INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

-ALANINE AND -fluoro--alanine (FAL) are the major products of uracil and 5-fluorouracil catabolism. This process occurs primarily in the liver and is specifically localized in parenchymal cells (15). In earlier studies by Diasio and co-workers (25) it was shown that high concentrations of FAL accumulated initially in kidney and liver, and later on it was found in several other tissues, including eyes and small intestine. These data suggested that an active concentrative process may be responsible. In unrelated studies by Miyamoto et al. (16) and Turner (23), the presence of an active Na⁺-dependent concentrative mechanism for -alanine and -amino acids was demonstrated in brush-border vesicles from small intestine and proximal renal tubules (23). Other studies in astrocytes and neurons also established the ability of these cells to concentrate -alanine by a Na⁺-dependent process (9, 14).

Our interest arises from studies on the localization of uridine degradation products in the liver (15). We confirmed that -[³H]alanine and [³H]FAL accumulated in the hepatocytes following incubation with [6-³H]uracil and [5,6-³H]fluorouracil. We also demonstrated these -amino acids were concentrated by a Na⁺-dependent process. Others have characterized the transport of taurine, another -amino acid, in rat hepatocytes, showing a single Na⁺-dependent transport system at low concentration of taurine (5, 8). The transport of taurine was shown to be competitively inhibited by -alanine and hypotaurine (8). Two distinct Na⁺-dependent -amino acid transport systems, one with high affinity and the other with low affinity and high capacity, have been characterized in two renal epithelial cell lines, LLC-PK and Madin-Darby canine kidney (MDCK), and in the proximal tubule cells from rabbit and human origin (10-12).

In this study we characterize the Na⁺- and Cl-dependent -alanine and FAL transporter and provide evidence that it is identical to the GABA transporter designated GAT-2 on the basis of kinetic, substrate, immunological, and inhibition studies.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

Chemicals. -[3-³H]alanine (87 Ci/mmol) and [³H]H₂O (18 Ci/mmol) were purchased from NEN (Boston, MA). [2,3-³H]GABA (91 Ci/mmol) and [carboxy-¹⁴C]inulin (10 mCi/mmol) were purchased from Amersham (Arlington Heights, IL). Taurine, hypotaurine, L- and D-alanine, -aminoisobutyric acid, and -aminoisobutyric acid were obtained from Sigma Chemical (St. Louis, MO). (R,S)-FAL was purchased from American Tokyo Kasei (Portland, OR). All other reagents were of the highest grade available.

Hepatocyte preparation. Rat hepatocytes, provided by the generous assistance of the Yale Liver Center, were isolated by the collagenase perfusion technique (2, 21). Male Sprague-Dawley rats (150-200 g) from Charles River Laboratories were anesthetized with 50 mg/kg pentobarbital by intraperitoneal injection. Heparin (500 U/100 g body wt) was injected into the abdominal vena cava, and the livers were perfused through the portal vein with calcium-free perfusion buffer (pH 7.4, 37°C) saturated with a gas mixture of 95% oxygen and 5% carbon dioxide at 38 ml/min for 10 min. Extending the preperfusion period beyond this time period reduced the metabolic and transport activity of both cell populations. The calcium-free perfusate was changed to buffer containing 0.025-0.05% collagenase (Boehringer Mannheim Biochemicals, Indianapolis, IN) and 0.5 mM CaCl₂. Perfusion was continued for 10 min at 38 ml/min. The liver was removed from the animal and combed into 50 ml of pH 7.4 L-15 medium (GIBCO, Grand Island, NY) saturated with O₂. The cell suspension was filtered through 45-µm mesh nylon fabrics (Tetko), and the filtrate was sedimented at 50 g. The hepatocyte pellet was washed three times with cold L-15 medium and resuspended at 4-6 × 10⁶ cells/ml in L-15 medium. Hepatocyte preparations obtained by Percoll centrifugation (21) were >95% pure, and the cell viability, determined by trypan blue exclusion, was in excess of 85% before use and at least 80% at the end of uptake experiments.

Transport studies. Hepatocyte suspensions were centrifuged at 750 g for 5 min in a Sorvall GLC-4 centrifuge. The pellet was washed twice with 150 mM Na⁺ or 150 mM choline (Na⁺ free) Hanks' balanced salt medium containing 5.5 mM D-glucose and 4 mM HEPES buffer (pH 7.4). In Cl-free transport experiments, NaCl and KCl were replaced by their corresponding gluconate salts. The pellet was resuspended in the appropriate medium to give a final cell density of 4-6 × 10⁶ cells/ml.

Western blot analysis. Rat hepatocytes and MDCK cells, transfected with GAT-2 cDNA (provided by Dr. Michael J. Caplan, Yale University School of Medicine, New Haven, CT), were solubilized in 2× SDS gel loading buffer (100 mM Tris · HCl, pH 6.8, 200 mM dithiothreitol, 4% SDS, 0.2% bromphenol blue, 20% glycerol) and heated to 100°C for 10 min. The lysate was separated on a discontinuous 7.5% SDS-polyacrylamide electrophoresis system (7, 13, 19). Separated proteins were transferred to polyvinylidene difluoride membranes (Hybond P, Amersham) and blocked with a 5% casein buffer (5% casein, 0.05% Triton X-100, 0.3 M NaCl, 50 mM citric acid, 0.3 M Tris base, pH 7.6). The membranes were incubated overnight at 4°C with polyclonal anti-GABA transporter antibody (R-22) diluted 1:100 in casein buffer (1). The membranes were washed three times (5 min each) with PBS buffer containing 0.05% Tween 20 (PBST) and incubated with 1:2,000 diluted secondary antibody (donkey anti-rabbit horseradish peroxidase conjugated antibody; Amersham) for 1 h at room temperature. The membranes were then washed with PBST (3 × 5 min) and detected with enhanced chemiluminescence reagents (Amersham).

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

Studies of the homeostasis and metabolic fate of uridine in the liver (15) indicated that -alanine, the primary breakdown product of uracil, remains in the hepatocytes in high concentrations when they are exposed to uracil, a process dependent on sodium (Fig. 1). Biphasic kinetics (Fig. 2) are observed with a high affinity concentrative component [Michaelis constant (K_m) 35.3 ± 4.4 µM and a maximal velocity (V_max) 22.7 ± 2.5 µM/min]. The apparent K_m of the concentrative transporter is significantly greater than circulating levels of -alanine and thus may be determinant under physiological conditions. A second Na⁺-dependent kinetic component is also present with an apparent K_m of at least 2 mM. The zero transport time course of accumulation of -alanine within the cells is also consistent with a very rapid passive diffusion component that leads to equilibration with the medium followed by active concentration, which only occurs in the presence of sodium. When sodium is replaced by either lithium or choline, -alanine still rapidly equilibrates with the intracellular water but is not concentrated (Fig. 1). In the presence of Na⁺, concentrative transport of -alanine, after an initial burst, remains linear for >20 min, and >90% of the radioactivity remained associated with -alanine.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Time course of -[3H]alanine uptake by rat hepatocytes. Cells were suspended in either Na+-containing medium () or Na+-free medium [Li+ replacement () and choline replacement (+)]. Final concentration of -alanine during incubation was 25 µM. After appropriate time intervals uptake was terminated by the oil-stop method as described in MATERIALS AND METHODS. Intracellular concentrations were based on intracellular water volume. Values are means ± SE of triplicate estimates.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Substrate dependence of total and Na+-dependent -alanine transport in rat hepatocytes. Hepatocytes were incubated in Na+-containing medium and in Na+-free medium with varying substrate concentrations. Transport was measured as described in MATERIALS AND METHODS. The Na+-dependent component of transport was obtained by subtracting, at each substrate concentration, the Na+-free (choline replacement) part from total (inset represents Eadie-Hofstee plot of -alanine uptake by hepatocytes at substrate concentrations up to 2 mM). Michaelis constant values for the high affinity process were determined as described.

To determine the stoichiometric relationship between sodium ion entry and the transport of uridine, the dependence of the rate of transport on sodium concentration was determined. A sigmoidal response was obtained (Fig. 3) which is consistent with a cooperative mechanism that requires at least two and probably three sodium ions for each mole of -alanine (Hill coefficient 2.6 ± 0.4). These results suggest a mechanism in which the binding of the first sodium ion increases the affinity of the carrier for the second sodium ion and the participation of a single sodium ion is insufficient to permit the active concentration of -alanine.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Effect of external Na+ concentrations on Na+-dependent -[3H]alanine transport. -alanine (25 µM) uptake in hepatocyte was determined in presence of varying Na+ concentrations. Isosmolarity was maintained by replacement of Na+ with choline. Each point is mean of triplicate determinations. Inset presents linear transformation of data as expressed by the Hill equation.

This Na⁺-dependent transport system was also Cl dependent. Substituting this anion with gluconate in the transport buffer reduced the rate of uptake and accumulation of -alanine to passive diffusion level (Fig. 4). Increasing concentrations of Cl resulted in a concentrative uptake of -alanine that exhibited a hyperbolic dependence, suggesting that one Cl is associated with the transport of one molecule of -alanine (data not shown).

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Time course of -[3H]alanine uptake by rat hepatocytes in presence and absence of Cl. Cells were suspended in either Cl-containing medium or Cl-free medium (gluconate replacement). Final concentration of -alanine during incubation was 25 µM. After appropriate time intervals uptake was terminated by oil-stop method as described in MATERIALS AND METHODS. Intracellular concentrations were based on intracellular water volume. Values are means of 2 experiments conducted in triplicate.

The Na⁺- and Cl-dependent transporter is highly substrate specific. L- or D-alanine, taurine, and -aminoisobutyric acid did not inhibit -alanine accumulations at concentrations up to 1 mM (Fig. 5). However, GABA and hypotaurine, the sulfonic acid analog of -alanine, were potent inhibitors of this process (Fig. 5 and Table 1). FAL, the primary catabolic product of 5-fluorouracil, also exhibits inhibitory properties on this transport system. All three compounds display competitive inhibition kinetics (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Substrate specificity of -alanine transport by rat hepatocytes. Hepatocytes were incubated in Na+-containing medium with -[3H]alanine (25 µM) for 20 min in absence or presence of -alanine structural analogs using range of inhibitor concentrations (0-1 mM). Relative velocities are expressed as percentage of rate of -alanine uptake determined in absence of -alanine analogs and represent the mean of triplicate determinations. -AIBA, -aminoisobutyric acid, FAL, -fluoro--alanine; HT, hypotaurine.

View this table: [in this window] [in a new window]   Table 1.   Inhibition of -alanine and GABA uptake by structural analogs in rat hepatocytes

Earlier studies using PCR (4) have demonstrated the existence of a specific GABA transporter, termed GAT-2, present in liver, kidney, as well as in brain. Because of its possible identity with the -alanine transporter, we have characterized the transport of GABA in hepatocytes and demonstrated an absolute dependence of the concentrative phase on the presence of Na⁺. A minor degree of concentration may have occurred when Na⁺ is replaced by Li⁺ but not choline (Fig. 6). It is apparent in this system as well as with -alanine that GABA rapidly enters cells by a Na⁺-independent system to approximately equilibrate with the medium. As with -alanine, the sodium-dependent uptake exhibits a high affinity for GABA (22.5 ± 2.8 µM) and a somewhat lower V_max (18.5 µM/min) (Fig. 6 and Table 1).

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Time course of [3H]GABA uptake by rat hepatocytes. Cells were suspended in l50 mM NaCl, l50 mM LiCl, or l50 mM choline chloride buffer and final concentration of GABA was 25 µM during uptake measurements. After appropriate time intervals uptake was terminated by oil-stop method as previously described. Values are means ± SE of triplicate determinations.

To establish the identity between the -alanine Na⁺- and Cl-dependent transporter in rat hepatocytes and the GAT-2 transporter, we probed protein extracts from hepatocytes with an anti-GABA transporter antibody R-22 (1). As shown in Fig. 7, an 80-kDa protein corresponding to the GAT-2 transporter was present in the hepatocytes extract and in the lysate from MDCK cells expressing GAT-2, used as positive control.

View larger version (55K): [in this window] [in a new window]   Fig. 7.   Western blot analysis of GAT-2 antigen in rat hepatocytes and nonparenchymal cells. Rat hepatocyte and MDCK cell proteins were separated on a 7.5% SDS-PAGE gel and analyzed by Western blot as described in MATERIALS AND METHODS. Lanes 1-3, hepatocyte lysates containing 100, 100, and 50 µg of total protein, respectively; lanes 4 and 5, extracts of MCDK cells transfected with GAT-2 cDNA.

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

The compartmentalization of uridine catabolism in liver resulting in the specific accumulation of -alanine and FAL, in the case of 5-fluorouracil, led to the characterization of this Na⁺- and Cl-dependent transporter for -amino acids and GABA, a system that had poor affinity for taurine. This transporter requires the presence of both specific cation and anion. Without one or the other no significant active transport of -alanine and GABA is observed. Thus to demonstrate the cation dependency of this transporter as in Fig. 1, the chloride ion must be present; similarly, as shown in Fig. 4, to prove anion dependency, the transport buffer must contain sodium ion.

The Hill coefficient of 2.6 ± 0.4 for Na⁺ and one Cl for -alanine is similar to other -alanine and -amino acid transport systems previously characterized in rat hepatocytes (5, 8), renal brush-border membranes (23), kidney cells derived from the proximal tubules (10, 11), and intestinal brush-border membranes (16). In all these transport systems, however, taurine was also transported and was able to efficiently inhibit the -alanine transporter in these tissues. In rat hepatocytes taurine had only a very limited inhibitory effect on -alanine uptake (~40%) even at 1 mM.

Kinetic characteristics, substrate specificity, and antibody recognition all suggest that the Na⁺- and Cl-dependent transporter here described is the GAT-2 transporter. Unlike other GABA transporters previously identified, studies from Borden et al. (4) have indicated that GAT-2 is present not only in the central nervous system but also in other nonneural tissues such as liver, kidney, and retina.

At the neural level GAT-2 is localized in the leptomeninges surrounding the brain, possibly suggesting its role in regulating GABA levels in the cerebrospinal fluid, therefore affecting GABAergic transmission indirectly, or playing a role in the osmoregulation. GAT-2 is absent in neuronal cultures or in neurons in vivo. The basolateral distribution of GAT-2 in MDCK cells (1) and its presence in the leptomeninges also suggest that this transporter is involved in the clearing of GABA from the cerebrospinal fluid in nonsynaptic regions (3).

Neurological toxicity has been reported following administration of 5-fluorouracil (17, 20), particularly for regimens using 24-h infusion of high-dose 5-fluorouracil (2,600 mg/m²) where approximately 6% of the patients developed treatment-related encephalopathy (24). The clinical manifestations include disorientation, confusion, agitation, seizure, stupor, and coma (24). The affinity of -alanine and FAL for the GAT-2 transporter raises the possibility that the neurotoxicity seen with 5-fluorouracil therapy may relate to an effect of FAL on the reuptake of GABA, an inhibitory neurotransmitter. Administration of FAL has been shown in cats and dogs to induce a drowsy and stuporous state and struggling that are similar to those produced by oral administration of 5-fluorouracil (18). Histologically two main changes were described: 1) vacuole formation and 2) necrotic lesions. The distribution of these effects was not diffuse but limited to the cerebellar nuclei and white matter, tectum, and tegumentum of the brain stem. The selective localization of the neuropathological changes may be due to the preferential accumulation of FAL and an inhibitory effect on the function of the GAT-2 transporter.