Chemokine-directed migration improves donor neuron integration into the retina in vivo

Introduction: Glaucoma is a leading cause of irreversible blindness worldwide, disproportionately impacting those over 40 and individuals of lower socioeconomic status. This condition, along with other optic neuropathies, results in the gradual loss of retinal ganglion cells (RGCs), the neurons responsible for transmitting visual information from the eyes to the brain. While current treatments primarily aim to reduce intraocular pressure and slow disease progression, they cannot restore lost vision or replace lost RGCs.
Cell transplantation offers a promising avenue for replacing RGCs lost in advanced retinal disease. However, a significant challenge to successful RGC integration into the existing neural circuitry is the migration of cells to their natural position in the retina, the ganglion cell layer (GCL). To address this hurdle and enhance donor neuron integration, we have developed a novel methodology for controlling stem cell-derived neurons in vivo by manipulating their microenvironment post-transplantation.

Materials and

Methods: To identify cues that can be used to modify the retinal microenvironment to control donor neuron behavior, we first analyzed the single-cell transcriptome of the developing human retina in silico. Based on their changes in gene expression and enrichment, we identify six receptor-ligand candidates that play a role in directing neuron migration: ADGRG1, CXCR4, DCC, FZD3, NR2F1, and ROBO2 (Fig 1.1). These lead candidates were then tested in functional in vitro transwell assays for their ability to drive stem cell-derived RGC recruitment. RGCs were differentiated from Brn3b-tdTomato hESC for this and other experiments using retinal organoid cultures. After identifying SDF1 as the most potent chemokine for RGC recruitment (Fig. 1.2), we transplanted human stem cell-derived RGCs subretinally and delivered recombinant SDF1 protein intravitreally to establish a chemokine gradient across the retina (Fig 1.3a).
To investigate the mechanisms of chemoattraction and the modes of donor RGC migration, we applied small-molecule inhibitors to donor RGCs before transplantation for 1) the CXCR4 receptor – AMD3100, 2) somal translocation (ST) – CK666, and 3) multipolar migration (MP) – roscovitine. Retinas were stained for donor and host neurons three days post-transplantation and evaluated by tracking the position of each donor RGC in 3D reconstructions of retinal flat mounts (5 – 9 mice/group).

Results, Conclusions, and Discussions: An average of 546 ± 260 donor RGCs were detected in each neural retina with no statistical differences between groups. Treatment with SDF1 significantly increased donor RGC integration into the GCL from 16 ± 11% to 44 ± 11% (Fig 1.3b). We confirmed the direct effect of SDF1 on donor cells by inhibiting in response to SDF1 by blocking SDF1-CXCR4 binding in donor cells with AMD3100 (17 ± 5.9%). Further, our results show that inhibiting MP migration by roscovitine limits the percentage of donor RGCs that migrate to the GCL in response to SDF1 (No inhibition: 45 ± 15%; MP inhibition: 20 ± 3.7%), whereas inhibiting ST by CK666 does not affect donor RGC migration (ST inhibition: 47 ± 12%), indicate that MP migration is the primary mode of migration for surviving human donor RGCs in the retina. Lastly, we demonstrate that only donor RGCs that migrate into the GCL extend their axons toward the optic nerve head (Fig. 1.3c) and express mature RGC markers – steps critical for RGC replacement therapies to restore vision.
In conclusion, we present a novel strategy for guiding donor RGC integration in the retina through chemokine-directed migration, with SDF1 co-treatment improving the structural integration of donor RGCs 2.7-fold. Furthermore, our study provides valuable insights into the mechanisms of chemoattraction and the modes of donor RGC migration, with our results suggesting that MP migration is the primary mode of migration for surviving human donor RGCs in the retina. We also show the importance of improving donor neuron structural integration, which has implications for restoring lost visual function in glaucoma and other retinal diseases.
Altogether, our findings have significant implications for restoring lost visual function in glaucoma and other retinal diseases and highlight the potential of chemokine-directed migration as a promising approach for future research and clinical applications. In addition, the established workflow using an “in silico—in vitro—in vivo” approach (Fig. 1) to identify microenvironment modifiers can also be adapted to control other aspects of donor neuron behavior, including functional connectivity, within the central and peripheral nervous systems.

Acknowledgements (Optional): NIH/NEI–F32EY033211 (Soucy), NIH/NEI–L70EY034355 (Soucy), NIH/NEI–T32EY007145 (MBED fellowship, Soucy), NIH/NEI–5U24EY029893-03 (Baranov), NIH/NEI–P30EY003790 (Core Facility Grant), Bright Focus Foundation (Baranov), Gilbert Family Foundation (Baranov)