The generation of Plppr5 knockout mice was conducted in the GemPharmatech Co., Ltd. (Nanjing, China) using CRISPR/Cas9 technology as has been reported before (Sun et al., 2021). KO and littermate wildtype (WT) mice of 10 days were used in our experiments and only male mice were used. All experiments were conducted with approval of the University of Soochow University Animal Care and Use Committee (ethical code:SUDA20201020A01). Mice were tested blindly for genotype. Hypoxic-Ischemic Brain Damage (HIBD) is based on the mature Rice Vannucci model (Rice et al., 1981). Briefly, 10-day old mice were anesthetized, then, the left common carotid artery was carefully separated and cut in the middle after double ligation with a 5–0 suture. Lastly, the skin was sutured. The entire procedure took no more than 6 min. The mice were returned to the dam to recover for one hour and then placed in a hypoxic chamber for two hours (37°C, 8% oxygen, 92% nitrogen). The sham-operated (Sham) mice were only treated by isolating the left common carotid artery without ligation, and placed in a similar equipment without hypoxic (Figure 1). A total of thirty male WT mice and thirty Plppr5^–/– mice were randomly selected and divided into the following four groups (fifteen in each group): WT group (wild-type sham operation group, n = 15), KO group (Plppr5^–/– mice with sham operation group, n = 15), WT + HIBD group (wild-type mice with hypoxic-ischemic group, n = 15), KO + HIBD group (Plppr5^–/–mice with hypoxic-ischemic group, n = 15).

Three mice in each group were anesthetized and executed 24 h after HIBD. The brains were removed and sliced coronally into 2-mm-thick sections and incubated in 1% 2,3,5-triphenyltetrazolium chloride (TTC) (Sigma, St. Louis, MO, United States) at 37°C in the dark for 10 min. Then fixed overnight in 4% formaldehyde solution. The TTC-stained brain sections were photographed and the infarct volumes were measured by Image J (an image analysis software, United States). The degree of cerebral infarction is presented as the percentage of infarction volume to total brain volume (Wang et al., 2016; Wong et al., 2018; Omur et al., 2019).

To explore the difference of the seizure latency (seizure threshold) between hypoxic-ischemic brain injury and sham operation, and the effect of Plppr5 knockout.

Four weeks after HIBD, six mice in each group were randomly selected to test. The specific operation refers to the literature (Jin et al., 2018): the mice were injected with penicillin (5.1 × 10⁶U/kg/d, i.p.), and the time from the injection of penicillin to the convulsive seizure of the mouse were recorded. The observation time was 90 min. The intensity of epilepsy was determined according to the Racine classification (Racine, 1972). Briefly, stage 1, stereotype mouth movement, eye blinking and/or mild facial clonus; stage 2, head nodding and/or sever facial clonus; stage 3, unilateral forelimb clonic twitching, but no hindlimb erection; stage 4, clonic convulsions in the forelimbs with rearing; and stage 5, generalized tonic-clonic seizures with falls, loss of postural control. Moreover, after the seizure started, the mouse was injected with 4% chloral hydrate immediately.

The neo-Timm’s staining is based on the sulfide/silver method. Three mice were randomly selected from each group four weeks after HIBD. After anesthesia, they were fixed in the supine position. The chest was opened along the midline and the heart was exposed. Then, PBS, 0.4% Na₂S, 4% paraformaldehyde, 0.4% Na₂S were injected into the heart in turn. Then removed the brain and put into 4% paraformaldehyde for fixation, and selected according to the hippocampal positioning of Wiqas et al. (2020). The section with the largest area of the dorsal hippocampus (Bregma-1.3 to -3.7 mm, The mouse brain in stereotaxic coordinates. Amsterdam, the Netherlands, Boston, MA: Elsevier/Academic Press.). Incubate the Timm staining solution (90 ml of 50% gum Arabic with 7.65 g of citric acid, 7.05 g of sodium citrate in 22 ml ddH₂0, 1.5 ml of 17% AgNO₃, 5.3 g of hydroquinone in 90 ml ddH₂O) at 37°C for 50 min in the dark. After the incubation, rinse it in running water and pass the 70%, 80%, 90% and 100% staining solution one by one. It is dehydrated in ethanol solution and mounted with neutral resin (Hsu and Buzsáki, 1993; Ni et al., 2009b; Yang et al., 2013). After drying, observe and take pictures under a light microscope. Semi-quantitative scoring of mossy fiber sprouting (MFS) in hippocampal DG and CA3 regions was performed (the scoring was done by persons who did not know the experimental group), and 5 consecutive brain slices were observed for each brain tissue CA3 area and DG area (Holmes et al., 1999; Ni et al., 2009b).

HT22 cells (neuronal line of mouse hippocampus) were obtained from the Cell Bank of the Institute of Cell Biology, Chinese Academy of Sciences. Cells were cultured in medium composed of DMEM, 10% FBS, 100U/mL penicillin, and 100mg/mL streptomycin in the cell culture incubator at 37°C with 5% CO₂.

For viral infection in vitro, when cells grew to 60–70% confluency, the HT22 cells were transfected with medium containing LV-Plppr5-RNAi (MOI = 100), and empty plasmids LV-control-RNAi (normal control;MOI = 100; Genechem). After 24 hours of transfection, serum-free transfer solution was replaced by complete medium to culture for 48 h. The cells were cultured for 3 days and then subjected to OGD/R. Oxygen and glucose deprivation/reoxygenation (OGD/R) is an accepted in vitro model for simulating HIBD. First, HT22 cells were cultured in glucose-free DMEM in a hypoxia incubator chamber (Billups-Rothenberg, Inc.) with 1% O₂, 5% CO₂, and 37°C for 6 h. Then, the medium was replaced with complete DMEM and the cells were incubated in an incubator with 5% CO₂ at 37°C for another 4h to simulated the reperfusion. Cells in the control group were treated identically except that they were not exposed to OGD (Wei et al., 2017). Briefly, it was divided into four groups: Sh-control, Sh-Plppr5, Sh-control + OGD/R, Sh-Plppr5 + OGD/R.

Brain tissues from the left hippocampus region of the mice or cells were collected and added with RIPA lysis buffer (Beyotime Biotechnology, China) (Rafało-Ulińska et al., 2020). Lyse on ice for 1 h after shaking. The supernatant was collected after centrifugation at 4°C and 12,000 × g for 30 min. Protein concentrations were determined with BCA protein assay kit (Beyotime Biotechnology, China). Next, the protein extracts were boiled at 100°C for 10 min. Equal amount of protein from each sample was resolved on SDS–PAGE, trans-blotted onto polyvinylidene fluoride membranes (Millipore), and subjected to immunoblot assay by primary antibodies followed by secondary antibodies. The following antibodies were used for western blotting: rabbit polyclonal anti-ZnT1 (SLC30A1) antibody (1:1000; allomone labs, AZT-011), and mouse monoclonal beta actin antibody (1:5000; proteintech,66009-1-Ig) (Rafalo-Ulinska et al., 2016). The bands were visualized by an ECL detection kit (Beyotime Biotechnology, China) using AmerSham Imager 600 (GE, United States). β-actin was used as a loading control.

Index of oxidative injury (lipoperoxidative damage) was determined by malondialdehyde (MDA) and antioxidant capacity was estimated by superoxide dismutase (SOD). After treatment, the cells were collected to detect MDA and SOD levels using a MDA (A003-2) or SOD (A001-1) kit (Nanjing Jiancheng Biotechnology Research Institute, Jiangsu, China) according to the manufacturer recommendations (Dai et al., 2018; Yan et al., 2019).

The intramitochondrial production of ROS in live HT22 cells was detected using the fluorescent MitoSox probe (M36008;Invitrogen) (Roelofs et al., 2015; Chen et al., 2019). The cells were reacted with a 5 μM working solution of MitoSOX Red for 10 min at 37°C, and then carefully washed twice with HBSS. Then, cells were stained with the Hoechst 33342 (1 mg/ml, Beyotime Biotechnology, China) dye for 10 min at 37°C and washed twice with HBSS. Images were obtained with a Laser Scanning Confocal Microscope (OLYMPUS, Japan), and the fluorescence intensity was evaluated with the Image J software (National Institutes of Health, MD, United States). The data are shown as the mean intensities.

MMP was estimated by flow cytometry after staining with JC-1 fluorescent dye (Beyotime Biotechnology, China). MMP is high and JC-1 predominantly appears as red fluorescence when the cell is in a normal state. When the cell is in an apoptotic or necrotic state, the MMP is reduced and JC-1 appears as a monomer indicated by green fluorescence. Approximately 1 × 10⁵ cells in 6-well plates were treated separately. The cells were then washed with PBS and incubated with JC-1 working solution for 20 min at 37°C in the dark. After the incubation, the supernatant was discarded after centrifugation at 600 g/min at 4°C for 3 min and resuspended in 500 μl JC-1 staining buffer. The stained cells were analyzed by flow cytometry to determine the change in the florescence from red to green (Jin et al., 2018).

Zinquin ethyl ester is thought to be TSQ analog and can form a fluorescent complex with zinc ions to detect intracellular free zinc (Grabrucker et al., 2014). The cells were cultured in confocal microplates and treated separately 500 μl of 2.4 μM Zinquin ethyl ester solution (diluted in DMEM, Dojindo Molecular Technologies, Kumamoto, Japan) was added to the cells and incubated for 60 min at 37°C and washed twice with HBSS. Images were obtained with a Laser Scanning Confocal Microscope, and the fluorescence intensity was evaluated with the Image J software (National Institutes of Health, MD, United States). The data are shown as the mean intensities.

All data in this study were statistically analyzed using SPSS 23.0 software. The measurement data is represented by the mean ± standard deviation, and the normality of the Shapiro-Wilk test is used. T test or analysis of variance was used for comparison between groups for normal distribution; the rank sum test was used for comparison between groups for abnormal distribution. The chi-square test was used to compare the count data between groups. *P < 0.05, **P < 0.01 indicates that the difference is statistically significant. The data are all obtained by more than three independent experiments. All statistical images are produced using GraphPad Prism version 8.0 software.

Both WT and Plppr5^–/– mice showed obvious infarct on the left side of the brain 24 h after HIBD, and the infarct volume of Plppr5^–/–mice was larger than WT mice (Figures 2A,B).

Results show that mice in each group had seizures successively. No mice died in the WT group, two mice died in KO group and WT + HIBD group, respectively (mortality rate of 33.3%), three mice died in KO + HIBD group (mortality rate of 50%). Fisher exact test (2 × C) showed that there was no significant difference in mortality among the four groups (P = 0.392). The seizure threshold of Plppr5^–/– mice was significantly lower than that of WT mice (P = 0.0102), which shows that Plppr5 knockout decreased the seizure threshold induced by penicillin in mice. After HIBD treatment, both WT mice and Plppr5^–/– mice had a lower seizure threshold than the corresponding sham group (P < 0.01). Hypoxic-ischemic injury can also reduce the seizure threshold and Plppr5^–/– mice were more likely to have convulsion (P < 0.01), Plppr5 knockout can aggravate the lower degree of convulsion threshold induced by HIBD. However, there was no significant difference in the intensity of seizures among all groups (P > 0.05) (Figures 2C,D).

As can be seen from the results of Timm staining, for WT mice, four weeks after HIBD, the mossy fiber sprouting (MFS) scores in bilateral hippocampus CA3 and DG were higher than the sham group (P < 0.05). For Plppr5^–/– mice, four weeks after HIBD, the MFS scores in bilateral hippocampus DG were higher than the sham group (P < 0.05). Furthermore, the MFS scores in bilateral hippocampus CA3 were higher in KO + HIBD group than in WT + HIBD group (KO + HIBD vs. WT + HIBD, P < 0.05) (Figure 3).

Western blot results showed that the expression level of ZnT1 was significantly decreased in Plppr5^–/– mice with sham treatment than WT mice (P < 0.05). And the expression level of ZnT1 in WT mice and Plppr5^–/– mice four weeks after HIBD treatment were significantly higher than those in mice treated with sham operation (P < 0.05). Four weeks after HIBD treatment, Plppr5^–/– mice had higher expression of ZnT1 than WT mice (KO + HIBD vs. WT + HIBD, P < 0.05) (Figure 4A and Supplementary Figures 1S1,S2). We found the same results at the cellular level (Figure 4B and Supplementary Figures 1S3,S4).

In Sh-Plppr5 group, the intracellular free zinc content was significantly higher than those in Sh-control. OGD/R treatment can increase the intracellular free zinc in both groups. However, the content of intracellular free zinc in Sh-Plppr5 + OGD/R group was higher than that in Sh-control + OGD/R group, but the difference was not statistically significant (Figure 4C).

Plppr5 knock out can significantly increase the content of ROS in HT22 cells (Sh-Plppr5 vs. Sh-control, P < 0.01). Meanwhile, after OGD/R treatment, the content of ROS was significantly higher than that in the respective control group in Sh-Plppr5 and Sh-control group (P < 0.01). What’s more, after OGD/R treatment, the content in Sh-Plppr5 group increased more significantly than that in Sh-control group (Figure 5A).

The experimental results showed that the mitochondrial membrane potential of Sh-Plppr5 group was significantly higher than that of Sh-control group. After OGD/R treatment, the mitochondrial membrane potential was significantly lower than that in the respective control group in Sh-Plppr5 and Sh-control group. Furthermore, the decrease degree of Sh-Plppr5 group after OGD/R was more obvious than that of Sh-control group (Figure 5B).

Results showed that there was no statistical difference in MDA content between Sh-control group and Sh-Plppr5 group (P = 0.188). However, OGD/R treatment can significantly increase the MDA content in both groups, especially in Sh-Plppr5 group (Figure 5C, sh-control vs. sh-control + OGD/R, P = 0.023; sh-Plppr5 vs. sh-Plppr5 + OGD/R, P = 0.0017).

Knockdown of Plppr5 decreased the activity of SOD. Compared with Sh-control group, SOD activity in Sh-Plppr5 group was significantly decreased (P = 0.0154). Meanwhile, the SOD activity were decreased after OGD/R treatment in both groups and the decrease was more obvious in the Sh-Plppr5 group (Figure 5D, sh-control vs. sh-control + OGD/R, P = 0.022; sh-Plppr5 vs. sh-Plppr5 + OGD/R, P = 0.0013).