Summary Retinal ganglion cell (RGC) replacement holds potential for restoring vision lost to optic neuropathy. Transplanted RGCs must undergo neuroretinal integration to receive afferent visual signals for processing and efferent transmission. To date, retinal integration following RGC transplantation has been limited. We sought to overcome key barriers to transplanted human stem cell-derived RGC integration. Following co-culture ex vivo on organotypic mouse retinal explants, human RGCs cluster and extend bundled neurites that remain superficial to the neuroretina, hindering afferent synaptogenesis. To enhance integration, we increased the cellular permeability of the internal limiting membrane (ILM). Extracellular matrix digestion using proteolytic enzymes achieved ILM disruption while minimizing retinal toxicity and preserving glial reactivity. ILM disruption is associated with dispersion rather than clustering of co-cultured RGC bodies and neurites, and increased parenchymal neurite ingrowth. The ILM represents a significant obstacle to transplanted RGC connectivity and its circumvention may be necessary for functional RGC replacement.

Purpose To delineate responses of optic nerve head astrocytes to sustained intraocular pressure (IOP) elevation in mice. Methods We elevated IOP for 1 day to 6 weeks by intracameral microbead injection in 4 strains of mice. Astrocyte alterations were studied by transmission electron microscopy (TEM) including immunogold molecular localization, and by laser scanning microscopy (LSM) with immunofluorescence for integrin β1, α-dystroglycan, and glial fibrillary acidic protein (GFAP). Astrocyte proliferation and apoptosis were quantified by Ki67 and TUNEL labeling, respectively. Results Astrocytes in normal optic nerve head expressed integrin β1 and α-dystroglycan by LSM and TEM immunogold labeling at electron dense junctional complexes that were found only on cell membrane zones bordering their basement membranes (BM) at the peripapillary sclera (PPS) and optic nerve head capillaries. At 1–3 days after IOP elevation, abnormal extracellular spaces appeared between astrocytes near PPS, and axonal vesical and mitochondrial accumulation indicated axonal transport blockade. By 1 week, abnormal spaces increased, new collagen formation occurred, and astrocytes separated from their BM, leaving cell membrane fragments. Electron dense junctional complexes separated or were absent at the BM. Astrocyte proliferation was modest during the first week, while only occasional apoptotic astrocytes were observed by TEM and TUNEL. Conclusions Astrocytes normally exhibit junctions with their BM which are disrupted by extended IOP elevation. Responses include reorientation of cell processes, new collagen formation, and cell proliferation.

Purpose To study aquaporin channel expression in astrocytes of the mouse optic nerve (ON) and the response to IOP elevation in mice lacking aquaporin 4 (AQP4 null). Methods C57BL/6 (B6) and AQP4 null mice were exposed to bead-induced IOP elevation for 3 days (3D-IOP), 1 and 6 weeks. Mouse ocular tissue sections were immunolabeled against aquaporins 1(AQP1), 4(AQP4), and 9(AQP9). Ocular tissue was imaged to identify normal AQP distribution, ON changes, and axon loss after IOP elevation. Ultrastructure examination, cell proliferation, gene expression, and transport block were also analyzed. Results B6 mice had abundant AQP4 expression in Müller cells, astrocytes of retina and myelinated ON (MON), but minimal AQP4in prelaminar and unmyelinated ON (UON). MON of AQP4 nulls had smaller ON area, smaller axon diameter, higher axon density, and larger proportionate axon area than B6 (all p≤0.05). Bead-injection led to comparable 3D-IOP elevation (p = 0.42) and axonal transport blockade in both strains. In B6, AQP4 distribution was unchanged after 3D-IOP. At baseline, AQP1 and AQP9 were present in retina, but not in UON and this was unaffected after IOP elevation in both strains. In 3D-IOP mice, ON astrocytes and microglia proliferated, more in B6 than AQP4 null. After 6 week IOP elevation, axon loss occurred equally in the two mouse types (24.6%, AQP4 null vs. 23.3%, B6). Conclusion Lack of AQP4 was neither protective nor detrimental to the effects of IOP elevation. The minimal presence of AQP4 in UON may be a vital aspect of the regionally specific phenotype of astrocytes in the mouse optic nerve head.

Glaucoma is the leading cause of irreversible blindness worldwide. Elevated intraocular pressure (IOP) is one of the major risk factors for glaucoma onset and progression, and available pharmaceutical interventions are exclusively targeted at IOP lowering. However, degeneration of retinal ganglion cells (RGCs) may continue to progress despite extensive lowering of IOP. A complementary strategy to IOP reduction is the use of neuroprotective agents that interrupt the process of cell death by mechanisms independent of IOP. Here, we describe an ion complexation approach for formulating microcrystals containing ~50% loading of a protein kinase inhibitor, sunitinib, to enhance survival of RGCs with subconjunctival injection. A single subconjunctival injection of sunitinib-pamoate complex (SPC) microcrystals provided 20 weeks of sustained retina drug levels, leading to neuroprotection in a rat model of optic nerve injury. Furthermore, subconjunctival injection of SPC microcrystals also led to therapeutic effects in a rat model of corneal neovascularization. Importantly, therapeutically relevant retina drug concentrations were achieved with subconjunctival injection of SPC microcrystals in pigs. For a chronic disease such as glaucoma, a formulation that provides sustained therapeutic effects to complement IOP lowering therapies could provide improved disease management and promote patient quality of life.