Proteomic Alterations in Retinal Müller Glial Cells Lacking Interleukin-6 Receptor: A Comprehensive Analysis

Abstract

Purpose: Interleukin-6 (IL-6) is an inflammatory cytokine implicated in various retinal pathologies and functions primarily through two signaling pathways: cis-signaling via IL-6 binding to its membrane-bound receptor (IL-6Rα), and trans-signaling via IL-6 binding to soluble IL-6 receptor (sIL-6R). Because the differential effects of IL-6 signaling in retinal Müller glial cells (MGCs) remain unclear, we generated an MGC-specific Il6ra^–/– knockout (KO) mouse to eliminate IL-6Rα and, consequently, IL-6 cis-signaling in MGCs. In this study, we examined the proteomic changes in MGCs isolated from KO mice lacking a functional IL-6Rα.

Methods: The proteomes of MGCs isolated from wild-type (WT) and KO mice were analyzed using liquid chromatography–tandem mass spectrometry (LC-MS/MS) and validated by parallel reaction monitoring (PRM). Relevant biological functions and pathways were examined using Gene Ontology and Ingenuity Pathway Analysis.

Results: LC-MS/MS detected 1866 proteins, of which 81 were significantly altered (41 upregulated, 40 downregulated). PRM analysis confirmed differential expression of Ptgis (fold change [FC] = 3.63), Dpep1 (FC = 2.79), Fmo1 (FC = 2.77), Igfbp7 (FC = 2.07), Rpb1 (FC = 1.73), Pygp (FC = 1.46), Niban 1 (FC = 0.58), Mest (FC = 0.48), and Aldh3a1 (FC = 0.30). The significantly altered proteins are involved in oxidative stress balance, inflammation, mitochondrial dysfunction, and regulation of vascular endothelial growth factor (VEGF) signaling.

Conclusions: The absence of IL-6Rα in KO MGCs corresponded to significant changes in their proteomic profile, highlighting the impact of autocrine IL-6 signaling on MGC function. This study provides a basis for future research evaluating distinct roles of IL-6 in MGCs and subsequent effects on retinal pathology.

Interleukin-6 (IL-6) is a multifunctional cytokine with critical roles in inflammation, wound healing, and immune system regulation.^1–3 Its diverse functions are mediated through distinct signaling pathways involving two different forms of the IL-6 receptor (IL-6R). Primarily considered anti-inflammatory, cis-signaling occurs via the membrane-bound receptor (IL-6Rα); trans-signaling functions via IL-6 binding to its soluble receptor (sIL-6R) and is pro-inflammatory.^1,2,4–6 The underlying molecular mechanisms that differentiate these pathways are not completely understood, but evidence suggests the resulting downstream response may vary depending on the localized concentrations of IL-6 and sIL-6R.^2,3 These differences are especially important in cell types that express IL-6Rα and can therefore be activated by both cis-signaling and trans-signaling. Thus, it is important to distinguish the specific effects of cis- versus trans-signaling to fully understand the functions of IL-6 within a particular type of cell or disease state. 

The majority of IL-6Rα expression within the retina is localized to Müller glial cells (MGCs), which are highly specialized macroglia crucial for regulation of the retinal microenvironment, neuroprotection, and the integrity of the blood–retinal barrier.^2,5,7–12 In diabetic retinopathy (DR), studies have implicated IL-6 as both a protective and harmful factor.^5,13–16 We and others have hypothesized that these differences could be due to the differing IL-6 signaling modalities, and our recent studies have shown that selective inhibition of IL-6 trans-signaling can protect against inflammation, diabetic oxidative damage, and endothelial barrier dysfunction.^4,17–21 Because MGCs express IL-6Rα, distinguishing the specific functions of IL-6 signaling modalities in the retina presents a unique challenge, as the elevated IL-6 and sIL-6R levels seen in pathologic states such as DR can potentially activate both pathways.^5,19,22–26 Our lab has recently developed and characterized a novel MGC-specific Il6ra^–/– knockout (KO) mouse to differentiate these pathways in vivo.²⁷ The cytokine-binding region of the Il6ra gene is deleted specifically in MGCs, meaning these cells are no longer capable of cis-signaling. However, because the Il6ra gene remains intact in other tissues, the mouse is still able to generate a functional IL-6Rα in other cell types to allow for both cis-signaling to occur outside the retina as well as sIL-6R to be produced systemically for activation of trans-signaling. We have shown that these mice retain normal retinal morphology and function in vivo,^27,28 and further studies in these KO mice will significantly enhance our current understanding of the role of IL-6 signaling in retinal diseases. 

Isolated murine MGCs are routinely used for mechanistic and molecular retinal research. Previous studies have suggested that MGCs can secrete IL-6 as well as shed membrane-bound IL-6Rα to form sIL-6R, which indicates that wild-type (WT) cells likely maintain a level of autocrine IL-6 signaling through both cis- and trans-signaling mechanisms.² In this study, we conducted proteomic profiling of MGCs isolated from MGC-specific Il6ra^–/– and WT mice to identify differentially expressed proteins and associated pathways altered by the loss of functional IL-6 receptor in vitro. 

Methods

Animals

Male and female WT C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA), and MGC-specific Il6ra^–/– mice were generated as previously described.²⁷ Briefly, male homozygous floxed Il6ra (B6;SJL-Il6ra^tm1.1Drew/J) mice were crossed with female Pdgfra-Cre mice (C57BL/6-Tg[Pdgfra-cre]1Clc/J), and female Cre^± offspring were backcrossed to the paternal strain. Subsequent generations of MGC-specific Il6ra^–/– KO mice were maintained in a hemizygous state for Cre. This study was conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and approved by the Institutional Animal Care and Use Committee at Augusta University (protocol no. 2014-0676). All efforts were made to ensure the safe and humane treatment of animals used in the study to minimize any possible suffering during experimental procedures, animal handling, and euthanasia. 

Isolation and Culture Of Murine MGCs

Primary MGCs from individual WT and KO mice were isolated and cultured using previously published protocols,^29,30 yielding primary cells in active division and proliferation that retain their phenotype through passage 5. Cells were isolated from mice at the age of postnatal day 7 (P7), as increasing gliogenic factors during the period from P4 to P12 show committal differentiation of a proportion of glial precursors into MGCs.³¹ Briefly, enucleated eyes were incubated overnight in 5 mL Gibco Dulbecco's Modified Eagle Medium (DMEM; 1-g/L glucose, 25-mM HEPES, 1-mM sodium pyruvate, 4-mM l-glutamine; Thermo Fisher Scientific, Waltham, MA, USA) with 0.1% penicillin/streptomycin (Cytiva; Thermo Fisher Scientific), washed with warm Dulbecco's Phosphate-Buffered Saline (DPBS), and then digested for 30 to 60 minutes with 0.25% trypsin/EDTA (Corning, Corning, NY, USA) and 2000 units/mL collagenase IV (LS004186; Worthington Biochemicals, Lakewood, NJ, USA) in a humidified incubator at 37°C and 5% CO₂. Retinas were isolated under a dissecting microscope and transferred to 5-cm cell culture dishes containing complete DMEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Cultures were incubated at 37°C with 5% CO₂ for 7 to 10 days. When they became confluent, the MGCs were further passaged two to three times to remove any retinal pigment epithelial cells or other cellular contaminants. Cell purity was assessed after the third passage by verifying positive immunofluorescence staining for MGC markers, including vimentin, glutamine synthetase, and glutamate/aspartate transporter protein. MGCs isolated by these methods have minimal staining for glial fibrillary acidic protein (GFAP), with no significant difference observed between WT and KO cultures.²⁷ As a quantitative measure of purity, flow cytometry analysis was performed to confirm >95% positive staining for anti-CRALBP (sc-376082, AF647; Santa Cruz Biotechnology, Dallas, TX, USA) and anti-NGFR p75 (sc-271708, AF488; Santa Cruz Biotechnology) (Supplementary Fig. S1). 

Sample Preparation for Liquid Chromatography–Tandem Mass Spectrometry Analysis

WT and KO MGCs were seeded and cultured in complete DMEM overnight, then incubated with serum-free DMEM for 18 hours. MGCs were harvested as cell pellets (∼160,000 cells per well) and stored at –80°C. Cells were lysed in radioimmunoprecipitation assay (RIPA) buffer (Cell Biolabs, San Diego, CA, USA) containing protease and phosphatase inhibitors and centrifuged at 12,000g for 15 minutes at 4°C. The protein concentration of cell lysates was determined by Bradford assay, and 50 µg of total protein per sample was used for analysis by mass spectrometry. Lysates were lyophilized, reconstituted in 8-M urea in 50-mM Tris-HCl (pH 8). Then, 10-mM dithiothreitol (DTT) and 55-mM iodoacetamide were added to the samples to reduce and alkylate cysteine residues, respectively. The urea concentration was further reduced to <1 M by the addition of 50-mM ammonium bicarbonate. For protein digestion, mass spectrometry–grade trypsin (90057; Thermo Fisher Scientific) was added at a ratio of 1:20 (w/w) trypsin to protein, and the samples were incubated overnight at 37°C. Digested samples were subsequently cleaned using C18 spin columns (744101; Harvard Apparatus, Holliston, MA, USA), lyophilized, and then analyzed on the Orbitrap Fusion Tribrid Mass Spectrometer coupled to an UltiMate 3000 RSLC nano UHPLC System (Thermo Fisher Scientific). 

Liquid Chromatography–Tandem Mass Spectrometry

For liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis, peptides reconstituted in 2% acetonitrile in water with 0.1% formic acid were loaded onto an Acclaim PepMap 100 C18 trap column (5 µm, 0.3 × 5mm; Thermo Fisher Scientific) at 20 µL/min for 10 minutes and then separated on an Acclaim PepMap 100 C18 RSLC column (2.0 µm, 75 µm × 150 mm; Thermo Fisher Scientific) using a gradient of 2% to 40% acetonitrile with 0.1% formic acid over 120 minutes at a flow rate of 300 nL/min and a column temperature of 40°C. The eluted peptides were introduced into the Orbitrap Fusion MS via a nano-electrospray ionization source at 300°C and spray voltage of 2000 V. The peptides were then analyzed by data-dependent acquisition in positive mode using the Orbitrap MS analyzer for precursor scans at 120,000 full width at half maximum (FWHM) from 400 to 2000 m/z and using the ion-trap MS analyzer for MS/MS scans at its top speed mode (3-second cycle time) with dynamic exclusion settings (repeat count 1 and exclusion duration 15 seconds). For LC-MS/MS analysis, higher energy collisional dissociation (HCD) fragmentation was used with a normalized collision energy of 30%. 

Protein Identification and Quantification

Raw MS data were processed via Proteome Discoverer 1.4 and submitted for the Sequest HT algorithm against the SWISS-PROT mouse protein database. Sequest HT search parameters were set as 10 ppm precursor and 0.6 Da product ion mass tolerance, with static carbamidomethylation (+57.021 Da) for cysteine and dynamic oxidation for methionine (+15.995 Da). The percolator peptide spectrum matching (PSM) validator algorithm was used for PSM validation. Proteins that could not be distinguished based on the database search results were grouped to satisfy the principles of parsimony. A protein report was generated containing the identities and number of PSM for each group, which were further used for spectral counting–based semiquantitative analysis. 

Targeted PRM Analysis to Validate the Top 20 Altered Proteins

Parallel reaction monitoring (PRM) analysis was conducted to validate the top 20 significantly altered proteins between KO and WT MGCs.^32,33 The digested samples were spiked with a heavy isotope–labeled peptide (NIQSLEVIGK) to serve as an internal standard and normalize the endogenous peptide levels across samples. Separation, elution, and ionization source parameters remained the same. The eluted peptides were analyzed by targeted acquisition in positive mode using quadrupole to isolate ions (isolation window = 1.2 m/z) and the Orbitrap MS analyzer for fragment scans at 30,000 FWHM from 200 to 1600 m/z. HCD fragmentation (27% collision energy) was used for PRM analysis. Peak areas of the peptide ion fragments were extracted and integrated across the elution profile using Skyline software for PRM-based targeted MS quantification. 

Gene Ontology and Ingenuity Pathway Analysis of Differentially Expressed Proteins

Gene Ontology (GO) enrichment analysis of differentially expressed proteins was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID 6.8) to determine the biological processes, cellular components, and molecular functions. Ingenuity pathway analysis (IPA) was performed to discern the biological mechanisms and pathways regulating the significantly altered proteins. 

Measurement of Secreted IL-6 Levels in MGCs

Purified WT and KO MGCs were seeded at 80,000 cells/well onto 12-well culture plates, grown in DMEM with 10% FBS and 1% penicillin/streptomycin, and allowed to adhere overnight. The culture media from each well were collected, and levels of endogenous secreted IL-6 were measured by ELISA kit (RK00008; ABclonal, Woburn, MA, USA) following the manufacturer's protocol. 

Western Blotting Analyses

WT and KO MGCs were treated with exogenous IL-6 (50 ng/mL; PeproTech, Rocky Hill, NJ, USA), harvested, and lysed as previously described. Then, 30 µg of protein per sample was separated on 4% to 15% SDS-PAGE gels by electrophoresis (Bio-Rad, Hercules, CA, USA). Proteins were transferred to nitrocellulose membranes and blocked using 5% milk prior to overnight incubation with vascular endothelial growth factor A (VEGFA) Rabbit pAb antibody (A0280, 1:1000; ABclonal). Membranes were washed with Tris-buffered saline with Tween 20 (TBST), incubated for 2 hours with Goat anti-Rabbit IgG (H+L) Secondary Antibody, HRP (31460; Thermo Fisher Scientific), and developed with enhanced chemiluminescence substrate (Thermo Fisher Scientific). Membranes were also probed for β-actin (AC028, 1:5000; ABclonal). Images were taken on a ChemiDoc Imaging System (Bio-Rad), and band intensity was quantified using ImageJ (National Institutes of Health, Bethesda, MD, USA). VEGFA levels were normalized to β-actin, and values were reported as fold change relative to expression in untreated WT MGCs. 

Assessment of Oxidative Stress Response Using MitoSOX Staining

Mitochondrial superoxide generation was examined by live-cell imaging of MGCs stained with MitoSOX Red (Molecular Probes, Eugene, OR, USA). Briefly, WT and KO MGCs were seeded at 10,000 cells/well onto 96-well plates and allowed to adhere overnight. The cells were serum-starved for 4 hours in serum-free DMEM prior to exposure to H₂O₂ (100 µM, 18 hours) to induce oxidative stress. Cells were stained with 5-µM MitoSOX Red dye according to the manufacturer's protocol. Images were acquired on a Leica STELLARIS confocal microscope (Leica Microsystems, Wetzlar, Germany) at 510/580-nm excitation/emission, and fluorescence intensity was quantified using ImageJ. 

Seahorse Mitochondrial Stress Assay

Mitochondrial function was assessed in WT and KO MGCs using a Seahorse XF Cell Mito Stress Test Kit (Agilent Technologies, Santa Clara, CA). Cells were seeded at 10,000 cells/well on a 24-well Seahorse cell culture microplate until confluency and then incubated in serum-free media for 18 hours with or without exogenous IL-6 (50 ng/mL). Cells were washed with Seahorse XF DMEM (4-mM l-glutamine, 1-mM pyruvate, and 5.5-mM glucose) and placed for 30 minutes at 37°C in a non-CO₂ incubator. The Seahorse XFe24 Analyzer measured the oxygen consumption rate (OCR) every 8 minutes after obtaining baseline readings, with kit reagents (1.5-µM oligomycin, 2.0-µM FCCP, and 0.5-µM Rot/AA) injected as appropriate for the Seahorse Mito Stress Test kit. Spare respiratory capacity, maximal respiration, and adenosine triphosphate (ATP) production were calculated using the Seahorse XF Cell Mito Stress Test report generator (Wave 4.0). 

Statistical Analysis

Statistical analyses were conducted using R 4.1.1 (R Foundation for Statistical Computing, Vienna, Austria). PSM count data from LC-MS/MS were log₂ transformed, and the batch effect was removed using the ComBat function in the SVA package in R. Differential expression analyses were conducted using the limma package,³⁴ and P values were adjusted using the false discovery rate (FDR) method. Differential expression analysis results were further visualized with volcano plots, and the expression profile of significantly altered proteins was visualized by heat map and principal component analysis (PCA) using the gplots package in R. All data for molecular studies are presented as mean ± standard error of the mean (SEM), and the statistical analyses were performed by two-way ANOVA with Tukey's multiple comparisons test using Prism (GraphPad, Boston, MA, USA). 

Results

Proteomic Profiling of Murine MGCs In Vitro

A total of 1866 unique proteins were identified through LC-MS/MS analysis of MGCs isolated from both WT and MGC-specific Il6ra^–/– mice. The top 50 proteins, ranked by averaging PSM counts across each group, are shown in Table 1. Notably, vimentin (Vim, average PSM = 807.80), actin cytoplasmic 1 (Actb, average PSM = 758.41), myosin-9 (Myh9, average PSM = 424.98), filamin-A (Flna, average PSM = 409.21), and actin alpha skeletal muscle (Acta1, average PSM = 334.81) emerged as the most abundant proteins across both experimental groups. Among the top 50 proteins identified in the MGC proteome, 10 proteins were members of the tubulin family, three belonged to the heat shock protein family, and an additional three were classified under the filamin protein family. 

Table 1.

 

Top 50 Proteins With the Most Abundance in WT and KO MGCs

 

Loss of Membrane-Bound IL-6 Receptor in MGCs Leads to Significant Proteomic Alterations

Our analysis identified 81 proteins that exhibited significant differential expression in knockout MGCs compared to wildtype (Fig. 1A). Principal component analysis (PCA) also revealed distinct clustering patterns between wildtype and KO MGCs (Fig. 1B). Additionally, unsupervised clustering highlighted two clusters delineating distinct expression patterns between both groups (Fig. 1C). The relative abundance of all 81 significantly altered proteins in MGCs following Il6ra knockout are displayed in Figure 2, and the top 20 proteins exhibiting the largest degree of fold change are listed in Table 2. Notably, the top five upregulated proteins included mammalian ependymin-related protein 1 (Epdr1, FC = 7.39), retinol-binding protein 1 (Rbp1, FC = 4.12), dimethylaniline monooxygenase [N-oxide-forming] 1 (Fmo1, FC = 3.22), prostacyclin synthase (Ptgis, FC = 2.85), and protein lin-7 homolog C (Lin7c, FC = 2.80) (Fig. 2A). Conversely, the top five downregulated proteins included aldehyde dehydrogenase dimeric NADP-preferring (Aldh3a1, FC = 0.24), apolipoprotein D (Apod, FC = 0.28), mesoderm-specific transcript protein (Mest, FC = 0.29), tenascin (Tnc, FC = 0.29), and CD63 antigen (Cd63, FC = 0.35) (Fig. 2B). Raw data and results of the differential expression analysis for all 1866 proteins are provided in Supplementary Table S1. 

Figure 1.

 

Proteomic alterations in MGCs after loss of IL-6Rα. (A) Volcano plot comparing protein expression in MGC-specific Il6ra^–/– KO MGCs relative to WT, highlighting 41 upregulated proteins (red) and 40 downregulated proteins (blue). (B) PCA demonstrated distinct separation between MGCs from WT (blue) and KO (red) mice. Individual points represent separate biological replicates for each group. (C) Heatmap of differentially expressed proteins between the two groups using unsupervised clustering analysis.

Figure 2.

 

Differentially expressed proteins in MGCs from MGC-specific Il6ra^–/– mice. LC-MS/MS analysis identified 81 proteins significantly altered after loss of IL-6Rα. Bar plots compare average PSM counts of significantly upregulated (A) and downregulated (B) proteins in KO MGCs compared to WT. Data are represented as mean PSM ± SEM (n = 7 or 8/group). *FDR-adjusted P < 0.05 versus WT.

Table 2.

 

Top 20 Upregulated and 20 Downregulated Proteins in MGCs After Loss Of IL-6 Receptor

 

Targeted PRM Confirmation of the Top Differentially Expressed Proteins

The proteins exhibiting the greatest fold change (10 upregulated and 10 downregulated) were selected for targeted PRM analysis to validate findings from initial LC-MS/MS analyses. Among the 10 upregulated proteins, six were confirmed through PRM analysis: Rbp1 (FC = 1.73, P = 0.001), Ptgis (FC = 3.63, P = 0.038), dipeptidase 1 (Dpep1; FC = 2.79, P = 0.010), glycogen phosphorylase brain form (Pygb; FC = 1.46, P = 0.049), dimethylaniline monooxygenase [Noxide-forming] 1 (Fmo1; FC = 2.77, P = 0.017), and insulin-like growth factor-binding protein 7 (Igfbp7; FC = 2.07, P = 0.035) (Fig. 3A). Beta-glucuronidase (Gusb; FC = 1.49, P = 0.155) exhibited an upward trend in KO MGCs but did not achieve statistical significance by PRM. Epdr1, Lin7c, and Osmr were below the detection limit in the targeted PRM panel. Of the top 10 downregulated proteins selected for confirmation, three were validated by PRM: Aldh3a1 (FC = 0.30, P = 0.001), Mest (FC = 0.48, P = 0.048), and Niban 1 (FC = 0.58, P = 0.001) (Fig. 3B). Four proteins, including ERO1-like protein alpha (Ero1a; FC = 0.68, P = 0.333), lysosome-associated membrane glycoprotein 1 (Lamp1; FC = 0.69, P = 0.341), Apod (FC = 0.60, P = 0.125), and Tnc (FC = 0.60, P = 0.323), exhibited decreasing trends in KO MGCs but did not reach statistical significance. Cd63, charged multivesicular body protein 5 (Chmp5), and DnaJ homolog subfamily C member 3 (Dnajc3) were below the detection limit in the PRM analysis. The peptide sequences and PRM data from the detected proteins are presented in Supplementary Tables S2 and S3, respectively. 

Figure 3.

 

PRM validation of the top upregulated and downregulated proteins in KO MGCs. (A) Six of the top 10 upregulated proteins were confirmed by PRM analysis: Rbp1, Ptgis, Dpep1, Pygb, Fmo1, and Igfbp7. (B) Three of the top 10 downregulated proteins were also validated by targeted PRM analysis: Aldh3a1, Mest, and Niban 1. Data are represented as mean ± SEM (n = 7 or 8/group). *P < 0.05, **P < 0.01, ***P < 0.001 versus WT.

Figure 3.

 

Continued.

GO Analysis of Significantly Altered Proteins in MGCs After Loss of IL-6 Receptor

GO analysis was performed using DAVID 6.8 to identify the biological processes, cellular components, and molecular functions impacted by IL-6 receptor KO in MGCs. The 81 differentially expressed proteins were analyzed to reveal enrichment across multiple categories (Fig. 4). In terms of biological processes, the altered proteins were prominently involved in metabolic processes (56 proteins), localization (35 proteins), anatomical structure development (28 proteins), cell communication (27 proteins), and biosynthetic processes (25 proteins) (Fig. 4A). Cellular component analysis highlighted enrichment in cytoplasm (68 proteins), membrane (45 proteins), endomembrane system (34 proteins), protein-containing complexes (27 proteins), and cell periphery (26 proteins) (Fig. 4B). Molecular function assessment indicated roles in protein binding (50 proteins), catalytic activity (37 proteins), ion binding (26 proteins), protein-containing complex binding and hydrolase activity (17 proteins), and nucleotide binding (16 proteins) (Fig. 4C). These findings highlight the diverse functional impacts of IL-6 receptor loss on MGC proteomics, implicating changes across fundamental biological processes and molecular interactions. 

Figure 4.

 

Characterization of MGC proteins altered after loss of IL-6Rα. GO analysis identified (A) biological processes, (B) cellular components, and (C) molecular functions associated with the 81 proteins that were significantly altered (FDR-adjusted P < 0.05) in MGC-specific Il6ra^–/– Müller glial cells relative to WT.

IPA of Differentially Expressed Proteins in MGC-Specific Il6ra^–/– MGCs

Significant proteins identified in our differential expression analysis were further examined using IPA to reveal major interaction networks, pathways of predicted activation or inhibition, and IL-6 signaling–related regulatory hubs (Fig. 5). For upregulated proteins, seven major hubs were identified, including mitogen-activated protein kinase 1 (Mapk1, 10 connections), RAC-beta serine/threonine-protein kinase (Akt2, eight connections), beta-glucuronidase (Gusb, six connections), lamina-associated polypeptide 2 isoforms beta/delta/epsilon/gamma (Tmpo, six connections), fascin (Fscn1, six connections), tropomodulin-3 (Tmod3, five connections), and thrombospondin-1 (Thbs1, five connections). Similarly, among the downregulated proteins the major hubs within the network were endoplasmic reticulum chaperone BiP (Hspa5, 13 connections), Lamp1 (nine connections), voltage-dependent anion-selective channel protein 1 (Vdac1, eight connections), proteasome subunit alpha type-3 (Psma3, eight connections), coronin-1C (Coro1c, seven connections), ruvB-like 1 (Ruvbl1, six connections), Dnajc3 (six connections), ras-related protein Rab-7a (Rab7a, five connections), and brain acid soluble protein 1 (Basp1, five connections). Furthermore, the proteins altered following the loss of IL-6Rα were associated with various pathways, including activation of VEGF signaling, mitochondrial depolarization, carbohydrate metabolism, and inhibition of endothelial cell binding. 

Figure 5.

 

IPA of proteins altered in MGCs from MGC-specific Il6ra^–/– mice. IPA software was utilized to identify connections among proteins altered due to the loss of IL-6Rα. Red nodes represent upregulated proteins, and green nodes indicate downregulated proteins.

Effect of IL-6Rα Loss on IL-6 Secretion, VEGF Levels, Oxidative Stress, and Mitochondrial Function in MGCs

The conditioned media from WT and KO MGCs were collected to quantify the total IL-6 release after the loss of IL-6Rα (Fig. 6A). The secreted IL-6 in the WT MGC media was found to be 95.68 pg/mL, indicating that these cells are capable of baseline autocrine IL-6 signaling. In contrast, IL-6 levels in the media of KO MGCs were significantly reduced, by ∼12-fold (8.11 pg/mL, P < 0.0001) compared to WT. 

Figure 6.

 

Loss of IL-6 receptor alters IL-6 secretion, VEGF levels, and mitochondrial function in MGCs. (A) IL-6 levels (pg/mL) in culture media were significantly decreased in KO MGCs as compared to WT. (B) Baseline VEGFA levels were comparable between WT and KO cells; however, IL-6 treatment significantly increased VEGFA protein levels in WT MGCs with no significant response observed in KO cells. (C) Mitochondrial superoxide production measured using MitoSOX Red staining showed no significant baseline differences between untreated WT and KO MGCs. However, H₂O₂ exposure (100 µM) significantly increased superoxide production in both WT MGCs (1.28-fold) and KO MGCs (1.54-fold), with KO MGCs exhibiting significantly higher superoxide levels than WT MGCs (n = 11/group). (D) OCR (pmol/min) was measured in WT and KO MGCs with and without IL-6 treatment by using the Seahorse Mito Stress test. Both spare and maximal respiratory capacity were significantly reduced in KO MGCs as compared to WT, whereas the ATP levels were unchanged. Data are represented as mean ± SEM (n = 5–11/group). *P < 0.05, ***P < 0.001, ****P < 0.0001; ns, not significant.

Given the role of MGCs as major VEGF producers in the retina, VEGFA protein levels were assessed in WT and KO MGC lysates both at baseline and after IL-6 treatment (50 ng/mL). Baseline VEGFA levels were comparable between WT and KO MGCs; however, IL-6 treatment significantly increased VEGFA levels in WT MGCs (1.90-fold, P = 0.021), with no significant response observed in KO MGCs (Fig. 6B). 

Several differentially expressed proteins were associated with oxidative stress and mitochondrial function. Mitochondrial superoxide production, assessed using MitoSOX Red staining (Fig. 6C), showed no significant baseline differences between untreated WT and KO MGCs (fold change = 1.14, P = 0.375). However, H₂O₂ exposure (100 µM) significantly increased superoxide production in both WT MGCs (1.28-fold, P = 0.014) and KO MGCs (1.54-fold, P = 0.0003), with KO MGCs exhibiting significantly higher superoxide levels than WT (P = 0.025). 

A Seahorse Mito Stress assay revealed impaired mitochondrial function in KO MGCs, with significantly reduced spare respiratory capacity (63.37 vs. 204.71 pmol/min, P < 0.0001) and maximal respiration (121.34 vs. 271.42 pmol/min, P < 0.0001) compared to WT MGCs (Fig. 6D). ATP production, however, was comparable between WT and KO MGCs (42.50 vs. 51.78 pmol/min, P = 0.140), suggesting potential compensatory activation of alternative metabolic pathways. 

Discussion

Müller glial cells provide crucial metabolic support for maintaining retinal homeostasis.^35,36 MGCs naturally express membrane-bound IL-6 receptor (IL-6Rα), enabling them to respond to both IL-6 cis- and trans-signaling.^2,5 Our recently generated MGC-specific Il6ra^–/– mouse eliminates IL-6 cis-signaling in MGCs to allow for the delineation of specific effects of the trans-signaling pathway on MGC function.²⁷ Proteomic characterization of MGC-specific Il6ra^–/– cells provides critical insights into how IL-6 trans-signaling impacts the function of Müller glia. Using LC-MS/MS, we identified 1866 unique proteins in both WT and KO mouse MGCs, revealing significant proteomic alterations following the loss of IL-6 receptor. 

Our findings in this study confirm previous reports that proteins essential for retinal integrity and oxidative defense are highly abundant in MGCs.^37,38 Specifically, we observed high levels of cytoskeletal proteins, such as filamins, actins, and tubulins, along with Vim, Actb, annexin A2 (Anxa2), glyceraldehyde-3-phosphate dehydrogenase (Gapdh), and alpha enolase (Eno1), in both WT and KO MGCs. Additionally, the presence of heat shock proteins (Hspa5, Hsp90b1, Hspa8, Hsp90ab1, and Hsp90aa1) highlights their role in supporting retinal neurons and regulating cellular stress. 

Among the proteins significantly upregulated in knockout MGCs, Ptgis exhibited the greatest fold change (3.63-fold) in validation PRM studies. Ptgis is linked to neovascularization and endothelial function modulation^39,40 and is integral to the synthesis of prostacyclin (Pgi2), a modulator of inflammation and vascular tone. The upregulation of Ptgis may be a compensatory response to the loss of anti-inflammatory properties linked to IL-6 cis-signaling. Studies have shown that upregulated Ptgis inhibits the macrophage switch to the M1 phenotype, suggesting a mechanism to stabilize the retinal microenvironment in response to increased inflammation. Dpep1, validated by PRM analysis, showed the second highest fold change (2.79-fold). The role of Dpep1 in metabolizing β-lactam antibiotics and glutathione,⁴¹ as well as its association with neutrophil adhesion factors and inflammatory responses,^42,43 suggests a potential involvement in ocular inflammation. However, further research is necessary to clarify its specific role in the retina. Fmo1 was also significantly upregulated in knockout MGCs. FMOs, known for their role in detoxifying xenobiotics and metabolites,^44,45 enhance resistance to oxidative stress⁴⁶ and convert hypotaurine to taurine, a potent antioxidant.^47–50 

Pygb, Igfbp7, and Rbp1 were other proteins significantly upregulated in MGCs lacking IL-6Rα. Pygb inhibits reactive oxygen species production and provides an alternative energy source during metabolic stress, which may be crucial for maintaining retinal function when IL-6 signaling is disrupted.^51–54 Igfbp7 is a known biomarker of oxidative cell stress⁵⁵ and is shown to be upregulated in the retina of oxygen-induced retinopathy models.^56,57 Moreover, Igfbp7 is known to regulate retinal development and is implicated in cellular senescence and apoptosis.^58,59 Its increased expression in KO MGCs possibly suggests an adaptive response to maintain retinal integrity under stress conditions. Rbp1 has an established function in modulating retinoid availability in cells.^60,61 Retinoids exhibit pro-apoptotic and antioxidant functions and are commonly used in the treatment of inflammatory diseases.⁶² Additionally, previous studies have also reported Rbp1 to be upregulated in aging⁶³ and Parkinson's disease.⁶⁴ 

The most significantly downregulated protein in knockout MGCs was Aldh3a1, and reduced levels of Aldh3a1 have been associated with increased oxidative damage and decreased cell proliferation.⁶⁵ This decrease highlights the potential vulnerability of MGCs to oxidative stress without functional IL-6Rα. Mest (0.48-fold) and Niban 1 (0.58-fold) are other proteins with downregulated expression in knockout MGCs. Though primarily associated with adipocyte differentiation, Mest has also been linked with neuronal migration, and Mest-specific knockdown in murine models disrupted the transition of bipolar neurons during embryonic development.^66,67 Niban 1 has been associated with modulating cell death signaling during cellular stress responses.⁶⁸ 

This study suggests a potential role of autocrine IL-6 in regulating VEGF signaling in murine MGCs. Mitochondrial function studies revealed that the loss of functional IL-6Rα impairs the baseline spare and maximal respiratory capacity of MGCs. Further, our mechanistic studies demonstrate that mitochondrial superoxide generation is enhanced in KO MGCs after inducing oxidative stress, whereas the IL-6–mediated production of VEGF is not observed in the absence of autocrine IL-6 signaling. These results suggest that an equilibrium of IL-6 cis- and trans-signaling (the “IL-6 cis-trans balance”) plays a critical role in maintaining MGC metabolic and mitochondrial integrity, as well as VEGF production. Several downstream targets of IL-6 signaling identified in this study warrant further investigation to elucidate their specific roles in mediating these effects. 

In summary, this study demonstrates the critical role of IL-6Rα–mediated autocrine signaling in maintaining the function and homeostasis of MGCs. The absence of IL-6Rα in MGCs disrupts critical pathways related to oxidative stress response, inflammation, metabolism, and mitochondrial function. Proteomic analysis of Il6ra^–/– MGCs identified potential candidates for further investigation of IL-6 signaling in MGC function and uncovering novel therapeutic targets for retinal diseases. 

Acknowledgments

The authors thank Garrett N. Jones for performing flow cytometry analyses for this study. 

Supported by grants from the National Eye Institute, National Institutes of Health (P30-EY031631 to Augusta University; R01-EY035995 to SS). 

Disclosure: J. Glass, None; R. Robinson, None; N. Edupuganti, None; J. Altman, None; G. Greenway, None; T.J. Lee, None; W. Zhi, None; A. Sharma, None; S. Sharma, None