TBN and NXY-059 were synthesized and purified in our laboratory as described previously 323334. TMP, edaravone 2,2-diphenyl-1-picrylhydrazyl (DPPH), N,N-dimethyl-4-nitrosoaniline (p-NDA), H₂O₂, pyrogallol, 2-amino-2-hydroxymethyl-propane-1,3-diol (Tris), 3-aminophthalhydrazide (luminol), tert-butyl hydroperoxide (t-BHP), 3-morpholinosydnonimine hydrochloride (SIN-1) and diethylamine NONOate diethylammonium salt (DEA NONOate) were purchased from Sigma Aldrich (St Louis, Mo, USA). Hoechst 33342, MitoProbe JC-1 assay kit, MitoSOX™ Red Mitochondrial Superoxide Indicator, 2',7'-dichlorodihydrofluorescein diacetate (H₂DCF-DA), dihydrorhodamine 123 (DHR 123), 4-amino-5-methylamino-2',7'-difluorofluorescein diacetate (DAF-FM DA), Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), 0.25% Trypsin-EDTA, Penicillin-Streptomycin (10,000 U/ml), phosphate buffered saline (PBS), were obtained from Invitrogen, Life technologies (Carlsbad, CA, USA). All other reagents were purchased from Sigma Aldrich unless otherwise specified.

The comparison of free radical-trapping activities among TBN, TMP, NXY-059 and edaravone (5, 20, 80, 320 μM used for all compounds) against 2,2-diphenyl-1-picrylhydrazyl radical (DPPH), hydroxyl radical (OH), superoxide anion (O₂⁻) and peroxynitrite (ONOO⁻) were determined by cell-free assays according to the previously published procedures with minor modification 3135.

DPPH radical-scavenging activity. One hundred microliter methanol (control) or a methanolic solution of each test compound was added in 96-well plates, and then 100 μl methanolic solution of DPPH (final concentration 50 μM) was added into each well. The plates were incubated in the dark at room temperature for 50 min. The measurement at 517 nm was measured using a plate reader (BioTek Synergy 4, Winooski, Vermont, USA). The clearance of the DPPH radical was calculated as follows: Clearance (%)= [(A_ctrl–A_t) /A_ctrl] × 100. Where A_ctrl was the absorbance of the control, and A_t was the absorbance of each sample solution at the time t = 50 min.

Hydroxyl radical-scavenging activity. The p-NDA, FeSO₄ and H₂O₂ were freshly prepared in N₂-purged, double-distilled H₂O (ddH₂O) to get a concentration of 1.0 mM, 2.0 mM and 1.0 mM respectively. While all of the test compounds were freshly dissolved in ddH₂O before the experiment, then 300 μl ddH₂O (control) or different concentration of each test compound, 50 μl p-NDA, 125 μl H₂O₂ and 125 μl Fe²⁺ were added to48-well plate in order. Hydroxyl free radical was generated by the Fenton reaction (Fe²⁺ + H₂O₂ → Fe³⁺ + OH^‒ + OH). The bleaching of p-NDA was monitored as the loss in absorbance at 440 nm for 100 s on a BioTek Synergy 4. The Clearance was calculated as follows: Clearance (%)=[1 ‒ (A₀ ‒ A₁₀₀)/A₀] ×100, where A₀ is the absorbance at the time t=0 s, and A₁₀₀ is the absorbance at t=100 s.

Superoxide anion-scavenging activity. Pyrogallol was dissolved in 10 mM HCl solution while Tris-HCl was prepared in N₂-purged ddH₂O to get a concentration of 2.0 mM and 50 mM, respectively. In each well of a 48-well plate, 250 μl Tris-HCl buffer (pH 8.2), 300 μl ddH₂O or different concentration of each test compound, 50 μl pyrogallol (final concentration 0.17 mM) was added one after another. The autoxidation rate of pyragallol was presented as the increase of absorbance per second at 320 nm (dA/dt) for 300 s on a BioTek Synergy 4. The inhibition ratio of pyrogallol autoxidation was calculated according to the formula: Clearance (%) = (dA_ctrl/dt ‒ dA_sample/dt) /dA_ctrl/ dt ×100.

Peroxynitrite-scavenging activity. Luminol and SIN-1 were dissolved in ice-water bathed PBS separately to get a concentration of 1.0 mM and 3 mg/ml. Firstly, 150 μl PBS, 250 μl of PBS (control) or each test compound, 50 μl luminol were pipetted into the luminometer tube before been positioned into the measurement chamber. Finally the reaction was initiated by adding 50 μl ice-colded SIN-1 with an injector. Then the luminescence was measured on an ultra-sensitive tube luminometer (Berthold Lumat LB 9507, Bad Wildbad, Germany) at the time interval of 100 s for 2000 s. The clearance at peak height represented the compounds’ ability of trapping ONOO^–. The clearance was calculated as follows: Clearance (%) = [(A_ctrl ‒ A_sample) /A_ctrl] × 100.

Murine H9c2 cardiomyoblasts (ATCC CRL-1446, Rockville, MD, USA) were maintained in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin (100 U/ml), at 37 ^oC in a humidified atmosphere with 5% CO₂. Cells were passaged regularly and sub-cultured to ∼80% confluence, then starved with serum free DMEM for 24 h before experimental procedures.

H9c2 cells were seeded in the 96-well plate (5×10³ cells/well) for 24 h growth, then exposed to increasing doses of t-BHP, DEA NONOate or SIN-1 for 24 h. The impairment of cell viability was determined by MTT method as described previously36.

In order to determine the effects of target compounds on t-BHP-induced cell death, the H9c2 cells were seeded in the 96-well plate (5×10³ cells/well) for 24 h growth, then pretreated with NXY-059, TMP, TBN (10, 300, 100, 300 μM) or edaravone (5 μM) for 2 h, respectively, before t-BHP was added for another 24 h incubation to induced cell damage. Cell viability was measured by MTT method as described previously36.

The H9c2 cells were cultured on cover slips (1×10⁴ cells) in 6-well plates and starved for 24 h, pretreated with 300 μM TBN or edaravone for 2 h, and then 150 μM t-BHP was added to the culture for another 24 h. Following fixation with 1% formalin for 10 min at room temperature, Hoechst 33342 (10 µg/ml) was added to the cover slips for 30 min under dark condition to stain the nuclei. Cells were observed and photographed under a fluorescence microscope (Olympus IX71, Shibuya-ku, Tokyo, Japan). Tetraplicate cells were prepared in each experimental condition. Apoptotic indices were determined by direct visualization and counting of a minimum of at least 200 cells from five randomly selected fields in each treatment, and expressed as a percentage of the total number of nuclei counted. The apoptotic index was calculated as the ratio of number of apoptotic cells to total cells counted in the field ×40 objective37.

The mitochondrial membrane potential (Δψm) was determined by a MitoProbe™ JC-1 assay kit (Invitrogen). Briefly, H9c2 cells seeded in Costar 96-well clear bottom black side microplate (5×10³ cells/well) were pretreated with or without 0.1, 1, 10, 100 μM target compounds for 2 h and then exposed with 150 μM t-BHP for another 1 h. Subsequently, the medium was replaced with 2 μM JC-1 and incubated for 30 min at 37 ^oC, 5% CO₂. The unbound dye was removed by washing with PBS twice, while the fluorescence of J-monomer and J-aggregates were quantified under area scan mode on a BioTek Synergy 4. The result was presented as the ratio of red/green fluorescence values relative to the untreated control. Carbonyl cyanide m-chlorophenylhydrazone (CCCP, 10 μM) served as a positive control for this experiment.

H9c2 cells seeded in Costar 96-well clear bottom black side microplate (5×10³ cells/well) were pretreated with or without 0.1, 1, 10, 100 μM target compound for 2 h, then 150 μM t-BHP was added to the culture for another 1 h. hydrogen peroxide scavenging activity was detected by employing 10 μM H₂DCF-DA incubated in the dark at 37 ^oC for 30 min. The relative fluorescence intensity was analyzed under area scan mode on BioTek Synergy 4.

MitoSOX Red mitochondrial superoxide indicator, a highly selective fluorogenic dye for detection of superoxide in the mitochondria of live cells, exhibits red fluorescence upon oxidation by superoxide 38. H9c2 cells seeded in Costar 96-well clear bottom black side microplate (5×10³ cells/well) were pretreated with or without 0.1, 1, 10, 100 μM target compound for 2 h, followed by exposing to 150 μM t-BHP for another 1 h. Subsequently, the culture medium was replaced by 5 μM MitoSOX Red in the dark at 37 ^oC for 30 min. Detection of oxidative MitoSOX Red was performed on a BioTek Synergy 4. The fluorescence intensity of control served at 100%, while the increased ratio of the relative fluorescence intensity presented as the amount of superoxide in the cells.

DEA NONOate can release NO in a controlled manner under physiological conditions nothing to do with the cell type and cellular enzyme 3940, while SIN-1 generates NO and O₂^– in equimolar amounts, and then forms the very reactive ONOO^–4142. DAF-FM DA and Dihydrorhodamine 123 (DHR 123) were used as selective fluorogenic dye for detection of NO 4344 and ONOO^–4546 in this experiment, respectively.

DEA NONOate and SIN-1 were dissolved in water at a final concentration of 50 mM each and store in small aliquots under - 80 ^oC until use. H9c2 cells seeded in Costar 96-well clear bottom black side microplate (5×10³ cells/well) were pretreated with or without 25, 50, 100, 200 μM target compounds for 2 h, then co-incubated with 1500 μM DEA NONOate or 1600 μM SIN-1 in serum-free DMEM for another 1 h. Following this, the medium was replaced with 5 μM DAF-FM DA or 10 μM DHR 123 in the dark at 37 ^oC for 30 min. After washing twice with PBS, the fluorescent intensity were quantified under area scan mode on a BioTek Synergy 4. The increased ratio of relative fluorescence intensity compared to control group presented as the encasement amount of RNS in the cells.

Each assay was carried out three times in order to determine the reproducibility. Data were expressed as mean ± standard deviation (SD) and statistical calculations were performed using prism 6.0 GraphPad software (GraphPad, San Diego, CA, USA). Comparisons between the different groups were performed by one-way analysis of variance (ANOVA) followed by Turkey’s test. p < 0.05 were considered significant.

The results showed that the effect on various radical-trapping among four different compounds was concentration-dependent from 5 to 320 μM. The radical-scavenging effects of TBN were more potent than TMP and NXY-059 against DPPH Figure 2A, O2– Figure 2B and ONOO^– Figure 2C at the same concentration, even close to edaravone. However, TBN was slightly better than TMP, while weaker than NXY-059 and edaravone in terms of OH-trapping activity Figure 2D.

As shown in Figure 3A, all the three inducers impaired H9c2 cells viability in a concentration dependent manner. A reduction of 50.5 ± 3.6% was observed with 150 μM t-BHP, 46.4 ± 5.5% with 1600 μM DEA NONOate, and 48.0 ± 3.5% with 1600 μM SIN-1. We chose the concentration of t-BHP 150 μM, DEA NONOate 1500 μM and SIN-1 1600 μM in our subsequent experiments. TBN significantly ameliorated the impairment of 150 μM t-BHP on H9c2 cells Figure 3B. Compared to TMP and NXY-059, TBN was more efficient than TMP and NXY-059 against t-BHP induced cell death. Edaravone (5 μM) served as a positive control, also mildly increased the cell viability.

Apoptotic cells were identified as chromatin condensation or nuclear fragmentation with nucleus exhibiting brightly stain of Hoechst 33342. Most nuclei in the control displayed uniform blue chromatin with organized structure, while cells stimulated with t-BHP were identified as those with a nuclei exhibiting brightly stain of condensed chromatin or nuclear fragments Figure 4A. NXY-059, TMP, TBN or edaravone pretreatment all markedly ameliorated t-BHP-induced apoptosis Figure 4B.

CCCP (10 μM) caused quick mitochondrial membrane depolarization, created a strong, single positive green fluorescence, served as a control. The pretreatment of all the four compounds could efficiently and concentration dependently reverse the Δψm loss caused by t-BHP. Compared to TMP and NXY-059, TBN was more potent in reversing the Δψm loss impaired by t-BHP Figure 5.

All four compounds significantly decreased the elevation of intracellular H₂O₂ and mitochondrial O₂^– stimulated by t-BHP in H9c2 cells. TBN was superior to NXY-059 and TMP against intracellular H₂O₂ and mitochondrial O₂^– induced by t-BHP at the same concentration, it was even better than the effect of edaravone at higher concentration Figure 6A and B.

In the NO-scavenging assay, the fluorescent intensity of the DEA NONOate treated alone group increased 29.3 ± 5.3% compared to the untreated control. Before exposure to DEA NONOate, 2 h pretreatment with all the four target compounds intensively decreased the NO level in H9c2 cells. TBN was weaker to edaravone but stronger to TMP and NXY-059 in NO-trapping activity Figure 7A. In the ONOO^– trapping activity, the fluorescent intensity of the SIN-1 treated alone group increased 287.6 ± 12.1% compared to the untreated control. Pretreatment with target compounds significantly and concentration-dependently decreased the relative fluorescence intensity of DHR 123. Similar to the NO-scavenging activity, TBN was also weaker to edaravone but stronger than TMP and NXY-059 in ONOO^–-trapping activity Figure 7B.

This work is partially supported by grants from China’s ‘12.5’ Innovative Drug Project (2012ZX09103101-055 to YW), the Science and Technology Project of Guangdong (2012A080201009 to YW, 2013A022100030 to PY and 2015A030313317 to ZZ), as well as the Science and Technology Program of Guangzhou (2014J4100097 to ZZ)