Controlling donor and newborn neuron migration and maturation in the eye through microenvironment engineering

Significance The “in silico–in vitro–in vivo” funnel holds significant potential for identifying targets to control cellular processes in research and clinical applications. In this report, we describe a framework for identifying, selecting, and applying chemokines to direct retinal neuron migration in vivo within the adult mouse retina. To reach a broader audience, we demonstrate this phenomenon using dissociated mouse and human stem cell–derived retinal neurons, retinal organoids, and endogenously reprogrammed retinal neurons. Last, we show that only neurons that migrate into their proper lamina after transplantation express mature cell markers, indicating the importance of driving structural integration for neuron transplantation. Abstract Ongoing cell therapy trials have demonstrated the need for precision control of donor cell behavior within the recipient tissue. We present a methodology to guide stem cell–derived and endogenously regenerated neurons by engineering the microenvironment. Being an “approachable part of the brain,” the eye provides a unique opportunity to study neuron fate and function within the central nervous system. Here, we focused on retinal ganglion cells (RGCs)—the neurons in the retina are irreversibly lost in glaucoma and other optic neuropathies but can potentially be replaced through transplantation or reprogramming. One of the significant barriers to successful RGC integration into the existing mature retinal circuitry is cell migration toward their natural position in the retina. Our in silico analysis of the single-cell transcriptome of the developing human retina identified six receptor-ligand candidates, which were tested in functional in vitro assays for their ability to guide human stem cell–derived RGCs. We used our lead molecule, SDF1, to engineer an artificial gradient in the retina, which led to a 2.7-fold increase in donor RGC migration into the ganglion cell layer (GCL) and a 3.3-fold increase in the displacement of newborn RGCs out of the inner nuclear layer. Only donor RGCs that migrated into the GCL were found to express mature RGC markers, indicating the importance of proper structure integration. Together, these results describe an “in silico–in vitro–in vivo” framework for identifying, selecting, and applying soluble ligands to control donor cell function after transplantation.

Approximately 3.5% of the world’s population over 40 has glaucoma, the most common optic neuropathy (1). By 2040, the global prevalence of glaucoma will exceed 110 million people as the population ages (2), making it a high-priority target for therapy development. Although several neuroprotective approaches focused on preserving existing cells and their axons are being explored (3, 4), this strategy alone will not restore vision already lost due to cell death. The mammalian retina has a limited capacity to regenerate; thus, retinal neuron death leads to irreversible vision loss (5). Retinal ganglion cell (RGC) replacement is needed to recover sight loss to glaucoma. RGC replacement remains an unsolved challenge in regenerative ophthalmology. Success would help bring vision back to millions of advanced-state glaucoma patients. Recent advancements enable transplantation of primary rodent RGCs (6–8), differentiation of RGCs from human pluripotent stem cells (9, 10), reprogramming Müller glia to RGCs (11), and functional axon regeneration to the brain suggest that transplantation-mediated vision repair may be feasible (12). We demonstrated the robust survival of induced pluripotent stem cell (iPSC)-derived RGCs following intravitreal transplantation into healthy and damaged retinas (13). These studies became possible due to robust RGC differentiation and isolation from iPSC cultures established in our lab (14–16). Despite those successes, the survival rate for individual RGCs remains low, and most donor neurons remain above the inner limiting membrane that defines the neural retinal border with the vitreous cavity without integrating (17). The highly conserved organization of the retina across species suggests a relationship between retinal structure and function (18). This challenge is not unique to RGC replacement. Poor structural integration is a significant obstacle to neuron transplantation (e.g., photoreceptors, dopaminergic, and motor neurons) and transdifferentiation (e.g., glia to cerebral and retinal neurons) (11, 19). There are nonsurgical techniques for inner limiting membrane disruption and removal of this existing anatomical barrier (20, 21); however, integration critically depends on the modulation of the host’s adult retinal environment, and inner limiting membrane disruption may not be appropriate in a clinical setting (22). We hypothesize that early guided migration can significantly improve the structural and functional integration of donor and newborn RGCs. Neurogenesis in the mammalian retina completes shortly after birth (23–25). Thus, the transplantation of stem cells and their progeny relies on the recapitulation of the development and/or regenerative pathways. During development, RGCs, like most early-born neurons, migrate via somal translocation, but confocal traces in zebrafish demonstrate that RGCs migrate through multipolar migration if somal translocation is inhibited (26). Multipolar migration does not rely on the extension and attachment of neural processes to reach their final location and is the preferred migratory mode for late-born neurons that navigate through developed tissues (27, 28). It is unknown whether RGCs are capable of multipolar migration in mammals, and the exact mechanism by which stem cell–derived donor RGCs migrate within the mature retina remains unknown. It is also unclear whether newborn and stem cell–derived RGCs can respond to chemokines, previously identified in the studies of retinal and cerebral development and neuron migration in the brain out of the subventricular zone during postnatal regeneration (29). Here, we describe a framework to identify, select, and apply chemokines to direct cell migration in vivo within the retina. We performed an in silico analysis of the single-cell transcriptome of developing human retinas and identified six receptor-ligand candidates to guide stem cell–derived or newborn neurons. The lead candidates were then tested in the functional in vitro assays for their ability to guide stem cell–derived RGCs, with stromal cell–derived factor-1 (SDF1) identified as the most potent chemokine for RGC recruitment. For this and other experiments, we differentiated RGCs from mouse and human stem cells using retinal organoid cultures or stimulated glial reprogramming to neurons with proneuronal transcription factors (11, 30, 31). We then transplanted these stem cell–derived RGCs subretinally and delivered recombinant SDF1 protein intravitreally to establish a chemokine gradient across the retina. Using a quantitative approach to transplantation, we confirmed that donor cell behavior is controllable by modulating the tissue microenvironment. Furthermore, we demonstrate that an SDF1 gradient across the host retina enhances the structural integration of mouse and human stem cell–derived donor RGCs via multipolar migration. Interestingly, only RGCs that successfully integrated into the ganglion cell layer (GCL) were shown to express RNA-binding protein with multiple splicing (RBPMS), a mature RGC marker. Last, we demonstrate that intravitreal delivery of SDF1 increases the displacement of newborn RGCs out of the inner nuclear layer and toward their natural connecting points in the retina. Altogether, this is a demonstration of the universal nature and applicability of neurokine-directed controlled migration of donor stem cell–derived and endogenously regenerated neurons. Moreover, the established workflow to identify microenvironment modifiers can be ported to control other aspects of donor neuron behavior. Results and Discussion Donor Stem Cell–Derived RGCs Fail to Migrate into the Retina Spontaneously. We and others have demonstrated the feasibility of cell replacement therapy with RGCs isolated from the developing retina and stem cell–derived RGCs. Our grafts survived in healthy and damaged retinas following xeno- and allo-transplantation and sent projections into the optic nerve (13). Cell survival does not equal functional integration; therefore, better structural integration is needed to achieve vision restoration. While more than 60% of recipients have donor RGCs at 2 wk to 1 y in our syngeneic transplantation study (13), the proportion of surviving cells typically remains low (<5%), and even fewer cells migrate completely into the GCL (<1%). Recently, Zhang et al. showed that stem cell–derived human RGCs cultured on the neuroretina explant could not migrate through the inner limiting membrane to graft into the GCL, and disruption of this barrier resulted in increased structural integration (20). However, while it is possible to degrade or surgically remove the inner limiting membrane, these procedures pose a significant risk of damaging the retina (22). Moreover, the inner limiting membrane has been shown to play an essential role in the proper lamination of the retina (32). Therefore, we sought to explore subretinal delivery as an alternative approach for transplanting donor neurons while leaving the inner limiting membrane completely intact. Subretinal delivery has proven an effective strategy for gene, photoreceptor, and retinal pigment epithelium delivery. However, unlike these transplantation paradigms, where the cells are delivered directly to their final position, RGCs must migrate more than 200 μm into the GCL. While our previous work suggests that the subretinal space supports donor RGC survival, we did not observe any migration toward the GCL using this approach (13). During development, RGCs are born on the apical surface and migrate toward the basal side into what will become the GCL, but we do not yet know whether this is due to an intrinsic RGC capacity or a response to the developmental retinal microenvironment. To investigate this phenomenon, we delivered mouse stem cell–derived RGCs using intravitreal and subretinal approaches to study the positions of the donor cells within the neural retina. Two weeks posttransplantation, mice were euthanized according to the Schepens Eye Research Institute Institutional Animal Care and Use Committee guidelines, and retinas were stained and mounted to access donor RGC distribution (Fig. 1A). We observed no spontaneous migration through the inner limiting membrane for RGCs delivered intravitreally (Fig. 1B) but demonstrated limited migration through the neural retina for RGCs delivered subretinally (Fig. 1C). Despite having a limited capacity for spontaneous migration following subretinal cell delivery and very few RGCs integrating into the GCL, these results suggest that the subretinal delivery route is a viable alternative to intravitreal delivery so that RGCs can circumvent the inner limiting membrane. Moreover, because spontaneous migration only occurred by mimicking development with our subretinal delivery approach, we hypothesize that neuron migration in the retina must be driven by a response to the developmental microenvironment.

Author(s): Jonathan R. Soucy https://orcid.org/0000-0003-1236-1279, Levi Todd https://orcid.org/0000-0003-2561-7675, Emil Kriukov, +4, and Petr Baranov

*Experimental Paper of the Month manager:** Anthony Khawaja

Editorial Board: Humma Shahid, Karl Mercieca, Francisco Goni

Editors in Chief: Francesco Oddone, Manuele Michelessi