Rat models and related experimental operations

Healthy male SD rats weighing approximately 250 g were used for the experiment. All animal studies were approved by the Ethics Committee of the Tianjin Eye Hospital and were conducted by skilled experimenters according to an approved protocol in accordance with the principles and procedures outlined in the NIH Guide for the Care and Use of Laboratory Animals. Under the microscope, corneal epithelium was mechanically scraped with corneal epithelial scrapers used for LASEK. CXCL14 antibody (Abcam, Cambridge, UK) and PAX6 antibody (Abcam, USA) were used for immunohistochemistry staining.

To verify whether the central corneal epithelium (CCE) has a repairing effect during corneal injury, we established 3 corneal injury models by mechanical curettage (with 8 rats in each model group): the CCE curettage group, the LE curettage group, and the whole corneal epithelium (WCE) curettage group. The calculation formula of corneal healing rateis: healing rate = repair area/damage area × 100%. Haze was assessed according to the evaluation method of the corneal injury experiment by Yoeruek et al. [20]. Corneal haze was divided into 5 grades: 0, completely transparent; 1, extremely mild haze, with the iris and pupil clearly visible; 2, mild haze, with the iris and pupil still visible; 3, moderate haze, with the iris and pupil faintly visible; and 4, severe haze, with the iris and pupil not clearly seen. Neovascularization was assessed by using a slit lamp and statistical results were obtained by analyzing tissue sections under a microscope (20 ×). These characteristics were observed at 0 h, 12 h, 24 h, 48 h, 72 h, 96 h, 1 w, 1 m, 3 m, and 6 m after operation under the slit lamp with normal light or cobalt blue light after fluorescein sodium staining.

To investigate which genes promote the repair of corneal injury in CCE, we established a model of LE scrape injury in rats (3 cases). After 24 h, CCE of the operative eye and the contralateral normal eye in the same rat were taken, and the genomes of these tissues were sequenced and analyzed.

To verify the expression of CXCL14 in the cornea, the corneal and kidney tissues of the same normal rat were evaluated by immunohistochemistry (4 cases), real-time polymerase chain reaction (PCR) (8 cases), and Western blot analysis (8 cases).

We also constructed stable corneal models with CXCL14 overexpression or downregulation by using lentivirus eye drops in rats, and the results of immunohistochemistry showed that the model was constructed successfully and was stable (Supplementary Fig. 1). Then, we scraped the LE in each model, we observed the healing and corneal state under the slit lamp at 0 h, 12 h, 24 h, 48 h, 72 h, 96 h, 1 w, 1 m, etc., after operation, under the conditions of common illumination and cobalt blue light after fluorescein sodium staining (Fig. 2F–H).

Cell culture related experimental operations

HCE-2 (50, B1) cells and rat primary corneal epithelial cells were cultured in Dulbecco’s Modified Eagle’s Medium (Nutrient Mixture F-12 (DMEM/F12)), supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 mg/mL streptomycin (Gibco, Grand Island, NY, USA) at 37 °C and 5% CO2. 293 T cells were cultured in DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 mg/mL streptomycin.

We construct stable cell lines with CXCL14 overexpression or downregulation in human corneal epithelial cells (HCECs) (HCE-2 [50, B1]), with corresponding control cells, by the method of stable transfection with lentiviral packaging, and we verified the success of the cell line construction by Western blot analysis (Fig. 2I). Then, we used the stable HCEC lines to carry out the wound healing assay. We took photos of the cells at the time points of 0 h, 24 h, and 48 h under the microscope, and we measured the healing area.

To explore the mechanism by which CXCL14 promotes corneal repair after injury, we used HCE- 2 (50, B1) cells with CXCL14 overexpression or downregulation to perform soft agar assay. We inoculated 6-well plates with 800 cells per well, with 3 holes for each group. We fixed and stained the cells with crystal violet after 7 days of culture, counted the cells, and performed statistical analysis.

To confirm whether CXCL14 mediates corneal epithelial cell repair after injury through NF-κB signaling, we stimulated HCE-2 with recombinant CXCL14 (20 ng/mL) for 4 h. After co-culture for 0 h, 0.5 h, 1 h, 2 h, and 4 h, the protein was extracted and detected by Western blot analysis.

To investigate how CXCL14 regulates the NF-κB signaling pathway, we analyzed the above RNA sequencing data and screened out several genes, including SDC1, STAT1, STAT3, JAK1, and ERK1/2, which expression increased with CXCL14 overexpression in CECs. Then, the co-immunoprecipitation experiment was carried out, and we found that SDC1 binded to CXCL14 significantly among these proteins (Fig. 5B). Flag-tagged SDC1 and V5-tagged CXCL14 vectors were transfected into HCE-2 cells, and immunoprecipitation was performed with the indicated antibodies. Cell lysates and immunoprecipitated proteins were analyzed by immunoblot (IB) with the indicated antibodies.

Specific experimental processCoimmunoprecipitation assay

HCE-2 (50, B1) cells were seeded in a 10-cm dish and transiently transfected with 3 μg of pFLAG-plasmid and 3 μg of MYC-plasmid by Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). After 24 h of post-transfection, cells were lysed by sonication in Immunoprecipitation (IP) buffer. IP assay was carried out by incubating the supernatants with an anti-FLAG antibody (Sigma, #SAB4301135) or an anti-MYC antibody (Cell Signaling Technology, #2276), followed by incubation with protein G agarose (Thermo-Fisher Scientific) at 4 °C overnight. Then, proteins bound to the beads were eluted by boiling in Sodium dodecyl sulfate (SDS) loading buffer and separated by SDS-PAGE for Western blot analysis to detect CXCL14 and SDC1.

Chromatin immunoprecipitation assay

Chromatin immunoprecipitation assays were performed by using a ChIP kit (Millipore) according to the manufacturer’s instructions. Briefly, HCECs were cross-linked with 1% formaldehyde for 10 min at room temperature. After sonication and centrifugation, the supernatant was collected for anti-Flag immunoprecipitation. Anti-RNA polymerase and anti-rabbit IgG were also used as positive and negative controls, respectively. Semi-quantitative real-time PCR was performed to detect DNA fragments in the CXCL14 promoter region. The PCR primers were as follows:

Dual-luciferase reporter gene detection assay

Vectors overexpressing PAX6 (pcDNA-PAX6) and control vectors (pcDNA-vector), along with luciferase expression vectors carrying mutations in the BM_c region (pGL3.1-CXCL14-HRE.MUT(c)), the BM_e region (pGL3.1-CXCL14-HRE.MUT(e)), and both BM_c and BM_e regions (pGL3.1-CXCL14-HRE.MUT(c + e)) of the CXCL14 gene promoter, were co-transfected into HCE-2 cells. pGL3.1-CXCL14 was used as a positive control, and pGL3.1 EV as a negative control, to analyze the impact of PAX6 overexpression on the transcriptional activity of the CXCL14 promoter.

HCE-2 cells were routinely cultured and replated at 50% density; when cells reached 70% confluency, they were transfected with the necessary plasmids for the luciferase report, along with corresponding transfection reagents. 48 h post-transfection, cells were lysed according to the lysis buffer instructions provided in the kit, and supernatants were collected post-centrifugation. The appropriate substrate was diluted as per the manual and protected from light. The prepared working solution was added to the cell supernatants to be tested. A luminometer was used to measure and compare the data, analyzing the impact on the binding sites.

Immunofluorescence staining

Cells grown on coverslips were fixed with 4% paraformaldehyde and permeabilized in 0.25% Triton X-100 for 10 min at room temperature. After incubation with blocking buffer for 1 h, the cells were incubated at 4 °C overnight with anti-NF-κB antibodies (CST, #8242). Coverslips with cells were washed 3 times and then incubated with secondary antibodies (goat anti-rabbit IgG or goat anti-mouse IgG antibodies conjugated with Alexa488) for 1 h at room temperature. Cells were then washed and stained with 0.1 μg/mL DAPI (Sigma-Aldrich) and mounted onto glass slides with ProlongGold Antifade Reagent (Thermo-Fisher).

Western blot analysis

Whole-cell extracts were lysed in cold RIPA lysis buffer supplemented with a protease inhibitor cocktail, and the lysate was subjected to SDS-PAGE. Proteins were transferred to nitrocellulose membranes (Millipore, Billerica, MA, USA) and blotted with antibodies against CXCL14, p63, NK-κB (p65), NK-κB (p50), Ki67, SDC1, Flag, PAX6, and β-actin.

Real-time polymerase chain reaction

The RAN of central corneal tissue, limbal tissue, and kidney tissue of rats were extracted with TRIzol (Invitrogen) according to the manufacturer’s instructions. Then, the mRNA was reverse transcribed to cDNA using an real-time PCR system (TaKaRa), and real-time PCR were performed with Hieff qPCR SYBR Green Master Mix (Yeasen Biotech) and LightCycler 96 (Roche, Basel, Switzerland). The primers were as follows, and β-actin was used as a loading control.

Wound healing assay

Approximately 5 × 105 cells were seeded in a 6-well plate. Wounds were generated by a conventional pipette tip when the cells were confluent. Images were taken immediately or at 6 h, 12 h, and 24 h post-wounding. Migration distance was determined by using Image J software and the wound healing rate is calculated as follows, the width of the unhealed wound/the initial width of the wound × 100%

Cell colony formation assay

Cells were seeded in 6‐well plates at a density of 800 cells/well. After approximately 1 week, the cells grew to visible colonies and were stained with crystal violet. The colonies were counted under the microscope.

Cell cycle assay

Cells were seeded in 6-well plates and fixed with 70% ethanol at 4 °C overnight. After washing, cells were resuspended and concomitantly treated with RNaseA (Sigma-Aldrich) and stained with 50 μg/mL of propidium iodide for 30 min. Cell cycles were analyzed by using a Beckman Coulter EPICS flow cytometer (Krefeld, Germany).