Neuroprotective and Antioxidant Properties of CholesteroNitrone ChN2 and QuinolylNitrone QN23 in an Experimental Model of Cerebral Ischemia: Involvement of Necrotic and Apoptotic Cell Death

Ischemic stroke is the leading cause of disability and the second leading cause of death worldwide. However, current therapeutic strategies are scarce and of limited efficacy. The abundance of information available on the molecular pathophysiology of ischemic stroke has sparked considerable interest in developing new neuroprotective agents that can target different events of the ischemic cascade and may be used in combination with existing treatments. In this regard, nitrones represent a very promising alternative due to their renowned antioxidant and anti-inflammatory effects. In this study, we aimed to further investigate the neuroprotective effects of two nitrones, cholesteronitrone 2 (ChN2) and quinolylnitrone 23 (QN23), which have previously shown great potential for the treatment of stroke. Using an experimental in vitro model of cerebral ischemia, we compared their anti-necrotic, anti-apoptotic, and antioxidant properties with those of three reference compounds. Both ChN2 and QN23 demonstrated significant neuroprotective effects (EC50 = 0.66 ± 0.23 μM and EC50 = 2.13 ± 0.47 μM, respectively) comparable to those of homo-bis-nitrone 6 (HBN6) and N-acetylcysteine (NAC) and superior to those of α-phenyl-N-tert-butylnitrone (PBN). While primarily derived from the nitrones’ anti-necrotic capacities, their anti-apoptotic effects at high concentrations and antioxidant powers—especially in the case of QN23—also contribute to their neuroprotective effects.


Introduction
Stroke is a is a medical condition caused by the disruption of blood flow to one or several brain regions [1]. It is currently the leading cause of disability and the second leading cause of death worldwide [2]. Its incidence rate continues to grow, driven by the increase in life expectancy in developed countries [3] and the accumulation of risk factors in increasingly younger populations [1,4]. greatest therapeutic potential in both in vitro and in vivo studies. Our goal was to comparatively analyze the contributions of the inhibition of necrosis and apoptosis and of the production of reactive oxygen species (ROS) to the neuroprotective effects of these nitrones against cerebral ischemia. We also juxtaposed their effectiveness against that of HBN6 ( Figure 1C) (a nitrone developed by our research group and used as an internal reference) and two standards with proven efficacy, namely, PBN ( Figure 1D) (the parent nitrone that provides their structural base) and N-acetylcysteine (NAC), a well known antioxidant [3].

Treatment of Neuroblastoma Cell Cultures with Oligomycin-Rotenone (O-R)
To induce oxidative stress, SH-SY5Y cells were treated with oligomycin and rotenone, inhibitors of complexes I and V of the respiratory chain, respectively. A mixture of 10 μM of oligomycin and 30 μM of rotenone was added to the wells, followed by the addition of various concentrations of the tested compounds (0.001-1000 μM). The plates were then kept for 24 h in an aerobic culture chamber (95% air and 5% CO2) before the cell viability assays were performed. A control condition was included, maintaining cells in standard Dulbecco's medium without adding O-R or any neuroprotective compound.

Treatment of Neuroblastoma Cell Cultures with Oligomycin-Rotenone (O-R)
To induce oxidative stress, SH-SY5Y cells were treated with oligomycin and rotenone, inhibitors of complexes I and V of the respiratory chain, respectively. A mixture of 10 µM of oligomycin and 30 µM of rotenone was added to the wells, followed by the addition of various concentrations of the tested compounds (0.001-1000 µM). The plates were then kept for 24 h in an aerobic culture chamber (95% air and 5% CO 2 ) before the cell viability assays were performed. A control condition was included, maintaining cells in standard Dulbecco's medium without adding O-R or any neuroprotective compound. Additionally, vehicle-only control conditions were included, employing ethanol (the solvent used for ChN2) and dimethyl sulfoxide (the solvent used for the rest of the molecules). To induce experimental ischemia (I), SH-SY5Y neuroblastoma cells were subjected to OGD for 3 h. This was achieved by adding glucose-free Dulbecco's medium (Gibco, Life Technologies, Madrid, Spain) to the wells and placing the plates inside an anaerobic chamber in the presence of a gas mixture of 95% N 2 /5% CO 2 which was humidified at 37 • C and at a constant pressure of 0.15 bar. After the OGD period, the glucose-free medium was replaced by the standard oxygenated medium oxygen, the compounds were added to the wells at concentrations ranging from 0.001 to 1000 µM, and the cells were maintained in normoxic conditions for 24 h to simulate reperfusion (IR). It is important to emphasize that although the term "ischemia-reperfusion" (IR) is used in this paper as an abbreviation, a more accurate description when using a cellular model would be "oxygen and glucose resupply" (OGD-R). Controls were performed by preserving SH-SY5Y cells in glucose-containing Dulbecco's medium in an incubator in normoxic conditions for 3 h, after which the medium was changed, and the cells were kept in the same conditions for an additional 24 h. Sodium nitroprusside (SNP) was also included at concentrations of 2 mM and 5 mM as a positive control for necrotic and apoptotic cell death. Moreover, we included the same concentrations of the vehicle used to dissolve the tested compounds (final concentration < 1%) in all experiments.

Evaluation of Cell Viability
To assess the neurotoxicity and neuroprotective effects of the compounds, cell viability was measured in SH-SY5Y human neuroblastoma cells seeded in 96-well culture plates at a density of 0.5 × 10 5 cells/well. After subjecting the cells to O-R, OGD-R, or a control treatment, ChN2, QN23, HBN6, PBN, and NAC were added at concentrations ranging from 0.001 to 1000 µM. Then, cell viability was evaluated using the Cell Proliferation Kit II (XTT) (Sigma Aldrich, Madrid, Spain), which quantifies cellular respiration based on the ability of metabolically active cells to reduce a yellow tetrazolium salt to an orange formazan dye [23]. After incubating the cells with an XTT solution (0.3 mg/mL) for 2 h at 37 • C with 5% CO 2 and 95% air (v/v), the absorbances at 450 nm and 650 nm used as references were measured with a spectrophotometer (Power-Wave XS microplate-reader, BioTek Instruments, Madrid, Spain) [21,23]. A viability of 100% was set by the control normoxic cells treated only with standard Dulbecco's medium. The effects of ChN2, QN23, HBN6, PBN and NAC were analyzed in four independent experiments, each carried out in triplicate.

Assessment of LDH Activity
To determine the neuroprotective effects of the compounds against necrotic cell death, SH-SY5Y cells were cultured in 48-well plates at a density of 1-1.5 × 10 5 cells/well and exposed to OGD-R conditions (including ischemia and reperfusion controls), adding the tested compounds at 0.001-1000 µM. To assess extracellular lactate dehydrogenase (LDH) levels, media from every well were collected and stored at −20 • C until measurement. Moreover, a lysis buffer containing 0.5% Triton X-100 in 0.1 M of phosphate buffer at a pH of 7.5 was added to each well, and the cells were scratched from the bottom. These samples were centrifuged at 13,000 rpm, and intracellular LDH levels were measured in the supernatants. All measurements were obtained using a spectrophotometer reader (Power-Wave XS microplate-reader, BioTek Instruments, Madrid, Spain), and the LDH activity was calculated as the degree of absorbance decline at 340 nm, indicating the oxidation of NADH to NAD + , as described elsewhere [10,21]. The results were expressed as LDH activity in the extracellular medium normalized to the intracellular LDH activity obtained after cell lysis. The anti-necrotic effects of ChN2, QN23, HBN6, PBN, and NAC were analyzed in four independent experiments, and each experiment was carried out in triplicate.

Measurement of Caspase-3 Activity
To evaluate the anti-apoptotic effects of the drugs, human neuroblastoma cells were seeded in 48-well plates at a density of 1-1.5 × 10 5 cells/well and subjected to OGD-R or standard glucose and oxygen levels as previously described. After a 24 h recovery period in normoxic conditions, the compounds were added at concentrations ranging from 0.001 to 1000 µM. The cells were then lysed in a buffer containing 5 mM of Tris/HCl (pH 8.0), 20 mM of ethylenediaminetetraacetic acid, and 0.5% Triton X-100, and the lysates were centrifuged at 13,000 rpm for 10 min. The caspase-3 activity was assessed by using the caspase substrate N-Acetyl-Asp-Glu-Val-Asp-amido-4-methylcoumarin (Ac-DEVD-AMC) (Merck, Madrid, Spain), which generates a fluorogenic compound after it is metabolized. Fluorescence was measured using a spectrofluorometer (Bio-Tek FL 600, BioTek Instruments, Madrid, Spain), using excitation at 360 nm and emission at 480 nm as previously described [10,21,23]. The Bradford assay was used to determine protein levels prior to the full assessment of caspase-3 activity. The anti-apoptotic effects of ChN2, QN23, HBN6, PBN, and NAC were evaluated in five independent experiments carried out in triplicate.

Determination of Reactive Oxygen Species Formation
To evaluate the antioxidant effects of the compounds, SH-SY5Y cells were seeded in 48-well plates at a density of 1-1.5 × 10 5 cells/well. The cultured cells were subjected to OGD for 3 h, after which the glucose-free medium was replaced with oxygenated Dulbecco's medium. Then, the compounds were added at concentrations ranging from 0.001 to 1000 µM, and the plates were maintained in normoxic conditions at 37 • C for 2.5 h. Dihydroethidium (Hydroethidine) (DHE (HEt; Molecular Probes, ThermoFisher Scientific, Madrid, Spain)), a fluorogenic molecule that reacts with the superoxide radical anion, was added to the medium, and fluorescence was recorded every 15-30 s over a 20 min period via a spectrofluorometer (Bio-Tek FL 600, BioTek Instruments, Madrid, Spain). the wavelengths of excitation and emission were 535 and 635 nm, respectively [21]. The fluorescence data were fitted using linear regression (expressed in arbitrary fluorescence units, AFUs), and the slopes were deemed indicators of mitochondrial superoxide production, as described previously [10,23,24]. The antioxidant effects of the tested compounds were analyzed in four independent experiments, and each experimental condition was tested in triplicate.

Statistical Analyses
The data obtained from cell cultures were expressed as the means ± SEMs of the results from at least three independent experiments, each performed in triplicate. Statistical comparisons among the different experimental groups were performed using a one-way analysis of variance (ANOVA), followed by a Holm-Sidak post hoc test. A p-value < 0.05 was considered statistically significant. The fitting curves for the effective dose 50 (EC 50 ) and maximal activity determinations were generated using SigmaPlot v.11 (Systat Software INC., Palo Alto, CA, USA, 2012). Curve adjustments to estimate the EC 50 values were carried out via a non-linear regression analysis of minimal squared using logistic curves f1 = min + (max − min)/(1 + (x/EC50)ˆ(−Hillslope)). Regression analyses and statistics were carried out via the Pearson correlation test, also using SigmaPlot v.11 (Systat Sofware INC., Palo Alto, CA, USA, 2012).

Results
This study aimed to increase our understanding of the neuroprotective and antioxidant effects of cholesteronitrone 2 (ChN2) and quinolylnitrone 23 (QN23) in SH-SY5Y human neuroblastoma cell cultures. This cell line is widely used as an in vitro model to investigate neuronal damage in ischemic stroke [25,26] and neurodegenerative diseases [27,28].
3.1. Basal Neurotoxicities of ChN2, QN23, HBN6, PBN, and NAC As a prior step to correctly determining the neuroprotective effects of the tested compounds in experimental models of ischemia and oxidative stress, we set out to determine Antioxidants 2023, 12, 1364 6 of 18 their basal neurotoxicities in the absence of any toxic insult via the XTT assay. While ChN2 showed an increase in cell death at a starting concentration of 100 µM, QN23 displayed toxic effects at concentrations of 500 µM and higher (Figure 2). At 1 mM, ChN2 and QN23 showed reductions in cell viability of approximately 50% and 20%, respectively. Consequently, the basal neurotoxicities of these two nitrones were considered when evaluating the results of the neuroprotection assays. On the other hand, no substantial neurotoxic effects were observed for the three other compounds used as standards (HBN6, PBN, and NAC) at any of the tested concentrations. These data indicate the compounds' lack of neurotoxic effects at a wide range of concentrations, making them suitable for subsequent neuroprotective assays. Their potential in neuroprotective studies is further supported by their compatibility with the typical concentration range used in compound screening assays for hit discovery (1-10 µM). dant effects of cholesteronitrone 2 (ChN2) and quinolylnitrone 23 (QN23) in SH-SY5Y man neuroblastoma cell cultures. This cell line is widely used as an in vitro model vestigate neuronal damage in ischemic stroke [25,26] and neurodegenerative dis [27,28].

Basal Neurotoxicities of ChN2, QN23, HBN6, PBN, and NAC
As a prior step to correctly determining the neuroprotective effects of the tested pounds in experimental models of ischemia and oxidative stress, we set out to deter their basal neurotoxicities in the absence of any toxic insult via the XTT assay. While C showed an increase in cell death at a starting concentration of 100 μM, QN23 displ toxic effects at concentrations of 500 μM and higher (Figure 2). At 1 mM, ChN2 and Q showed reductions in cell viability of approximately 50% and 20%, respectively. C quently, the basal neurotoxicities of these two nitrones were considered when evalu the results of the neuroprotection assays. On the other hand, no substantial neuro effects were observed for the three other compounds used as standards (HBN6, PBN NAC) at any of the tested concentrations. These data indicate the compounds' lack of rotoxic effects at a wide range of concentrations, making them suitable for subseq neuroprotective assays. Their potential in neuroprotective studies is further supporte their compatibility with the typical concentration range used in compound screenin says for hit discovery (1-10 μM). The stat analysis (one-way ANOVA) reveals significant differences in viability compared to C24h (* p < ** p < 0.01; *** p < 0.001). Statistical analysis of results exceeding 100% is not displayed.

Effect on Cell Viability
Upon confirming the lack of toxic effects of ChN2 and QN23 within drug disco concentration ranges, we proceeded to investigate the neuroprotective properties o tested compounds. We employed two in vitro models: an oxidative stress model ( involving the treatment of cells with oligomycin and rotenone, and an ischemia-rep sion model (IR) wherein cells were subjected to OGD followed by reperfusion. The statistical analysis (one-way ANOVA) reveals significant differences in viability compared to C24h (* p < 0.05; ** p < 0.01; *** p < 0.001). Statistical analysis of results exceeding 100% is not displayed.

Effect on Cell Viability
Upon confirming the lack of toxic effects of ChN2 and QN23 within drug discovery concentration ranges, we proceeded to investigate the neuroprotective properties of the tested compounds. We employed two in vitro models: an oxidative stress model (O-R) involving the treatment of cells with oligomycin and rotenone, and an ischemia-reperfusion model (IR) wherein cells were subjected to OGD followed by reperfusion. Both models depict events within the recanalization process, which is particularly relevant at present as it represents the ideal timeframe for the intervention of neuroprotective agents in the treatment of ischemic stroke [15,16]. An XTT assay was used to evaluate cellular metabolic capacity after exposure to the toxic insults and the addition of the compounds under study.

Figure 3. Neuroprotective effects of ChN2, QN23, HBN6, PBN, and NAC against metabolic capacity loss induced by (A) oligomycin-rotenone (O-R) treatment and (B) oxygen-glucose deprivation (OGD) (3 h) followed by oxygen and glucose resupply (OGD-R), indicated as IR (24 h) in SH-SY5Y human neuroblastoma cells.
The bar graphs represent the % of cell viability at the indicated compound concentrations. Values shown are the means ± SEMs of four experiments performed in triplicate. Black asterisks (*): significant differences in viability between the compounds and the exposure to the O-R/ischemia-reperfusion (IR) conditions only. Red asterisks (*): significant differences in cell viability between O-R/IR models and controls (C3h and C24h) (* p < 0.05; ** p < 0.01; *** p < 0.001; one-way ANOVA followed by Holm-Sidak post hoc test).

Figure 3. Neuroprotective effects of ChN2, QN23, HBN6, PBN, and NAC against metabolic capacity loss induced by (A) oligomycin-rotenone (O-R) treatment and (B) oxygen-glucose deprivation (OGD) (3 h) followed by oxygen and glucose resupply (OGD-R), indicated as IR (24 h) in SH-SY5Y human neuroblastoma cells.
The bar graphs represent the % of cell viability at the in-dicated compound concentrations. Values shown are the means ± SEMs of four experiments per-formed in triplicate. Black asterisks (*): significant differences in viability between the compounds and the exposure to the O-R/ischemia-reperfusion (IR) conditions only. Red asterisks (*): significant differences in cell viability between O-R/IR models and controls (C3h and C24h) (* p < 0.05; ** p < 0.01; *** p < 0.001; one-way ANOVA followed by Holm-Sidak post hoc test).
The five tested compounds exhibited potent neuroprotective effects, reversing the O-R/IR-induced decrease in cell viability (decline in metabolic activity) in a dose-dependent manner at all or most of the concentrations. Interestingly, ChN2 and QN23 demonstrated the strongest neuroprotective effects against O-R toxicity ( Figure 3A), even at the lowest concentration of 0.001 µM. In the IR model, ChN2 and QN23 displayed a broad range of efficacy (0.1-1000 µM), although their neuroprotective effects at the higher concentrations seemed to be lower than those of the rest of the compounds ( Figure 3B).
To further explore the pharmacodynamic parameters of the tested compounds, we determined their EC 50 and maximal neuroprotective activity (MNA) values as indicators of their efficacy and potency ( Figure 4). The EC 50 values of the compounds against O-R toxicity from the lowest to the highest were QN23 ≤ ChN2 ≤ NAC ≤ HBN6 <<< PBN ( Figure 4A,C). These data show that in the oxidative stress model, ChN2 and QN23 have similar EC 50 values to those of HBN6 and NAC and significantly lower values than PBN; thus, PBN displays the lowest efficacy of all the tested drugs. On the other hand, the MNA values showed the following order: NAC = QN23 ≥ HBN6 ≥ ChN2 ≥ PBN, implying the absence of statistically significant differences between the potencies of the five assayed compounds. On the other hand, under IR conditions ( Figure 4B,D), the EC 50 values, from the lowest to the highest, follow the order ChN2 ≤ NAC ≤ HBN6 ≤ QN23 <<< PBN, showing again that ChN2, QN23, HBN6, and NAC have similar EC 50 values and significantly lower values than PBN. The comparison of the MNA values shows the following order of potency: NAC > PBN ≥ HBN6 > ChN2 > QN23, indicating that under the conditions of the ischemia-reperfusion model, ChN2 and QN23 are the least effective compounds in terms of their maximal neuroprotective activities. All these results confirm the efficacies of ChN2 and QN23 at low concentrations, while their capacities to enhance cell viability decrease at high concentrations. They also show that they are effective and potent drugs, particularly against oxidative stress toxicity.
R/IR-induced decrease in cell viability (decline in metabolic activity) in a dose-depen manner at all or most of the concentrations. Interestingly, ChN2 and QN23 demonst the strongest neuroprotective effects against O-R toxicity ( Figure 3A), even at the lo concentration of 0.001 μM. In the IR model, ChN2 and QN23 displayed a broad ran efficacy (0.1-1000 μM), although their neuroprotective effects at the higher concentra seemed to be lower than those of the rest of the compounds ( Figure 3B).
To further explore the pharmacodynamic parameters of the tested compound determined their EC50 and maximal neuroprotective activity (MNA) values as indic of their efficacy and potency ( Figure 4). The EC50 values of the compounds agains toxicity from the lowest to the highest were QN23 ≤ ChN2 ≤ NAC ≤ HBN6 <<< PBN ure 4A,C). These data show that in the oxidative stress model, ChN2 and QN23 have ilar EC50 values to those of HBN6 and NAC and significantly lower values than PBN; PBN displays the lowest efficacy of all the tested drugs. On the other hand, the values showed the following order: NAC = QN23 ≥ HBN6 ≥ ChN2 ≥ PBN, implyin absence of statistically significant differences between the potencies of the five as compounds. On the other hand, under IR conditions ( Figure 4B,D), the EC50 values, the lowest to the highest, follow the order ChN2 ≤ NAC ≤ HBN6 ≤ QN23 <<< PBN, s ing again that ChN2, QN23, HBN6, and NAC have similar EC50 values and signific lower values than PBN. The comparison of the MNA values shows the following ord potency: NAC > PBN ≥ HBN6 > ChN2 > QN23, indicating that under the conditions ischemia-reperfusion model, ChN2 and QN23 are the least effective compounds in of their maximal neuroprotective activities. All these results confirm the efficacies of C and QN23 at low concentrations, while their capacities to enhance cell viability dec at high concentrations. They also show that they are effective and potent drugs, pa larly against oxidative stress toxicity.  and MNA values of the tested compounds. Statistical comparisons were made com-paring the compounds to PBN and NAC. * p < 0.05, ** p < 0.01, and *** p < 0.001; ns: no statisti-cally significant difference (one-way ANOVA + Holm-Sidak post hoc test).

Effects on Necrotic Cell Death
To further explore the neuroprotective effects of the tested compounds, we decided to examine their anti-necrotic capacities. Necrosis is characterized by cellular swelling, Antioxidants 2023, 12, 1364 9 of 18 the rupture of intracellular organelles, and ultimately, plasma membrane breakdown [29]. Lactate dehydrogenase (LDH), a soluble cytosolic enzyme, easily crosses the damaged cell membrane. The comparison of its activity in the culture media to its intracellular activity makes it possible to infer the magnitude of necrotic death after cell exposure to IR conditions [27]. As shown in Figure 5, OGD cell exposure induced a percentage of LDH release of about 12% (12.35 ± 0.79%, mean ± SEM; n = 4), with a recovery of up to 7% after 24 h of reperfusion, while under basal conditions, the value of LDH release did not exceed 5.5% (5.07 ± 0.92% for C3h and 2.54 ± 0.18% for C24h, mean ± SEM; n = 4). All the assayed compounds significantly decreased the release of LDH in a dose-dependent manner except for the lowest concentrations of QN23 (0.001 µM) and PBN (0.001-0.1 µM) ( Figure 5).
ues. Data represent the means ± SEMs of four experiments performed in triplicate. (C,D) EC50 and MNA values of the tested compounds. Statistical comparisons were made comparing the compounds to PBN and NAC. * p < 0.05, ** p < 0.01, and *** p < 0.001; ns: no statistically significant difference (one-way ANOVA + Holm-Sidak post hoc test).

Effects on Necrotic Cell Death
To further explore the neuroprotective effects of the tested compounds, we decided to examine their anti-necrotic capacities. Necrosis is characterized by cellular swelling, the rupture of intracellular organelles, and ultimately, plasma membrane breakdown [29]. Lactate dehydrogenase (LDH), a soluble cytosolic enzyme, easily crosses the damaged cell membrane. The comparison of its activity in the culture media to its intracellular activity makes it possible to infer the magnitude of necrotic death after cell exposure to IR conditions [27]. As shown in Figure 5, OGD cell exposure induced a percentage of LDH release of about 12% (12.35 ± 0.79%, mean ± SEM; n = 4), with a recovery of up to 7% after 24 h of reperfusion, while under basal conditions, the value of LDH release did not exceed 5.5% (5.07 ± 0.92% for C3h and 2.54 ± 0.18% for C24h, mean ± SEM; n = 4). All the assayed compounds significantly decreased the release of LDH in a dose-dependent manner except for the lowest concentrations of QN23 (0.001 μM) and PBN (0.001-0.1 μM) ( Figure 5). Concentration-response curves ranging from 0.001 to 1000 μM were fitted to a nonlinear regression to calculate the EC50 and MNA values ( Figure 6). The EC50 values ranked the compounds from higher potency to lower potency as follows: ChN2 ≤ PBN ≤ NAC ≤ HBN6 < QN23 ( Figure 6B). According to their MNA values, the compounds showed this order in terms of efficacy: QN23 ≥ NAC ≥ HBN6 > ChN2 ≥ PBN ( Figure 6B). Consequently, ChN2 had a similar EC50 value to those of HBN6, PBN, and NAC and a significantly lower value than the values of QN23. QN23 displayed the highest MNA value, similar to the values of HBN6 and NAC and significantly higher than those of ChN2 and PBN. These results underscore the remarkable anti-necrotic effects of ChN2 at low concentrations and QN23 at high concentrations. Concentration-response curves ranging from 0.001 to 1000 µM were fitted to a nonlinear regression to calculate the EC 50 and MNA values ( Figure 6). The EC 50 values ranked the compounds from higher potency to lower potency as follows: ChN2 ≤ PBN ≤ NAC ≤ HBN6 < QN23 ( Figure 6B). According to their MNA values, the compounds showed this order in terms of efficacy: QN23 ≥ NAC ≥ HBN6 > ChN2 ≥ PBN ( Figure 6B). Consequently, ChN2 had a similar EC 50 value to those of HBN6, PBN, and NAC and a significantly lower value than the values of QN23. QN23 displayed the highest MNA value, similar to the values of HBN6 and NAC and significantly higher than those of ChN2 and PBN. These results underscore the remarkable anti-necrotic effects of ChN2 at low concentrations and QN23 at high concentrations.

Effects on Apoptotic Cell Death
To comprehensively investigate the neuroprotective effects of the tested compounds, we examined their anti-apoptotic properties. Caspases-specifically Caspase-3-are the main proteins involved in this type of cell death in mammals [30]. To assess Caspase-3 activity, we used Ac-DVED-AMC, a substrate with a specific amino acid sequence that is hydrolyzed by Caspase-3, forming AMC, a fluorescent product that can be measured in the spectrofluorometer. As depicted in Figure 7, exposure to OGD resulted in a Caspase-3 activity of 27.33 ± 3.13 AFU/h × μg protein (mean ± SEM; n = 5), which decreased to 21.10 ± 1.90 (mean ± SEM; n = 5) after reperfusion. Under the basal conditions C3h and C24h, the activity was 4.21 ± 0.14 (mean ± SEM; n = 5) and 5.69 ± 0.70 AFU/h × μg (mean ± SEM; n = 5), respectively. Although the lowest concentrations (0.001 and 0.01 μM) of the assayed compounds did not exhibit significant anti-apoptotic effects, the five drugs demonstrated some degree of anti-apoptotic activity at higher doses ( Figure 7). ChN2 and NAC had statistically significant neuroprotective effects across a wide range of concentrations (1-1000 and 0.1-1000 μM, respectively). QN23 and HBN6 exhibited anti-apoptotic effects at concentrations of 50-1000 μM, while PBN demonstrated anti-apoptotic activity only at 100, 500 and 1000 μM (Figure 7).

Effects on Apoptotic Cell Death
To comprehensively investigate the neuroprotective effects of the tested compounds, we examined their anti-apoptotic properties. Caspases-specifically Caspase-3-are the main proteins involved in this type of cell death in mammals [30]. To assess Caspase-3 activity, we used Ac-DVED-AMC, a substrate with a specific amino acid sequence that is hydrolyzed by Caspase-3, forming AMC, a fluorescent product that can be measured in the spectrofluorometer. As depicted in Figure 7, exposure to OGD resulted in a Caspase-3 activity of 27.33 ± 3.13 AFU/h × µg protein (mean ± SEM; n = 5), which decreased to 21.10 ± 1.90 (mean ± SEM; n = 5) after reperfusion. Under the basal conditions C3h and C24h, the activity was 4.21 ± 0.14 (mean ± SEM; n = 5) and 5.69 ± 0.70 AFU/h × µg (mean ± SEM; n = 5), respectively. Although the lowest concentrations (0.001 and 0.01 µM) of the assayed compounds did not exhibit significant anti-apoptotic effects, the five drugs demonstrated some degree of anti-apoptotic activity at higher doses ( Figure 7). ChN2 and NAC had statistically significant neuroprotective effects across a wide range of concentrations (1-1000 and 0.1-1000 µM, respectively). QN23 and HBN6 exhibited antiapoptotic effects at concentrations of 50-1000 µM, while PBN demonstrated anti-apoptotic activity only at 100, 500 and 1000 µM (Figure 7).
Concentration-anti-apoptotic effect curves were fitted to a non-linear regression to calculate the EC 50 and MNA values to rank the compounds for their potency and efficacy in preventing apoptotic cell death (Figure 8). The order of EC 50 values, from highest to lowest anti-apoptotic capacity, was as follows: NAC ≤ HBN6 < ChN2 ≤ QN23 < PBN ( Figure 8B). As for their MNA values, the compounds showed the following potency order: NAC ≥ HBN6 ≥ ChN2 ≥ PBN > QN23 ( Figure 8B). Thus, ChN2 showed a similar MNA value to HBN6, PBN and NAC and, although it had a higher EC 50 value than NAC and HBN6, it still was significantly lower than that of PBN. On the other hand, QN23 displayed a EC 50 value comparable to that of ChN2, but it also showed the lowest MNA value out of the five compounds. Therefore, ChN2 exhibited pronounced anti-apoptotic properties at medium to high concentrations, while QN23 showed less efficacy against this type of cell death, although it surpasses its precursor nitrone, PBN. Concentration-anti-apoptotic effect curves were fitted to a non-linear regression to calculate the EC50 and MNA values to rank the compounds for their potency and efficacy in preventing apoptotic cell death (Figure 8). The order of EC50 values, from highest to lowest anti-apoptotic capacity, was as follows: NAC ≤ HBN6 < ChN2 ≤ QN23 < PBN (Figure 8B). As for their MNA values, the compounds showed the following potency order: NAC ≥ HBN6 ≥ ChN2 ≥ PBN > QN23 ( Figure 8B). Thus, ChN2 showed a similar MNA value to HBN6, PBN and NAC and, although it had a higher EC50 value than NAC and HBN6, it still was significantly lower than that of PBN. On the other hand, QN23 displayed a EC50 value comparable to that of ChN2, but it also showed the lowest MNA value out of the five compounds. Therefore, ChN2 exhibited pronounced anti-apoptotic properties at medium to high concentrations, while QN23 showed less efficacy against this type of cell death, although it surpasses its precursor nitrone, PBN.   Concentration-anti-apoptotic effect curves were fitted to a non-linear regressi calculate the EC50 and MNA values to rank the compounds for their potency and eff in preventing apoptotic cell death (Figure 8). The order of EC50 values, from highe lowest anti-apoptotic capacity, was as follows: NAC ≤ HBN6 < ChN2 ≤ QN23 < PBN ure 8B). As for their MNA values, the compounds showed the following potency o NAC ≥ HBN6 ≥ ChN2 ≥ PBN > QN23 ( Figure 8B). Thus, ChN2 showed a similar M value to HBN6, PBN and NAC and, although it had a higher EC50 value than NAC HBN6, it still was significantly lower than that of PBN. On the other hand, QN23 played a EC50 value comparable to that of ChN2, but it also showed the lowest MNA out of the five compounds. Therefore, ChN2 exhibited pronounced anti-apoptotic p erties at medium to high concentrations, while QN23 showed less efficacy agains type of cell death, although it surpasses its precursor nitrone, PBN.

Antioxidant Effects
To assess the impact of the tested compounds' antioxidant properties on their neuroprotective capacity, we examined the rate of ROS production in SH-SY5Y cells following exposure to OGD (3 h) and reperfusion (2.5 h). To measure superoxide radical anion (O 2 •− ) levels, we used dihydroethidium (DHE) as a fluorogenic probe. As shown in Figure 9, O 2 •− production after IR (100 ± 9.29%, mean ± SEM; n = 4) was similar to O 2 •− production after 3 h of OGD exposure (102.18 ± 10.68%, mean ± SEM; n = 4). ChN2 and QN23 were effective at most or all the tested concentrations while HBN6, NAC and, especially, PBN, displayed a narrower range of potency. However, they showed better concentration-dependence and more significant antioxidant effects at the highest concentrations (100-1000 µM) (Figure 9).

Antioxidant Effects
To assess the impact of the tested compounds' antioxidant properties on their neuroprotective capacity, we examined the rate of ROS production in SH-SY5Y cells following exposure to OGD (3 h) and reperfusion (2.5 h). To measure superoxide radical anion (O2 •− ) levels, we used dihydroethidium (DHE) as a fluorogenic probe. As shown in Figure 9, O2 •− production after IR (100 ± 9.29%, mean ± SEM; n = 4) was similar to O2 •− production after 3 h of OGD exposure (102.18 ± 10.68%, mean ± SEM; n = 4). ChN2 and QN23 were effective at most or all the tested concentrations while HBN6, NAC and, especially, PBN, displayed a narrower range of potency. However, they showed better concentration-dependence and more significant antioxidant effects at the highest concentrations (100-1000 μM) (Figure 9). Concentration-antioxidant response curves and the EC50 values and the maximal antioxidant activities (MAA) of the tested compounds are shown in Figure 10. The order of EC50 values, from higher to lower antioxidant potency, was as follows: NAC ≤ HBN6 ≤ PBN ≤ QN23 ≤ ChN2 ( Figure 10B). Regarding the MAA values, the order of potency was as follows: HBN6 ≥ NAC ≥ QN23 ≥ PBN ≥ ChN2 ( Figure 10B). Notably, there were no statistically significant differences among the EC50 and MAA values of the five assayed compounds. Concentration-antioxidant response curves and the EC 50 values and the maximal antioxidant activities (MAA) of the tested compounds are shown in Figure 10. The order of EC 50 values, from higher to lower antioxidant potency, was as follows: NAC ≤ HBN6 ≤ PBN ≤ QN23 ≤ ChN2 ( Figure 10B). Regarding the MAA values, the order of potency was as follows: HBN6 ≥ NAC ≥ QN23 ≥ PBN ≥ ChN2 ( Figure 10B). Notably, there were no statistically significant differences among the EC 50 and MAA values of the five assayed compounds.

Correlation Analyses between the Different Neuroprotective Properties of ChN2, QN23, HBN6, PBN and NAC
To establish a comparative relationship between the different types of neuroprotective activities of the compounds and determine their contributions to neuroprotection, we performed correlation analyses. Individual analyses with data from all nine concentrations (0.001-1000 µM) of each compound indicated a statistically significant correlation between all neuroprotective mechanisms. We conducted linear correlation between the neuroprotective activity and the anti-apoptotic, anti-necrotic or antioxidant effects for all the compounds, comparing their EC 50 to be able to group them according to their mechanisms of action ( Figure 11). First, we observed a correlation between the neuroprotective and antioxidant activity of QN23, HBN6 and NAC ( Figure 11A). Moreover, the anti-necrotic effects of these three compounds were also significantly correlated with their antioxidant ( Figure 11D) and anti-apoptotic ( Figure 11F) capacity. ChN2, QN23, HBN6 and NAC showed a significant correlation between their neuroprotective and anti-necrotic activity ( Figure 11B) and between their antioxidant and anti-apoptotic activity ( Figure 11E). Finally, we observed a clear correlation between the neuroprotective and anti-apoptotic effects of the five studied compounds ( Figure 11C). These results indicate that, while the antioxidant, anti-necrotic and anti-apoptotic effects of QN23, HBN6 and NAC seem to be equally relevant for their overall neuroprotective capacity, ChN2's protective properties seem to derive especially from its anti-necrotic and anti-apoptotic activity, and PBN depends mainly on its anti-apoptotic capacity.

Correlation Analyses between the Different Neuroprotective Properties of ChN2, QN23 HBN6, PBN and NAC
To establish a comparative relationship between the different types of neuropr tive activities of the compounds and determine their contributions to neuroprotectio performed correlation analyses. Individual analyses with data from all nine conce tions (0.001-1000 μM) of each compound indicated a statistically significant correl between all neuroprotective mechanisms. We conducted linear correlation betwee neuroprotective activity and the anti-apoptotic, anti-necrotic or antioxidant effects f the compounds, comparing their EC50 to be able to group them according to their m nisms of action ( Figure 11). First, we observed a correlation between the neuroprote and antioxidant activity of QN23, HBN6 and NAC ( Figure 11A). Moreover, the an crotic effects of these three compounds were also significantly correlated with their oxidant ( Figure 11D) and anti-apoptotic ( Figure 11F) capacity. ChN2, QN23, HBN6 NAC showed a significant correlation between their neuroprotective and anti-necrot tivity ( Figure 11B) and between their antioxidant and anti-apoptotic activity ( Figure  Finally, we observed a clear correlation between the neuroprotective and anti-apop effects of the five studied compounds ( Figure 11C). These results indicate that, whi antioxidant, anti-necrotic and anti-apoptotic effects of QN23, HBN6 and NAC seem equally relevant for their overall neuroprotective capacity, ChN2′s protective prop seem to derive especially from its anti-necrotic and anti-apoptotic activity, and PBN pends mainly on its anti-apoptotic capacity.

Discussion
The present study aimed to delve deeper into the analysis of the neuroprotective and antioxidant mechanisms of two synthetic nitrones, cholesteronitrone 2 (ChN2) and quinolylnitrone 23 (QN23), which have shown promising therapeutic potential for the treatment of ischemic stroke in previous in vitro and in vivo studies [4,18,20,22]. Our objectives were to expand the knowledge of their neuroprotective effects at a wider range of

Discussion
The present study aimed to delve deeper into the analysis of the neuroprotective and antioxidant mechanisms of two synthetic nitrones, cholesteronitrone 2 (ChN2) and quinolylnitrone 23 (QN23), which have shown promising therapeutic potential for the treatment of ischemic stroke in previous in vitro and in vivo studies [4,18,20,22]. Our objectives were to expand the knowledge of their neuroprotective effects at a wider range of concentrations and against two of the main types of cell death involved in ischemia, namely, apoptosis and necrosis. Notably, the anti-necrotic effects of these nitrones were previously unexplored. Additionally, we aimed to compare the protective and antioxidant capacities of ChN2 and QN23 with those of HBN6, another nitrone studied by our research group [10], as well as PBN and NAC, which were used as standards.
In order to do so, we decided to use cultures of the human neuroblastoma cell line SH-SY5Y as our study model. While experiments conducted using cell lines have their limitations as they do not fully replicate the physiological characteristics of a real biological system, they also offer several advantages. Cell line models are accessible, relatively easy to maintain, provide a readily available and abundant supply of cells, and allow for control over all environmental variables, thereby enhancing reproducibility [26,31]. In fact, human in vitro systems are commonly employed for cost-effective high-throughput screening assays to assess the efficacy of therapeutic compounds [26,31]. The SH-SY5Y cell line in particular is a well-established model frequently employed in neuroprotection studies related to ischemic stroke [25,26] and neurodegenerative diseases [27,28].
Our findings revealed that ChN2 and QN23 had basal neurotoxic effects at high concentrations (100-1000 µM), which was not observed with any of the reference compounds ( Figure 2). Nevertheless, these two nitrones also showed remarkable neuroprotective potencies in oxidative stress and ischemia-reperfusion models (Figures 3 and 4). ChN2 and QN23 both displayed a wide range of efficacy (0.001-1000 µM) under both the O-R treatment and the OGD-R models. While they both exhibited exceptional effectiveness in increasing cell viability at high concentrations in the oxidative stress model, their efficacies were not as high in the IR model (Figures 3 and 4). These results were further supported by the statistical analysis of the EC 50 and MNA values. ChN2 (EC 50 = 0.29 ± 0.08 µM) and QN23 (EC 50 = 0.20 ± 0.08 µM) were, at low concentrations, the most neuroprotective compounds against oxidative stress (Figure 4). ChN2 also exhibited a notable potency in the ischemia model (EC 50 = 0.66 ± 0.23 µM), while QN23 (EC 50 = 2.13 ± 0.47 µM) displayed a slightly higher EC 50 value, although it was significantly lower than that of PBN (EC 50 = 8.24 ± 1.49 µM) (Figure 4). On the other hand, ChN2 and QN23 were the compounds with the lowest MNA values, which can be attributed to their demonstrated basal neurotoxicities at high concentrations (also observed in other works, such as that of Ayuso et al. in 2015) [18]. However, it should be noted that the observed MNA values are higher-and therefore denote greater neuroprotective efficacies-than the values previously reported for many other nitrones investigated by our research group. As an example, HBN1 showed an MNA value of 68.55 ± 3.96%, while ChN2 and QN23 always reached values greater than 100% [10]. Therefore, the results obtained in this study underscore the notable neuroprotective capacities of ChN2 and QN23, particularly at lower concentrations, which holds significant implications at the clinical level since much lower doses of ChN2 and QN23 would be needed to achieve the same therapeutic effects when compared to other nitrones.
Next, we aimed to investigate the contributions of the anti-necrotic and anti-apoptotic effects of ChN2 and QN23 to their neuroprotective effects. Both nitrones exhibited notable anti-necrotic properties, with ChN2 demonstrating a wide range of efficacy (0.001-1000 µM), while QN23 displayed the highest protective capacity against necrotic cell death at the highest tested concentrations (100-1000 µM) ( Figure 5). In terms of their EC 50 values, ChN2 (EC 50 = 1.77 ± 0.17 µM) showed a similar value to HBN6, PBN, and NAC, while its MNA value (MNA = 113.73 ± 4.50%) was significantly lower than that of HBN6 and NAC but comparable to PBN (Figure 6). Conversely, QN23 had the highest EC 50 value (EC 50 = 26.78 ± 8.39 µM) among all tested compounds yet displayed a similar MNA value (MNA = 146.63 ± 6.80%) to those of HBN6 and NAC and a significantly higher value than PBN (Figure 6). Regarding their anti-apoptotic effects, both ChN2 and QN23, along with the three standards, showed more limited efficacies. ChN2 and QN23 did not have neuroprotective effects against apoptosis at the lowest concentrations (0.001-10 µM), which was the same as HBN6 (Figure 7). However, ChN2 seemed to be as effective as NAC at the highest concentrations tested (500 and 1000 µM). Both ChN2 and QN23 displayed higher EC 50 values (EC 50 = 64.84 ± 9.19 and 82.50 ± 13.86, respectively) than HBN6 and NAC but significantly lower values than that of PBN ( Figure 8). Interestingly, the EC 50 values of the anti-apoptotic effects in this study were significantly higher compared to previous studies conducted by our research group (e.g., PBN and NAC had EC 50 values of 1.12 ± 0.27 µM and 0.05 ± 0.01 µM, respectively) [21]. This discrepancy could be attributed to several factors, including the use of lower concentrations of the compounds in this study (as ChN2 and QN23 demonstrated very promising results at 0.001 and 0.01 µM on the cell viability test), which may have impacted the shapes of the concentration-response curves, as well as the use of a different batch of cells, which may have introduced additional variability. In terms of the MNA value, ChN2 exhibited a similar value (MNA = 117.32 ± 6.50%) to HBN6, PBN, and NAC, while QN23 displayed the lowest MNA activity (MNA = 92.30 ± 5.59%) among all the tested compounds.
When evaluating the antioxidant properties of ChN2 and QN23, we found that they both showed a wide range of effectiveness. They notably had the widest ranges of efficacy, with QN23 being able to reduce O 2 •− production at all the tested concentrations ( Figure 9). These results are consistent with previous findings from our research group [18,20]. Notably, both nitrones showed similar EC 50 and MAA values to those of HBN6, PBN, and NAC ( Figure 10). However, they displayed better antioxidant profiles than other previously studied synthetic nitrones, such as HBN3 (EC 50 = 38.52 ± 5.25 µM) and QN2 (MAA= 82.22 ± 5.37%), with ChN2 and QN23 exhibiting EC 50 values below 15 µM and both compounds surpassing 100% activity [10,21].
Our final objective was to determine the contribution of different neuroprotective mechanisms to the overall neuroprotective capacity of each compound. Correlation analyses indicated that the neuroprotective effects of ChN2 were particularly dependent on its anti-necrotic and anti-apoptotic activities. In the case of QN23, HBN6, and NAC, their protective capacities seemed to be equally due to anti-necrotic, anti-apoptotic, and antioxidant capacities. Conversely, the neuroprotection provided by PBN was primarily dependent on its anti-apoptotic activity ( Figure 11). It is worth noting that these results are consistent with the maximal activity data, which were not included in the correlation analyses. Notably, the MNA values of ChN2 for its anti-necrotic (MNA = 113.73 ± 4.50%) and anti-apoptotic (MNA = 117.62 ± 6.50%) effects significantly exceeded its MAA values (101.82 ± 3.85%). Conversely, the maximal anti-apoptotic activity of PBN (MNA = 114.16 ± 3.96%) greatly surpasses its MNA values associated with its anti-necrotic and anti-apoptotic properties (MNA = 102.91 ± 3.79% and MNA = 102.10 ± 3.32%, respectively). One possible explanation for the difference in the neuroprotective mechanisms of ChN2 and QN23 can be found in their structural differences. ChN2 is a steroid nitrone with a cholesterol skeleton, while QN23 is a nitrone inserted into a quinoline heterocyclic ring. Additionally, ChN2 bears an N-methyl substituent at the nitrogen of the nitrone motif, whereas QN23 features a tert-butyl substituent at the nitrogen of the nitrone functional group.
In conclusion, ChN2 and QN23 demonstrated significant antioxidant, anti-necrotic, and anti-apoptotic properties. We examined their effects across a broad range of concentrations, which is particularly valuable as it allows for the determination of the optimal concentrations to be employed in future study models (e.g., organoids) or when investigating their impact on other types of cell death associated with cerebral ischemia (autophagy, pyroptosis, and ferroptosis) [11,12], mitochondrial membrane potential, and ATP production. Such investigations could provide deeper insight into the mechanisms underlying the neuroprotective effects of these compounds, elucidating the factors influencing their distinct neuroprotective activities.

Conclusions
Here, we conducted a comprehensive examination of the neuroprotective profiles of ChN2 and QN23, two synthetic nitrones that have demonstrated considerable potential for the treatment of ischemic stroke in previous in vitro and in vivo studies conducted by our research group. Our aim was to extend the existing knowledge by investigating their neuroprotective and antioxidant mechanisms across a wider range of concentrations, comparing them with three standard compounds (HBN6, PBN and NAC), and exploring their impacts on necrosis and apoptosis, which had not been previously explored.
In conclusion, our findings highlight the promising therapeutic potential of ChN2 and QN23 as neuroprotective agents, particularly at low concentrations. Notably, ChN2 and QN23 displayed significant anti-necrotic properties, acted as effective antioxidant molecules, and exhibited anti-apoptotic effects, albeit with more limited efficacies and only at high concentrations. Furthermore, ChN2's effect primarily relied on its anti-necrotic and anti-apoptotic activities for neuroprotection, while QN23's protective capacity against necrosis and apoptosis was equally relevant to its antioxidant effects. These findings underscore the potential of ChN2 and QN23 as promising candidates for the treatment of ischemic stroke, as their neuroprotective effects are evident even at low concentrations. The elucidation of their mechanisms of action and their comparison with other nitrones contribute to a better understanding of their efficacy and position them as viable candidates for further exploration and potential clinical applications.