Regulation of Amyloid Precursor Protein Secretion by Glutamate Receptors in Human Ntera 2 Neurons (NT2N)*

The amyloid precursor protein (APP) can be cleaved by a β-secretase to generate a β-amyloid peptide, which has been implicated in the pathogenesis of Alzheimer’s disease. However, APP can also be cleaved by an α-secretase to form a non-amyloidogenic secreted form of APP (APP-S). APP-S secretion can be physiologically regulated. This study examined the glutamatergic regulation of APP in the human neuronal Ntera 2 (NT2N) cell line. Metabotropic glutamate receptor subtypes 1α/β and 5α were identified in the NT2N neurons by reverse transcription-polymerase chain reaction. Stimulation of these phosphatidylinositol-linked receptors with glutamate or specific receptor agonists resulted in a dose- and time-dependent increase in the secretion of the amyloid precursor protein (APP-S), measured by the immunoprecipitation of APP-S from the medium of [35S]methionine-labeled NT2N neurons. The glutamate-induced APP-S secretion was maximal at 30 min and at a concentration of 1 mm glutamate. Glutamate-induced APP-S secretion required activation of phospholipase C, which resulted in inositol 1,4,5-trisphosphate production, as shown by the rapid glutamate-induced accumulation of inositol 1,4,5-trisphosphate. Glutamate also caused an increase in intracellular Ca2+. The protein kinase C activator phorbol 12-myristate 13-acetate, a phorbol ester, as well as 1-oleoyl-2-acetoyl-3-glycerol, a cell-permeable diacylglycerol analog, also stimulated APP-S secretion. These findings suggest that APP-S secretion from NT2N neurons can be regulated by the activation of phosphatidylinositol-linked metabotropic glutamate receptor signaling pathway.

The amyloid precursor protein (APP) can be cleaved by a ␤-secretase to generate a ␤-amyloid peptide, which has been implicated in the pathogenesis of Alzheimer's disease. However, APP can also be cleaved by an ␣-secretase to form a non-amyloidogenic secreted form of APP (APP-S). APP-S secretion can be physiologically regulated. This study examined the glutamatergic regulation of APP in the human neuronal Ntera 2 (NT2N) cell line.
Metabotropic glutamate receptor subtypes 1␣/␤ and 5␣ were identified in the NT2N neurons by reverse transcription-polymerase chain reaction. Stimulation of these phosphatidylinositol-linked receptors with glutamate or specific receptor agonists resulted in a doseand time-dependent increase in the secretion of the amyloid precursor protein (APP-S), measured by the immunoprecipitation of APP-S from the medium of [ 35 S]methionine-labeled NT2N neurons. The glutamateinduced APP-S secretion was maximal at 30 min and at a concentration of 1 mM glutamate. Glutamate-induced APP-S secretion required activation of phospholipase C, which resulted in inositol 1,4,5-trisphosphate production, as shown by the rapid glutamate-induced accumulation of inositol 1,4,5-trisphosphate. Glutamate also caused an increase in intracellular Ca 2؉ . The protein kinase C activator phorbol 12-myristate 13-acetate, a phorbol ester, as well as 1-oleoyl-2-acetoyl-3-glycerol, a cell-permeable diacylglycerol analog, also stimulated APP-S secretion. These findings suggest that APP-S secretion from NT2N neurons can be regulated by the activation of phosphatidylinositol-linked metabotropic glutamate receptor signaling pathway.
Alzheimer's disease is characterized by the deposition of ␤-amyloid (A␤) 1 into senile plaques, the formation of neurofi-brillary tangles, and neuronal death. Senile plaques consist of a A␤ protein surrounded by dystrophic neuritic processes, astrocytes, and microglia. The ␤-amyloid protein is a 39 -43amino acid cleavage product of ␤-amyloid precursor proteins (APP) (1)(2)(3)(4). APP is a family of transmembrane glycoproteins with a large extracytoplasmic domain, a membrane-spanning domain containing the A␤ peptide, and a short intracytoplasmic domain (4). APP exists as three alternatively spliced isoforms, ranging from 695 to 770 amino acids in length and is expressed in mammalian neuronal and nonneuronal cells and tissues (5). APP is most abundant in the brain and APP 695 is the major APP in the human neuronal cell line NTera 2/cl.D1 (NT2) (6). APP is thought to be processed through various pathways. APP processed by a constitutive secretory pathway (or ␣-secretase) results in a large secreted form (APP-S ␣ ) excluding A␤ formation (7,8). Processing by a ␤-secretase or by an endosomal and lysosomal pathway produces the 4-kDa A␤ protein (9,10). Mutations in the APP gene have been shown to result in the abnormal processing of APP, which may lead to subsequent A␤ deposition and accumulation (11,12). Recent evidence for this was provided by a human familial Alzheimer's disease mutant APP transgenic mouse that expresses high levels of APP, as well as A␤, which increases in an age-dependent manner. This results in the formation of amyloid plaques, dendritic and synaptic loss, and astrocytosis, as seen in Alzheimer's disease (13,14). Mutations in the presenilin 1 and 2 genes (responsible for many cases of familial Alzheimer's disease) alter APP processing by increasing A␤ generation (15)(16)(17)(18). This supports the idea of a primary role for APP and A␤ in the pathogenesis of Alzheimer's disease.
Many populations of neurons are affected in Alzheimer's disease, in particular the cholinergic and glutamatergic neurons of the hippocampus and cerebral cortex, therefore it is important to examine these receptor signaling pathways. The m1 and m3 muscarinic receptors have been shown previously to stimulate the secretion of APP, and recently the activation of metabotropic glutamate receptors was shown to cause an increase in APP secretion (19,20). Metabotropic glutamate receptors are seven transmembrane-spanning proteins that consist of eight receptor subtypes. The mGluR1 and R5 subtypes are linked to phosphatidylinositol (PI) hydrolysis, whereas the other six subtypes are negatively coupled to adenylate cyclase. The degeneration and loss of neurons seen in Alzheimer's disease and a recent study demonstrating the involvement of metabotropic glutamate receptors in the regulation of APP secretion suggest that a defect in the receptors or in their signaling pathways might lead to abnormal processing of APP and subsequent A␤ deposition. The regulatory mechanism of APP processing and secretion is unknown; however, recent findings suggest that the control is through neurotransmitter receptor coupling to signal transduction pathways (21).
In order to examine the glutamatergic signaling pathway, the human teratocarcinoma cell line NTera 2/cl.D1 (NT2N) was used. NT2N cells are post-mitotic, terminally differentiated neurons that possess cell surface markers consistent with neurons of the central nervous system. Pure cultures of neurons are obtained through differentiation of the stem cell population with retinoic acid and treatment of the replated neuronal population with mitotic inhibitors to clear any contaminating nonneuronal cells. NT2N neurons provide a strong model for examining APP secretion because they endogenously express APP and secrete both APP and A␤ in response to physiological stimuli. In contrast to non-neuronal cells, NT2N neurons express APP 695 , the major APP isoform expressed in the brain, and NT2N neurons process APP differently (6,22). For this study, NT2N neurons were used to determine if glutamate is involved in the regulation of APP secretion.

EXPERIMENTAL PROCEDURES
Materials-The media and serum used to maintain the NT2N culture were from Life Technologies, Inc., and all of the mitotic inhibitors and antibiotics used in the cell culture were purchased from Sigma. Matrigel was from Collaborative Research (Bedford, MA). The DAG kinase and lipase inhibitors were from Biomol (Plymouth Meeting, PA), and the glutamate receptor agonists and antagonists were purchased from Tocris Cookson (Ballwin, MO). The phorbol ester PMA was from Sigma and the calcium ionophore A23187 was from Boehringer Mannheim. [  Products. Trichloroacetic acid, trichlorotrifluorethane, and trioctylamine used in the inositol extraction procedure were also from Sigma. SAX Amprep minicolumns were obtained from Amersham Pharmacia Biotech. RNA isolation kit was from CLONTECH, RT-PCR reagents were from Promega (Madison, WI), and the mGluR1 primers were from the Wistar Institute (University of Pennsylvania, Philadelphia, PA), and mGluR5 primers were from National Biosciences Inc. (Beverly, MA).
Measurement of APP-S Secretion by Metabolic Labeling and Immunoprecipitation-In order to metabolically label APP, NT2N cells were serum-starved for 20 min in DMEM methionine-free medium and then labeled with 100 -400 Ci/ml [ 35 S]methionine in DMEM methioninefree medium (1% penicillin/streptomycin, 5% fetal bovine serum) for 3 h. The cells were then treated as described above. After treatment, the supernatant was removed and centrifuged for 15 min at 15,000 ϫ g to remove any remaining cells. The supernatant was precleared with protein A-Sepharose beads and then immunoprecipitated overnight with 10 -20 g of an anti-APP-S polyclonal antibody (Karen) and protein A-Sepharose beads as described previously (6). Immunoprecipitates were analyzed on 7.5% SDS-PAGE mini-gels, and the radioactivity quantitated using Image Quant software on a Molecular Dynamics PhosphorImager. For data analysis, the control was designated as 100% within each experiment due the differences in radioactive counts between the controls in separate experiments. The glutamate-induced APP-S secretion was determined to be specific for APP after demonstrating that glutamate did not have an effect on the amount of total secreted proteins (data not shown).
RT-PCR Analysis of Metabotropic Glutamate Receptor Subtypes mGluR1␣a/␤b and mGluR5␣a in NT2N Cells-60 -90 g of total RNA was isolated from NT2N cells as described previously (24). Reverse transcription was then performed as follows. For reverse transcription, RNA was incubated at 65°C for 5 min and then placed on ice for 10 min. For each reaction of 31 l, less than 30 g of total RNA was added to the reaction mixture of: RNase inhibitor (40 units/ml), first strand cDNA 5 ϫ buffer, Moloney murine leukemia virus reverse transcriptase (200 units/ml), bovine serum albumin (1 mg/ml), 0.1 M dithiothreitol, and oligo(dT) (1.5 g). The reverse transcription reaction was performed in a thermal cycler for 60 min at 42°C, followed by a 10-min incubation at 72°C. The cDNA generated was then amplified in a 50-l PCR using 8.5 l of the cDNA, 10 ϫ PCR buffer, 1 l of dNTP mix (10 mM each of dATP, dCTP, dGTP, dTTP), 0.25 l of Taq polymerase (5 units/l), and 2.5 l of each specific primer (20 M each) (designed from the published sequence of human mGluR1 (5Ј CTGCATGTTCACTCCCAAGATGTA-CAT, 3Ј CACGCGCCTGTGCACCACCATGGAAG) and mGluR5 (5Ј TGTGCCCAGCTAGTGATTGC, 3Ј TGCTCTTCTCATTCTGGGC) (25,26). The reaction mixtures were layered with mineral oil, placed in the thermal cycler, and programmed for 2 min at 94°C, followed by 40 cycles at 94°C for 1 min, 60°C for 2 min, 72°C for 3 min, and then a final 7 min at 72°C. The tubes were placed at 4°C until the PCR products were analyzed for purity and size on a 1.5% agarose gels. The products were excised from the gels and purified with the Geneclean II kit and the purified products subjected to a restriction enzyme digest with EcoRI for the identification of mGluR1␣/␤ and BglI for mGluR5␣.
Cytosolic Free Ca 2ϩ Measurement-Cells were loaded with fura-2 during a 40-min incubation at 37°C in 2 ml of Krebs-Hepes buffer (115 mM NaCl, 24 mM NaHCO 3 , 5 mM KCl, 1 mM MgCl 2 , 2.5 mM CaCl 2 , 25 mM glucose, and 25 mM Hepes, pH 7.40) supplemented with 2.5 M fura-2 acetoxymethylester (Molecular Probes, Eugene, OR) and 0.2 mg/ml pluronic F-127 (Molecular Probes), which was used to increase the loading. The coverslip with the loaded cells was then mounted in a perifusion chamber placed on the homeothermic platform of an inverted Zeiss microscope. The cells were superfused with Krebs-Hepes buffer at 37°C at a flow rate of 1.5 ml/min. The microscope was used with a ϫ 40 oil objective. Fura-2 was successively excited at 334 and 380 nm by means of two narrow band-pass filters. The emitted fluorescence was filtered through a 520-nm filter, captured with an Attofluor CCD video camera at a resolution of 512 ϫ 480 pixels, digitized into 256 gray levels, and analyzed with version 6.00 of the Attofluor RatioVision software (Atto Instrument, Rockville, MD). The concentration of Ca 2ϩ was calculated by comparing the ratio of fluorescence at each pixel to an in vitro 2-point calibration curve. The Ca 2ϩ concentration is presented by averaging the values of all pixels in the body of a differentiated NT2N cell (neuron soma), excluding the Ca 2ϩ data of neurites. Data points were collected at an interval of 4.5 s.
Data Analysis-Student's t test was performed when two groups were compared. Analysis of variance was used, followed by the Student-Newman-Keuls method when multiple groups were compared. In cases where the data did not have a normal distribution, Kruskal-Wallis one-way analysis of variance on ranks was used, followed by Dunn's method of multiple comparison. Differences were considered significant for p Ͻ 0.05.

Glutamate Stimulates APP-S Secretion from NT2N Neurons in a Time-and Dose-dependent Manner-NT2N
neurons were metabolically labeled with [ 35 S]methionine and then stimulated with 1 mM glutamate for 0 -90 min or various concentrations of glutamate for 30 min. APP-S was immunoprecipitated from the medium and the proteins separated on 7.5% SDS-PAGE gels. Fig. 1 (top panel) shows the resulting APP-S bands. The lower APP-S band is the 695 form, while the higher APP-S band is the 751 and/or 770 form, the result of any remaining non-neuronal cells within the culture. Glutamate treatment resulted in a 2-fold increase in APP-s secretion between 5 and 30 min, with maximal secretion at 30 min (p Ͻ 0.05 versus control) (Fig. 1, bottom panel). Between 30 and 90 min, glutamate-induced secretion decreased to the control level of secretion. Glutamate-induced APP-S secretion was dose-dependent (Fig. 2, bottom panel). Fig. 2 (top panel) shows a representative gel depicting an increase in APP-S 695 secretion with increasing concentrations of glutamate. Glutamate also caused a rapid and sustained increase in cytosolic Ca 2ϩ (Fig. 3, top panel). Likewise, the glutamate-induced increase in cytosolic Ca 2ϩ was also dose-dependent (Fig. 3, bottom panel).
Metabotropic Glutamate Receptor Agonists Stimulate APP-S Secretion from NT2N Neurons-Glutamate receptors can be categorized into two distinct groups, ionotropic and metabo-tropic receptors (27,28). Ionotropic receptors are selectively permeable ion channels and are subdivided into three groups according to agonist selectivity, NMDA, kainate, and AMPA receptors. The metabotropic receptors are seven transmembrane-spanning proteins coupled to intracellular second messengers. The metabotropic receptor family consists of at least eight different subtypes that are subdivided into three groups according to sequence similarities, agonist selectivity, and intracellular signaling machinery. Therefore, the effects of both metabotropic and ionotropic glutamate receptor agonists and antagonists were examined in NT2N cells. The effects of both metabotropic and ionotropic glutamate receptor agonists on APP-S secretion are shown in Table I. The ionotropic glutamate receptor agonist NMDA (10 M) did not have any effect on APP-S secretion; however, the non-NMDA glutamate receptor agonists AMPA (100 M) and kainate (100 M) caused a 1.5-fold and a 2-fold increase, respectively, in APP-S secretion. AMPA and kainate also caused an increase in intracellular Ca 2ϩ (Fig.  4). The nonspecific glutamate receptor agonist quisqualate (100 M) had a slight effect on APP-S secretion; however, the specific metabotropic glutamate receptor group I and II agonists, ACPD (100 M) and CCG-I (100 M) both resulted in a 1.5-fold increase in APP-S secretion, suggesting the glutamate-induced APP-S secretion is mediated in part by metabotropic glutamate receptors.  (Table II) (29 -31). These results suggest that glutamate-induced APP-S secretion is mediated in part by metabotropic glutamate receptors.

Metabotropic Glutamate Receptor Antagonists Reduce Glutamate-induced APP-S Secretion in NT2N Cells
Identification of Metabotropic Glutamate Receptor Subtypes mGluR1␣/␤ and mGluR5␣ in NT2N Neurons-The metabotropic glutamate receptor subtypes, mGluR1␣/␤ and mGluR5␣, were identified in NT2N cells by RT-PCR analysis. For mGluR1␣/␤, RT-PCR yielded the expected products of 463 and 547 bp, respectively (Fig. 5), while for mGluR5␣, a PCR product of the expected size of 600 bp was obtained (Fig. 5). The products were excised from the gels, purified with the Geneclean II kit, and subjected to a restriction enzyme digest with EcoRI for the identification of mGluR1␣/␤ and BglI for mGluR5␣. Re-  Fig. 1 and then pretreated for 30 min with various glutamate receptor antagonists. The cells were then treated for 30 min with the control, 1 mM glutamate, the antagonist alone, or 1 mM glutamate ϩ the antagonist. APP-S secretion was quantitated by PhosphorImager analysis as the amount of radioactivity in the APP-S band and is expressed as a percentage of the control within each experiment. Results are shown as the mean Ϯ S.E. of APP-S secretion from 3 to 14 separate observations/condition.  mM glutamate for 2 min. The inositol phosphates were extracted and purified by strong anion exchange chromatography. Glutamate caused an increase in inositol 1,4,5-trisphosphate at 2 min in NT2N neurons (p Ͻ 0.05 versus control) (Fig.  6). This suggests that glutamate activates the PI-linked metabotropic glutamate receptors resulting in the activation of phospholipase C and the subsequent increase in IP 3 .

TABLE II Effect of glutamate receptor antagonists on APP-S secretion in NT2N cells Cells were labeled for 3 h with [ 35 S]methionine as in
Increases in Calcium and Diacylglycerol Increase APP-S Secretion-Because glutamate-induced APP-S secretion involves the phospholipase C pathway, the contribution of downstream signaling events was next examined. NT2N neurons were treated with the phorbol ester PMA, the cell-permeable DAG analog OAG, the calcium ionophore A23187, thapsigargin, an inhibitor of the endoplasmic reticulum Ca 2ϩ -ATPase, the DAG kinase inhibitor R59949, or the DAG lipase inhibitor RG80267. Both PMA and OAG caused a robust increase in APP-S secretion (p Ͻ 0.05 versus control), comparable with that of glutamate (Fig. 7). These results suggest that protein kinase C is involved in the secretion of APP-S. Furthermore, the DAG kinase inhibitor R59949, which in other systems increases DAG levels, caused a large increase in the secretion of APP (Table III) (32). However, DAG metabolism does not seem to be involved in APP-S secretion, since the DAG lipase inhibitor RG80267 did not have any effect on glutamate-induced APP-S secretion (Table III). The calcium ionophore A23187 also caused an increase in APP-S secretion (3.5-fold), and thapsigargin caused a slight decrease in glutamate-induced APP-S secretion. These results suggest that both calcium and DAG activation of protein kinase C mediate glutamate-induced APP-S secretion in NT2N neurons. DISCUSSION We have shown that the metabotropic glutamate receptors mGluR1 and mGluR5 are present in NT2N cells and that upon stimulation with glutamate, increase the accumulation of the second messenger IP 3 and the secretion of APP-S in a time-and dose-dependent manner. This is a novel demonstration of glutamatergic regulation of APP-S secretion in a nontransfected human neuronal cell line.
The regulation of receptor-mediated APP-S secretion has been studied extensively. However, a complete model of the signaling pathway has yet to be developed. APP-S secretion has previously been shown to be regulated by the muscarinic re-ceptor signaling pathway (19). Muscarinic receptor activation results in the activation of a phospholipase C-linked pathway, as seen by increases in intracellular calcium, DAG, and IP 3 , and subsequent secretion of APP-S (19). The stimulation of APP-S secretion by glutamate receptors has also been shown recently (20,(33)(34)(35). Glutamate, quisqualate, and ACPD were shown to increase APP-S secretion from rat hippocampal neurons, as well as from rat cortical and hippocampal brain slices. Glutamate, quisqualate, and ACPD have also been shown to increase intracellular IP 3 accumulation in cultured rat hippocampal neurons, cortical and hippocampal rat brain slices, and astrocytes (20,34,35). Ionotropic glutamate receptor agonists NMDA, AMPA, and kainate did not have any effect on APP-S secretion from rat hippocampal neurons or from rat   Fig. 1 and then pretreated for 30 min with thapsigargin, RG80267, or R59949. The cells were then treated for 30 min with the control, 1 mM glutamate, A23187, the inhibitor alone, or 1 mM glutamate ϩ the inhibitor. APP-S secretion was quantitated by PhosphorImager analysis as the amount of radioactivity in the APP-S band and is expressed as a percentage of the control within each experiment. Results are shown as the mean Ϯ S.E. of APP-S secretion from 4 to 13 separate observations/condition. cortical and hippocampal brain slices. However, it is not known which glutamate receptors are directly involved in the stimulation of APP-S secretion. It has been demonstrated that metabotropic glutamate receptors are involved in glutamatemediated APP-S secretion. These findings support our data and conclusion that metabotropic glutamate receptors are involved in the regulation of APP-S secretion. For our model, the human neuronal NT2N cells were used. NT2N cells provide a unique model system because they are human cells that endogenously express APP and A␤. These cells also express the muscarinic receptor subtypes m2 and m3 and have been previously shown to express ionotropic glutamate receptors (36,37). We currently have demonstrated the presence of the metabotropic glutamate receptor subtypes mGluR1 and mGluR5 in NT2N cells. There are at least eight different subtypes of metabotropic glutamate receptors, and these are subdivided into three groups according to sequence similarities, agonist selectivity, and intracellular signaling machinery. The metabotropic glutamate receptors mGluR1 and mGluR5 are the group I receptors, coupled to PI hydrolysis, and upon stimulation with specific receptor agonists resulted in an increase in secretion of APP-S from these cells. Specific metabotropic glutamate receptor antagonists inhibited receptor-mediated APP-S secretion, suggesting that the agonists were specific for the metabotropic receptors. In addition, but contrary to previous reports in non-neuronal cells and rat neurons, the ionotropic glutamate receptor agonists AMPA and kainate also caused an increase in the secretion of APP-S from NT2N cells. This we believe is the result of the increase in intracellular calcium levels upon activation of these receptors (Fig. 4). The calcium ionophore A23187, which results in an increase in intracellular calcium concentrations, also caused an increase in the secretion of APP-S, suggesting that calcium is a potentiator of APP-S secretion. Furthermore, stimulation of these cells with glutamate resulted in the activation of a phospholipase C-linked pathway, as seen by an increase in IP 3 accumulation. We also examined the effect of OAG, a cellpermeable DAG analog, and DAG lipase and kinase inhibitors to determine if DAG metabolism or generation is involved in the glutamate-induced APP-S secretion. The DAG lipase inhibitor RG80267 did not have any effect on glutamate-induced APP-S secretion, suggesting that DAG metabolism is not involved. However, both OAG and the DAG kinase inhibitor R59949, which in other systems has been shown to increase DAG levels (38), caused an increase in glutamate-induced APP-S secretion, suggesting that DAG is a downstream mediator in glutamate receptor signaling. The effect of the protein kinase C activator PMA on APP-S secretion mimicked the effect seen by glutamate, suggesting that protein kinase C is also involved as a downstream mediator in the regulation of APP-S secretion. Protein kinase C has extensively been shown to be involved in the regulation of APP-S secretion (21,39,40); however, it has not been demonstrated how protein kinase C is involved or which isoform is involved in APP-S secretion.
Metabotropic glutamate receptors of Groups II and III are negatively coupled to cAMP, and in astrocytes, forskolin and dibutyryl cyclic AMP inhibited mGluR agonist-induced APP-S secretion and increased APP gene expression and the cellular APP holoprotein. (35). These studies demonstrated an inhibition of both glutamate-and phorbol ester-stimulated APP-S secretion with forskolin, an activator of adenylate cyclase, and dibutyryl cyclic AMP, suggesting that activation of the cAMP pathway inhibits APP-S secretion. Additionally, activation of adrenergic receptors coupled to cAMP also increased APP gene expression and APP holoprotein in astrocytes (42). In C6 cells transfected with APP 751 , forskolin, dibutyryl cyclic AMP, and isoproterenol increased the concentration of cAMP, and forskolin inhibited both the constitutive and phorbol ester-mediated secretion of nexin II, the ␣-secretase cleavage product of APP 751 (43). These studies also demonstrate an involvement of cAMP in APP-S secretion. In contrast, Xu et al. (44) showed that forskolin stimulated the secretion of APP-S from PC12 cells. Therefore, the effect of the cAMP pathway on APP-S secretion still remains unclear and may be cell-specific. In NT2N cells, preliminary experiments examining the effect of forskolin on APP-S secretion did not demonstrate a change in APP-S secretion after the addition of forskolin (APP-S secretion of 105% Ϯ 11% of control in the presence of 10 M forskolin, n ϭ 6). However, the metabotropic glutamate receptor agonist CCG-I (a Group II agonist) did stimulate APP-S secretion, although this could be due to its reported stimulation of mGluR1 and phosphoinositide hydrolysis (41). In addition, Group II/III antagonists partially inhibited glutamate-induced APP-S secretion (Table II). These conflicting results suggest that activation of the cAMP pathway may also be an important regulatory mechanism of APP-S secretion in NT2N cells and will require further investigation.
In conclusion, we have demonstrated that glutamate stimulates the metabotropic glutamate receptor signal transduction pathway in human neurons causing an increase in the second messenger IP 3 and the secretion of APP-S.