Lipoxin B₄ Mitigates TRPV4-Activated Müller Cell Gliosis During Ocular Hypertension

Abstract

Purpose

Müller glia play dual roles in glaucoma, contributing to retinal homeostasis and neuroinflammation; activation by elevated intraocular pressure through mechanosensitive transient receptor potential vanilloid 4 (TRPV4) promotes a reactive state that drives retinal ganglion cell loss. Lipoxin B₄ (LXB₄), an endogenous lipid mediator produced by retinal astrocytes, suppresses glial reactivity and protects retinal ganglion cells. This study investigated whether LXB₄ modulates TRPV4-driven Müller glial activation and whether Müller glia contribute to the lipoxin pathway.

Methods

Ocular hypertension was induced in mice via a silicone oil model, and reactive Müller glia were isolated by magnetic sorting for transcriptomics. In vitro, Müller glia cultures were treated with a TRPV4 agonist with or without LXB₄. Glial reactivity was assessed by flow cytometry, immunostaining, qPCR, and Western blotting. Lipidomics quantified pathway metabolites, and single-cell RNA sequencing examined transcriptional responses to LXB₄.

Results

Bulk RNA sequencing and qPCR revealed Müller glia express 5- and 15-lipoxygenase. Lipidomics confirmed a functional pathway, with endogenous LXB₄ production, identifying Müller glia as a source of neuroprotective LXB₄. TRPV4 activation induced gliosis with increased glial fibrillary acidic protein, IL-6, and signal transducers and activators of transcription 3 (STAT3) expression, and lipoxin production, indicating biomechanical stress triggers reactivity and protective signaling. LXB₄ suppressed TRPV4-induced gliosis in vitro by downregulating IL-6 and STAT3 and in vivo by reducing Stat3, Il6, and Tnf-α and attenuating TRPV4 upregulation during ocular hypertension.

Conclusions

Müller glia are a significant source of LXB₄ in the retina. This neuroprotective Müller glia pathway is amplified during chronic TRPV4 activation to counter-regulate gliosis. These findings support the targeting of the TRPV4–lipoxin pathway as a potential approach to protect against ocular hypertension–induced neurodegeneration in glaucoma.

Glaucoma, a group of optic neuropathies, is characterized by the progressive degeneration of retinal ganglion cells (RGCs) and their axons. Elevated IOP, a major risk factor, disrupts retinal homeostasis and activates inflammatory pathways that accelerate RGC loss.^1,2 Although IOP-lowering therapies slow disease progression, they do not fully prevent RGC degeneration, underscoring the need for neuroprotective strategies that target underlying mechanisms. Emerging evidence suggests that glia–neuron interactions, including those involving Müller glia, play a critical role in glaucomatous neurodegeneration.³ However, the molecular mechanisms underlying the contributions of glia to RGC loss remain unclear. Elucidating these pathways is essential for developing therapies that preserve RGCs and slow disease progression.

Müller glia are the principal macroglia cells of the retina and extend from the inner limiting membrane to the outer limiting membrane.^3,4 This unique architecture enables them to detect and respond to mechanical, metabolic, and inflammatory stress across all retinal layers.⁵ Under physiological conditions, Müller glia regulate osmotic balance, support neuronal metabolism, and maintain ion and water homeostasis.^3,6 After retinal injury, these cells become reactive, characterized by glial fibrillary acidic protein (GFAP) upregulation.³ Chronic gliosis can disrupt tissue repair, promote inflammation, and accelerate RGC degeneration.^5,7

A key pathway driving Müller glial reactivity involves the nonselective, mechanosensitive cation channel transient receptor potential vanilloid 4 (TRPV4).^8,9 TRPV4 is expressed in both RGCs and Müller glia and responds to mechanical insults, such as those caused by elevated IOP, by facilitating calcium influx and triggering downstream inflammatory signaling.^10,11 Sustained TRPV4 activation in Müller glia induces a reactive phenotype characterized by proinflammatory cytokine release and upregulated GFAP expression.^8,9,12 This process involves Janus kinases (JAK) and signal transducers and activators of transcription (STAT) signaling, including IL-6/gp130 activation, and direct STAT3 binding to the GFAP promoter, ultimately contributing to RGC apoptosis in glaucoma.^13–15 Pharmacological activation of TRPV4 in Müller glia recapitulates key features of IOP‐induced insults, making it a robust in vitro model for glaucomatous neuroinflammation.^8,9,12

Lipoxins (lipoxin A₄ [LXA₄] and lipoxin B₄ [LXB₄]) are endogenous eicosanoids whose canonical role is to counter-regulate chronic inflammation. They are synthesized through cellular interactions between selective cell types expressing 5-lipoxygenase (5-LOX) and 15-LOX.^16–18 The anti-inflammatory action are well-established for LXA₄ and LXA₄ stable analogs and include inhibiting leukocyte recruitment, JAK/STAT signaling, and nuclear factor-κB–driven cytokine production.^19–21 LXB₄’s mechanisms of action is distinct from LXA₄, not mediated by the LXA₄ receptor, and remains to be fully defined. We recently identified LXB₄ as an endogenous lipid mediator in the healthy retina and as a promising neuroprotective target to prevent RGC death in glaucoma models.^22,23 We have established that LXB₄ treatment in rat and mouse models of glaucomatous injury is neuroprotective in the retina, stopping death and preserving RGC function and astrocyte homeostatic function, and in the optic nerve maintaining a microglia functional homeostatic phenotype.^23–25 In the retina, astrocytes generate LXB₄ as a part of their homeostatic function.²³ Our recent findings also revealed that treatment with LXB₄ reduces glial reactivity in both astrocytes and Müller glia after induced ocular hypertension (OHT).²⁴ However, whether Müller glia can endogenously produce LXB₄ or if LXB₄ modulates TRPV4-induced reactivity remains unclear. Given Müller glia's shared homeostatic roles with astrocytes and their abundance (accounting for approximately 90% of all retinal macroglia), it is plausible that Müller glia serve as a major source of retinal LXB₄ and/or their metabolic precursors.²⁶

In this study, we investigated whether Müller glia express LOXes required for LXB₄ biosynthesis (5-LOX and 15-LOX) and whether they generate LXB₄. We also examined whether LXB₄ treatment counter regulates TRPV4-induced Müller glial reactivity in both in vitro (TRPV4 agonist–treated Müller glia) and in vivo (mouse OHT) models. This study defines a novel glia–lipoxin regulatory mechanism and provides new insight into potential neuroprotective strategies for glaucoma.

Materials and Methods

Mice

All procedures were conducted according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). All animal procedures were approved by the Institutional Animal Care and Use Committee at the University of California, Berkeley. The mice were housed under a 12-hour light/12-hour dark cycle with ad libitum access to food and water.

Mouse Model of OHT

Silicone oil–induced OHT was induced in 8-week-old mice via an established protocol.^27–29 The mice were anesthetized by intraperitoneal (i.p.) ketamine (100 mg/kg) and xylazine (10 mg/kg), and topical ocular anesthesia was achieved with 0.5% proparacaine hydrochloride (Sandoz, Princeton, NJ, USA). Under a surgical microscope, pharmaceutical-grade silicone oil (Alcon, Fort Worth, TX, USA) was injected into the anterior chambers of both eyes using a 33G Hamilton glass syringe (Reno, NV, USA). Chronic mild OHT was induced by injecting 1.2 µL of silicone oil. Acute severe OHT was induced by injecting 1.8 µL of silicone oil. Sham-injected control mice received a corneal incision created with a sterile 31G paracentesis needle without silicone oil. After each procedure, 0.3% tobramycin ophthalmic solution (Tobrex; Alcon, Fort Worth, TX, USA) was applied topically. The IOP was measured via a rebound tonometer, as described previously.^23,24 OHT was sustained for up to 8 weeks, as previously validated.^28–30 In prior studies, IOP was measured once every 2 weeks using a rebound tonometer, confirming persistent IOP elevation throughout the experimental period. In the present study, repeated IOP measurements were not performed because pupil dilation required for tonometry can disrupt silicone oil–induced anterior chamber blockage and alter OHT severity. Instead, silicone oil droplet size was monitored as an established indicator of sustained OHT. Consistent with prior validation demonstrating a direct correlation between droplet size and IOP elevation, droplet size was used as an inclusion/exclusion criterion to ensure stable, nonleaking oil droplets.^28–30 Tissues from the chronic mild OHT group were collected at 2, 4, 6, and 8 weeks post injection. Eyes were enucleated and fixed in 4% paraformaldehyde (PFA) for immunohistochemistry (IHC). For the acute severe OHT group, mice were euthanized at 1 week post injection, and eyes were enucleated and fixed in 4% PFA for IHC.

Quantitative PCR

Total RNA was isolated from mouse retinas and Müller glia cultures via TRIzol reagent (Invitrogen, Waltham, MA, USA) according to the manufacturer's instructions and quantified using our previously published protocol.²⁴ cDNA was synthesized from 1 µg of RNA via iScript Reverse Transcription Supermix (Bio-Rad, Hercules, CA, USA). The transcripts for Il6, Stat3, TNF-α, Alox5, and Alox15 were quantified via GoTaq PCR master mix (Promega, Madison, WI, USA) in a OneStep Plus qPCR system (Applied Biosystems, Waltham, MA, USA). Gapdh served as the control for normalization. The mouse primers used in this study are listed in Table 1. Primers for rat genes used in this study are listed in Table 2.

Table 1.

Sequence of Mouse Forward and Reverse Primers

Primer	Sequence (5′ → 3′)
Alox5 Forward	ACA GCT TAT CTG CGA GTA TGG
Alox5 Reverse	GGG AAACAC AGG GAG GAA TAG
Alox15 Forward	TGG GTT CTC TGC CTT AGT GG
Alox15 Reverse	CAC TCA GGG TTG TCA CCT CA
Il6 Forward	CCC CAA TTT CCA ATG CTC TCC A
Il6 Reverse	CGC ACT AGG TTT GCC GAG TA
TNF- α Forward	TGA TCG GTC CCC AAA GGG AT
TNF- α Reverse	TGT CTT TGA GAT CCA TGC CGT
Stat3 Forward	ACC AACGAC CTGCAG CAA TA
Stat3 Reverse	TCC ATG TCA AAC GTG AGC GA

Table 2.

Sequence of Rat Forward and Reverse Primers

Primer	Sequence (5′ → 3′)
Alox5 Forward	AGA GTC AAG AAT CTG GTG GGC
Alox5 Reverse	GGT GAC AGT GTA GGA AGG CA
Alox15 Forward	GGG ACT CGG AAG CAG AAT TCA A
Alox15 Reverse	GCC CTG AAC CCA TCG GTA A
Il6 Forward	TCC TAC CCC AAC TTC CAA TGC TC
Il6 Reverse	TTG GAT GGT CTT GGT CCT TAG CC

Bulk RNA Sequencing (RNA-seq)

Retinal Dissociation and Müller Glia Enrichment

Three weeks after silicone oil–induced OHT, the mice were euthanized, and neural retinas (n = 4 per condition with each n consisting of 4 pooled retinas) were dissected free of the RPE. Retinal tissue was dissociated into single cells via a papain dissociation system (Worthington, Columbus, OH, USA), passed through a 40 µm cell strainer, and washed with PBS containing 0.5% BSA. Dead cells and debris were depleted with the Dead Cell Removal Kit (Miltenyi Biotech, Bergisch Gladbach, Germany). The remaining cell suspension was incubated on ice for 10 minutes with anti-GLAST (ACSA-1)-APC antibody at a 1:50 dilution (Miltenyi Biotech) protected from light. After washing and centrifugation, cell were incubated with anti-APC microbeads (Miltenyi Biotech) at 4°C for 15 minutes. After a wash, the cells were passed through an LS column (Miltenyi Biotech) in a QuadroMACS separator. Unlabeled (GLAST⁻) cells were collected as flow-through; the column was washed three times with 3 mL of MACS buffer. GLAST⁺ Müller glia were eluted with 5 mL of MACS buffer via a plunger. Müller glia purity was validated by flow cytometric quantification of GLAST⁺ cells (88% purity) after MACS isolation, and further confirmed by bulk RNA-seq analysis showing distinct PCA clustering and enrichment of canonical Müller glia marker genes (Slc1a3, Gfap, Rlbp1, ApoE, and Ccl2) relative to whole retina controls.

RNA Isolation and Library Preparation

Total RNA from enriched Müller glia was extracted via the Arcturus PicoPure RNA Isolation Kit (Thermo Fisher, Waltham, MA, USA). RNA quality was assessed via an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and samples with an RNA integrity number of 7.0 were used for library construction. cDNA was synthesized and amplified via the SMARTer v4 Ultra-Low Input RNA Kit (Clontech, Mountain View, CA, USA). cDNA was fragmented on a Diagenode Bioruptor Pico (Diagenode, Denville, NJ, USA), and sequencing libraries were prepared with the KAPA HyperPrep DNA Kit (Roche, Basel, Switzerland). Libraries were pooled and sequenced on an Illumina NovaSeq 6000 S4 flow cell (QB3 Genomics Core, UC Berkeley; RRID: SCR_022170) to generate 150-bp paired end reads.

Bioinformatic Analysis

The raw base call files were demultiplexed via Illumina bcl2fastq2 (v2.20). Adapter trimming and quality filtering were performed with Trim Galore (v0.6.6). Quality metrics were assessed with FastQC (v0.11.9). Cleaned reads were aligned to the mouse genome (mm39, UCSC) via STAR (v2.7.1a). Gene counts were obtained via featureCounts (version 1.5.3). Downstream analysis was conducted via RStudio (version 4.2.0). Transcript abundance was expressed as transcripts per million using the normalizeTPM() function, which normalizes for both gene length and sequencing depth. Differential gene expression was determined with DESeq2 (P_adj < 0.05, |Log₂FC| > 1). Gene Set Enrichment Analysis was performed with fgsea (P_adj < 0.05, |Log₂FC| > 1), and the results were visualized in ggplot2²⁴ Bulk RNA-seq datasets were deposited in GEO (GSE306785, RNAseq).

Cell Culture

Primary Müller Glia

Primary mouse Müller glia were isolated from postnatal days 8 to 10 C57BL/6J pups.^31–33 Retinas were dissociated with papain (Worthington) into a single-cell suspension. The cells were plated on six-well plates coated with 0.1% gelatin in Dulbecco's modified Eagle's medium (Gibco, Waltham, MA, USA) supplemented with 10% fetal bovine serum (Avantor Seradigm; Radnor, PA, USA) and 1% penicillin–streptomycin (Gibco). Cultures were maintained at 37°C in a humidified 5% CO₂ incubator. Müller glia identity was confirmed after two passages by immunofluorescence staining for vimentin, glutamine synthetase, and Sox9. All the assays were conducted using cells after passage 3.

Immortalized Rat Müller Glia (rMC-1) Müller Glia

The immortalized rMC cell line (rMC-1; T0576) was obtained from the UC Berkeley Cell Culture Facility. The cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 1% MEM NEAA (Gibco), 1% sodium pyruvate (Gibco), and 1% penicillin–streptomycin (Gibco). Immortalized cells were used in place of primary cells for experiments where abundant protein and RNA were needed. All experiments were performed using cells at passages 2 through 10.

In Vitro Treatments

Cells (both primary and immortalized) were treated with a specific TRPV4 agonist at a concentration of 10 µM GSK1016790A (G0798; Millipore Sigma, Burlington, MA, USA) or with an equivalent volume of vehicle (DMSO, <0.02%) in culture medium. GSK1016790A has been extensively validated as a highly selective and potent TRPV4 agonist, with its effects validated by TRPV4 antagonists or genetic deletion in prior studies.^34,35 For LXB₄ treatment, cells were incubated in 1 µM LXB₄ (Cayman Chemical, Ann Arbor, MI, USA) or the corresponding vehicle (ethanol, <0.01%). For cotreatment experiments, the cells were pretreated with LXB₄ for 20 minutes before the addition of GSK1016790A.

Immunofluorescence Staining

Enucleated eyes were fixed overnight in 4% PFA at 4°C and then dehydrated in a sucrose gradient (10%, 20%, and 30% sucrose in PBS) until the tissues sank. Eyes were embedded in optimal cutting temperature medium and frozen at −80°C for IHC. Ten-micron sections were cut on a Leica CM1900 cryostat (Leica Microsystems, Wetzlar, Germany) and mounted on Superfrost Plus slides (Thermo Fisher Scientific, Waltham, MA, USA). Immunocytochemistry (ICC) was performed on primary and immortalized Müller glia were grown on glass coverslips, washed with PBS, and fixed in 4% PFA for 15 minutes at room temperature (RT). The sections and coverslips were permeabilized and blocked for 1 hour in PBS containing 10% donkey serum and 0.25% Triton X-100 at RT. Primary antibodies against GFAP (1:1000; Abcam ab53554, Cambridge, UK), vimentin (1:500; Abcam ab8978, Cambridge, UK), Sox9 (1:200; Abcam ab185230, Cambridge, UK), Iba-1 (1:200; Cell Signaling E4O4W, Danvers, MA), and TRPV4 (1:400; Invitrogen PA5-41066) were diluted 1:1 (blocking buffer: PBS) and applied overnight at 4°C. The slides and coverslips were washed three times in PBS (5 minutes each) and then incubated for 1 hour at RT (in the dark) with Alexa Fluor 488- and 594-conjugated secondary antibodies (Invitrogen; 1:1000 dilution). Nuclei were counterstained with DAPI (1:3000), and samples were mounted with FluorSave mounting media (Sigma Aldrich, St. Louis, MO, USA). Retinal sections and cell coverslips were imaged via a Nikon Ti-Eclipse fluorescence microscope (Nikon Instruments, Melville, NY, USA) with a 20× objective. Both central and peripheral retina regions were captured for quantitative analyses.

Image Analysis

The fluorescence images were analyzed via ImageJ (v1.53c; National Institutes of Health, Bethesda, MD, USA). Regions of interest were defined based on DAPI staining. Binary masks of each region of interest were generated, and the mean fluorescence intensity was measured in the corresponding channels. Negative control sections were used to define the background signal, which was subtracted from all measurements.

Liquid Chromatography‒Tandem Mass Spectrometry (LC-MS/MS)

Primary and immortalized Müller glia were collected after 24 hours of incubation with or without the TRPV4 agonist. Müller glial cultures and conditioned media were collected in LC-MS-grade methanol for analysis and quantification of the eicosanoid and PUFA pathways via LC-MS/MS via our previously published protocol.³⁶ The supernatants containing lipids were extracted via C18 solid-phase columns and analyzed via a triple-quadrupole linear ion-trap LC-MS/MS system (AB SCIEX 4500 QTRAP) equipped with a Kinetex C18 mini-bore column. The mobile phase was a linear gradient of A (water, acetonitrile, and acetic acid [72:28:0.01 by volume]) and B (isopropanol and acetonitrile [60:40 by volume]), with a flow rate of 450 µL/min. MS/MS analysis was performed in negative ion mode. Polyunsaturated fatty acids and LOX pathway metabolites were identified and quantified via scheduled multiple reaction monitoring using four to six specific transitions for each analyte. Established diagnostic fragment ions were used for quantification: arachidonic acid (303→259 m/z), 5-hydroxyeicosatetraenoic acid (5-HETE; 319→115 m/z), 15-HETE (319→175 m/z), 12-HETE (319→179 m/z), LXA₄ (351→115 m/z), LXB₄ (351→221 m/z), and DHA (327→283 m/z) EPA (301→257). Peaks were identified based on the integration criterion of a signal-to-noise ratio of at least 5:1. Class-specific deuterated internal standards were used to validate the retention times of the analytes in each sample. Quantification was based on analyte-specific calibration curves prepared from synthetic standards (Cayman Chemical).³⁷ The results were normalized to the number of cells (1 × 10⁶ cells/sample).

Flow Cytometry

Primary Müller glia were treated with a TRPV4 agonist and/or LXB₄ for 24 hours and then harvested with TrypLE Express (Gibco). The cells were washed in ice-cold PBS and fixed/permeabilized. Fixed cells were incubated overnight at 4°C. The cells were washed in wash buffer and then stained with primary antibody for 1 hour at RT in the dark, followed by washing with cell staining wash buffer. The cells were incubated with secondary antibody for 1 hour at RT, washed, and resuspended in wash buffer for analysis via flow cytometry. The cells were analyzed via an Attune Acoustic Focusing Flow Cytometer (Thermo Fisher Scientific). GFAP⁺ cells were identified by comparison to isotype controls. The data were analyzed with FlowJo v10 (BD, Ashland, OR, USA). The fraction of GFAP-positive cells was used for quantitative comparisons.

PhosphoKinase Assay

For phosphokinase profiling, the Proteome Profiler Human Phospho-Kinase Array (R&D Systems, Minneapolis, MN, USA) was used. rMC-1 cells were treated with or without 1 µM LXB₄ for 30 minutes and assay was performed according to the manufacturer's instructions. Protein concentrations were determined using a BCA assay (Pierce, Thermo Fisher Scientific), and 200 µg of total protein per sample was used for each assay. Membranes imaged on a LI-COR Odyssey system. Spot intensities were quantified using LI-COR software after background subtraction and were normalized to the array's positive control signals.

LXB₄ Treatment

For single-cell transcriptomic analysis, the mice were treated i.p. with LXB₄ methyl ester (Cayman Chemical) (1 µg in 100 µL of PBS) once daily and topically (1 µg in 5 µL of PBS per eye) three times per day for 3 days. Vehicle (ethanol) for LXB₄ methyl ester was removed under a stream of nitrogen, and LXB₄ methyl ester was resuspended in sterile PBS immediately before injection or topical treatment. On day 4, mice were euthanized for retinal dissociation and single-cell library preparation. For IHC analysis, the mice received daily LXB₄ methyl ester (1 µg in 100 µL of PBS) via i.p. injection and topical ocular administration for 1 week. Control mice received equivalent volumes of sterile PBS. LXB₄ was administered 15 minutes before OHT induction.

Single-Cell Transcriptomics

Single-cell transcriptomic data were reanalyzed for the Müller glial population from a previously published dataset from our laboratory.²⁴ Retinas were dissociated into single‐cell suspensions via a papain dissociation kit (Worthington), followed by rod photoreceptor depletion in a magnetic column via CD133-biotin labeling and anti-biotin magnetic beads (Miltenyi Biotech, Bergisch Gladbach, Germany). Single cells were loaded on the 10× Chromium Single Cell 3′ v3.1 platform (10x Genomics, Pleasanton, CA, USA) to generate barcoded gel bead emulsions, and captured RNA was reverse transcribed and amplified into cDNA libraries, which were quality checked on an Agilent Bioanalyzer. Libraries were sequenced on an Illumina NovaSeq S1 100SR flow cell, and raw BCL files were demultiplexed via Illumina bcl2fastq2 at the QB3 Genomics Core, UC Berkeley (Berkeley, CA, USA) (RRID: SCR_022170). The sequencing reads were aligned to the mm10 (mouse) genome with Cell Ranger software (10x Genomics). Downstream analyses were performed in Seurat (v4.1.0) as detailed in our recent publications.^24,38 Retinal cell types were annotated by identifying cluster-specific markers with FindAllMarkers() and comparing them with known markers.³⁸ Differential gene expression in Müller glia was determined via FindMarkers() and visualized with VlnPlot() and DotPlot(). The single-cell RNA-seq data were deposited in GEO (GSE251716, single-cell RNA-seq).

Western Blot

Treated Müller glia were lysed in RIPA buffer (Thermo Fisher Scientific) containing a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific) at a 1:100 dilution. Protein concentrations were determined via a BCA assay (Pierce, Thermo Fisher Scientific). Equal amounts of protein (20 µg) were denatured, separated on 10% SDS‒PAGE gels, and then transferred to nitrocellulose membranes (Bio-Rad Laboratories). The membranes were blocked in Intercept Blocking Buffer (LiCOR, Lincoln, NE, USA) for 1 hour at RT and then probed overnight at 4°C with the following antibodies: anti-phospho-STAT3 (Ser727; rabbit monoclonal; 1:1000; Abcam ab32143, Cambridge, MA, USA), total anti-STAT3 (rabbit monoclonal; 1:2000; Abcam ab109085, Cambridge, MA, USA), and anti-GAPDH (mouse monoclonal; 1:10,000; Cell Signaling Technology, Danvers, MA, USA). After washing in TBS-T, the membranes were incubated with IRDye 800CW donkey anti-rabbit IgG and IRDye 680RD donkey anti-mouse IgG (1:10,000; LI-COR Biosciences) for 1 hour at RT in the dark. The blots were imaged on a LI-COR Odyssey system. Band intensities were quantified with ImageJ (National Institutes of Health). Phospho-STAT3 signals were normalized to total STAT3 and GAPDH signals and normalized to those of untreated controls for fold-change calculations.

Statistical Analysis

All the data are presented as the means ± standard errors of the means. Biological replicates (n) represent the number of animals (for in vivo experiments) or independent cell culture experiments (for in vitro assays). Statistical analyses were conducted in GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA). For two-group comparisons, an unpaired Student t test was used. For multiple-group comparisons, one-way ANOVA followed by Tukey's post hoc test was performed, with the assumption of a normal distribution of variance. Statistical significance was defined as a P value of less than 0.05.

Results

Müller Glia Express a Functional Lipoxin Biosynthetic Pathway

Our previous studies established that astrocytes generate lipoxins and that LXB₄ treatment reduces Müller glia reactivity.^23,39 We next asked whether Müller glia themselves synthesize lipoxins. Primary mouse Müller glia were isolated and validated by ICC for canonical markers, including vimentin, glutamine synthetase, and SOX9, and were confirmed to be negative for GFAP and IBA1 (Fig. 1A). Immortalized rMC-1 cells were validated via the same markers and expressed all the same positive and negative markers (Supplementary Fig. S1A). Bulk RNA-seq analyses of cultured Müller glia revealed robust expression of Alox5, along with significant expression of Alox5ap, Alox12, and Alox15 (Fig. 1B), indicating the presence of key LOX pathway enzymes for lipoxin generation. Targeted lipidomic profiling revealed the presence of arachidonic acid and its LOX metabolites 5-HETE, 12-HETE, and 15-HETE in both cell lysates and conditioned media from both primary (Fig. 1C) and rMC-1 cells (Supplementary Fig. S1B). qPCR analysis established robust expression of the Alox5 (Ct: 27.18) and Alox15 (Ct: 28.05) genes in primary Müller glia, confirming the RNA-seq data (Fig. 1D). More important, Müller glia (both primary and immortalized) generated significant amounts of 5-HETE (1210 ng/mL) and 15-HETE (1502 ng/mL) and endogenous LXB₄ (114 pg/mL) (Fig. 1C; Supplementary Fig. S1B; Supplementary Fig. S2A). No LXB₄ was detected in media with serum and no cells that served as a control for cell specific formation. Together, these findings establish that primary Müller glia functionally express the LXB₄ biosynthetic pathway (Alox15 and Alox5), revealing a previously unrecognized glial lipid mediator pathway.

Figure 1.

Functional expression of the lipoxin pathway in Müller glia. (A) Primary mouse Müller glia were stained for canonical markers (vimentin [VIM], glutamine synthetase [GS], and SOX9) and negative control markers (GFAP and IBA1). (B) Bulk RNA-seq of primary Müller glia for LOX pathway genes (n = 3 biological replicates). (C) LC-MS/MS quantification of released eicosanoid substrate (arachidonic acid [AA]) and corresponding LOX products in primary Müller glia (n = 3 biological replicates). (D) qPCR analysis of Alox5 and Alox15 expression in primary Müller glia (n = 3 biological replicates).

OHT Induces a Reactive Müller Glial Phenotype and Increases TRPV4 Expression

To model sustained mechanical insult, we induced chronic OHT in C57BL/6J mice via anterior chamber silicone oil injection.^23,24,28 Mice were euthanized, and the retinas were collected at 2, 6, and 8 weeks post injection to assess Müller glia reactivity by IHC with GFAP as an established marker (Fig. 2A). Imaging revealed an 80% increase in GFAP signal intensity in the ganglion cell layer and inner nuclear layer, specifically in astrocytes and Müller glia, at 2 and 6 weeks (Fig. 2B). In parallel, TRPV4 signal intensity increased significantly at every time point examined (Fig. 2C), with a 114.6% increase in TRPV4 expression relative to that in age-matched normotensive controls at 6 weeks (Fig. 2D). qPCR analysis at 4 weeks OHT revealed the upregulation of Il-6 (6.2-fold), Tnf-α (20-fold), and Stat3 (25.5-fold) when compared with controls (Fig. 2E), which was consistent with proinflammatory activation. Gene Set Enrichment Analysis of RNA-seq data from MACS-isolated Müller glia supported these findings, revealing significant upregulation of cytokine-related pathways, including Il-6 and Tnf-α production. In addition, pathways related to visual perception and the response of sensory stimuli to light stimuli were also downregulated for the primary supportive cell of neuronal cells, indicating possible negative changes in Müller glia supportive functional pathways (Fig. 2F). These data demonstrate that chronic OHT induces gliosis-associated molecular changes, TRPV4 upregulation, and activation of inflammatory signaling pathways, highlighting the role of Müller glia as central responders to mechanical insult.

Figure 2.

OHT induces GFAP and TRPV4 expression. (A) Representative IHC images of retinal sections from mice with sustained OHT at 2, 6, and 8 weeks post injection and age-matched normotensive controls stained for GFAP and DAPI. (B) Quantification of GFAP immunoreactivity, expressed as the percent change from the control (n = 5–10). (C) Representative immunofluorescence images from mice with OHT and normotensive controls stained for TRPV4 and DAPI. (D) Quantification of TRPV4 immunoreactivity, expressed as the percent change from control (n = 3–7). (E) qPCR analysis of Il6, Tnf-α, and Stat3 in whole retinas collected 4 weeks post OHT (n = 3). Data were normalized to Gapdh and matched with normotensive controls. (F) Gene Set Enrichment Analysis of MACS-isolated Müller glia from chronic OHT conditions. Images acquired at 20× magnification; scale bar, 50 µm. GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Statistical analysis: one-way ANOVA with Tukey's post hoc test (*P < 0.05, **P < 0.01, ***P < 0.001) or unpaired Student t test (*P < 0.05, **P < 0.01).

TRPV4 Activation Drives Müller Glia Reactivity and Amplifies LXB₄ Production

To isolate the specific contribution of TRPV4 signaling, primary and immortalized Müller glia were treated for 24 hours with the selective TRPV4 agonist GSK. Cells from control and treated conditions were analyzed by flow cytometry, and GFAP⁺ cells were selectively gated (Fig. 3A). Quantification revealed that TRPV4 activation caused a 118% increase in GFAP-positive cells compared with vehicle-treated controls (DMSO) (Fig. 3B). ICC further confirmed the increased expression of GFAP and vimentin along Müller glial filaments after TRPV4 activation (Fig. 3C). Lipidomic analyses revealed that TRPV4 activation triggered a marked increase in the release of the LOX substrate arachidonic acid and 12-HETE a downstream signature 12/15-LOX metabolite. Notably, reactive Müller glia increased LXB₄ production by 2.3-fold relative to controls (Fig. 3D). Thus, TRPV4 activation is sufficient to drive Müller glial reactivity and activate LOX pathways to amplify intrinsic LXB₄ production.

Figure 3.

Activation of TRPV4 induces Müller glial reactivity and enhances LXB₄ formation. (A, B) Flow cytometry gating and quantification of GFAP⁺ cells following TRPV4 agonist (GSK) vs vehicle (DMSO, ctrl) treatment (n = 4 biological replicates). (C) Representative ICC images of GFAP and vimentin in GSK- or vehicle-treated (ctrl) Müller glia. (D) LC-MS/MS quantification of released polyunsaturated fatty acids, 5-LOX, 12/15-LOX metabolites, and lipoxins (LXA₄, LXB₄) by Müller glia after GSK treatment (n = 5 biological replicates). Images acquired at 20× magnification; scale bar, 50 µm. Statistical analysis: unpaired Student t test (*P < 0.05, **P < 0.01, ***P < 0.001).

LXB₄ Suppresses JAK-STAT Signaling in Homeostatic Müller Glia

Because IL-6/JAK/STAT is a central driver of reactive gliosis and was upregulated in our OHT model (Fig. 2E), we next investigated whether exogenous LXB₄ modulates this pathway. When immortalized Müller glia (rMC-1) were treated with LXB₄, a phosphokinase array revealed that LXB₄ selectively reduced the phosphorylation of STAT2-Y690 (18%), STAT5a/b-Y699 (16.8%), and STAT3-S727 (19%) (Fig. 4A). To assess the transcriptional consequences of LXB₄ signaling in vivo, retinas from LXB₄ treated mice were analyzed by single-cell RNA-seq. UMAP analysis revealed a distinct Müller–glial cluster (Fig. 4B), within which Stat1 and Stat3 transcript levels were reduced in LXB₄-treated mice (Fig. 4C). Differential gene expression analysis from single-cell RNA-seq data further revealed downregulation of Jak2, Il6st, Stat5b, and Stat2 (Fig. 4D), indicating broad suppression of the inflammatory JAK-STAT network. Collectively, these findings demonstrate that LXB₄ broadly attenuates JAK-STAT signaling at both the protein and transcriptional levels in homeostatic Müller glia.

Figure 4.

LXB₄ regulates JAK-STATs in homeostatic Müller glia in vitro and in vivo. (A) Phosphokinase array of immortalized Müller glia (rMC-1) treated with or without LXB₄ (n = 4). (B) UMAP projection of single-cell RNA-seq data from retinas treated with or without LXB₄, highlighting the Müller glia (MG) populations. (C) Violin plots comparing Stat1 and Stat3 expression in MG populations between groups. (D) Differential gene expression of Jak2, Stat2, Stat5b, Il6st, and Stat1 in MGs from the single-cell RNA-seq dataset. Statistical analysis: unpaired Student t test (*P < 0.05).

LXB₄ Inhibits TRPV4-Induced Müller Glial Reactivity via IL-6/STAT3 Suppression

To determine whether LXB₄ can prevent TRPV4-induced Müller glial reactivity, we pretreated primary Müller glia with LXB₄ before stimulation with the TRPV4 agonist GSK. After 24 hours, flow cytometry analysis revealed that LXB₄ pretreatment significantly reduced the number of GFAP-positive cells by 84.6% compared with that in the GSK-treated controls (Fig. 5A, 5B). ICC further confirmed that the GFAP and vimentin filament intensities significantly decreased, returning to near-baseline levels (Fig. 5C). qPCR analysis revealed a significant decrease in Il6 expression in LXB₄-treated primary Müller glia (Fig. 5D). This reduction in Il6 (1.6-fold) was accompanied by decreased STAT3 phosphorylation (Ser727) by 75%, as determined by Western blotting and normalized to total STAT3 and GAPDH (Fig. 5D), which is consistent with Il6 acting upstream of STAT3 activation. Together, these results demonstrate that LXB₄ effectively suppresses TRPV4-induced inflammatory activation in immortalized Müller glia by downregulating Il6 expression, inhibiting STAT3 phosphorylation, and preventing the upregulation of cytoskeletal reactivity markers.

Figure 5.

LXB₄ suppresses the IL-6–STAT pathway in vitro. (A, B) Flow cytometry of GFAP⁺ cells in control (vehicle), GSK, or GSK + LXB₄ treated primary Müller glia (n = 4 biological replicates). (C) ICC images of GFAP and vimentin in the GSK, or GSK + LXB₄–treated cells. (D) qPCR results for Il6 expression in control, GSK, or LXB₄ + GSK–treated Müller glia (n = 5). (E, F) Western blot analysis of phospho-STAT3 (S727) and total STAT3. Representative blots are shown; GAPDH was used as a loading control (n = 5). Images acquired at 20× magnification; scale bar, 50 µm. Statistical analysis: one-way ANOVA with Tukey's post hoc test (*P < 0.05, **P < 0.01).

LXB₄ Inhibits Inflammatory Gene Expression and TRPV4 Upregulation in Chronic and Acute OHT Models

Finally, we evaluated the therapeutic potential of LXB₄ in both chronic and acute models of OHT.^23,24 In the chronic model, qPCR analysis of whole retina RNA from mice treated with LXB₄ after 4 weeks of OHT revealed marked reductions in Stat3 (1.95-fold), Il6 (3.7-fold), and TNF-α (3.6-fold) expression when directly compared to vehicle-treated mice with OHT (Fig. 6⁷A). In a separate acute OHT model (1 week post induction), IHC demonstrated that LXB₄ treatment prevented the OHT-induced increase in TRPV4 expression, particularly in the ganglion cell layer and inner nuclear layer (Fig. 6B). TRPV4 signal intensity was reduced by 63% in LXB₄-treated mice when directly compared with vehicle-treated mice with OHT (Fig. 6C). These in vivo results reinforce the therapeutic neuroprotective potential of LXB₄, as it suppresses inflammation and TRPV4 upregulation in both acute and chronic OHT models of glaucomatous neurodegeneration.

Figure 6.

LXB₄ treatment suppresses proinflammatory markers and TRPV4 expression in vivo. (A) qPCR analyses of Stat3, Il6, and TNF-α expression in whole retinas of normotensive eyes and at 4 weeks of chronic OHT with or without LXB₄ treatment (n = 4 biological replicates). (B) Representative IHC images of TRPV4 IHC in normotensive controls and those subjected to acute OHT for 1 week with or without LXB₄ treatment. (C) Quantification of TRPV4 protein expression in control retinas and OHT-treated retinas with or without LXB₄ treatment (n = 3–5 biological replicates). Images acquired at 20× magnification; scale bar, 50 µm. Statistical analysis: one-way ANOVA with Tukey's post hoc test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Figure 7.

Model scheme for the Müller glia TRPV4-LXB₄ circuit. Proposed scheme for the Müller Glia LXB₄ pathway. Homeostatic Müller express the full LXB₄ biosynthetic pathway and generate neuroprotective LXB₄. OHT leads to activation of mechanosensitive TRPV4 and upregulation of the ion channel in Müller Glia. This chronic and amplified activation induces an inflammatory reactive Müller Glia phenotype, but in parallel amplifies neuroprotective LXB₄ as a counter-regulatory pathway. LXB₄ treatment downregulates TRPV4 and counter regulates TRPV4 signaling by inhibiting the STAT pathway and expression of IL-6 and GFAP.

Discussion

The role of Müller glia in glaucoma pathogenesis is complex and multifaceted, transitioning from an initially protective response to a chronic, maladaptive phenotype that accelerates RGC loss.⁴⁰ Building on prior work showing that astrocyte-derived LXB₄ protects RGCs, we extended this paradigm to Müller glia.^23,41 We show that these cells are both sensors of biomechanical insult via TRPV4 and active producers of an intrinsic, lipoxin-based anti-inflammatory circuit. These findings suggest that Müller glia not only contribute to neuroinflammation, but also possess a self-regulatory mechanism that could be amplified to suppress gliosis and protect RGCs (Fig. 7).

We discovered that Müller glia express a functional lipoxin biosynthetic circuit characterized by 5-LOX and 15-LOX expression and the endogenous production of LXB₄. Although previous studies have identified retinal astrocytes as a source of lipoxins,²³ our findings extend this paradigm by demonstrating that Müller glia, the most abundant macroglia cell type in the retina, also is a key contributor to this neuroprotective retinal pathway. LC-MS/MS-based lipidomics confirmed that these cells not only express the necessary enzymes, but also actively produce key lipoxin precursors (5-, 12-, and 15-HETE) and release neuroprotective LXB₄ under homeostatic conditions, revealing a previously unrecognized biosynthetic capacity.

In vivo, chronic OHT robustly upregulated TRPV4 in Müller glia, leading to the activation of canonical gliosis hallmarks, including IL-6/STAT3 signaling and GFAP expression. These findings align with those of previous studies in other central nervous system models, which demonstrated that TRPV4 activation leads to Ca²⁺-dependent p-STAT3 induction and that inhibiting STAT3 mitigates neurodegeneration.^42–44 Our in vitro data further support this link, showing that pharmacological TRPV4 activation in Müller glia recapitulates gliotic changes while increasing lipoxin production. These findings suggest a direct functional relationship between the biomechanical stress sensor TRPV4 and the gliotic response, highlighting a key regulatory point for Müller glia function.

TRPV4-activated Müller glia reactivity induced both inflammatory (IL-6 and TNF-α) and neuroprotective (LXB₄) pathways. The LOX pathways (5-LOX and 12/15-LOX) that were amplified, in addition to generating protective lipoxins, can also generate proinflammatory lipid mediators.^16,45 We found that 5-LOX, a rate-limiting enzyme for lipoxin generation, can also generate proinflammatory leukotrienes. However, in vitro primary and immortalized Müller glia did not generate any detectable amounts of leukotrienes under any conditions. Consistent with our previous lipidomic analyzes of healthy and glaucomatous retinas,^23,41 these findings are consistent with our previous in vivo studies that detected no leukotrienes in the retina or optic nerve in healthy eyes or eyes with OHT.^22,23 A primary product of the mouse 12/15-LOX (Alox15) but with the human 15-LOX (ALOX15), 12-HETE was increased in reactive rodent Müller glia.⁴⁶ In patients with diabetic retinopathy, 12-HETE has been reported and is known to activate and upregulate inflammatory markers, including IL-6 and TNF-α, in Müller glia.^46,47 These findings indicate that TRPV4-induced gliosis triggers the activation of downstream LOX pathways, which potentially amplifies LXB₄ neuroprotective signaling but may also initiate an inflammatory pathway (i.e., 12-HETE). Hence, TRPV4 activation triggered by OHT-induced biomechanical stress triggers gliosis, but can also increase lipoxin biosynthesis in Müller glia, which indicates a potential self-regulating mechanism where TRPV4-driven gliosis is counterbalanced by amplified endogenous lipoxin signaling.

LXB₄ treatment disrupted the inflammatory circuit and restored homeostatic balance under chronic OHT. Our experiments revealed that LXB₄ treatment suppresses STAT3 phosphorylation, downregulates TRPV4 expression, and reverses gliosis (Figs. 4⁵–6), effects that parallel those observed with pharmacological TRPV4 inhibition.⁹ Our findings align with studies that have shown treatment with omega-3 pro-resolving lipid mediators, such as Maresin-1, has overlapping anti-inflammatory actions by decreasing TRPV4 expression in other central nervous system models.⁴⁸ Lipoxins, unlike Maresins, are an endogenous protective lipid mediator pathway in the retina. Hence, therapeutic amplification of LXB₄ signaling selectively targets the resident lipid mediator pathway in the retina. Stable analogs of lipoxin have been developed and especially the protective action of LXA₄ mimetics are well-characterized in multiple models of central nervous system dysfunction.⁴⁹ In models of posterior uveitis, a form of retinal inflammation, both LXA₄ and LXB₄ reduce macroglial reactivity, microglial activation, and macrophage infiltration.⁵⁰ However, LXB₄’s neuroprotective actions in OHT and neurotoxic models of glaucomatous injury are superior those of LXA₄.⁴¹ Whereas prior studies have focused on lipoxin activity in modulating astrocyte and microglial inflammatory responses, our current findings are the first to identify Müller glia as both a source and a target of lipoxin signaling, establishing a unique self-regulatory inflammatory circuit specific to the retina.^23,24,39,41

The identification of a counter-regulatory lipoxin circuit in Müller glia has potential translational implications for glaucoma treatment. Because Müller glia are the major glial cell type in the retina and originate from the same progenitor cells as RGCs do, they represent an ideal target for therapeutic intervention.^5,7 Therefore, strategies that augment this intrinsic protective circuit could be a targeted approach to prevent the maladaptive gliosis that precedes RGC degeneration. For example, this could be achieved pharmacologically with established stable LXB₄ analogs or genetically via macroglia specific adeno-associated virus–mediated amplification of the lipoxin pathway. This targeted approach is particularly promising, because it leverages the retina's own neuroprotective mechanisms to restore homeostatic balance.

This study has several limitations. Although our findings provide mechanistic insight into Müller glia reactivity, longer-term studies are needed to evaluate the impact of LXB₄ treatment on the homeostatic function of Müller glia. Although we focused on TRPV4 and STAT3 signaling, other inflammatory pathways, such as the nuclear factor-κB or mitogen-activated protein kinase pathways, likely contribute to the gliotic response and deserve further exploration.^49,50 Additionally, much of our mechanistic work was performed in primary and immortalized Müller glia cultures. Although these systems are invaluable for dissecting signaling pathways in a controlled environment, they inevitably simplify the cellular complexity of the retina. Important interactions between glia, neurons, and the vasculature are not fully understood in vitro. Future work using conditional knockout models and live imaging will help to clarify the temporal dynamics of this lipid signaling axis in vivo.

Conclusions

We identified the intrinsic neuroprotective and anti-inflammatory lipoxin circuit in Müller glia, which are the primary and most abundant supportive glia in the retina. We established that chronic activation of the mechanosensitive TRPV4 ion channel in Müller glia drives sustained gliosis and IL-6/STAT3 pathway activation. This TRPV4-induced gliosis is counter-regulated by the amplification of Müller glia LXB₄ formation. LXB₄ targets both OHT-induced TRPV4 overexpression and TRPV4-induced STAT3 phosphorylation, thereby reversing the inflammatory reactivity of Müller glia. Hence, targeting the Müller glia lipoxin pathway could provide new therapeutic opportunities for glaucoma and other retinal neurodegenerative diseases associated with chronic activation of TRPV4 and gliosis of Müller cells.