Animals

Male Wistar rats (weighing 180–190 g) were obtained from Jinan Peng-yue Experimental Animal Breeding Co., Ltd. Male C57/BL 6J mice (weighing 25–30 g) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. All experimental procedures were approved by the Ethics Committee of Shandong University (ECSBMSSDU2020-2-017), and in compliance with the international guidelines for animal research formulated by the Council of International Medical Organizations. The rats and mice were both housed under standard laboratory conditions for one week before experimental procedures, with free access to food and water. All efforts were made to reduce the animals’ suffering in the experiments.

Regents and antibodies

SB203580 (0.5 mg/kg) was purchased from MedChemExpress Co., Ltd. (USA). Streptozocin (STZ) and 0.1 M citrate buffer (pH 4.5) were purchased from Solarbio Co., Ltd. Lipopolysaccharide (LPS) and dimethyl sulfoxide (DMSO) were purchased from Sigma-Aldrich Co. (St Louis, MO, USA). The polyclonal rabbit anti-CD11b (ab184307) and polyclonal rabbit anti-CD45 (ab10558) were purchased from Abcam Co. (Cambridge, UK). The polyclonal rabbit anti-NF-κB p65 (BS90940) was purchased from BioWorld Technology, Inc. The polyclonal rabbit anti-GAPDH (10494-1-AP) purchased from Proteintech Technology and the polyclonal rabbit anti-P38 (9212s) was purchased from Cell Signaling Technology. Constructed adeno-associated virus (AAV9-CMV-eGFP-miR-26a-3p-sponge vector) was purchased from Gene-Chem Co. (Shanghai, China).

Chronic unpredicted mild stress (CUMS) model

The rats were acclimated in the experimental animal room for a week. In brief, the rats were individually housed for 5 weeks. These rats were subjected to chronic stressors including clipping tails (2 min), cold swimming (5 min, 4 ℃), cage shaking (5 min), physical restraints (2 h), food and water deprivation (24 h), wet bedding (24 h), cage tilting (24 h), overnight illumination. Each rat was given one stimulation in a random order daily [25].

LPS-induced depression model

The rats received intraperitoneal (i.p.) injections of LPS (0.5 mg/kg) once a day and the process lasted for 10 days. LPS was prepared freshly prior to injection. Rats in the control group were injected with the same dose of normal saline every day [26].

Diabetic encephalopathy model

STZ was freshly prepared in cold 0.1 M citrate buffer (pH 4.5) and then injected intraperitoneally into rats using a single dose of 60 mg/kg body weight, and the feeding was continued for 12 weeks to develop diabetic encephalopathy (DE) model [27].

Acute restraint stress mice model

Acute restraint stress (ARS) was used to induce disordered behaviors in mice. Male C57/BL 6 J-wild-type mice (8 weeks) were immobilized for 5 h using an individual retainer to limit all physical movement but causing minimum pain to them. Mice were deprived of water and food during the ARS experiment [28].

Drug treatments

SB203580 (0.5 mg/kg) or DMSO (0.1%) were administered via an intraperitoneal injection for two weeks after viral injection. LPS (0.5 mg/kg) was dissolved in 0.9% saline at the concentration of 10 mg/ml before injection. The dose and route of SB203580 administration is based upon previous study with minor modifications [29].

Behavioral tests

All behavioral tests of rats were conducted during the dark circadian period (19:00–24:00). The tests and analysis were performed by an experimenter blind to the treatment group.

Novel object recognition test

The novel object recognition (NOR) test is a widely used behavioral test to assess hippocampus-dependent recognition memory. The experiment was based on previous research with minor modifications [30]. To test recognition, rats were placed in an empty box (100 × 100 × 50 cm) for habituation 1 d before the test. On the first day of the test (familiarization phase), the rats received 5 min to explore the two identical objects in the recognition box. On the second day (24 h later, test phase), the researcher returned the rats to the open-field recognition box with two objects, the initial object explored during the familiarization phase and a newly introduced novel object. In this phase, each animal received 5 min to explore the two objects freely. A video system was used to record the rats' movements in the open field, and researchers measured the amount of time the rats spent in contacting with new and familiar objects by watching videos. The discrimination index was calculated according to the following formula: discrimination index (time spent on novel objects/total time spent on both exploring objects) × 100%. Exploration of an object was defined as the animal placing its nose within 2 cm of the zone where the object located.

Sucrose preference test

Sucrose preference test (SPT) was used as the classical method to define anhedonia [31]. Individually rat from different groups was given two water bottles during the test period. On the first day, two bottles were both filled with tap-water. Twenty-four hours later, one bottle was filled with tap-water and the other one was replaced with 2% sucrose solution. The two water bottles position were changed every twelve hours (right one to left, left one to right) to ensure that the rats did not have a side preference. On the third day, rats were deprived of water and food for 24 h. In the next day, two water bottles (one bottle was filled with tap-water and the other one was 2% sucrose solution) were placed in each cage. The position of the bottles was switched after one hour during the test to prevent the possible effects of a side preference. The amount of liquid remaining in bottles was measured to calculate the liquid consumption. The calculation formula of sucrose preference (%) is as follows: sucrose consumption/(sucrose consumption + water consumption).

Forced swimming test

Forced swimming test (FST) was used to define behavioral despair [32]. In the first training phase, rats were placed individually in a plastic bucket (80 cm high, 30 cm diameter) filled with water 50 cm high at suitable temperature for 15 min of forced swimming. The testing phase was twenty-four hours after training. The rats were orderly placed in the bucket for 5 min. The immobility time and swimming time were recorded during the test. The new water replaced dirty water before the next rat swimming test. The rats cannot touch the bucket bottom or escape in test period.

Open field test

Open field test (OFT) was used to assess basic activities and anxiety-like behavior [33]. The open field is a black wooden open field box (100 × 100 × 50 cm). The inner side surface was cleaned with 75% ethanol to be free of all dirt. The rats were individually placed in the center of the open field and allowed to move freely for five minutes. The video tracking software (SMART 2.5, Panlab Harvard Apparatus, Spain) was used to record the activities of rats. The total distance of horizontal locomotion and the time spent in the central area were analyzed to assess anxiety-like behavior.

The elevated plus maze test

The elevated plus maze (EPM) was used to assess anxiety in rats [33]. The EPM consists of two closed arms (50 cm × 10 cm), two open arms (50 cm × 10 cm) and a central platform (10 cm × 10 cm) at the intersection of the arms. The animal was firstly placed on the central platform with their heads facing to the open arms during 5-min test period. Time spent in the open arms and the number of entries into open arms in 5 min were recorded by video tracking software (SMART 2.5, Panlab Harvard Apparatus, Spain).

Tail suspension test

The experiment was based on previous research with minor modifications [34]. The rat's tail was cleaned and taped to the hook of the tail hanging device. Rats were suspended 40 cm above the floor for 5 min. The immobility time was recorded during the test. Immobility was defined as a lack of attempt to move their limbs and staying in the vertical posture during suspension. At the end of each rat's experiment, the device was cleaned with 75% alcohol.

Virus injections

Rats were deeply anesthetized and placed in a stereotaxic apparatus (Stoelting, USA). The AAV9-CMV-eGFP-miR-26a-3p-sponge vector virus, which acts as a sponge to inhibit the functions of endogenous miR-26a-3p, was bilaterally injected into the hippocampus region (bregma: − 3.24 mm; medial/lateral: ± 1.5 mm; dorsal/ventral: − 4.5 mm) with use of the electric microinjection pump (Stoelting, USA) at the speed of 0.11 µl/min. After the required volume has been injected, the micro syringe remained in the place for at least 5 min and was slowly withdrawn from the tissue. Rats were placed in warm conditions to resuscitate slowly and rest for three days for the following study. All injection sites were verified by immunofluorescence slice before further experiments and only rats of correct injection site were used for analysis.

Immunofluorescence staining

After all behavioral tests, the rats were anesthetized and then perfused with heparin sodium saline and 4% paraformaldehyde (PFA). Brains of rats were fixed in PFA overnight followed by a graded dehydration (10%, 20% and 30% sucrose solution) at 4 °C and then were made into slices (30 μm) using the frozen slicer. The slices were washed 3 × 5 min in 1 × PBS and incubated in the blocking solution (1 × PBS + 5% goat serum + 2.5% BSA + 0.2% Triton X-100) for 1 h. Slices were incubated with primary antibodies in the blocking solution at 4 °C overnight. On the second day, slices were incubated with matched secondary antibody (1:1000, Invitrogen) in the blocking solution at 4 °C for 1 h and 5 min in DAPI (Beyotime Biotechnology C1002). Images were obtained from high-speed confocal platform (Dragonfly 200).

Transmission electron microscopy (TEM)

Fresh sample of hippocampal tissue (size: 1 mm × 1 mm × 1 mm) was dissected and placed in 2.5% glutaraldehyde at room temperature for 0.5 h and then at 4 °C overnight. The samples were washed 3 × 10 min in 0.1 M PBS (pH 7.4) and then fixed with 1% osmium for 1.5 h. After that, samples were washed 3 × 10 min in ultrapure water followed by dehydration and infiltration. Ultrathin section (Leica UC7) of tissues were performed after embedding in resin. Pictures were finally captured by Talos F200C TEM (JEM-1200EX TEM) and randomly selected from each rat for analysis.

Western blot analysis

Rats were anesthetized with sodium pentobarbital (150 mg/kg, i.p.) and the hippocampus regions were carefully dissected for western blot analysis at 24 h after behavioral tests. Briefly, hippocampus was homogenized in ice-cold RIPA buffer (catalog R0020, Solarbio) with a cocktail of protease/phosphatase inhibitors (catalog P1260, Solarbio). The homogenate was centrifuged at 12,000×g for 25 min at 4 °C, and supernatants were collected. Protein concentrations in hippocampus was determined using BCA Protein Assay Kits (catalog CW0014s, CWBIO). Equal amounts of protein samples (30 μg) were separated by SDS-PAGE and transferred onto PVDF membranes, which were blocked in 5% nonfat milk for 1 h, and then incubated overnight at 4 °C with the appropriate primary antibodies including polyclonal rabbit anti-CD11b (1:1000, catalog ab184307, Abcam), polyclonal rabbit anti-NF-κB p65 (1:1000, catalog BS90940, BioWorld), monoclonal rabbit anti-NLRP3 (1:1000, catalog ab263899, Abcam), polyclonal rabbit anti-GAPDH (1:5000, catalog 10494-1-AP, Proteintech), polyclonal rabbit anti-CD45 (1:500, catalog ab10558, Abcam) and polyclonal rabbit anti-P38 (1:1000, catalog 9212s, Cell Signaling Technology). The membranes were incubated with secondary horseradish peroxidase-conjugated antibodies goat anti-rabbit IgG (1:5000, catalog ZB-2301, Zhongshan Golden Bridge Biotechnology) and Peroxidase-conjugated goat anti-mouse (1:2000, catalog ZB-2305, Zhongshan Golden Bridge Biotechnology). Blots were detected using an enhanced chemiluminescence kit (catalog E412-01, Vazyme). Protein band densities were quantified using Image-J software and the experiment with the samples of each rat was replicated at least three times.

Reverse transcription PCR (RT-PCR) and real-time quantitative PCR

Total RNA was extracted from tissues of hippocampus using the RNA extraction kit (catalog AP-MN-MS-RNA-50, Axygen) according to the manufacturer’s instructions.

For mRNA

Reverse transcribe RNA into cDNA using SureScript™ First-Strand cDNA Synthesis Kit (catalogQP056, GeneCopoeia™). The reverse transcription reaction system was prepared using the following components: 1 µl SureScript RTase Mix (20x), 4 µl SureScript RT Reaction Buffer (5x), 1ug or 10ug total RNA. The whole system was replenished to 20 µl with ddH2O (RNase/DNase free). Expression levels of mRNA were determined using BlazeTaq™ SYBR Green qPCR Mix 2.0 (catalog QP031, GeneCopoeia™). QPCR reaction system was prepared using the following components: 2 µl 5xBlazeTaq qPCR Mix, 2 µl specific primers, 1-2 µl cDNA (diluted 1:20), the whole system was replenished to 10 µl with ddH2O (RNase/DNase free).

For microRNA

Reverse transcribe RNA into cDNA using All-in-OneTM miRNA qPCR Detection Kit 2.0 (catalog QP115, GeneCopoeia™). The reverse transcription reaction system was prepared using the following components: 1 µl Poly-A Polymerase, 1 µl SureScript™ RTase Mix (20x), 4 µl 5xPAP/RT Buffer, 100 ng–1ug Total RNA. The whole system was replenished to 20 µl with ddH2O (RNase/DNase free). The reverse transcription process was performed on PCR thermal cycler (Hangzhou Jingle Scientific Instrument Co, Ltd). Expression levels of miRNA were determined using All-in-One™ miRNA qPCR Detection Kit 2.0 (catalog QP115, GeneCopoeia™). QPCR reaction system was prepared using the following components: 5 µl 2xAll-in One TM qPCR MIX, 1 µl Universal Adaptor PCR Primer (diluted 1:24), 1 µl specific primer, 1–2 µl cDNA (diluted 1:7.5). The whole system was replenished to 10 µl with ddH2O (RNase/DNase free.

Real-time quantitative PCR analysis was performed on a Bio-Rad iCycler system (Bio-Rad). GAPDH served as a loading control for the sample to test for mRNA and Rno-U6 served as a loading control for the sample to test for miRNA. The mRNA expression levels and miRNA expression levels were evaluated using the 2 − (ΔΔCt) method. All samples were repeated twice to reduce the error and all the special primers are obtained from BGI Genomics Co., Ltd (Additional file 1: Table S1).

Golgi staining

Golgi staining was performed as described previously [35]. Fresh brain tissues of rats were collected for the Golgi–Cox staining test according to the user manual provided by the FD Rapid Golgi Stain™ kit (PK401, FD Neuro Technologies. INC). Brains were sectioned into 150-μm coronal slices with the vibratome (Leica VT1200S, Germany), and collected on the pre-gelatin-coated microscope slides, dehydrated with a gradient series of alcohols, finally cleared in xylene and covered slip. Slides were kept at room temperature, away from light. Stained pyramidal neurons in hippocampal region were captured with use of panoramic digital section scanning microscope (Olympus microscope VS120, Japan). Dendritic spine densities were calculated as the number of dendritic spines per 10 μm of dendrite length. All images were Fiji (Image J, NIH) processed, including Sholl analysis.

Molecular docking

The protein structures of p38 and p65 derived from rat and human were obtained, respectively, from UniProt database (https://www.uniprot.org/), which is a database of sequence and function information for protein where a large amount of protein information derived from the research literature. Gramm-x server (http://vakser.bioinformatics.ku.edu/resources/gramm/grammx) was used for rigid molecular docking between proteins to evaluate the possibility of their interaction. P38 is defined as a ligand while P65 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.

Electrophysiology recording

Hippocampal slice preparations and electrophysiological recordings were performed according to procedures described previously [21]. Animals were decapitated and brains were removed transferred to the ice immediately. All brain slices (300 mm) were prepared with use of the microtome (VT1200s, Leica, Germany) at rate of 0.18 mm/s velocity in 1 × cutting solution (119 mM choline chloride, 30 mM glucose, 26 mM NaHCO3, 7 mM MgSO4, 2.5 mM KCl, 1 mM NaH2PO4, 1 mM CaCl2, 3 mM sodium pyruvate, 1.3 mM sodium l-ascorbate and 1 mM kynurenic acid). Brain slices were maintained for 30 min at 33 ℃ in the recovery solution containing 8 mM NaCl, 24 mM NaHCO3, 4 mM MgCl2, 2.5 mM KCl, 1.2 mM NaH2PO4, 0.5 mM CaCl2, 25 mM glucoses and 50 mM sucrose, and then at room temperature for at least 30 min. All solutions during the experiments were continuously infused with 95% O2/5% CO2. Whole cell recordings were performed by use of MultiClamp 700B amplifier and the glass pipettes (4–6 MΩ, World Precision Instruments) filled with the internal solution (115 mM CsMeSO3, 20 mM CsCl, 10 mM HEPES, 2.5 mM MgCl2, 4 mM Na2-ATP, 0.4 mM Na-GTP, 10 mM Na-phosphocreatine, and 0.6 mM EGTA). For mEPSC recordings, neurons of rats were clamped at − 70 mV in the presence of TTX (1 μM) and picrotoxin (100 μM) in the ACSF (artificial cerebral spinal fluid: 120 mM NaCl, 3.5 mM KCl, 2.5 mM CaCl2.2H2O, 1.3 mM MgSO4, 1.25 mM NaH2PO4, 26 mM NaHCO3, 10 mM glucose). Data were filtered at 2 kHz and sampled at 10 kHz using Digidata 1440A. For sEPSC recordings, neurons were recorded at − 70 mV in the presence of picrotoxin (100 μM) in the ASCF. Data were analyzed with Mini Analysis Program (Mini 60, Synaptosoft).

Statistical analysis

All data were present as the means ± SEMs, and statistical analyses were performed using GraphPad Prism software version 8.0. Independent group Student’s t-tests (two-tailed) were used to analyze in comparisons between two independent groups. One-way or two-way analysis of variance (ANOVA) was used to analyze in comparisons between three or more groups followed by Tukey’s post hoc test. All experiments were repeated at least three times comprising a minimum of 6 animals/group. P value < 0.05 was required for results to be considered statistically significant.