Effect of Parkin on methamphetamine-induced α-synuclein degradation dysfunction in vitro and in vivo.

Abstract Introduction Methamphetamine (METH) is a psychostimulant drug with complicated neurotoxicity, and abuse of METH is very common. Studies have shown that METH exposure causes alpha‐synuclein (α‐syn) accumulation. However, the mechanism of α‐syn accumulation has not been determined. Methods In this study, we established cell and animal models of METH intoxication to evaluate how METH affects α‐syn expression. In addition, to explore METH‐induced neurotoxicity, we measured the level of Parkin and the phosphorylation levels of α‐syn, Polo‐like kinase 2 (PLK2), the proteasome activity marker CD3δ, and the apoptosis‐related proteins Caspase‐3 and PARP. Parkin is a key enzyme in the ubiquitin–proteasome system. In addition, the effect of Parkin on METH‐induced neurotoxicity was investigated by overexpressing it in vitro and in vivo. Results METH exposure increased polyubiquitin and α‐syn expression, as did MG132. Furthermore, the level of Parkin and the interaction between Parkin and α‐syn decreased after METH exposure. Importantly, the increases in α‐syn expression and neurotoxicity were relieved by Parkin overexpression. Conclusions By establishing stable cell lines and animal models that overexpress Parkin, we confirmed Parkin as an important factor in METH‐induced α‐syn degradation dysfunction in vitro and in vivo. Parkin may be a promising target for the treatment of METH‐induced neurotoxicity.

. Furthermore, METH can lead to the damage of dopaminergic neurons and Parkinson-like pathology (Biagioni et al., 2019;Li et al., 2017). Despite much evidence of METH-induced neuronal damage, the molecular mechanisms of METH neurotoxicity have not been completely revealed.
Parkin is an extremely conserved protein and an E3 ubiquitin ligase that degrades proteins of the ubiquitin-proteasome system (UPS) (Martinez, Ramirez, Osinalde, Arizmendi, & Mayor, 2018). Parkin in cooperation with E1 activating enzyme and E2 binding enzyme promotes the ubiquitination of substrate proteins to be degraded (Corsa et al., 2019). Deletion or functional defects of Parkin may lead to the accumulation of substrate proteins that cannot be efficiently degraded (Brahmachari et al., 2019). Previous studies have found that α-syn is a substrate protein of Parkin (Liu, Hebron, Shi, Lonskaya, & Moussa, 2019). In addition, the UPS plays a vital role in the degradation of α-syn . However, the contribution of Parkin to the degradation of α-syn following METH-induced α-syn accumulation is unclear.
The study aimed to determine the effect of Parkin on α-syn degradation dysfunction after METH exposure in vitro and in vivo. First, we tested whether impairment of the UPS pathway and Parkin level contributes to the dysfunction of α-syn degradation after METH exposure. We measured α-syn and polyubiquitin expression after exposure to METH and MG132 (a proteasome inhibitor). Then, we established cell and animal models of METH intoxication to investigate how METH affects Parkin and α-syn expression and their interaction. In addition, the phosphorylation levels of α-syn (P-α-syn), Polo-like kinase 2 (PLK2), proteasome activity marker CD3δ, and the apoptosis-related proteins Caspase-3 and PARP were measured to detect METH-induced neurotoxicity. Furthermore, a thorough investigation of the effect of Parkin on α-syn degradation dysfunction was conducted by overexpressing Parkin in vitro and in vivo. The results indicated that METH can increase polyubiquitin and α-syn expression, as can MG132. Furthermore, the level of Parkin and the interaction between Parkin and α-syn were found to be decreased after METH exposure. Importantly, the increases in α-syn and neurotoxicity were relieved after overexpressing Parkin. By establishing stable cell lines and animal models that overexpress Parkin, we confirmed Parkin as an important factor in α-syn degradation dysfunction after METH exposure in vitro and in vivo.

| Animal protocol
Mice (C57 BL/6) were obtained from Laboratory Animal Center of Southern Medical University (Guangzhou, China) and housed in a dedicated animal room. The mice were divided randomly into two groups: a control group and a subacute METH exposure group (n = 3/group). Mice in the subacute exposure group received intraperitoneal (i.p.) injections at 12-hr intervals for 4 days with 15 mg/ kg METH diluted with saline (>99% purity; National Institutes for Food and Drug Control, Guangzhou, China). Mice in the control group received an equivalent amount of saline. We selected this exposure paradigm after referring to the concentrations of shortterm METH exposure in humans (Du et al., 2017). At 24 hr after the last injection, the mice were euthanized, and midbrain and striatum tissues of brain were dissected. The midbrain and striatum were selected based on our previous trials and a previous study (Jakowec, Donaldson, Barba, & Petzinger, 2001). The Institutional Animal Care and Use Committee at Southern Medical University approved the animal experiments, and the experiments were performed according to the guidelines of the National Institutes of Health (NIH).

| Immunofluorescence analysis
We performed immunofluorescence labeling to determine ubiquitin and α-syn expression levels in SH-SY5Y cells. Cells were fixed in 4% paraformaldehyde and then washed three times with phosphate-buffered saline (PBS). The cells were then blocked with 10% BSA containing 0.05% Triton X-100 at room temperature for Then, the cells were incubated with secondary antibody for 1 hr at room temperature. DAPI (H-1200; Vector) was used to stain the nuclei. Images were obtained via a fluorescence microscope or a N-SIM Nikon Super-Resolution Microscope (Nikon).

| Immunohistochemistry analysis
The brains were fixed in 4% paraformaldehyde and embedded in paraffin. Sections (3 μm in thickness) were made, pretreated in citrate buffer (0.01 M, pH 6.0) for antigen retrieval via hydrated autoclaving in a humid atmosphere for 10 min, and then blocked at room temperature with goat serum for 30 min. Then, the sections were incubated with the primary antibody anti-Parkin antibody (1:200, Abcam, Cat. #ab77924) overnight at 4°C. 3,3′-diaminobenzidine (DAB) kits (CW Bio, Cat. #CW2069) were used to develop the slices. Images were obtained using Zeiss Metasystem (Zeiss).

| Coimmunoprecipitation
With the protease inhibitors, protein samples were extracted in RIPA buffer. The mixtures were treated with anti-Parkin antibody (1:50, Abcam, Cat. #ab77924) or control IgG (Beyotime, Cat. # A7016) at 4°C overnight followed by incubation with Protein A/G (Beyotime, Cat. #P2012) at 4°C for 2 hr. After the samples were centrifuged at 1,500 g for 5 min, the supernatant was removed. Subsequently, the beads were washed three times with PBS to remove the proteins.

| Virus transfection in cells
The Parkin gene lentivirus was purchased from the GK Genes Company. LV-NC was used as the control virus. According to preliminary experiments, we determined the optimal infection conditions: virus concentration 1 × 10 7 Tu/ml for 48 hr. SH-SY5Y cells were infected with LV-Parkin or LV-NC at approximately 80% density for 48 hr. Then, the cells were treated with METH or PBS for the following experiment.

| Virus transfection in mice
Mice were randomly divided into four groups (n = 3/group): an LV-NC group, an LV-Parkin group, an LV-NC + METH group, and an LV-Parkin + METH group. A stereotaxic injection protocol was performed as previously described by our laboratory Zhu et al., 2018). Mice were anesthetized by 1% pentobarbital and fixed in a stereotaxic frame (Domitor). A total of 2 µl of LV-NC or LV-Parkin was injected into the right striatum at an injection rate of 1 µl/min. The specific coordinates were as follows: The bite bar was set at zero, 0.38 mm rostral to Bregma, 1.78 mm right lateral to the midline, and 3.25 mm ventral to the dura. The striatum was selected because it was the major target site for METH according to our previous study (Chang et al., 2005). Mice were exposed to METH or saline after 2 weeks.

| Statistical analysis
All data are presented as the mean ± standard deviation (SD). SPSS version 20.0 (IBM Corporation) was used to perform all data analysis. Comparison between groups was conducted by one-way ANOVA followed by the Bonferroni post hoc analysis or the Mann-Whitney U test for two independent samples. p < .05 was considered statistically significant. Details on the statistics are shown in Table 1. Each experiment was repeated three times.

| MG132 (a proteasome inhibitor) increases the expression of α-syn and polyubiquitin in SH-SY5Y cells
To test the role of the UPS pathway in α-syn degradation, we treated SH SY5Y cells with different concentrations of F I G U R E 1 MG132 (proteasome inhibitor) increases the levels of alpha-synuclein (α-syn) and polyubiquitin in SH-SY5Y cells. SH-SY5Y cells were exposed to a range of MG132 doses for 24 hr. Western blot (a) and quantitative analyses (b) showed that MG132 increased α-syn and polyubiquitin expression in SH-SY5Y cells. Western blot (a) and quantitative analyses (b) were performed to determine the levels of α-syn and ubiquitin. β-actin was used as a loading control. *p < .05 compared with the control group. The data were analyzed using one-way ANOVA followed by the Bonferroni post hoc analysis. DMSO was used as the drug carrier at a concentration of <0.2%. Each experiment was repeated three times ubiquitin proteasome inhibitor MG132 (0-125 nM) cells for 24 hr. Western blot results showed that the levels of α-syn and polyubiquitin were significantly increased after exposure to MG132 (Figure 1a,b). The results suggest that impairment of the UPS pathway may be of great importance to α-syn degradation dysfunction.
F I G U R E 2 Methamphetamine (METH) increases polyubiquitin protein expression in vitro and in vivo. Mice were randomly divided into two groups: a control group and a subacute METH group (n = 3/group). Western blot (a) and quantitative analyses (b) showed that METH increased polyubiquitin expression in the midbrain and striatum of male C57 mice. SH-SY5Y cells were exposed to 2 mM METH for 24 hr. Fluorescence microscopy (c) results showed ubiquitin and α-syn were colocalized and expressed at higher levels in METH-treated SH-SY5Y cells than in control cells. Ubiquitin was stained with an anti-Ubiquitin antibody (green), α-syn was stained with an anti-α-syn antibody (red), and nuclei were counterstained with DAPI (blue). β-actin was used as a loading control. *p < .05 compared with the control group. The data were analyzed using the Mann-Whitney U test. Each experiment was repeated three times

| METH increases polyubiquitin protein expression in vitro and in vivo
To test the effect of ubiquitin on METH-induced toxicity, we determined the levels of ubiquitin in vitro and in vivo. Western blot results showed no difference in monoubiquitin between the two groups. However, polyubiquitin protein expression in the midbrain and striatum was significantly higher in the METH exposure group than in the control group (Figure 2a,b). In addition, the immunofluorescence results showed that the levels of ubiquitin and α-syn proteins were increased and exhibited colocalization after METH exposure (Figure 2c). Taken together, the results suggest that METH exposure can increase the level of polyubiquitin in vitro and in vivo.

| METH Influences Parkin and α-syn Protein Expression in vitro and in vivo
To determine the effects of METH on Parkin and α-syn proteins,  Figure 3a,b).

F I G U R E 3 Methamphetamine (METH) increases α-syn expression, decreases Parkin protein expression, and reduces the interaction between
Parkin and α-syn in SH-SY5Y cells. SH-SY5Y cells were treated with 0.5-3.0 mM METH for 24 hr or 2.0 mM METH for 2-24 hr. Western blot (a) and quantitative analyses (b) showed that METH increased the expression of α-syn and decreased the expression of Parkin in dose-and time-dependent manners. Coimmunoprecipitation results (c) showed that the interaction between Parkin and α-syn was reduced after METH exposure. Cells were immunoprecipitated with an anti-Parkin antibody and then with an anti-α-syn antibody and analyzed with Western blot. IgG was used as a negative control, and β-actin was used as a loading control. *p < .05 compared with the control group. The data were analyzed using one-way ANOVA followed by the Bonferroni post hoc analysis. Each experiment was repeated three times

F I G U R E 4 Methamphetamine (METH) increases α-syn expression and decreases
Parkin protein expression in the midbrain and striatum of mice. Mice were randomly divided into two groups: a control group and a subacute METH group (n = 3/group). Western blot (a) and quantitative analyses (b) showed that METH increased α-syn expression and decreased Parkin protein expression in vivo. β-Actin was used as a loading control. *p < .05 compared with the control group. The data were analyzed using the Mann-Whitney U test. Each experiment was repeated three times F I G U R E 5 Parkin was overexpressed after the transfection of LV-Parkin in SH-SY5Y cells and C57 mice. SH-SY5Y cells and male C57 mice were transfected with LV-NC or LV-Parkin (n = 3/group). Immunofluorescence (a) and immunohistochemistry analyses (b) showed that the distribution and extent of Parkin were significantly increased in cells and striatum transfected with LV-Parkin. Parkin was stained with an anti-Parkin antibody (red), and nuclei were counterstained with DAPI (blue). Each experiment was repeated three times | 9 of 12 MENG Et al.
To verify the results obtained in vitro, we analyzed Parkin and α-syn protein expression in the midbrain and striatum of mice. The Western blot results indicated that in the subacute exposure group, Parkin and α-syn protein expression in the midbrain and striatum was increased (Figure 4a,b). These results are consistent with the results in vitro.
The results presented above suggest that the levels of α-syn and Parkin protein were affected by METH exposure. Moreover, the coimmunoprecipitation results showed that the interaction between Parkin and α-syn was decreased in SH-SY5Y cells after METH exposure ( Figure 3c).

| The increase in α-syn induced by METH is relieved when Parkin is overexpressed
To investigate the effect of Parkin on METH-induced α-syn degradation, we used LV-Parkin to overexpress Parkin in SH-SY5Y cells and mice. The level of Parkin protein was significantly increased after virus injection (Figure 5a,b). In the saline-exposed group, there was no difference in the levels of α-syn, P-α-syn, PLK2, PARP, Caspase-3, and CD3δ between the LV-NC and LV-Parkin groups. However, after exposure to METH, the levels of α-syn, P-α-syn, PLK2, PARP, Caspase-3, and CD3δ were increased. More importantly, transfection with LV-Parkin relieved these increases (Figure 6a,b).
The results were verified in mice. We injected LV-Parkin or LV-NC into the right striatum of mice to verify the effect of Parkin on the degradation of α-syn induced by METH. The results showed that the levels of α-syn, P-α-syn, PLK2, PARP, Caspase-3, and CD3δ in the midbrain and striatum of mice treated with LV-NC + METH were increased significantly relative to the corresponding levels in the LV-NC + Saline group (Figure 6a,b). In addition, the increases were mitigated to large extent after transfection with LV-Parkin ( Figure 6a,b). The results of in vivo and in vitro experiments revealed that the increases in α-syn and α-syn degradation dysfunction induced by METH can be relieved when Parkin is overexpressed.

| D ISCUSS I ON
In this study, we found that Parkin level decreased and α-syn level increased after exposure to METH in vitro and in vivo. Furthermore, we found that polyubiquitin was increased by METH and MG132, which suggests that impairment of the UPS pathway may contribute to the dysfunction of α-syn degradation after METH exposure.
The reason for the increase in polyubiquitin may have been that Parkin, as an E3 ubiquitin ligase that regulates the level of α-syn ubiquitination, decreased after exposure to METH. When Parkin is overexpressed, proteasome activity increases, and α-syn levels and METH-induced apoptosis are significantly reduced. Our results and previous studies  show that Parkin has an important role in the METH-induced dysfunction of α-syn degradation.
α-syn is a neuroprotein widely expressed in the brain and is . The balance is broken when cells are exposed to oxidative stress, poison, and other stress conditions. Fibrils rapidly aggregate into insoluble macromolecules to form LBs (Auluck, Chan, Trojanowski, Lee, & Bonini, 2002). In the study, we verified that the level of α-syn increases with increasing METH concentration and exposure time, which is consistent with previous studies (Qiao et al., 2019;Zhu et al., 2018). The accumulation of α-syn can lead to apoptosis and even death of neurons. Most misfolded, damaged, and incompletely assembled proteins in the body are cleared by the UPS Yuan et al., 2019). However, our results indicate that polyubiquitin is increased after METH exposure, as it is after MG132 exposure, which suggests impairment of the UPS in degrading α-syn. In the subacute METH exposure model, proteasome activity is decreased; thus, the degradation of ubiquitinated proteins decreases, which leads to an increase in polyubiquitination. We propose the hypothesis that Parkin is of great significance to the METH-induced dysfunction of α-syn degradation.
Parkin functions as an E3 ubiquitin ligase in degrading proteins of the UPS. The function of Parkin is impaired when stimulated by external stimuli such as drugs (Moszczynska & Yamamoto, 2011).
As a result of this impaired functioning, substrate proteins are not fully degraded, and they consequently accumulate in cells. Studies have shown that Parkin specifically recognizes α-syn and promotes its degradation . Our study found that the level of Parkin showed a trend of first increasing and then decreasing in SH-SY5Y cells in a dose-dependent and time-dependent manner after METH exposure. A possible reason for this pattern is that Parkin undergoes a compensatory increase and then decreases with increasing exposure time and METH concentration. METH increases α-syn levels in a dose-and time-dependent manner. In addition, the F I G U R E 6 The increase in α-syn induced by METH is relieved when Parkin is overexpressed. SH-SY5Y cells and male C57 mice were transfected with LV-NC or LV-Parkin (n = 3/group). Then, cells and mice were treated or not with METH. Western blot (a) and quantitative analyses (b) showed that the levels of α-syn, P-α-syn, PLK2, PARP, Caspase-3, and CD3δ in the groups treated with LV-NC + METH were significantly increased relative to the corresponding levels in the LV-NC + Saline groups. These increases were mitigated to large extent after transfection with LV-Parkin. β-Actin was used as a loading control. *p < .05 compared with the LV-NC + Saline group; #p < .05 compared with the LV-NC + METH group. The data were analyzed using one-way ANOVA followed by the Bonferroni post hoc analysis. Each experiment was repeated three times coimmunoprecipitation results showed that the interaction between Parkin and α-syn was reduced significantly by METH exposure. The results suggest that Parkin is of great significance to the dysfunction of α-syn degradation induced by METH.
The study also revealed that the serine 129-site phosphorylation modification of α-syn (P-α-syn) increased significantly in METHtreated cells. Studies have shown that under physiological conditions, <4% of α-syn in brain neurons is in phosphorylated α-syn forms, whereas more than 90% α-syn is phosphorylated in LBs (Anderson et al., 2006;Fujiwara et al., 2002;Oueslati, 2016). Our results confirm that METH induces an increase in the α-syn phosphorylation in SH-SY5Y cells, which is of great significance to the development of METH neurotoxicity. According to the literature, CD3δ is the most characteristic substrate for ER-related degradation. When E3 ligase cannot degrade CD3δ, the protein level of CD3δ is increased, and the stability is greatly improved. CD3δ is closely related to proteasome activity (Garcillan et al., 2014). The results showed the CD3δ was increased after METH exposure, indicating that proteasome activity was reduced after METH treatment. Such reduction could result in the lack of degradation of α-syn, which then accumulates in cells. Furthermore, compared with the environment of cells in brain tissue, the environment of cells cultured in vitro is simple, and the response to stimulation is large. When cells are stimulated, the microenvironment around the cells has a buffer effect, which may lead to the degradation of CD3d. There is an inhibitory effect on its increase, so the increase in CD3d in the brain is not as large as it is in cells (Seo et al., 2019). Apoptosis is an autonomous procedure of cell death that is of great significance to the development of the nervous system and to maintaining cellular homeostasis. Many studies have reported that neuronal apoptosis occurs during the development of neurodegenerative diseases such as AD, PD, and ALS (Ghavami et al., 2014). In this study, we found that the expression of the apoptosis-related proteins Cleaved Caspase-3 and Cleaved PARP was increased significantly after METH exposure, which indicated that apoptosis was increased by METH. These results are consistent with previous studies Xu et al., 2018).
In the present study, we overexpressed Parkin protein by transfection of human Parkin gene recombinant lentivirus in vitro and in vivo. Compared with LV-NC + METH, the LV-Parkin + METH group exhibited a significant decrease in α-syn, demonstrating that Parkin can promote the degradation of α-syn. Increased P-α-syn is an important pathological feature of PD, and α-syn phosphorylation produces many insoluble oligomers. The phosphorylation modification at this site is mainly catalyzed by PLK2. Compared with the corresponding levels in the LV-NC + METH group, PLK2 was significantly decreased in METH-treated cells and mice, and P-α-syn was significantly reduced. In addition, the LV-Parkin + METH group had higher proteasome activity and significantly reduced apoptosis relative to the LV-NC + METH group. These results indicate that the overexpression of Parkin protein can promote the degradation of α-syn and alleviate the neurotoxicity induced by METH, which is consistent with previous studies (Liu, Traini, Killinger, Schneider, & Moszczynska, 2013).
In summary, we verified that Parkin, as a E3 ubiquitin ligase, plays a critical role in METH-induced dysfunction of α-syn degradation in vitro and in vivo. The overexpression of Parkin significantly promoted the degradation of α-syn and relieved METH-induced neurotoxicity. It is necessary to study the mechanisms underlying the neuroprotective effect of Parkin under METH exposure and the role of the UPS pathway in α-syn degradation after METH exposure.

ACK N OWLED G M ENTS
This work was supported by a grant from National Natural Science Foundation of China (81671865) to PMQ.

CO N FLI C T O F I NTE R E S T
None declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data supporting the results of this study are publicly available.