Ocular-specific ER stress reduction rescues glaucoma in murine glucocorticoid-induced glaucoma

Abstract

Administration of glucocorticoids induces ocular hypertension in some patients. If untreated, these patients can develop a secondary glaucoma that resembles primary open-angle glaucoma (POAG). The underlying pathology of glucocorticoid-induced glaucoma is not fully understood, due in part to lack of an appropriate animal model. Here, we developed a murine model of glucocorticoid-induced glaucoma that exhibits glaucoma features that are observed in patients. Treatment of WT mice with topical ocular 0.1% dexamethasone led to elevation of intraocular pressure (IOP), functional and structural loss of retinal ganglion cells, and axonal degeneration, resembling glucocorticoid-induced glaucoma in human patients. Furthermore, dexamethasone-induced ocular hypertension was associated with chronic ER stress of the trabecular meshwork (TM). Similar to patients, withdrawal of dexamethasone treatment reduced elevated IOP and ER stress in this animal model. Dexamethasone induced the transcriptional factor CHOP, a marker for chronic ER stress, in the anterior segment tissues, and Chop deletion reduced ER stress in these tissues and prevented dexamethasone-induced ocular hypertension. Furthermore, reduction of ER stress in the TM with sodium 4-phenylbutyrate prevented dexamethasone-induced ocular hypertension in WT mice. Our data indicate that ER stress contributes to glucocorticoid-induced ocular hypertension and suggest that reducing ER stress has potential as a therapeutic strategy for treating glucocorticoid-induced glaucoma.

Figure 1. Topical ocular dexamethasone induces glaucoma in mice.

(A) Elevated IOP in dexamethasone-treated C57BL/6 mice. Topical ocular vehicle (sterile PBS) or dexamethasone (0.1%) was administered 3 times daily for up to 20 weeks. IOP measurements of dexamethasone-treated (n = 20–24) and vehicle-treated (n = 20) mice are shown from week 0 to 6 of treatment. **P < 0.005, ***P < 0.0001, unpaired t test. (B) Progressive RGC functional loss in dexamethasone-treated mice. PERG amplitudes (P50-N95) in vehicle and dexamethasone-treated mice at 5 (n = 5), 15 (n = 10), and 20 (n = 6) weeks of treatment. (C and D) Loss of RGCs in dexamethasone-treated mice. (C) Representative images of Nissl-stained whole-mount retinas from mice treated for 20 weeks with vehicle or dexamethasone. (D) Remaining cells in ganglion layer were counted in the periphery of the retina. n = 5 (vehicle); 10 (dexamethasone). **P = 0.0045, unpaired t test. (E and F) Progressive optic nerve degeneration in mice treated with dexamethasone for 10 or 15 weeks. Optic nerve sections were stained with PPD (E), and mean axon counts (F) were compared in dexamethasone- (n = 8) and vehicle-treated (n = 7–10) mice. *P = 0.0174, **P = 0.0013 vs. vehicle, unpaired t test. Scale bar: 10 μm.

Figure 2. Topical ocular dexamethasone increases MYOC and actin levels in the TM.

Immunostaining for MYOC and actin merged with nuclear stain DAPI is shown. MYOC and actin were localized to the TM and CB of mice treated with vehicle or dexamethasone for 3 weeks. MYOC and actin levels were increased in TM and CB of dexamethasone-treated mice (n = 4) compared with vehicle-treated mice (n = 3). Arrow shows TM. Scale bar: 50 μm.

Figure 3. Dexamethasone induces ER stress and activates the UPR in human TM cells.

(A) Splicing of XBP-1. Primary human TM cells were treated with dexamethasone (100 nM) for 10 days, and total RNA prepared was subjected to RT-PCR with XBP1 primers. The resulting products were subjected to 2% agarose gel electrophoresis. PCR products were sequenced to further confirm the presence of spliced and unspliced XBP-1. Spliced product was observed at 257 bp, while unspliced products were observed at 283 bp. Spliced XBP-1 was only observed in dexamethasone-treated samples. (B) Dexamethasone induced ER stress and activated the UPR in human TM cells. Primary human TM cells obtained from normal human donors were treated with dexamethasone for 48 hours. GRP78, GRP94, CHOP, and phosphorylation of IRE1 and eIF2α were examined by Western blot analysis (n = 3). (C) Human TM cells were treated with vehicle or dexamethasone for 48 hours and stained with GRP78 and phalloidin (which stains F-actin) (n = 3). Dexamethasone treatment increased actin and GRP78 levels. Scale bars: 50 μm.

Figure 4. Topical dexamethasone induces ER stress and activates the UPR in TM tissues of WT mice.

(A) GRP78, activated ATF-6α, ATF-4, spliced XBP-1, CHOP, MYOC, and GAPDH (loading control) were examined by Western blot analysis in anterior segment tissues of mice treated for 1 week with vehicle or dexamethasone (n = 4 per group). (B) Representative immunostaining for GRP78 in the anterior segment tissues of mice treated for 8 weeks with vehicle or dexamethasone (n = 5 per group), showing increased GRP78 in the TM (arrows) of dexamethasone-treated mice. CE, corneal endothelium. Scale bars: 50 μm.

Figure 5. Dexamethasone withdrawal reduces dexamethasone-associated elevations in IOP and ER stress.

WT mice were given topical ocular vehicle or dexamethasone for 3 weeks (IOP measurements confirmed elevated IOP with dexamethasone treatment). At this time point, dexamethasone-treated mice were randomly divided into 2 groups: one received topical dexamethasone, and the other vehicle (i.e., dexamethasone withdrawal), for another 2 weeks. (A) 2 weeks of dexamethasone withdrawal significantly reduced IOP (n = 10 per group). P values were determined by 1-way ANOVA. (B) Western blot analysis of GRP78, MYOC, and FN in anterior segment tissue lysates. (C–E) Densitometric analysis of GRP78 (C), MYOC (D), and FN (E) levels, normalized to loading control GAPDH, revealed that dexamethasone withdrawal reduced the dexamethasone-associated elevations in these proteins. *P < 0.05, **P < 0.005, 1-way ANOVA.

Figure 6. Deletion of Chop protects against dexamethasone-induced ocular hypertension.

(A) Chop knockout protected mice from dexamethasone-induced ocular hypertension. WT and Chop knockout mice were given topical ocular vehicle or dexamethasone for 3 weeks. IOP measurements showed that dexamethasone elevated IOP significantly compared with vehicle in WT mice, but not in Chop knockout mice (n = 10 per group). P values were determined by 1-way ANOVA. (B and C) Western blot analysis for GRP78, CHOP, and MYOC in anterior segment tissues of vehicle- and dexamethasone-treated WT (B) and Chop knockout (C) mice (n = 4 per group). (D) ER stress and MYOC levels, shown as fold change relative to vehicle control, were reduced with dexamethasone treatment in Chop knockout compared with WT mice. **P < 0.005, unpaired t test.

Figure 7. Decreasing ER stress by administration of the chemical chaperone PBA reduces IOP elevation by dexamethasone.

(A–C) WT mice were given topical ocular vehicle or dexamethasone for 3 weeks. Dexamethasone-treated mice were divided into 2 groups: one received water, the other received 20 mM PBA in drinking water. (A) PBA treatment significantly protected from dexamethasone-induced IOP elevation (n = 20 per group). **P < 0.05, ***P < 0.005, 1-way ANOVA. (B and C) Western blot (B) and densitometric analysis (C) of ER stress markers in anterior segment tissues revealed that combined dexamethasone and PBA treatment reduced ER stress markers compared with dexamethasone treatment alone (n = 5 per group). (D) PBA reduced ER stress associated with dexamethasone in human TM cells. Human TM cells were treated with dexamethasone with or without 5 mM PBA. Total cell lysates were subjected to Western blot analysis for GRP78, GRP94, phosphorylated and total eIF2α, MYOC, and GAPDH.