Potentiation of Methylmercury-Induced Death in Rat Cerebellar Granular Neurons Occurs by Further Decrease of Total Intracellular GSH with BDNF via TrkB in Vitro

Abstract

Brain-derived neurotrophic factor (BDNF) is a principal factor for neurogenesis, neurodevelopment and neural survival through a BDNF receptor, tropomyosin-related kinase (Trk) B, while BDNF can also cause a decrease in the intracellular glutathione (GSH) level. We investigated the exacerbation of methylmercury-induced death of rat cerebellar granular neurons (CGNs) by BDNF in vitro. Since methylmercury can decrease intracellular GSH levels, we hypothesized that a further decrease of the intracellular GSH level is involved in the process of the exacerbation of neuronal cell death. In the present study, we established that in CGN culture, a decrease of the intracellular GSH level was further potentiated with BDNF in the process of the methylmercury-induced neuronal death and also in GSH reducer-induced neuronal death. BDNF treatment promoted the decrease in GSH levels induced by methylmercury and also by L-buthionine sulfoximine (BSO) and diethyl maleate (DEM). The promoting effect of BDNF was observed in a TrkB-vector transformant of the rat neuroblastoma B35 cell line but not in the mock-vector transformant. These results indicate that the exacerbating effect of BDNF on methylmercury-induced neuronal death in cultures of CGNs includes a further decrease of intracellular GSH levels, for which TrkB is essential.

Intracellular glutathione (GSH) is a small peptide and one of the principal substances to detoxify reactive oxygen spices (ROS) to maintain the intracellular redox balance.^1,2) Both GSH depletion and the formation of ROS causes an insufficiency of the intracellular GSH level to detoxify ROS and results in cell degeneration due to oxidative stress.^3,4)

Brain-derived neurotrophic factor (BDNF) is a critical factor for neurogenesis, neurodevelopment and especially neuron survival. BDNF exerts its biological effects through binding to tropomyosin-related kinase (Trk) B, a neurotrophin receptor with high affinity, after which TrkB processes signal transduction, primarily through three signaling pathways, the mitogen-activated protein kinase (MAPK), the phospholipase C (PLC) γ, and the phosphoinositide 3-kinase (PI3K) pathways.^5,6) BDNF also acts as a potentiate factor to exacerbate the cell death in in vitro models of neuronal degradation and glutamate- and low-potassium-induced neuronal death.^7,8)

Methylmercury is a well-known environmental neurotoxin affecting individuals from embryos to adults. We have shown that methylmercury-induced and glutamate-induced neuronal deaths are exacerbated by BDNF binding to TrkB.^9,10) During the process of methylmercury- and glutamate-induced cell deaths, a decrease of the intracellular GSH level and increase of ROS production are generally observed, and these changes are known to play a key role in the toxicological process.^11–18) Therefore, we hypothesized that BDNF potentiates the methylmercury-induced death of neuronal cells by decreasing the intracellular GSH level via TrkB. To elucidate our hypothesis, we investigated the intracellular GSH levels in cultures of cerebellar granular cells from rat pups after treatment by methylmercury and/or BDNF. Furthermore, to clarify whether BDNF exacerbates GSH depletion-induced cell death, we investigated whether BDNF stimulates the cell death that is induced by employing two GSH reducers, L-buthionine sulfoximine (BSO) and diethyl maleate (DEM) in primary cultured neurons, and whether further depletion of intracellular GSH level by BDNF is shown in the cells before death.

MATERIALS AND METHODS

Cell Cultures and Treatments

Cerebellar granular neurons (CGNs) for primary culture were prepared from rat pups (Jcl:Wistar; Clea Co., Tokyo, Japan) within the day of birth, as described previously.^9,17,19,20) Cerebella from the pups after meninges were trimmed in alpha modified Eagle’s minimal essential medium (αMEM), and they were placed in Hank’s balanced salt solution (HBSS, Life Technologies Japan, Tokyo, Japan) containing 1% trypsin (Life Technologies) and 0.05% DNase I (Roche Diagnostics K.K., Tokyo, Japan), and incubated for 13 min at room temperature. The cerebella were replaced in HBSS containing 0.05% DNase I and 12 mM MgSO₄ after washing with HBSS and mildly triturated with a Pasteur pipette to get dispersed neurons. The neurons after rinsing in HBSS were suspended in αMEM containing 1 mg/mL bovine serum albumin, 10 µg/mL bovine insulin, 0.1 nM thyroxine, 0.1 mg/mL human transferrin, 1 µg/mL aprotinin, 30 nM Na₂SeO₃, 0.25% glucose, 100 units/mL penicillin and 135 µg/mL streptomycin, and seeded on poly-L-lysine-coated dishes. These supplements for cell culture were purchased from Sigma (St. Louis, MO, U.S.A.). After pre-incubation for 2 d, the cultures were treated with various reagents.

Rat neuroblastoma B35 cells were purchased from the American Type Culture Collection (ATC C, Manassas, VA, U.S.A.). The B35 cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum (HyClone, South Logan, UT, U.S.A.), 100 U/mL penicillin and 135 µg/mL streptomycin. The cells were passaged or fed with fresh medium every second day. B35 cells stably overexpressing mRNA of TrkB (B35^TrkB) were selected after transfection with a TrkB expression vector as utilized in our previous reports,^9,10) to determine the involvement of TrkB in the accelerating effect of BDNF on methylmercury cytotoxicity. B35 cells transfected with a mock vector were used as the negative control in that experiment, B35^Mock cells that do not express TrkB even endogenously.

The CGNs and B35 cells were pre-incubated for 2 d and 24 h before treatment, respectively. BDNF (Peprotech, Rochy Hill, NJ, U.S.A.), U0126 (Millipore, Billerica, MA, U.S.A.) and PD98059 (PD) (Millipore) were added to the medium 30 min before the treatment with methylmercuric chloride (MeHg) (Tokyo Kasei Kogyo Co., Ltd., Tokyo, Japan), BSO (Sigma) or DEM (Sigma). To investigate whether the MAPK signaling cascade was included in BDNF effects, U0126 and PD were added to the medium 30 min before BDNF treatment.

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium Bromide (MTT) Assay

To estimate cell viability in cultures, a MTT (Sigma) assay was used, as described in our previous report.²¹⁾ MTT solution was added to culture medium and incubated for 2 h. The medium was completely exchanged with cell lysis buffer (5.32 M N,N-dimethylformamide, 10% sodium dodecyl sulfate, 2% acetic acid and 4.5 mM HCl) and incubated overnight to dissolve intracellular formazan. The optical absorption of the cell lysate was measured at 570 nm with a Bio-Rad microplate reader, model 550 (Bio-Rad Laboratories, Hercules, CA, U.S.A.). The absorbance was divided by that of control cells, and the calculated mean±standard deviation (S.D.) are shown.

Total Intracellular GSH Assay

Total intracellular GSH levels of CGNs and B35 cells were determined according to a previously reported method.¹⁷⁾ CGNs and B35 cells were collected and rinsed with phosphate-buffered saline (PBS, pH 7.4). The cells were frozen and thawed twice to burst the cells in 10 mM HCl, and centrifuged at 15000 rpm for 10 min. Protein concentration in the supernatant was determined using a Bio-Rad Protein Assay Kit for standardization of GSH levels. The supernatant was mixed with 5-sulfosalicylic acid (5% final concentration), and centrifuged to remove proteins. The supernatant was used as a sample to determine total GSH levels. Total levels of GSH in each sample were measured by the 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB)-GSH reductase recycling method as follows: reaction solution was made of the adequately diluted supernatant, 6.25 mM phosphate buffer (pH 7.5), 0.31 mM ethylenediaminetetraacetic acid, 0.25 mM reduced nicotinamide adenine dinucleotide phosphate, 0.75 U/mL GSH reductase and 0.625 mM DTNB as final concentrations. The optical absorbance was determined at 415 nm with the microplate reader after incubation for 20 min at room temperature. The GSH content in the sample was quantified by comparing the absorption with a standard curve constructed with reference to the known level of GSH and standardized by protein concentration.

Statistical Analysis

Statistical analysis was performed using unpaired Student’s t-test or two-way ANOVA with Fisher’s protected least significant difference (PLSD) test as a post hoc test. The differences were considered to be statistically significant at the level of p<0.05.

RESULTS

Alteration of Intracellular GSH Level in CGNs by BDNF Treatment

After 48-h incubation with methylmercury, the viability of CGNs was decreased to 54.8±3.5% compared to the control group, which was again significantly decreased to 37.3±1.1% by BDNF treatment at 100 ng/mL, while BDNF treatment alone increased CGN cell viability to 122.8±8.2% (Fig. 1A) as we previously reported.⁹⁾ While, there were no statistically significant difference among groups in cell viability, even at 24-h incubation (Fig. S1). Therefore, to investigate the effect of BDNF on intracellular GSH levels, cells treated with methylmercury and/or BDNF were harvested after 24 h of incubation and the GSH level was determined. The GSH level of each group, BDNF-treated, methylmercury-treated and BDNF-plus-methylmercury-treated, was significantly declined compared to that of the control group (Fig. 1B). The values of BDNF-treated and methylmercury-treated groups were decreased to 80.2±7.6 and 57.4±4.4% of the control group. The treatment combination of BDNF+methylmercury resulted in significantly further decrease of the GSH level in CGNs, to 37.8±1.3%. The GSH level time-dependently decreased from 12 to 24 h of incubation time in both methylmercury-treated and BDNF-plus-methylmercury-treated groups (Fig. 1C), with the decrease of the BDNF-plus-methylmercury-treated group being significantly greater.

Fig. 1. BDNF Exacerbates the Cell Death and Intracellular GSH Decrease in Methylmercury-Treated Cultures from Rat Cerebella

The CGNs were treated with methylmercury (MeHg) and/or BDNF at 100 nM and 100 ng/mL, respectively. (A) The effect of BDNF on cell viability after incubation in MeHg-containing medium for 48 h. The viability was estimated by MTT assay compared to the control. (B) The effect of BDNF on intracellular total GSH level after 24-h incubation with MeHg. (C) Time–course-dependent changes of intracellular total GSH level after MeHg treatment. Plaid and filled columns indicate MeHg-treatment and BDNF-plus-MeHg-treatment groups, respectively. Note that BDNF treatment further lowered the MeHg-induced decrease of cell viability and GSH level. Asterisks indicate a statistical difference between the indicated groups (p<0.05). Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test. Data represent the mean±S.D. from three independent wells.

Effect of BDNF on GSH-Reducer-Induced Cell Death and Intracellular GSH Level

In order to examine whether methylmercury-induced lowering of the GSH level was involved in the exacerbation of methylmercury-induced cell death by BDNF, two reagents—BSO, which is an inhibitor of a rate-determining enzyme for GSH synthesis, gamma-glutamylcysteine synthetase,²²⁾ and DEM, which irreversibly conjugates glutathione²³⁾— were utilized to induce intracellular GSH depletion in CGN cultures and thereby to induce cell death; then, the effect of BDNF was investigated in this setting. BSO treatment significantly decreased the cell viability of CGNs after 72-h incubation in a manner dependent on BSO concentration (open column, Fig. 2A). The viabilities in BSO-treated and BDNF-plus-BSO-treated groups were respectively 99.2±3.7 and 84.1±4.3% at 100 µM BSO, and then 77.8±1.9 and 50.3±1.9% at 300 µM BSO, indicating a statistically significant difference between cell viabilities of the BSO-treated and BDNF-plus-BSO-treated groups at each BSO concentration (Fig. 2A). BSO treatment also time-dependently decreased the total GSH level of CGNs. After 24- and 36-h incubation in BSO-containing medium at 300 µM, the intracellular GSH level in the BDNF-plus-BSO-treated group was significantly declined compared with that in the BSO-treated group. The GSH values in BSO-treated and BDNF-plus-BSO-treated groups were 100±7.3 and 54.5±8.7% after 24-h incubation, and 70.5±9.8 and 37.3±0.8% after 36-h incubation, as relative values to the GSH level of the BSO-treated group after 24-h incubation (Fig. 2B). The BSO treatment at 100 and 300 µM had no effect on cell viability even after 48-h incubation (Fig. S2). DEM treatment reduced the CGN viability after 24-h incubation in a concentration-dependent manner (Fig. 2C, open column). The relative value of the cell viability in DEM-treated and BDNF-plus-DEM-treated groups were 99.1±6.8 and 117.5±4.9% at 100 µM of DEM and 73.7±1.4 and 34.4±1.9% at 200 µM of DEM, respectively, indicating statistically significant differences between the viabilities in the DEM-treated and BDNF-plus-DEM-treated groups at each DEM concentration (Fig. 2C). The intracellular decrease in the GSH level of CGNs by DEM treatment was also time dependent (Fig. 2D). The GSH level of the BDNF-plus-DEM-treated group was significantly decreased compared with that of DEM-treated group after 6-h or 12-h incubation, with the relative values of DEM-treated and BDNF-plus-DEM-treated groups being100±11.0 and 57.0±6.7% at 6-h incubation time, and 62.4±10.3 and 30.7±2.6% at 12-h incubation time, respectively (Fig. 2D). This result indicated that BDNF caused the BSO- and DEM-induced decrease of cell viability and GSH level in the cultures of CGNs to decrease significantly further; this was an additive decline.

Fig. 2. BDNF Exacerbates the Cell Death and Intracellular GSH Decrease in BSO-Treated Primary Cultures from Rat Cerebella

The CGNs were treated with BDNF at 100 ng/mL 30 min before BSO- or DEM-treatment. Viability of cells was estimated by MTT assay. (A) Exacerbation of BSO-induced cell death by BDNF. The cell viability was estimated in comparison to the control (no BDNF, 0 µM BSO; open column) after 72-h incubation with BSO. Filled and open columns show groups treated with and without BDNF, respectively. (B) Time–course-dependent changes of intracellular total GSH level after BSO treatment at 300 µM. Open and filled columns indicate the groups treated with BSO and with BSO-plus-BDNF, respectively. (C) Exacerbation of DEM-induced cell death by BDNF. The viability of cells incubated for 24 h in DEM-containing medium was also estimated compared to the group treated with nether DEM or BDNF. Filled and open columns indicate groups treated with and without BDNF, respectively. (D) Time–course-dependent decrease of intracellular total GSH level after DEM treatment. The intracellular total GSH level was estimated at 6 h and 12 h. Open and filled columns show the DEM and DEM+BDNF groups, respectively. Asterisks indicate a statistical difference between the indicated groups (p<0.05). Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test. N.S., no significance. Data represent the mean±S.D. from three independent wells.

Involvement of TrkB in the Decrease of Intracellular GSH Level and Cell Viability through BDNF Treatment

Our previous study found that TrkB is involved in the mechanisms by which BDNF exacerbates methylmercury-induced neuronal death⁹⁾ using a TrkB stable transformant of B35 cells, B35^TrkB cells. In the present study, we used the B35^TrkB cells to investigate whether the further decline by BDNF in GSH-reducer and methylmercury-induced GSH lowering were through TrkB. The total level of intracellular GSH in B35^TrkB cells was significantly decreased by methylmercury treatment, of which the relative value to the control level was 74.3±9.1% after 12-h incubation with methylmercury. The BDNF treatment further reduced the methylmecury-induced decrease of total GSH level in B35^TrkB cells into 52.6±3.9% (Fig. 3A), while there was no significant difference between methylmercury- and methylmercury-plus-BDNF treatment cells in B35^Mock (Fig. 3B). Incubation with methylmercury at 600 nM did not affect the viability of B35^TrkB cells at 24 and 36 h (Fig. S3). Furthermore, we investigated whether BDNF has any effect on GSH reducer-induced cell death through TrkB in B35 cells. The cell viabilities of B35^Mock and B35^TrkB cells were decreased to 89.2±0.5 and 89.3±6.4% after 24-h incubation with BSO, respectively, with no statistically significant difference between B35^Mock and B35^TrkB cells. BDNF treatment further stimulated the BSO-induced decrease of the cell viability in B35^TrkB cells to 54.3±7.6 and 46.7±10.6% at 10 and 100 ng/mL BDNF, respectively. The B35^TrkB cell viability was significantly less than that of B35^Mock cells as treated with BDNF-plus-BSO, while the B35^Mock cell viability was not altered by BDNF treatment (Fig. 4A). In addition, the intracellular GSH level was determined after 12-h incubation. The GSH level was significantly decreased in B35^TrkB cells by BSO treatment, and the relative value of the level was 43.1±2.8% of negative control group. The BSO-induced decrease of GSH level in B35^TrkB cells was additionally exacerbated to 20.5±2.6% by BDNF treatment (Fig. 4B), but any treatments did not show a statistically significant reduction in cell viability compared to the control group (Fig. S4). Furthermore, BDNF treatment had no significant effect on intracellular GSH level in B35^Mock cells. The relative values of BSO and BSO-plus-BDNF treated groups were 37.6±3.6 and 36.3±2.0% compared to the control group, respectively, with no statistical difference between the treatment groups (Fig. 4C).

Fig. 3. Essential Role of TrkB in Acceleration of Methylmercury-Induced Lowering of Intracellular Total GSH Level in Rat B35 Cells

Intracellular total GSH level was assayed in B35 cells transfected with TrkB expression vector (B35^TrkB, A) or a mock vector (B35^Mock, B) after 12 h-incubation with methylmercury (MeHg) at 600 nM and/or BDNF treatment at 100 ng/mL. Data represent the mean±S.D. from three independent wells. Asterisks indicate statistically significant differences. Statistical analysis was performed using the unpaired Student’s t-test. N.S., no significance.

Fig. 4. Essential Role of TrkB in BDNF’s Acceleration of BSO-Induced Lowering of Cell Viability and Intracellular Total GSH Level in Rat B35 Cells

The cells were treated with BDNF 30 min before incubation with 50 µM of BSO. (A) Effect of BDNF on BSO-induced alteration of B35 cells with TrkB overexpression (filled column) or not (open column). The viability was determined after BSO and/or BDNF incubation for 24 h. The viability was estimated by MTT assay compared to the group untreated with BDNF or BSO. (B, C) Alteration of intracellular total GSH level via TrkB in B35 cells with TrkB overexpression (B, B35^TrkB) or not (C, B35^Mock). The samples for total GSH level were collected after 12-h incubation with BSO. Note that the further decrease of cell viability and intracellular total GSH level in the BSO-plus-BDNF-treated group compared to BSO treated group was observed in B35^TrkB (B), but not in B35^Mock (C). Data represent the mean (% vs. control of each cell lines)±S.D. from three independent wells. Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test (A) or the unpaired Student’s t-test (B, C). Asterisks and N.S. indicate statistical differences (p<0.05) and no significance, respectively.

Effect of BDNF on Methylmercury-Induced Decrease of Intracellular GSH Level

We previously reported that the kinase inhibitor U0126 suppressed the stimulating effect of BDNF on methylmercury-induced cell death,⁹⁾ leading to the hypothesis that intracellular GSH lowering was a key action of BDNF in this setting and was contributed to by a factor inhibited by U0126. Therefore, firstly, we sought to determine the modulating effect of U0126 on BDNF’s effect on the intracellular GSH level in cultures of rat CGNs (Fig. 5). The cell viability in the methylmercury-treated group decreased to 59.9±4.4% of the control group cell viability, and BDNF further lowered the level to 55% of the control group cell viability. The reductions of the cell viabilities in BDNF-plus-methylmercury-treated groups were significantly recovered to 104.9±4.1% by U0126 treatment, of which the relative value was statistically different from that of control group. Unexpectedly, U0126 treatment even recovered the cell viability reduction in the methylmercury-treatment group to 82.7±4.9% of the control group’s viability (Fig. 5A).

Fig. 5. Suppressive Effect of U0126 on BDNF-Induced Acceleration of Cell Viability and Intracellular Total Glutathione Level in Methylmercury-Treated CGNs

Rat CGNs were treated with methylmercury (MeHg) at 100 nM, U0126 at 1 µM, PD 98059 (PD) at 30 µM and/or BDNF at 100 ng/mL. U0126 or PD were added into the culture 30 min before BDNF or MeHg treatment. BDNF was added to the cell cultures 30 min before MeHg treatment. (A) U0126 inhibited the accelerative effect of BDNF on MeHg-induced cell death after 48-h incubation with MeHg. The viability incubated in MeHg-containing medium for 48 h was estimated by MTT assay compared to control. (B) Effect of U0126 on BDNF-induced decrease of intracellular total GSH level after 24-h incubation with MeHg. Data represent the mean±S.D. from three independent wells. Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test. Columns with different lowercase letters are significantly different (p<0.05) from one another.

The intracellular GSH level of rat CGNs after 24-h incubation in methylmercury-containing medium were determined (Fig. 5B). The GSH level was significantly lowered by BDNF and methylmercury treatment to 74.3±4.5 and 59.9±4.4% of the control value; then this decrease was further lowered to 47.21±7.1% of the control value with the combination of BDNF and methylmercury treatments. The reduction of the GSH level was restored to 74.3±4.4% of the control by U0126 treatment. To determine which mechanism U0126 inhibited, the BDNF or methylmercury effect, in causing the restoration of GSH, the effect of U0126 on BDNF-induced or methylmercury-induced decrease of GSH level was investigated in CGNs. U0126 treatment suppressed the GSH level decrease through methylmercury treatment (Figs. S5(B)) but not through BDNF treatment (Fig. S5(C)). Meanwhile, PD, an MEK inhibitor like U0126, did not show an inhibitory effect on methylmercury-induced cell death and lowering of intracellular GSH level in CGNs, or on BDNF-induced GSH decrease (Figs. S5(A–C)).

DISCUSSION

Methylmercury induces the decrease of the intracellular GSH level in the process of cell death. In the present study, we revealed two clues about the BDNF-exacerbating effect on methylmercury-induced neuronal death. First, methylmercury decreased the intracellular GSH level and BDNF accelerated the GSH lowering in a culture of rat CGNs. BDNF also further exacerbated the GSH reducer-induced cell death and decrease of intracellular GSH level. BDNF has the reducing effect on the intracellular GSH level in cultures of rat auditory neurons,²⁴⁾ and we showed that BDNF treatment lowered the intracellular GSH level in CGNs and B35 cells even without any reduction of cell viability in the present study. These results indicate that BDNF reduces the intracellular GSH level, and further stimulates the methylmercury-induced decrease of the intracellular GSH level. Second, the further lowering effects of BDNF on the intracellular GSH level and exacerbating the cell viability reduction were observed in B35^TrkB cells treated with methylmercury or GSH reducer but not in B35^Mock cells. Again, we had known that TrkB was involved in dark side effect of BDNF.⁹⁾ These results indicate that TrkB expression is necessary for the BDNF effects in B35 cells, and also suggest possibility that TrkB is involved in BDNF effects even in CGNs.

BDNF was shown to induce cell death only when methylmercury and GSH reducers created cytotoxic conditions for the CGNs and B35 cells. The methylmercury-induced cell death was inhibited with U0126 treatment. Actually, U0126 treatment also inhibited the methylmercury-induced lowering of the intracellular GSH level, but did not inhibit the BDNF-induced lowering of the GSH level (Figs. S5(B, C)). Previously, treatment with PD, an MEK inhibitor like U0126,^25,26) did not inhibit the methylmercury-induced lowering of the GSH level. Paradoxically, BDNF behaves as an exacerbator and neurotrophic factor at concentrations of methylmercury that show neurotoxicity and at those that do not, respectively.⁹⁾ U0126 can inhibit the kinase activation of both MEK1 and MEK2 with IC₅₀ values of 0.07 and 0.06 µM, respectively,²⁵⁾ while PD inhibits MEK1 more potently than MEK2, at 2–7 µM for MEK1 and 50 µM for MEK2, and shows higher IC₅₀ than U0126.²⁶⁾ Therefore, the difference of the inhibitory effect on MEK2 may contribute to the U0126 effect; otherwise, the methylmercury-induced neuronal death may involve a pathway of the intracellular kinase cascade that is an unknown target of U0126 except for ERK1/2-including pathways. We have no data in the present study to explain the reason for the U0126 effect. The suppressive effect of U0126 on methylmercury-induced cell death may be able to explain the inhibitory effect of U0126 on the exacerbating effect of BDNF. BDNF does not show an exacerbating effect at non-cytotoxic concentrations of methylmercury and GSH reducers. Thus, as U0126 completely inhibits methylmercury-induced cell death, the suppression of methylmercury-induced cell death by U0126 should make the culture condition mimic that of non-cytotoxic concentration of methylmercury, and result in the suppression of the BDNF effect even when the cultures were treated with methylmercury at cytotoxic concentration.

In the present study, the treatment of GSH level reducers, BSO and DEM, induced neuronal cell death in the CGN culture, and BDNF further reduced the intracellular GSH level and additionally promoted cell death. Treatment with BDNF alone reduced the intracellular GSH level in the present study. Actually, BDNF treatment increases intracellular production of reactive oxygen species in neuron cultures of rat cerebral cortex cells,²⁷⁾ and removal of BDNF from culture medium results in the increase of intracellular GSH levels in cultured auditory neurons.²⁴⁾ The intracellular GSH level in neuron culture is decreased by exposure to methylmercury.^14,15,19) This lowering of the intracellular GSH level is one of the most important factors for methylmercury-induced cell death in neuron cultures, as methylmercury induces production of reactive oxygen species, and the methylmercury-induced cell death is inhibited by treatment with antioxidative enzymes and antioxidants such as GSH, Trolox and vitamin E.^{12,13,16,28,29)} The present results show that BDNF alone induced a slight lowering of the intracellular GSH level and further increased the methylmercury-induced lowering of the intracellular GSH level. Actually, BDNF treatment additionally lowered the intracellular GSH level and promoted cell death in CGN and B35^TrkB-cell cultures treated with GSH reducers, BSO or DEM, but not in B35^Mock. These results indicate that BDNF acts as an exacerbator of GSH reducer-induced cell death through further GSH lowering via TrkB, which strongly suggests the additional promotion by BDNF should stimulate the increase of methylmercury-induced cell death via TrkB. On the other hand, treatment with BDNF alone had no action as an inducer of cell death, as BDNF did not show any affect on cell viability except on the intracellular GSH level in CGNs and B35^TrkB cells.

Decreases of the intracellular GSH level are detected in neuronal death models such as N-methyl-D-aspartate (NMDA) receptor-mediated, beta-amyloid-induced and low potassium-induced neuronal death.^30–32) Among the neuronal death models, NMDA receptor-mediated neuronal death is exacerbated by treatment with BDNF.⁷⁾ These reports support the present study’s results of the BDNF effect on cell deaths with GSH decreasing. On the other hand, beta-amyloid-induced and low potassium-induced neuronal deaths are suppressed by BDNF.^32,33) BDNF has been well-known to be a survival factor against obvious apoptosis of neurons, while also to be a stimulator of neuron necrosis,^7,27) which may explain the difference of the BDNF effect for the models of cell death, depending on the type of cell death. Furthermore, the type of methylmercury-induced cell death, apoptosis or necrosis, depends on the exposure concentration and length of methylmercury exposure in vitro³⁴⁾; at the concentrations used in this study, methylmercury should induce apoptosis. However, some factor of necrotic cell death may be included in the mechanism of the cell death even at concentrations of methylmercury lower than the µM-order concentration, i.e., in the culture conditions of the present study, because BDNF exacerbated methylmercury-induced cell death in the present study. Anyway, the determining factor as to which effect of BDNF occurs in the neuron as a survival factor or a death stimulator must not be the receptor TrkB, as BDNF needs to bind this receptor to exert either of its effects.

In conclusion, the results of the present in vitro study demonstrate that the exacerbation of methylmercury-induced death of rat CGNs by BDNF is due to a further decrease in the intracellular GSH level with BDNF during the process of neuronal death, for which TrkB is essential.

Acknowledgments

This research was supported by a Grant (No. 22780271, to M.S.) from the Ministry of Education, Culture, Sports, Science and Technology of Japan, by a research project Grant awarded by the Azabu University Research Services Division (to M.S.), and by the Promotion and Mutual Aid Corporation for Private Schools of Japan, Grant-in-Aid for Matching Fund Subsidy for Private Universities (to M.S.).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

Fig. S1. Cell Viability in Primary Cultures of Rat Cerebellar Granular Neurons after Methylmercury (MeHg) and/or BDNF for 24 h. The cells were incubated in the medium containing MeHg at 100 nM with or without BDNF at 100 ng/mL for 24 h. The viability was estimated by MTT assay compared to control. Data are shown as the mean±S.D. from four independent wells. Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test. MeHg, methylmercury; BDNF, brain-derived neurotrophic factor; N.S., no significance compared to the group without both MeHg and BDNF.

Fig. S2. Cell Viability in Primary Cultures of Rat Cerebellar Granular Neurons after BSO and/or BDNF Treatments for 48 h. The viability was estimated by MTT assay compared to control. The cells were treated with BSO at the concentrations indicated in the figure and BDNF at 100 ng/mL. BDNF was added to the medium 30 min before BSO treatment. Data are shown as the mean±S.D. from four independent wells. Statistical analysis was performed using unpaired Student’s t-test. Asterisks indicate statistical differences (p<0.05). Open and filled columns show the values of groups treated without and with BDNF, respectively.

Fig. S3. Effect of Methylmercury on Cell Viability of TrkB-Stable Transformant of Rat B35 Cells. Cell viability of the TrkB-stable transformant, B35^TrkB, was estimated by MTT assay after 24-, 36- and 48-h incubations with methylmercury (MeHg) at concentrations indicated in the figure. Data are shown as the mean±S.D. from six independent wells. Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test. Asterisks indicate statistical difference in the treated groups of each incubation time from each negative control group treated with MeHg at 0.000 μM (p<0.05). Note there is statistically no difference in the group treated with MeHg at 0.625 μM compared to the control group after each incubation time.

Fig. S4. Effect of BSO on Cell Viability of TrkB-Stable Transformant of Rat B35 Cells. Cell viability of the stable transformant, B35^TrkB, was estimated by MTT assay after 12-h incubation with BSO at 50 μM. BDNF at 100 ng/mL as a final concentration was added to the culture 30 min before the BSO treatment. Data are shown as the mean±S.D. from four independent wells. Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test. Asterisks indicate statistical differences between indicated groups (p<0.05). N.S., no significance. Note that BSO and/or BDNF treatment did not decrease the cell viability compared to control group after 12-h incubation.

Fig. S5. Effect of MAP-Signaling Inhibitors on BDNF-Induced Alterations of Methylmercury-Induced Cell Death and Intracellular Total GSH Level in Rat Primary Culture of Cerebellar Granular Neurons. The cultures were treated with U0126 or PD98059 (PD) 30 min before methylmercury (MeHg) and/or BDNF. MeHg, BDNF, U0126 and PD were added to the cultures at 100 nM, 100 ng/mL, 1 and 30 μM, respectively. (A) No effect of PD on BDNF-induced acceleration of MeHg-induced cell death. The cell viability was estimated by MTT assay after 48h-incubation with MeHg. Effect of U0126 or PD on MeHg-induced (B) or BDNF-induced (C) alteration of intracellular total GSH level. The cultures in (B) and (C) were sampled for GSH assay after 24-h incubation with MeHg or BDNF. Data are shown as the mean±S.D. from four independent wells. Statistical analysis was performed using two-way ANOVA with Fisher’s PLSD test as a post hoc test. Columns with different lowercase letters are significantly different (p<0.05) from one another. Note that PD had no effect on the established effects of MeHg and BDNF.

REFERENCES

• 1) Ross D. Glutathione, free radicals and chemotherapeutic agents. Mechanisms of free-radical induced toxicity and glutathione-dependent protection. Pharmacol. Ther., 37, 231–249 (1988).
• 2) Orrenius S, Moldéus P. The multiple roles of glutathione in drug metabolism. Trends Pharmacol. Sci., 5, 432–435 (1984).
• 3) Coyle JT, Puttfarcken P. Oxidative stress, glutamate, and neurodegenerative disorders. Science, 262, 689–695 (1993).
• 4) Halliwell B. Oxidants and human disease: some new concepts. FASEB J., 1, 358–364 (1987).
• 5) Bibel M, Barde YA. Neurotrophins: key regulators of cell fate and cell shape in the vertebrate nervous system. Genes Dev., 14, 2919–2937 (2000).
• 6) Reichardt LF. Neurotrophin-regulated signalling pathways. Phil. Trans. R. Soc. B, 361, 1545–1564 (2006).
• 7) Koh JY, Gwag BJ, Lobner D, Choi DW. Potentiated necrosis of cultured cortical neurons by neurotrophins. Science, 268, 573–575 (1995).
• 8) de Luca A, Weller M, Frei K, Fontana A. Maturation-dependent modulation of apoptosis in cultured cerebellar granule neurons by cytokines and neurotrophins. Eur. J. Neurosci., 8, 1994–2005 (1996).
• 9) Sakaue M, Mori N, Makita M, Fujishima K, Hara S, Arishima K, Yamamoto M. Acceleration of methylmercury-induced cell death of rat cerebellar neurons by brain-derived neurotrophic factor in vitro. Brain Res., 1273, 155–162 (2009).
• 10) Maki T, Arishima K, Yamamoto M, Sakaue M. TrkB is involved in the mechanisms by which BDNF accelerates the glutamate-induced death of rat neuroblastoma B35 cells. Neurol. Res., 37, 30–34 (2015).
• 11) Murphy TH, Miyamoto M, Sastre A, Schnaar RL, Coyle JT. Glutamate toxicity in a neuronal cell line involves inhibition of cystine transport leading to oxidative stress. Neuron, 2, 1547–1558 (1989).
• 12) Yee S, Choi BH. Oxidative stress in neurotoxic effects of methylmercury poisoning. Neurotoxicology, 17, 17–26 (1996).
• 13) Sanfeliu C, Sebastià J, Kim SU. Methylmercury neurotoxicity in cultures of human neurons, astrocytes, neuroblastoma cells. Neurotoxicology, 22, 317–327 (2001).
• 14) Sarafian T, Anthony Verity M. Oxidative mechanisms underlying methylmercury neurotoxicity. Int. J. Dev. Neurosci., 9, 147–153 (1991).
• 15) Sarafian TA, Vartavarian L, Kane DJ, Bredesen DE, Verity MA. Bcl-2 expression decreases methyl mercury-induced free-radical generation and cell killing in a neural cell line. Toxicol. Lett., 74, 149–155 (1994).
• 16) Kaur P, Aschner M, Syversen T. Glutathione modulation influences methyl mercury induced neurotoxicity in primary cell cultures of neurons and astrocytes. Neurotoxicology, 27, 492–500 (2006).
• 17) Sakaue M, Mori N, Okazaki M, Kadowaki E, Kaneko T, Hemmi N, Sekiguchi H, Maki T, Ozawa A, Hara S, Arishima K, Yamamoto M. Vitamin K has the potential to protect neurons from methylmercury-induced cell death in vitro. J. Neurosci. Res., 89, 1052–1058 (2011).
• 18) Kromidas L, Trombetta LD, Jamall IS. The protective effects of glutathione against methylmercury cytotoxicity. Toxicol. Lett., 51, 67–80 (1990).
• 19) Sakaue M, Mori N, Okazaki M, Ishii M, Inagaki Y, Iino Y, Miyahara K, Yamamoto M, Kumagai T, Hara S, Yamamoto M, Arishima K. Involvement of independent mechanism upon poly(ADP-ribose) polymerase (PARP) activation in methylmercury cytotoxicity in rat cerebellar granule cell culture. J. Neurosci. Res., 86, 3427–3434 (2008).
• 20) Sakaue M, Okazaki M, Hara S. Very low levels of methylmercury induced cell death of cultured rat cerebellar neurons via calpain activation. Toxicology, 213, 97–106 (2005).
• 21) Sekiguchi H, Hemmi N, Maki T, Ozawa A, Kadowaki E, Kamiie J, Yamamoto M, Arishima K, Sakaue M. Culture on a fragmin/protamine-coated plate suppresses the collagen type I alpha I and TGF-beta1 mRNA expression of rat hepatic stellate RI-T cells. J. Vet. Med. Sci., 75, 553–559 (2013).
• 22) Griffith OW. Mechanism of action, metabolism, and toxicity of buthionine sulfoximine and its higher homologs, potent inhibitors of glutathione synthesis. J. Biol. Chem., 257, 13704–13712 (1982).
• 23) Plummer JL, Smith BR, Sies H, Bend JR. Chemical depletion of glutathione in vivo. Methods Enzymol., 77, 50–59 (1981).
• 24) Gabaizadeh R, Staecker H, Liu W, Van De Water TR. BDNF protection of auditory neurons from cisplatin involves changes in intracellular levels of both reactive oxygen species and glutathione. Brain Res. Mol. Brain Res., 50, 71–78 (1997).
• 25) Duncia JV, Santella JB 3rd, Higley AH, Pitts WJ, Wityak J, Frietze WE, Rankin FW, Sun JH, Earl RA, Tabaka CA, Teleha CA, Blom KF, Favata MF, Manos EJ, Daulerio AJ, Stradley DA, Horiuchi K, Copeland RA, Scherle PA, Trzaskos JM, Magolda RL, Trainor GL, Wexler RR, Hobbs FW, Olson RE. MEK inhibitors: the chemistry and biological activity of U0126, its analogs, and cyclization products. Bioorg. Med. Chem. Lett., 8, 2839–2844 (1998).
• 26) Alessi DR, Cuenda A, Cohen P, Dudley DT, Saltiel AR. PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J. Biol. Chem., 270, 27489–27494 (1995).
• 27) Kim SH, Won SJ, Sohn S, Kwon HJ, Lee JY, Park JH, Gwag BJ. Brain-derived neurotrophic factor can act as a pronecrotic factor through transcriptional and translational activation of NADPH oxidase. J. Cell Biol., 159, 821–831 (2002).
• 28) Kaur P, Evje L, Aschner M, Syversen T. The in vitro effects of Trolox on methylmercury-induced neurotoxicity. Toxicology, 276, 73–78 (2010).
• 29) Park ST, Lim KT, Chung YT, Kim SU. Methylmercury-induced neurotoxicity in cerebral neuron culture is blocked by antioxidants and NMDA receptor antagonists. Neurotoxicology, 17, 37–45 (1996).
• 30) Wallin C, Weber SG, Sandberg M. Glutathione efflux induced by NMDA and Kainate: implications in neurotoxicity? J. Neurochem., 73, 1566–1572 (1999).
• 31) Müller WE, Romero FJ, Perovic S, Pergande G, Pialoglou P. Protection of flupirtine on beta-amyloid-induced apoptosis in neuronal cells in vitro: prevention of amyloid-induced glutathione depletion. J. Neurochem., 68, 2371–2377 (1997).
• 32) Tabuchi A, Funaji K, Nakatsubo J, Fukuchi M, Tsuchiya T, Tsuda M. Inactivation of aconitase during the apoptosis of mouse cerebellar granule neurons induced by a deprivation of membrane depolarization. J. Neurosci. Res., 71, 504–515 (2003).
• 33) Sun Z, Ma X, Yang H, Zhao J, Zhang J. Brain-derived neurotrophic factor prevents beta-amyloid-induced apoptosis of pheochromocytoma cells by regulationg Bax/Bcl-2 expression. Neural Regen. Res., 7, 347–351 (2012).
• 34) Castoldi AF, Barni S, Turin I, Gandini C, Manzo L. Early acute necrosis, delayed apoptosis and cytoskeletal breakdown in cultured cerebellar granule neurons exposed to methylmercury. J. Neurosci. Res., 59, 775–787 (2000).