Using Stem Cells to Rebuild the Outer Neural Retina

Unlike the neural retina of amphibians and many lower vertebrates, the human retina is largely incapable of endogenous regeneration. For this reason, death of the light-sensing rod and cone photoreceptor cells, a hallmark of retinal degenerative diseases such as age-related macular degeneration and retinitis pigmentosa, results in irreversible vision loss. One of the most promising approaches currently under development for the treatment of retinal degenerative blindness is stem cell–based photoreceptor cell replacement. Experimentally, a variety of different donor cell types—ranging from embryonic and induced pluripotent stem cells to fetal-derived retinal progenitors and postmitotic photoreceptor precursors—have been used with varying degrees of success.¹ Although the postmitotic photoreceptor precursor cell has been demonstrated to be at the ideal point in retinal development for optimal cellular integration and posttransplant maturation into new functional photoreceptor cells,² for a variety of ethical, immunologic, and logistical reasons it is difficult to envision a strategy in which these cells could be directly isolated from postmortem donors and transplanted into patients. With that in mind, pluripotent stem cells, which can be used to generate photoreceptor precursors in vitro prior to transplantation, seem the most logical option. 

In a study published in this issue of IOVS, Santos-Ferreria and colleagues³ report subretinal transplantation of embryonic stem cell–derived photoreceptor precursor cells that were generated using a novel three-dimensional retinal differentiation protocol, purified via anti-CD73–based magnetic bead immunopanning, and identified post transplant via expression of a rhodopsin–green fluorescent protein reporter.³ Unlike previous transplant studies using models of rod-selective disease, in this study two different models of cone–rod dystrophy were used: a prominin1-deficient mouse (Prom1^−/−) with mild, slowly progressive disease, and a novel rhodopsin-deficient congenic mouse (Cpfl1/Rho^−/−, developed for this study) exhibiting a severe phenotype that includes cone loss.³ Following transplantation in the Prom1^−/− model the authors convincingly demonstrate migration of donor cells into the host outer nuclear layer, morphologic photoreceptor cell maturation, and synaptic integration.³ In contrast, when transplantation experiments were performed in the Cpfl1/Rho^−/− model of end-stage disease, in which the structural support provided by the host outer nuclear layer was no longer present, donor cells that expressed mature rod-specific markers were identified as a disorganized mass within the subretinal space.³ Likewise, examples of synaptic connectivity between the remaining host retina and donor cells were rare in this model.³ Although the Prom1^−/− data presented were encouraging, it is highly unlikely that patients with a near-full-thickness photoreceptor cell layer would ever be treated using a stem cell–based photoreceptor cell replacement approach. For patients like this, therapeutic approaches designed to slow or prevent extant photoreceptor cells from dying would be more appropriate. Rather, it is more likely that patients with late-stage disease, such as that seen in the Cpfl1/Rho^−/− mouse model, would be seeking stem cell–based treatments. Models like this will be very useful to scientists seeking to develop effective stem cell–based photoreceptor replacement therapies.