Freshly isolated embryonic cortical cells obtained from day 17–19 pregnant rats were cultured and maintained in serum-free neurobasal medium containing B27 supplements for 7 days as described under "Materials and Methods". Phase contrast microscopy indicated that the cells had characteristic morphology of neurons and their cellular extensions (dendrites) were clearly visible (Figure 1B). Immunocytochemistry using anti-rat neuronal specific enolase (NSE), and anti-glial fibrillary acidic protein (GFAP) indicates, as can be seen in Figure 1A and Figure 1C, that the cells stained positively only for neuronal enolase (green) while GFAP staining (red) was not detectable. These data further indicate that almost all cells seen by phase-contrast microscopy (Figure 1B) were positive for neuronal enolase (Figure 1A). Thus, the cultures used in our studies are purely neuronal.

Cortical neuronal cells (4 × 10⁴) cultured for 7 days in 96-well plates were treated with increasing concentrations (0.1 to1000 μM) of glutamate for 20 min and cell viability was measured after 24 hrs and compared with vehicle-treated samples. As depicted in Figure 2, 0.1 to 5 μM glutamate resulted in minimal cell death. As the amount of glutamate increased between 5 to 100 μM, the cell death increased gradually in a dose-dependent manner. Thus, treatment with 10 μM glutamate and 50 μM glutamate decreased cell viability by 35.6 ± 3.6% and 54 ± 2.5% respectively. Higher concentrations (>50 μM) of glutamate did not result in any further decrease in cell viability (Figure 2). Thus cell death induced by acute glutamate treatment reached a plateau in these rat primary cortical cells with concentrations of glutamate between 50 and 1000 μM. For all subsequent experiments 200 μM glutamate was used to induce cell death and this concentration resulted in a mean cell death of 48.7 ± 0.7%.

In order to determine the kinetics of glutamate excitotoxicity, rat primary cortical neurons were cultured for 7 days and then treated with 200 μM glutamate for 20 min. Cell death was then assessed at 2, 4, 8, and 24 hrs as described under "Materials and Methods".

The results depicted in Figure 3 indicated that acute glutamate treatment resulted in a time-dependent increase in LDH release compared to vehicle treated cells. Measurement of LDH release at 2 hours resulted in no observable difference in LDH release compared to vehicle treated samples, however after 4 hours, an increase in LDH release of 13 ± 2.7% (p < 0.05) compared to control was observed. At 8 hours LDH release increased by 40 ± 6.8% (p < 0.05) over control. No significant increase in LDH release was observed thereafter (Figure 3). The change in LDH release ratio between glutamate and the control cells at 8 hours (40 ± 6.8%) compared to 24 hrs (42 ± 3.4%) was not statistically significant. Thus, it appears that the majority of cell death in rat primary cortical cells from acute glutamate exposure occurred between 2 to 8 hours post glutamate treatment.

To determine whether the neurotoxic effect of glutamate can be prevented by estrogen treatment, cells were pretreated for 24 hrs with various concentrations, 0.01 to 10 μM, of 17β-E₂ or Δ⁸, 17β-E₂, glutamate treatment was then initiated and cell viability was measured after 24 hrs. An acute (20 minute), 200 μM glutamate exposure resulted in a 50% decrease in cell viability. As depicted in Figure 4, pretreatment with 17β-E₂ or Δ⁸, 17β-E₂ for 24 hrs resulted in a dose-dependent increase in cell viability against glutamate-induced neurotoxicity. No significant protection was observed with 0.01 to 0.05 μM 17β-E₂, however, 0.1 to 5 μM 17β-E₂ resulted in a 8.16 ± 1.93% to 20.94 ± 2.13% increase in protection (p < 0.05) respectively. Similarly, pretreatment with 0.01 to 0.5 μM Δ⁸, 17β-E₂ resulted in an increase in protection from 0.63 ± 1.74% to 6.85 ± 2.09% respectively, however, statistical significance at these concentrations was not reached. In contrast, pretreatment with 1 and 5 μM Δ⁸, 17β-E₂ resulted in a significant (p < 0.05) increase in protection of 9.24 ± 2.23% and 12.12 ± 2.11% respectively (Figure 4). Higher concentration (10 μM) of both estrogens did not result in any further increase in protection. Although both 17β-E₂ and Δ⁸, 17β-E₂ were able to significantly reduce the glutamate-induced cell death complete protection was not achieved with the concentrations tested. However, the results do indicate that 17β-E₂ was more effective in preventing glutamate-induced cell death than Δ⁸, 17β-E₂ (Figure 4) and this was found to be significant with the 0.1 to 10 μM concentrations of estrogen tested (p < 0.05). Pretreatment of cells with estrogen for 15 minutes did not protect the cells against glutamate neurotoxicity, (data not shown).

To determine whether the neurotoxic effect of acute glutamate treatment on rat primary cortical neurons is mediated through the NMDAR or NO, cell viability was determined following inhibition of NMDAR function or inhibition of NOS activity in cells treated acutely with 200 μM glutamate. As can be seen in Figure 5, a 20 minute exposure with 200 μM glutamate in rat primary cortical cells resulted in a 44.43 ± 1.34% decrease in cell viability compared to the vehicle treated control (p < 0.001). A 15 minute pretreatment with 0.1 μM MK-801, a non-specific antagonist of the NMDAR, completely prevented cell death (p < 0.001) compared to glutamate alone treated samples. Similarly, a 15 minute pretreatment with 10 mM NOS inhibitor L-NAME also completely prevented the cell death induced by glutamate (p < 0.001) compared to glutamate alone treated samples. These results clearly indicate that cell death induced by glutamate involves both NMDAR and NO.

To evaluate the effect of glutamate on [Ca²⁺]_i concentrations in cortical neurons various glutamate concentrations, (0.1 to 200 μM), were added to wells and changes in relative fluorescence units (RFU) were recorded. Baseline readings were taken every 15 seconds for a total of 60 seconds prior to an injection (10 μl) of glutamate into each well (indicated by glu Figure 6A). Readings were continued for another 2 minutes post-injection. The kinetics of glutamate induced Ca²⁺ influx are depicted in Figure 6A, and as can be seen, glutamate treatment (Figure 6A) results in a rapid increase in [Ca²⁺]_i with all concentrations of glutamate (0.1 to 200 μM) tested. This initial rapid rise in [Ca²⁺]_i occurs at 15 seconds after glutamate addition and is followed by a reduced rate of Ca²⁺ influx. The kinetics of glutamate-induced Ca²⁺ influx following the initial rise in [Ca²⁺]_i remain relatively stable, but are elevated at glutamate concentration between 0.1 and 1 μM. However, at higher glutamate concentration (5 to 200 μM), [Ca²⁺]_i continues to increase following the initial rise in [Ca²⁺]_i. In addition, glutamate treatment results in a dose dependent increase in the initial rise in [Ca²⁺]_i levels measured at 15 seconds, and reaches significance (p < 0.05) with 0.5 μM to 200 μM glutamate (Figure 6B). Maximal [Ca²⁺]_i levels were observed with 10 μM glutamate and further increase in glutamate concentration did not result in any further increase in [Ca²⁺]_i (Figure 6B).

In order to evaluate the contribution of glutamate-induced Ca²⁺ influx on cell viability the effect of NMDAR antagonist MK-801 on Ca²⁺ influx and cell viability was measured 24 hours later in the same cell population after glutamate treatment. The maximum initial Ca²⁺ influx was reached 15 seconds after glutamate (200 μM) injection and percent of maximum initial Ca²⁺ influx was calculated for each concentration of MK-801 (0.001 to 10 μM). As can be seen in Figure 7, a 15 minute pretreatment with 0.001 to 10 μM MK-801 resulted in a significant dose-dependent reduction in the initial (15 seconds) Ca²⁺ influx by 14.17 ± 2.56% to 40.30 ± 2.28% respectively. Thus, between 0 μM and 10 μM MK-801, the concentration of [Ca²⁺]_i drops from an initial high level of 350 nM (glutamate alone) at 15 seconds to 200 nM (glutamate +10 μM MK-801). This drop in Ca²⁺ influx was associated with a dose-dependent increase in cell viability compared to glutamate alone (Figure 8). MK-801 mediated protection reached significance (p < 0.05) with 0.01 μM MK-801 and 0.1 μM - 10 μM MK-801, completely prevented glutamate-induced cell death (Figure 8).

The effect of estrogen on acute, glutamate (200 μM) induced Ca²⁺ influx was measured at various concentrations (0.01 to 10 μM) of 17β-E₂ and Δ⁸, 17β-E₂. Representative traces for the effect of 5 μM 17β-E₂ and Δ⁸, 17β-E₂ on the glutamate-induced Ca²⁺ influx are presented in Figure 9. Baseline [Ca²⁺]_i measurements were made every 15 seconds for a total of 60 seconds and measurements were continued every 15 seconds following glutamate addition (indicated by glu in Figure 9). As shown in Figure 9, prior to glutamate addition [Ca²⁺]_ilevels were below 20 nM and remained relatively constant. Addition of 200 μM glutamate resulted in an immediate increase in concentration of [Ca²⁺]_i to levels above 300 nM (Figure 9A and 9B). Pretreatment with 5 μM 17β-E₂ or Δ⁸, 17β-E₂ was not associated with any significant decrease in the levels of [Ca²⁺]_i induced by glutamate (Figures 9A and 9B). However, pretreatment with estrogens resulted in a significant (p < 0.05) but modest increase (up to 20%) in cell protection compared to glutamate alone. Similarly, no effect on the initial glutamate induced Ca²⁺ influx was observed at all other concentrations (0.01 to 1 μM) of estrogen. These results indicate that protection against glutamate-induced cell death in rat primary cortical neuronal cells by estrogen does not involve a reduction in the glutamate-induced, NMDAR associated, Ca²⁺ influx, and thus estrogen mediated protection appears to be mediated by a Ca²⁺ independent mechanism.

Day 17–19 pregnant Sprague Dawley rats were anesthetized and sacrificed by cervical dislocation. The fetuses were delivered, decapitated and the brains were dissected out. Each cortex was isolated and individual cells were cultured essentially according to Brewer et al 3334 with the following modifications. Briefly, the tissue was maintained in Hank's balanced salt solution (HBSS) (Invitrogen, Life Technologies, Burlington, ON, Canada) until enough tissue to give sufficient cell numbers for plating was collected. The tissue was then dispersed by mechanical dissociation through a series of constricted fire polished, siliconized Pasteur pipettes to obtain a suspension of single cells. The suspension was then centrifuged and the cells were washed to remove red blood cells and other impurities. An aliquot was stained with 4% trypan blue, live cells were counted using a hemocytometer and the suspension was then diluted accordingly with neurobasal defined medium (Invitrogen, Life Technologies, Burlington, ON, Canada) supplemented with 2% of B27 containing antioxidants (Invitrogen, Life Technologies, Burlington, ON, Canada), 0.5 mM glutamine, penicillin G (10 U/mL) and streptomycin (13 μg/mL) purchased all from Sigma, Canada. This medium was labeled as neurobasal complete (NBC).

Cells, 40,000 per well, were seeded onto poly-D-lysine coated 96 well plates (BD Biosciences, Discovery Labware, Bedford, MA, USA) and were maintained in NBC. Under these culture conditions, it has been previously demonstrated (16,23,33,34) that only neurons survive while glia are essentially absent. The purity of our cortical neuronal cultures is described separately. After day 1 of plating the media was completely replaced and on subsequent days 50% of the media was replaced. Cell cultures were maintained for 7–12 days at which time glutamate experiments were initiated. Preliminary experiments carried out during the standardization of the methods indicated that the cell viability decreased and was variable after day 7 in culture. Although some neurons were viable up to day 12, all experiments described in the manuscript were carried out in cultures maintained for only 7 days.

Cortical cells on poly-D-lysine coated plates were cultured in NBC medium as described above. After seven days, cells were washed twice with PBS and then fixed with 4% formaldehyde in PBS for 15 minutes. Cells were then washed three times with PBS and non-specific binding was blocked for 5 mins with PBS containing 1% BSA (Sigma, Canada), 5% normal goat serum (Sigma, Canada), 0.05% Triton-X100 (BDH, Canada). Cells were incubated for 2 hrs at room temperature with rabbit anti-rat neuron specific enolase (anti NSE; Polyscience, PA, USA) 1:1000 dilution in block solution. After rinsing 4 times with PBS, the cells were incubated at room temperature with Alexa Fluor 488 F(ab')₂ fragment of goat anti-rabbit IgG (H+L) secondary antibody (Molecular Probes, Invitrogen, Canada) 1:400 dilution in block solution. Cells were then rinsed 4 times with PBS and the fluorescence labeling was visualized and photographed using Olympus IMT-2 microscope, equipped with a mercury light source and spot slider digital camera (Diagnosis Instruments, MI, USA). The neuronal cells displayed characteristic green fluorescence (Figure 10A).

The neuronal cells maintained in NBC are devoid of glial cells and to obtain cultures containing glia, the cortical cells were prepared as described above but were maintained in Dubelcco's Modified Eagle Medium (DMEM, Sigma Canada) containing 10% Fetal Bovine Serum (FBS; Invitrogen Canada Inc; On, Canada). After 7 days in culture, the mixture of cells were processed for immunocytology by incubating with the primary antibody mouse anti-GFAP (1:4000 dilution; Sigma Canada) and then with Texas Red dye conjugated AffiniPure F (ab')₂ fragment goat anti-mouse IgG (H+L) (1:100 dilution; Jackson Immunoresearch Laboratories, PA, USA) as described above for NSE, (data not shown). The GFAP positive glial cells stain red as seen in Figure 10B. For double staining, cells maintained in NBC medium were first incubated with primary and secondary antibodies for GFAP and after rinsing four times with PBS, the cells were incubated with antibodies for NSE. Cells were then visualized and photographed. The dual staining indicated that the cells were only positive for neuronal NSE and negative for glial GFAP (Figure 1A and 1C).

For determination of the effect of glutamate on the viability of rat primary cortical cells, the cells were washed twice with a modified Locke's buffer (154 mM NaCl, 5.6 mM KCl, 3.6 mM NaHCO₃, 2.3 mM CaCl₂, 5.6 mM Glucose, 5 mM HEPES, 10 μM Glycine, pH 7.4) and incubated for 24 hours at 37°C and 5% CO₂ in NBC medium, plus 0.5 mM glutamine and 2% B27 minus antioxidants (NBC AO-). Following this, the cells were washed 3 times with modified Locke's buffer and glutamate exposure was initiated. All glutamate treatments were conducted in modified Locke's buffer. Various concentrations of glutamate were added to a final concentration of 0.1, 0.5, 1, 5, 10, 50, 100, 200, 500 and 1000 μM for a total of 20 minutes and changes in [Ca²⁺]_i were recorded. The cells were then washed with modified Locke's buffer and NBC AO- was added and cells were incubated at 37°C, 5% CO₂ for 24 hours, following which cell death/viability was measured. Measurement for Ca²⁺ and cell death/viability were performed on the same culture of cells.

The kinetics of glutamate-induced cell death following a 20 minute exposure to glutamate was also measured using a concentration of glutamate found to be toxic to 50% of the cells. Cell death measurements were made at 2, 4, 8 and 24 hours post glutamate treatment and comparisons between glutamate treated samples and vehicle (modified Locke's buffer) treated samples were made.

For the determination of the effects of estrogen on glutamate induced changes, cells were washed twice with modified Locke's buffer and then fresh NBC AO- and estrogens were added to final concentrations of 0.01, 0.05, 0.1, 0.5, 1 and 10 μM and the cells were incubated for 24 hours at 37°C and 5% CO₂. Estrogens were dissolved in ethanol and diluted to working concentrations with NBC AO- medium. The final concentration of ethanol was 0.2% for control and estrogen treated samples. Following incubation with estrogen and prior to glutamate exposure cells were prepared for [Ca²⁺]_i measurements as outlined below. The cells were washed and incubated in Locke's buffer containing sufficient (200 μM) glutamate (pre-determined) to give 50% cell death. Incubations were carried out at 37°C for 20 minutes and changes in [Ca²⁺]_i were recorded. Following glutamate treatment cells were washed with fresh Locke's buffer and NBC AO- was added. The cells were again incubated for 24 hours at 37°C and 5% CO₂, following which cell viability was determined.

In some experiments various concentrations of MK-801, an NMDAR antagonist, and 10 mM of L-NAME 53, a NOS inhibitor, were added 15 minutes prior to glutamate exposure following which changes in [Ca²⁺]_i and cell viability were measured.

The Lactate dehydrogenase (LDH) cytotoxicity assay (Promega G1780, VWR, Toronto, Ontario, Canada) and the CellTiter 96^® AQ_ueous Non-Radioactive cell proliferation assay (MTS method) (Promega G5430) were used to assess cell death and viability respectively. The LDH assay measures the amount of lactate dehydrogenase that is released from dead cells while the MTS method measures the formation of a formazan product in live cells. Changes in optical density (O.D.) at 490 nm were measured using the SpectraMax 340 microplate reader and SOFTmax^® PRO software (Molecular Devices, California, U.S.A.) and comparisons of treated to control values were made. Both of these methods provide a reliable measure of cell death/viability.

Intracellular Calcium levels were assessed using the fluorescent probe Fluo-3 AM (Molecular Probes, Eugene, OR, USA). Changes in relative fluorescence units (RFU) were monitored with the Fluoroskan Ascent FL fluorescent plate reader equipped with a micro-injection syringe pump (Labsystems, Helsinki, Finland). The protocol for intracellular dye loading was based on the method provided by the manufacturer (Molecular Probes and Labsystems Fluo-3 AM Application Note) with minor modifications. Briefly, cells were loaded with 5 μM Fluo-3 AM for 2 hours and then washed three times with Locke's buffer to remove excess dye. Cell plates were then placed in the Fluoroskan and incubated for an additional 15 minutes at 37°C to allow for complete de-esterification of dye. Measurements of relative fluorescence units (RFU) were made using an excitation and emission wavelength of 485 and 527 nm respectively. Change in [Ca²⁺]_i (nM) was calculated by multiplying the ratio of ΔF-Fmin and Fmax-ΔF by Kd. Where ΔF is the change in RFU, Fmax is the maximum observed RFU following NP40 treatment, Fmin is the minimum observed RFU following EGTA Ca²⁺ chelation and Kd is the dissociation constant for Fluo-3 AM, 390 nM, provided by supplier (Molecular Probes).

The dose-response curves for effects of glutamate and estrogen on cell viability and [Ca²⁺]_i were developed using Slide Write Plus V6 Software (Advanced Graphics Software, Inc., Encinitas, CA) using the non-linear curve fitting functions. Each estrogen experiment was carried out a minimum of six times with eight replicates per treatment condition and all values shown in figures are the mean ± standard error of the mean (SEM). All treatment conditions were compared to the respective controls and an effect was determined significant if p < 0.05. All statistical analyses were performed using Instat^® Software (Graph Pad, San Diego, CA). To establish the effect of estrogen against glutamate induced cell death and [Ca²⁺]_i, nonparametric statistical analyses were performed using the Kruskal-Wallis nonparametric analysis of variance test with the Dunn post hoc test to compare each estrogen concentration against the glutamate alone control.