Ocular Hypertension Results in Hypoxia within Glia and Neurons throughout the Visual Projection

Abstract

The magnitude and duration of hypoxia after ocular hypertension (OHT) has been a matter of debate due to the lack of tools to accurately report hypoxia. In this study, we established a topography of hypoxia in the visual pathway by inducing OHT in mice that express a fusion protein comprised of the oxygen-dependent degradation (ODD) domain of HIF-1a and a tamoxifen-inducible Cre recombinase (CreERT2) driven by a ubiquitous CAG promoter. After tamoxifen administration, tdTomato expression would be driven in cells that contain stabilized HIF-1α. Intraocular pressure (IOP) and visual evoked potential (VEP) were measured after OHT at 3, 14, and 28 days (d) to evaluate hypoxia induction. Immunolabeling of hypoxic cell types in the retina and optic nerve (ON) was performed, as well as retinal ganglion cell (RGC) and axon number quantification at each time point (6 h, 3 d, 14 d, 28 d). IOP elevation and VEP decrease were detected 3 d after OHT, which preceded RGC soma and axon loss at 14 and 28 d after OHT. Hypoxia was detected primarily in Müller glia in the retina, and microglia and astrocytes in the ON and optic nerve head (ONH). Hypoxia-induced factor (HIF-a) regulates the expression of glucose transporters 1 and 3 (GLUT1, 3) to support neuronal metabolic demand. Significant increases in GLUT1 and 3 proteins were observed in the retina and ON after OHT. Interestingly, neurons and endothelial cells within the superior colliculus in the brain also experienced hypoxia after OHT as determined by tdTomato expression. The highest intensity labeling for hypoxia was detected in the ONH. Initiation of OHT resulted in significant hypoxia that did not immediately resolve, with low-level hypoxia apparent out to 14 and 28 d, suggesting that continued hypoxia contributes to glaucoma progression. Restricted hypoxia in retinal neurons after OHT suggests a hypoxia management role for glia.

Destabilizing COXIV in Müller Glia Increases Retinal Glycolysis and Alters Scotopic Electroretinogram

“…Additionally, inhibiting glycolysis in MG in vivo did not affect their survival [ 12 ]. These data and our observations about MG sensitivity to hypoxia [ 31 , 32 ] suggest that MG may not be as reliant on glycolysis as previously reported.…”

“…Metabolic dysfunction has been reported in RGCs from glaucoma patients and in animal models [ 2 , 4 , 5 , 40 , 41 , 42 ], and we have reported the sensitivity of MG to OHT-associated hypoxia [ 31 , 32 ], so we examined the impact of MG-specific COX10 knockout on RGC survival following 4 weeks of ocular hypertension. Significant IOP elevation was observed 7 days after microbead injection and lasted for 3 additional weeks ( Figure 5 A); there was no difference in IOP between the groups that underwent ocular hypertension (OHT).…”

“…Our goal with this work was two-fold, to understand why MG are sensitive to hypoxia as shown in our previous work [ 31 , 32 ], and to shed additional light on fundamental MG metabolism. By knocking out COX10, a heme farnesyltransferase, in GLAST-expressing cells, we sought to destabilize OXPHOS in the retina and evaluate the impact on MG morphology and function.…”

“…Magnetic microbeads (1.5 μL, 8 μm diameter; COMPEL COOH-Modified, UMC4001; Bangs Laboratories, Inc., Fishers, IN, USA) were injected into the anterior chamber of both eyes via glass pulled micropipette attached to a microinjection system (World Precision Instruments, Sarasota, FL, USA), and a neodymium magnet was used to guide the microbeads into the trabecular meshwork to block the outflow of aqueous humor and increase IOP [ 32 ]. We used the magnetic bead model successfully to elevate IOP 4–5 weeks.…”

“…We recently observed that MG mount a strong hypoxia response to ocular hypertension [ 31 , 32 ]. This sensitivity to low levels of oxygen in the retina led us to examine what is known about mitochondria and OXPHOS in MG.…”