D -Aspartate Is Stored in Secretory Granules and Released through a Ca 2 1 -dependent Pathway in a Subset of Rat Pheochromocytoma PC12 Cells*

D -Aspartate in mammalian neuronal and neuroendo- crine cells is suggested to play a regulatory role(s) in the neuroendocrine function. Although D -aspartate is known to be released from neuroendocrine cells, the mechanism underlying the release is less understood. Rat pheochromocytoma PC12 cells contain an appreciable amount of D -aspartate (257 6 31 pmol/10 7 cells). Indirect immunofluorescence microscopy with specific antibodies against D -aspartate indicated that the amino acid is present within a particulate structure, which is co-localized with dopamine and chromogranin A, markers for secretory granules, but not with synaptophysin, a marker for synaptic-like microvesicles. After sucrose density gradient centrifugation of the postnuclear particulate fraction, about 80% of the D -aspartate was recov- ered in the secretory granule fraction. Upon the addition of KCl, an appreciable amount of D -aspartate (about 40 pmol/10 7 cells at 10 min) was released from cultured cells on incubation in the presence of Ca 2 1 in the medium. The addition of A23187 also triggered D -aspartate , aliquots (40 m l) were subjected to SDS-polyacrylamide gel electrophoresis followed by Western blotting with antibodies against chromogranin for secretory granules and synaptophysin for synaptic-like microvesicles. B , fractions 1–4 , which contained secretory granules, and fractions 5–9 , which contained synaptic-like microvesicles, were combined, respec- tively, and then their D -aspartate and dopamine content were determined as described under “Experimental Procedures.” Three independ- ent experiments gave essentially similar results. P1, P2, and S2 were prepared as described under mental Procedures.” The results were expressed as percentage of D aspartate and dopamine found in whole cells, which corresponded to 250 6 31 pmol of D -aspartate/10 7 cells and 32.75 6 6.10 nmol of dopamine/10 7 cells. In a separate experiment, PC12 cells were incu- bated with 10 m M 2 9 ,7 9 -bis-(2-carboxyethyl)-5,6-carboxylfluorescein ace-toxymethylester (BCECF-AM) for 30 min and washed with PBS. Then the cells trapped with BCECF (a cytosolic marker) were fractionated as described above. Major BCECF fluorescence (93%) was recovered in the S2 fraction, whereas only 5% was found in the P2 fraction, indicating that contamination of cytoplasm into the P2 fraction is scarce.

Although the L-enantiomers of amino acids are predominant in living organisms, substantial levels of free D-amino acids have recently been detected in mammals. Among these Damino acids, D-serine and D-aspartate are of special interest because neuronal and endocrine cells contain high levels of these amino acids (1-3; for review, see Ref. 4). D-Serine was found to potentiate the N-methyl-D-aspartate receptor through binding to the glycine site on the receptor, suggesting that D-serine might be a modulator of the N-methyl-D-aspartate receptor (5). Consistently, indirect immunofluorescence microscopy with specific antibodies against D-serine demonstrated the localization of D-serine in astrocytes (6). Specific antibodies against D-aspartate revealed the localization of D-aspartate in a subset of stellate and basket cells of the cerebellum, adrenal chromaffin cells, pituitaries, and pinealocytes (7). In the pineal gland, it has been shown that D-aspartate is present in pinealocytes, endocrine cells for melatonin (8,9). Upon incubation of pinealocytes with exogenous D-aspartate, melatonin synthesis is strongly inhibited through the inhibition of N-acetyltransferase activity (9,10). Thus, D-aspartate seems to be a modulator of melatonin synthesis. Furthermore, exogenous Daspartate stimulates the release of luteinizing hormone and growth hormone in the anterior pituitary (11).
As chemical transmitters, D-serine and D-aspartate should be secreted from neuroendocrine cells. However, the mechanism by which these amino acids are secreted from neuroendocrine cells is less understood. D-Serine is released from astrocytes upon stimulation by glutamate (5). Because D-serine is present in the cytoplasm, reversed D-serine transport through a Na ϩdependent serine transporter at the plasma membrane was proposed (5). Similarly, D-aspartate is present in the cytoplasm of pinealocytes and is released from the cells (8,9). Pinealocytes express the Na ϩ -dependent glutamate transporter, which recognizes D-aspartate as a substrate, and its inhibition by various antagonists decreases release of D-aspartate (9,10), suggesting that the Na ϩ -dependent glutamate transporter is involved in the release of D-aspartate in pinealocytes.
Here we present another type of mechanism of secretion of D-aspartate in neuroendocrine cells. A subset of rat pheochromocytoma PC12 cells contains an appreciable amount of Daspartate (12). We have extensively investigated the localization and release of D-aspartate in PC12 cells and found that PC12 cells store D-aspartate in secretory granules and secrete it through a Ca 2ϩ -dependent exocytotic mechanism. glucose, 0.1 mg/liter streptomycin, 100 units/ml penicillin G, and 0.25 mg/liter Fungizone and incubated at 37°C under 5% CO 2 (13). When necessary, 20 ng/ml nerve growth factor was included in the medium. The dispersed cells were washed three times with the above medium, placed in a 100-mm culture dish coated with poly(L-lysine) to give 1 ϫ 10 7 cells/dish, and cultured in the above medium at 37°C under 5% CO 2 . These cells were maintained for 5 days, washed with culture medium, cultured further for 1 h, and then used for experiments.
Subcellular Fractionation-Subcellular fractionation of PC12 cells was performed according to a published procedure (14) with slight modifications. In brief, cultured PC12 cells (1.0 ϫ 10 8 cells) were extensively washed with PBS 1 containing 5 g/ml leupeptin and 5 g/ml pepstatin A and homogenized by passage through a 25-gauge needle. The homogenate was then centrifuged at 800 ϫ g for 10 min, and the pellets (P1) and the supernatant (S1) were obtained. Then the S1 was centrifuged at 100,000 ϫ g for 1 h, and the supernatant (S2) and the pellet (P2) were obtained. The P2 was suspended in 20 mM MOPS-Tris (pH. 7.0) containing 0.3 M sucrose, 5 mM EDTA, 5 g/ml leupeptin, and 5 g/ml pepstatin A (SME buffer), applied to a continuous sucrose gradient (0.6 -1.6 M), and then centrifuged at 78,000 ϫ g for 3.5 h. Then samples were collected in nine tubes.
Assay After cells had been incubated in 5 ml of the above medium at 37°C, the release of D-aspartate was stimulated by the addition of 50 mM KCl (15) as specified. When necessary, various compounds were added to the incubation medium. Aliquots (10 l) were taken at time intervals, and the amount of D-aspartate was determined by HPLC according to published procedures (16,17). Simultaneously, their dopamine contents were measured by HPLC combined with amperometric detection as described previously (18).
Treatment with BoNT/E-The intoxication of PC12 cells with BoNT/E was performed using a procedure similar to that previously described (15). The cultured cells (1 ϫ 10 7 cells/dish) were incubated at 37°C for 24 h in a low ionic strength buffer consisting of 5 mM NaCl, 4.8 mM KCl, 2.2 mM CaCl 2 , 1.2 mM MgSO 4 , 20 mM HEPES-NaOH, 10 mM glucose, 220 mM sucrose, and 0.5% bovine serum albumin (pH 7.4) in the presence or absence of 10 or 50 nM BoNT/E. Then the cells were washed with fresh culture medium and incubated for an additional 12 h at 37°C. Finally, the KCl-evoked release of D-aspartate and dopamine was measured in ϩCa 2ϩ -Ringer's solution as described above.
Measurement of Intracellular [Ca 2ϩ ]-For the determination of intracellular [Ca 2ϩ ], an Argus 20/CA ratio imaging system (Hamamatsu Photonics Co., Hamamatsu, Japan) was used. Cells were cultured for 3 days on a thin glass coverslip precoated with (poly)L-lysine (0.12 mm thick and 40 mm in diameter; 2.5 ϫ 10 4 cells/coverslip). After exchange with fresh culture medium, the cells were treated with 10 M Fura-2/AM (Dojindo Co., Kumamoto, Japan) plus 0.1% Pluronic F-127 (Molecular Probes) for 2 h at 37°C and then washed twice with the same medium. The cells were perfused with the warmed ϩCa 2ϩ -Ringer's solution or ϪCa 2ϩ -Ringer's solution. Images were continuously taken at 37°C with a silicon-intensified camera (C2741-08; Hamamatsu Photonics Co.). The velocity of data acquisition for F 334 by F 380 images was 4 s at a resolution of 256 ϫ 256 pixels/image. A personal computer with appropriate software (U4469; Hamamatsu Photonics Co.) was used to control the optical equipment, recording, and data analysis. The software enabled subtraction of background fluorescence, pixel-to-pixel division of F 334 by F 380 images, fitting of the F 334 /F 380 ratios to a [Ca 2ϩ ] calibration curve prepared separately, and digital averaging of the Ca 2ϩ concentration in multiple cells (15).
Antibodies Specific to D-Aspartate-Antibodies against D-aspartate were prepared according to Schell et al. (7) and Lee et al. (19). The antisera were further purified by affinity chromatography with CNBractivated Sepharose 4B conjugated with D-aspartate. The purified antibodies against D-aspartate were divided into small portions and frozen at Ϫ85°C until use. These antibodies are active for at least 1 year after preparation. The specificity of the purified antibodies was examined by enzyme-linked immunosorbent assay as follows. Various amino acids and peptides (10 mol each) were coupled to a SUMILON MS-3696F microplate with glutaraldehyde. After blocking with PBS containing 10% goat serum, the plates were incubated with the purified antibodies (10 to 1250 dilutions) for 2 h at room temperature and then washed four times with PBS containing 0.1% Tween 20. Then the immunoreactivities were visualized with peroxidase secondary antibodies and quantified using a microplate reader (Bio-Rad, Model 550). The antibodies gave immunoreactivities of less than 1.0% (taking those of D-aspartate to be 100%) to the following compounds: L-aspartate, L-glutamate, Dglutamate, L-asparagine, D-asparagine, L-serine, D-serine, glycine, ␤-alanine, taurine, carnosine, Asp-Ala, Asp-Asp, Asp-Glu, serine-O-sulfate, Glu-Asp, and Gly-Asp. Furthermore, the purified antibodies stained some chromaffin cells in the section of rat adrenal glands and almost all pinealocytes in the primary culture, which is consistent with previous reports (7,9,19).
Immunoblotting-A membrane fraction of PC12 cells was denatured with SDS sample buffer containing 1% SDS and 10% ␤-mercaptoethanol and then electrophoresed on a 12% polyacrylamide gel in the presence of SDS (20). Following electrotransfer at 0.3 A for 2 h, the nitrocellulose filters were blocked in a buffer consisting of 20 mM Tris-HCl (pH 7.6), 5 mM EDTA, 0.1 M NaCl, and 0.5% bovine serum albumin for 4 h and then probed with 50 g of antibodies in the above buffer. The filters were washed with 20 mM Tris-HCl buffer (pH 7.6) containing 5 mM EDTA, 0.1 M NaCl, and 0.05% Tween 20, decorated with either peroxidase-labeled anti-rabbit IgG or anti-mouse IgG at a dilution of 1:2000 for 30 min, washed further with the same buffer, and then subjected to ECL amplification according to the manufacturer's manual (Amersham Pharmacia Biotech).
Immunohistochemistry-The published procedure was used (21). In brief, PC12 cells on (poly)L-lysine-coated glass coverslips were washed three times with PBS and incubated in 80 mM PIPES-KOH buffer (pH 6.8) containing 1 g/ml digitonin, 5 mM EGTA, and 1 mM MgCl 2 for 5 min at room temperature. Then the cells were fixed with 4% paraformaldehyde in PBS for 15 min. After washing with PBS again, the cells were further permeabilized with 50 g/ml digitonin in PBS for 5 min. Then ammonium chloride at 50 mM was added to the solution followed by incubation for 10 min after which the cells were washed with 0.1% gelatin in PBS for 5 min. Finally, the fixed cells were incubated for 1 h with antibodies at 10 g/ml diluted in PBS containing 0.1% gelatin. The samples were washed three times with PBS containing 0.1% gelatin and then incubated with the secondary antibodies conjugated with either fluorescein or Texas Red. After washing the cells with PBS containing 0.1% gelatin, immunoreactivity was observed under an Olympus confocal laser microscope (FLUOVIEW). D-Aspartate Uptake by Digitonin-permeabilized Cells-PC12 cells were rinsed with 1 ml of buffer composed of 20 mM MOPS-Tris (pH 7.0), 0.3 M sucrose, 2 mM magnesium acetate, and 4 mM KCl. The cells were then permeabilized for 10 min at 37°C in 0.5 ml of the buffer containing 10 M digitonin (22,23). The medium was then replaced with fresh buffer containing Tris-ATP at 2 mM in the absence of digitonin. After incubation for 10 min, uptake of D-aspartate was immediately started by the addition of radioactive D-aspartate (2.5 Ci, 0.1 mM) at 37°C. Uptake was terminated by washing the cells twice with 1 ml of ice-cold 20 mM MOPS-Tris (pH 7.0) containing 0.3 M sucrose. Then the cells were lysed with 1 ml of 1% SDS, and the radioactivity was counted with a liquid scintillation counter.

Subcellular Localization of D-Aspartate-PC12 cells contain
an appreciable amount of D-aspartate (257 Ϯ 31 pmol/10 7 cells). To reveal the subcellular localization of D-aspartate in PC12 cells, indirect immunofluorescence microscopy was performed with specific antibodies against D-aspartate. Fig. 1A shows that the antibodies stained vesicular structures. The D-aspartate immunoreactivity was co-localized with dopamine ( Fig. 1, A-C) and chromogranin A (Fig. 1, D-F), markers for secretory gran-FIG. 2. D-Aspartate is recovered in the secretory granule fraction. The P2 particulate fraction of PC12 cells was suspended in SME buffer and then applied to a continuous sucrose density gradient. After centrifugation, nine fractions were collected from the bottom. A, aliquots (40 l) were subjected to SDS-polyacrylamide gel electrophoresis followed by Western blotting with antibodies against chromogranin for secretory granules and synaptophysin for synaptic-like microvesicles. B, fractions 1-4, which contained secretory granules, and fractions 5-9, which contained synaptic-like microvesicles, were combined, respectively, and then their D-aspartate and dopamine content were determined as described under "Experimental Procedures." Three independent experiments gave essentially similar results.  D-aspartate and dopamine in whole cells and the subcellular fractions of PC12 cells P1, P2, and S2 fractions were prepared as described under "Experimental Procedures." The results were expressed as percentage of Daspartate and dopamine found in whole cells, which corresponded to 250 Ϯ 31 pmol of D-aspartate/10 7 cells and 32.75 Ϯ 6.10 nmol of dopamine/10 7 cells. In a separate experiment, PC12 cells were incubated with 10 M 2Ј,7Ј-bis-(2-carboxyethyl)-5,6-carboxylfluorescein acetoxymethylester (BCECF-AM) for 30 min and washed with PBS. Then the cells trapped with BCECF (a cytosolic marker) were fractionated as described above. Major BCECF fluorescence (93%) was recovered in the S2 fraction, whereas only 5% was found in the P2 fraction, indicating that contamination of cytoplasm into the P2 fraction is scarce. ules, but was not co-localized with synaptophysin, a marker for synaptic-like microvesicles (Fig. 1, G-I). The pre-adsorbed antibodies with D-aspartate-glutaraldehyde conjugate as well as preimmune serum did not stain the cells (Fig. 1, J and K).
These results suggest the localization of D-aspartate in secretory granules. D-Aspartate immunoreactivity is not present in all PC12 cells: about 10% of cells exhibited D-aspartate immunoreactivity among the more than 10 4 cells examined, although more than 95% of the cells exhibited synaptophysin and chromogranin immunoreactivities. The percentage of D-aspartatepositive cells did not change during culture in the presence or absence of nerve growth factor (data not shown). Subcellular fractionation was conducted to confirm the localization of D-aspartate in secretory granules. After homogenization of PC12 cells (1 ϫ 10 8 cells), 74 Ϯ 5% of D-aspartate and 62 Ϯ 4% of dopamine were recovered in the P2 fraction that contains secretory granules (Table I). Then secretory granules were separated from the P2 fraction by sucrose density gradient centrifugation. As shown in Fig. 2, secretory granules were recovered in fractions 1-7 (peak fraction, 2), whereas synapticlike microvesicles were recovered in fractions 2-9 (peak fractions, 6 -8) as revealed by the distribution of marker proteins. Consistent with the distribution of marker proteins, about 80% of each of D-aspartate and dopamine in the P2 fraction was recovered in the secretory granule fraction (Fig. 2B). Together with the immunohistochemical localization of D-aspartate shown above, it is concluded that D-aspartate is present in secretory granules.
Ca 2ϩ -dependent D-Aspartate Release-The localization of Daspartate in secretory granules suggests secretion of D-aspartate through regulated exocytosis. In fact, when PC12 cells were cultured under the standard culture conditions, an appreciable amount of D-aspartate appeared in the medium, suggesting the presence of a D-aspartate-releasing system(s) (12). To obtain evidence of D-aspartate exocytosis, we measured the D-aspartate concentration in the medium upon stimulation of the PC12 cells with KCl, which is known to trigger exocytosis of secretory granules following depolarization of plasma membrane.
An appreciable amount of D-aspartate (40.0 Ϯ 7.1 pmol/10 7 cells, which corresponds to 16% of the total D-aspartate; four determinations) was released by 10 min upon the addition of KCl in the presence of Ca 2ϩ (Fig. 3A). Essentially no L-aspartate (Ͻ Ͻ0.1 pmol/10 7 cells) was released under these conditions. The amount of released D-aspartate depended on time (Fig. 3A) and the concentration of KCl (Fig. 3B): the D-aspartate concentration was saturated at 10 min and increased at least by 50 mM KCl. In the absence of CaCl 2 , KCl-evoked D-aspartate release was significantly reduced (Fig. 3, A and B). A23187, a Ca 2ϩ ionophore, caused rapid release of D-aspartate by the cells only in the presence of Ca 2ϩ (Fig. 3C). Furthermore, upon treatment with EGTA-AM to remove intracellular free Ca 2ϩ , the cells lost the ability of KCl-evoked release of D-aspartate (Fig. 3D). Likewise dopamine was released from PC12 cells with similar kinetics and Ca 2ϩ dependence (Fig. 3, E-H). These results indicate that D-aspartate is released from PC12 cells depending on Ca 2ϩ and suggested that the entry of extracellular Ca 2ϩ following depolarization is necessary for the D-aspartate release as in the case of dopamine. tion or when cells were treated with EGTA-AM. Furthermore, the cadmium ion, a nonspecific inhibitor of voltage-gated Ca 2ϩ channels (26), at 10 M inhibited 60% of the KCl-and Ca 2ϩdependent D-aspartate release and blocked 75% of the KClevoked increase in intracellular [Ca 2ϩ ]. These results suggested that voltage-gated Ca 2ϩ channels, at least in part, are involved in the KCl-and Ca 2ϩ -dependent D-aspartate release by PC12 cells.
Evidence of Exocytosis of D-Aspartate-We further characterized the K ϩ -and Ca 2ϩ -dependent D-aspartate release by PC12 cells by examining the effect of temperature. The KCl-and Ca 2ϩ -dependent D-aspartate release was affected by the temperature: it was not observed at 4°C but appeared gradually with increasing temperature and reached the maximum at 37°C. Furthermore, once D-aspartate had been secreted the successive stimulation by KCl within 1 h was not effective. The KCl-evoked release of D-aspartate was gradually restored upon incubation and had recovered completely by incubation for 12 h, suggesting that charged and discharged processes are involved in the KCl-and Ca 2ϩ -dependent D-aspartate release. These properties are similar to those of the exocytosis of dopamine, supporting that D-aspartate is secreted through exocytosis.
The sensitivity to BoNT/E constitutes evidence of Ca 2ϩ -dependent regulated exocytosis because this neurotoxin splits SNAP25 and inhibits the late postdocking steps, resulting in inhibition of the exocytosis of secretory granules in PC12 cells (27)(28)(29). As shown in Fig. 4A, BoNT/E cleaved SNAP25, yielding a low molecular weight fragment (Fig. 4A, asterisk) and inhibited the KCl-and Ca 2ϩ -dependent release of D-aspartate (Fig. 4B) and dopamine (Fig. 4C). The inhibitory potency of BoNT/E was essentially the same as that in the exocytosis of dopamine. Under the same assay conditions, the addition of K ϩ increased intracellular [Ca 2ϩ ] in BoNT/E-treated cells to an extent similar to that in control cells (data not shown). These results indicated that SNAP25 is involved in the KCl-and Ca 2ϩ -dependent D-aspartate release.
␣-Latrotoxin, a component of black widow spider venom, triggers Ca 2ϩ -dependent exocytosis from neurons and neuroendocrine cells (30). The neurotoxin, therefore, stimulates dopamine exocytosis in PC12 cells (31). As shown in Fig. 5A, ␣-latrotoxin at 1 nM increased intracellular [Ca 2ϩ ] to 405 Ϯ 13 nM (n ϭ 54) in the presence of Ca 2ϩ . In the absence of Ca 2ϩ increase of intracellular [Ca 2ϩ ] was not observed. Under these conditions, ␣-latrotoxin caused the release of D-aspartate and dopamine to a similar extent in the presence of Ca 2ϩ (Fig. 5, B and C). In the absence of Ca 2ϩ , however, neither D-aspartate nor dopamine was released on the addition of ␣-latrotoxin (Fig. 5, B and C). These results indicated that the ␣-latrotoxin-evoked stimulation of the release of D-aspartate is Ca 2ϩ -dependent as in the case of dopamine.
It is known that bafilomycin A1, a specific inhibitor of vacuolar H ϩ -ATPase (32), effectively inhibits the exocytosis of neurotransmitters (33,34) because the compound dissipates the electrochemical proton gradient necessary for neurotransmitter uptake into vesicles. As expected, bafilomycin A1 at 1 M dissipated the transmembrane pH gradient of acidic organelles in the cells as revealed on acridine orange vital staining (Fig. 6,  A and B). Under similar assay conditions, bafilomycin A1 inhibited 40% of the KCl-and Ca 2ϩ -dependent D-aspartate release (Fig. 6C) and 55% of the dopamine release (Fig. 6D). The results suggest that an electrochemical proton gradient is necessary at least in part for the KCl-and Ca 2ϩ -dependent Daspartate release by PC12 cells as in the case of dopamine.
Uptake of D-Aspartate-Sensitivities to bafilomycin A1 suggest the presence of systems for the accumulation of D-aspartate in secretory granules. In the final part of the study, we investigated whether or not such a transport system is present in the cells. As shown in Fig. 7, radiolabeled D-aspartate was taken up by the digitonin-permeabilized cells depending on ATP. The omission of Mg 2ϩ reduced the ATP-dependent D-aspartate uptake to the control level. Consistent with the Ca 2ϩ -dependent release of D-aspartate, bafilomycin A1 at 1 M inhibited the ATP-dependent D-aspartate uptake. SF6847 (3,5-di-tert-butyl-4-hydroxybenzylidene-malononitrile), a proton conductor that dissipates an electrochemical proton gradient, also inhibited the ATP-dependent D-aspartate uptake. These results suggest that the D-aspartate transporter energetically coupled with vacuolar H ϩ -ATPase is responsible for the storage and release of D-aspartate in PC12 cells.

DISCUSSION
PC12 cells develop two different kinds of secretory machinery: one is secretory granules, which are responsible for the secretion of monoamines such as dopamine, and the other is synaptic-like microvesicles, counterparts of neuronal synaptic vesicles that contain acetylcholine (14). Here we showed that D-aspartate is stored in secretory granules but not in synapticlike microvesicles. Furthermore, D-aspartate is secreted FIG. 4. BoNT/E cleaved SNAP25 and inhibited the KCl-and Ca 2؉ -dependent release of D-aspartate and dopamine. A, cells were treated with BoNT/E at the indicated concentrations, washed with PBS containing 0.01% DNase, 5 mM MgCl 2 , and 10 g/ml each of leupeptin and pepstatin A, and then dissolved in the sample buffer containing 10% SDS and ␤-mercaptoethanol. After the dissociation of proteins, samples were applied to a 12.5% polyacrylamide gel containing SDS, electrophoresed, and then analyzed by immunoblotting with antibodies against SNAP25. The resultant digestive fragment of SNAP25 is indicated by *. B and C, after BoNT/E treatment as above, the KCl-evoked release of D-aspartate (B) and dopamine (C) at 10 min was measured in ϩCa 2ϩ -Ringer's solution as described under "Experimental Procedures." The results are presented as means Ϯ S.E. (four independent experiments). 100% corresponds to 41.3 Ϯ 5.1 pmol/10 7 cells and 3.8 Ϯ 0.3 nmoles/10 7 cells, respectively. through a Ca 2ϩ -dependent exocytotic mechanism in a parallel fashion to dopamine secretion.
Indirect immunofluorescence microscopy and subcellular fractionation studies revealed the localization of D-aspartate in secretory granules. The storage of D-aspartate seems to be stereospecific because essentially no accumulation of the Lform was observed in secretory granules on indirect immunofluorescence microscopy, 2 and the release of the L-form from cells was under the detection limit as revealed on HPLC analysis as described under "Experimental Procedures." Because the accumulation of transmitters in secretory granules is generally mediated through an active transport mechanism by secondary transporters energized with vacuolar H ϩ -ATPase, our results suggest the presence of a D-aspartate transporter(s) in secretory granules in PC12 cells (Fig. 7). The involvement of the brain-specific Na ϩ -dependent inorganic phosphate transporter (BNPI), recently identified as a vesicular glutamate transporter in neuronal cells (23,35), in the accumulation of D-aspartate is unlikely because a vesicular glutamate transporter does not recognize D-aspartate as a substrate (36,37) and is not expressed in PC12 cells (35). The absence of BNPI in PC12 cells was confirmed by immunoblotting using the sitespecific polyclonal antibodies against BNPI (data not shown). Thus, it is possible that a novel vesicular D-aspartate transporter functions in the secretory granules in a subset of PC12 cells. Identification and characterization of the putative D-aspartate transporter is now in progress in our laboratory.
The presence of D-aspartate in secretory granules raises an important issue of the origin of D-aspartate in PC12 cells. The D-aspartate content in the culture medium was below the detection limit (Ͻ Ͻ1 fmol). The D-aspartate content in the cells did not change with the duration of culture or during 5-10 passages (12). Furthermore, PC12 cells do not express a plasma membrane-type Na ϩ -dependent glutamate transporter that recognizes D-aspartate as a substrate. Thus, it is likely that D-aspartate is not of exogenous origin but is de novo synthesized by D-aspartate synthase in the cells. In this respect, it is noteworthy that D-aspartate racemase-like activity was detected in rat pinealocytes (9) and adrenal glands (38).
Several criteria are known for regulated exocytosis. We found that the properties of the release of D-aspartate in PC12 cells satisfied the criteria. At first, PC12 cells secrete the amino acid depending on Ca 2ϩ . The sensitivity to Cd 2ϩ suggests the involvement of voltage-gated Ca 2ϩ channels in the process. Our preliminary experiments indicated that antagonists for voltage-gated L-type Ca 2ϩ channels such as nifedipine inhibited the Ca 2ϩ -dependent D-aspartate release and blocked the KClevoked increase in intracellular [Ca 2ϩ ], but antagonists for N-type or P/O-type channels did not have such effects. These results suggest that L-type Ca 2ϩ channels are responsible for the entry of Ca 2ϩ from the extracellular space into PC12 cells. Consistently, the presence of L-type Ca 2ϩ channels in PC12 cells has been demonstrated (39 -41).
The second line of evidence is the temperature sensitivity and the requirement of an appropriate duration of the response for a second stimulation. These properties may reflect complex membrane dynamics including the charging and discharging of neurotransmitters. Similar phenomena were observed for Ca 2ϩdependent exocytosis found in various endocrine cells and neuronal cells (15,(42)(43)(44). The relatively slow rate of D-aspartate release (minute order as shown in Fig. 3A) is also similar to that of the release of dopamine by PC12 cells and glutamate by rat pinealocytes (15). The third line of evidence is the sensitivity to ␣-latrotoxin, BoNT/E, and bafilomycin A1. The sensitivity to these compounds strongly suggests the involvement of the ␣-latrotoxin receptor, SNAP receptor complex, and vacuolar H ϩ -ATPase, which are important components of Ca 2ϩ -dependent regulated exocytosis, in the secretion of D-aspartate.  7. ATP-dependent uptake of radiolabeled D-aspartate by the digitoninpermeabilized cells. Uptake of radiolabeled D-aspartate by the permeabilized cells was carried out as described under "Experimental Procedures" in the presence or absence of the listed compounds. In some experiments, magnesium acetate (ϪMg 2ϩ ) or ATP (ϪATP) was omitted. The results are the means Ϯ S.E. of four independent experiments.