Emulsifier safe cocktail ( PerkinElmer, Waltham, MA, USA). The radioactivity was measured by a β-scintillation counter (LSC-3100; Aloka, Tokyo). For the inhibition experiments, the uptake of 1 μM L-[ 14 C]leucine or L-[ 14 C]alanine was examined in the presence or absence of 1 mM non-radiolabeled test compounds. For T3 and T4, 10 μM of the compounds with 0.1% DMSO were used to examine their inhibitory effect on the uptake of L-[ 14 C]leucine and L-[ 14 C]alanine (1 μM) due to the low solubility of the compounds. The uptake was measured for 1 min because it was within the linear rage of the time dependent uptake of L-[ 14 C]leucine and L-[ 14 C]alanine. To obtain K m and V max values of the transport, the uptake of substrate was measured at 1, 10, 100, 300, 500, 1,000, 1,500, and 2,000 μM. Kinetic parameters, K m , V max , and K i , were determined by a nonlinear regression program in SigmaPlot 11 (Systat Software, Chicago, IL, USA).

All experiments were performed with 3 -5 replications. The data are expressed as means ± S.E.M. Statistical differences were determined using the unpaired Student's t-test. Differences were considered significant at P < 0.05.

The human cell lines established in this study, HEK293-hLAT1 and HEK293-hLAT2, stably expressing human LAT1 and human LAT2, respectively, were cloned under the selections for higher uptake of To confirm that the endogenous 4F2hc of HEK293 cells formed a heterodimer with exogenously expressed LAT1 or LAT2, the 4F2hc was immunoprecipitated by an anti-human 4F2hc monoclonal antibody (Fig. 2C ). As shown in the right panel of Fig. 2C , the anti-4F2hc antibody successfully precipitated the endogenous 4F2hc from the HEK293-mock, HEK293-hLAT1, and HEK293-hLAT2 cells (lanes 4 -6). Co-immunoprecipitated LAT1 was detected in the mock cells and predominantly in HEK293-hLAT1 cells ( 1 and 3) , the accumulation of LAT1 by co-immunoprecipitation made LAT1 in both cells detectable (Fig. 2C: left panel, lanes 4 and 6) . LAT2 was co-precipitated only from HEK293-hLAT2 cells, whereas the HEK293-mock and HEK293-hLAT1 cells showed no LAT2 band (Fig. 2C: middle panel, lanes 4 -6) . LAT1, LAT2, and 4F2hc were not co-precipitated with control IgG (lanes 7 -9 in all panels of Fig. 2C) .

The levels of expression of LAT1 and LAT2 on the plasma membrane were compared between stable cell lines established in the present study from HEK293 cells and those generated previously from S2 cells. As shown in Fig. 2D , HEK293-hLAT1 cells showed larger amount of LAT1 on the plasma membrane compared with S2-hLAT1 cells, whereas the level of LAT2 is similar between HEK293-hLAT2 cells and S2-hLAT2 cells (Fig.  2D ).

The properties of the uptake of L-[ 14 C]leucine and L-[ 14 C]alanine were compared among HEK293-mock, HEK293-hLAT1, and HEK293-hLAT2 cells. L-[ 14 C]Leucine uptake of HEK293-hLAT1 and HEK293-hLAT2 cells was higher than that of HEK293-mock cells because the stable clones that exhibited higher L-[ 14 C]leucine uptake were selected in the screening of clones (Fig. 3A) . The L-[ 14 C]leucine uptake of HEK293mock, HEK293-hLAT1, and HEK293-hLAT2 cells was all inhibited by a classic system L inhibitor, 2-aminobicyclo-(2,2,1)-heptane-2-carboxylic acid (BCH) (Fig.  3A ). In the present study, we, furthermore, measured L-[ 14 C]alanine uptake because L-alanine is a good substrate for LAT2 but not for LAT1 (3, 40) . As indicated in Fig.  3B , HEK293-hLAT2 cells showed high level of L-[ 14 C]alanine uptake. In contrast, the L-[ 14 C]alanine uptake of HEK293-mock and HEK293-hLAT1 cells was quite low (Fig. 3B) . The uptake of L-[ 14 C]leucine (1.0 μM) by HEK293-hLAT1 cells and that of L-[ 14 C]alanine (1.0 μM) by HEK293-hLAT2 cells both exhibited a linear dependence on the incubation time during the first 2 -3 min (data not shown). Thus, all the uptake measurements were conducted for 1 min. The uptake of L-[ 14 C]leucine in HEK293-hLAT1 cells and that of L-[ 14 C]alanine in HEK293-hLAT2 cells were saturable and followed Michaelis-Menten kinetics (Fig. 3C) . The kinetic parameters of the uptake were listed in Table 1 Michaelis-Menten curve and showed a similar K m value (56.3 μM) to that in HEK293-hLAT1 cells.

The uptake of L-[ 14 C]leucine by HEK293-hLAT1 cells and that of L-[ 14 C]alanine (1.0 μM) by HEK293-hLAT2 cells were measured in the presence of 1.0 mM nonradiolabeled L-amino acids (Fig. 4) . The uptake of L-[ 14 C]leucine by HEK293-hLAT1 cells was inhibited at higher levels by L-methionine, L-leucine, L-isoleucine, L-valine, L-phenylalanine, L-tyrosine, L-tryptophan, Lhistidine, and BCH and at lower levels by L-alanine, Lserine, L-threonine, L-cysteine, L-asparagine, and Lglutamine (Fig. 4A) . Glycine, L-proline, L-cystine, basic amino acids such as L-lysine and L-arginine, and acidic amino acids such as L-glutamate and L-aspartate caused little or no inhibition on L-[ 14 C]leucine uptake in HEK293-hLAT1 cells. The similar inhibition profile was obtained in the mock cells (data not shown). The L-[ 14 C]alanine uptake of HEK293-hLAT2 cells was strongly inhibited by all the neutral amino acids, Lhistidine and BCH, whereas L-proline, L-cystine, L-lysine, L-arginine, L-glutamate, and L-aspartate did not inhibit L-[ 14 C]alanine uptake (Fig. 4B) . It is noteworthy that both stably-transfected cell lines showed the same profiles of inhibition by amino acids even after 10 passages or recovering from frozen stocks in liquid nitrogen (data not shown).

Amino acid derivatives were also examined for the inhibitory effects on HEK293-hLAT1, and HEK293-hLAT2 cells. As shown in Fig. 5A , the ones with bulky side chains such as 3,3′,5-triiodothyronine (T3) and thyroxine (T4) (10 μM) inhibited the uptake of L-[ 14 C]leucine (1 μM) by HEK293-hLAT1 cells, whereas they did not inhibit the uptake of L-[ 14 C]alanine (1 μM) by HEK293-hLAT2 cells. The α-methyl amino acids such as α-methyl-L-tyrosine and α-methyl-L-phenylalanine (1.0 mM) strongly inhibited the uptake of L-[ 14 C]leucine (10 μM) in HEK293-hLAT1 cells, whereas they weakly inhibited the uptake of L-[ 14 C]alanine (10 μM) in HEK293-hLAT2 cells (Fig. 5B) . The inhibition of L-[ 14 C]alanine uptake by α-methyl-L-DOPA in HEK293-hLAT2 cells was prominently less than that by L-DOPA (Fig. 5B) . The level of inhibition was similar to that for α-methyl-L-tyrosine and α-methyl-L-phenylalanine in HEK293-hLAT2 cells, although the inhibition of L-[ 14 C]leucine uptake by α-methyl-L-DOPA in HEK293-hLAT1 cells was relatively less than that by α-methyl-L-tyrosine and α-methyl-L-phenylalanine (Fig. 5B) . The effects of α-methyl-alanine and α-ethyl-L-alanine were examined on LAT2 because L-alanine is a good substrate of LAT2. The inhibition of L-[ 14 C]alanine uptake by α-methyl-alanine and α-ethyl-L-alanine in HEK293-hLAT2 cells was less than that by L-alanine (Fig. 5B) . Furthermore, the inhibition by α-ethyl-L-alanine was significantly less than that by α-methyl-alanine (P < 0.01). Figure 6 illustrates the concentration-dependence of the inhibition of L-[ 14 concentration-dependent manner (Fig. 6A) . Similarly, L-alanine, BCH, and α-methyl-L-tyrosine concentrationdependently inhibited LAT2-mediated L-[ 14 C]alanine uptake (Fig. 6B) . IC 50 values of L-leucine, BCH, and α-methyl-L-tyrosine to inhibit L-[ 14 C]leucine uptake in HEK293-hLAT1 were 48.5, 53.9, and 75.4 μM, respectively. As for HEK293-hLAT2 cells, IC 50 values of Lalanine, BCH, and α-methyl-L-tyrosine to inhibit L-[ 14 C]alanine uptake were 151.7, 289.9, and 312.9 μM, respectively. The effects of BCH and α-methyl-L-tyrosine were further examined on the concentration dependence of LAT1-mediated L-[ 14 C]leucine uptake and LAT2mediated L-[ 14 C]alanine uptake. As shown in Fig. 6 , C -F, BCH and α-methyl-L-tyrosine shifted the concentration-dependence curves for both LAT1-mediated L-[ 14 C]leucine uptake and LAT2-mediated L-[ 14 C]alanine uptake. The inhibition kinetics of both BCH and α-methyl-L-tyrosine as well as L-leucine and L-alanine fit to the competitive inhibition by non-linear regression analysis, which was also confirmed by Lineweaver-Burk plots (data not shown). K i values are listed in Table 1 . K i value of α-methyl-L-tyrosine for LAT2 was approximately 9-fold of that for LAT1.

In this study, to establish the human cell lines stably expressing human LAT1 and human LAT2, we used HEK293 cells as host cells because endogenous LAT1 expression of HEK293 cells is lower than that of most human cell lines generally available in laboratories (data not shown). This property is beneficial to reduce the backgrounds of amino acid transport and to avoid the contamination of endogenous LAT1 activity particularly in LAT2-transfected cells. It was confirmed by RT-PCR and western blotting using antibodies against LAT1 and LAT2 that the exogenously transfected LAT1 and LAT2 were expressed at high level in HEK293 cells (Fig. 2: A  and B ). In fact, the LAT1 protein level of HEK293-hLAT1 cells was higher than that of HeLa S3 (Fig. 2B) , one of the cell lines with the highest LAT1 expression. Furthermore, predominant bands corresponding to the heterodimers of exogenously introduced LAT1 or LAT2 with endogenous 4F2hc of HEK293 cells were detected in non-reducing condition (Fig. 2B ). LAT1 and LAT2 were co-immunoprecipitated using the anti-4F2hc antibody (Fig. 2C ). This suggests that exogenously introduced LAT1 and LAT2 appear on the cell surface by forming heterodimers with endogenous 4F2hc, which is essential for cell surface localization of LAT1 and LAT2 (2, 3, 15) . Because of the overexpression of exogenously introduced LAT1 and LAT2, endogenous LAT1 of HEK293 cells would be competed out and not be able to form heterodimers with 4F2hc, so that exogenously introduced LAT1 or LAT2 would be predominant as partners of 4F2hc on the plasma membrane. Therefore, the cell lines we generated in the present study are proposed to almost exclusively express exogenously introduced LAT1 or LAT2 as the partner of endogenous 4F2hc on the cell surface with avoiding surface expression of endogenous LAT1 of HEK293 cells. This is also advantageous considering the formation of heterodimers between human 4F2hc and human LAT1 or LAT2, compared with S2-hLAT1 and S2-hLAT2 cells generated previously using a mouse cell line in which heterodimers of mouse 4F2hc and human LAT1 or LAT2 are formed (30) . Although the contribution of 4F2hc to the function of partner transporters has not been defined at the moment, the heterodimers of human LAT1 or LAT2 with human 4F2hc would be more desirable to evaluate the functions of human LAT1 and LAT2.

The clones of HEK293-hLAT1 and HEK293-hLAT2 cells that we selected in screening for higher uptake of substrates showed high levels of expression of LAT1 or LAT2. The level of expression of LAT1 in the plasma membrane fraction of HEK293-hLAT1 cells was much higher than that of S2-hLAT1 cells (Fig. 2D) . The level of expression of LAT2 in the plasma membrane fraction of HEK293-hLAT2 cells was similar to that of S2-hLAT2 cells which highly express LAT2 (Fig. 2D ). In the screening of cell clones with higher substrate uptake function, we were eventually able to obtain the clones with quite high 4F2hc protein levels (Fig. 2B) , which could explain the reason why the cell clones with higher surface LAT1 and LAT2 levels were obtained. In addition to the aforementioned benefit of forming heterodimers between human 4F2hc and human LAT1 or LAT2, HEK293-hLAT1 and HEK293-hLAT2 cells established in the present study, compared with previously generated S2-hLAT1 and S2-hLAT2 cells (30) , have the benefits of lower background of LAT1 in host HEK293 cells, higher level of expression of transfected LAT1 on the cell surface, easier handling of the established cell lines due to simple conventional culture medium required for the maintenance of the cells, and somewhat more stable expression of transfected LAT1 and LAT2. Such properties would be advantageous when HEK293-hLAT1 and HEK293-hLAT2 cells are used for the screening of chemical compounds interacting with LAT1 or LAT2 as well as in the kinetic analysis of their interactions.

In the previous studies conducted on S2-hLAT1 and S2-hLAT2 cells, we used L-[ 14 C]leucine as a substrate to characterize the functional properties of the stably expressing cell lines and to examine the interaction of chemical compounds with both LAT1 and LAT2 (18, 30) . Because the host or mock cells take up L-[ 14 C]leucine due to the endogenous LAT1-4F2hc complex on the plasma membrane and also because LAT2-specific inhibitors were not presently available, it was difficult for LAT2-expressing cells to evaluate the contribution of exogenously expressed LAT2 to L-[ 14 C]leucine uptake of the cells. In this study, we have found that L-[ 14 C]alanine is an appropriate substrate to evaluate the function of exogenously expressed LAT2 in HEK293 cells. L-[ 14 C]alanine uptake of the mock or host HEK293 cells was quite low, whereas HEK293-hLAT2 cells exhibited remarkably high L-[ 14 C]alanine uptake (Fig. 3B) . In contrast, L-[ 14 C]alanine uptake by HEK293-hLAT1 cells was low (Fig. 3B) . This clearly indicates that L-[ 14 C]alanine is specific to LAT2 in HEK293 cells in a Na + -free condition so that L-[ 14 C]alanine is a desirable substrate for evaluating the function of LAT2 introduced into HEK293 cells. By using HEK293-hLAT2 cells and L-[ 14 C]alanine as a substrate in the present study, we have succeeded for the first time to evaluate directly the function of human LAT2 expressed in mammalian cells and to obtain its reliable kinetic parameters for the transport.

The K m value of LAT2 for L-alanine uptake determined in this study using HEK293-hLAT2 cells (151 μM) agrees with that reported for LAT2 expressed in Xenopus oocyte (187 μM) (3) . Similarly, the K m value for L-leucine measured on HEK293-hLAT2 cells (153 μM), rather than that reported for S2-hLAT2 cells (318.0 μM) (30) , corresponds more to that obtained using the Xenopus oocyte expression system (119 μM) (3) . We propose that L-[ 14 C]alanine uptake measurement on HEK293-hLAT2 cells should enable the accurate functional evaluation of LAT2 in culture cells and the reliable estimation of the interaction of chemical compounds with LAT2. Inhibition profiles of HEK293-hLAT1 cells and HEK293-hLAT2 cells obtained in this study (Fig. 4: A and B) are basically similar to those reported previously for LAT1 and LAT2 examined in Xenopus oocytes (2, 3, 5) . This is consistent with the idea that endogenous 4F2hc of HEK293 cells is associated with exogenously introduced LAT1 and LAT2 in the HEK293-hLAT1 and HEK293-hLAT2 cells, respectively. Thus, the transport function of these stably transfected HEK293 cell lines reflects the properties of exogenously introduced LAT1 and LAT2.

As reported for S2-hLAT1 and S2-hLAT2 cells and for Xenopus oocyte expression systems (21, 30) , thyroid hormones, T3 and T4, and α-methyl aromatic amino acids such as α-methyl-L-tyrosine, α-methyl-Lphenylalanine, and α-methyl-L-DOPA are more selective to LAT1 compared with LAT2 in HEK293-hLAT1 and HEK293-hLAT2 cells (Fig. 5: A and B) . In the experimental scheme that we validated in the present study, we measured kinetic parameters of α-methyl-L-tyrosine for human LAT1 and human LAT2. α-Methyl-L-tyrosine inhibited both human LAT1 and human LAT2 in a concentration-dependent manner with K i values of 90.7 μM and 839.9 μM, respectively ( Fig. 6 and Table 1 ). This demonstrates for the first time in a quantitative manner that α-methyl-L-tyrosine is selective to LAT1. In this study, we furthermore, examined the effect of a wider spectrum of α-alkyl amino acids including α-methylalanine and α-ethyl-L-alanine, which have not been studied so far. We confirmed that LAT1 interacts with the α-methyl moiety (Fig. 5B) . In contrast, LAT2 less interacted with α-alkyl moieties. Although L-alanine strongly inhibited LAT2-mediated uptake, its α-alkyl derivatives such as α-methyl-alanine and α-ethyl-Lalanine inhibited LAT2 very weakly (Fig. 5B) . α-Ethyl-L-alanine interacted less with LAT2 compared with α-methyl-alanine, suggesting that bulkiness at the α-position interferes with the interaction with LAT2. Thus, the nature of LAT2 which does not accept the α-methyl moiety was further confirmed by using α-alkyl derivatives of LAT2 substrates. It was also the first demonstration that the α-ethyl group interferes with the interaction of amino acids with LAT2.

In this study, we also measured K i values of BCH on LAT1 and LAT2 in HEK293-hLAT1 and HEK293-hLAT2 cells to show that BCH is more selective to LAT1 compared with LAT2 ( Fig. 6 and Table 1 ). Since BCH was originally developed as a system L inhibitor by Christensen and co-workers (1), the differential sensitivity of system L isoforms to BCH has not been considered quantitatively. BCH has been in fact regarded as a general inhibitor of system L (1). In the previous study using S2-hLAT1 and S2-hLAT2 cells, we examined the effect of BCH on LAT1 and LAT2 and obtained the K i values to indicate that BCH is more selective to LAT1 than LAT2 (30) . Now we have confirmed it in the experimental scheme validated in the present study and concluded the LAT1-selectivity of BCH. In in vivo studies in which BCH was used to inhibit LAT1 expressed in cancers, BCH suppressed cancer growth with less side effects (12) . This could be explained by the LAT1-selective nature of BCH.

It has been shown that LAT1 is upregulated in tumor cells to supply tumor cells with amino acids to support their continuous growth and proliferation (2, 7 -20, 41) . In contrast, LAT2 is essential for the amino acid supply to non-cancer cells and for the epithelial transport of amino acids in the small intestine and kidney (3, 25) . Therefore, the development of LAT1-specific inhibitors or substrates that do not affect LAT2 should have important therapeutic and diagnostic implications. The LAT1selective inhibitors would suppress tumor cell growth by reducing amino acid supply to tumor cells so that they could be used as anti-tumor agents with less side effects on non-tumor cells. Furthermore, LAT1-selective substrates are expected to be selectively taken up and accumulate in tumor cells so that it could be used as probes for cancer diagnosis in PET and single photon emission computed tomography (SPECT). L-[3-18 F]Fluoro-αmethyl-tyrosine ([ 18 F]FAMT) (9) and L-[3-123 I]iodo-αmethyl-tyrosine ([ 123 I]IMT) (42, 43) , LAT1-selective α-methyl aromatic amino acids, have been used for PET and SPECT, respectively (24) . These LAT1-selective radiotracers are quite beneficial compared with the conventionally used glucose analogue PET tracer 2-[ 18 F]fluoro-2-deoxy-D-glucose ([ 18 F]FDG) and other amino acid PET tracers in the differentiation of malignant tumors and benign lesions as well as in reducing physiological backgrounds in the imaging (22, 24, 44, 45) . LAT1-selective substrates could also be used for the treatment of cancers if the compounds are taken up by the cells via LAT1 and exert anti-tumor effects: melphalan, a phenylalanine mustard, and borono-phenylalanine used for boron neutron capture therapy (19, 23) . For the delivery of amino acid-related drugs though the blood-brain barrier, it would be advantageous to design drugs to be transported by both LAT1 and LAT2 which exist in the barriers. The in vitro simple assay systems developed and validated in this study would be useful for generating such drugs interacting with LAT1 and/or LAT2.

In summary, we have established human cell lines stably expressing high level of heterodimeric complexes of human 4F2hc and human amino acid transporter LAT1 or LAT2 on the cell surface. We have, furthermore, established reliable assays to investigate the interaction of the transporters with chemical compounds. These cell lines that can be maintained in the conventional culture medium could also be used for the screening of compounds to develop drugs specific to LAT1 and/or LAT2 with desired pharmacodynamics and pharmacokinetics. In addition, the strategy we took in this study to establish cell lines stably expressing heterodimeric transporters would also be applicable to the other heterodimeric transporters. The drugs targeting xCT1 and Asc-1, members of the heterodimeric transporters, would also have important clinical implications (46, 47) .

Role of amino acid transport and countertransport in nutrition and metabolism

Expression cloning and characterization of a transporter for large neutral amino acids activated by the heavy chain of 4F2 antigen (CD98)

Identification and functional characterization of a Na + -independent neutral amino acid transporter with broad substrate selectivity

CATs and HATs: the SLC7 family of amino acid transporters

Human L-type amino acid transporter 1 (LAT1): characterization of function and expression in tumor cell lines

Expression of a system L neutral amino acid transporter at the blood-brain barrier

Prognostic significance of L-type amino acid transporter 1 (LAT1) and 4F2 heavy chain (CD98) expression in stage I pulmonary adenocarcinoma

Prognostic significance of L-type amino acid transporter 1 expression in resectable stage I-III nonsmall cell lung cancer

Comparison of L-type amino acid transporter 1 expression and L-[3-18 F]-alpha-methyl tyrosine uptake in outcome of non-small cell lung cancer

18 F-FMT uptake seen within primary cancer on PET helps predict outcome of non-small cell lung cancer

L-type amino-acid transporter 1 as a novel biomarker for highgrade malignancy in prostate cancer

L-type amino acid transporter 1 as a potential molecular target in human astrocytic tumors

Enhanced tumor growth elicited by L-type amino acid transporter 1 in human malignant glioma cells

Amino acid transporters ASCT2 and LAT1 in cancer: partners in crime?

Heterodimeric amino acid transporters: molecular biology and pathological and pharmacological relevance

System L amino acid transporter inhibitor enhances anti-tumor activity of cisplatin in a head and neck squamous cell carcinoma cell line

Bidirectional transport of amino acids regulates mTOR and autophagy

L-type amino acid transporter 1 inhibitors inhibit tumor cell growth

Characterization of the system L amino acid transporter in T24 human bladder carcinoma cells

Functional properties of multispecific amino acid transporters and their implications to transporter-mediated toxicity

Transport of amino acid-related compounds mediated by L-type amino acid transporter 1 (LAT1): insights into the mechanisms of substrate recognition

Fluorine-18-alpha-methyltyrosine positron emission tomography for diagnosis and staging of lung cancer: a clinicopathologic study

L-amino acid transporter-1 and boronophenylalanine-based boron neutron capture therapy of human brain tumors

Transport of 3-fluoro-L-α-methyl tyrosine by tumor-upregulated amino acid transporter LAT1: a cause of the tumor uptake in PET

LAT2, a new basolateral 4F2hc/CD98-associated amino acid transporter of kidney and intestine

Amino acid transport across mammalian intestinal and renal epithelia

Regulation of placental amino acid transporter activity by mammalian target of rapamycin

Expression of thyroid hormone transporters in the human placenta and changes associated with intrauterine growth restriction

Molecular and functional identification of large neutral amino acid transporters LAT1 and LAT2 and their pharmacological relevance at the blood-brain barrier

Establishment and characterization of mammalian cell lines stably expressing human L-type amino acid transporters

Characterization of the system L amino acid transporter in T24 human bladder carcinoma cells

Role of amino acid transporter LAT2 in the activation of mTORC1 pathway and the pathogenesis of crescentic glomerulonephritis

Regulation by protein kinase C of organic anion transport driven by rat organic anion transporter 3 (rOAT3)

Characteristics of a human cell line transformed by DNA from human adenovirus type 5

Interactions of human organic anion transporters with aristolochic acids

Murine coronavirus requires lipid rafts for virus entry and cell-cell fusion but not for virus release

Induction of c-Src in human blood monocytes by anti-CD98/FRP-1 mAb in an Sp1-dependent fashion

A novel role of the C-terminus of b 0,+ AT in the ER-Golgi trafficking of the rBAT-b 0,+ AT heterodimeric amino acid transporter

Human L-type amino acid transporter 1 (LAT1): characterization of function and expression in tumor cell lines

Alanine metabolism, transport, and cycling in the brain

High expression of L-type amino acid transporter 1 in infiltrating glioma cells

Isoform selectivity of 3-125 I-iodo-alpha-methyl-L-tyrosine membrane transport in human L-type amino acid transporters

Characterization of 3-[ 125 I]iodo-alpha-methyl-L-tyrosine transport via human L-type amino acid transporter 1

Evaluation of thoracic tumors with 18 F-FMT and 18 F-FDG PET-CT: a clinicopathological study

False-positive FDG PET uptake--the role of PET/CT

Identification and characterization of a Na + -independent neutral amino acid transporter that associates with the 4F2 heavy chain and exhibits substrate selectivity for small neutral D-and L-amino acids

CD44 variant regulates redox status in cancer cells by stabilizing the xCT subunit of system xc(-) and thereby promotes tumor growth

This work was supported in part by a Grant-in-Aid for Scientific Research on Priority Areas of 'Transportsome' from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science, the Osaka Medical Research Foundation for Intractable Diseases, Hyogo COE Program Promotion Project, and Ajinomoto Amino Acid Research Program. The authors are grateful to Michiko Minobe for technical assistance.