Abstract
Tetrahydroxy stilbene glucoside (TSG) from Polygonum multiflorum exerts neuroprotective effects after ischemic stroke. We explored whether TSG improved ischemic stroke injury via PTEN-induced kinase 1 (PINK1)/Parkin-mediated mitophagy. Oxygen glucose deprivation/reoxygenation (OGD/R) in vitro model and middle cerebral artery occlusion (MCAO) rat model were established. Cerebral injury was assessed by neurological score, hematoxylin and eosin staining, 2,3,5-triphenyltetrazolium chloride staining, and brain water content. Apoptosis, cell viability, and mitochondrial membrane potential were assessed by flow cytometry, cell counting kit-8, and JC-1 staining, respectively. Colocalization of LC3-labeled autophagosomes with lysosome-associated membrane glycoprotein 2-labeled lysosomes or translocase of outer mitochondrial membrane 20-labeled mitochondria was observed with fluorescence microscopy. The ubiquitination level was determined using ubiquitination assay. The interaction between molecules was validated by coimmunoprecipitation and glutathione S-transferase pull-down. We found that TSG promoted mitophagy and improved cerebral ischemia/reperfusion damage in MCAO rats. In OGD/R-subjected neurons, TSG promoted mitophagy, repressed neuronal apoptosis, upregulated Y-box binding protein-1 (YBX1), and activated PINK1/Parkin signaling. TSG upregulated ubiquitin-specific peptidase 10 (USP10) to elevate YBX1 protein. Furthermore, USP10 inhibited ubiquitination-dependent YBX1 degradation. USP10 overexpression activated PINK1/Parkin signaling and promoted mitophagy, which were reversed by YBX1 knockdown. Moreover, TSG upregulated USP10 to promote mitophagy and inhibited neuronal apoptosis. Collectively, TSG facilitated PINK1/Parkin pathway-mediated mitophagy by upregulating USP10/YBX1 axis to ameliorate ischemic stroke.
Significance Statement
Ischemic stroke is one of the leading causes of disability and death worldwide. Previous studies have demonstrated a neuroprotective role of tetrahydroxy stilbene glucoside (TSG) in ischemic stroke, while the underlying mechanism is still not fully understood. Here, this study confirmed that TSG relieved cerebral ischemia/reperfusion injury in vivo and in vitro via facilitated PINK1/Parkin-mediated mitophagy. In addition, we further identified the molecular mechanism by which TSG regulates mitochondrial autophagy. Our study provided new insights into the protective role of TSG in ischemic stroke via regulating mitophagy.
Introduction
Ischemic stroke, induced by a sudden occlusion of the cerebral artery, is one of the leading causes of disability and death around the world. Recent therapeutic method includes rapid reperfusion using mechanical and pharmacological thrombolysis, both of which are time-sensitive. Furthermore, the recanalization of blood flow always elicits ischemia/reperfusion (I/R) damage (Tuo et al., 2022). Studies have confirmed that neuronal injury is an important pathogenesis of ischemic stroke (Zhao et al., 2022). Therefore, finding novel therapeutic medicines to rescue neuronal damage is significant for protecting the brain following stroke.
Mitophagy, as a selective autophagy, is a crucial biological process to remove redundant or dysfunctional mitochondria, thus maintaining energy metabolism (Shao et al., 2020). Mitophagy is a benefit for the function maintenance and survival of neurons, while excessive mitophagy can also cause cell death under certain circumstances; therefore, mitophagy plays double-edged roles in ischemic stroke (Song et al., 2022). PTEN-induced kinase 1 (PINK1)/Parkin signaling is an important mitophagy signaling in neurological diseases (J. Li et al., 2023a). PINK1 modulates the translocation of Parkin in damaged mitochondria and drives their elimination through mitophagy (Quinn et al., 2020). Activating mitophagy mediated by PINK1/Parkin signaling can improve neuronal damage caused by cerebral I/R (M. Wu et al., 2021, Mao et al., 2022). Thus, PINK1/Parkin-mediated mitophagy is an important therapeutic target against ischemic stroke.
Tetrahydroxy stilbene glucoside (TSG) is purified from medicinal herb Polygonum multiflorum and has been verified to exert powerful free radical scavenging, antioxidative and antiapoptotic activities (J. Wu et al., 2017). A current study has shown that TSG eliminated brain injury in a mice model of bilateral common carotid artery (CCA) occlusion (M. Li et al., 2022b). The neuroprotective effect of TSG has also been validated on a cell model of oxygen glucose deprivation/reoxygenation (OGD/R) and a rat I/R model of middle cerebral artery occlusion (MCAO; Ruan et al., 2021). TSG promoted neuronal mitophagy and inhibited apoptosis in ischemic stroke (Y. Li et al., 2021). Moreover, TSG promoted mitophagy via regulating PINK1/Parkin pathway to alleviate inflammatory damage (Gao et al., 2020). Nevertheless, the specific molecular mechanism through which TSG regulates mitophagy in ischemic stroke is still unclear.
Ubiquitin-specific peptidase 10 (USP10), one member of the ubiquitin-specific protease family, is a deubiquitination enzyme implicated in numerous essential cellular activities (Ye et al., 2022). Previous study has reported that USP10 suppressed apoptosis and inflammation to protect against cerebral ischemia damage (L. Wang et al., 2019b). USP10 alleviated neuroinflammation in ischemic stroke via repressing nuclear factor-κB signaling (C. Xie et al., 2023). Furthermore, USP10 was reported to participate in regulating autophagy in neurodegenerative diseases (Bhattacharya et al., 2020). Resveratrol, a TSG analog, has been reported to enhance USP10-mediated deubiquitination of p53 (Oi et al., 2015). Does TSG also play a role in promoting deubiquitination function of USP10?
Here, the molecular mechanism of TSG regulating mitophagy and neuronal injury following ischemic stroke was explored through the in vitro and in vivo experiments. We hypothesized that TSG upregulated USP10 to improve brain injury in ischemic stroke via promoting PINK1/Parkin-mediated mitophagy. Our study provided a better understanding of the molecular mechanism of TSG as a treatment for ischemic stroke.
Materials and Methods
PC12 cell culture and OGD/R
Rat PC12 cells were received from American Type Culture Collection. Cells were cultured in DMEM (Invitrogen) supplemented with 10% FBS, penicillin (100 U/mL), and streptomycin (100 U/mL), in a humidified incubator with 5% CO2 at 37°C.
PC12 cell subjected to OGD/R was constructed to mimic cerebral ischemic conditions. In brief, cells were maintained in a hypoxia chamber in glucose-free DMEM under 95% N2 and 5% CO2 for 2 h (Deng et al., 2022). Then cells were moved to normal culture conditions. After 24 h cultivation, cells were applied for the following assays.
For the treatment of TSG, cells were added with TSG (2, 4, or 8 μM, Yuanbaofeng Medical Technology) for 24 h incubation at the time of reperfusion. Cells were added with the proteasome inhibitor MG132 (25 μM, Sigma-Aldrich) or protein synthesis inhibitor cycloheximide (CHX; 50 μg/ml; Cell Signaling Technology, CST) and incubated for the indicated time periods.
MCAO model construction and drug treatment
A total of 74 adult Sprague Dawley rats (male, 230–250 g, 6–8 weeks) were obtained from the Hunan University of Medicine. Animals were given water and food ad libitum and housed in a room with a 12 h light/dark cycle at 22°C. All animal procedures were performed in accordance with the National Institutes of Health Guidelines for Animal Research animal care committee's regulations and approved by our hospital.
To establish MCAO model, 54 animals were anesthetized with 5% isoflurane, followed by maintaining with 1.5% isoflurane. Warming pad was applied to maintain the body temperature at 37°C. After a ventral midline incision, the left CCA, external carotid artery (ECA) and internal carotid artery (ICA) were isolated. Then a monofilament nylon suture coated with silicone was gently inserted into the ICA through the ECA. After 2 h, the filament was withdrawn to allow reperfusion. Rats in the sham group received the same operation without artery occlusion. Next, the rat's neck was subcutaneously injected with TSG (10, 20, or 40 mg/kg) dissolved in saline at 4 h after reperfusion and then subcutaneously administered with TSG for 7 d, twice a day (Ruan et al., 2021). The same volume of saline was injected into the rats in model and sham groups. Nevertheless, four of the animals died during the experiment, which were excluded. After experiment, rats were killed, and brain tissues were immediately collected for subsequent detections. According to experimental requirements, samples were divided into four cohorts, which were used for neurological function/2,3,5-triphenyltetrazolium chloride (TTC) measurements, brain water content, pathological examination, and biochemical test, respectively.
Cell transfection
Specific shRNA targeting USP10 (sh-USP10) or Y-box binding protein-1 (YBX1; sh-YBX1) was employed to knock down the USP10 or YBX1 gene, respectively. To overexpress USP10, we subcloned the pcDNA3.1 vector with the full length of USP10 gene for establishing pcDNA3.1-USP10. The plasmids were obtained from GeneChem. Scramble shRNA (sh-NC) and oe-NC vectors were used as negative controls. Then, Lipofectamine 2000 (Invitrogen) was utilized to transfect these vectors into the cells.
Neurological score
A five-point scale was used to quantify the neurological deficits as previously described (Ruan et al., 2021): 4, unable to move autonomously and unconsciousness, serious; 3, lean to the left, severe; 2, circling to the left, moderate; 1, unable to extend the left forepaw, slight; and 0, no observable deficit, normal. Two evaluators blinded to the experiment design were responsible for assessing the neurological function. For each rat, the score was calculated within 60 s and repeated three times. Then Kruskal–Wallis H test was used to analyze the scores (D. Yang et al., 2019).
Cerebral infarct volume
Brain tissues were cut into coronal sections (2 mm thickness). Next, the sections were stained for 0.5 h at 37°C with 2% TTC solution (Thermo Fisher Scientific). Following immersed with paraformaldehyde (4%) overnight, the Image-Pro Plus software was applied to obtain the infarct sizes of the brain slices. Infarct volume (%) = [(contralateral hemisphere volume) − (ipsilateral undamaged volume)] × 100 / (contralateral hemisphere volume).
Brain water content
Brain tissues were gathered from the rats after indicated treatments. Next, the dry weight approach was applied to determine the brain water content. In brief, brains collected from the rats were sectioned into coronal slices (5 mm thickness) and divided into the ischemic and nonischemic hemispheres. Then an analytical balance was employed to measure the wet weight of the ischemic hemisphere samples. Afterward, the brain tissue was maintained for 24 h at 105°C in an oven. Then an analytical balance was also used to determine the dry weight of the brain tissue (wet weight − dry weight) × 100 / wet weight = brain water content (%) (Hatashita et al., 1988).
Hematoxylin and eosin staining
Brain tissues were incubated overnight at 4°C with paraformaldehyde (4%), before being sectioned into paraffin slices (5 μm). Later, following deparaffinized in xylene and rehydrated in gradient ethanol, slices were then stained by hematoxylin (2 g/l) for 5 min and then washed by distilled water. Slices were then incubated with eosin solution (1%) for 2 min, followed by washing with distilled water. Afterward, samples were sealed with neutral resin. Light microscopy (Olympus, Tokyo, Japan) was used to obtain the pictures of neuronal morphological alterations in the cerebral cortex.
Immunohistochemistry
Brain tissues were placed in phosphate-buffered saline (PBS), followed by sectioning into frozen slices (5 µm). After incubation for 0.5 h in H2O2 (3%) for blocking peroxidase activity, slices were treated with 3% BSA at 37°C for 0.5 h. Next, they were immersed at 4°C in anti-LC3 (PA1-46286, 1:200, Invitrogen) overnight. Following incubation, slices were washed three times, prior to immersion for 1 h at 37°C in secondary antibody. Finally, after staining with diaminobenzidine and hematoxylin, sections were photographed by microscopy (BX51, Olympus).
RT-qPCR
TRIzol reagent (Takara Bio) was used to extract the total RNA of brain tissues or cells. NanoDrop Spectrophotometer (BioDrop) was applied for measuring RNA concentration and purity. Subsequently, PrimeScript RT reagent kit (Takara Bio) was applied to operate reverse transcription reaction for obtaining cDNA. ABI 7300 Real-Time PCR System was used to perform quantitative RT-PCR with SYBR Green PCR premix (Thermo Fisher Scientific). The comparative threshold cycle method (2-DeltaDeltaCt) was used to calculate RNA expression levels normalized to GAPDH. The primers are displayed in Table 1.
The primer sequences used in RT-qPCR
Western blot
Total proteins were extracted from brain tissues and cells with RIPA lysis buffer containing protease inhibitors and phosphatase inhibitors (Sigma-Aldrich). For the protein analysis of mitochondria, mitochondria were isolated from brain tissues (C3606, Beyotime) and cells (C3601, Beyotime) based on the kit instructions. The pellet containing separated mitochondria was resuspended in mitochondrial lysate with phenylmethanesulfonyl fluoride (1 mM). BCA assay (Solarbio) was used to determine protein concentrations. Next, equal amounts (30 μg) of protein samples were separated with SDS–PAGE, followed by movement to PVDF membranes. After blocking with skimmed milk (5%) for 1 h, membranes were immersed at 4°C overnight in primary antibodies against LC3II/I (PA1-46286, 1:1,000, Invitrogen), p62 (ab109012, 1:1,000, Abcam), PINK1 (PA1-16604, 1:500, Invitrogen), Parkin (702785, 1:200, Invitrogen), translocase of outer mitochondrial membrane 20 (Tomm20, ab186735, 1:1,000, Abcam), YBX1 (ab76149, 1:1,000, Abcam), GAPDH (ab181602, 1:10,000, Abcam), cytochrome C oxidase subunit IV (COX IV; ab202554, 1:2,000, Abcam), and USP10 (PA5-117656, 1:500, Invitrogen). Membranes were immersed for 1 h in secondary antibodies combined with horseradish peroxidase, followed by visualization with enhanced chemiluminescence reagent (Thermo Fisher Scientific). COX IV and GAPDH were used as internal references for mitochondrial and total fractions, respectively.
Cell counting kit-8 (CCK-8) assay
The viability of PC12 cells was assessed using CCK-8 assay (C0037, Beyotime). Cells (2 × 103 cells/100 μl) were planted to a 96-well plate, followed by the indicated treatments. Next, CCK-8 reagent (10 μl) was added into the cells for 2 h incubation at 37°C. Lastly, a microplate reader (Thermo Fisher Scientific) was used to detect the absorbances at 450 nm.
Flow cytometry
Apoptosis in PC12 cells was detected by using Annexin V-FITC apoptosis detection kit (Beyotime). In brief, after indicated treatments, Annexin V-FITC (5 μl) and propidium iodide (10 μl) were added into the cells for 20 min incubation in darkness. Flow cytometer (ACEA Biosciences) was used to detect the apoptotic cells.
Mitochondria membrane potential (MMP)
The alteration in MMP was detected by staining with JC-1 dye (Beyotime). In brief, cells were planted in 12-well plates. Following the corresponding treatment, 2 mmol/l JC-1 was added into the cells, before being incubated for 20 min with 5% CO2 at 37°C in an incubator. Next, after being rinsed in PBS, cells were observed under a fluorescence microscope, and the fluorescence images were obtained. An emission wavelength of 530 or 590 nm and an excitation wavelength of 490 or 525 nm were used for detecting JC-1 monomers or aggregates, respectively. A fluorescence spectrophotometer was applied to examine the fluorescent intensity. MMP was reflected by calculating the percentage of JC-1 aggregates (red) to monomer (green).
Immunofluorescence
Brain sections were incubated at 4°C overnight with primary antibodies against NeuN (ab104224, Abcam) and YBX1 (ab76149, Abcam), following fixation, embedding, slicing, permeabilization, and blocking. Following incubation at 37°C for 1 h with secondary antibodies, sections were stained by DAPI. Fluorescence pictures were obtained utilizing fluorescence microscope (Olympus). PC12 cells in vitro were coincubated in primary antibodies against lysosome-associated membrane glycoprotein 2 (LAMP2, L0668, Sigma-Aldrich) and LC3 (83506, CST) or Tomm20 (ab186735, Abcam) and LC3, and other procedures were the same as above.
Ubiquitination assay
HA-Ub plasmid (2 μg) was transfected into cells. After 48 h, cells were pretreated for 6 h with 30 μM of MG132, followed by resuspension in IP lysis buffer. Samples were incubated for 5 min at 4°C before being sonicated on ice. Following centrifugation for 15 min at 12,000 rpm, supernatant was incubated overnight at 4°C with anti-YBX1 (ab76149, Abcam) and then immersed at 4°C with protein A/G-Sepharose beads for another 2 h. Immunoprecipitated protein complexes on beads were subsequently eluted for 10 min at 95°C. Next, they were examined using Western blot with anti-YBX1 and anti-HA (H6908, Sigma-Aldrich).
Coimmunoprecipitation
Cell lysates from PC12 cells were centrifugated at 4°C for 10 min at 13,000 × g. Next, after precleared for 2 h in protein A/G agarose beads (20 μl), the supernatants were immersed at 4°C in anti-USP10 (PA5-117656, 1:200, Invitrogen) or IgG (ab172730, 1:100, Abcam) overnight, prior to incubation for 2 h at 4°C in protein A/G plus agarose. The immunoprecipitated complexes were eluted through boiling with SDS loading buffer, after being washed in IP buffer. Next, primary antibodies including anti-USP10 and anti-YBX1 were used for the immunoblot analyses.
Glutathione S-transferase (GST) pull-down
Recombinant glutathione transferase-USP10 (GST-USP10) was prepared in E. coli for GST pull-down assay. GST proteins were purified using glutathione Sepharose 4B microspheres (Solarbio) and then incubated with PC12 cells for 2 h. Then GST-USP10 fusion proteins or GST (Proteintech) on microspheres were added to the total lysate of PC12 cells and incubated overnight at 4°C on a rolling shaker. The bound proteins were eluted from the beads. Anti-USP10 and anti-YBX1 were used to carry out Western blot analysis.
Statistical analysis
GraphPad Prism 7.0 was applied for performing statistical analyses. Each experiment was repeated at least three times. All data were exhibited as mean ± SD. One-way ANOVA followed by the Tukey's post hoc test was used to carry out the comparison of multiple groups, while Student's t test was used to operate the comparison between two groups. p < 0.05 was considered as significance.
Results
TSG improved cerebral I/R injury and promoted mitophagy in MCAO rats
First, we explored the neuroprotective effect of TSG on ischemic stroke. Male SD rats were used to construct MCAO model. Following 2 h of occlusion, the filament was withdrawn for reperfusion. Rats were then treated with TSG (10, 20, or 40 mg/kg) dissolved in saline via subcutaneous injection at 4 h after reperfusion and then for another 6 d (twice a day). Besides, rats in model and sham groups were administrated with the same volume of saline. A timeline image of the experimental design is showed in Figure 1A. Compared with the sham group, MCAO rats exhibited significant cerebral infarction, while treatment with TSG (10, 20, or 40 mg/kg) decreased the infarction volume (Fig. 1B). Besides, in a dose-dependent manner, TSG treatment significantly reduced the neurological scores after MCAO (Fig. 1C), indicating that TSG alleviated MCAO-induced neurological deficit. According to the above data, 40 mg/kg of TSG had the strongest effects, which was selected for subsequent experiments. Hematoxylin and eosin staining showed that MCAO caused neuronal vacuolation in the infarct area, whereas TSG ameliorated the damage of cortical neurons (Fig. 1D). MCAO caused significant brain edema, while TSG treatment reduced brain water content in MCAO rats (Fig. 1E). Immunofluorescence showed that MCAO caused decreases in the immunofluorescence intensity of YBX1 in the cerebral cortex, which were increased after TSG treatment (Fig. 1F). We then explored whether mitophagy was implicated in the protective impact of TSG. Immunohistochemistry showed that MCAO downregulated the expression of LC3 (an autophagy marker) in brain tissues, while TSG treatment upregulated LC3 expression (Fig. 1G). It was found that MCAO downregulated LC3 and upregulated p62 protein levels in brain tissues, which was reversed by TSG treatment (Fig. 1H). It is known that PINK1/Parkin signaling is activated during mitophagy, and PINK1 recruits Parkin to the mitochondria surface to activate Parkin E3 ubiquitin ligase (Nguyen et al., 2016). Parkin can label mitochondrial components such as Tomm20 through ubiquitination, which results in the degradation of these mitochondrial proteins by proteasome (Stolz et al., 2014). To further verify the effect of TSG on mitophagy, we isolated the mitochondria, and the results displayed that MCAO decreased mitochondrial Parkin and PINK1 protein levels but upregulated mitochondrial Tomm20 in brain tissues, while TSG abolished the effects caused by MCAO (Fig. 1H). These results indicated that TSG ameliorated MCAO-induced cerebral ischemic injury and facilitated mitophagy.
TSG improved cerebral I/R damage and promoted mitophagy in MCAO rats. A, Schematic diagram of the experimental design. B,C, Rats were used to construct MCAO model, followed by subcutaneous injection of TSG (10, 20, or 40 mg/kg). B, TTC staining for determining infarction volume. C, Neurological scores. D–H, Rats were subjected to MCAO, followed by treatment with TSG (40 mg/kg). D, Hematoxylin and eosin staining for observing the morphological changes of cortical cells. E, Brain edema determined with the ratio of dry/wet weight. F, Immunofluorescence for detecting the YBX1 expression. G, Immunohistochemistry for detection of LC3 expression. H, Western bolt for detecting mitophagy-associated proteins (LC3II/I, p62, PINK1, Parkin) and Tomm20 (a mitochondrial marker). n = 5 per group. *p < 0.05; **p < 0.01; ***p < 0.001. Data are displayed as mean ± SD.
TSG promoted mitophagy and inhibited apoptosis by upregulating YBX1 and activating PINK1/Parkin signaling in OGD/R-induced neurons
We then investigated the potential mechanism underlying the neuroprotection of TSG. PC12 cells were treated with TSG (0, 2, 4, 8 μM). The data showed that no obvious alteration was observed in cell viability following the treatments of different doses of TSG, suggesting that TSG had no toxic impact on PC12 cells in the range of 0–8 μM (Fig. 2A). Then PC12 cells were exposed to OGD/R, followed by the treatment of TSG (2, 4, or 8 μM). Compared with the control group, cell viability was decreased by OGD/R, while TSG treatment elevated cell viability in a dose-dependent manner (Fig. 2B). OGD/R promoted apoptosis of PC12 cells, while TSG repressed apoptosis in a dose-dependent manner (Fig. 2C). Mitophagy-mediated removal of injured mitochondria is critical for cell homeostasis and mitochondrial function (Tian et al., 2022). Here, we examined mitochondrial function by assessing MMP using JC-1 staining. PC12 cells were exposed to OGD/R, prior to the addition of TSG (8 μM). Compared with the control group, OGD/R decreased MMP, whereas TSG treatment increased MMP (Fig. 2D). Moreover, OGD/R decreased the colocalization of Tomm20 and LC3 in PC12 cells, which was increased after the treatment of TSG (Fig. 2E). PINK1 activation induced by mitochondrial injury motivates phosphorylation-dependent Parkin activation and ubiquitin-dependent mitochondrial removal through mitophagy (Agarwal and Muqit, 2022). Next, whether PINK1/Parkin pathway participated in TSG-caused mitophagy was explored through measuring the expressions of mitophagy-related proteins. Compared with the control group, OGD/R decreased LC3II/I and YBX1, increased p62, downregulated mitochondrial PINK1 and Parkin, and increased mitochondrial Tomm20 in PC12 cells; however, TSG treatment overturned the protein expressions in OGD/R-subjected PC12 cells (Fig. 2F). These outcomes demonstrated that TSG could facilitated mitophagy and suppressed apoptosis in OGD/R-induced neurons through upregulating YBX1 and activating PINK1/Parkin signaling.
TSG promoted mitophagy and inhibited apoptosis by upregulating YBX1 and activating PINK1/Parkin signaling in OGD/R-induced neurons. A, Cell viability was determined with CCK-8 in PC12 cells treated with TSG (0, 2, 4, or 8 μM). B, C, PC12 cells were subjected to OGD/R, followed by the treatment of TSG (2, 4, or 8 μM). B, CCK-8 for measuring cell viability. C, Apoptosis detected using flow cytometry. D–F, PC12 cells were exposed to OGD/R, prior to TSG (8 μM) incubation. D, JC-1 staining for detecting MMP. E, Immunofluorescence for measuring the colocalization of LC3 and Tomm20. F, Western blot for detecting autophagy-associated proteins (LC3II/I, p62, PINK1, Parkin), Tomm20 (a mitochondrial marker), and YBX1. n = 3. *p < 0.05; **p < 0.01; ***p < 0.001. Data are displayed as mean ± SD.
TSG increased YBX1 protein expression in OGD/R-subjected neurons via upregulating USP10
Next, we examined the possible mechanisms by which TSG upregulates YBX1. YBX1 protein levels in PC12 cells after the treatment of protein synthesis inhibitor (CHX) were detected using Western blot. Compared with the control group, YBX1 protein degradation was rapid in OGD/R-exposed neurons, while TSG treatment slowed down the degradation of YBX1 protein (Fig. 3A). Furthermore, YBX1 protein levels in PC12 cells after the treatment of proteasome inhibitor (MG132) were detected using Western blot. The YBX1 protein level was downregulated in OGD/R-exposed cells, while it was significantly upregulated after TSG treatment; however, MG132 treatment further elevated the YBX1 protein level (Fig. 3B). The above results indicated that TSG could stabilize OGD/R-induced degradation of YBX1 protein. Moreover, ubiquitination assay showed that OGD/R promoted the ubiquitination of YBX1 in PC12 cells, which was inhibited after TSG treatment (Fig. 3C). Compared with the control group, the USP10 protein level was decreased in OGD/R-exposed cells, whereas TSG treatment reversed OGD/R-induced decrease of USP10 (Fig. 3D). In PC12 cells, the direct interaction between USP10 and YBX1 was validated using coimmunoprecipitation (Fig. 3E). Moreover, GST pull-down further confirmed the relationship of USP10 and YBX1 (Fig. 3F). Subsequently, PC12 cells were transfected with sh-USP10, and the silencing efficiency was validated, as indicated by the data that sh-USP10 transfection downregulated USP10 at mRNA and protein levels (Fig. 3G,H). Then PC12 cells that transfected with sh-USP10 were subjected to OGD/R and treated with TSG. In OGD/R-exposed cells, TSG treatment upregulated USP10 at mRNA and protein levels; however, USP10 downregulation reversed this effect induced by TSG (Fig. 3I,J). Moreover, in OGD/R-subjected cells, knockdown of USP10 reversed the upregulated the YBX1 protein level caused by TSG (Fig. 3J). Collectively, TSG inhibited YBX1 ubiquitination degradation by upregulating USP10 in OGD/R-exposed neurons.
TSG increased YBX1 protein expression in OGD/R-subjected neurons via upregulating USP10. A–D, PC12 cells were exposed to OGD/R, prior to TSG (8 μM) incubation. A, Western blot for determining the YBX1 protein level after the treatment of CHX. B, Western blot for testing the YBX1 protein level after the treatment of MG132. C, Detection of the YBX1 ubiquitination level. D, Western blot for detecting the USP10 protein level. E, F, Coimmunoprecipitation (E) and GST pull-down (F) for determining the interaction of USP10 and YBX1. G, H, sh-USP10 was transfected into PC12 cells. RT-qPCR and Western blot for detecting USP10 expressions. I, J, PC12 cells were transfected by sh-USP10, before being exposed to OGD/R or treated with TSG. I, RT-qPCR for determining USP10 mRNA expression. J, Western blot for testing USP10 and YBX1 protein levels. n = 3. *p < 0.05; **p < 0.01; ***p < 0.001. Data are displayed as mean ± SD.
USP10 overexpression inhibited ubiquitination-dependent degradation of YBX1 in OGD/R-subjected neurons
Our study further examined how USP10 regulated YBX1 expression in PC12 cells. Knockdown of USP10 in PC12 cells did not affect the YBX1 mRNA level (Fig. 4A); however, USP10 silencing downregulated the protein level of YBX1 (Fig. 4B). With the treatment of CHX, knockdown of USP10 accelerated the degradation of YBX1 protein (Fig. 4C). Furthermore, compared with the sh-NC group, knockdown of USP10 reduced the YBX1 protein level, while YBX1 protein degradation induced by USP10 knockdown was repressed in the presence of MG132 (Fig. 4D). Meanwhile, ubiquitination assay exhibited that knockdown of USP10 promoted the ubiquitination of YBX1 in neurons (Fig. 4E). Subsequently, cells were transfected with pcDNA 3.1-USP10, and the transfection efficiency was validated. As shown in Figure 4, F and G, pcDNA 3.1-USP10 upregulated USP10 mRNA and protein levels (Fig. 4F,G). Then PC12 cells were transfected with pcDNA 3.1-USP10, followed by OGD/R exposure. OGD/R downregulated USP10 at mRNA and protein levels, while overexpression of USP10 upregulated USP10 (Fig. 4H,I); besides, OGD/R declined the protein level of YBX1, whereas USP10 overexpression increased the YBX1 protein level (Fig. 4I). Moreover, OGD/R promoted the ubiquitination of YBX1, which was reversed after overexpression of USP10 (Fig. 4J). The above indicated that in OGD/R-exposed neurons, USP10 suppressed YBX1 protein ubiquitination degradation.
USP10 overexpression inhibited ubiquitination-dependent degradation of YBX1 in OGD/R-subjected neurons. A–E, sh-USP10 was transfected into PC12 cells. A, RT-qPCR for determining YBX1 mRNA expression. B, Western blot for detecting the YBX1 protein level. C, Western blot for testing the YBX1 protein level after treatment with CHX. D, Western blot for determining the YBX1 protein level after MG132 treatment. E, The ubiquitination level of YBX1. F, G, pcDNA 3.1-USP10 was transfected into PC12 cells. RT-qPCR and Western blot were applied for measuring USP10 levels. H–J, pcDNA-3.1-USP10 was transfected into PC12 cells, followed by OGD/R exposure. H, RT-qPCR for detecting the USP10 mRNA level. I, Western blot for determining USP10 and YBX1 protein levels. J, The ubiquitination level of YBX1. n = 3. *p < 0.05; **p < 0.01; ***p < 0.001. Data are displayed as mean ± SD.
Knockdown of YBX1 reversed the effects of USP10 overexpression on PINK1/Parkin signaling and mitophagy in OGD/R-exposed neurons
Then, we investigated whether USP10 affected PINK1/Parkin signaling and neuronal mitophagy via YBX1. sh-YBX1 transfection in PC12 cells reduced YBX1 mRNA and protein levels (Fig. 5A,B), indicating successful knockdown of YBX1. Subsequently, PC12 cells were transfected with sh-YBX1 or pcDNA-3.1-USP10, followed by exposure to OGD/R. The YBX1 protein level was decreased in OGD/R-subjected neurons, which was upregulated by USP10 overexpression; however, the effect induced by USP10 overexpression was reversed after knockdown of YBX1 (Fig. 5C). The decreased cell viability induced by OGD/R was elevated after overexpression of USP10, whereas YBX1 silencing abolished the effect of USP10 upregulation (Fig. 5D). The promoted cell apoptosis caused by OGD/R was repressed by USP10 overexpression, which was reversed after knockdown of YBX1 (Fig. 5E). Moreover, overexpression of USP10 elevated MMP under OGD/R conditions, which was offset by YBX1 silencing (Fig. 5F). Immunofluorescence showed that USP10 overexpression enhanced the colocalization of LC3 and Tomm20, whereas this effect was abolished by overexpression of USP10 (Fig. 5G). In addition, upregulation of USP10 elevated LC3II/I and mitochondrial PINK1 and Parkin but downregulated the protein levels of p62 and mitochondrial Tomm20 under OGD/R conditions, whereas knockdown of YBX1 reversed the effects of USP10 overexpression (Fig. 5H). The results suggested that USP10 activated PINK1/Parkin signaling and promoted neuronal mitophagy via upregulating YBX1 in OGD/R-exposed neurons.
Knockdown of YBX1 reversed the effects of USP10 overexpression on PINK1/Parkin signaling and mitophagy in OGD/R-exposed neurons. A–B, sh-YBX1 was transfected into PC12 cells. RT-qPCR and Western bolt for detecting YBX1 expression. C–H, PC12 cells were transfected with sh-NC, sh-YBX1, pcDNA- 3.1, or pcDNA-3.1-USP10, followed by exposure to OGD/R. C, Western blot for detection of the YBX1 protein level. D, CCK-8 assay for cell viability measurement. E, Flow cytometry for determining apoptosis. F, JC-1 staining for measuring MMP. G, Immunofluorescence for detecting the colocalization of LC3 and Tomm20. H, Western blot for determining autophagy-associated proteins (LC3II/I, p62, PINK1, Parkin) and Tomm20 (a mitochondria marker). n = 3. *p < 0.05; **p < 0.01; ***p < 0.001. Data are displayed as mean ± SD.
TSG heightened mitophagy and inhibited neuronal apoptosis through upregulating USP10 in OGD/R-induced neurons
To study whether USP10 was implicated in the neuroprotection of TSG on regulating mitophagy and neuronal apoptosis, cells were transfected with sh-USP10, followed by exposure to OGD/R and treated with TSG. We found that knockdown of USP10 reversed the promoting effect of TSG on OGD/R-induced PC12 cell viability (Fig. 6A). TSG treatment rescued apoptosis in OGD/R-induced PC12 cells, whereas USP10 knockdown reversed the effect induced by TSG (Fig. 6B). Furthermore, TSG treatment elevated MMP in OGD/R-exposed cells, which was abolished after knockdown of USP10 (Fig. 6C). The enhanced colocalization of LC3 and Tomm20 caused by TSG treatment in OGD/R-exposed cells was reversed by USP10 knockdown (Fig. 6D). As expected, knockdown of USP10 abolished the upregulation of LC3II/I and mitochondrial PINK1 and Parkin, and downregulation of p62 and mitochondrial Tomm20 caused by TSG in OGD/R-subjected cells (Fig. 6E). LAMP2 is one of the major protein components of the lysosomal membrane (Qiao et al., 2023). The colocalization of LC3B-II with LAMP2 indicates fusion of autophagosome with lysosomes (H. Yang et al., 2020). Herein, we found that TSG increased the colocalization puncta of LC3 and LAMP2 in OGD/R-exposed cells, which was reversed after knockdown of USP10 (Fig. 6F). Taken together, TSG upregulated USP10 to promote mitophagy and inhibit neuronal apoptosis in OGD/R-induced neurons.
TSG heightened mitophagy and inhibited neuronal apoptosis through upregulating USP10 in OGD/R-induced neurons. sh-NC or sh-USP10 was transfected into PC12 cells, before being exposed to OGD/R and treated with TSG. A, CCK-8 assay for cell viability detection. B, Apoptosis measured with flow cytometry. C, JC-1 staining for determining MMP. D, Immunofluorescence for detecting the colocalization of LC3 and Tomm20. E, Western blot for testing autophagy-associated proteins (LC3II/I, p62, PINK1, Parkin) and Tomm20 (a mitochondria marker). F, Immunofluorescence for detecting the colocalization of LAMP2 and LC3. n = 3. Data are shown as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001.
Discussion
Ischemic stroke is a major cause of neurological disability and mortality worldwide (Ahsan et al., 2021). Given that mitophagy is linked with many central nervous system disorders including cerebral ischemic injury (Guan et al., 2018), this research aims to investigate whether TSG plays a neuroprotective role in cerebral ischemic damage via regulating mitophagy and explores the potential mechanism. Here, we uncovered that TSG improved cerebral I/R injury and promoted mitophagy in MCAO rats. Meanwhile, TSG inhibited neuronal apoptosis and promoted mitophagy in OGD/R-induced neurons. Mechanically, TSG upregulated USP10 to elevate YBX1 protein expression, thereby facilitating PINK1/Parkin-mediated mitophagy and improving ischemic stroke injury.
Mitophagy controls the number and the quality of mitochondria, clearing excessive or dysfunctional mitochondria which may result in cell death (L. Li et al., 2023b). In the context of autophagy, the cytosolic LC3 form (nonlipidated LC3I) becomes linked to phosphatidylethanolamine, resulting in the lipidated LC3 (LC3II) form, which is aggregated on the autophagosomal membrane as an autophagy marker (M. Li et al., 2016). The colocalized dots of LC3 with LAMP2 (a marker of lysosome) are known as autolysosomes, reflecting the quantity of cargo to be degraded by autophagy (Liao et al., 2022). The overlap of LC3-labeled autophagosomes with Tomm20-labeled mitochondria indicates the level of mitophagy (X. Yang et al., 2018). Previous studies have indicated reperfusion following cerebral ischemia is an essential turning point for autophagy/mitophagy from the protecting mechanism to the detrimental process (Feng et al., 2017, 2018a). Moreover, existing evidence suggests the dual roles of mitophagy in ischemic stroke (Guan et al., 2018). During cerebral ischemic injury, mitophagy can be enhanced (L. Yang et al., 2022, Zhou et al., 2023) or blocked (Feng et al., 2018b, L. Li et al., 2023b) following the treatments of different drugs, while these drugs may exert neuroprotective functions; however, the precise mechanism remains unclarified. Here, whether TSG improved ischemic stroke injury through mitophagy was explored. Notably, concurrently with the neuroprotection of TSG, LC3II/I was increased, and the expression of p62 was decreased. Besides, the colocalized dots of LC3 with Tomm20 (a mitochondrial marker) in the cytoplasm were elevated after TSG treatment. Our research also found that MMP was increased with TSG treatment under OGD/R in vitro. These discoveries indicated that the enhanced mitophagy induced by TSG might be a crucial mechanism underlying the protective function.
PINK1/Parkin-mediated mitophagy has been reported to play a critical role in the neuroprotection following ischemic stroke (H. Wang et al., 2019a, Mao et al., 2022). When mitophagy occurs, PINK1/Parkin signaling is activated, and PINK1 recruits Parkin to the mitochondria surface to activate Parkin E3 ubiquitin ligase; Parkin can label mitochondrial components such as Tomm20 through ubiquitination, which results in the degradation of these mitochondrial proteins by proteasome (Stolz et al., 2014). For example, Lixuan Yang et al. have discovered that mitophagy was repressed in LPS-induced astrocytes, and Parkin in the cytoplasm failed to transfer to the mitochondria, which contributed to the stability of Tomm20; Morinda officinalis leads to ubiquitination and degradation of Tomm20 through promoting the enrichment of Parkin in mitochondria, thereby facilitating mitophagy (L. Yang et al., 2023). In addition, it has been revealed that BDNF induced mitophagy via reducing Tomm20 and increasing LC3II/LC3I, thereby alleviating high-glucose–induced brain microvascular endothelial cell injury (Jin et al., 2019). Our study discovered reduction of Parkin and increase of Tomm20 in mitochondria under OGD/R conditions. However, TSG could promote the transfer of Parkin to mitochondria, which led to the downregulation of Tomm20 in mitochondria, indicating activation of PINK1/Parkin-mediated mitophagy might be implicated in the neuroprotection of TSG on ischemic stroke.
Ubiquitination is usually mediated by ubiquitin proteasome system, while ubiquitin, ubiquitin protein ligase, ubiquitin-conjugating enzyme, ubiquitin-activating enzyme, deubiquitinase, and the proteasome are essential components of the ubiquitin proteasome system (Kane et al., 2021). USP10, a critical member of the deubiquitinase family, has been verified to be involved in many cellular processes and multiple neurodegenerative disorders (Bhattacharya et al., 2020). USP10 modulates pivotal signals involved in cell proliferation, apoptosis, and autophagy (H. Li et al., 2022a). Increase in USP10 level induced by vagus nerve stimulation was associated with the deubiquitination of IkappaB kinase (IKK)-γ in ischemic stroke (C. Xie et al., 2023). Encouraging USP10-dependent Smad4 deubiquitination aggravated cardiac fibrosis post myocardial I/R damage (S. Xie et al., 2022). Herein, we uncovered that the promoted mitophagy and inhibited neuronal apoptosis induced by TSG treatment could be reversed following knockdown of USP10 in OGD/R-exposed neurons. USP10 may be a molecular target for the neuroprotection of TSG, and this hypothesis can be further validated at the animal level in the future. Nevertheless, how TSG leads to USP10 activation is still unclear. It has been found that resveratrol, a TSG analog, disrupted the interaction between Ras-GTPase–activating protein SH3 domain-binding protein 1 (G3BP1) and USP10 through directly targeting G3BP1, thereby enhancing USP10-mediated p53 deubiquitination (Oi et al., 2015). G3BP1 is one of the main components of stress granules (SGs; Q. Wang et al., 2024). SGs are cytoplasmic aggregates of stalled translational preinitiation complexes induced by various environmental stresses such as heat shock, hypoxia, endoplasmic reticulum stress, and oxidative stress (An et al., 2019). In I/R injury, regulating the formation of SGs is a promising intervention for improving self-protection capability of the cells and inhibiting apoptosis (Si et al., 2020). As a resveratrol analog, whether TSG affects the activation of USP10 through directly interacting with G3BP1 protein or regulates the function of USP10 via combining with other proteins remains to be further explored.
YBX1, a DNA/RNA-binding protein, is one member of the cold-shock protein family (Mordovkina et al., 2020). It has been demonstrated that upregulating YBX1 protein with dexmedetomidine could protect neuronal cells from I/R-induced apoptosis (Yan et al., 2023); YBX1 protected against apoptosis induced by OGD/R in PC12 cells (Tuerxun et al., 2021). Besides, YBX1 carried by neural stem cell extracellular vesicles inhibited neuronal pyroptosis in ischemic stroke (Peng et al., 2023). Furthermore, YBX1 knockdown decreased PINK1 and inhibited mitophagy during embryonic development (Jiang et al., 2023). Additionally, YBX1 enhanced PINK1/Parkin-mediated mitophagy, thereby promoting brown adipogenesis (R. Wu et al., 2022). Here, we found that USP10 overexpression inhibited ubiquitination-dependent degradation of YBX1 in OGD/R-exposed neurons. Furthermore, USP10 overexpression promoted PINK1/Parkin signaling activation and neuronal mitophagy; however, these effects could be reversed by knockdown of YBX1, indicating that USP10 upregulated YBX1 to activate PINK1/Parkin signaling and promote neuronal mitophagy.
To sum up, TSG played a vital protective role in cerebral ischemic damage via upregulating USP10/YBX1 axis. USP10 may be a molecular target for the neuroprotection of TSG. Inevitably, there are also some limitations. First, no further intervention studies have been designed to verify the role of USP10/YBX1 axis in IS. Additionally, only the effects on neurons were explored, whether the effects on other cells types within the brain such as microglia are worthy of further exploration. Our research provides a theoretical basis for applying natural component TSG in cerebral ischemic injury, suggesting that the therapeutic effect of TSG may benefit from upregulating USP10 and YBX1 levels.
Ethics Approval and Consent to Participate
The procedures were performed according to the National Institutes of Health Guidelines for Animal Research and approved by the Animal Ethics Committee of Hunan University of Medicine [202211026].
Availability of Data and Materials
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Consent for Publication
N/A.
Footnotes
The authors declare no competing financial interests.
This work was supported by Joint project of Hunan Provincial Natural Science Foundation (2023JJ50437) and Joint project of Hunan Provincial Natural Science Foundation (2024JJ7322).
↵*Y.L., K.H., and J.L. are the co-first authors.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.
References
In this issue
