2.1 Primary Neuron Cultures and Meth Treatment

Primary cultures of rat hippocampus tissues were dissected from fetal Sprague–Dawley (SD) rats (embryonic day 18), and the neurons were maintained in neurobasal medium (21103049, Gibco, USA) supplemented with 2% B27 (17504044, Gibco, USA) at 37°C in a 5% CO₂ incubator. Half of the medium was replaced every 3 days. After 7 days, the neurons were exposed to Meth obtained from the National Institutes for Food and Drug Control (Beijing, China).

2.2 In Vivo Meth Administration

Male C57BL/6 mice were obtained from the Experimental Animal Center of Nanjing Medical University. The animals were housed in cages with temperature (23°C ± 2°C), humidity (45% ± 5%), and a 12 h dark/light cycle. During the experiment, the animals were provided free access to food and water. All experiments were conducted under the control of the Ethics Committee of Animal Care and Experimentation of Europe and approved by the Institutional Animal Care and Use Committee (IACUC) at Nanjing Medical University (approval no. 2302024). Meth (10 mg/kg) or saline was administered to the mice every 2–3 h in four successive intraperitoneal (i.p.) injections within 24 h. Behavior tests were performed after the last injection of Meth or saline.

2.3 RNA Sequencing (RNA-Seq) and Data Analysis

RNA extracted from primary neurons was used to prepare mRNA libraries (Illumina platform, LC-Bio Technology Co. Ltd, Hangzhou, China). After cluster generation, transcriptome sequencing was carried out on an Illumina Novaseq 6000 platform that generated raw reads. The mapped clean-read number was normalized to RPKM (reads per kilo of per million mapped reads). The edgeR package was used to determine the StringTie genes. The threshold of significant difference was |log2foldchange| ≥ 2, p < 0.05. The Gene Ontology (GO) enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, Gene Set Enrichment Analysis (GSEA), and Volcano Plot analysis were conducted.

2.4 Proteomics

The mouse brain tissues were treated with Meth or Saline and then collected with lysis solution. The samples were homogenized using a tissue lyser followed by 15 min centrifugation (20,000 × g, 4°C) to collect the supernatant. Mass spectrometry analysis was performed and analyzed at LC Sciences (Hangzhou, China). Volcano Plot analysis and KEGG pathway analysis were conducted.

2.5 Stereotaxic Injection of Adeno-Associated Virus

Mice were anesthetized with isoflurane and placed in a stereotactic apparatus (RWD Life Science). The rAAV-hSyn-PARL-P2A-EGFP-WPRE-hGH (PARL) and the rAAV-U6-shPGAM5-CMV-mCherry-SV40 (ShPGAM5) were built by Brain VTA (Wuhan, China). The adeno-associated virus (AAV) titer was 5.88E+12 vg/mL and the sequence of shPGAM5 was GATCTTCATATGCCATGCCAA. A total of 1.6 μL of the virus was injected bilaterally into the hippocampus of mice. The bregma and posterior coordinates were used as the reference points. To overexpress PARL or knockdown PGAM5, the virus particle was bilaterally injected in 4-week-old mice at the following target coordinates: anteroposterior (AP) 2.1 mm (CA1)/2.9 mm (CA3), mediolateral (ML) 1.7 mm (CA1)/3 mm (CA3), and dorsoventral (DV) 2.1 mm (CA1)/3.8 mm (CA3) from bregma. The sham-operated mice were injected with the empty vector using the same procedure. Three weeks later, the mice underwent Meth treatment and behavioral tests.

2.6 Nissl Staining

The paraffin-embedded sections were immersed in a Nissl stain solution, then the brain slides were dehydrated before being sealed with neutral balsam and imaged under a light microscope. The number of Nissl bodies in stained tissue sections was calculated, and surviving neurons were assessed based on the presence or absence of Nissl bodies.

2.7 Neurobehavioral Assay

For each test, mice were habituated in the testing room for 2 h. Each box was cleaned with 75% alcohol between trials to eliminate olfactory cues.

2.8 Adenoviral Vector Transfection

Adenoviral vectors carrying 3 × Flag-PARL, 3 × Flag-ShPGAM5, and mCherry-GFP-LC3 (HB-AP2100001) were purchased from Hanbio (Shanghai, China), and the transfection was carried out according to the manufacturer's instructions. After adenovirus (Ad) infection for 12 h, the neurons were treated with Meth for 24 h and harvested for further analysis.

2.9 Transmission Electron Microscopy

The hippocampus was cut into blocks (Leica EM UC7, Germany), fixed with 2.5% glutaraldehyde overnight at 4°C, postfixed in 1% osmium tetroxide for 2 h at 37°C, and dehydrated in a graded series (20%–100%) of ethanol. The pellets were subsequently embedded in Epon 812 for 48 h at 60°C. Images were obtained by Transmission Electron Microscopy (TEM) (HT7700, Hitachi, Japan).

2.10 Western Blotting

The cells and tissue samples were lysed in RIPA lysis buffer containing a protease inhibitor cocktail. The protein concentrations of all samples were determined using a BCA protein assay kit (23,227, Thermo Fisher Scientific, USA). Equal amounts of protein (20 μg) were electrophoresed by 10%–12% SDS-PAGE gradient gel (PG113, Epizyme Biotechnology, China) and transferred to PVDF membranes. Then, the membranes were blocked by 3% BSA (07-248, Sigma-Aldrich, Germany) for 1 h at room temperature. After that, the membranes were incubated with the primary antibody diluent at 4°C overnight and incubated with the horseradish peroxidase (HRP)-conjugated secondary antibodies (A0208, Beyotime, China) at room temperature for 1 h. Protein bands were visualized using enzyme-catalyzed chemiluminescence and quantified using ImageJ software. β-actin was used as a reference. The primary antibodies used were p62 (ab109012, 1:10,000) antibody was obtained from Abcam; LC3 (12,741, 1:1000), RIP1 (3493, 1:1000), and MLKL (37,705, 1:1000) antibodies were obtained from Cell Signaling Technology; PARL (NBP1-80878, 1:500) and PINK1 (BC100-494, 1:1000) antibodies were obtained from Novus Biologicals; RIP3 (sc-374,639, 1:500) and β-actin (sc-47,778, 1:500) antibodies were obtained from Santa Cruz Technology; PGAM5 (28445-1-AP, 1:2000), Fis1 (10956-1-AP, 1:4000), Drp1 (12957-1-AP, 1:2000), and Parkin (14060-1-AP, 1:2000) antibodies were obtained from Proteintech.

2.11 Lactic Dehydrogenase Release Measurement

After Meth treatment, the supernatant of neurons was collected, and Lactic Dehydrogenase (LDH) levels were detected using the LDH Activity Assay Kit (C0016, Beyotime, China) according to the manufacturer's protocol. The absorbance at 490 nm was measured with a microplate reader (Infinite M200, TECAN, China).

2.12 Immunofluorescence Staining

The primary neurons were fixed with 4% paraformaldehyde for 30 min and then rinsed with 0.3% Triton X-100 for 15 min. After 5% bovine serum albumin blockage, the cells were incubated at 4°C overnight with the following primary antibodies: LC3 (12741, 1:200), RIP1 (3493, 1:200), and Tom20 (42406S, 1:200) antibodies were obtained from Cell Signaling Technology; PARL (NBP1-80878) antibody was obtained from Novus Biologicals, RIP3 (sc-374,639, 1:200), LC3 (sc-398,822, 1:200), Tom20 (sc-17,764, 1:200), and PINK1 (sc-514,353, 1:200) antibodies were obtained from Santa Cruz Technology; p-MLKL (ab196436, 1:200) antibody was obtained from Abcam; PGAM5 (68116-1-Ig, 1:1000), MLKL (66675-1-Ig, 1:200), RIP3 (17563-1-AP, 1:200), LAMP1 (67300-1-Ig, 1:200), and Parkin (14060-1-AP, 1:200) were obtained from Proteintech. Then, the cells were incubated with a fluorescently labeled secondary antibody Alexa Fluor 488 goat anti-mouse IgG (A21206, Invitrogen, USA) or Alexa Fluor 568 goat anti-rabbit (A10036, Invitrogen, USA) at a 1:500 ratio for 2 h, and the nuclei were stained with DAPI (C1006, Beyotime, China). Images were captured with a fluorescence microscope (LSM900, Zeiss, Germany), and Pearson's correlation was shown to evaluate the merged colocalization level by Image J software.

2.13 Calcein-AM/Propidium Iodide Staining

2.14 Measurement of Mitochondrial Reactive Oxygen Species

The Mitochondrial Reactive Oxygen Species (mtROS) was measured by MitoSOX Red (M36008, ThermoFisher Scientific, USA). The primary neurons were incubated with MitoSOX (5 μM) for 20 min at 37°C. Then the fluorescence intensity was detected by confocal microscopy (LSM900, Zeiss, Germany). The mtROS level was semi-quantitatively analyzed by ImageJ software using the following formula:

2.15 Mitochondrial Membrane Potential Assessment

The mitochondrial membrane potential (MMP) was detected by a JC-1 Assay Kit (C2006, Beyotime, China) according to the instructions of the manufacturer. Briefly, the primary neurons were incubated with JC-1 working solution for 20 min, then washed with JC-1 staining buffer, and the images were captured under a fluorescence confocal microscope (LSM900, Zeiss, Germany), and the ratio of JC-1 aggregates to monomers was detected.

2.16 Mitotracker Staining

Mitotracker Red CMXRos (C1049B, Beyotime, China) was used to label mitochondria according to the manufacturer's instructions. In brief, the neurons were incubated with Mitotracker Red working solution (100 nM) at 37°C for 20 min and subjected to confocal microscopy (LSM900, Zeiss, Germany). The length of mitochondria was determined and calculated by ImageJ software.

2.17 Enzyme-Linked Immunosorbent Assay

After Meth treatment, the culture medium of primary neurons was obtained, and the tumor necrosis factor-α (TNF-α) was detected by an Enzyme-Linked Immunosorbent Assay (ELISA) kit (MM-0180R2, Maisha, China) according to the manufacturer's instructions.

2.18 Reverse Transcription-Quantitative Polymerase Chain Reaction

Total RNA from primary neuron lysates was extracted with TRIzol reagent following the manufacturer's protocol (TrizolTM Reagent, Thermo Fisher Scientific, USA). The extracted RNA was reverse-transcribed to cDNA using the TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix kit (AQ601-04, TransGen Biotech, China). Relative mRNA levels were detected using SYBR Green Supermix (11201ES08, Yeasen Biotech Co. Ltd., China) on the LightCycler 480 (Roche, USA). The β-actin was used as the internal control. The sequences of the primers used for qPCR are listed in Table S1.

2.19 Statistical Analysis

All data were presented as the mean ± SEM and analyzed using GraphPad Prism (version 8). Normality was assessed by applying Shapiro–Wilk tests, and data sets that failed these tests were analyzed nonparametrically. Subsequently, comparisons between two groups were made using two-tailed Student's t tests, while comparisons among multiple groups were conducted using one-way ANOVA followed by Tukey's post hoc tests, and p < 0.05 was considered statistically significant.

3.1 Meth Exposure Contributed to Mitophagy and Necroptosis in Primary Neurons

Previous studies have found that binge administration of Meth ranging from 250 mg to 1 g results in brain concentration levels of Meth between 164 and 776 μM [35]. Thereafter, a dose of 900 μM Meth was used in the subsequent experiments were based on our previous studies [20, 21]. The transcriptome results identified 184 upregulated genes and 301 downregulated genes in primary neurons after Meth exposure (Figure 1A). GO enrichment analysis revealed these differentially expressed genes (DEGs) mainly participate in neuron death, autophagy, and mitophagy pathways (Figure 1B). Similarly, KEGG enrichment analysis showed substantial changes in autophagy, mitophagy, and TNF signaling pathways (Figure 1C). The importance of mitophagy and mitochondrial fission was further confirmed by GSEA analysis (Figure 1D). Mitophagy-related genes PARL, PGAM5, PINK1, and Parkin, as well as necroptosis-related genes MLKL, RIP1, and RIP3, all significantly increased in primary neurons after Meth treatment (Figure 1E). Besides, the proteomics analysis revealed 184 upregulated proteins and 121 downregulated proteins in the hippocampus tissue after Meth treatment. Of note, the mitophagy and necroptosis signaling pathways were obviously activated after Meth exposure (Figure 1F). Overall, PARL was one of the most significantly decreased proteins after Meth treatment (Figure 1G), which was validated at the gene and protein levels in primary neurons. In parallel, the protein levels of PGAM5, PINK1, Parkin, and the expression of MLKL, RIP1, and RIP3 in primary neurons were examined as well. Compared with the control group, these proteins were obviously increased after Meth exposure (Figure 1H,I), implying that mitophagy and necroptosis are involved in Meth-induced neurotoxicity.

3.2 PARL Reversed the Meth-Induced Neuronal Necroptosis in Primary Neurons

Since the critical role of PARL in PGAM5 modulation and the dramatic decline in PARL levels mediated by Meth, we investigated whether neuronal necroptosis can be rescued by PARL overexpression. Manifestly, as the Calcein-AM/PI double staining displayed, PARL overexpression dramatically attenuated the Meth-induced neuronal death (in red) (Figure 2A,B) and alleviated the membrane injury and TNF-α excretion, since the marked decrease in LDH and TNF-α levels in the cell supernatant (Figure 2C,D). The necrosome plays a critical role in necroptosis; therefore, it is reasonable to examine the necrosome formation. The colocalization of RIP1, RIP3, and MLKL was substantially increased in neurons subjected to Meth, an effect that could be markedly reversed by PARL overexpression (Figure S1G–J). Moreover, the RIP1, RIP3, and MLKL at gene and protein levels were significantly enhanced in the Meth-treated group; after PARL overexpression, the increased levels of RIP1, RIP3, and MLKL were strikingly ameliorated (Figure S1A–F and Figure 2E,F). The activated MLKL (p-MLKL) plays a vital role as it shuttles from the cytoplasm to the membrane, leading to membrane rupture. Intriguingly, p-MLKL notably colocalized with the mitochondrial membrane marker Tom20 after Meth treatment, and the phenomenon was reversed by PARL overexpression (Figure 2G,H). Hence, these results suggest that Meth exposure promotes necrosome generation and activates the MLKL, leading to mitochondrial membrane permeabilization, whereas PARL plays a critical role in alleviating these adverse outcomes.

3.3 PARL Attenuated Neuronal Mitochondrial Dysfunction Induced by Meth

Given the critical role of ROS in triggering necrosome formation [9], Mitosox, a probe specific for mitochondrial ROS, was employed. As expected, Meth treatment resulted in excessive mitochondrial ROS production; however, this effect could be substantially impeded by PARL overexpression, suggesting the beneficial roles of PARL in mitochondrial ROS inhibition (Figure 3A,D). Since mtROS is closely associated with both mitochondrial depolarization and fragmentation [36], the alterations in mitochondrial morphology by Mitotracker were examined. It showed that Meth significantly shortened the mitochondria, leading to mitochondrial fragmentation, whereas PARL overexpression dramatically suppressed this process and restored mitochondrial morphology (Figure 3C,F). To address the phenomenon of the Meth-induced alterations of mitochondrial morphology, the expression of mitochondrial fission proteins Drp1 and Fis1 at gene and protein levels was explored (Figure S1E,F; Figure 2E,F); in line with the phenotype, Drp1 and Fis1 levels were obviously increased, an effect that was ameliorated by PARL overexpression. Mitochondrial fragmentation was reported to affect mitochondrial membrane potential (MMP); thus, JC-1, a probe for MMP detection, was applied. As shown in Figure 3B,E, compared with the control neurons (normal MMP with red fluorescence), the Meth-treated neurons displayed significantly decreased red fluorescence and reciprocally enhanced green fluorescence, and the phenomenon could be reversed by PARL overexpression. Taken together, PARL overexpression prevents the Meth-induced mitochondrial injury by decreasing mtROS generation, mitochondrial fragmentation, and impeding mitochondrial membrane depolarization.

3.4 PARL Regulates Differential Cleavage of PINK1 and PGAM5 After Meth Exposure

The cleavage of PINK1 and PGAM5 by PARL is dependent on the status of mitochondria [29]. To decipher the cleavage properties modulated by PARL, the colocalization among PARL, PINK1, and PGAM5 was examined after Meth treatment. As shown in Figure 4A,B; Figure S3A,B, the colocalization between PARL and PINK1 was dramatically decreased and reciprocally markedly increased between PARL and PGAM5 in the Meth group. Besides, the full-length PINK1, full-length PGAM5, and cleaved PGAM5 protein levels were significantly increased, mediated by Meth, an effect that was substantially rescued by PARL overexpression (Figure 4E,F). Intriguingly, an increased colocalization of PINK1 and PGAM5 with Tom20 was observed after Meth exposure (Figure 4C,D; Figure S3C,D), suggesting a plausible connection between mitochondrial dysfunction and neuronal injury. Since PGAM5 acts as an exacerbating factor for mitochondrial fission and neuronal death, it is important to determine whether the PARL-mediated cleavage of PGAM5 plays a critical role in the regulation of necroptosis. As shown in Figure 4G,H, PGAM5 exhibited a significant increase in colocalization with p-MLKL following Meth exposure, and this effect was strikingly ameliorated by PARL overexpression. These results indicate that PARL overexpression ameliorates the Meth-induced PGAM5 cleavage and PGAM5-p-MLKL complex formation in the mitochondria outer membrane.

3.5 PARL Ameliorated Excessive Mitophagy and Autophagic Flux Defects in Primary Neurons

Since both PINK1 and cleaved PGAM5 stimulate the process of mitophagy, we then sought to investigate whether PARL overexpression ameliorates the excessive mitophagy induced by Meth. The immunofluorescence (IF) staining showed a significant increase in the colocalization of Parkin and Tom20 in primary neurons after Meth exposure. Moreover, LC3, a marker protein of the autophagosome, was strongly colocalized with Tom20, indicating excessive mitophagy. Notably, these effects were strikingly reversed by PARL overexpression (Figure 5C,D,G,H). Consistent with this phenomenon, the protein levels of Parkin, LC3II, and p62 markedly increased after Meth challenge, along with enhanced conversion from LC3I to LC3II, and these effects can be strikingly reversed by PARL overexpression (Figure 5A,B; Figure S3E,F). To further verify the autophagic defects, an mCherry-GFP-LC3-expressing adenovirus commonly used to examine autophagic flux was applied. It showed that Meth treatment exhibited a massive accumulation of colocalized mCherry-GFP yellow puncta in primary neurons, indicating defective autophagic flux. Noteworthily, PARL overexpression obviously improved the autophagic flux, evidenced by the decreased numbers of yellow puncta (Figure 5E,I). Consistently, Meth also reduced the colocalization of LC3 and LAMP1, confirming the blockage of autophagosome-lysosome fusion, and this process, similarly, can be strikingly reversed by PARL overexpression (Figure 5F,J).

3.6 Knockdown of PGAM5 Alleviated Necroptosis and Mitochondrial Dysfunction in Primary Neurons

PGAM5 was demonstrated to be the converging signal of multiple necroptosis pathways [32], and it is logical to investigate whether PGAM5 knockdown alleviated the Meth-induced mitochondrial injury and neuronal necroptosis. As expected, the knockdown of PGAM5 obviously decreased PI-positive cells, LDH, and TNF-α release (Figure 6A–D). Moreover, PGAM5 knockdown significantly impeded the Meth-induced necrosome formation, MLKL, RIP3, and RIP1 expression (Figure 6E,F; Figure S2B–D,G–J) and retarded MLKL transferring to the mitochondrial membrane (Figure 6G,H), thereby attenuating mtROS generation, MMP decrease, and mitochondrial fragmentation (Figure 7A–F). For the mitochondrial dynamics analysis, Meth obviously increased Drp1 and Fis1 at the mRNA and protein levels (Figure S2E,F; Figure 6E,F), whereas PGAM5 knockdown obviously reversed this effect. These data underscored the key roles of PGAM5 in Meth-induced mitochondrial dysfunction and neuronal necroptosis.

3.7 PARL Overexpression and PGAM5 Knockdown Ameliorated Meth-Induced Neuronal Necroptosis and Cognitive Defects in Mice

Having determined the crucial roles of the PARL-PGAM5 signaling axis in Meth-induced neuronal necroptosis in vitro, we then explored whether these effects can be validated in the in vivo study. Therefore, the PARL-AAV or ShPGAM5-AAV was stereotaxically injected into the CA1 and CA3 regions of the mice hippocampus (Figure S3K–N). After 3 weeks, mice were injected with binge doses of Meth (4 × 10 mg/kg, 2–3 h intervals) or saline (Figure S3O). The Y-maze test revealed that there was a significant decrease in the number of entries and duration of time spent in the novel arm induced by Meth compared to mice with PARL overexpression or PGAM5 knockdown (Figures 8A–C and 9A–C). Additionally, the new object recognition test showed that Meth exposure markedly reduced the time spent and number of explorations of the novel object, with a lesser recognition index. As expected, both PARL overexpression and PGAM5 knockdown could rescue the Meth-induced cognitive impairments (Figures 8D–G and 9D–G). Pathologically, the neuronal cells were loosely arranged and decreased, and Nissl bodies were lightly stained or even dissolved compared with PARL overexpression and PGAM5 knockdown cells (Figures 10A and 11A; Figure S3I,J), suggesting that PARL overexpression or PGAM5 knockdown exerts a salutary effect on cognitive and spatial learning against Meth-induced memory impairments. Meanwhile, TEM was used to evaluate the formation and accumulation of mitophagosomes in the mouse hippocampus. As shown in Figures 10B and 11B, mitochondria in Meth-treated neurons displayed fragmentation with reduced or no cristae structures and ruptured membranes, which were engulfed by the accumulated autophagosomes, while PARL overexpression or PGAM5 knockdown remarkably restored the Meth-induced subcellular structure impairments. On the basis of the aforementioned results, the expression levels of mitophagy-related proteins, including PINK1, PGAM5, Parkin, LC3II/I, and p62 in the hippocampus were examined. In line with the results represented in TEM, these proteins were increased in the Meth group, an effect that was dramatically reduced by PARL overexpression (Figure 10C,D; Figure S3G,H). Additionally, mitochondrial dynamics play a key role in the response of cells to oxidative stress, and mitochondrial fission is necessary to trigger mitophagy. Unsurprisingly, PARL overexpression or PGAM5 knockdown significantly suppressed the Meth-induced increase of mitochondrial fission proteins, Drp1 and Fis1. Furthermore, the necroptosis proteins, including MLKL, RIP1, and RIP3, were significantly inhibited as well (Figures 10C,D and 11C,D). Taken together, these findings indicate that targeting PARL-PGAM5 signaling plays pivotal salutary effects against Meth-mediated brain injury.