pH-dependent Heterogeneity of Acidic Amino Acid Transport in Rabbit Jejunal Brush Border Membrane Vesicles*

Initial rates of Na+-dependent L-glutamic and was- partic acid uptake were determined at various substrate concentrations using a fast sampling, rapid fil- tration apparatus, and the resulting data were ana-lyzed by nonlinear computer fitting to various transport models. At pH 6.0, L-glutamic acid transport was best accounted for by the presence of both high (K, = 61 PM) and low ( K , = 7.0 mM) affinity pathways, whereas D-aSpartiC acid transport was restricted to a single high affinity route (K, = 80 p ~ ) . Excess D- aspartic acid and L-phenylalanine served to isolate L-glutamic acid flux through the remaining low and high affinity systems, respectively. Inhibition studies of other amino acids and analogs allowed us to identify the high affinity pathway as the XiG system and the low affinity one as the intestinal NBB system. The pH dependences of the high and low affinity pathways of L-glutamic acid transport also allowed us to establish some relationship between the NBB and the more classical ASC system. Finally, these studies also revealed a heterotropic activation of the intestinal X ~ G transport system by all neutral amino acids but glycine through an apparent activation of V,,,.

Initial rates of Na+-dependent L-glutamic and waspartic acid uptake were determined at various substrate concentrations using a fast sampling, rapid filtration apparatus, and the resulting data were analyzed by nonlinear computer fitting to various transport models. At pH 6.0, L-glutamic acid transport was best accounted for by the presence of both high (K, = 61 PM) and low ( K , = 7.0 mM) affinity pathways, whereas D-aSpartiC acid transport was restricted to a single high affinity route (K, = 80 p~) .
Excess Daspartic acid and L-phenylalanine served to isolate Lglutamic acid flux through the remaining low and high affinity systems, respectively.
Inhibition studies of other amino acids and analogs allowed us to identify the high affinity pathway as the XiG system and the low affinity one as the intestinal NBB system. The pH dependences of the high and low affinity pathways of L-glutamic acid transport also allowed us to establish some relationship between the NBB and the more classical ASC system. Finally, these studies also revealed a heterotropic activation of the intestinal X~G transport system by all neutral amino acids but glycine through an apparent activation of V,,,.
The early studies of Gibson and Wiseman (1) have provided the first evidence that, in the rat small intestine, the uptake of acidic amino acids is a carrier-mediated process. Since this work, a substantial body of evidence, as recently reviewed (2,3), has accumulated showing a multiplicity of Na+-dependent and Na+-independent transport pathways for this class of amino acids in mammalian cells and tissues. Heterogeneity in Na+-dependent acidic amino acid transport systems has also been described in numerous cell types (2,3). For example, both high and low affinity systems for L-glutamic acid transport have been reported in membrane preparations of the central nervous system (2-4), in cultured human skin fibroblasts (5) and rat hepatocytes (6), and in rat (7) and rabbit (8) renal brush border membrane vesicles.
In the brush border membrane of the small intestine, contradictory results have appeared with regard to the heterogeneity of Na+-dependent acidic amino acid transport (2). For example, Lerner and Steinke (9) have reported a single transport system with a K , of  intestine. In this study however, whereas the uptake of 50 WM L-glutamic acid was markedly inhibited by D-aspartic acid and L-y-methylglutamic acid, it was also partially inhibited by a number of neutral amino acids such as L-proline, Lleucine, and L-alanine. More recent studies by Wingrove and Kimmich (10) did show the presence of both high ( K , = 16 WM) and low affinity ( K , = 2.7 mM) routes for the Na+dependent transport of L-aspartic acid in a preparation of isolated chick intestinal epithelial cells. Similarly, in studies using intestinal brush border membrane vesicles, Corcelli et al. (11) found a single transport system in the rat with a K , of 1.5 mM for L-glutamic acid and 1.0 mM for L-aspartic acid whereas, in human, Rajendran et al. (12) demonstrated the presence of a high affinity transport for L-glutamic acid with a K , of 91 p~ with no evidence for a low affinity transport pathway.
Transport studies that have examined the pH dependence of acidic amino acid transport have also led to contradictory results when performed under Na' gradient conditions alone but seem to agree quite closely when done under optimum gradient conditions of both inward Na+ and outward K+ (2). This observation, in conjunction with the fact that a transport system for neutral amino acids with properties similar to those described for system ASC, according to Christensen's nomenclature (13), may serve upon protonation as a low affinity pathway for acidic amino acids in those cell types for which inhibitor specificity and pH dependence have been studied (14)(15)(16), could actually provide a rationale for the lack of consistency in the characterization of intestinal acidic amino acid transport pathways. Accordingly, the presence in the brush border membrane of the small intestine of both an ASC-like transporter for neutral amino acids and a more specific, XiG type (13) of acidic amino acid transporter could have led to different results in the characterization of the acidic amino acid transport routes depending on the pH conditions of the uptake assay.
In this study, we have tested this hypothesis by determining the kinetics of both L-glutamic and D-aspartic acids under slightly acidic conditions (pH 6.0). The choice for these substrates was dictated quite naturally by the demonstration that the high affinity acidic amino acid transport system XiG displays no stereospecificity between L-and D-aspartic acids, whereas ASC-like systems never accept D-aspartic acid as substrate (13). Since transport heterogeneity was observed in the case of L-glutamic acid only, we then tried to delineate the two transport pathways by inhibition studies. Finally, we report the pH dependence of the two transport pathways. All together, these results do support the presence of both ASCand Xic-like systems in the rabbit jejunal brush border membrane.

Preparation of Brush Border Membrane
Vesicles-Two large batches of rabbit jejunal brush border membrane vesicles were pre-pared, as described recently (17), using a modified homogenate media in combination with Mg2+ precipitation to ensure both vesicle preequilibration and stabilization. Briefly, for each batch of vesicles, the jejunum of 16 rabbits (male, New Zealand White, 2.0-2.5 kg) was removed and flushed with ice-cold saline. The mucosal scrapings were homogenized at a 20:1 scrapings ratio (v/w) in the modified homogenate media containing a wide spectrum of phospholipase inhibitors (17). MgC1, was then added to give a final concentration of 10 mM, and the vesicles were prepared down to the second pellet (Pz). These were resuspended in a minimum volume of 50 mM Hepesl-Tris buffer, pH 7.0, containing 300 mM mannitol, combined, mixed, and divided into 500-pl aliquots that were frozen in liquid N,. On the day of vesicle preparation, a suitable number of aliquots were thawed and resuspended in the media required for the particular experiment (see descriptions in the figure and table legends) and prepared down to the final vesicle pellet (Pl). The vesicles, resuspended in the same media to give a final concentration of about 25 mg of protein/ml, were incubated overnight at 4 "C to ensure complete equilibration of the components of the resuspension media (17). On the next morning, the vesicles were divided into 25-pl aliquots suitable for individual uptake assays and frozen in liquid N:! until the time of assay to ensure complete stabilization of the specific activity of substrate uptake over the course of an experiment (17). Under these conditions, very similar uptake rates were routinely obtained from the same batch of vesicles when the same conditions of uptake were employed in separate experiments on different days. Also, since large batches of vesicles tend to reduce variations in uptake data due to animal differences, quite comparable results were obtained between the two batches used in these studies.
Transport Assays-Initial rates of Na'-dependent L-glutamic and D-aspartic acid uptakes were determined using the fast sampling, rapid filtration apparatus recently developed in our laboratory (18). For each assay, 20 p1 of vesicles were loaded into the apparatus, and uptake was initiated by injecting the vesicles into 480 pl of the uptake media required for the particular assay (see the description in the figure and table legends). Tracer uptakes were determined at 35 "C by a nine-point automatic sequential sampling of the uptake mixture at 1.5-s intervals. At each time point, the apparatus automatically injected 50 pl of the uptake mixture into 1 ml of ice-cold stop solution (see the description in the figure and table legends), filtered each stopped sample through 0.65-pm cellulose nitrate filters, and washed the filters three times with 1 ml of ice-cold stop. Radioactivities on the filters were then determined by liquid scintillation counting as described previously (19).
Data Analysis-Initial rates of [3H]~-glutamic acid and [%IDaspartic acid uptake were determined by linear regression over the nine-point time course of each assay, as described previously (18,20). The kinetic parameters of acidic amino acid transport or inhibition were estimated by nonlinear regression analysis using the standard errors of regression on the initial rates as weighting factors (20). Curve fitting of various transport or inhibition models to the nontransformed data was performed after proper transformation of the corresponding rate equations as justified recently (21) and described previously (20). Both linear and nonlinear regression analyses were performed using the Enzfitter program (R. J. Leatherbarrow;1987) and an IBM-compatible microcomputer. Only the best model fit to the data is reported in the figures, together with an Eadie-Hofstee transformation of the carrier-mediated process for visual appraisal of the goodness of fit (21).
Corp. Unlabeled D-aspartic and L-glutamic acids were purchased from Protein was measured with the BCA (bicinchoninic) assay kit from Pierce Chemical Co., using bovine serum albumin as a standard.

RESULTS
Kinetics of L-Glutamic and D-Aspartic Acids at pH 6.0-The initial rates of tracer L-glutamic acid and D-aspartic acid uptakes have been determined, as for all other experiments to be reported in this paper, under conditions of isotonicity and isoosmolarity and in the presence of an inwardly directed The abbreviations used are: Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; Mes, 4-morpholineethanesulfonic acid; Me-AIB, methylaminoisobutyric acid. Na+ gradient with the membrane potential clamped to 0 mV, using equal concentrations of the highly permeant anion iodide on both sides of the membrane (22).
A direct plot of the initial rates of [3H]~-glutamic acid uptake at pH 6.0 as a function of varying concentrations of unlabeled L-glutamic acid in the uptake media is shown in Fig. 1. The best model fit to these data was that of two transport systems with high (K,,, = 61 p~) and low (K,,, = 7.0 mM) affinities for the substrate working in parallel with diffusion. Fig. 1 (inset) shows that the curved nature of the Eadie-Hofstee transformation of the transport-mediated component is indeed compatible with the presence of more than one transport system and clearly demonstrates the goodness of fit.
Under the same assay conditions, Fig. 2 shows that the direct plot of the [3H]D-aspartic acid data has a completely different shape when compared with Fig. 1. In fact, the initial rates of tracer uptake remained constant at D-aspartic acid concentrations of greater than 1 mM and actually matched the values of unspecific transport of L-glutamic acid. The best model fit to the data in that case was that of a single high affinity system (K,,, = 80 p M ) plus a diffusional component with no evidence to support more complex models of substrate uptake as shown by the Eadie-Hofstee plot in Fig. 2 Effect of D-Aspartic Acid on the Kinetics of L-Glutamic Acid atpH 6.0-The good agreement between the K,,, for D-aspartic acid transport and that for L-glutamic acid transport through the high affinity pathway seems to indicate that these two amino acids might share a common transport system. Accordingly, it should be possible to block [3H]~-glutamic acid flux through the high affinity system by incorporating saturating concentrations of D-aspartic acid in the uptake media. The results of such an experiment are presented in Fig. 3, where the assay conditions were those of Fig. 1   of 5 mM D-aspartic acid in the uptake media. Under these conditions, tracer uptake rates of L-glutamic acid remained constant at unlabeled substrate concentrations of less than 1 mM, and the best model fit to the data was that of a single low affinity system ( K , = 9.4 mM) plus a diffusional component, as is also obvious from the Eadie-Hofstee plot in Fig. 3 (inset). The kinetic parameters of this system are comparable with the parameters of the low affinity system for L-glutamic acid uptake obtained in the absence of D-aspartic acid.
Inhibition of t-Glutamic Acid Uptake atpH 6.0 by D-Aspartic Acid and L-Phenylalanine-Using the kinetic parameters given in Fig. 1 and a substrate concentration of 50 PM, it can be calculated that about 20% of the total flux of L-glutamic acid should occur through the high affinity system, whereas the remainder should occur through the low affinity pathway and by diffusion across the membrane. As demonstrated in Fig. 4, increasing concentrations of D-aSpartiC acid in the uptake media caused a n inhibition of L-glutamic acid uptake that reached a maximum of 25% of the rate obtained in the absence of inhibitor. Moreover, the estimated Ki for competitive inhibition was 75 PM, in close agreement with the K, for   Fig. 1, except that 50 mM Lphenylalanine was incorporated into each uptake media in place of 52 mM acetate-Tris, pH 6.0. D-aspartic acid transport at pH 6.0 (Fig. 2). On the other hand, increasing concentrations of L-glutamic acid progressively reduced the initial rates of D-aspartic acid uptake down to the level of diffusion, and the estimated Kt for competitive inhibition was 129 ptM (results not shown), in good agreement also with the K,,, for L-glutamic acid transport through the isolated high affinity pathway at pH 6.0 (Fig. 5). These results thus demonstrate quite clearly that D-aspartic and L-glutamic acids share a common high affinity transport route in the rabbit small intestine.
The nature of the low affinity pathway for L-glutamic acid transport cannot be inferred from these studies, however. Assuming that such a low affinity pathway could actually be represented by a neutral, ASC-like system in the brush border membrane, as has been found in other cell types (14-161, we thus tried to inhibit L-glutamic acid uptake by phenylalanine, a neutral amino acid with large specificity for neutral amino acid carriers in the rabbit intestinal brush border membrane (23, 24). As shown in Fig. 4, addition of L-phenylalanine in the uptake media on top of saturating concentrations of Daspartic acid caused a further inhibition of L-glutamic acid uptake rates down to the level of diffusion. The estimated Ki for competitive inhibition by L-phenylalanine was 0.60 mM in that experiment.
If both L-glutamic acid and L-phenylalanine share the same transport system with low affinity for the former substrate, it should be possible to block [3H]~-glutamic acid flux through the low affinity pathway by incorporating saturating concentrations of L-phenylalanine in the uptake media. The results of such an experiment are presented in Fig. 5, where the assay conditions were those of Fig. 1 but for the addition of 50 mM L-phenylalanine in the uptake media. Under these conditions, tracer uptake rates of L-glutamic acid remained constant at unlabeled substrate concentrations in excess of 1 mM, and the best model fit to the data was that of a single high affinity system ( K , = 87 KM) plus a diffusional component, as was also obvious from the Eadie-Hofstee plot in Fig. 5 (inset). The K , of this system is identical with the K , of the high affinity pathway for L-glutamic acid transport obtained in the absence of L-phenylalanine (Fig. 1). However, the V,,, in the presence of L-phenylalanine was 8.85 k 1.35 pmol/mg of protein/s, a value that represents an apparent 4.2-fold activation of the maximal velocity of the high affinity transport system.
Effect of Acidic Amino Acids on L-Phenylalanine Uptake at pH 6.0 and 8.0-Complete inhibition of the low affinity system for L-glutamic acid transport at pH 6.0 by L-phenylalanine implies a shared route for these amino acids under these conditions. If this were the case, it should be possible to inhibit L-phenylalanine uptake by incorporating excess concentrations of L-glutamic acid in the uptake media. Fig. 6 shows the effects of 100 mM L-glutamic or D-aspartic acid on the initial rates of L-phenylalanine uptake at pH 6.0 and 8.0. L-Glutamic acid caused a 70% inhibition of uptake at pH 6.0 and had a small but nonsignificant effect at pH 8.0. In contrast, D-aspartic acid had no effect on the initial rates of uptake at either pH.
Effect of Amino Acids and Analogs on Uptake Rates of L -Glutamic and o-Aspartic Acids at p H 6.0-In order to get a better appraisal as to the nature of the transport systems involved in the high and low affinity routes for L-acidic amino acid transport, we have studied the effect of different classes of amino acids and analogs on the initial uptake rates of both D-aspartic and L-glutamic acids. Tables I and I1 show the results of such an experiment using 50 mM concentrations of these agents in the uptake media in comparison to a control run in the presence of mannitol. Using D-aspartic acid (Table  I), it appears that the acidic compounds, with the possible exception of D-glutamic acid, were all potent inhibitors, reducing uptake rates down to the level of diffusion obtained with 50 mM D-aspartic acid in the uptake media. None of the other compounds tested, however, could demonstrate any capacity to inhibit D-aspartic acid uptake. On the contrary, and quite interestingly in fact, it would appear that all of the neutral amino acids but glycine stimulated the uptake rates in excess of controls, thus suggesting a possible activation of the high affinity system by neutral amino acids. Both threonine and isoleucine proved the most efficient in this respect, with mean activations of 60% above controls.
With L-glutamic acid (Table 11), the situation is more complex, but it is quite clear that the addition of D-aspartic, L-cysteinesulfinic, and L-cysteic acids to the uptake media caused about 30% inhibition of 50 KM L-glutamic acid uptake, in accordance with the flux expected through the high affinity system at pH 6.0 and this substrate concentration (Fig. 1). L-Aspartic acid, however, caused a 65% inhibition of uptake, thus suggesting an inhibition of both the high and low affinity   Effect of amino acids and analogs on L-glutamic acid uptake rates at pH 6.0 Conditions for vesicle resuspension and uptake assays were the same as described in the legend to Table I of the diffusion-corrected initial rates of transport. The D-aspartic acid-sensitive component of total transport is the difference between total and D-aspartic acid-insensitive transport rates.
pathways. All of the neutral amino acids tested were found to inhibit uptake. Since these same compounds did not inhibit D-aspartic acid uptake, it would appear that the effect is best explained by an inhibition of the low affinity pathway for Lglutamic acid transport. However, since these same amino acids also stimulated D-aspartic acid transport, their effects may not have been maximum. The remaining imino acids, basic amino acids, and Me-AIB had no obvious effects on Lglutamic acid uptake rates.
pH-dependent Heterogeneity of L-Glutamic Acid Transport-In Fig. 7 are shown the initial rates of 50 p M L-glutamic acid transport in the presence and absence of 5 mM D-aspartic acid at varying pH from 6.0 to 8.0. Total flux at any given pH represents L-glutamic acid flux through both the high and low affinity transport systems, whereas the D-aspartic acid-insensitive component of total flux represents the isolated flux of L-glutamic acid through the low affinity system. By subtracting the mean uptake rates obtained in the presence of Daspartic acid from the total flux, one can thus estimate the fraction of L-glutamic acid transport occurring through the high affinity system. It is quite clear from Fig. 7 that the low and high affinity routes for L-glutamic acid transport have quite different pH dependences. Although the former is most active at the acidic pH of 6.0 and declines steadily with increasing pH such that it is almost inactive at pH 7.5 and 8.0, the latter showed an optimum pH around 7.0 with somewhat lower rates in moving to the basic or acidic sides of neutrality. This apparent "bell-shaped" pH dependence for acidic amino acid transport through the high affinity route was also obtained when D-aspartic acid was used as a specific substrate (Fig. 8), thus indicating that the D-isomer of aspartic acid behaves as a comparable substrate to L-glutamic acid, at least with regard to determining the pH dependence of the high affinity system.
In a first attempt to determine the effect of pH on the kinetics of the high affinity transport pathway for acidic amino acids, we have measured the initial rates of [%IDaspartic acid uptake at pH 8.0 as a function of varying concentrations of unlabeled D-aspartic acid. Fig. 9 shows the results of this experiment, and the best fit model was that of a single high affinity system ( K , = 79 pM) plus diffusion, with no evidence of a more complex model, as evidenced by the Eadie-Hofstee plot in the inset. Comparing the kinetic parameters for D-aspartic acid uptake at pH 6.0 (Fig. 2) and 8.0 (Fig. 9) and considering the pH dependence of transport (Fig. 8), it would appear that the pH effect might be entirely explained through an effect on V,,, with no effect on K,,,. DISCUSSION Using stabilized and fully equilibrated preparations of rabbit jejunal brush border membrane vesicles (17) and following a protocol of nonlinear regression analysis of nontransformed true initial rates of tracer uptakes (20,21), as determined with the fast sampling, rapid filtration apparatus (18), we found evidence for both high and low affinity Na+-dependent transport systems for L-glutamic acid at pH 6.0 (Fig. 1). Under the same conditions, however, the uptake of D-aspartic acid was restricted to a single high affinity system (Fig. 2) that was  Fig. 1 and 2, except for the buffer in the vesicle resuspension media and the uptake media, which consisted of 50 m M Hepes-Tris buffer, pH 8.0.
All other solutions were also adjusted to pH 8.0 when required.
shown to be identical with the high affinity route of L-glutamic acid transport (Figs. 3 and 4). Moreover, this high affinity pathway was completely inhibited by L-aspartic, L-glutamic, L-cysteinesulfinic, and L-cysteic acids (Tables I and 2) but was not inhibited by any other amino acids or analogs when using its specific substrate, D-aspartic acid (Table I). These basic properties are the same as those originally described for system XiG in the hepatocyte plasma membrane (6,(13)(14)(15).
Therefore, it is reasonable to identify the high affinity acidic amino acid transporter in the rabbit intestinal brush border membrane as a member of the system X~G family of transporters.
The transport of 50 p~ L-glutamic or D-aspartic acids by system Xic in the vesicles showed a pH optimum of 7.0 with declining rates of transport observed in shifting the pH toward 6.0 or 8.0 (Figs. 7 and 8). Bell-shaped pH-dependent profiles for L-glutamic acid have already been reported in brush border membrane vesicles from rabbit (25) and human (26) jejunum and from rat kidney (7), and in the hamster kidney cell line BHK21-Cl3 (27). Since the major substrate species at pH 7.0 is the anionic form of these amino acids (-99%) and since its concentration is increased by less than 1% when the pH is increased to 8.0 and drops by 7-9% when the pH is decreased to 6.0, it is quite clear from Fig. 8 that the pH-dependent variations in the initial rates of transport (22-28% from pH 7.0 to 8.0 and 38-44% from pH 7.0 to 6.0) cannot be ascribed solely to changes in the ionization state of the substrates. It must then be concluded that the anionic form of the substrates is the transported species and that the pH dependence of transport reflects changes in the ionization state of the carrier molecule itself. Accordingly, Berteloot and Maenz (2) have recently speculated that system XiG in the intestine and kidney possesses two H+-titratable regulatory sites with separate pK values and that optimal transporter function occurs when one site is protonated while the other site is in the deprotonated state. Moreover, it would appear that the pH effect may be entirely due to a V,,, effect (compare Figs. 2  and 9). Kinetic parameter determinations on a wider range of p H values may, however, be required to ascertain this conclusion.
The nature of the low affinity transport system for acidic amino acids was investigated using L-phenylalanine as a potential inhibitor since this amino acid is known to be a substrate for the PHE (phenylalanine) and NBB (neutral brush border) systems present in the intestinal brush border membrane (23, 24). In our vesicle preparation, at pH 6.0, Lphenylalanine did block the flux of L-glutamic acid through the low affinity pathway (Fig. 5) with kinetics compatible with those expected from a competitive inhibition (Fig. 4), and excess concentrations of L-glutamic acid caused a marked inhibition of L-phenylalanine uptake (Fig. 6). In addition, all of the neutral amino acids tested caused a partial inhibition of L-glutamic acid uptake at pH 6.0, whereas proline, arginine, methylaminoisobutyric acid, and lysine had no inhibitory effects (Table 11). These results rule out the possibility that systems IMINO, y+, A, or L, as defined by Stevens et al. (23) in the small intestine, represent the low affinity route for Lglutamic acid transport. Instead, these results provide good evidence that the low affinity pathway may actually represent L-glutamic acid flux through the NBB system.
Transport of 50 p~ L-glutamic acid through the low affinity system was marginal at the basic pH values of 8.0 and 7.5 but increased steadily thereafter as the pH decreases to 6.0 (Fig.  7). In agreement with these results, excess L-glutamic acid caused little or no inhibition of L-phenylalanine uptake at pH 8.0 (Fig. 6). This pH dependence is similar to that described for L-acidic amino acid flux through the neutral amino acid transport system ASC in the hepatocyte plasma membrane (13-15) for which it was hypothesized that a H+-titrable site regulates the functional capacity to transport acidic amino acids (13). These results thus indicate that the NBB system of the intestinal brush border membrane may belong to the ASC family of transport systems despite a substrate specificity that was found not to conform to the classical ASC pathway (24).
An unexpected finding during these studies was that most of the neutral amino acids tested, with the exception of glycine, produced an activation of the high affinity transport of acidic amino acids (Tables I and 11). At least at pH 6.0 and in the presence of 50 mM L-phenylalanine in the uptake media with L-glutamic acid as substrate, it would appear that this heterotropic activation occurs through a V,,, effect only (Fig.  5). Such an activation of system XiG by neutral amino acids has never been reported before in any cell type, and this property may well be specific to intestinal (and maybe renal) cells. It should be noted however that leucine has already been proposed as an allosteric modulator of the lysine transporter in the rat intestinal basolateral membrane (28). We are currently characterizing the effects of neutral amino acids on system XiG in our vesicle preparation using D-aspartic acid as a specific substrate.
In conclusion, this study demonstrates that L-glutamic acid is transported by both high and low affinity Na+-dependent transport systems at acidic pH, whereas D-aspartic acid serves as a specific substrate for the high affinity system. These findings bring to question many of the results obtained on the effects of varying pH and imposing H+ gradients across membranes using L-acidic amino acids as substrates, and they establish the necessity of using a specific substrate such as Daspartic acid in future research on characterizing system X~G in the intestinal epithelium.