AnimalsFour 5–6-week-old New Zealand rabbits, weighing 1.5–2 kg, were obtained from the Department of Laboratory Animal Resources, Yonsei Biomedical Research Institute, Yonsei University College of Medicine. All animals were raised in accordance with the guidelines specified by the Institutional Animal Care and Use Committee, at a temperature of 25 °C and a 12-h light-dark cycle. The animal study protocol was approved by the Institutional Ethics Committee of the Department of Laboratory Animal Resources, Yonsei Biomedical Research Institute, Yonsei University College of Medicine (approval no. 2021-0054). This study was carried out in strict accordance with the recommendations in the ARRIVE guidelines. All surgery was performed under anesthesia, and every effort was made to minimize suffering.Form deprivation myopia inductionWe used the same method for form deprivation myopia induction as in our previous study7. Anesthesia was induced through subcutaneous injection of Zoletil (0.3 mL/kg, Virbac, Carros, France) and Rompun (0.2 mL/kg, Elanco, Indiana, US). To manage local pain and minimize bleeding, additional subcutaneous injections of Xylocaine (lidocaine HCl 2%: epinephrine mix = 1:100,000, Yuhan, Seoul, Korea) were administered above and below the eye. A horizontal incision was then made on the right eyelid to form two layers—inner and outer. The left eye was left untreated to serve as a control, while the right eye underwent light deprivation to induce myopia for comparison of the experimental effects. The inner layer was sutured with 6 - 0 Vicryl (Ethicon, Raritan, NJ, USA) and the outer layer with 5 - 0 Nylon (Ethicon, Raritan, NJ, USA). Postoperatively, Baytril (0.1 mL/kg, Elanco, Indiana, USA) and Keromin (0.1 ml, Hana Pharm, Hwaseong-si, Korea) were injected for 7 days to manage pain and prevent inflammation. Additionally, Tarivid (Santen, Osaka, Japan) was applied topically to the surgical site during this period for the same purposes. The Nylon sutures in the outer layer were removed after 7 days. The sutured eyelids were maintained for 4 weeks. Thereafter, the same anesthesia protocol was followed, and the eyelid surgery site was incised to examine the eyeball. Axial length measurements were conducted using an A-scan ultrasound (UD-6000, Tomey, Nagoya, Japan). After local anesthesia of the eye with proparacaine (Hanmi, Seoul, Republic of Korea), measurements were taken by placing the device in contact with the cornea. Once the alignment was accurately set from the cornea to the retina, more than seven measurements were taken, and the average value was calculated. The refractive error was measured using streak retinoscopy both before and after myopia induction.Retinal tissue preparationIn this study, rabbits were initially anesthetized using Zoletil (0.3 mL/kg, Virbac, Carros, France) and Rompun (0.2 mL/kg, Bayer Animal Health, Germany) administered intramuscularly to ensure adequate sedation for the procedures. After confirming deep anesthesia by verifying the absence of reflexes, potassium chloride (KCl) at a dosage of 1 mL/kg was administered intravenously to induce cardiac arrest. The use of KCl under anesthesia ensured that the animals experienced a humane and painless death. This method adhered to institutional ethical guidelines and followed the AVMA Guidelines for the Euthanasia of Animals. After euthanasia, the eyeballs were promptly enucleated. The eyeballs from one rabbit were designated as whole-retinal block preparations for immunofluorescence staining. The remaining three pairs of eyeballs were used for retinal tissue preparation. The cornea, lens, and vitreous humor were excised, and the remaining eyeball was cut into a petaloid configuration. The retina was segmented into central and peripheral regions based on the distance from the visual streak to the limbus. Each region was separately immersed in Dulbecco’s phosphate-buffered saline (DPBS; Thermo Fisher Scientific, Waltham, MA, USA).Isolation of Müller cells from rabbit retinaFollowing segmentation, the retinal tissue was enzymatically dissociated into single cells using the Neural Tissue Dissociation Kit – Postnatal Neurons (Miltenyi Biotec, Bergisch Gladbach, German), as described in a prior study39. To eliminate microglia and vascular endothelial cells, the cell suspension was treated with CD11b-FITC (6 µg/ml, MA110081, Invitrogen, Waltham, MA, USA) and CD31-FITC (4 µg/ml, NBP2-33136 F, Novus Biologicals, Minneapolis, MN, USA) antibodies at 4 °C for 30 min. Subsequent separation involved the application of Anti-FITC Microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) for 15 min at 4 °C, followed by magnetic sorting using an MS column. For selective enrichment of Müller cells, CD29-FITC (1:50, bs-0486R-FITC, Bioss, Woburn, MA, USA) was added and incubated at 4 °C for another 30 min. The cells were then washed with DPBS, mixed with microbeads, and again passed through an MS column to segregate the CD29+ and CD29- cell populations.Protein extraction and processingProteins were extracted from Müller cells obtained from three distinct retinal samples. Each sample was lysed in 8 M Urea and 100 mM ammonium bicarbonate, followed by sonication. The lysate was centrifuged at 14,000 rpm for 15 min to collect the supernatant containing proteins. Protein concentration was quantified using a micro BCA protein assay according to the manufacturer’s instructions (23235, Thermo Fisher Scientific, Waltham, MA, USA). The Seppro IgY Depletion Kit (Sigma-Aldrich, St Louis, MO, USA) was used to eliminate antibodies and reduce protein abundance. Protein denaturation and reduction were achieved by adding 10 mM DTT, and alkylation was performed using 30 mM iodoacetamide (IAA). Digestion was performed overnight at 37 °C with trypsin and stopped by adding 0.4% trifluoroacetic acid. The peptides were purified using a Pierce C18 Spin Column, dried, and stored at -80 °C until further analysis.Protein identification using LC-MS/MSFor Label-Free Quantification (LFQ) analysis, peptides were reconstituted in 0.1% formic acid (FA) in water and analyzed using a Q Exactive HF-X mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) coupled to a nano EASY-nLC 1000 system (Thermo Fisher Scientific, Seattle, WA, USA). The peptides (2 µg) were loaded onto a trap column (acclaim PepMap 100, 75 μm × 2 cm, 3 μm, C18, 100 Å) in buffer A (0.1% formic acid) and ionized via a spray column (PepMap RSLC C18, 50 cm × 75 μm ID) packed with 2 μm C18 particles at an electric potential of 1.8 kV. Peptides were eluted using a gradient of buffer B (0.1% FA in acetonitrile) ranging from 5 to 80% at a flow rate of 300 nL/min. Full MS data were acquired in a scan range of 400–2000 Th at a resolution of 60,000. The automated gain control (AGC) target value was set at 3.0 × 106, with a maximum ion injection time of 100 ms for the MS scans. The top 20 most abundant ions with an isolation window of 2.0 m/z were fragmented via data-dependent MS/MS experiments with an exclusion duration of 30 s and a normalized collision energy of 27 for HCD. The maximum ion injection time for MS/MS scans was 100 ms, and the AGC target value was set at 1.0 × 106.Raw data processing and proteomics data analysisFor data analysis profiling, the MS/MS spectra were searched using the proteome discoverer software version 2.5 (Thermo Fisher Scientific, Waltham, MA, USA) against the UniProt rabbit proteome database, which was last updated in November 2023. The criteria for protein identification included the detection of at least one unique peptide per protein. Specific attention was paid to unique peptides that were altered by carbamidomethylation of cysteine residues, N-acetylation, or methionine oxidation. To maintain rigorous quality control, a false discovery rate cutoff of 1% was used at both the peptide spectrum match and protein levels. Log2 transformation was applied to all LFQ intensities to minimize the impact of outliers, and sum normalization was used to correct for biological and technical variations in the dataset. Missing protein values in any group were excluded to enhance data reliability.Protein intensities were normalized by total sum scaling to make them comparable, and the ratio of the mean value of normalized protein intensities was calculated for each protein. Statistical and bioinformatics evaluation of the proteomics data, utilizing normalized protein abundance, was conducted using R (version 4.4.0, as of May 2024). This included clustering analyses such as heatmap and hierarchical clustering, which employed Euclidean distances and Ward’s linkage methods to elucidate the variances and distinctions between the samples. Differential expression was determined based on proteins exhibiting at least a 1.2-fold change in LFQ intensity and a p-value < 0.1, as calculated via Student’s t-test. However, the number of DEPs identified using these criteria was too small to perform a robust Gene Ontology (GO) analysis (Table 1).GO enrichment and network analysesTo address this limitation, we expanded our analysis to include all the identified proteins, categorizing them based on whether they showed an increase or decrease in fold change under the myopic condition compared to that under the control conditions, regardless of their p-values. This approach allowed for a more inclusive analysis of proteomic shifts between conditions and facilitated the comprehensive exploration of GO annotation and network interactions.GO biological processes associated with the identified proteins and DEPs were examined using the Database for Annotation, Visualization, and Integrated Discovery, along with the g: Profiler online tool40. Protein networks and interactomes relating to the DEPs were explored using the public database STRING 12.0 and were visualized using Cytoscape. Furthermore, the ClueGO plugin for Cytoscape (version 2.5.10, available at https://apps.cytoscape.org/apps/cluego, accessed March 2024) was employed to organize and visualize GO terms, drawing on UniProt GO annotations for this analysis.Confirmation of reactive gliosis via immunofluorescence stainingTo confirm the presence of reactive gliosis in the retina following myopia induction, GFAP expression was assessed using immunofluorescence staining. The whole eyeball was fixed in 4% paraformaldehyde for 3 days, embedded in paraffin, and sectioned into 4 μm-slides. Slides were deparaffinized using xylene and rehydrated using a graded series of ethanol. Antigen retrieval involved heating the sections at 95 °C for 30 min in citrate buffer for antigen retrieval. Sections were blocked at room temperature for 1 h using 10% normal goat serum (50062Z, Thermo Fisher Scientific, Waltham, MA, USA) to prevent nonspecific antibody binding. Subsequently, the slides were incubated overnight at 4 °C with an anti-GFAP antibody (ab7260, Abcam, Cambridge, UK). After incubation with the primary antibody, the slides were washed several times with PBS to remove any unbound antibodies and then incubated with the anti rabbit alexa fluor 647 antibody (711-605-152, Jackson ImmunoResearch, West Grove, PA, USA) at room temperature for 1 h. Nuclei were stained with DAPI during the final wash step for 5 min. After staining, the slides were washed with PBS to remove excess DAPI and residual antibodies. The prepared slides were mounted using anti-fade mounting medium and covered with coverslips. Immunofluorescence images of the stained sections were captured using an Axio Imager M2 microscope (Carl Zeiss, Oberkochen, Land Baden-Württemberg, Germany) and analyzed using the Zen software (Carl Zeiss, Oberkochen, Land Baden-Württemberg, Germany).Immunofluorescence staining of the isolated CD29+ cellsCD29+ cell fractions were plated onto 4-well chamber slides to facilitate cell adherence. Once the cells had adhered, the slides were gently washed with PBS and fixed with 4% paraformaldehyde (PFA; Biosesang, Yongin, Korea) at room temperature. For permeabilization, 0.1% Triton X-100 in PBS was added to the samples. To block nonspecific binding, the cells were incubated with 10% normal goat serum (50062Z, Thermo Fisher Scientific, Waltham, MA, USA) at room temperature for 1 h. Overnight incubation at 4 °C was carried out with primary antibodies against GLUL (ab73593, Abcam, Cambridge, UK) and GFAP (G3893, Sigma Aldrich, St Louis, MO, USA), which were specifically selected to verify the identity of Müller cells38,41. Following incubation with primary antibodies, the slides were washed and then incubated with the anti rabbit alexa fluor 647 (711-605-152, Jackson ImmunoResearch, West Grove, PA, USA) and anti mouse alexa fluor 488 (A11001, Invitrogen, Waltham, MA, USA) at room temperature for an additional hour. Nuclei were stained with DAPI. Immunofluorescence images were captured using an Axio Imager M2 microscope (Carl Zeiss Oberkochen, Land Baden-Württemberg, Germany). Fluorescence intensity was quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Following the acquisition of fluorescence images, regions of interest (ROIs) were manually selected for each sample. The mean fluorescence intensity within each ROI was measured, and background fluorescence was subtracted to correct for non-specific signal. The quantified fluorescence intensities were averaged across multiple independent samples to ensure the reliability and reproducibility of the results.Validation of selected Müller cell enriched genes using qRT-PCRDue to limited quantity of isolated Müller cells, direct validation using these cells was not feasible. As an alternative approach, we utilized whole retinal tissues from the central and peripheral regions to validate our findings. qRT-PCR was performed for LDHA and PKM, which were consistently upregulated in both regions according to our proteomic analysis. Whole RNA was extracted from the central and peripheral regions of the rabbit retina using Trizol reagent (15596026, Invitrogen, Waltham, MA, USA). The extracted RNA was synthesized into cDNA using a premix. RT-qPCR reactions were performed on a QuantStudio 3 system (Applied Biosystems, Waltham, MA, USA) using Power SYBR™ Green PCR Master Mix (4367659, Applied Biosystems™, Waltham, MA, USA). The relative expression of the target genes was calculated using the comparative threshold method, normalized to the GAPDH housekeeping gene. The primer sequences are listed in Supplementary Table S2.
