Glutamate Acting on N-Methyl-d-aspartate Receptors Attenuates Insulin-like Growth Factor-1 Receptor Tyrosine Phosphorylation and Its Survival Signaling Properties in Rat Hippocampal Neurons*

Impairing intracellular signaling induced by survival factors and excess glutamate have recently been suggested to play important role in neurodegenerative processes. However, the underlying mechanism(s) and interrelationships between these factors mostly remain to be established. In the present study, we show that glutamate attenuates the tyrosine phosphorylation of the insulin-like growth factor-1 (IGF-1) receptor and the survival effect of IGF-1 (100 nm) in hippocampal cultured neurons. Pretreatment of cultured hippocampal neurons with glutamate concentration dependently inhibited the tyrosine phosphorylation of IGF-1 receptors as well as that of IRS-1 and Shc, two IGF-1 receptor adapter proteins. The effect of glutamate was also evident on the phosphorylation of Akt, as well as its upstream kinase PI3K/PDK1 and downstream targets, GSK3β and FOXO3a. The inhibitory effect of glutamate (1 mm) was blocked by antagonists of the N-methyl-d-aspartate (NMDA) receptor, including MK801 (20 μm) and AP5 (100 μm), but not by blockers of other ionotropic or metabotropic glutamate receptor sub-types demonstrating the involvement of the NMDA receptor. This hypothesis is supported further by the observation that treatment with NMDA concentration dependently inhibited the activation and phosphorylation of IGF-1 receptors and downstream targets induced by IGF-1 (100 nm). These findings demonstrate that glutamate can block the effect of IGF-1 by decreasing IGF-1 receptor signaling and responsiveness, hence attenuating the survival properties of this trophic factor in neuronal cells. Our results also suggest a novel mechanism by which glutamate can reduce cell viability and induce neurotoxicity.

Insulin-like growth factor-1 (IGF-1) 2 is a trophic factor with multiple biological functions, including important roles during development, and in the maintenance of cellular integrity throughout the organism (1,2). Various studies have demonstrated the existence of both IGF-1 and IGF-1 receptors (IGF-1Rs) in different brain regions, including the hippocampal formation, a region known to play important roles in learning processes and that is severely affected in Alzheimer disease (2,3). IGF-1 possesses trophic effects in the hippocampus and promotes survival of cultured hippocampal neurons against various insults (3)(4)(5)(6)(7)(8). Consistent with its neuroprotective role, levels of IGF-1 are up-regulated following various brain insults supporting the notion that IGF-1 is a key neuroprotective/neurorescuing factor (3,9).
Glutamate is the major excitatory neurotransmitter in the mammalian brain, responsible for basal excitatory synaptic transmission and many forms of synaptic plasticity such as long term potentiation and long term depression associated with cognitive processes. The various effects of glutamate are mediated by specific receptors belonging to two major families, namely ionotropic and metabotropic receptors (24,25). Ionotropic glutamate receptors are ligand-gated ion channels consisting of three subtypes known as AMPA(␣-amino-3-hydroxy-5-methy1-4-isoxazoxazole propionate), NMDA (N-methyl-Daspartate), and kainate receptors on the basis of the affinity of their preferential agonist (24,25). Metabotropic (mGlu) receptors are G-protein-coupled receptors divided into three main classes. Group I mGlu receptors are coupled to phospholipase C and intracellular Ca 2ϩ signaling, whereas group II and III receptors are negatively coupled to adenylyl cyclase and the production of cAMP. Preferential metabotropic receptor agonists include the excitatory amino acid analogue L-quisqualic acid (group I), DCG-IV (group II), and L-AP4 (group III).
Glutamate receptors are essential for the normal functioning of the brain. However, their excessive activation can lead to neuronal damage ranging from acute hypoxic-ischemic brain injuries to chronic neurodegenerative diseases (24,25). Interestingly, it has been suggested that excessive glutamate levels can block the activation of Akt, and hence alter cell survival, linking glutamate excitotoxicity to impaired cell survival signaling (26). However, it is not clear from this earlier study if trophic factor signaling pathways are involved. We report here that glutamate is able to attenuate the survival promoting effects of IGF-1 by interrupting IGF-1R survival signaling. The effect of glutamate is mediated by the NMDA receptor sub-type as mimicked by NMDA itself and blocked by an antagonist, MK-801. Taken together, these data reveal a novel mechanism by which glutamate and NMDA receptors regulating trophic factor signaling may be involved in a variety of neurotoxic events leading to neurodegenerative diseases.
Hippocampal Neuronal Cultures-Hippocampal cultured neurons were prepared as described by Zheng et al. (27) with minor modifications and from fetuses (embryonic day 20) of pregnant Sprague-Dawley rats (Charles River Breeding Laboratories, St. Constant, Quebec, Canada). Animal care was according to protocols and guidelines approved by McGill University Animal Care Committee and the Canadian Council for Animal Care. Hippocampi were dissected in Ca 2ϩ -and Mg 2ϩ -free Hanks' balanced salt solution (HBSS) supplemented with 15 mM HEPES, 10 units/ml penicillin, and 10 g/ml streptomycin. Tissues were collected and washed 4 or 5 times with HBSS and then submitted to an enzymatic digestion at 37°C with 0.25% trypsin in HBSS media for 10 min. The reaction was stopped by the addition of fetal bovine serum (final concentration 10%), and tissue was rinsed with HBSS 4 -5 times to remove fetal bovine serum. The cellular suspension was then obtained by repeating aspirations through a pasteur pipette. Following a centrifugation at 800 ϫ g for 10 min, the medium was removed and cells were resuspended in a chemically defined serum-free Neurobasal medium supplemented with 2% B27, 20 M L-glutamine, 15 mM HEPES, 10 units/ml penicillin, and 10 g/ml streptomycin. Neurons were plated at density of 5-8 ϫ 10 5 cells/ml in culture plates (coated with 10 g/ml poly-D-lysine) under serum-free conditions and grown at 37°C with 5% CO 2 humidified atmosphere. On the day following the plating, the medium was replaced with fresh culture medium. Medium was changed again with either the medium as above or Neurobasal supplemented with 1% N 2 after 4 -5 days. Experimental treatments were performed on the seventh day after plating.
Treatments-Before each experiment, the culture medium was replaced with Neurobasal 2 h before adding the desired reagents. To study the signaling pathways induced by IGF-1 in cultured hippocampal neurons, cells were treated with 100 nM IGF-1 for 6 -10 min. To study the effect of glutamate, neurons were pretreated for 1 h with various concentrations of glutamate or NMDA before exposure to IGF-1 as above. Alternatively, cells were first exposed to MK801 (20 M, 20 min), AP-5 (50 M, 20 min), 6,7-Dinitroquinoxaline-2,3-dione (DNQX) (10 M, 20 min), LY 341495 (300 nM, 20 min), and CPCCOEt (100 M, 20 min) followed by a pretreatment with glutamate and then stimulation with 100 nM IGF-1. All experiments were repeated at least three to four times.
Transfection of Cultured Hippocampal Neurons with SHP-2 and siRNA for SHP-2 by Nucleofector TM -Primary hippocampal neurons from fetuses (embryonic day 20) were prepared as above and the transfections of SHP-2, and its siRNA was performed by Nucleofector (Amaxa, Germany) with 96-well shuttle rat neuron Nucleofector kit according to the manufacturer's instructions.
Determination of Tyrosine Phosphorylation of IGF-1 Receptor, IRS-1, Shc, and P13K by Immunoprecipitation-Cultured hippocampal neurons were treated as described above except that exposure to IGF-1 was 8 min. Cells were then rinsed with cold phosphate-buffered saline twice and collected. After centrifugation at 1000 ϫ g for 5 min at 4°C, cell pellets were lysed on ice in pre-cold RIPA buffer for 20 min. Cell lysates were then pelleted at 13,000 ϫ g for 10 min, and the concentration of protein in each sample was determined using the Bio-Rad dyebinding method with bovine serum albumin as standard. The supernatant with equal amounts of protein was incubated overnight at 4°C with either anti-IGF-1R, anti-IRS-1, anti-Shc, or anti-PI3K antibodies. Formed immunocomplexes were isolated by protein A/G PLUS-agarose (Santa Cruz Biotechnology), separated by 4 -20% SDS gel, and then tyrosine phosphorylation was determined by Western blot using a mixture of anti-phosphotyrosine antibodies 4G10 and PY99. Blots were striped and reprobed with antibodies to ensure equal protein loading.
Potential interaction of IGF-1R with NMDA receptors was determined by coimmunoprecipitation of IGF-1 and NMDA receptors using anti-IGF-1R and anti-NMDA receptor antibodies as described above (2). No direct interaction was detected (data not shown).
Statistical Analysis-Data are expressed as mean Ϯ S.E. A one-way analysis of variance with a Newman-Keuls test was used to calculate difference between means, and p Ͻ 0.05 was considered significant.

Glutamate Attenuates the Tyrosine Phosphorylation of IGF-1 Receptors Induced by IGF-1 and IGF-1-promoted Survival Effects in Cultured Hippocampal
Neurons-Impaired IGF-1 intracellular signaling and excess glutamate have been suggested to be involved in neurodegenerative processes (24). To investigate the effects of exposure to glutamate on IGF-1R signaling and its survival properties, cultured hippocampal neu-rons were pretreated with 1 mM glutamate, and the tyrosine phosphorylation of IGF-1Rs and the survival effects of IGF-1 were determined (Fig. 1). IGF-1 caused the tyrosine phosphorylation of IGF-1R (Fig. 1, A and B) and protected cultured hippocampal neurons from B27 deprivation-induced cell death (Fig. 1C). Pretreatment with glutamate (1 mM) significantly attenuated the tyrosine phosphorylation of IGF-1Rs (Fig. 1, A  and B) and the survival effects of IGF-1 (Fig. 1C).
Glutamate Decreases Tyrosine Phosphorylation of IRS-1, Shc, and PI3K-We studied next the effect of glutamate on IGF-1R downstream signaling. Fig. 2A shows that glutamate significantly blocked the phosphorylation of Shc-and IRS-1, two main adaptor proteins of the IGF-1R and that of PI3K, a main downstream signaling protein of IGF-1 ( Fig. 2A). Moreover, the effect of glutamate in the tyrosine phosphorylation of the IGF-1R and its adaptor protein IRS1 was clearly concentrationdependent (Fig. 2B).
Glutamate Attenuates IGF-1-induced Activation of the PI3K/ Akt Pathway While Promoting the Phosphorylation of IGF-1stimulated MAPK-We have previously shown that the PI3K/ Akt pathway is essential for the survival effects of IGF-1, whereas the MAPK pathway plays a rather minimal role (27,28). Accordingly, we studied the comparative effect of a pretreatment with glutamate on IGF-1-induced activation of the PI3K/Akt and MAPK pathways. Consistent with results in Fig.  2A showing that a pretreatment with glutamate (1 mM) attenuated the tyrosine phosphorylation of PI3K, it also inhibited the activation of PDK1 (an upstream kinase of Akt), the phosphorylation of both Thr-308 and Ser-473 of Akt, and the phosphorylation of GSK3␤ (Fig. 3) and FOXO3a (Fig. 4). In contrast, the same treatment enhanced IGF-1-induced activation of MAPK and its upstream kinase MEK. Glutamate (1 mM), by itself, only had a minor effect on the Akt pathway while more powerfully stimulating the phosphorylation of MEK and MAPK (Fig. 3). In addition to Akt and MAPK pathways, a pretreatment with glutamate (1 mM) also attenuated the activation of different iso- forms of PKC, including PKC␣, PKC␤, PKC␥, and PKC␦, and had a minor effect on Src kinase (Fig. 3).
The Inhibitory Effect of Glutamate on IGF-1 Receptor Signaling Is Mediated by the NMDA Receptor-Knowing that glutamate is able to attenuate the IGF-1R/PI3K/Akt signaling survival pathway, we examined next the role of various glutamate receptor sub-types on this effect. Cultured hippocampal neurons were pretreated with the selective NMDA receptor antagonists MK801 or AP-5, DNQX (antagonist for the non-NMDA receptors, AMPA and kainate receptors), CPCCOEt (selective, non-competitive mGluI antagonist), and LY 341495 (mGluII/ III antagonist), and the effect of glutamate (1 mM) on IGF-1R phosphorylation and signaling was determined. MK801 (20 M), a non-competitive NMDA receptor antagonist, inhibited the effect of glutamate, whereas AP-5, a competitive one, also blocked it but to a lesser extent. In contrast, DNQX at a con-FIGURE 2. Glutamate reduces the tyrosine phosphorylation of IGF-1R substrate proteins IRS-1, Shc, and PI3K in cultured hippocampal neurons. Cultured hippocampal neurons were pretreated with 1 mM glutamate for 1 h, and then exposed to 100 nM IGF-1 for 8 min, and the tyrosine (ty) phosphorylation of IRS-1, Shc, and PI3K was determined as described under "Experimental Procedures." As shown in A, IGF-1 dramatically increased the tyrosine phosphorylation of these proteins, whereas a pretreatment with glutamate significantly inhibited the effect of IGF-1. B, the effect of glutamate on the tyrosine phosphorylation of IGF-1Rs and IRS-1 is concentration-dependent. Blots represent prototypical examples of experiments replicated at least three times.  centration (20 M) known to block non-NMDA receptors, had no effect (Fig. 4A). The metabotropic glutamate receptor subtypes antagonists CPCCOEt and LY 341495 also had no effect on the inhibitory effect of glutamate on IGF-1 signaling (data not shown).
Consistent with the role of NMDA receptor on the effect of glutamate on IGF-1R signaling, Fig. 4B showed that 1 mM glutamate attenuates IGF-1-induced phosphorylation of Akt and its downstream targets GSK3␤ and FOXO3a, whereas pretreatment of hippocampal cultured neurons with this NMDA receptor antagonist reversed the effect of glutamate. On another hand, antagonists of non-NMDA receptors and metabotropic glutamate receptors had no effect (Fig. 4B).
NMDA Impairs the Signaling of IGF-1 Receptors in Cultured Hippocampal Neurons-To verify further the role of the NMDA receptor subtype, the effect of NMDA itself was tested on IGF-1-induced receptor signaling.  (Fig. 6).
Glutamate Stimulates the Phosphorylation/Activation of the Tyrosine Phosphatase SHP-2 While Its Inhibitor NaVO 4 Attenuates the Effect of Glutamate-It was recently reported that the activation of the tyrosine phosphatase SHP-2 is able to dephosphorylate TrkB (29). We thus investigated the role of SHP-2 on the action of glutamate on IGF-1R signaling. Fig. 7A shows that glutamate is able to time dependently activate the phosphorylation and activation of SHP-2 while the tyrosine phosphatase inhibitor NaVO 4 attenuated the effect of glutamate on the phosphorylation of Akt (Fig. 7B) and IGF-1R (Fig. 7C).

siRNA for SHP-2 Attenuated the Effect of Glutamate in Hippocampal Neurons Enhanced Its
Effect-To further confirm the role of SHP-2, hippocampal neurons were transfected with a mixture of four siRNA for SHP-2 using Nucleofector technology. Fig. 8 shows that siRNA for SHP-2 inhibited the expression of SHP-2 and attenuated the effect of glutamate on the phosphorylation of IGF-1R.

DISCUSSION
In the present study, we showed that glutamate, at concentrations not inducing excitotoxicity, is able to attenuate the survival effects of IGF-1. Glutamate acted by impairing IGF-1R coupling and signaling to the PI3K/Akt/FOXO pathway while possibly promoting IGF-1 action on the MAPK pathway. The effects of glutamate are mediated by the NMDA receptor subtype.
Glutamate Inhibits the Survival Effect of IGF-1 in Hippocampal Neurons-The survival effect of IGF-1 was attenuated by a treatment with glutamate at concentrations as low as 100 M. At such low concentrations, glutamate, by itself, had no apparent effect on cell survival suggesting that its effects were not due to excitotoxicity but to the inhibition of the survival properties of IGF-1. Our data also suggest that cell death inducers, such as glutamate, are not only able to activate pathways leading to cell death, but can also inhibit survival pathways hence amplifying their effects. Comparable findings have been reported from another group (37) and for tumor necrosis factor-␣, a pro-inflammatory cytokine known to mediate various forms of experimental neurodegeneration and to induce insulin/ IGF-1 resistance in various tissues (30 -33). Similar results were also obtained with prion proteins, ethanol, and methylmercury (34 -37).
Glutamate Impairs the Survival Signaling of IGF-1 at the Level of the IGF-1 Receptor-Exposure to glutamate significantly blocked the survival effects of IGF-1 by reducing IGF-1R  signaling. This hypothesis is based on the following experimental results. First, glutamate, concentration dependently inhibited the tyrosine phosphorylation of the IGF-1R and of its adaptor protein IRS-1. Second, glutamate reduced the phosphorylation of PI3K and PDK1 (an upstream kinase of Akt), leading to reduced Akt activity via reduced phosphorylation. This event results in the activation of death effectors such as FOXO3a by reducing their phosphorylation. Third, the glutamate treatment preferentially affected the PI3K/Akt over the MAPK pathway. Hence, a pre-exposure to glutamate can reduce IGF-1 survival effect by directly altering IGF-1R phosphorylation and signal transduction. Consistent with our results, a recent report showed that glutamate was able to activate the tyrosine phosphatase SHP-2 in cerebellar neurons, which enhanced the association of SHP-2 with TrkB, leading to the inhibition of the phosphorylation of this receptor and diminished trophic properties (29). In accordance with these findings, preliminary data also showed that glutamate is able to attenuate the phosphorylation of Akt induced by brain-derived neurotrophic factor in hippocampal neurons. 3

NMDA Receptors Mediate the Effect of Glutamate on IGF-1
Receptor Signaling and Survival-Specific antagonists of the various glutamate receptors were used to establish which subtype was responsible for the effects of glutamate on IGF-1R signaling and survival. Our data demonstrate that the NMDA receptor subtype likely plays a key role. Indeed, only the NMDA receptor antagonists MK801 and AP-5 inhibited the effects of glutamate on the tyrosine phosphorylation of the IGF-1R and its downstream signaling targets. Moreover, the specific agonist, NMDA, concentration dependently mimicked the effect of glutamate. Taken together, these data show that the NMDA receptor sub-type is involved in mediating the action of glutamate on IGF-1R signaling and cell survival.
The one or more precise mechanisms underlying the effect of NMDA receptor activation on IGF-1R signaling and IGF-1mediated cell survival are not clear. NMDA and IGF-1Rs do not directly interact with each other as coimmunoprecipitation using various NMDA and IGF-1R antibodies failed to reveal any direct interaction (data not shown). Accordingly, glutamate 3 W. Zheng and R. Quirion, unpublished results.  acting on NMDA receptors likely activates downstream signaling kinases that phosphorylate serine residues on IGF-1Rs leading to altered receptor tyrosine phosphorylation (38). Alternatively, glutamate could activate phosphatases inducing IGF-1R dephosphorylation. The first hypothesis appears rather unlikely, because glutamate decreased the phosphorylation of classic PKC (this study). PKCs have been shown to be involved in the phosphorylation of serine residues of the IGF-1R in other cells (38). Consistent with this observation, the inhibition of MAPK, a downstream effector of PKC, had no effect on the action of glutamate on IGF-1R signaling, although glutamate seems to enhance the activation of MAPK kinase. Further studies will be required to establish the functional significance of the effect seen here on the MAPK pathway.
Phosphatases may be better candidates to explain the effects observed in the present study. Indeed, it has recently been shown that the activation of the tyrosine phosphatase SHP-2 can dephosphorylate TrkB (29). SHP-2 may also be involved in the phosphorylation of the IGF-1R (39 -40). In vitro, SHP-2 binds to both activated IGF-1 and dephosphorylated IGF-1Rs (39), whereas in vivo, activated IGF-1Rs phosphorylate SHPS-1 (40). The phosphorylation of SHPS-1 promotes the recruitment of SHP-2 to the IGF-1R leading to receptor dephosphorylation (40). Interestingly, glutamate-activated SHP-2 could lead to the dephosphorylation of the IGF-1Rs and reduced signaling efficacy (29). This hypothesis is supported by the following results: (a) glutamate stimulated the phosphorylation of SHP-2; (b) knocking down the expression of SHP-2 by siRNA attenuated the inhibitory effect of glutamate on the phosphorylation of IGF-1R induced by IGF-1; and (c) the tyrosine phosphatase inhibitor, sodium vanadate, blocked the effect of glutamate on IGF-1-induced phosphorylation of IGF-1R and Akt.
In summary, the present study reveals that glutamate, acting via the NMDA receptor sub-type, is capable of impairing IGF-1R signaling and survival effects in hippocampal neurons. These findings suggest a novel mechanism by which glutamate can regulate neuronal viability by altering trophic factor receptor signaling.