(490b) Modifying the Retinal Microenvironment with Slow-Release Neurotrophic Factors Enhances Donor Retinal Neuron Survival in Mice

The loss or degeneration of retinal ganglion cells (RGCs) in glaucoma is a leading cause of irreversible blindness worldwide, with no current therapies available to restore vision once these cells are lost.¹ While RGCs, the neurons in the retina that relay all visual information from the eye to the brain, can regenerate in some lower vertebrates, they have no capacity for regeneration in the mammalian retina.² Cell transplantation has been proposed as a potential solution to replace RGCs lost in advanced retinal disease, but poor cell survival rates post-transplantation have limited the effectiveness and further development of this strategy.^3–5

The glaucomatous retina is typically characterized by increased interocular pressure (IOP), which can cause inflammation, immune cell infiltration, and host cell reactivity.^6,7 Even after the IOP normalizes this, toxic microenvironment can persist,⁷ making it challenging for transplanted donor cells to survive in the host retina. Further, after transplantation, donor cells encounter an environment that lacks adequate oxygenation and metabolic support until they structurally integrate into the respective positons within host retina. This deficiency can intensify the stress experienced by the donor cells, leading to further decreased survival rates. To overcome these limitations and establish a more favorable microenvironment, we identified and delivered pro-survival factors into the retina to improve donor RGC survival after transplantation.

Methods:

We performed an in silico analysis of the single-cell transcriptome of developing human retinas to determine which molecular features were present during development but drop off during adulthood. We examined changes in gene expression of various neurotropic factors over time and identified Brain-Derived Neurotrophic Factor (BDNF) and Glial Cell-Derived Neurotrophic Factor (GDNF) as potential proneuronal factors to modify the retina microenvironment. In addition to studying their role in reducing apoptosis pathway gene expression, we examined included ligand-receptor interactions using the CellChat package to distinguish between autocrine and paracrine pathways.⁸ Since optic neuropathies lead to RGC dead, no proneuronal autocrine singling supports donor RGCs post-transplantation.

For transplantation experiments, RGCs were differentiated from Thy1-GFP mouse iPSCs and Brn3b-tdTomato human ESCs in 3D retinal organoid cultures.^5,9,10 Mouse and human RGCs were isolated by magnetic microbeads against CD90 on day 21 and day 45 of development, respectively. To support donor RGCs and establish a proneuronal microenvironment, we formulated our mouse or human donor RGCs with selected pro-survival factors. To avoid repeated intraocular injections, which could cause additional inflammation and reduce donor RGC viability, we formulated donor RGCs with slow-release neurotropic factors (GDNF- and BDNF-loaded polyhedrin-based particles (PODs)). Donor RGCs were transplanted (10,000 – 20,000 RGCs/retina) into a RGC-deficient mouse retina (inherited loss or neurotoxic model) with and without PODs. Three to fourteen days after transplantation, whole mounted retinas were stained for host and donor cells to assess RGC integration and survival (8 – 11 mice/group). Eyes were enucleated after three days for human donor RGC transplantation experiments to limit the immune reaction associated with xenotransplantation and fourteen for mouse donor RGCs to allow for neurite outgrowth.

Results:

Our in silico analysis revealed the critical role of BDNF and GDNF as neurotrophic factors during development. These factors are essential during the early stages of RGC development, but they are not present in adulthood (Fig. 1A), indicating we may need to supplement the retina with these factors to establish an environment to support regeneration. Further, the negative correlation between BDNF/GDNF receptor expression and apoptosis pathway gene expression during development indicates that upregulation of these factors can improve RGC survival (Fig. 1B). Lastly, our finding that RGCs are the primary cell type that secretes BDNF and GDNF highlights the importance of replacing these factors after RGC death in glaucoma to establish a more favorable microenvironment for transplantation (Fig. 1C).

Our transplantation experiments demonstrated that delivering BDNF/GDNF-PODS with human and mouse donor RGCs significantly improved the transplantation outcome. The success rate of mouse donor RGC transplantations, defined as a transplant with more than 0.5% of delivered donor RGCs surviving, increased to 73% from 37% with the co-treatment of BDNF/GDNF-PODs. Moreover, the PODs formulation led to a 2.7-fold increase in donor RGC coverage across the retina, with several neurites extending towards the optic nerve head (Fig. 1D-E). This finding is critical as it demonstrates that the transplanted cells have the potential to connect with their postsynaptic targets in the brain, a necessary step for functional recovery after injury.

Although PODs co-treatment did not affect the success rate of human donor RGC transplantation, the PODs formulation led to a 15-fold increase in the total number of human donor RGCs detected in the mouse retina (Fig. 1F-G). Additionally, 50% of all transplants containing PODs showed donor RGC neurites extending toward the optic nerve head, with an average length of 1008.44 ± 264.12 µm compared to 10% in the control group. Critically, these results demonstrate the potential of PODs to improve the survival and integration of transplanted human RGCs, which could lead to functional recovery in patients with retinal degeneration.

Conclusion:

Modulating the host retinal microenvironment with slow-release growth factor co-treatment is an effective tool for improving transplantation outcomes by improving donor RGC morphology and axon outgrowth. While further research is necessary to improve donor RGC integration, using neuroprotective factors in combination with other approaches to modify the retinal microenvironment and stem cell-derived RGCs delivery may offer a viable solution to the challenge of RGC replacement. Depending on the model or state of the disease, the retinal microenvironment may also need to be altered to support donor RGC growth, maturation, axon extension, and increased metabolic demands. Altogether, these findings provide promising insights into potential strategies for restoring vision lost due to retinal ganglion cell damage or loss.

References:

1. Tham, Y.-C. et al. Global Prevalence of Glaucoma and Projections of Glaucoma Burden through 2040 A Systematic Review and Meta-Analysis. Ophthalmology 121, 2081–2090 (2014).

2. Erskine, L. & Herrera, E. Connecting the Retina to the Brain. Asn Neuro 6, 1759091414562107 (2014).

3. Zhang, K. Y., Aguzzi, E. A. & Johnson, T. V. Retinal Ganglion Cell Transplantation: Approaches for Overcoming Challenges to Functional Integration. Cells 10, 1426 (2021).

4. Venugopalan, P. et al. Transplanted neurons integrate into adult retinas and respond to light. Nat Commun 7, 10472 (2016).

5. Oswald, J., Kegeles, E., Minelli, T., Volchkov, P. & Baranov, P. Transplantation of miPSC/mESC-derived retinal ganglion cells into healthy and glaucomatous retinas. Mol Ther - Methods Clin Dev 21, 180–198 (2021).

6. Storgaard, L., Tran, T. L., Freiberg, J. C., Hauser, A. S. & Kolko, M. Glaucoma Clinical Research: Trends in Treatment Strategies and Drug Development. Frontiers Medicine 8, 733080 (2021).

7. Tehrani, S. et al. Astrocyte Structural and Molecular Response to Elevated Intraocular Pressure Occurs Rapidly and Precedes Axonal Tubulin Rearrangement within the Optic Nerve Head in a Rat Model. Plos One 11, e0167364 (2016).

8. Jin, S. et al. Inference and analysis of cell-cell communication using CellChat. Nat Commun 12, 1088 (2021).

9. Wahlin, K. J. et al. CRISPR Generated SIX6 and POU4F2 Reporters Allow Identification of Brain and Optic Transcriptional Differences in Human PSC-Derived Organoids. Frontiers Cell Dev Biology 9, 764725 (2021).

10. Sluch, V. M. et al. Enhanced Stem Cell Differentiation and Immunopurification of Genome Engineered Human Retinal Ganglion Cells. Stem Cell Transl Med 6, 1972–1986 (2017).

Figure 1. Co-treatment with slow-release GDNF and BDNF improves RGC transplantation outcomes. (A) Expression levels of BDNF and GDNF in the retina during different stages of development. (B) Correlation between BDNF receptor expression and apoptotic genes. (C) Chord diagram showing cell-cell communication for the BDNF pathway. (D) Representative retinal whole mount showing mouse donor RGC outgrowth (green) and (E) quantification of coverage area. (F) Representative high magnification images from retinal whole mounts showing human donor RGC outgrowth (red) and (G) quantification of the number of donor RGCs detected in the host retina.