Tomatidine provides mitophagy‐independent neuroprotection after ischemic injury

Cerebral ischemia is one of the leading causes of human mortality and disability worldwide. The treatment of cerebral ischemia is refractory due to its short therapeutic window and lack of effective clinical drugs. Mitophagy, the autophagic elimination of damaged mitochondria, attenuates neuronal injury in cerebral ischemia, indicating the potential of mitophagy inducers as therapies for cerebral ischemia. We previously determined that, by enhancing autophagy flux, the steroidal alkaloid tomatidine can function as a neuroprotective agent against ischemic injury. However, its effects on mitophagy remain unknown. For this purpose, neuroblastoma cell lines Neuro‐2a and SH‐SY5Y were subjected to ischemic injury induced by oxygen–glucose deprivation/reperfusion (OGD/R) and then treated with tomatidine. OGD/R induced a general decrease of cellular contents, and this study revealed that tomatidine had no impact on mitophagy. In addition, tomatidine did not affect mitochondrial contents, including translocase of outer mitochondrial membrane 20 and voltage‐dependent anion channel 1, in either OGD/R‐treated or intact SH‐SY5H cells. Our results indicate that tomatidine exhibits its neuroprotective effects by enhancing autophagy, but in a potentially mitophagy‐independent manner, and provide insights for further investigation into its mechanism(s) and potential therapeutic use against cerebral ischemia.

Cerebral ischemia is one of the leading causes of human mortality and disability worldwide. The treatment of cerebral ischemia is refractory due to its short therapeutic window and lack of effective clinical drugs. Mitophagy, the autophagic elimination of damaged mitochondria, attenuates neuronal injury in cerebral ischemia, indicating the potential of mitophagy inducers as therapies for cerebral ischemia. We previously determined that, by enhancing autophagy flux, the steroidal alkaloid tomatidine can function as a neuroprotective agent against ischemic injury. However, its effects on mitophagy remain unknown. For this purpose, neuroblastoma cell lines Neuro-2a and SH-SY5Y were subjected to ischemic injury induced by oxygen-glucose deprivation/reperfusion (OGD/R) and then treated with tomatidine. OGD/R induced a general decrease of cellular contents, and this study revealed that tomatidine had no impact on mitophagy. In addition, tomatidine did not affect mitochondrial contents, including translocase of outer mitochondrial membrane 20 and voltage-dependent anion channel 1, in either OGD/R-treated or intact SH-SY5H cells. Our results indicate that tomatidine exhibits its neuroprotective effects by enhancing autophagy, but in a potentially mitophagy-independent manner, and provide insights for further investigation into its mechanism(s) and potential therapeutic use against cerebral ischemia.
Cerebral ischemia represents one of the refractory diseases worldwide because it causes mortality and permanent adult disability [1], while the underlying pathological mechanisms are not fully elucidated. Only until recent years has autophagy, the intracellular catabolic process that delivers cytosol and organelles to lysosomes for macromolecule turnover and recycling [2], drawn increasing attention in the context of cerebral ischemia, as well as a number of neurodegenerative disorders such as Parkinson's disease [3,4]. For cerebral ischemia, mounting evidence indicated that autophagy is activated in ischemic neurons and protects against neuronal injury in multiple cerebral ischemic models [5][6][7]. By eliminating damaged mitochondria (termed as mitophagy), autophagy prevents mitochondria-dependent apoptosis and attenuates ischemia-reperfusioninduced brain injury [3,8]. Alternatively, reinforced mitophagy exhibited additional neuroprotection against ischemic injury [9][10][11][12][13]. These findings suggest the therapeutic potential of mitophagy activation against cerebral ischemic injury.
Emerging studies indicated that tomatidine, a steroidal alkaloid abundant in the Solanaceae family, could activate autophagy in varying species [14,15]. Known as an anti-inflammatory agent by blocking nuclear factor of kappa light polypeptide gene enhancer in B cells and c-jun N-terminal kinase signaling [16], tomatidine also exhibits immunostimulatory and cardioprotective effects and the ability to prevent muscle atrophy [17,18]. To further investigate its therapeutic potential against cerebral ischemia, we previously reported tomatidine-induced neuroprotection against ischemic injury by enhancing autophagic flux in Neuro-2a (N2a) cells [19]. This was suggested to be achieved rather by autophagosome generation, but instead mainly through facilitating lysosomal degradation via activation of the transcription factor EB [19]. Even though it has the ability to promote lysosomal function, the effects of tomatidine on mitophagy were not determined in neuronal cells, although it was identified that tomatidine induces mitophagy in multiple other models [14,15]. This study aims to identify whether mitophagy is involved in the neuroprotection of tomatidine in ischemic neuronal cells.

Oxygen-glucose deprivation/reperfusion
To mimic ischemia/reperfusion-like conditions, an oxygenglucose deprivation/reperfusion (OGD/R) model was used as previously reported with minor modifications [20]. In brief, seeded cells were rinsed with PBS and glucose-free DMEM, placed in a sealed chamber (MIC-101; Billups-Rothenberg, Del Mar, CA, USA), loaded with oxygen-free atmosphere (95% N 2 + 5% CO 2 , 25 LÁmin À1 , 6 min), followed by incubation (4 h, 37°C) and refreshed with O 2and glucose-free DMEM (prebalanced in an O 2 -free chamber at 37°C). Subsequently, cells were refreshed with highglucose DMEM with or without treatment in standard atmosphere. In contrast, cells refreshed with high-glucose DMEM and incubated in standard atmosphere were used as an oxygen-glucose deprivation (OGD)-negative control. See specific values for different assays detailed in the following subsections.
After OGD/R, cells were observed under a confocal microscope [Leica TCS SP8, 63 9 (NA-1.4) oil-immersion; Shanghai, China] in standard atmosphere, and Z-stack images were acquired (resolution, 1024 9 1024 pixels; step size, 0.3-1 µm). The average area of mito-GFP puncta and Mander's overlap efficiency were measured and analyzed as described previously [21] using Image Pro-Plus software (version 7.0; Media Cybernetics, Rockville, MD, USA). At least 50 cells were analyzed per condition, and data were standardized on the negative control from three independent assays.

Quantification of cytotoxicity and cytoprotection
SH-SY5Y cells were seeded (5 9 10 3 cells per well, 100 lLÁper well, 96-well plate) and left to adhere overnight for WST-1, lactate dehydrogenase (LDH) or protein assay. After OGD, as described earlier in this chapter, reperfusion was performed by replacing media with high-glucose DMEM (100 lLÁper well) containing tomatidine (0, 1, 3 and 10 lM, 24 h). To indicate viability, we used NAD(P)H level as a surrogate marker measured by supplementing WST-1 reagent (10 lLÁper well, 1 h, 37°C, absorbance at 450 nm) according to the manufacturer's instructions. To indicate cellular membrane integrity, we used extracellular LDH level as a surrogate marker using an LDH release assay kit. Culture media were transferred to a separate plate (80 lL per well, 96-well plate) and supplemented with reagent mixture according to the manufacturer's instructions (40 lLÁper well, 1 h, room temperature, absorbance at 490 nm). After complete removal of media, cells were washed with PBS (100 lL) and permeabilized with radioimmunoprecipitation assay (RIPA) lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 0.5% SDC, 0.1% SDS, 1% Triton X-100, 1 mM EDTA, 20 mM NaF; 50 lL, 10 min). Protein concentration of cell lysates (20 lL) was determined using a BCA protein assay kit according to the manufacturer's instructions (140 lL per well, 30 min, 37°C, absorbance at 570 nm). A standard curve of BSA (0-2 mgÁmL À1 ) was used for protein quantification. OGD-negative controls followed by tomatidine treatment at the maximal test concentration (10 lM, 24 h) was used to indicate inherent toxicity of tomatidine. Data were standardized on the OGDnegative controls (100%) and expressed as mean AE standard error of mean (SEM) from at least three independent experiments with four parallel wells per experiment.

Statistical analysis
All data were collected and analyzed in a blinded manner and expressed as mean AE SEM. GRAPHPAD  of multiple comparisons, and P < 0.05 was considered statistically significant.

Results and Discussion
Tomatidine protects SH-SY5Y cells against ischemic injury without cytotoxicity In this study, we aimed to deepen the understanding of tomatidine neuroprotective mechanism(s) against ischemic neuronal injury. First, we investigate the toxicity of tomatidine. In our test model, OGD/R significantly reduced the rate of cell proliferation, whereas tomatidine treatment had no effect. Moreover, as the dose of tomatidine (1, 3 and 10 µM) increased, the cell proliferation rate increased (Fig. 1A). Consistent with these results, we used LDH to detect cell viability, and OGD/R significantly reduced the rate of cell viability, whereas tomatidine treatment reverses that phenomenon (Fig. 1B).

Tomatidine does not reinforce mitophagy in ischemic N2a cells
Based on our previous report that tomatidine induces autophagy [19] and that tomatidine induces mitophagy in mammal cells and Caenorhabditis elegans [14], tomatidine was further assessed for its inducing potential of mitophagy. On the accumulation of mCherry-LC3 in autophagosomes and mito-GFP in mitochondria, autophagosomes appear as red puncta compared with mitochondria as green puncta. As such, the colocalization of autophagosomes and mitochondria appear as yellow puncta that is merged from the two fluorescences. Therefore, the number of yellow puncta indicates the level of mitophagy. In our test model, OGD/ R significantly reduced the area of mito-GFP-labeled mitochondria in N2a cells (P < 0.01), which suggested mitophagy activation (Fig. 2A). Nevertheless, no significant differences of the area of mito-GFP-labeled mitochondria were observed between OGD/R-treated and tomatidine-treated groups (1, 3 and 10 µM) (Fig. 2B). Consistent with these results, OGD/R significantly increased Mander's overlap coefficient of mito-GFP-labeled mitochondria and mCherry-LC3-labeled LC3 puncta in N2a cells (P < 0.01), suggesting mitophagy activation (Fig. 2C). Again, no significant differences of the Mander's overlap coefficient was observed between OGD/R-treated and tomatidinetreated groups (1, 3 and 10 µM) (Fig. 2C).

Tomatidine does not affect mitochondrial content of SH-SY5Y cells
To further verify whether tomatidine affects mitochondrial contents, we further determined the expression of the widely used mitochondrial markers TOM20 and VDAC1 OGD/R-treated and intact SH-SY5Y cells by western blot. The oxidative phosphorylation uncoupler carbonyl cyanide 3-chlorophenylhydrazone was used to induce mitophagy as a positive control. The lysosome inhibitor CQ was used to inhibit autophagy by altering lysosomal pH upon the diffusion into the lysosome and conversion into the diprotonated form [23]. The results showed that carbonyl cyanide 3chlorophenylhydrazone treatment and OGD/R significantly reduced TOM20 and VDAC1 expression, whereas neither of the two markers was further reduced by tomatidine at all test concentrations (Fig. 3A,B). Similarly, none of the test concentrations of tomatidine altered TOM20 and VDAC1 expression even in intact or CQ-treated SH-SY5Y cells (Fig. 3C,  D). Other blots were included in Fig. S1.

Conclusions
The development of effective therapeutics against cerebral ischemia has been challenging worldwide. As a novel therapeutic candidate, tomatidine is able to cross the blood-brain barrier identified with neuroprotection from ischemic injury in the context of neurodegenerative disorders such as Alzheimer's disease [22,23]. Studies using varying tissues and/or organs provide mounting evidence and indicate that tomatidineinduced cytoprotection against ischemic injuries was associated with antioxidant, anti-inflammatory and anticancer activities [16,17,22]. Furthermore, we reported that tomatidine induced autophagic flux against ischemic injury by activating lysosomal functions in vitro [19], suggesting the pharmacological basis of tomatidine and its use as a promising therapy against cerebral ischemia. Based on our knowledge that autophagy confers neuroprotection by promoting mitophagy [3], this study aimed to further characterize the ability of tomatidine to induce mitophagy using the same test model. Although our test model successfully mimicked ischemia-reperfusion-like conditions, to our surprise, tomatidine did not reinforce mitophagy in differentiated N2a cells, as supported by analysis of mitochondria alone and the colocalization of mitochondria and autophagosomes. Furthermore, by quantifying the most representative mitochondrial markers VDAC1 and TOMM20, our results suggested that tomatidine was also devoid of impact on mitochondrial contents in neither ischemic nor intact cells, despite inhibition of autophagy by CQ. These negative observations could be attributed by the high level of reactive oxygen species in our test model, which was also reported in ischemia-reperfusion-treated neurons [9]. Alternatively, the effects of tomatidine on mitochondria could be exhibited in a delayed manner and specific timescale in neurons [24] that may not be detectable in the model used for this study, which will require further in-depth investigations.
Taken together, this study suggests that mitochondria may not be associated with tomatidine-conferred neuroprotection against ischemic injury and provided insights into deeper understanding of the mechanism underlying its neuroprotection. Although the causality of tomatidine on mitophagy remains to be elucidated in future investigations, tomatidine significantly induces neuroprotection against ischemic injury by enhancing autophagy. Together with the favorable bioactivities and safety as a natural product-derived compound, tomatidine exhibits its promising therapeutic potential against cerebral ischemia.