Phactr4 promotes oxidative stress and behavioral disorder caused by chronic stress via regulating PP1/GSK3-β pathway

Abstract

Oxidative stress, defined as a process triggered by an imbalance between the accumulation of free radicals and antioxidant defences, has been considered implicated in many neurological disorders, including major depressive disorder (MDD). In the present study, we demonstrated that the expression of phosphatase and actin regulatory factor 4 (Phactr4) was increased within the dentate gyrus (DG) region of the hippocampus of the chronic stress-induced depressive mice. Phactr4 has been shown to enhance oxidative stress in the brain by interacting with protein phosphatase 1 (PP1) and synergistically reducing the level of phosphorylation of glycogen synthase kinase 3β (GSK3β), thus enhance the susceptibility to stress stimulation in mice. Knocking down phactr4 in the hippocampal DG regions can suppressed GSK3β activation, alleviate oxidative stress, and further improve the depression-like behaviors in mice. More interestingly, we further found physical exercise can downregulate the level of Phactr4, reduce the accumulation of reactive oxygen species (ROS) in the brain, ameliorate neuronal damage, and reverse depressive-like behaviors in mice. These findings suggest that physical exercise may promote the restoration of oxidative stress in brain and ameliorates depressive behaviors in mice by down-regulating the Phactr4-PP1-GSK3β pathway.

Graphical abstract

1. Introduction

As one of the most disabling mental diseases in the world, the clinical manifestations of major depressive disorder (MDD) include persistent depressed mood, anhedonia and cognitive dysfunction, which severely impair patients' quality of life and increase the risk of suicide [1,2]. The complexity of depression is considered mainly attribute to its multidimensional pathophysiological mechanisms, for example including the disorder of the neurotransmitter system, hypothalamic-pituitary-adrenal (HPA) axis dysregulation, neuroinflammation, neurotrophic factor deficiency or gut-brain axis dysbiosis [[3], [4], [5], [6]]. However, no effective strategies currently exist for the treatment of depression in clinic. Therefore, the exploration of novel therapeutic targets and the integration of intervention strategies have become the critical direction of current investigation of depression.

The relationship between oxidative stress and depression has been validated in both preclinical and clinical studies [7,8]. Elevated levels of reactive oxygen species (ROS) have been observed in the peripheral blood and brain tissue of depressed patients [9,10]. Furthermore, Animal experimental studies have further revealed that chronic stress can induce oxidative damage in the prefrontal cortex and hippocampus, which is accompanied by an increased release of pro-inflammatory factors [11,12]. Notably, oxidative stress exhibits complex interactions with other pathological mechanisms, such as HPA axis dysregulation and neuroinflammation, indicating its potential role as a key node in the multi-mechanistic network of depression [8]. Therefore, targeting oxidative stress is likely to serve as a critical means to disrupt this pathological network.

Among the non-pharmacological interventions, exercise therapy is considered a promising intervention for the treatment of depression due to its multi-target modulatory properties [[13], [14], [15]]. Meta-analyses have shown that aerobic exercise significantly improves depressive symptoms [16,17]. Mechanistically, the antioxidant effect induced by exercise is of great importance: long-term exercise can reduce the accumulation of ROS in the brain and alleviate oxidative stress within the brain [18,19]. Although the antidepressant efficacy of exercise-induced antioxidant modulation is empirically validated, the exact mechanisms underlying these benefits still remain to be elucidated.

Therefore, in this study, with use of the chronic restraint stress (CRS)-induced mouse model of depression, we revealed that Phactr4 expression in the hippocampal dentate gyrus (DG) region of depressed mice was increased by high-throughput sequencing analysis. Additionally, it was demonstrated that oxidative stress in DG region was mediated via the Phactr4-PP1-GSK3β pathway. Concurrently, the exercise intervention was observed to result in a downregulation of Phactr4 expression in the hippocampus, this downregulation was found to be inversely associated with the activation of GSK3β and the levels of oxidative stress markers. In conclusion, this study elucidates the role of Phactr4 in the oxidative stress imbalance underlying depression and clarifies the potential mechanism by which exercise ameliorates oxidative stress through targeting Phactr4.

2. Methods

Animals: All experiments were approved by the Shandong University Institutional Animal Care and Use Committee (ECSBMSSDU-2022-2-65) and were guided by the Association for Assessment and Accreditation of Laboratory Animal Care International. Unless otherwise indicated, adult (6–8 weeks old) male C57BL/6 mice were randomly divided into control and experimental groups and housed in a 22 ± 2 °C, 12h light-dark cycle with food and water supplied ad libitum. The experiment was initiated after the mice had acclimated to the environment for one week.

Chronic restraint stress (CRS): Mice were restrained daily for 4–5 h for 21 days as previously described [20,21]. Mice were individually placed in ventilated restraint tubes, thus ensuring that they were unable to move freely within the tube. After 4–5 h, the mice were released from the restraint tubes and returned to their home cages.

Exercise paradigm: After the mice acclimated to the treadmill apparatus, the exercise training was conducted as described previously [22]. The mice were run for 1 h at 10 m/min, restraints were then applied 1 h after the run.

Drug Treatments: mice were gavaged MK-2206 dihydrochloride (MedChem Express, 1:10 DMSO: normal saline, 100 mg/kg) or solvent by every other day for 10 days [23]. Mice were intraperitoneally injected lithium chloride (LiCl) (Sigma, normal saline, 100 mg/kg) or solvent by daily for 3 weeks [24,25].

Behavioral assays: Prior to all behavioral tests, mice were transported to the experimental room for a period of acclimatisation. Before each test, all behavioral boxes were wiped with 75 % ethanol and then air-dried. All behavioral experiments were conducted in an isolated behavioral testing room, and operators were blinded as to the experimental treatment group.

Sucrose preference test (SPT): The SPT was performed as previously described with minor modifications [26]. Briefly, mice were individually housed in cages equipped with two bottles of 2 % sucrose solution for the first 24 h, the sucrose solution was replaced with water for the next 24 h. After this acclimation period, mice were deprived water for 12 h. Then, the mice were permitted free access to two bottles for 6h, one bottle contained a 2 % sucrose solution, and the other contained water. Sucrose preference was defined as sucrose consumption/[water consumption + sucrose consumption] × 100 % during the 6h test.

Forced swimming test (FST): The forced swimming test was performed under normal light as described previously [26]. Mice were placed in a cylinder of water (temperature 23–25 °C; diameter 20 cm, height 25 cm) for 6 min. The water depth was set to prevent the mice from touching the bottom with their tails or hind limbs. Immobility times were calculated, with immobility being defined as floating or remaining motionless with no active movement except to maintain a buoyant position in water.

Open field test (OFT): Motor activity and anxiety behaviors of mice were recorded using an open field arena (50 × 50 × 40 cm). Mice were placed in the center of the arena with dim lighting and allowed to explore for 10 min. An infrared camera positioned above the box recorded the movements of mice. Total spontaneous activity and time spent in this central area were analyzed offline using Topscan software.

Elevated plus maze test (EPM): The elevated plus maze used in this study consisted of two open arms (35 × 6 cm), two closed arms (35 × 6 × 20 cm) and a central platform area (6 × 6 cm). Under standard lighting conditions, each mouse was individually placed on the central platform and allowed to freely explore the maze for 5 min. Use the video tracking system Top scan and its supporting software to record and analyze the movement data.

2.1. Proteomics analysis

Label-free quantitative proteomics analysis was carried out on hippocampal tissues in the Control and Depression group. Protein Extraction and Quality Control: each sample was lysed and sonicated. The lysate was centrifuged and the supernatant was collected for protein quantification and SDS-PAGE quality control. Enzymatic Digestion and Desalting: After reduction and trypsin digestion, the resulting peptide mixtures were lyophilized. LC-MS/MS Analysis: Lyophilized peptides were reconstituted in mobile phase. Sample was injected for LC-MS/MS analysis using a Q Exactive HF-X mass spectrometer equipped with a Nanospray Flex™ (NSI) ion source. The spray voltage was set to 2.4 kV, and the ion transfer tube temperature was 275 °C. Raw data were acquired in data-dependent mode. Label-free mass spectrometry analysis was processed using Proteome Discoverer2.4 software. The qualified MS/MS data were then subjected to database searching for protein identification and label-free quantification to determine the differentially expressed proteins between the two groups.

Differential protein expression analysis between different groups was conducted via DESeq2 packages (v3.21). DEPs with |log2FC | ≧ 1 and FDR ≤0.05(DESeq2) or FDR ≤0.001(DEPseq) were considered to be significantly different expressed proteins. Heatmaps were constructed to visualize the expression patterns of differentially expressed proteins across samples using the R package pheatmap (v1.0.12). The list of preranked proteins was subjected to GO pathway analysis. Significantly enriched terms were identified using a false discovery rate (q value) of less than 0.05. The differentially expressed proteins were defined as those meeting dual thresholds: P value < 0.05 coupled with a minimum 1.5-fold change in expression. GO functional enrichment and volcano plot were carried out by the ClusterProfiler R package (v4.6.0) and R package ggplot2 (v3.4.2).

Stereotaxic surgery and virus injection: Procedures of stereotaxic surgery was performed as previously described with minor modifications [27]. Mice were anaesthetized (sodium pentobarbital at 35 mg/kg, i. p.) and placed in a stereotaxic apparatus. Viruses were either bilaterally injected into the DG region (AP, −2.1 mm; ML, ±1.4 mm; DV, −2.1 mm, relative to bregma) using an electric microinjection pump (Stoelting, USA) at an infusion rate of 30–50 nl/min. The glass pipette remained at the injection site for 10–15 min after infusion and was then slowly removed to avoid viral backflow. Individual animals failing to achieve an AAV-mediated gene transfer into the target region were excluded from the experiments.

Cell culture and treatment: PC12 cells were cultured in DMEM supplemented with 10 % FBS and 1 % penicillin, and incubated at 37 °C in an atmosphere containing 5 % CO2 in air. PC12 cells were cultured for 2 days, after which they were divided into 6 groups, including Control, CORT (400 μM), CORT + LiCL (5 mM), CORT + LiCL + ML385 (5 μM), Control + LiCL, Control + ML385. After 24 h of intervention, subsequent experiments were performed. For experiments examining PP1 inhibition, cells were cultured for two days and then divided into the following groups: Ctrl, Ctrl + DMSO, Ctrl + Okadaic acid (OA) (20 nM).

Plasmid transfection: Full-length and truncated sequences of Phactr4 and PP1 were amplified and subsequently cloned into the pcDNA3.1-Flag or pcDNA3.1-MYC vector. Plasmid transfection was conducted using the PEI Transfection Reagent (Proteintech) according to the manufacturer's instructions. Briefly, 5 μl reagent and 4 μg plasmid were diluted into 250 μl Serum - free medium and incubated at 37 for 15 min, then incubated the cells after mixing.

Immunochemistry: After anesthesia, mice were transcardially pefused with 0.9 % normal saline and immobilized with 4 % PFA. Brain tissues were removed, postfixed overnight with 4 % PFA at 4 °C, and then dehydrated in PBS with a sucrose gradient (10–30 %). Brain slice were sectioned on a cryogenic sectioning machine and incubated with blocking buffer at room temperature for 3 h and overnight with the primary antibody at 4 °C. Cell slides were fixed with 4 % paraformaldehyde for 10 min. Subsequently, the cell slides were incubated with 0.5 % Triton-X 100 for 30 min and blocked with 5 % BSA for 1 h. Then, incubated overnight at 4 °C with primary antibodies. All images were captured with an Olympus VS120 high-throughput fluorescent, HOOKE S3000 (HOOKE Instruments Ltd, China) or high-speed confocal platform (Dragonfly 200). Fluorescent intensity analyses were performed using Image J (v.7.4.2) or Imaris viewer (10.2.0).

Western blotting: Brain tissues were homogenized in RIPA Buffer supplemented with protease and phosphatase inhibitors. The Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime) was used for the separation of nuclear and cytoplasmic proteins. The supernatants of tissue lysates were aspirated and centrifuged (20 min, 12000 rpm, 4 °C). PC12 cells were lysed in IP buffer containing a cocktail for 30min, followed by centrifugation at 12000 rpm for 15 min. For immunoprecipitation (IP), supernatants were collected and incubated with specific antibody at 4 °C for 6 h. Protein A + G agarose IP (Biowold, BD0048) were then added to the mix overnight at 4 °C. After incubation, bead‐linked immune complexes were washed with IP buffer 5 times, eluted by boiling with 1 % SDS sample buffer for 10 min. Western blotting was subsequently performed for further analysis. Equal amounts of proteins were separated by SDS-PAGE, transferred to PVDF membranes, then incubated in 5 % BSA diluted in 1 × Tris-Buffered Saline (TBS)-Tween 20 for 2 h. The membranes were incubated with primary antibodies overnight at 4 °C. The polyclonal goat anti‐rabbit secondary antibody and polyclonal goat anti‐mouse secondary antibody were used as labelled secondary antibodies to detect the proteins of interest. Image-J software was used to perform pixel quantifications of the images. The antibodies used were: anti-Phactr4 (1:1000; Proteintech), anti-PP1 (1:1000; Santa), anti-Nrf2 (1:1000; Proteintech), anti–HO–1 (1:1000; Abcam), anti-Phospho-GSK3β(Ser9) (1:1000; Affinity), anti-GSK3β (1:1000; Affinity), anti-Phospho-GSK3α(Ser21) (1:1000; Bioworld), anti-GSK3α (1:1000; Bioworld), anti-Phospho-P65 (1:1000; CST), anti-P65 (1:1000; CST), anti-GAPDH (1:1000, Bioworld) and anti-LaminB 1(1:1000, Affinity).

Molecular Docking: The protein structures of Phactr4 and PP1 derived from mice were obtained from UniProt database (https://www.uniprot.org/). Gramm-x server (http://vakser.bioinformatics.ku.edu/resources/gramm/grammx) was used for rigid molecular docking between proteins to evaluate any possibility of their interactions. PP1 is defined as a ligand while Phactr4 as a receptor. Among the output models, the first one was used as the final model and then PyMOL and LigPlot + were used for visualization.

2.2. Oxidative stress measures

ROS measurement: To measure the ROS production in tissue, frozen slices were incubated with 10 μM dihydroethidium (DHE) for 30 min at 37 °C and then stained with DAPI (Beyotime) for 10 min. To measure the ROS production in cells, PC12 cells were cultured in 24-well plates with coverslips prior to staining. 10 μM DHE (Beyotime) or 10 μM MitoSOX Red fluorescent dye (ThermoFisher) was applied to the cells. Cells were incubated with the dye for 15 min at 37 °C. Subsequently, the cells were stained with DAPI (Beyotime) for 10 min. Images were captured on a confocal microscope.

Assessment of DNA damage markers: Oxidative DNA damage was detected using an immunofluorescent assay. Brain slices were incubated with the primary mouse anti-DNA/RNA Damage antibody (ab62623 Abcam) and rabbit anti-NeuN overnight at 4 °C followed by Alexa Fluor 568 or Cyanine5 conjugated (ThermoFisher). DAPI (Beyotime) was used for nuclear staining.

Oxidative stress products: The lipid peroxidation product resulting from oxidative stress was measured using immunofluorescent staining with anti-4-hidroxynonenal (4-HNE, Abcam). In addition, the malondialdehyde (MDA) content in the DG region or PC12 cells were determined by using assay kits from Jiancheng Inc. (Nanjing, China).

Antioxidant enzyme activities: Activity of antioxidant enzymes in DG region was measured using the superoxide dismutase (SOD), total antioxidant capacity (T-AOC) activity assay kits according to the manufacturers’ guidelines. All assay kits were purchased from Jiancheng Inc. (Nanjing, China).

Statistics: All mice were randomly divided into control and experimental groups. All data were collected and processed randomly or in a counterbalanced manner. All data analyses were performed blindly. Use a sample size similar to those as reported in previous publications. All data are presented as Means ± SD. We tested the normality of the data using the Shapiro–Wilk test. The two-tailed unpaired Student's t-test to evaluate the data from both groups, one way ANOVA with a Bonferroni post hoc test were used to evaluate data from more than two groups. A P value of <0.05 was required for results to be considered as statistically significant. Statistical analyses and graph generation were performed using GraphPad Prism 10.0.

3. Results

3.1. CRS-induced depression-like mice with oxidative stress damage in the hippocampal DG region

Depression is usually accompanied by neuronal damage and abnormal accumulation of ROS in specific brain regions. These may synergistically induce oxidative stress and neuroinflammatory responses [28]. After 3 weeks of restraint treatment, the results obtained from the sucrose preference test and the forced swimming test indicated that the mice displayed anhedonia and behavioral despair (Fig. 1A–C). These results suggest that CRS induces a typical depression-like behaviors in mice. Anxiety-like behaviors of the mice were assessed by the OFT and the EPM. It was observed that the time CRS mice spent in the open arms of the EPM (Fig. 1D) and in the central area of the OFT was significantly reduced (Fig. 1E–G). DHE staining analysis revealed that the level of ROS in the DG region of the hippocampus was significantly elevated (Fig. 1H–J). Meanwhile, 4-HNE immunostaining revealed substantial accumulation of lipid peroxidation products in DG region (Fig. 1I–K). The MDA level, an indicator of oxidative damage, was significantly elevated in the DG (Fig. 1L). In contrast, the activities of key antioxidant enzymes decreased synchronously. Both the activity of SOD and the T-AOC showed a significant downward trend (Fig. 1M–N).

Fig. 1.

Oxidative damage as observed in the hippocampal DG region of CRS mice. (A) Schematic of the experimental design for the CRS model. (B–G) Depressive- and anxiety-like behaviors in the different experimental groups of the CRS model (n = 12 animals/group). SPT (B), FST (C), EPM (D) and OFT (E-G). (H) Representative images of DHE staining (red) within the DG area from each group of mice. Scale bar: 50 μm. (I) The levels of 4-HNE detected by immunofluorescence staining in each group of mice. Scale bar: 50 μm. (J) ROS relative intensity analysis in each group (n = 6 per group). (K) The quantification of the 4-HNE in each group (n = 6 per group). (L–N) The levels of MDA (L), SOD (M) and T-AOC (N) in each group (n = 6/group). For all statistical tests: two-tailed unpaired Student's t-test. ∗, P < 0.05; ∗∗, P < 0.01; ∗∗∗, P < 0.001; NS, no significant difference. Data are presented as Means ± SD. Dots represent individual mice.

3.2. The expression of Phactr4 was upregulated in the hippocampus DG of CRS mice

The volcano plot results revealed a large number of differentially expressed proteins between the hippocampal of the two groups (Fig. 2A). Detailed protein profiles was visualised using heat maps (Fig. 2B). Volcano plot analysis identified multiple differentially expressed proteins (DEPs) in the depression group versus controls. Consistently, the heatmap revealed a significant upregulation of Phactr4 in the depression group compared to controls. Gene ontology (GO) enrichment analysis confirmed that the differentially expressed proteins in the CRS group were closely associated with the oxidative stress pathway (Fig. 2C). Protein-protein interaction (PPI) network analysis further revealed the direct interaction between Phactr4 and PP1 (Fig. 2D). In addition, we verified the co-localization between Phactr4 and PP1 in the DG region of the hippocampus by immunofluorescence (Fig. 2E). CoIP assays demonstrated physical interactions between Phactr4 and PP1 in the hippocampal DG of both control and CRS model mice (Fig. 2F). Western blot results showed the increased protein expression levels of Phactr4 which verified the accuracy of proteomics (Fig. 2G). The above research results suggest that elevated Phactr4 expression in the hippocampal DG region may play a key role in the pathogenesis of CRS-induced depression, which may be mediated by its functional interaction with PP1 and form a key regulatory pathway in the stress.

Fig. 2.

CRS exposure increased the protein expression of Phactr4. (A-B) Volcano plots and heatmaps showing the protein expression levels of the Phactr4 in the datasets. (C) Gene Ontology enrichment analysis of the differentially expressed protein of control and CRS mice. (D) Protein-protein interaction network of Phactr4 and PP1. (E) Representative images of Phactr4 (red) and PP1 (green) staining. Scale bar: 50 μm. (F) Representative Western blots of the co-immunoprecipitation of Phactr4 and PP1 of control and CRS mice. (G) Representative western blots and quantification of protein expression levels of Phactr4 of control and CRS mice (n = 6/group). For all statistical tests: two-tailed unpaired Student's t-test. ∗∗, P < 0.01. Data are presented as Means ± SD. Dots represent individual mice.

3.3. Phactr4 directly interacted with PP1

To further validate Phactr4 and PP1 interactions, we detected the binding of Phactr4 and PP1 proteins by prediction of protein-protein interactions and CoIP assay. After downloading the protein information of mouse PP1 (PDB: P62137) and Phactr4 (PDB: Q501J7) from the PDB database, PyMOL software was used to remove excess amino acid sequences, and docking prediction was carried out. Dock with the lowest binding energy was selected and potential binding sites were predicted (Fig. 3A). The results found that PP1 depicted high compatibility with Phactr4 (639–694) protein binding pockets, and the binding energy was −5.6 kcal/mol, which was able to bind under natural conditions. Endogenous CoIP revealed that Phactr4 protein could directly bind to PP1 protein in PC12 cells (Fig. 3B), with this interaction predominantly localized to the cytoplasm (Fig. S1A). To further investigate the interaction between Phactr4 and PP1, PC12 cells were transfected with Flag-PP1 and Myc-Phactr4, respectively, and control cells were respectively transfected with empty plasmids labelled with Flag or Myc. The results showed exogenous binding of Flag-PP1 and Myc-Phactr4 proteins (Fig. 3C). Co-localization analysis of double fluorescent staining of PC12 cells showed some overlap between endogenous Phactr4 and PP1 (Fig. 3D). These results suggest that Phactr4 can directly bind neuronal PP1 protein (Fig. 3E). The full length of Phactr4 protein is composed of 694 amino acids. Truncations of different Phactr4 structural domain deletions were constructed and labelled as P1 (1–639), P2 (640–694) (Fig. 3F). In cells, each Myc-tagged Phactr4 short plasmid or full-length Phactr4 plasmid was cotransfected with the Flag-PP1 plasmid, which showed that sequences close to the C terminus were sufficient to abolish the interaction with PP1 as compared to full-length Phactr4 (Fig. 3G). To further determine whether the interaction of Phactr4 with PP1 regulates the phosphorylation of GSK3β, we transfected Myc-Phactr4 overexpression plasmid into PC12 cells and examined the expression of PP1 and p-GSK3β in the transfected cells. Western blotting results showed that, after transfection with Myc-Phactr4 overexpression plasmid, PP1 protein levels were significantly increased. Meanwhile, p-GSK3β (Ser9) (Fig. S1B–C) and p-GSK3α (Ser21) expression (Fig. S1D–E) were reduced. These results suggest that Phactr4 overexpression increases p-GSK3β (Ser9) and p-GSK3α (Ser21) protein levels in neurons. Following treatment with okadaic acid (OA), a specific inhibitor of PP1 activity, CoIP results demonstrated that the interaction between Phactr4 and PP1 remained intact (Fig. S1F). Concomitantly, the expression level of p-GSK3β, a downstream effector molecule, was significantly increased (Fig. S1G–H).

Fig. 3.

Phactr4 directly interacted PP1 protein and affect its level. (A) Molecular docking revealed potential binding sites between Phactr4 and PP1. (B) The interaction between endogenous phactr4 and endogenous PP1 in PC12 cells was analyzed by CoIP and immunoblotting. (C) The interaction between exogenous PP1 and exogenous Phactr4 was analyzed by CoIP. (D–E) Line intensity plots showed colocalization of endogenous PP1 and endogenous Phactr4 in PC12 cells. (F–G) CoIP assay was performed to explore the binding between the PP1 protein and the Phactr4 domain. (H) The expression levels of PP1 was analyzed by Western blot after Phactr4 domain plasmid was transfected into PC12 cells.

3.4. Knockdown of Phactr4 ameliorated depressive-like behavior in CRS mice

To determine whether Phactr4 is a pivotal risk factor for CRS-induced depression, AAV-sh-Phactr4-eGFP was bilaterally injected into the hippocampus DG region in mice of the CRS group (Fig. 4A). The accuracy of the viral injection site was confirmed by fluorescence imaging analysis (Fig. 4B). Behavioural testing was conducted three weeks after injection. The results of the SPT (Fig. 4C) and FST (Fig. 4D) showed that the knockdown of Phactr4 enhanced the sucrose consumption and reduced the immobility time in CRS rats. Meanwhile, knocking down Phactr4 can significantly increase the time for mice to enter the open arms in EPM (Fig. 4E) and the central area in OFT (Fig. 4F–G). These findings indicate that downregulating Phactr4 expression can alleviate depression-like and anxiety-like behaviors in CRS mice. Next, we delved into the downstream mechanisms by which Phactr4 participates in the regulation of oxidative stress within the DG region (Fig. 4H). Given that Phactr4 directly regulates the expression of PP1, we measured the protein levels of PP1 in the DG region of mice treated with the AAV-sh-Phactr4. Western blot results demonstrated that the expression of PP1 was significantly decreased in CRS mice treated with the AAV-sh-Phactr4 (Fig. 4I–J). Previous literature reports have indicated that PP1, a protein phosphatase, can dephosphorylate GSK3β (Ser9) which functions as its substrate [29]. Moreover, the expression of p-GSK3β has been found to be positively correlated with the expression of Nrf2 [30]. Consequently, through Western blot analysis, we found that CRS led to increased expression levels of Phactr4 and PP1 in the DG region of mouse brains (Fig. 4I–J), while the expression of p-GSK3β (Fig. 4K–L) was significantly downregulated, accompanied by decreased expression of Nrf2 and HO-1. In contrast, knockdown of Phactr4 resulted in reduced GSK3β expression, thereby enhancing the expression of Nrf2 and HO-1. Meanwhile, the exacerbation of oxidative stress was accompanied by the occurrence of neuroinflammation, and Western blot results also showed that the expression of p-p65 in the brains of CRS mice was significantly increased (Fig. 4M–N). These results suggest that Phactr4 may exert an effect on oxidative stress through the PP1-GSK3β pathway.

Fig. 4.

Knockdown of Phactr4 expression in hippocampal DG ameliorated behavioral disorders of CRS mice. (A–B) Experimental paradigm for viral construction and injection in CRS mice. (C–G) Depressive- and anxiety-like behaviors in the different experimental groups of the CRS model (n = 15 animals/group). SPT (C), FST (D), EPM (E) and OFT (F-G). (H) Schematic of the Phactr4-PP1-GSK3β pathway. (I–J) Representative western blots and quantification of protein expression levels of Phactr4, PP1, Nrf2 and HO-1 within the DG region. (K–L) Representative western blots (K) and quantification (L) of p-GSK3β/GSK3β. (M–N) Representative western blots (M) and quantification (N) of P-p65/p65 within the DG region. For all statistical tests: one-way analysis of variance (ANOVA) with Bonferroni post-hoc test. ∗, P < 0.05; ∗∗, P < 0.01; ∗∗∗, P < 0.001; NS, no significant difference. Data are presented as Means ± SD.

3.5. Phactr4 modulates depression-like behaviors and oxidative stress in mice via the PP1-GSK3β pathway

We separately used MK-2206 to indirectly enhance the activity of GSK3β and LiCl, a selective inhibitor of GSK3β, to enhance GSK3β (ser9), aiming to evaluate the contribution of Phactr4-PP1-GSK3β pathway to the display of depression. The behavioral results showed that knockdown of Phactr4 did not improve depression-like and anxiety-like behaviors in CRS + AAV-sh-Phactr4 group mice treated with MK-2206 (Fig. 5A–B). Specifically, under the treatment of MK2206, the percentage of sucrose consumption in the SPT of the mice increased, while the immobility time in the FST decreased (Fig. 5C–D). Treatment with MK2206 failed to reverse anxiety-like behaviors in CRS + AAV-sh-Phactr4 group mice as observed in the OFT and EPM (Fig. 5E–G). Meanwhile, the content of MDA in the DG was increased (Fig. 5H), the content of SOD was decreased (Fig. 5I), and the T-AOC was diminished after MK2206 treatment (Fig. 5J). The expressions of the DNA damage marker 8-OHdG (Fig. 5K–M) and the lipid peroxidation product 4-HNE (Fig. 5L–N) was increased in the hippocampal DG region, as well as the content of ROS (Fig. 5O–P). In addition, the mice treated with the inhibitor LiCl showed a significant reduction in depression- and anxiety-like behaviors induced by CRS (Fig. 6A–G). Meanwhile, the content of MDA in the DG was decreased, the content of SOD and the T-AOC was increased after LiCl treatment (Fig. 6H–J). The expressions of 8-OHdG and 4-HNE was decreased in the hippocampal DG region (Fig. 6K–N). Inhibition of GSK3β significantly reduced the oxidative stress damage in the hippocampus of CRS mice (Fig. 6O–P). These findings provide further support for the hypothesis that Phactr4 may directly target the PP1-GSK3β pathway to induce depression and anxiety behaviors in mice, and that the inhibition of this pathway can ameliorates behavioral of CRS mice.

Fig. 5.

Inhibition of the GSK3β pathway reversed the improvement of behavioral disorders mediated by the down-regulation of Phactr4. (A–B) Schematic of the experimental design. (C–G) Depressive- and anxiety-like behaviors in the different experimental groups (n = 12 animals/group). SPT (C), FST (D), EPM (E) and OFT (F-G). (H–J) The levels of MDA (H), SOD (I) and T-AOC (J) in each group (n = 6/group). (K) The quantification of the 8-OHdG with each group (n = 6/group). (L) The quantification of the 4-HNE with each group (n = 6/group). (M–N) The levels of 8-OHdG (M) and 4-HNE (N) detected by immunofluorescence staining in each group of mice (n = 6 per group). Scale bar: 50 μm. (O–P) Representative images (O) and the quantification (P) of DHE staining within the DG from each group of mice (n = 6 per group). Scale bar: 50 μm. For all statistical tests: one-way analysis of variance (ANOVA) with Bonferroni post-hoc test. ∗, P < 0.05; ∗∗, P < 0.01; ∗∗∗, P < 0.001; NS, no significant difference. Data are presented as Means ± SD.

Fig. 6.

Activation of the GSK3β pathway improved behavioral disorders and oxidative damage in CRS mice. (A–B) Schematic of the experimental design. (C–G) Depressive- and anxiety-like behaviors in the different experimental groups (n = 12 animals/group). SPT (C), FST (D), EPM (E) and OFT (F-G). (H–J) The levels of MDA (H), SOD (I) and T-AOC (J) in each group (n = 6/group). (K) The quantification of the 8-OHdG with each group (n = 6/group). (L) The quantification of the 4-HNE with each group (n = 6/group). (M–N) The levels of 8-OHdG (M) and 4-HNE (N) detected by immunofluorescence staining in each group of mice (n = 6 per group). Scale bar: 50 μm. (O–P) Representative images (O) and the quantification (P) of DHE staining within the DG from each group of mice (n = 6 per group). Scale bar: 50 μm. For all statistical tests: one-way analysis of variance (ANOVA) with Bonferroni post-hoc test. ∗, P < 0.05; ∗∗, P < 0.01; ∗∗∗, P < 0.001; NS, no significant difference. Data are presented as Means ± SD.

3.6. Upregulating the p-GSK3β can effectively improve the oxidative stress of PC12 cells

In vitro experiments were conducted to further investigate the effects of GSK3β phosphorylation on cellular oxidative stress injury. Specifically, PC12 cells were treated with CORT to simulate neuronal oxidative damage in the brain (Fig. 7A–B). Following CORT treatment, the content of MDA was significantly increased (Fig. 7C), while SOD activity was obviously decreased (Fig. 7D). Western blotting revealed that p-GSK3β levels were significantly reduced, accompanied by decreased expression of Nrf2 and HO-1 (Fig. 7E–G). However, pretreatment with LiCl significantly attenuated MDA accumulation, enhanced SOD activity, upregulated p-GSK3β expression, and increased Nrf2 and HO-1 protein levels. In addition, ML385, a specific Nrf2 transcriptional inhibitor, partially reversed the protective effects of LiCl. Cellular immunofluorescence assays demonstrated that LiCl treatment significantly reduced intracellular ROS levels (Fig. 7H–I) and alleviated mitochondrial damage (Fig. 7J–K). These results indicate that GSK3β modulates neuronal oxidative stress responses by regulating its phosphorylation status and subsequently mediating Nrf2 transcriptional activity.

Fig. 7.

Enhancing the expression of GSK3β (Ser9) improved the oxidative stress of PC12 cells. (A–B) Schematic of the experimental design. (C–D) The levels of MDA (C) and SOD (D) in each group. (E–G) Representative western blots (E) and quantification (F-G) of protein expression levels of Nrf2, HO-1 and p-GSK3β/GSK3β. (H–I) Representative images (H) and the quantification (I) of DHE staining within PC12 cells. Scale bar: 50 μm. (J–K) Representative images (J) and the quantification (K) of Mito-Sox staining. Scale bar: 50 μm. For all statistical tests: one-way analysis of variance (ANOVA) with Bonferroni post-hoc test. ∗, P < 0.05; ∗∗, P < 0.01; ∗∗∗, P < 0.001; NS, no significant difference. Data are presented as Means ± SD.

3.7. Exercise alleviates depression-like behaviors and oxidative stress in mice via the Phactr4-PP1-GSK3β pathway

As described in our previous experimental protocol, during the CRS modeling period, mice were subjected to treadmill exercise for 1h (speed: 10 m/min) of 21 days (Fig. 8A). After three weeks of exercise, behavioral tests revealed that the mice showed significantly decreased depressive- and anxiety-like behaviors. Specifically, these exercised CRS mice demonstrated an enhanced preference for sucrose water in SPT, reduced immobility time in the FST, and a significant increase in the time entries into the open arm of EPM as well as the time spent in the central area of the OFT (Fig. 8B–F). Subsequently, we explored whether exercise could ameliorate oxidative stress via the Phactr4-pp1 pathway. Western blotting revealed that, in association with elevated Phactr4, the level of Nrf2 protein was decreased under CRS and was potentiated by exercise under CRS. Conversely, exercise was able to reduce the expression of Phactr4 in the DG region and enhance the expression of Nrf2 by regulating Phactr4-PP1-GSK3β (Fig. 8G–M). DHE staining results demonstrated that exercise could significantly decrease the levels of ROS in hippocampus DG region (Fig. 8N–O). Meanwhile, the content of 8-OHdG, a marker of DNA damage, in the DG region was also significantly reduced, indicating that exercise effectively alleviate oxidative stress (Fig. 8P–Q). These results converged to imply the contribution of the Phactr4 pathway to exercise-mediated stress resilience and the alleviation of oxidative stress.

Fig. 8.

Exercise alleviates behavioral disorders and oxidative stress in mice via the Phactr4-PP1-GSK3β pathway. (A) Schematic of the experimental design. (B–F) Depressive- and anxiety-like behaviors in the different experimental groups (n = 12 animals/group). SPT (B), FST (C), EPM (D) and OFT (E-F). (G–H) Representative western blots (G) and quantification (H) of protein expression levels of Phactr4 and PP1 within the DG region (n = 6/group). (I–J) Representative western blots (I) and quantification (J) of p-GSK3β/GSK3β within the DG region (n = 6/group). (K–M) Representative western blots (K) and quantification (L-M) of protein expression levels of Nrf2, HO-1 and p-P65/P65 within the DG region (n = 6/group). (N–O) Representative images (N) and the quantification (O) of DHE staining in each group of mice. (P) The levels of 8-OHdG detected by immunofluorescence staining in each group of mice. Scale bar: 50 μm. (Q) The quantification of the 8-OHdG in each group (n = 6 per group). For all statistical tests: one-way analysis of variance (ANOVA) with Bonferroni post-hoc test. ∗, P < 0.05; ∗∗, P < 0.01; ∗∗∗, P < 0.001; NS, no significant difference. Data are presented as Means ± SD.

4. Discussion

As a pervasive mental health disorder, MDD severely impairs the quality of patients’ life, affecting over 300 million individuals worldwide [31,32]. Dysregulation of redox signaling plays a critical role in the pathophysiology and neurological progression of MDD [33]. Here, we report that oxidative stress exists in the DG region of the hippocampus in depressed model mice, accompanied by upregulated protein expression of Phactr4. Inhibition of Phactr4 significantly reduces the content of ROS and the accumulation of peroxidation products in neurons of depressed mice, alleviates oxidative stress damage, and improves depression-like behaviors. COIP assays confirm that the C-terminal domain of Phactr4 directly interacts with PP1, thereby regulating PP1 expression levels. Mechanistic studies show that PP1 modulates the phosphorylation of GSK3β and further regulating the Nrf2 signaling pathway to induce oxidative stress. In addition, exercise intervention downregulates the activity of the Phactr4-PP1-GSK3β pathway, attenuate oxidative stress, and significantly alleviate the depressive behavior of mice.

The brain, owing to its high oxygen consumption and rich content of oxidizable lipids, is more vulnerable to increased ROS damage than any other organ in the body [34,35]. Imbalances between ROS and antioxidant defense system trigger brain dysfunction and abnormal neuronal signaling [[36], [37], [38]]. Impairments in these antioxidant systems accelerates the pathophysiology of depression by reducing the protective effect against ROS and reactive nitrogen species (RNS), thereby potentially increasing the risk of persistent damage [39]. Using CRS for depression modeling, our results showed that CRS effectively induced depression- and anxiety-like behaviors in mice, while simultaneously causing oxidative damage in the DG region of the brain.

Phosphatase and actin regulators (Phactrs) represent a family of proteins specifically expressed in the brain, which have both regulatory activity on PP1 and actin-binding capabilities [40]. Comprising four members (Phactr1-4), these proteins display distinct expression patterns in the adult mammalian brain. It participates in multiple neural functions such as cell cycle regulation, neuroinflammatory responses, and synaptic plasticity by regulating PP1 activity, and plays a critical role in the development of neural tubes and the pathological mechanisms of brain-related diseases [[41], [42], [43]]. Among them, Phactr4 is highly expressed in specific brain regions, such as the subventricular zone (SVZ), subiculum (SE), and DG of the hippocampus [44]. According to the sequencing data, Phactr4 was identified as the significantly upregulated protein with the greatest expression difference between depression models and controls, and exhibited specific interaction with PP1. Further functional validation indicated that down-regulating the expression of Phactr4 in the hippocampal DG region could significantly alleviate depressive-like behaviors in mice and reduce oxidative stress damage in the DG region. These findings suggest that the Phactr4-PP1 signaling axis may serve as a potential intervention target in the pathological mechanisms of depression.

We further investigated the mechanisms by which Phactr4 regulates oxidative stress in the hippocampal DG of mice. GSK3β was identified as a critical integrative node in the regulation of Nrf2 activity [[45], [46], [47]]. As a ubiquitously expressed and constitutively active multifunctional protein, GSK3β participates in diverse physiological processes, including glycogen metabolism, embryonic development, tissue repair, immune regulation, and redox homeostasis [[48], [49], [50]]. Recent evidence has highlighted the pivotal role of GSK3β in the regulatory pathway of Nrf2, with its modulation of Nrf2 being closely associated with pathological conditions such as aging, type 2 diabetes, and neurodegenerative diseases [[51], [52], [53]]. Specifically, Phactr4 enhances GSK3β activity by activating PP1, which induces dephosphorylation of GSK3β and reduces its inhibitory phosphorylation. Consistently, treatment with a PP1 activity inhibitor significantly increases the inhibitory phosphorylation level of GSK3β. The activated GSK3β amplifies phosphorylation of its substrate Nrf2, promoting nuclear export and degradation of Nrf2, thereby attenuating the Nrf2-mediated antioxidant response. We verified the critical role of GSK3β in regulating oxidative stress through targeted intervention of GSK3β activity. In Phactr4-knockdown depression model mice, administration of the GSK3β phosphorylation inhibitor MK-2206 failed to alleviate depression- and anxiety-like behaviors or reduce oxidative damage in the brain. Conversely, daily injection of the GSK3β dephosphorylation inhibitor LiCl during CRS modeling significantly improved behavioral abnormalities and attenuated oxidative stress in the hippocampal DG region. Further in vitro experiments demonstrated that LiCl protected neurons from CORT-induced oxidative injury, whereas this protective effect was reversed by the Nrf2 transcription inhibitor ML385.

Depression is a multifactorial and complex psychiatric disorder, with its pathogenesis involving intricate interactions among genetic, environmental factors, as well as gene-sex dimorphisms [54,55]. In the present study, we observed that both GSK3β and GSK3α are regulated by Phactr4 and PP1. Accumulating evidence has further confirmed that GSK3α plays a pivotal role in brain function and the pathogenesis of neurodegenerative diseases [50,56]. However, the potential differences and interconnections in the regulatory roles of GSK3β and GSK3α in oxidative stress during depression remain to be further elucidated. Additionally, considering the possible diverse impacts of stress or exercise on systemic energy metabolism, hormonal homeostasis (HPA axis), and brain structure-function remodeling [57,58], the specific upstream molecular mechanisms by which exercise and chronic restraint stress regulates the expression and activity of Phactr4 and PP1 merit in-depth investigation.

In this study, we found that Phactr4 expression was significantly upregulated in the DG of the hippocampus in depression model mice, whereas downregulation of Phactr4 significantly alleviated depression-like behaviors and reduced oxidative stress damage in the DG region. Mechanism studies have shown that the binding of Phactr4 to PP1 regulates neuronal oxidative stress levels by enhancing the dephosphorylation of GSK3β. Additionally, exercise can downregulate Phactr4 expression and promote GSK3β phosphorylation thereby inhibiting its activity, effectively alleviating brain oxidative stress responses. Collectively, this study identifies the Phactr4-PP1-GSK3β pathway as critical in depression-associated neuronal oxidative stress and suggests it may as a potential intervention strategy for neuronal oxidative damage in depressed patients. The mechanism by which exercise exerts antidepressant and neuroprotective effects through regulation of this pathway provides new theoretical insights for clinical non-pharmacological interventions.

5. Study approval

All experimental procedures were approved by the Shandong University Animal Care and Use Committee (ECSBMSSDU-2022-2-65) and conform to the NIH Guide for the Care and Use of Laboratory Animals (National Academies Press, 2011).

CRediT authorship contribution statement

Tian Lan: Writing – review & editing, Writing – original draft, Software, Project administration, Methodology, Investigation, Formal analysis, Data curation. Ye Li: Writing – review & editing, Software, Methodology, Investigation, Formal analysis, Data curation. Wanzhe Zhang: Methodology, Investigation, Formal analysis. Xiao Chen: Methodology, Investigation. Mengni Chang: Methodology, Investigation. Wenjing Wang: Methodology, Investigation. Changmin Wang: Methodology, Investigation. Shihong Chen: Writing – review & editing, Supervision, Software, Resources, Project administration, Data curation, Conceptualization. Linghua Kong: Writing – review & editing, Supervision, Software, Resources, Project administration, Funding acquisition, Data curation, Conceptualization. Shuyan Yu: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of competing interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix A. Supplementary data

The following are the Supplementary data to this article.

Data availability

Data will be made available on request.