Introduction

Top Abstract Introduction Methods Results Discussion References

Glutamate is the main excitatory neurotransmitter in mammals; however, excessive activation of glutamate receptors is neurotoxic, which leads to neuronal death. Excitatory amino acid neurotoxicity has been proposed to contribute to the pathogenesis of different neurodegenerative situations, including ischemia, amyotrophic lateral sclerosis, Huntington's disease and Alzheimer disease. There are several types of glutamate receptors, some of them (ionotropic) are associated with the opening of ion channels; the ionotropic glutamate receptors include NMDA and kainate/(±)--amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors. Other glutamate receptors (metabotropic) are coupled to G-proteins and modulate the activity of certain enzymes (e.g., phospholipase C or adenylate cyclase).

In many systems, glutamate neurotoxicity is mediated by activation of the NMDA type of glutamate receptors (Choi, 1987; Novelli et al., 1988). This leads to the opening of the associated ion channel, allowing the entry of extracellular Ca⁺⁺ and Na⁺ into the neurons. The molecular mechanism of glutamate neurotoxicity is not well understood. Increased intracellular Ca⁺⁺ is an essential step leading to neuronal death (Choi, 1987; Manev et al., 1989), but the subsequent steps leading to neuronal death remain unclear.

It has been reported that tACPD, a selective agonist of mGluRs, attenuates NMDA neurotoxicity in cortical cultures (Koh et al., 1991), in cerebellar cultures (Pizzi et al., 1993; Felipo et al., 1994) and in rat retina in vivo (Siliprandi et al., 1992). This indicates that activation of mGluRs interferes with the neurotoxic process initiated by activation of NMDA receptors.

Eight different subtypes of mGluRs have been described; tACPD is able to activate mGluR1, mGluR2, mGluR3 and mGluR5 (Birse et al., 1993; Bockaert and Fagni, 1993). These receptors are coupled to G-proteins that modulate the activity of different enzymes, mGluR1 and mGluR5 are mainly associated with activation of phospholipase C, whereas mGluR2 and mGluR3 are mainly associated with inhibition of adenylate cyclase (Pin and Duvoisin, 1995; Schoepp, 1993). Therefore the molecular mechanism of the protective effect of tACPD would be different depending on the mGluR responsible for the protection. Cerebellar granule neurons in primary culture express mGluR1a, mGluR2, mGluR3, mGluR4 and mGluR5 (Santi et al., 1994). It was therefore considered of interest to study which of the mGluRs activated by tACPD is responsible for the protective effect against glutamate neurotoxicity in these cells.

We have tested the protective effects of selective agonists for different subtypes of mGluRs against glutamate-induced neuronal death in primary cultures of cerebellar neurons. We have also tested the ability of selective antagonists of different mGluRs to prevent the protective effect of tACPD. The results obtained indicate that activation of mGluR5 prevents glutamate and NMDA neurotoxicity. The results obtained do not allow excluding the possibility that activation of mGluR1 could prevent glutamate neurotoxicity. To test this possibility we should use an agonist that activates mGluR1 but not mGluR5. As far as we know such an agonist is not currently available.

	Methods

Top Abstract Introduction Methods Results Discussion References

Primary neuronal cultures. Primary cultures of neurons were prepared from cerebellum of 8-day-old Wistar rats. Tissue was mechanically dissociated with a pipette. The cell suspension was filtered through a mesh with a pore size of 90 µm and resuspended in 4.6 ml/brain of basal Eagle's medium supplemented with 10% heat-inactivated fetal bovine serum, 2 mM glutamine, 100 µg/ml gentamycin, 5 µg/ml fungizone and 25 mM KCl. Cells were seeded onto polylysine-coated plates, after 15 min at 37°C the medium containing unattached cells was removed and fresh medium was added. To prevent proliferation of nonneuronal cells, cytosine arabinoside (10 µM) was added to the culture medium 20 h after seeding. Cells were incubated at 37°C in 5% CO₂ atmosphere.

Assay of glutamate neurotoxicity and of its prevention by compounds. Determination of glutamate neurotoxicity was carried out essentially as described previously (Felipo et al., 1993). Experiments were carried out 7 days after seeding. Monolayers were washed with Locke's solution without magnesium (154 mM NaCl, 5.6 mM KCl, 3.6 mM NaHCO₃, 2.3 mM CaCl₂, 5 mM HEPES, pH 7.4), containing 5.6 mM glucose. Cells were preincubated with different compounds for 15 min in Locke's solution without magnesium and then incubated with glutamate (1 mM) in the same solution for 4 h at 37°C. Cell viability was determined immediately as described below.

Intravital staining of the culture. Monolayers were washed with the Locke's solution without magnesium and stained for 5 min at 23°C with a mixture of fluorescein diacetate (15 µg/ml) and propidium iodide (4.6 µg/ml). The stained cells were immediately examined with a fluorescence microscope. Fluorescein diacetate crosses the cell membranes and is hydrolyzed by intracellular esterases to produce a green-yellow fluorescent compound (Krause et al., 1984; Novelli et al., 1988). Neuronal injury curtails fluorescein diacetate staining and facilitates propidium iodide penetration and interaction with DNA to yield a bright red fluorescent complex (Favaron et al., 1988). The percentage of surviving neurons was calculated by assessing the ratio fluorescein diacetate to propidium iodide (green/red) staining directly under the microscope. Several randomly chosen fields were counted in each plate. At least 600 cells were counted for each plate.

Materials. MCPG (racemate) and AP-3 were from RBI (Natick, MA). All the other antagonists and agonists were from Tocris Cookson (Bristol, UK). All compounds were dissolved in distilled water and neutralized with NaOH. The concentrations of each compound to be used were determined empirically. Initial experiments were carried out using successive 1-× dilutions of each compound ranging from 1 µM to 1 mM. The concentrations for subsequent experiments were chosen according to the results obtained.

	Results

Top Abstract Introduction Methods Results Discussion References

For the present studies different compounds acting as selective agonists or antagonists of some of the mGluRs have been used. The effects of each compound on the different mGluRs glutamate receptor are summarized in table 1.

The protective effects of different concentrations of the agonists tACPD, DHPG, S4C3HPG and tADA against glutamate-induced neuronal death in primary cultures of cerebellar neurons are shown in figure 1. All these agonists prevent glutamate neurotoxicity nearly completely; the concentrations required to afford protection are different. The concentrations required to prevent the death of one half of the neurons dying as a consequence of glutamate receptors activation were 0.5 µM for tACPD, 4 µM for DHPG, 0.05 µM for S4C3HPG and 1 µM for tADA.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Protective effect of different concentrations of tACPD, DHPG, S4C3HPG or of tADA against glutamate-induced neuronal death. Primary cultures of cerebellar neurons were used 7 days after seeding. Neurons were treated with 1 mM glutamate after preincubation for 15 min with the indicated concentrations of tACPD: agonist of mGluR1, mGluR2, mGluR3 and mGluR5; DHPG: agonist of mGluR1 and mGluR5; S4C3HPG: agonist of mGluR5 and antagonist of mGluR1; or tADA: agonist of mGluR5. Incubation with glutamate was for 4 h and cell viability was quantified immediately as indicated under "Methods." Values are the mean ± standard deviations of at least three different plates from different cultures. At least 600 neurons were counted for each point. For all compounds the neuronal survival in the presence of agonist + glutamate was significantly higher (P < .001) than for neurons treated only with glutamate for all agonists concentrations tested, except for those indicated as N.S. Dotted area represents the mean ± S.D. of the survival of control neurons.

The effect of L-CCG-I was also tested; this compound did not afford protection at any of the large concentrations (500 µM).

To further confirm which of the mGluRs is/are mediating the protective effects of the agonists of mGluRs, we tested whether antagonists selective for some of the mGluRs are able to prevent the protective effect. As shown in figure 2, the protective effect afforded by tACPD against glutamate-induced neuronal death is prevented in a dose-dependent manner by MCPG. In contrast, MTPG is not able to prevent the protection at any of the concentrations tested.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Prevention of the protective effect of tACPD by antagonists of mGluRs. Primary cultured neurons from cerebellum were used 7 days after seeding. Neurons were treated with the indicated concentrations of MCPG (antagonist of mGluR1, mGluR2 and mGluR5) or MTPG (antagonist of mGluR2 and mGluR3). After 15 min, 10 µM tACPD was added to all plates and 1 mM glutamate was added 15 min later.  indicates the sample which was treated only with glutamate. The dotted area shows the survival of neurons treated only with tACPD and glutamate; it can be seen that tACPD prevents neuronal death. Incubation with glutamate was for 4 h, and cell viability was quantified immediately as indicated under "Methods." Values are the mean ± standard deviations of at least three different plates from different cultures. At least 600 neurons were counted for each point. Neuronal survival in the presence or the absence of MTPG was not statistically different. For samples treated with MCPG + tACPD + glutamate, neuronal survival was significantly decreased with respect to those treated only with tACPD + glutamate (P < .001) at all MCPG concentrations tested except when indicated as N.S.

To be sure that the protective agonists are preventing the toxicity mediated by activation of the NMDA receptor, we tested whether they can prevent neuronal death induced by NMDA. As shown in figure 3, NMDA neurotoxicity is completely prevented by MK-801, a selective antagonist of NMDA receptors, and also by tACPD, DHPG and tADA. As is the case for glutamate neurotoxicity, L-CCG-I also did not prevent NMDA neurotoxicity at the 500 µM concentration.

View larger version (54K): [in this window] [in a new window]   Fig. 3.   Protective effects of MK-801, tACPD, S4C3HPG and tADA against NMDA-induced neuronal death. Primary cultures of cerebellar neurons were used 7 days after seeding. Neurons were treated with 1 mM NMDA after preincubation for 15 min with MK-801 (20 nM, antagonist of NMDA receptors), tADA (1 µM, agonist of mGluR5), tACPD (10 µM, agonist of mGluR1, mGluR2, mGluR3 and mGluR5) or DHPG (20 µM, agonist of mGluR1 and mGluR5). Incubation with glutamate was for 4 h, and cell viability was quantified immediately as indicated under "Methods." Values are the mean ± standard deviations of at least three different plates from different cultures. At least 600 neurons were counted for each point. Neuronal survival was significantly higher (P < .001) for samples treated with MK-801 or mGluRs agonists and NMDA than for samples treated only with NMDA.

These results confirm that activation of mGluR5 (and maybe of mGluR1) prevents glutamate toxicity in primary cultures of cerebellar neurons. It could be therefore considered that activation of mGluR5 by glutamate could reduce glutamate neurotoxicity when the cells are incubated with glutamate for 4 h. We tested whether blocking mGluR5 (and mGluR1) with MCPG (0.1 mM) increases glutamate neurotoxicity. Neuronal survival was 25 ± 3% for neurons treated with glutamate and 31 ± 5% for neurons treated with MCPG and glutamate. A similar effect (slight increase in survival) has been reported for another antagonist of mGluR5 and of mGluR1, AP-3. Survival of neurons treated with AP-3 and glutamate is slightly higher than for those treated only with glutamate (Felipo et al., 1994).

	Discussion

Top Abstract Introduction Methods Results Discussion References

It has been shown that tACPD, an agonist of mGluRs 1, 2, 3 and 5, attenuates glutamate and NMDA neurotoxicity in different systems including cultures of cortical (Koh et al., 1991) or cerebellar (Pizzi et al., 1993; Felipo et al., 1994) neurons, as well as in rat retina in vivo (Siliprandi et al., 1992). It has also been reported that agonists of mGluRs inhibit NMDA receptor function in cerebellar (Courtney and Nicholls, 1992) and neostriatal (Colwell and Levine, 1994) neurons.

The aim of this work was to discern which of the mGluRs mediates the protective effect against NMDA-receptor-mediated glutamate neurotoxicity. As shown in figure 1, tACPD, an agonist at mGluRs 1, 2, 3 and 5, completely prevents glutamate-induced neuronal death, which indicates that the protective effect is mediated by activation of one or more of these receptors. DHPG is an agonist at mGluRs 1 and 5 and also affords complete protection (fig. 1), whereas L-CCG-I, that activates mGluRs 2 and 3, did not prevent glutamate neurotoxicity. This indicates that the protective effect is mediated by mGluRs 1 and/or 5.

To discern which of these receptors mediates the protective effect we tested the effects of tADA, which activates mGluR5 but not mGluR1, and of S4C3HPG, which activates mGluR5 and blocks mGluR1 (table 1). Both compounds completely prevent glutamate-induced neuronal death, which indicates that activation of mGluR5 is able to prevent glutamate neurotoxicity. These results do not allow excluding the possibility that activation of mGluR1 could also prevent glutamate neurotoxicity. To test this possibility we should use an agonist that activates mGluR1 but not mGluR5. As far as we know, such an agonist is not currently available.

It has been shown that AP-3, an antagonist of mGluR1, 2, 3 and 5, prevents the protective effect of tACPD (Felipo et al., 1994). To confirm that the protective effect of tACPD is mediated by activation of mGluR5, we tested whether antagonists of the different mGluRs are able to prevent the protective effect of tACPD. As shown in figure 2, MCPG, an antagonist of mGluR1 and mGluR5, prevents nearly completely the protective effect of tACPD, whereas MTPG, antagonist of mGluR2 and mGluR3, did not prevent at all the protection afforded by tACPD. These results support the idea that the protective effect of tACPD is mediated by activation of mGluR1 and/or mGluR5. Both types of mGluRs are expressed in the primary cultures of cerebellar neurons used (Santi et al., 1994).

The results shown in figure 3 confirm that activation of mGluR5 prevents NMDA-induced neuronal death. These results confirm the protective effects of mGluRs against NMDA receptor-mediated responses reported by different authors (Koh et al., 1991; Siliprandi et al., 1992; Pizzi et al., 1993; Felipo et al., 1994). However, other authors have reported that activation of mGluRs (also by tACPD) potentiates NMDA-induced brain injury and NMDA receptor-mediated responses (McDonald and Schoepp, 1992; Bleackman et al., 1992; Kinney and Slater, 1993; Harvey and Collingridge, 1993). The reasons for these differences are not clear. One possibility is that the different systems used for the assays contain different combinations of subtypes of mGluRs and/or of NMDA receptors. It is possible that the protective mGluRs are present in rat cerebellar neurons but not (or not at sufficient level) in other systems. It is also possible that the mGluRs that potentiate NMDA responses are not present in rat cerebellar neurons. In this regard, it should be noted that, in Xenopus oocytes, mGluR activation enhances NMDA toxicity except when the NR1 subunit is cotransfected with the NR2C subunit (Shen et al., 1995). Interestingly, cerebellar granule cells express the NMDA-R2C subunit, which is nearly absent in other brain regions (Akazawa et al., 1994). For cultured cerebellar granule neurons, it has been shown that NR2B is the predominant form of NR2 at 0 days in vitro. During early development (up to 5 days in vitro) NR2B expression decreases, NR2A is the predominant form and NR2C expression is very low. However, at a later phase of development in vitro, expression of NR2C increases markedly and similar amounts of NR2A and NR2C are found after 10 days in vitro (Resink et al., 1995; Vallano et al., 1996). The results presented here were obtained after 7 days in vitro, so that it should be expected that some NR2C would be present in the neurons used. A protective effect of tACPD against glutamate neurotoxicity in cultured cerebellar granule cells after 11 days in vitro (when NR2C is expressed) has been reported (Felipo et al., 1994).

In summary, the results reported show that activation of mGluR5 by tACPD, DHPG, S4C3HPG or tADA prevents glutamate- and NMDA-induced neurotoxicity in primary cultures of cerebellar neurons. Further studies to clarify the mechanism by which activation of mGluR5 interferes in the process by which activation of NMDA receptors leads to neuronal death could contribute to the understanding of the molecular mechanism of glutamate neurotoxicity.