2.1 Establishment of acute seizure epilepsy model in rats and lentivirus (LV) injection

A total of 32 male Wistar rats (8 weeks old) were purchased from Charles River Laboratories. All animals were used in strict accordance with national animal experiment requirements that were approved by the animal ethics association of the Third Xiangya Hospital (Hunan, China). Rats were randomly divided into four groups (n = 8): Control group, epileptic model group, LV-Negative control (NC) (control LV) group, and LV-prodynorphin (PDYN) (PDYN overexpression LV) group. Rats in the control group received an intraperitoneal injection with PBS. The epilepsy was induced by pilocarpine injection as described in our previous study (Dai et al., 2019; L. Liu et al., 2020). Briefly, the rats in the epilepsy model group were injected intraperitoneally (ip) with lithium chloride (127 mg/kg; Sigma-Aldrich) 18 h before the first administration of pilocarpine (30 mg/kg, ip, Sigma-Aldrich). Pilocarpine (10 mg/kg, ip) was then repeatedly injected every 30 min until the rats developed seizures. One hour after the onset of the status epileptics, rats were injected with diazepam (10 mg/kg, ip) to terminate seizures.

Rats in the LV-NC group and LV-PDYN group received a stereotaxic intrahippocampus injection of LV-NC and LV-PDYN (GeneChem), respectively, as described in our previous study (L. Liu et al., 2020). After LV injection, all of the rats were kept in a standard environment for 1 week, and then given pilocarpine to observe acute seizure features. Animals were killed by decapitation 24 h after status epilepticus. The hippocampus was removed and stored in liquid nitrogen for TdT-mediated dUTP-biotin nick end labeling (TUNEL) staining and extraction of RNA and protein.

Moreover, the baseline levels of PDYN expression before the pilocarpine administration are shown in Supporting Information: Figure 1. The animals were killed on the 7th day after LV injection (1 day before the establishment of epilepsy model). Compared with LV-NC group, the messenger RNA (mRNA) and protein of PYDN were upregulated 2.4-fold in LV-PDYN group.

2.2 TUNEL staining

TUNEL staining was performed to analyze the apoptosis in rat hippocampus. Briefly, the 5 μm-thick sections were dewaxed, rehydrated, and incubated with proteinase K (20 µg/ml) at 37°C for 20 min. After that, the sections were washed with phosphate-buffered saline (PBS) twice and then incubated with the TUNEL reaction mixture provided by the in situ cell death detection kit (Roche). After washing with PBS three times, the sections were stained with diaminobenzidine (Sigma-Aldrich) for 10 min. Finally, the sections were sealed and images were acquired by an Olympus microscope. TUNEL positive cells (brown) was counted and compared with the total cells using Image-Pro Plus 6.0 software and expressed as a percentage (from three mice, respectively).

2.3 Immunofluorescent staining

The sections were dewaxed, rehydrated, and then immersed into citrate buffer for antigen repair. After that, the sections were blocked with 2% bovine serum albumin and then incubated overnight at 4°C with the following primary antibodies: anti-LC3 (1:200; Invitrogen) and anti-NeuN (1:100; Abcam), followed by incubation with the secondary antibodies Alex Fluor® 488-labeled secondary antibody (green; 1:1000; Abcam), Alex Fluor®647-labeled secondary antibody (red; 1:1000; Abcam) and 4,6-diamino-2-phenyl indole (DAPI) (blue; 1:1000) for 1 h at room temperature. Slides were observed by a fluorescent microscope (Olympus IX-71) equipped with a Canon EOS digital camera. Using Image J software, the LC3 and NeuN positive cells were separately counted from three representative slides (from three mice, respectively).

2.4 Cell isolation, culture, and identification

Hippocampal tissue of normal rats was isolated, dissected, ground, and placed on the cell plate treated with poly-d-lysine. The neural cells were cultured in the medium supplemented with 2% B27 (Neuroblastal/B27) and 10% FBS (Gibco) in the 37°C and 5% CO₂. After plating for 3 days, the hippocampal neuron was treated with cytosine arabinoside (5 μm; sigma Aldrich) to inhibit astrocytes proliferation, and the medium was changed every 3 days. Hippocampal neurons cultured for 7−18 days in vitro were used for subsequent experiments.

Morphology of hippocampal neurons was observed at different cultured time points, and hippocampal neurons were identified by immunofluorescence staining of neuronal microtubule associated protein-2 (MAP-2), which suggested that hippocampal neurons in this study have been successfully isolated and have strong cell viability (Supporting Information: Figure 2).

2.5 Cell treatment

Epileptiform activity was induced in vitro by magnesium-free (Mg²⁺-free) physiological solution (PH7.3) supplemented with 2.5 mM KCl, 145 mM NaCl, 2 mM CaCl₂, 10 mm HEPES, 10 mM glucose, and 0.002 mM glycine. After being cultured in Mg²⁺-free solution for 3 h, the neurons were transferred to the conventional Neuroblastal/B27 medium for further culture.

To investigate the role of KOR activation in epilepsy or the effect of mTOR pathway in dynorphin-A mediated effect in epilepsy, we treated neurons exposed to Mg²⁺-free with Dyn-A (dynorphin-A, KOR agonist; Sigma-Aldrich; 1 µM, 1 h), Guanidinonaltrindole (GNTI) (KOR antagonist; R&D system; 2 nM, 1 h), or rapamycin (mTOR pathway inhibitor; Sigma-Aldrich; 20 nM, 1 h) alone or in combination.

2.6 Cell viability assay

Cell viability assay was performed using the Cell Counting Kit-8 (CCK-8; Beyotime). Briefly, the primary mouse hippocampal neurons were seeded in the 96-well plates and were exposed to Mg²⁺-free and/or treated with dynorphin-A, GNTI or rapamycin. After the treatments, CCK-8 reagent (10 μl) was added to each well and incubated at 37°C for 2 h. Optical delnsity values were evaluated at 450 nm using a microplate absorbance reader (Thermo Fisher Scientific). All experiments were repeated three times independently.

2.7 Flow cytometry analysis

To evaluate cell apoptosis in the primary mouse hippocampal neurons, flow cytometry analysis was performed using the FITC-conjugated Annexin V apoptosis detection kit (BD Pharmingen) according to the manufacturer's instructions. After Mg²⁺-free and/or treated with dynorphin-A, GNTI or rapamycin treatments, the hippocampal neurons were collected, washed, and then suspended in the binding buffer. Annexin V-FITC and PI (Sigma-Aldrich) were then added to the binding buffer and incubated in the dark at room temperature for 15 min. Finally, the apoptosis rate of hippocampal neurons was analyzed by flow cytometry and cell quest software (BD Biosciences). All experiments were repeated three times independently.

2.8 Real-time polymerase chain reaction (RT-PCR) analysis

For tissue RNA extraction, the hippocampal tissues of rats in each group were isolated, ground under liquid nitrogen, and then RNA was extracted. For cell RNA extraction, after the treatments, the hippocampal neurons were collected and then suspended in the PBS. Total RNA was extracted from hippocampus tissue or hippocampal neurons using TRIzol (Invitrogen). Then, RNA was reverse-transcribed to complementary DNA (cDNA) using a Transcriptor First Strand cDNA synthesis kit (Roche). The expression of PYDN was using the SYBR premix (Takara) in the Applied Biosystems 7500 PCR system. The glyceraldehyde-3-phosphate dehydrogenase was used as the internal control for mRNA. The relative expression of genes was analyzed using $${2}^{\mbox{--}\unicode{x02206}\unicode{x02206}{C}_{{\rm{T}}}}$$ method. All experiments were repeated three times independently.

The total protein of hippocampus tissue and cells were extracted by radio-immunoprecipitation assay lysis buffer (Sigma Aldrich). Bicinchoninic acid assay protein quantitative Kit (Invitrogen) was used to quantify the total protein concentration. The extracted proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to the poly vinylidene fluoride (PVDF) membrane. Tris-hcl buffer plus tween-20 buffer containing 5% skim milk was used to block for 1 h at room temperature. Then, the PVDF membranes were incubated with primary antibody overnight at 4°C. The primary antibodies were obtained from Abcam or Santa Cruz Biotechnology, including anti-PYDN (ab11137), anti-p-mTOR (ab109268), anti-mTOR (ab134903), anti-Beclin-1(ab210498), anti-LC3B (ab192890), anti-Caspase-3 (ab39754), anti-Bax (ab182733), and anti-Bcl-2 (ab194583). Then the secondary antibody was incubated at room temperature for 1 h. The β-actin and α-tubulin were used as internal references. Immuno Western chemiluminescence HRP substrate (EMD Millipore) was used to display the protein bands. The protein bands were quantified by ImageJ software.

2.9 Green fluorescent protein-light chain 3 (GFP-LC3) immunofluorescence

The GFP-LC3-expressing plasmid (pEGFP-LC3) was transiently transfected into the stably transfected neuron cells, with the help of Lipofectamine 2000 reagent (Invitrogen). LC3 protein was indicated by the aggregated green fluorescent particles. Fluorescence images were captured and analyzed under a fluorescence microscope (Nikon Corporation). Coverslips were mounted on glass slides, and the GFP-LC3 punctated dots in a total of >3 cells were counted. The number of GFP-labeled autophagosomes was presented as the dots' number in one cells. All experiments were repeated three times independently.

2.10 Statistical analysis

All statistical analyses were conducted using GraphPad Prism 7.0. A statistical comparison of different treatment groups was determined by one-way analysis of variance followed by Tukey's posttest. p < .05 were considered significant.

3.1 Overexpression of PDYN alleviated neuronal apoptosis in epileptic rats

Rats were randomly divided into four groups: Control, epileptic model, LV-NC (control LV injection), and LV-PDYN group (PDYN overexpression LV injection). At first, we have detected the PDYN expression in the hippocampus of epileptic rats, before details of LV stereotaxic injections. The baseline levels of PDYN expression before the pilocarpine administration are shown in Supporting Information: Figure 1. The animals were killed on the 7th day after LV injection (1 day before the establishment of epilepsy model). Compared with LV-NC group, the mRNA and protein of PYDN were upregulated 2.4-fold in LV-PDYN group.

Next, to investigate the role of overexpression PDYN on neuronal apoptosis in epileptic rats, the epilepsy model was induced by pilocarpine injection, and Rats in the LV-NC group and LV-PDYN group received a stereotaxic intrahippocampus injection of LV-NC and LV-PDYN, respectively, before the establishment of the epilepsy model. The PDYN expression in the hippocampus of epileptic rats was detected by quantitative reverse transcription-polymerase chain reaction and western blot. As shown in Figure 1a–c, the PDYN mRNA, and protein expression levels in the hippocampus of epileptic rats were significantly lower than that in the control group. As expected, compared with the LV-NC group, LV-PDYN administration significantly restored the reduction of PDYN expression in the epilepsy model rats. Then, the neuronal apoptosis in epileptic model rats was measured by TUNEL staining. As shown in Figures 1d,e, intrahippocampus injection of LV-PDYN markedly attenuated the pilocarpine-induced hippocampal neuronal apoptosis, which suggested that overexpression of PDYN could alleviate pilocarpine-induced neuronal apoptosis in rats.

3.2 Overexpression of PDYN activated mTOR signaling pathway and inhibited neuronal autophagy in epileptic rats

Autophagy is a catabolic pathway in which dysfunctional organelles and cellular components are degraded by lysosomes. During this process, cytoplasmic LC3 translocates to the autophagosome membrane. Therefore, cells undergoing autophagy can be identified by visualizing fluorescent labeled LC3. To visualize the distribution and cellular localization of LC3 in the hippocampus of rats, double-labeling immunofluorescence experiments using an antibody to LC3 (which does not distinguish LC3-I from LC3-II) in combination with NeuN marker were performed. LC3 labeling was localized in most NeuN-positive neuronal cells (red). We observed a marked LC3 immunoreactivity (green) in the epileptic model group, which was decreased by LV-PDYN administration (Figure 2a). Mammalian target of mTOR (a serine/threonine-protein kinase) is a major repressor of autophagy. Here，the p-mTOR and mTOR proteins expression were detected by western blot. As shown in Figures 2b,c, the ratio of p-mTOR/mTOR protein was decreased significantly in the epileptic model group, while LV-PDYN administration restored the reduction of p-mTOR protein, which demonstrated that overexpression of PDYN could activate the mTOR signaling pathway. Subsequently, western blot analysis data revealed that autophagy-related proteins (LC3-II/LC3-I and Beclin-1) were upregulated in epileptic model rats, but reduced by LV-PDYN administration (Figure 2d–f). These data indicated that overexpression of PDYN could activate the mTOR signaling pathway and inhibit neuronal autophagy in acute seizure epileptic rats.

In addition, to assess receptor specificity, the effects of PDYN overexpression have be tested in the presence of pharmacological selective agents. As shown in Supporting Information: Figure 2, we found the significantly decreased KOR in the hippocampus of epileptic rats. Overexpression of PDYN upregulated the expression of KOR, µ-opioid receptor (MOR) and δ-opioid receptor (DOR). According to this result, we can conclude that dynorphin/KOR inhibits neuronal autophagy by activating mTOR signaling pathway to exert antiepileptic effect.

3.3 Dyn-A activation of KOR inhibited Mg²⁺-free-induced seizure-like neuron apoptosis

To verify the effect of KOR activation on neurons, we first isolated and cultured rat hippocampal neurons. As shown in Supporting Information: Figure 3A, morphology of hippocampal neurons was observed at different cultured time points. Under the inverted microscope, it can be seen that all hippocampal neuronal cells were round at the beginning of implantation, the cell body was transparent, uniform in size, and suspended in the implantation solution. After 24 h of planting, the cells adhered well, the cell body increased, the refraction was good, spindle or irregular, and most of the cells showed short protrusions. 48 h after planting, the cell body was further enlarged, the processes were significantly elongated, the dendrites and axons were clearly visible, and the contact between adjacent cells was formed. Three days after planting, most of the cells had typical neuronal morphology, and the processes further extended and intertwined into a network. On the 7th day, the cell body was full and bright, the connection of processes was closer, and the cells showed a centralized distribution trend. In addition, on the 8th day of hippocampal neuron culture, neuronal MAP-2 was used for cellular immunofluorescence staining. Randomly select five visual fields under the fluorescence microscope, count the positive cells (representing neurons and red fluorescence) and nuclei (representing the total number of cells and blue fluorescence) respectively, and calculate the positive rate, that is, the purity of cultured neurons. Statistical analysis showed that the maximum purity of neurons reached 90%, the minimum was 85%, and the median (representing the average purity of neurons) was 88% on the 8th day. Supporting Information: Figure 3B shows the nuclei and MAP-2 positive cells under the fluorescence microscope. These images suggested that hippocampal neurons in this study have been successfully isolated and have strong cell viability.

Then the rat neuron cells were randomly divided into five groups: Control, Mg²⁺-free, Mg²⁺-free+ Dyn-A, Mg²⁺-free+GNT1, Mg²⁺-free+ Dyn-A + GNT1. As shown in Figure 3a, the cell viability decreased significantly in the Mg²⁺-free group. Dyn-A activation of KOR reversed the decreased cell viability, while GNT1 further inhibited the cell viability in rat neuron cells exposed to Mg²⁺-free solution. Most notably, GNTI significantly abrogated the Dyn-A-mediated cell viability change in rat neuron cells exposed to Mg²⁺-free solution. Then, the flow analysis results show remarkably increased cell apoptosis in the Mg²⁺-free group (Figure 3b,c). Dyn-A reversed the increased cell apoptosis, while GNT1 further enhanced the cell apoptosis in rat neuron cells exposed to Mg²⁺-free solution. On the contrary to the trend of cell viability, GNTI significantly abrogated the Dyn-A-mediated cell apoptosis change in rat neuron cells exposed to Mg²⁺-free solution. In addition, the apoptosis-related protein (Caspase-3, Bax and Bcl-2) expression was detected by western blot (Figure 3d,e). In line with the trend of apoptotic cell proportion, apoptotic protein (Caspase-3 and Bax) were increased significantly in the Mg²⁺-free group, and Dyn-A reversed the increased protein expression, while GNT1 further increased the protein expression in rat neuron cells exposed to Mg²⁺-free solution. The expression of antiapoptotic protein Bcl-2 was opposite to that of apoptotic protein Caspase-3 and Bax, showing a marked decline in the Mg²⁺-free group, and Dyn-A reversed the decreased Bcl-2 protein expression, while GNT1 further inhibited the Bcl-2 protein expression. These results indicated that Dyn-A activation of KOR inhibited Mg²⁺-free-induced seizure-like neuron apoptosis.

3.4 Dyn-A activation of KOR activated mTOR signaling pathway and inhibited seizure-like neuron autophagy

To investigate the effect of Dyn-A activation of KOR on mTOR signaling pathway and seizure-like neuron autophagy in rat neuron cells exposed to Mg²⁺-free solution, the p-mTOR, mTOR, LC3-I, LC3-II, and Beclin-1proteins expression were detected by Western blot. As shown in Figures 4a,b, the ratio of p-mTOR/mTOR protein significantly decreased in the Mg²⁺-free group, which was reversed by Dyn-A but further decreased by GNT1. And GNTI significantly abrogated the Dyn-A-mediated change on the ratio of p-mTOR/mTOR protein in rat neuron cells exposed to Mg²⁺-free solution. Next, the Western blot analysis data showed that the ratio of LC3-II/LC3-I and the Beclin-1 protein expression were remarkably upregulated in the Mg²⁺-free group, which was reversed by Dyn-A but further increased by GNT1. And GNTI significantly abrogated the Dyn-A-mediated change on the ratio of LC3-II/LC3-I and the expression of Beclin-1 protein in rat neuron cells exposed to Mg²⁺-free solution (Figure 4c–e). Moreover, we also found that the rat neuron cells following Mg²⁺-free exhibited stronger GFP-LC3 puncta and increased percentage of GFP-LC3 cells when compared with the control neuron cells. Dyn-A reversed the increased cell autophagy, while GNT1 further enhanced the cell autophagy in rat neuron cells exposed to Mg²⁺-free solution (Figure 4f). These data indicated that Dyn-A activation of KOR could activate the mTOR signaling pathway and inhibit seizure-like neuron autophagy.

3.5 Dyn-A activation of KOR inhibited Mg²⁺-free-induced seizure-like neuron autophagy by activating the mTOR signaling pathway

As we have known that Dyn-A activation of KOR could activate the mTOR signaling pathway and inhibit seizure-like neuron autophagy, so does Dyn-A activation of KOR inhibit seizure-like neuron autophagy by activating the mTOR signaling pathway. Here, we used the mTOR inhibitor, Rapamycin to prove that Dyn-A activation of KOR inhibited Mg²⁺-free-induced seizure-like neuron autophagy by activating the mTOR signaling pathway. As shown in Figure 5a–c, compared with the Mg²⁺-free group, the ratio of LC3-II/LC3-I and the Beclin-1 protein expression in group A were significantly decreased by Dyn-A but further increased by Rapamycin. And Rapamycin significantly abrogated the Dyn-A activation of KOR-mediated the ratio of LC3-II/LC3-I and the Beclin-1 protein expression change in rat neuron cells exposed to Mg²⁺-free solution. Accord to autophagy-related proteins changes, Rapamycin also significantly abrogated the Dyn-A activation of KOR-mediated cell viability and apoptosis changes in rat neuron cells exposed to Mg²⁺-free solution (Figures 5d–h). These data supported our hypothesis that Dyn-A activation of KOR inhibited Mg²⁺-free-induced seizure-like neuron autophagy by activating the mTOR signaling pathway.