To examine the potential of NIH-maintained human embryonic stem cell (hESC) lines TE03 and UC06 to differentiate into retinal progenitor cells (hESC-RPCs) using the noggin/Dkk-1/IGF-1/FGF9 protocol. An additional goal is to examine the in vivo dynamics of maturation and retinal integration of subretinal and epiretinal (vitreous space) hESC-RPC grafts without immunosuppression.
hESCs were neuralized in vitro with noggin for 2 weeks and expanded to derive neuroepithelial cells (hESC-neural precursors, NPs). Wnt ( Integration 1 and wingless) blocking morphogens Dickkopf-1 (Dkk-1) and Insulin-like growth factor 1 (IGF-1) were used to direct NPs to a rostral neural fate, and fibroblast growth factor 9 (FGF9)/fibroblast growth factor-basic (bFGF) were added to bias the differentiation of developing anterior neuroectoderm cells to neural retina (NR) rather than retinal pigment epithelium (RPE). Cells were dissociated and grafted into the subretinal and epiretinal space of young adult (4–6-week-old) mice (C57BL/6J x129/Sv mixed background). Remaining cells were replated for (i) immunocytochemical analysis and (ii) used for quantitative reverse transcription polymerase chain reaction (qRT–PCR) analysis. Mice were sacrificed 3 weeks or 3 months after grafting, and the grafts were examined by histology and immunohistochemistry for survival of hESC-RPCs, presence of mature neuronal and retinal markers, and the dynamics of in vivo maturation and integration into the host retina.
At the time of grafting, hESC-RPCs exhibited immature neural/neuronal immunophenotypes represented by nestin and neuronal class III β-tubulin, with about half of the cells positive for cell proliferation marker Kiel University -raised antibody number 67 (Ki67), and no recoverin-positive (recoverin [+]) cells. The grafted cells expressed eye field markers paired box 6 ( PAX6), retina and anterior neural fold homeobox ( RAX), sine oculis homeobox homolog 6 ( SIX6), LIM homeobox 2 ( LHX2), early NR markers (Ceh-10 homeodomain containing homolog [ CHX10], achaete-scute complex homolog 1 [ MASH1], mouse atonal homolog 5 [ MATH5], neurogenic differentiation 1 [ NEUROD1]), and some retinal cell fate markers (brain-specific homeobox/POU domain transcription factor 3B [ BRN3B], prospero homeobox 1 [ PROX1], and recoverin). The cells in the subretinal grafts matured to predominantly recoverin [+] phenotype by 3 months and survived in a xenogenic environment without immunosuppression as long as the blood–retinal barrier was not breached by the transplantation procedure. The epiretinal grafts survived but did not express markers of mature retinal cells. Retinal integration into the retinal ganglion cell (RGC) layer and the inner nuclear layer (INL) was efficient from the epiretinal but not subretinal grafts. The subretinal grafts showed limited ability to structurally integrate into the host retina and only in cases when NR was damaged during grafting. Only limited synaptogenesis and no tumorigenicity was observed in grafts.
Our studies show that (i) immunosuppression is not mandatory to xenogenic graft survival in the retina, (ii) the subretinal but not the epiretinal niche can promote maturation of hESC-RPCs to photoreceptors, and (iii) the hESC-RPCs from epiretinal but not subretinal grafts can efficiently integrate into the RGC layer and INL. The latter could be of value for long-lasting neuroprotection of retina in some degenerative conditions and glaucoma. Overall, our results provide new insights into the technical aspects associated with cell-based therapy in the retina.