Propensity of Patient-Derived iPSCs for Retinal Differentiation: Implications for Autologous Cell Replacement.

Prior to use, newly generated induced pluripotent stem cells (iPSC) should be thoroughly validated. While excellent validation and release testing assays designed to evaluate potency, genetic integrity, and sterility exist, they do not have the ability to predict cell type-specific differentiation capacity. Selection of iPSC lines that have limited capacity to produce high-quality transplantable cells, places significant strain on valuable clinical manufacturing resources. The purpose of this study was to determine the degree and root cause of variability in retinal differentiation capacity between cGMP-derived patient iPSC lines. In turn, our goal was to develop a release testing assay that could be used to augment the widely used ScoreCard panel. IPSCs were generated from 15 patients (14-76 years old), differentiated into retinal organoids, and scored based on their retinal differentiation capacity. Despite significant differences in retinal differentiation propensity, RNA-sequencing revealed remarkable similarity between patient-derived iPSC lines prior to differentiation. At 7 days of differentiation, significant differences in gene expression could be detected. Ingenuity pathway analysis revealed perturbations in pathways associated with pluripotency and early cell fate commitment. For example, good and poor producers had noticeably different expressions of OCT4 and SOX2 effector genes. QPCR assays targeting genes identified via RNA sequencing were developed and validated in a masked fashion using iPSCs from 8 independent patients. A subset of 14 genes, which include the retinal cell fate markers RAX, LHX2, VSX2, and SIX6 (all elevated in the good producers), were found to be predictive of retinal differentiation propensity.

Development of a Molecularly Stable Gene Therapy Vector for the Treatment of RPGR-Associated X-Linked Retinitis Pigmentosa.

In a screen of 1,000 consecutively ascertained families, we recently found that mutations in the gene RPGR are the third most common cause of all inherited retinal disease. As the two most frequent disease-causing genes, ABCA4 and USH2A, are far too large to fit into clinically relevant adeno-associated virus (AAV) vectors, RPGR is an obvious early target for AAV-based ocular gene therapy. In generating plasmids for this application, we discovered that those containing wild-type RPGR sequence, which includes the highly repetitive low complexity region ORF15, were extremely unstable (i.e., they showed consistent accumulation of genomic changes during plasmid propagation). To develop a stable RPGR gene transfer vector, we used a bioinformatics approach to identify predicted regions of genomic instability within ORF15 (i.e., potential non-B DNA conformations). Synonymous substitutions were made in these regions to reduce the repetitiveness and increase the molecular stability while leaving the encoded amino acid sequence unchanged. The resulting construct was subsequently packaged into AAV serotype 5, and the ability to drive transcript expression and functional protein production was demonstrated via subretinal injection in rat and pull-down assays, respectively. By making synonymous substitutions within the repetitive region of RPGR, we were able to stabilize the plasmid and subsequently generate a clinical-grade gene transfer vector (IA-RPGR). Following subretinal injection in rat, we demonstrated that the augmented transcript was expressed at levels similar to wild-type constructs. By performing in vitro pull-down experiments, we were able to show that IA-RPGR protein product retained normal protein binding properties (i.e., analysis revealed normal binding to PDE6D, INPP5E, and RPGRIP1L). In summary, we have generated a stable RPGR gene transfer vector capable of producing functional RPGR protein, which will facilitate safety and toxicity studies required for progression to an Investigational New Drug application.

Generation of an immortalized human choroid endothelial cell line (iChEC-1) using an endothelial cell specific promoter.

Age-related macular degeneration (AMD) is a common cause of blindness worldwide. While recent studies have revealed that the loss of choroidal endothelial cells (ChECs) is critical to the disease pathogenesis of dry AMD, in vitro studies are needed to fully elucidate the disease mechanism. However, these studies remain hindered due to the lack of publically available human ChEC lines. To address this need, ChECs were harvested form donor tissue and enriched for by using magnetic cell separation using anti-CD31 conjugated microbeads. Next, lenti-viral vectors with endothelial-specific promoters driving genes necessary for immortalization, CDH5p-hTERT and CDH5p TAg, were generated. Stable integration of both gene cassettes allowed cells to maintain their proliferative state and yielded an immortalized cell line (iChEC-1). Immunocytochemical analysis of iChEC-1 confirmed the expression of important ChEC markers such as CA4, a marker of choriocapillaris endothelial cells, CDH5, and CD34, pan-endothelial cell markers. qRT-PCR analysis of expanded clones from iChEC-1 further showed that the line maintained expression of other important endothelial markers, vWF, PECAM1, and PLVAP, similar to primary cells. Functional responses were characterized by tube-forming assays and repopulation of decellularized choroid with the immortalized cell line. In conclusion, the iChEC-1 line presents a suitable immortalized human ChEC line for future in vitro studies of AMD.