Materials and methods

Animals

Male C57BL/6J mice, 8 weeks old and weighing 21–25 g, were purchased from Vital River (Beijing, China). All animals were acclimatized for one week in an SPF-grade facility with controlled conditions (24 ± 2 °C, 40 ± 5% humidity). Following random assignment to the experimental or control group, all animals were maintained on a 12-hour light/dark cycle. Mice were euthanized by CO2 inhalation in an induction chamber with a displacement rate of 30% chamber volume/min. Following respiratory arrest, animals were maintained in the chamber for at least 2 min, and death was subsequently confirmed by cervical dislocation. All experimental procedures were performed in accordance with the Guide for Care and Use of Laboratory Animals and the Chinese national standard GB/T 35,892 − 2018. The study was approved by Institutional Animal Care and Use Committee of Shenzhen Rongwan Biomedical Experimental Animal Center.

Primary microglia culture

Primary microglia were isolated from neonatal C57BL/6J mice at postnatal days 0–3. Pups were not selected or stratified according to sex, and cerebral cortices from three pups within the same litter were pooled for each culture flask. The cells were then maintained in DMEM with 10% FBS and 1% penicillin-streptomycin (P/S) under standard conditions. The cerebral cortices were retained and mechanically dissociated by gentle trituration with a 10 mL pipette. The cell suspension was transferred to T75 culture flasks with three brains per flask. We replaced the culture medium after 24 h and every 5 days thereafter. After 14 days in vitro, the flasks were placed on an orbital shaker and incubated overnight to detach microglia. The next day, the supernatant with primary microglia was gathered and spun at 200 g for 5 min. The cell pellet was then redissolved, and cells were counted and seeded as required for subsequent experiments.

HT-22 cell culture

HT-22 cells were maintained in DMEM supplemented with 10% FBS. Every 1–2 days, the culture medium was refreshed.

Oxygen-glucose deprivation (OGD) model

HT-22 cells were cultured to 80–90% confluence. Primary microglia were subjected to OGD 1–2 days after seeding. Cells were washed twice with DPBS, and the culture medium was replaced with glucose-free DMEM. The cells were then exposed to hypoxic conditions (1.5% O₂) for the indicated durations. An oxygen concentration of 1.5% was selected based on published studies using organotypic hippocampal slice OGD and primary microglial hypoxia models16,17. The duration of OGD was subsequently optimized in primary microglia. After OGD, the cells were washed with DPBS and returned to their respective culture media. Reoxygenation was performed under normoxic conditions at 37 °C with 5% CO₂ for 24 h. The cells were subsequently harvested for further analyses.

In vitro rCXCL16 treatment

Before treatment, primary microglia were deprived of serum in DMEM with 1% P/S for 2 h. For experiments under hypoxic conditions, cells were treated with increasing amounts of rCXCL16 (30, 60 and 120 ng/ml) as indicated and incubated overnight prior to OGD/R induction. For experiments involving LPS stimulation, microglia were co-treated with 120 ng/ml CXCL16 and 500 ng/ml LPS.

Co-culture system of primary microglia and HT-22 cells

Microglia were seeded into 0.4 μm transwell inserts, and neurons were maintained in the lower wells. Both cell types underwent OGD/R modeling following the same protocol as described in the drug treatment section. The timing was coordinated such that microglia completed reoxygenation just as neurons began reoxygenation. At this point, transwell inserts containing microglia were transferred to the neuronal wells and co-cultured for 24 h.

3-(4,5-dimethylthiazol-2-yl)−2,5-diphenyltetrazolium bromide (MTT) assay

Cell viability was assessed using the MTT assay. Active cells reduce tetrazolium salts to purple formazan crystals, forming the basis of this assay. Cells were exposed to 100 µl of MTT solution (5 mg/ml in PBS) at 37 °C for 4 h following OGD/R. The resulting crystals were solubilized in 1 ml of DMSO with gentle shaking for 5 min. Subsequently, a 200 µL portion of the dissolved solution from each well was moved to a 96-well microplate for spectrophotometric measurement of absorbance at 570 nm.

2,3,5-triphenyltetrazolium chloride (TTC) staining

The brains of mice were rapidly extracted and placed in a brain matrix 24 h after MCAO, following euthanasia. The brains were coronally sectioned into 2 mm-thick slices. Afterward, the sections were immersed in a 2% TTC solution and kept at 37 °C for 15 min. After staining, the slices were imaged, and the volume of the infarct was measured with ImageJ software. To correct for brain edema, infarct volume was determined using the indirect method: infarct volume = contralateral hemisphere volume − non-infarcted ipsilateral hemisphere volume.

Middle cerebral artery occlusion (MCAO) model establishment and drug administration

We fasted the mice before surgery but allowed them free access to water. We induced and maintained anesthesia with 1.5% isoflurane and maintained body temperature at 37 °C using a heating pad. Laser Doppler flowmetry with a flexible probe placed on the skull overlying the MCA territory monitored cerebral blood flow. After cervical disinfection, a midline skin incision was made to dissect the left carotid bifurcation. The distal common carotid artery (CCA) and the external carotid artery (ECA) were ligated. The internal carotid artery (ICA) was temporarily occluded with an arterial clip. To induce MCAO, we made a small incision in the CCA, inserted the filament (Doccol Corporation, USA), and advanced it into the ICA until resistance was detected. Forty-five minutes later, the filament was withdrawn, and the CCA was ligated at the incision site. After each use, the filaments were cleaned, sterilized, and stored in sterile saline at 4 °C. We performed all surgical procedures on the sham group mice except that the filament was not inserted.

Recombinant mouse CXCL16 (Peprotech) was solubilized in saline and delivered via intracerebroventricular (ICV) injection at 1 h post MCAO. For dose-response studies, mice received 10, 30, or 60 µg/kg of CXCL16 in 2 µL of saline. Stereotaxic frame placement (RWD Life Science, Shenzhen, China) was performed under anesthesia for each animal, and the injection was targeted to the left lateral ventricle (3 mm below the dura, 1 mm lateral to bregma). The solution was infused at a constant rate of 0.2 µL/min using a R462 high-precision microinjection pump (RWD Life Science, Shenzhen, China).

Modified neurological severity Score(mNSS)

mNSS was used to evaluate motor coordination, sensory reflexes, and postural stability. The scoring system ranges from 0 to 18, and higher scores correlate with more severe neurological dysfunction.

TUNEL staining

TUNEL staining was carried out using a One-step TUNEL In Situ Apoptosis Kit (Elabscience, Wuhan, China).

Immunofluorescence staining

After transcardial perfusion with ice-cold PBS and 4% paraformaldehyde (PFA), brains were collected and fixed overnight in 4% PFA at 4 °C. Tissues were dehydrated in a 30% sucrose solution until they subsided; this process was repeated once with fresh 30% sucrose. Primary microglia were seeded on coverslips, fixed with 4% PFA for 10 min, permeabilized with 0.25% Triton X-100 for 5 min, and blocked with 5% bovine serum albumin (BSA) at RT for 30 min. Brain sections and cultured cells were then incubated overnight at 4 °C in a humidified chamber with primary antibodies against Iba-1 (1:200, Oasis Biofarm), CXCR6 (1:200, Abclonal), CD86 (1:200, Affinity Biosciences), and CD206 (1:400, CST). After three washes with PBS, samples were incubated with the corresponding secondary antibodies for 1 h at room temperature in the dark. Nuclei were counterstained with DAPI. Coverslips and sections were mounted using anti-fade mounting medium and stored at 4 °C. Images were obtained using a Zeiss LSM900 confocal microscope (Carl Zeiss, Germany).

For quantitative analysis, brain sections from six mice per group (n = 6) were examined. For each mouse, three coronal sections were selected, and five randomly chosen fields per section were imaged. Quantification was performed in a blinded manner, and data from multiple fields and sections were averaged to obtain a single value per animal.

Western blotting

Total protein was extracted and quantified as described previously18. Electrophoresis was run at 80 V for 30 min, and then at 120 V until the dye front reached the bottom of the gel. Before transfer, the PVDF membrane was activated by immersing it in methanol for 3 s. Protein transfer was performed by semi-dry blotting at 1.0 A for 30 min. A blocking step was performed using TBST containing 5% BSA for 1 h. The membranes were then incubated overnight at 4 °C with the following primary antibodies: anti-CXCL16 (1:1000, Affinity Biosciences), anti-CXCR6 (1:1000, Affinity Biosciences), anti-Bax (1:1000, Proteintech), anti-Bcl-2 (1:1000, Proteintech), anti-Arg-1 (1:1000, HUABIO), anti-iNOS (1:1000, HUABIO), and anti-β-actin (1:5000, Proteintech). After washing the next day, the membranes were incubated with a 1:5,000 dilution of secondary antibody for 1 h at RT. After washing, signal detection was performed by incubating the membrane with ECL substrate, followed by imaging using the ChemiDoc imaging system (Bio-Rad, USA). Band intensities were quantified using ImageJ software.

Reverse transcription quantitative PCR (RT-qPCR)

Total RNA was extracted from primary microglia using the Universal RNA Extraction Kit (Accurate Biotechnology, Hunan, China). The concentration and purity of the extracted RNA were determined using a Metash B-600 ultra-micro spectrophotometer (Metash, Shanghai, China). cDNA was synthesized using the RT Mix Kit with gDNA Clean for qPCR (Accurate Biotechnology, Hunan, China). qPCR was carried out using the SYBR Green Premix Kit (Accurate Biotechnology, Hunan, China) on a LightCycler 480 II Real-Time PCR System (Roche, Switzerland). Primer sequences are listed in Supplementary Table 1. We normalized target gene expression to β-actin and calculated relative levels using the 2⁻ΔΔCt method.

Flow cytometry

Cell apoptosis was evaluated using a FITC-labeled Annexin V and PI Apoptosis Detection Kit (Sangon Biotech, Shanghai, China). A suspension of 1 × 10⁶ cells was incubated with Annexin V-FITC and PI for 15 min in the dark. Subsequent analysis was performed on a BD FACSCalibur flow cytometer. Early and late apoptotic cells were quantified and expressed as a percentage of total cells.

Single-cell RNA-sequencing analysis

Raw count matrices from the GSE227651 dataset, comprising sham and days 1, 3, and 7 after MCAO, were processed using Seurat in R. Seurat objects were generated for each sample with genes detected in at least three cells and cells containing at least 200 detected genes. After merging, cells with 300–7,000 detected genes and mitochondrial transcript content below 15% were retained, yielding 57,195 cells. Data were normalized using LogNormalize with a scale factor of 10,000, and 3,000 variable features were selected using the variance-stabilizing transformation method. The variable features were scaled and subjected to principal component analysis. The first 30 principal components were used for neighbor detection, clustering at a resolution of 0.5, and UMAP visualization. Cluster-enriched genes were identified using the Seurat FindAllMarkers function, and cell types were manually annotated based on these genes and established lineage markers. Feature plots were used to visualize Cxcl16 and Cxcr6 expression, and a dot plot was additionally generated for Cxcl16. Because only one sample was available at each time point, no statistical comparisons of temporal changes were performed.

Statistical analysis

Results are reported as the mean ± SD. Statistical significance was defined as P < 0.05. One-way or two-way ANOVA followed by Tukey’s post hoc test was used for comparisons among three or more groups. Data were analyzed using two-tailed paired Student’s t-test for comparisons between normoxia and OGD/R conditions within each independent experiment. Statistical analyses and graph preparation were conducted with Prism 10 (GraphPad Software, USA).