Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

N-Acetylaspartate occurs at very high concentrations in the mammalian brain (1.5-10 µmol/g of tissue) where it is the second most abundant free amino acid next to glutamate (Tsai and Coyle, 1995; Baslow, 1997; Clark, 1998). Regional studies in the mouse brain have demonstrated the presence of N-acetylaspartate in all brain areas with highest concentrations in cerebral gray matter (Tallan, 1957). During maturation, N-acetylaspartate is found in neurons as well as in glial cells, but it is present primarily in neurons in the fully developed brain (Simmons et al., 1991; Urenjak et al., 1992). The intraneuronal concentration is 10 to 15 mM, whereas the extracellular concentration in the brain interstitial space is 80 to 100 µM (Taylor et al., 1994; Sager et al., 1997).

The most likely function of N-acetylaspartate in neurons is in osmoregulation, protecting these cells against osmotic stress (Taylor et al., 1995; Sager et al., 1997). It also may function as a precursor of the neuromodulator N-acetylaspartylglutamate (Blakely and Coyle, 1988). Interestingly, the enzyme responsible for the synthesis of N-acetylaspartate, called L-aspartate-N-acetyl transferase, is present exclusively in neurons, whereas the enzyme responsible for the breakdown of N-acetylaspartate, called aspartoacylase II, is present predominantly in glial cells (Benuck and D'Adamo, 1968; Goldstein, 1976). Thus, N-acetylaspartate is synthesized in neurons but is hydrolyzed in glial cells. This compound is probably released continuously from neurons and is subsequently taken up into glial cells for hydrolysis. This is consistent with the findings that neurons contain high levels of N-acetylaspartate, whereas glial cells contain low levels of this compound. The glial metabolism of N-acetylaspartate has been shown to be important for myelin synthesis because this compound is a major source of acetyl groups for lipid synthesis during brain development (Burri et al., 1991). A genetic disorder leading to aspartoacylase II deficiency, called Canavan disease, is associated with mental retardation and spongy degeneration of brain white matter (Matalon and Michals-Matalon, 1999).

Aspartoacylase II is a cytoplasmic enzyme in glial cells and therefore the extracellular N-acetylaspartate has to be first transported into the glial cells for the hydrolysis to proceed. N-Acetylaspartate exists as a divalent anion at physiological pH due to the acetylation of the amino group. Diffusion of this compound into glial cells is very unlikely because of the inside-negative membrane potential. This suggests that the glial cells must possess a transport system for this compound. This is supported by recent studies that show the existence of a transporter for N-acetylaspartate in rat glial cells (Sager et al., 1999). This transporter is saturable (K_t ~80 µM), Na⁺- and Cl-dependent, and does not interact with any of the naturally occurring amino acids. However, the molecular identity of this novel transport system has not been established. Herein, we show that the Na⁺-coupled high-affinity dicarboxylate transporter NaDC3 is capable of transporting N-acetylaspartate. The transport process is Na⁺-dependent and electrogenic and exhibits a K_t value of 100 to 250 µM for N-acetylaspartate. The characteristics of N-acetylaspartate transport via NaDC3 are exactly the same as those of N-acetylaspartate transport described in glial cells by Sager et al. (1999). In addition, we provide evidence for the expression of a Na⁺-coupled high-affinity dicarboxylate transporter in rat primary astrocyte cultures and for the interaction of the transporter with N-acetylaspartate. These studies demonstrate that NaDC3 is the transporter responsible for the uptake of N-acetylaspartate into glial cells. The molecular identification of the N-acetylaspartate transporter in the brain is of physiological and clinical significance because the function of this transporter is obligatory for the clearance of N-acetylaspartate from the interstitial space and its subsequent hydrolysis. Because a deficiency of aspartoacylase II is known to cause serious neurological complications, it is possible that defects in the transporter function also may lead to similar neurological problems.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. The human retinal pigment epithelial (HRPE) cell line 165, used in expression studies, was originally provided by M. A. Del Monte (W.K. Kellog Eye Center, Department of Ophthalmology, Ann Arbor, MI) and was routinely maintained in Dulbecco's modified Eagle's medium-F12 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 µg/ml streptomycin as described before (Huang et al., 1997). Frogs (Xenopus sp.) were purchased from Nasco (Fort Atkinson, WI) and mMESSAGE mMACHINE kit for cRNA synthesis was obtained from Ambion (Austin, TX). SuperScript plasmid system for cDNA cloning, TRIzol reagent, oligo(dT)-cellulose, and Lipofectin were purchased from Life Technologies, Inc. (Grand Island, NY). Restriction enzymes were purchased from New England Biolabs (Beverly, MA). Magna nylon transfer membranes were purchased from Micron Separations, Inc. (Westboro, MA). [2,3-³H]Succinic acid (specific radioactivity, 37.5 Ci/mmol) was purchased from Moravek Biochemicals (Brea, CA). N-Acetylaspartate, aspartate, glutamate, and trans-pyrrolidine-2,4-dicarboxylate (PDC) were obtained from either Sigma (St. Louis, MO) or Research Biochemicals International (Natick, MA).

Functional Expression in HRPE Cells. This was done using the vaccinia virus expression system as described previously (Prasad et al., 1998; Kekuda et al., 1999; Wang et al., 1999). Subconfluent HRPE cells grown in 24-well culture plates were first infected with a recombinant (VTF_7-3) vaccinia virus encoding T7 RNA polymerase and then transfected with the plasmid carrying the full-length human and rat NaDC3 cDNAs. These cDNAs were originally cloned from human placenta and rat placenta, respectively (Kekuda et al., 1999; Wang et al., 2000). After 10 to 12 h post-transfection, uptake measurements were made at room temperature with [³H]succinate. The uptake medium was 25 mM HEPES/Tris (pH 7.5), containing 140 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl₂, 0.8 mM MgSO₄, and 5 mM glucose. When the influence of Na⁺ on the inhibition of succinate transport by N-acetylaspartate was investigated, the buffers containing 140 mM NaCl or 140 mM N-methyl-D-glucamine (NMDG) chloride were mixed to give uptake buffers of desired Na⁺ composition. In most experiments, the time of incubation was 1 min, which was found in previous studies to be appropriate for measurement of initial uptake rates (Kekuda et al., 1999; Wang et al., 2000). Endogenous transport was always determined in parallel using cells transfected with pSPORT vector alone.

Functional Expression in X. laevis Oocytes. Capped cRNA from the cloned human NaDC3 cDNA was synthesized using the mMESSAGE mMACHINE kit (Ambion) according to manufacturer's protocol. The cRNA was dissolved in sterile water and its integrity checked on denaturing formaldehyde-agarose gel. Mature oocytes from X. laevis were isolated by treatment with collagenase A (1.6 mg/ml), manually defolliculated, and maintained at 18°C in modified Barth's medium supplemented with 10 mg/l gentamycin (Mackenzie et al., 1996a,b). On the following day, oocytes were injected with 50 ng of cRNA. Oocytes injected with water served as control. The oocytes were used for electrophysiological studies 6 days after cRNA injection. Expression of NaDC3 in oocytes was confirmed by comparing the uptake of radiolabeled succinate between cRNA-injected oocytes and water-injected oocytes. Electrophysiological studies were done by the conventional two-microelectrode voltage-clamp method (Mackenzie et al., 1996a,b). Oocytes were superfused with a NaCl-containing transport buffer (100 mM NaCl, 2 mM KCl, 1 mM MgCl₂, 1 mM CaCl₂, 3 mM HEPES, 3 mM 4-morpholineethanesulfonic acid, and 3 mM Tris, pH 7.5) followed by the same buffer containing different substrates. The membrane potential was clamped at 50 mV. Voltage pulses between +50 and 150 mV, in 20-mV increments, were applied for 100-ms durations and steady-state currents measured. The difference between the steady-state currents measured in the presence and absence of substrates were considered as the substrate-induced currents. Kinetic parameters for the saturable transport of N-acetylaspartate were calculated using the Michaelis-Menten equation. Data were analyzed by nonlinear regression and confirmed by linear regression.

In Situ Hybridization. Four-week-old male whole rat brains were collected and immediately frozen in liquid nitrogen. Unfixed 12-µm serial sections were prepared on a cryostat, mounted on 2% 3-aminopropyltriethoxysilane-coated slides, air dried for 10 min, and then stored at 70°C until required. Nonradioactive in situ hybridization using digoxigenin UTP-labeled riboprobes was performed on tissue sections as described previously (Wu et al., 1998, 1999). Briefly, cryostat sections were fixed in 4% paraformaldehyde in 0.1 M PBS (pH 7.3) at 4°C for 20 min, washed, and permeabilized with proteinase K (10 µg/ml in PBS containing 0.1% Tween 20) for 10 min at room temperature. Sections were then refixed in 4% paraformaldehyde in PBS, prehybridized for 1 h at 65°C in 50% formamide, and incubated with sense or antisense riboprobes for 16 h at 65°C. Hybridization was followed by washing and the hybridization signals were detected immunologically using alkaline phosphatase-conjugated anti-digoxigenin antibody. The color reaction was developed with a commercially available detection kit (Boehringer Mannheim, Indianapolis, IN). These sections were compared with adjacent sections subjected to H&E staining.

Reverse Transcription-Polymerase Chain Reaction (RT-PCR) and Restriction Analysis. Poly(A)⁺ mRNA samples from rat brain and human brain were used for RT-PCR. The rat NaDC3-specific primers used in the analysis were 5'-CTGCTGCCCATTCTCTTC-3' (upstream) and 5'-GAGGAGGATGGCAAACAA-3' (downstream). These primers correspond to the nucleotide positions 109 to 126 and 1075 to 1092 in the full-length rat NaDC3 cDNA (Kekuda et al., 1999). The human NaDC3-specific primers used in the analysis were 5'-GACCACCCTGGGGAGACA-3' (upstream) and 5'-CCTGTTCGGCAAACTTGATG-3' (downstream). These primers correspond to the nucleotide positions 632 to 649 and 1030 to 1049 in the full-length human NaDC3 cDNA (Wang et al., 2000). The expected size of the RT-PCR product is 984 bp in the rat brain and 418 bp in the human brain. The resulting RT-PCR products were gene-cleaned and used for restriction analysis. Three enzymes were used in the restriction analysis of each RT-PCR product: AvaI, BsrDI, and XmnI for rat cDNA and AvaI, XhoI, and XmnI for human cDNA.

Screening of the Rat Brain cDNA Library. A whole-brain cDNA library was constructed using poly(A)⁺ RNA isolated from rat brain. The SuperScript plasmid cDNA library construction system (Life Technologies, Inc.) was used to clone the cDNA inserts into pSPORT vector. This library has been used in our earlier studies for successful isolation of full-length functional clones of the high-affinity peptide transporter (Wang et al., 1998) and type 1 -receptor (Seth et al., 1998). The screening of the cDNA library was done as described previously (Kekuda et al., 1999; Wang et al., 2000). The cDNA probe used for screening was a ~1.9-kbp-long EcoRI/BamHI restriction fragment of rat NaDC3 cDNA (Kekuda et al., 1999). The probe included the entire open reading frame, 30 nucleotides of the 5'-noncoding region and 69 nucleotides from the 3'-noncoding region. The cDNA probe was labeled with [-³²P]dCTP by random priming using the Ready-to-go oligolabeling beads (Amersham Pharmacia Biotech, Piscataway, NJ). Following overnight hybridization at 65°C, the filters were washed under low-stringency conditions. Positive clones were purified by secondary screening. The functional identity of one of the positive clones was established by demonstrating its ability to mediate Na⁺-dependent succinate transport in a heterologous expression system in HRPE cells using the vaccinia virus expression technique.

DNA Sequencing. Both the sense and antisense strands of the cDNA were sequenced by primer walking. Sequencing was done by Taq DyeDeoxy terminator cycle sequencing using an automated Perkin-Elmer Applied Biosystems 377 Prism DNA sequencer. The sequence was analyzed using the BCM Search Launcher server at http://dot.imgen.bcm.tmc.edu:9331/and NCBI server at http://www.ncbi.nlm.nih.gov/.

Primary Culture of Astrocytes. Primary cultures composed of mixed glial cells were prepared from newborn rat (0-1 day old) cerebral cortices as described previously (Cameron and Rakic, 1994) and type 1 astrocytes obtained according to the procedure described by Levison and McCarthy (1991). Cell cultures were maintained in 24-well plates in minimum essential medium (Earle's salts) containing 15% newborn calf serum at 37°C in a humidified atmosphere of 5% CO₂. The cells were grown for 2 weeks with change in medium twice a week. After this, the cells were cultured for another 1 week in the medium containing 25 µM forskolin. During this period, the medium was changed every other day. Forskolin was used to induce morphological differentiation of the astrocytes. Uptake of [³H]succinate in these cells was measured at 37°C for 15 min as described above for the measurement of the transport activity of NaDC3 in the heterologous expression system.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Interaction of N-Acetylaspartate with Cloned Rat and Human NaDC3s. NaDC3 is a high-affinity Na⁺-coupled transporter for succinate and other dicarboxylate intermediates of Krebs cycle and NaDC3-specific mRNA transcripts are found in the kidney, liver, placenta, and brain (Kekuda et al., 1999; Wang et al., 2000). This transporter interacts with succinate with a K_t value in the range of 2 to 20 µM in different animal species (Kekuda et al., 1999; Wang et al., 2000). The other high-affinity substrates for NaDC3 include -ketoglutarate, oxaloacetate, malate, and fumarate. All these substrates contain four or five carbon atoms, two carboxylate groups, and no positively charged groups. In addition, NaDC3 accepts -monomethylsuccinate, dimethylsuccinate, and dimercaptosuccinate as substrates, indicating that substitutions in the -carbon atoms of succinate are tolerated as long as the substituting groups are not ionized. Succinate differs from aspartate only in the presence of an amino group at the -carbon atom of the latter compound (Fig. 1). Because the amino group is ionized and consequently aspartate contains one positively charged group and two negatively charged groups, the affinity of aspartate for NaDC3 is very low. At a concentration of 2 mM, aspartate inhibits rat NaDC3-mediated succinate transport by 60% and human NaDC3-mediated succinate transport by 40% (Kekuda et al., 1999; Wang et al., 2000). In contrast to aspartate, N-acetylaspartate contains a nonionized substitution at the -carbon atom and thus possesses the structural features of an -carbon-substituted succinate derivative (Fig. 1). Therefore, we thought that N-acetylaspartate may be recognized with much higher affinity than aspartate by NaDC3. To test this, we expressed rat and human NaDC3s heterologously in HRPE cells and compared the ability of N-acetylaspartate, aspartate and glutamate to inhibit NaDC3-mediated succinate transport (Fig. 2). In both cases, all three compounds inhibited succinate transport in a dose-dependent manner. In the case of rat NaDC3, the IC₅₀ values (i.e., the concentration of the inhibitor causing 50% inhibition) for N-acetylaspartate, aspartate, and glutamate were 59 ± 4 µM, 1.49 ± 0.10 mM, and 12.2 ± 0.7 mM, respectively. The corresponding values for human NaDC3 were 232 ± 20 µM, 3.07 ± 0.17 mM, and >10 mM, respectively. Thus, acetylation of the amino group in aspartate was found to increase the affinity by 25-fold in the case of rat NaDC3 and by 13-fold in the case of human NaDC3.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Chemical structures of succinate, aspartate, and N-acetylaspartate.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Inhibition of NaDC3-mediated succinate transport by N-acetylaspartate, aspartate, and glutamate. Rat (A) and human (B) NaDC3s were expressed in HRPE cells heterologously using the vaccinia virus expression technique. The function of the expressed transporters was monitored by measuring the uptake of 20 nM [3H]succinate for 1 min in a NaCl-containing medium in the presence of increasing concentrations of N-acetylaspartate (), aspartate (), and glutamate (). Uptake measured in cells transfected with vector alone was subtracted from uptake in cells transfected with cDNA to calculate cDNA-specific uptake. The cDNA-specific uptake in the absence of inhibitors was taken as 100%. Results are mean ± S.E. from four independent transfections.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Kinetics of N-acetylaspartate-induced inhibition of succinate transport mediated by human NaDC3. Human NaDC3 was expressed heterologously in HRPE cells and its function was monitored by measuring succinate uptake. Concentration of succinate was varied over the range of 5 to 100 µM. Uptake was measured for 1 min in a NaCl-containing medium in the absence () or presence of 300 µM () or 500 µM () N-acetylaspartate. Uptake measured in vector-transfected cells was subtracted to determine the cDNA-specific uptake. Inset, Eadie-Hofstee plots. Results are mean ± S.E. from two separate transfection experiments, each done in duplicate.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Influence of Na+ on the interaction of N-acetylaspartate with human NaDC3. A, human NaDC3 was expressed heterologously in HRPE cells and its function was monitored by measuring the uptake of 20 nM [3H]succinate for 1 min. Concentration of Na+ in the uptake medium was varied over the range of 20 to 120 mM with isoosmotic addition of NMDG chloride in place of NaCl. Uptake was measured in the absence () or presence () of 1 mM N-acetylaspartate. Uptake measured in vector-transfected cells was subtracted to determine cDNA-specific uptake. Results are mean ± S.E. from two different transfection experiments, each done in duplicate. B, influence of Na+ on N-acetylaspartate-induced inhibition of human NaDC3-mediated succinate uptake. The data in A were used to calculate inhibition at each concentration of Na+.

Direct Evidence for the Transport of N-Acetylaspartate by NaDC3. The studies described thus far demonstrate that the cloned rat and human NaDC3s interact with N-acetylaspartate in a Na⁺-dependent manner and that N-acetylaspartate and succinate compete for the substrate-binding site. These studies suggest, but do not prove directly, that N-acetylaspartate is transported via the NaDC3s. It is possible that a structural analog may compete with a transportable substrate for binding to the substrate-binding site without itself being transported. Therefore, we used the X. laevis oocyte expression system to investigate whether N-acetylaspartate is a transportable substrate for NaDC3. These studies were done with the cloned human NaDC3. The Na⁺-coupled transport of succinate via rat or human NaDC3 is electrogenic, resulting from a 3:1 stoichiometry of Na⁺:succinate (Kekuda et al., 1999; Wang et al., 2000). If a similar mechanism operates for the transport of N-acetylaspartate via NaDC3, such a process can be monitored in NaDC3-expressing oocytes using electrophysiological methods. The transport process is expected to be associated with membrane depolarization and this can be detected as an inward current by the two-microelectrode voltage-clamp technique. The data given in Fig. 5 show that when human NaDC3-expressing oocytes were perifused with 5 mM N-acetylaspartate in a NaCl-containing medium, an inward current of ~150 nA was induced. This inward current was however not noticeable in a Na⁺-free medium in which NaCl was substituted with NMDG chloride. However, the magnitude of the inward current in a NaCl-containing medium was the same when Cl in the medium was replaced with gluconate. These data provide direct evidence for Na⁺-coupled transport of N-acetylaspartate via the cloned human NaDC3. The transport process is obligatorily dependent on Na⁺. Cl was however not necessary for this transport. Perifusion of water-injected oocytes with N-acetylaspartate did not induce any detectable inward currents, demonstrating that the observed currents in NaDC3- expressing oocytes were dependent on the expression of NaDC3.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Influence of Na+ and Cl on N-acetylaspartate-induced inward currents in X. laevis oocytes expressing human NaDC3. Oocytes were perifused with 5 mM N-acetylaspartate (NAA) in medium containing NMDG chloride (Na+-free), NaCl, or sodium gluconate (Cl-free). Currents were monitored at 50 mV using the two-microelectrode, voltage-clamp technique.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Kinetic analysis of N-acetylaspartate-induced inward currents in X. laevis oocytes expressing human NaDC3. Oocytes expressing human NaDC3 were perifused with increasing concentrations of N-acetylaspartate (NAA) in NaCl-containing medium. The membrane potential was clamped at 50 mV and the substrate-induced currents were measured at different testing membrane potentials. A, influence of N-acetylaspartate concentration on N-acetylaspartate-induced currents at various testing membrane potentials. Concentration of N-acetylaspartate was varied over the range of 0.025 to 2 mM. B, saturability of N-acetylaspartate-induced currents at different testing membrane potentials with respect to N-acetylaspartate concentration. C, influence of testing membrane potential on Imax. D, influence of testing membrane potential on K0.5 for N-acetylaspartate.

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Influence of Na+ on N-acetylaspartate-induced currents in X. laevis oocytes expressing human NaDC3. A, influence of Na+ concentration on N-acetylaspartate (1 mM)-induced currents at different testing membrane potentials. B, influence of testing membrane potential on INamax. C, influence of testing membrane potential on KNa0.5. D, Influence of testing membrane potential on Hill coefficient nHNa. Inset, Hill plot at a testing membrane potential of 50 mV.

Transport of PDC via Human NaDC3. PDC, a glutamate uptake blocker, has been shown to inhibit the transport of N-acetylaspartate in rat glial cell cultures (Sager et al., 1999). In the present study, we tested whether PDC is a transportable substrate for the cloned human NaDC3. As shown in Fig. 8A, perifusion of human NaDC3-expressing oocytes with PDC was associated with detectable inward currents and the magnitude of the currents increased as the concentration of PDC increased. The currents were however much smaller compared with the currents induced by N-acetylaspartate. The PDC-induced currents also were dependent on membrane potential (Fig. 8B), the magnitude of the currents increasing with hyperpolarization of the membrane potential.

View larger version (15K): [in this window] [in a new window]   Fig. 8.   Induction of inward currents in human NaDC3-expressing X. laevis oocytes when perifused with PDC in NaCl-containing medium. A, influence of PDC concentration on PDC-induced currents. B, influence of testing membrane potential on PDC-induced currents. Concentration of PDC was varied over the range of 0.1 to 1 mM.

Evidence for Expression of NaDC3 in Brain. We first investigated the expression of NaDC3 in rat brain by in situ hybridization and RT-PCR and in human brain by RT-PCR. In situ hybridization with rat NaDC3-specific antisense riboprobe revealed that NaDC3 mRNA is expressed strongly within the meningeal layers of supporting tissue that surround the brain and relatively weakly throughout the cerebral cortex, hippocampus, and cerebellum (Fig. 9). Higher power images of different areas of the brain were used to define the sites of mRNA expression in a greater detail. High levels of expression were observed within the arachnoid mater that overlies the subarachnoid space and within the pia mater. There was no detectable expression in the choroid plexus nor in the large and small blood vessels found within the subarachnoid space and choroid plexus. In the cerebellum, detectable levels of expression were found throughout the extremely cellular inner granular layer as well as within the Bergmann glial cells located at the level of Purkinje cell layer, but were absent from the molecular layer that contains basket and stellate neurons. In the cortex, the expression was detectable within the astrocytes scattered among the neural cells that themselves were negative for the expression. The outermost acellular layer of the cortex was also negative for the expression. Parallel experiments done with sense probe did not yield any hybridization-positive signals.

View larger version (113K): [in this window] [in a new window]   Fig. 9.   In situ hybridization for detection of NaDC3 mRNA in adult rat brain. a, low power image (12×) of an oblique lateral section of whole normal rat brain probed with antisense digoxigenin-labeled riboprobe specific for rat NaDC3 (Ctx, cortex; CA, hippocampus; CB, cerebellum). b, adjacent brain section probed with a sense riboprobe. Higher power images of boxed areas (b-f) are shown. c, higher power image (200×) showing high-level expression of NaDC3 mRNA within the arachnoid mater (arrowheads) and within the pia mater (asterisk). Large and small blood vessels (arrow) were negative for expression. d, higher power image (200×) showing high-level expression of NaDC3 mRNA within the pia mater (arrowheads). Choroid plexus (Ch) and large and small blood vessels (arrows) within the choroid plexus are negative for expression. e, higher power image (200×) showing high-level expression of NaDC3 mRNA within the arachnoid mater (arrowhead). Low-level expression is evident in granular layer (GL), but expression is absent in molecular layer (ML) and in blood vessels (arrow). f, higher power image (400×) showing strong expression of NaDC3 mRNA within the dura mater (arrowhead) and comparatively lower level of expression within the astroglia throughout the cortex (Ctx). Expression is absent in blood vessels (arrows) and within the acellular layer (double arrow).

View larger version (36K): [in this window] [in a new window]   Fig. 10.   RT-PCR and restriction analysis for the expression of NaDC3 mRNA in brain using rat (A) and human (B) brain mRNA. Rat and human brain mRNA was used for RT-PCR with primer pairs specific for rat NaDC3 and human NaDC3, respectively. Restriction analysis of the RT-PCR products was done with various restriction enzymes. M, molecular size markers for size determination of the RT-PCR products (uncut) and restriction fragments.

Uptake Studies in Rat Primary Cultures of Cortical Astrocytes. To provide unequivocal evidence for the expression of the Na⁺-coupled high-affinity dicarboxylate transporter in glial cells, we measured succinate uptake in rat primary cultures of cortical astrocytes. Uptake of succinate in these cultures was markedly Na⁺-dependent. Replacement of Na⁺ with NMDG reduced the uptake by more than 85% (data not shown). The Na⁺-dependent uptake of [³H]succinate (50 nM) was almost completely inhibited by 250 µM -ketoglutarate, malate, fumarate, and unlabeled succinate (Fig. 11A), showing the recognition of other dicarboxylates by the transporter. We also assessed the interaction of this transporter with N-acetylaspartate (Fig. 11B). The Na⁺-dependent uptake of [³H]succinate was inhibited by N-acetylaspartate in a dose-dependent manner with an IC₅₀ value of 188 ± 11 µM. Under similar conditions, unlabeled succinate inhibited the uptake of [³H]succinate with an IC₅₀ value of 13 ± 2 µM. These data show that the Na⁺-coupled high-affinity dicarboxylate transporter is functionally expressed in rat primary cultures of cortical astrocytes and that N-acetylaspartate is recognized as a substrate by the transporter.

View larger version (16K): [in this window] [in a new window]   Fig. 11.   Expression of the Na+-coupled high-affinity dicarboxylate transporter in primary cultures of rat cerebral astrocytes. A, inhibition of Na+-dependent uptake [3H]succinate (50 nM) by 250 µM unlabeled succinate, -ketoglutarate (-KG), malate, and fumarate. B, dose-response relationship for the inhibition of Na+-dependent [3H]succinate (50 nM) uptake by N-acetylaspartate () and unlabeled succinate (). In both A and B, uptake was measured at 37°C for 15 min in the presence of NaCl. Uptake measured in the absence of NaCl (NMDG chloride was used to substitute for NaCl iso-osmotically) was subtracted from total uptake to calculate Na+-dependent component. Results are mean ± S.E. from triplicate measurements.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

N-Acetylaspartate plays an important role in myelination during normal brain development. Such a role is supported by the following observations: the levels of N-acetlyaspartate in the brain increase markedly (~30-fold) during the perinatal period of active myelination and brain development (Mussini et al., 1967; Miyake and Kakimoto, 1981; Koller and Coyle, 1984) and N-acetylaspartate is a donor of acetyl group for de novo lipid synthesis during myelination and also is an essential factor necessary for the synthesis of cerebrosides, a major constituent of myelin (Patel and Clark, 1979; Shigematsu et al., 1983; Shimeno et al., 1984). The enzyme responsible for the release of the acetyl group from N-acetylaspartate, called aspartoacylase II, is found only in glial cells. Because glial cells are responsible for myelination and a genetic defect in the enzyme leads to the demyelinating disorder Canavan disease, intracellular hydrolysis of N-acetylaspartate in glial cells is essential for the function of this compound in the myelination process. N-Acetylaspartate in the extracellular space is the source of the substrate for aspartoacylase II in glial cells (Tsai and Coyle, 1995; Baslow, 1997; Clark, 1998). The origin of N-acetlyaspartate in the extracellular fluid is the release of this compound from neurons that are predominantly responsible for the biosynthesis of N-acetylaspartate in the brain. It also can arise extracellularly from the hydrolysis of N-acetylaspartylglutamate by the plasma membrane enzyme N-acetylated--linked-amino dipeptidase whose active site is exposed to the extracellular space (Slusher et al., 1990).

The present investigation was undertaken to test the hypothesis that the Na⁺-coupled high-affinity dicarboxylate transporter NaDC3 may be the transporter that is responsible for N-acetylaspartate transport in the glial cells. The rationale for this hypothesis is as follows: NaDC3 is a high-affinity transporter for Krebs cycle intermediates, including succinate. N-Acetylaspartate and succinate are structurally related, both of them being four-carbon chain-length divalent anions. Because aspartate, a zwitterion, is a low-affinity substrate for NaDC3, it is possible that N-acetylaspartate, a divalent anion like succinate, is accepted by NaDC3 as a substrate with comparatively much higher affinity. Furthermore, NaDC3 is expressed in brain as evidenced from Northern blot analysis (Kekuda et al., 1999; Wang et al., 2000). The results of the present investigation provide unequivocal evidence in support of the hypothesis. The cloned rat and human NaDC3s transport N-acetylaspartate. This function has been established in the present studies using two different heterologous expression systems, the vaccinia virus expression system in mammalian cells and the X. laevis oocyte expression system. The interaction of N-acetylaspartate with rat and human NaDC3s was tested in the mammalian cell expression system by assessing the ability of this compound to inhibit the transport of succinate mediated by NaDC3. These studies show that N-acetylaspartate is a competitive inhibitor of NaDC3-mediated succinate transport. Direct evidence for the Na⁺-coupled transport of N-acetylaspartate via NaDC3 was obtained in the X. laevis oocyte expression system. Exposure of oocytes expressing human NaDC3 heterologously to N-acetylaspartate led to the generation of Na⁺-dependent inward currents as assessed by the two-microelectrode, voltage-clamp technique. The generation of the inward currents associated with the transport process is explained by the observed Na⁺:N-acetylaspartate stoichiometry of 3:1. Because N-acetylaspartate is a divalent anion, cotransport of three Na⁺ ions with one molecule of N-acetylaspartate would result in the net transfer of one positive charge into the oocyte. Such a process is detected as inward currents by the two-microelectrode, voltage-clamp technique. The characteristics of rat NaDC3-mediated N-acetylaspartate transport are similar to those observed by Sager et al. (1999) for N-acetylaspartate transport in rat glial cells in culture. A minor difference between our present studies and the studies by Sager et al. (1999) is related to the role of Cl in the transport process. The transport in rat glial cells was found to be Cl-dependent. However, the present studies with cloned NaDC3s did not reveal any role for Cl in the transport process. The most likely explanation for the discrepancy is the difference in the experimental conditions used in the studies. The transport measurements in cultured glial cells were made without clamping the membrane potential (Sager et al., 1999). It is possible that extracellular Cl influences the steady-state membrane potential in these cells. This Cl-dependent change in the membrane potential might be responsible for the Cl-dependent stimulation of N-acetylaspartate transport in cultured glial cells. This notion also is supported by the findings by Sager et al. (1999) that a complete replacement of Cl with gluconate did not abolish the transport entirely but rather reduced the transport only by ~50%. If the transport process is obligatorily dependent on Cl as are the members of the Na⁺ plus Cl-dependent transporter family, removal of Cl would cause a much more marked decrease in the transport. Therefore, we conclude that NaDC3-mediated N-acetylaspartate transport is not a Cl-dependent process. Our previous studies showing that Cl does not play any role in NaDC3-mediated succinate transport support this conclusion (Kekuda et al., 1999; Wang et al., 2000).

To establish the relevance of the observed transport of N-acetylaspartate via the cloned NaDC3s to the transport of N-acetylaspartate in the brain, we have obtained supporting evidence for the expression of NaDC3 in the brain. The evidence was initially derived from in situ hybridization and RT-PCR studies. In situ hybridization studies show that NaDC3 mRNA is expressed in the meningeal layers that surround the brain and the ventricles and also diffusely throughout the brain. The pattern of expression in the brain is consistent with the expression in glial cells and in the blood-brain barrier. There is no evidence for NaDC3 mRNA expression in choroid plexus and in large and small blood vessels. To provide unequivocal evidence for the expression of a functional NaDC3 in the brain, we isolated a full-length NaDC3 cDNA from a rat brain cDNA library and established its functional identity in a mammalian cell heterologous expression system. We also provide herein direct functional evidence for the expression of the Na⁺-coupled high-affinity dicarboxylate transporter in rat primary cultures of cortical astrocytes. The evidence includes the demonstration of Na⁺-dependent high-affinity succinate uptake in these cells and the selective inhibition of this uptake by other known dicarboxylate substrates of NaDC3. The uptake of succinate in these cultures is inhibited by N-acetylaspartate with an IC₅₀ value of 188 ± 11 µM.

With regard to the transport of N-acetylaspartate in the brain, the expression of NaDC3 in glial cells is very relevant because these cells have been shown previously to express a transport system for N-acetylaspartate. There is however very little N-acetylaspartate in the systemic circulation and therefore the physiological function of NaDC3 expressed in the blood-brain barrier is something other than N-acetylaspartate transport. Several Krebs cycle intermediates such as succinate, oxaloacetate, -ketoglutarate, fumarate, and malate are high-affinity substrates for NaDC3. Collectively, the concentration of these intermediates reaches significant levels in the circulation (e.g., succinate, ~40 µM; oxaloacetate, ~10 µM; -ketoglutarate, ~50 µM) (Krebs, 1950; Grundig, 1961; Kaser, 1961). These metabolic intermediates are excellent energy sources and also direct precursors for the synthesis of aspartate and glutamate. It is possible that these intermediates are transported into the brain across the blood-brain barrier via NaDC3. Because this transporter also is expressed in the placenta, liver, and kidney, the physiological function of NaDC3 in these tissues is likely to be the transport of Krebs cycle intermediates rather than the transport of N-acetylaspartate.

The affinity of NaDC3 for N-acetylaspartate is severalfold lower than for succinate. However, the extracellular concentration of succinate in the brain is very low compared with that of N-acetylaspartate. Succinate is present in the cerebrospinal fluid at a concentration of ~20 µM (Diem and Lentner, 1970), whereas the concentration of N-acetylaspartate in the brain interstitial space is 80 to 100 µM (Taylor et al., 1994; Sager et al., 1997). These concentrations are comparable to the respective affinities of NaDC3 for succinate and N-acetylaspartate. Therefore, NaDC3 is most likely to be involved in the glial uptake of N-acetylaspartate under physiological conditions. Furthermore, NaDC1 and NaDC3 are the only two Na⁺-coupled dicarboxylate transporters described thus far in mammalian tissues at the molecular level (Pajor, 1999). Of these two transporters, only NaDC3 is expressed in the brain, supporting a physiological role for NaDC3 in this tissue.

In summary, the studies presented herein demonstrate convincingly that the Na⁺-coupled high-affinity dicarboxylate transporter NaDC3 is capable of mediating the transport of N-acetylaspartate and that functional NaDC3 is indeed expressed in the brain. These findings strongly suggest that the transport of N-acetylaspartate that is known to occur in glial cells is mediated by NaDC3. Because the transport of N-acetylaspartate into glial cells is a prerequisite for the intracellular hydrolysis of this compound, a process necessary for the physiological role of N-acetylaspartate in myelination, we speculate that impairment of NaDC3 function is likely to interfere with the myelination process. The gene for NaDC3 is located on human chromosome 20q12-13.1 (Wang et al., 2000) and the structure of this gene has already been elucidated in its entirety. The present findings that NaDC3 is responsible for N-acetylaspartate transport in the brain raise the possibility that genetic defects in the NaDC3 gene may lead to inheritable forms of disorders associated with demyelination.