Mouse glioma cells (GL261) were kindly provided by Professor Maria da Conceição Pedroso de Lima (Center for Neuroscience and Cell Biology, University of Coimbra, Portugal). This cell line was chosen since it is a very common model for GBM studies, presenting normal features of an astrocytoma (Strong et al., 2018) and is resistant to TMZ therapy similarly to human GBM (Haddad et al., 2021). Moreover, xenografts of these cells display histopathological features that closely resemble the human GBM (Yi et al., 2013). GL261 cells were cultured in DMEM (Dulbecco’s Modified Eagle’s Medium; Gibco) containing 4.5 g/L of glucose, Glutamax, 25 mM HEPES and supplemented with 10% of heat-inactivated (30 min, 56°C) fetal bovine serum (FBS; Biochrome) and 1% penicillin/streptomycin (Gibco) in a humidified incubator at 37°C and 5% CO₂.

Total cell lysates were prepared according to previous protocols (Branco et al., 2014). 1 × 10⁶ cells were plated in 10 mm² culture dishes incubated at 37°C until reaching 70–80% confluence. At this point, the culture medium was refreshed and TmHg and EtHg (0.5; 1; 2 µM) were added and incubated at 37°C for 24 h, after which cells were collected by trypsinization and washed with PBS 1x. Subsequently, the pellet was resuspended in lysis buffer (25 mM TrisCl, pH 7.5; 100 mM NaCl; 2.5 mM EDTA; 2.5 mM EGTA; 20 mM NaF, 1 mM sodium orthovanadate, 20 mM sodium pyrophosphate; 20 mM of sodium β-glycerophosphate, 0.5% Triton X-100, and 1 tablet of protease inhibitor cocktail per 10 ml; Roche) vortexed and frozen at −20°C. These lysates were compared to a control group and used to quantify TrxR and Trx activity and expression as described below.

To obtain the mitochondrial and cytosolic fractions, after exposure to Hg compounds cells were washed and suspended in mitochondrial isolation buffer (210 mM mannitol, 70 mM sucrose, 1 mM EDTA, 10 mM Hepes–NaOH, and pH 7.5), containing protease inhibitor cocktail (Roche). Cells were disrupted with a Teflon pestle and then centrifuged at 600 g for 10 min at 4°C. The resulting nuclear pellets were discarded, and the supernatant was further centrifuged at 13,000 g and 4°C for 15 min, to obtain the mitochondrial pellets and the supernatant soluble fraction. Mitochondrial pellets were treated with cell lysis buffer as described above for whole-cell lysates and the supernatant fraction was further centrifuged at 100,000 g and 4°C for 1 h, to produce the cytosolic fraction (Branco et al., 2014). These fractions were used to quantify TrxR and Trx activity and expression as described below.

The total protein present in the samples was quantified by the Bradford method. In a 96-well microplate (Nunc), samples were incubated with Coomassie dye (Bio-Rad; diluted 5 times) and subsequently, absorbance was measured at 595 nm in the microplate reader (Zenyth 3100, Anthos Labtec Instruments). Protein concentration in each sample was calculated from a calibration curve using BSA as a standard (Bradford, 1976).

The enzymatic activity of TrxR and Trx was determined according to the insulin end-point assay described by Arnér and Holmgren (2000) with modifications (Branco et al., 2014).

To determine TrxR activity, samples (50 µg for total cell lysates and 20 µg for subcellular fractions) were incubated in 96-well plates (Nunc) in TE buffer (50 mM Tris pH 7.5 + 2 mM EDTA) with 3 μM of human Trx (IMCO Corp, Sweden) (previously reduced with dithiothreitol, DTT, at 37°C and desalted in a NAP-5 column) and a master mix (1.6 mM insulin, 50 mM NADPH, 2 mM EDTA and 200 mM HEPES pH 7.6) for 20 min at 37°C. In parallel, control wells were prepared to contain the same reagents previously mentioned apart from Trx. After the incubation period, 250 µL of a 1 mM DTNB solution in 6 M guanidine-HCL (pH 8.0) was added to wells, and absorbance was measured at 412 nm in a microplate reader (Zenyth 3100, Anthos Labtec Instruments). The quantitation of TrxR activity was performed considering the difference in absorbance between the Trx-containing wells and the control wells. The same procedure was done for the determination of Trx activity, but the samples were incubated with 100 nM of recombinant rat TrxR (IMCO Corp, Sweden) instead of Trx (Branco et al., 2014).

The expression of thioredoxin system enzymes was evaluated by Western Blot in cell lysates and subcellular fractions after the separation of total soluble proteins (30 μg for both TrxR and for Trx) by SDS-PAGE on a 4–12% Bis-Tris gel with MES running buffer under reducing conditions (140 V for 1 h). Next, they were transferred to a nitrocellulose membrane (30 V for 2 h) and blocked with a 5% skimmed milk solution (1 h). The following antibodies were used: TrxR1 rabbit polyclonal IgG (sc-20147, Sta. Cruz), TrxR2 mouse monoclonal IgG (sc-376868, Sta. Cruz), Trx1 rabbit polyclonal IgG (ATRX8, IMCO Corp.), Trx2 rabbit polyclonal IgG (sc-50335, Sta. Cruz), GAPDH rabbit polyclonal IgG (sc-25778, Sta. Cruz), VDAC mouse monoclonal IgG (sc-390996, Sta. Cruz), goat anti-mouse IgG-HRP (sc-2005, Sta. Cruz), and mouse anti-rabbit IgG-HRP (sc-2357, Sta. Cruz). Expression levels were normalized for protein loading on the gel, which was assessed either by Ponceau-S staining prior to the blocking step or by evaluating the expression of housekeeping proteins (GAPDH, VDAC) (Branco et al., 2014).

The assessment of general ROS production was based on the 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) assay (Huang et al., 2014).

Cells (8 × 10⁴) were plated in black 96-well microplates and then incubated at 37°C for 24 h. Afterwards, the medium was removed, the cells were washed twice with PBS 1x, and culture media with DHCF-DA (50 mM) was added to each well followed by a 40 min incubation at 37°C. After the incubation period, the DHCF-DA containing media was removed and the wells were washed twice with PBS 1x before being replenished with fresh medium. This was followed by exposure of cells to Hg compounds (EtHg and TmHg 0.5; 1, 2.5; and 5 µM) for 3 h using 1 mM H₂O₂ as a positive control. The fluorescence signal was read in a microplate reader (Zenyth 3100, Anthos Labtec Instruments) using 495 nm as the excitation wavelength and 529 nm as the emission wavelength.

The redox state of Trx1 and Trx2 was evaluated by the quantification of the number of reduced -SH groups (Free -SH), according to the Protein Electrophoretic Mobility Shift Assay (Bersani et al., 2002; Branco et al., 2017).

Briefly, after exposure to the TmHg and EtHg for 24 h, cell pellets were resuspended in Sample Solution (SS - Tris 50 mM, Urea 8 M, pH 8.3) containing 30 mM of iodoacetic acid (IAA; Sigma) and incubated at 37°C for 30 min to alkylate free thiols. This was followed by centrifugation at 13,000 g for 10 min, and the resulting supernatant was washed 3 times with an acetone/1 M HCl mix (98:2 v/v), followed by centrifugation at 13,000 g for 5 min. After evaporation of the acetone mixture, the resulting pellet was incubated at 37°C for 30 min in a SS solution containing 5 mM of DTT to reduce thiols oxidized by TmHg and EtHg. Posteriorly, these thiols were alkylated with SS containing 10 mM iodoacetamide (IAM; Sigma) at 37°C for 30 min.

Next, the samples were loaded onto a Urea-PAGE gel (stacking gel: 4% acrylamide; run gel: 12% acrylamide; 8 M Urea) and run at 10 mA. After 2 h, the proteins in the gel were transferred to a nitrocellulose membrane blocked with skimmed milk and incubated with respective antibodies - Trx1 rabbit polyclonal IgG (ATRX8, IMCO Corp.), Trx2 rabbit polyclonal IgG (sc-50335, Sta. Cruz) and mouse anti-rabbit IgG-HRP (sc-2357, Sta. Cruz). The migration pattern will vary according to the oxidation state of Trx, with more reduced enzymes being enriched in negatively charged IAA, thus migrating more in the gel than more oxidized enzymes, enriched in IAM. Mouse Trx1 presents a total of 6 Cys residues (2 in the active site plus 4 structural) and, therefore, this protein can assume 7 possible oxidation states varying between the totally reduced enzyme (6 free -SH groups) and total oxidation (0 free -SH groups). On the other hand, Trx2 presents only 2 Cys residues and 3 possible oxidation states (Branco et al., 2017).

To evaluate the Prx2 redox state, cells were lysed in lysis buffer containing NEM (N-ethylmaleimide; Sigma) and, subsequently, the soluble protein fraction was separated by non-reducing SDS-PAGE electrophoresis, followed by transfer to a nitrocellulose membrane, after blockage with milk and incubation with the respective antibodies–Prx2 mouse monoclonal IgG (sc-515428, Sta. Cruz) and goat anti-mouse IgG-HRP (sc-2005, Sta. Cruz)—as previously described (Brown et al., 2008). The ratio between the bands of the Prx2 dimer and monomer was quantified for each sample using the Bio-Rad Quantity One software.

Caspase-3 activity was determined by measuring the enzymatic cleavage of chromophore p-nitroanilide (pNA) from the substrate N-acetyl-Asp-Glu-Val-Asp-pNA (DEVD-pNA) (Borralho et al., 2007).

This activity was determined based on a proteolytic reaction previously described (Borralho et al., 2007) with some modifications. In this assay total cell lysates were used, where cells were exposed to the TmHg and EtHg (1; 2 and 5 µM) for 24 h. This reaction was performed in an isolation buffer, containing 100 μg of cytosolic protein of cellular lysates, 50 μM of DTT, and 50 mM of DEVD-pNA (Sigma-Aldrich). Next, the reaction mixtures were incubated at 37°C for 3 h. Finally, the pNa formation was measured at 405 nm using a microplate reader (Zenyth 3100, Anthos Labtec Instruments) (Borralho et al., 2007).

The ViaCount Assay was used to distinguish viable and non-viable cells based on the differential permeability of two dyes in the Guava ViaCount Reagent. The membrane-permeant dye stains all nucleated cells, leaving the cellular debris unstained, while the membrane-impermeant dye brightly stains damaged cells, thus indicating apoptotic and dying cells. GL261 cells (6 × 10⁴) were seeded for 24 h in 24-well plates and thereafter were exposed to the TmHg and EtHg for 24 and 48 h. Following treatment, cells were collected from plates along with cell culture supernatants and centrifuged for 5 min (650 g). Supernatants were discarded and the cells were resuspended in phosphate-buffered saline (PBS) containing 2% FBS. Subsequently, 15 μL of cell suspension were mixed with 135 μL of Guava ViaCount reagent and incubated for 5 min at room temperature. Sample acquisition and data analysis were performed with the Guava easyCyte 5HT flow cytometer (GuavaTechnologies, Inc., Hayward, CA, United States) using the ViaCount software module.

Results in tables and figures are presented as the mean ± standard error (S.E.) of at least 3 independent experiments. Statistical differences between groups were determined by applying a t-test for independent samples and considered significant at p < 0.05 and very significant at p < 0.01.

The cytotoxicity of EtHg and TmHg to GL261 cells was evaluated by the MTT assay following 24, 48, and 72 h of exposure and confirmed by the LDH assay after 24 h of exposure (Figure 1).

Exposure to either TmHg or EtHg caused a concentration- and time-dependent inhibition of MTT reduction by GL261 cells which reflects a reduction in cellular viability (Figures 1A,B). As shown in Table 1 both mercury compounds caused an analogous effect on viability, with IC₅₀ values at 24 h in the range (2.5–2.7) µM.

TmHg and EtHg caused a similar LDH release and cytotoxicity (EC₅₀ of 1.5 µM) on GL261 cells.

Additionally, the MTT assay was also carried out in GL261 cells exposed to different concentrations of TMZ. However, after 24 h exposure, TMZ failed to reduce viability by greater than 40% (p < 0.05) excluding IC₅₀ calculation, which was only possible after 48 h and corresponded to (444 ± 60) µM (Table 1).

Based on these results the concentration of 200 µM of TMZ was used in the co-exposure assays with EtHg and TmHg.

Co-exposure assays showed (Figures 1A, B) that at all time-points, Hg compounds enhanced TMZ cytotoxicity in GL261 cells. This co-exposure effect was further confirmed in the LDH assay and the increase in LDH release is in good agreement with the results from the MTT assay.

On the other hand, as shown in Figure 1C, exposure to 200 µM of TMZ did not change the levels of LDH, but upon co-exposure with TmHg or EtHg (1 µM) LDH release increased by greater than 50% (p < 0.05).

EtHg and TM significantly inhibited the activity of TrxR and Trx in whole-cell lysates (p < 0.05) (Figures 2A,B) as well as in the subcellular fractions of cytosol and mitochondria (Figures 3A,B), in a concentration-dependent manner.

Considering the overall results, TrxR was more affected than Trx by both compounds. For instance, in whole-cell lysates, TrxR showed an IC₅₀ of 0.8 and 0.7 µM (p < 0.05) for EtHg and TmHg, respectively (Figure 2), whereas Trx was inhibited with an IC₅₀ of 1.3 µM (EtHg) and 0.7 µM (TmHg).

Thimerosal (1 µM) decreased TrxR activity by 60% (p < 0.01) in both subcellular fractions, whereas EtHg, while producing a strong inhibition over cytosolic TrxR (50% inhibition with 1 µM), had a lesser effect on TrxR activity in mitochondria (Figure 3A).

A comparable trend was observed for Trx activity, with thimerosal analogously affecting its activity in the cytosol (IC₅₀ = 1.2 µM) and mitochondria (IC ₅₀ = 1.6 µM), and EtHg affecting the cytosolic activity to a greater extent.

Additionally, the expression of both enzymes in the two subcellular fractions was comparable (Figure 3C).

Upon EtHg and TmHg exposure there was a concentration-dependent increase in ROS production after 3 h of exposure, as compared with the control (Figure 4A). The pro-oxidant effects were slightly more significant with TmHg exposure.

As shown in Figures 4B,C, exposure to 1 µM of either EtHg or TmHg altered the redox states of Trx1 and Trx2 increasing the oxidation of Cys residues in both isoforms.

Also, exposure for 24 h of GL261 to 1 and 2 μM of both EtHg and TmHg caused a 20–40% increase in Prx2 dimerization relative to the non-treated control (Figure 4D), which was significant for TmHg exposure (p < 0.05).

Caspase-3 activity was increased by all exposure levels of both EtHg and TmHg up to 50% at 5 µM (p < 0.05) showing a concentration dependence (Figure 5C).

These findings are consistent with the Guava ViaCount assay which showed an increase (>50% when compared to the control group) in the number of apoptotic cells after 24 and 48 h of exposure to both TmHg and EtHg (Figures 5A,B). The increase in apoptotic cell numbers was concentration and time dependent.