Final Report Summary - RET-IPSC (Exploiting the power of human induced pluripotent stem cells to generate synthetic retinae in vitro for cell based therapies, drug discovery and disease modelling)
Blindness represents an increasing global problem with social and economic impacts both for the patients and society in general. In Europe, approximately 1 in 30 individuals experience sight loss and 75% of those are unemployed, a social burden which is very likely to increase as the population of Europe ages. Novel early diagnostic tests and preventive treatments are key themes of the new “Vision 2020” strategies, but for those to be successful, a detailed understanding of the causes of blindness and development of cell replacement strategies are necessary. Diseases affecting the retina, the light sensitive extension of the central nervous system lining the back of the eye, account for approximately 26% of global blindness. To date, there are no treatments to restore lost retinal cells and visual function and there is an urgent need for new therapeutic approaches. There has been great excitement on a pioneering breakthrough from Sasai lab which supports generation of retinal organoids from human pluripotent stem cells. The research performed during the course of this ERC consolidator fellowship has enabled optimisation of two different methods of retinal organoid formation and has shown that both can be applied to a large number of pluripotent stem cells derived from unaffected subjects and patients with inherited and age related macular dystrophies. Our data show some variability in the efficiency of retinal organoid generation between human induced pluripotent stem cell (hiPSC) lines; nonetheless by month 5 of differentiation, retinal organoids composed of multiple retinal phenotypes, reminiscent of human foetal retina, and which generate light responses, can be generated from all lines. To understand and control process variability and increase the efficiency of 3D laminated retina from hiPSC, statistical modelling of the differentiation processes was carried out. This indicated that the differentiation process was highly dependent on seeding cell density and nutrient availability and enabled optimisation of conditions for production of retinal organoids from a large number of hiPSC. To enable large-scale toxicology and pharmacology studies we adapted the differentiation protocol to a multiwell plate format. This enhanced generation of retinal organoids with retinal pigment epithelium (RPE) and improved ganglion cell development and the response to physiological stimuli. We tested the response of hiPSC-derived retinal organoids to a known retinotoxin (Moxifloxacin) and showed that similar to in vivo adult mouse retina, the primary affected cell types were photoreceptors. Together our data indicate that light responsive retinal organoids derived from carefully selected and differentiation efficient hiPSC lines can be generated at the scale needed for pharmacology and drug screening purposes.
The extracellular matrix (ECM) plays an important role in numerous processes including cellular proliferation, differentiation, migration, maturation, adhesion guidance and axonal growth. To date, there has been no detailed analysis of the ECM distribution during retinal ontogenesis in humans and the functional importance of many ECM components is poorly understood. During the ERC funded study, the expression of key ECM components in adult mouse and monkey retina, developing and adult human retina and retinal organoids derived from human pluripotent stem cells was studied. Our data indicate a stage specific expression of ECM components during human retinal ontogenesis which was largely conserved in hiPSC-derived retinal organoids. By performing functional blocking experiments, we were also able to show that three key ECM components, namely Laminin Ƴ3, IMPG1 and CD44 play a critical role in many aspects of retinal organoid formation including their laminar organisation, photoreceptor development, interphotoreceptor matrix formation and retinal ganglion cell survival. To further improve the retinal tissue development within the hiPSC-derived retinal organoids, four different hydrogels (0.5% RGD-alginate; 1% RGD-alginate; hyaluronic acid (HA) or HA/gelatin) were tested. Of these, encapsulation of retinal organoids with 0.5% RGD-alginate hydrogel resulted in the most optimal retinal tissue formation, which highlights potential for the clinical application of alginate hydrogels for the derivation and subsequent transplantation retinal tissue. The interaction between neural retina and the adjacent retinal pigmented epithelium (RPE) is well documented in the literature and shown to be beneficial in a range of ex vivo culture system. To investigate this further, we supplemented the culture media of differentiating pluripotent stem cell derived retinal organoids with media conditioned by RPE cells, decellularised matrix from adult retina as well as RPE. Our data show that decellularised matrix obtained from adult retina and RPE enhances synapse formation and light responses of pluripotent stem cell derived retinal organoids.
To be able to properly assess the development of stem cell derived retinal organoids at the molecular level we need to fully understand human retinal development itself. The protracted period of retinal development (4th post conception week till five postnatal months) and the scarcity of human embryonic and fetal material, has meant that most of the insights gained in retinal development were based on animal models. Whilst this has been extremely valuable and has guided developmental and stem cell biology in many fronts, there are substantial differences with regard to the gross anatomy, cellular morphology, retinal microarchitecture, molecular content and gene expression between widely used animal models (mouse and rat) and human retina. Thus we undertook a systematic transcriptional and immunohistochemical study of developing human retina using embryonic and fetal specimens (4th-19th post conception weeks) obtained with appropriate ethics consent from human developmental biology resource. This integrated analysis revealed three developmental windows with specific gene expression patterns, which informed the sequential emergence of retinal cell types and enabled identification of stage-specific cellular and biological processes and transcriptional regulators. Each stage was characterised by a specific set of alternatively spliced transcripts which code for proteins involved in the formation of the photoreceptor connecting cilium, pre-mRNA splicing and epigenetic modifications. Most importantly, the transition from fetal to adult retina was characterised by a large increase in the percentage of mutually exclusive exons, which code for proteins involved in photoreceptor maintenance. The circular RNA population was also defined and found to increase during retinal development. Collectively these data represent a valuable resource for increasing our understanding of human retinal development, pre-mRNA splicing process, staging pluripotent stem cell-derived retinal organoids and identifying new candidate disease genes.
Given the complex nature of retinal organoids and the paucity of established cell surface markers for each of the emerging retinal cell types, it is almost impossible to purify and study each cell type in detail. To this end we have used the most recent gene editing techniques to tag fluorescent proteins to key transcription factors that characterise development of photoreceptor precursors. This approach allows live visualisation of cell types of interest within the retinal organoids as well as fluorescence activated cell sorting. We have applied both of these methods to isolate photoreceptor precursors and using single cell RNA-Seq we have looked into the molecular profile of CRX+ expressing photoreceptor precursors. Our data show that CRX+ cells are comprised of a dominant cell cluster comprising 72% of the cells, which displayed the hallmarks of early cone photoreceptor expression. When transplanted subretinally into animal models of retinal degeneration, the CRX+ cells settled next to the inner nuclear layer, made connections with the inner neurons of the host retina and about one third expressed the pan cone marker, Arrestin 3, indicating further maturation upon integration into the host retina. Following from the successful application of single cell RNA-Seq of purified photoreceptors, we investigated the feasibility of this technique in retinal organoids throughout the differentiation process. This analysis was able to reveal the presence of multiple cell types already identified by immunohistological analysis, which indicates the feasibility and applicability for large-scale studies during retinal development of characterisation of organoids derived from patients with inherited and age related retinal dystrophies. We have also combined our fluorescent tagging approach with bulk and single cell RNA-Seq as well as flow cytometric based cell surface profiling and we identified cell surface markers for photoreceptors, which are currently under validation.
Given the success of our differentiation protocols for generating 3D retinal organoids as well as RPE monolayers, we went on to perform disease modelling studies for the following: (1) Retinitis Pigmentosa caused by mutation in the pre-mRNA splicing factor, PRPF31; (2) Age related macular degeneration caused by polymorphism in the CFH gene; (3) patients with mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes caused by m.3243A > G mitochondrial DNA mutation; (4) X-linked Retinitis Pigmentosa caused by mutations in the RPGR gene. These disease modelling studies which are under different stages of publication process have yielded novel molecular insights into molecular pathology of the disease as well as have created in vitro disease models which can be used to test new therapeutic approaches before large and expensive clinical trials.
The extracellular matrix (ECM) plays an important role in numerous processes including cellular proliferation, differentiation, migration, maturation, adhesion guidance and axonal growth. To date, there has been no detailed analysis of the ECM distribution during retinal ontogenesis in humans and the functional importance of many ECM components is poorly understood. During the ERC funded study, the expression of key ECM components in adult mouse and monkey retina, developing and adult human retina and retinal organoids derived from human pluripotent stem cells was studied. Our data indicate a stage specific expression of ECM components during human retinal ontogenesis which was largely conserved in hiPSC-derived retinal organoids. By performing functional blocking experiments, we were also able to show that three key ECM components, namely Laminin Ƴ3, IMPG1 and CD44 play a critical role in many aspects of retinal organoid formation including their laminar organisation, photoreceptor development, interphotoreceptor matrix formation and retinal ganglion cell survival. To further improve the retinal tissue development within the hiPSC-derived retinal organoids, four different hydrogels (0.5% RGD-alginate; 1% RGD-alginate; hyaluronic acid (HA) or HA/gelatin) were tested. Of these, encapsulation of retinal organoids with 0.5% RGD-alginate hydrogel resulted in the most optimal retinal tissue formation, which highlights potential for the clinical application of alginate hydrogels for the derivation and subsequent transplantation retinal tissue. The interaction between neural retina and the adjacent retinal pigmented epithelium (RPE) is well documented in the literature and shown to be beneficial in a range of ex vivo culture system. To investigate this further, we supplemented the culture media of differentiating pluripotent stem cell derived retinal organoids with media conditioned by RPE cells, decellularised matrix from adult retina as well as RPE. Our data show that decellularised matrix obtained from adult retina and RPE enhances synapse formation and light responses of pluripotent stem cell derived retinal organoids.
To be able to properly assess the development of stem cell derived retinal organoids at the molecular level we need to fully understand human retinal development itself. The protracted period of retinal development (4th post conception week till five postnatal months) and the scarcity of human embryonic and fetal material, has meant that most of the insights gained in retinal development were based on animal models. Whilst this has been extremely valuable and has guided developmental and stem cell biology in many fronts, there are substantial differences with regard to the gross anatomy, cellular morphology, retinal microarchitecture, molecular content and gene expression between widely used animal models (mouse and rat) and human retina. Thus we undertook a systematic transcriptional and immunohistochemical study of developing human retina using embryonic and fetal specimens (4th-19th post conception weeks) obtained with appropriate ethics consent from human developmental biology resource. This integrated analysis revealed three developmental windows with specific gene expression patterns, which informed the sequential emergence of retinal cell types and enabled identification of stage-specific cellular and biological processes and transcriptional regulators. Each stage was characterised by a specific set of alternatively spliced transcripts which code for proteins involved in the formation of the photoreceptor connecting cilium, pre-mRNA splicing and epigenetic modifications. Most importantly, the transition from fetal to adult retina was characterised by a large increase in the percentage of mutually exclusive exons, which code for proteins involved in photoreceptor maintenance. The circular RNA population was also defined and found to increase during retinal development. Collectively these data represent a valuable resource for increasing our understanding of human retinal development, pre-mRNA splicing process, staging pluripotent stem cell-derived retinal organoids and identifying new candidate disease genes.
Given the complex nature of retinal organoids and the paucity of established cell surface markers for each of the emerging retinal cell types, it is almost impossible to purify and study each cell type in detail. To this end we have used the most recent gene editing techniques to tag fluorescent proteins to key transcription factors that characterise development of photoreceptor precursors. This approach allows live visualisation of cell types of interest within the retinal organoids as well as fluorescence activated cell sorting. We have applied both of these methods to isolate photoreceptor precursors and using single cell RNA-Seq we have looked into the molecular profile of CRX+ expressing photoreceptor precursors. Our data show that CRX+ cells are comprised of a dominant cell cluster comprising 72% of the cells, which displayed the hallmarks of early cone photoreceptor expression. When transplanted subretinally into animal models of retinal degeneration, the CRX+ cells settled next to the inner nuclear layer, made connections with the inner neurons of the host retina and about one third expressed the pan cone marker, Arrestin 3, indicating further maturation upon integration into the host retina. Following from the successful application of single cell RNA-Seq of purified photoreceptors, we investigated the feasibility of this technique in retinal organoids throughout the differentiation process. This analysis was able to reveal the presence of multiple cell types already identified by immunohistological analysis, which indicates the feasibility and applicability for large-scale studies during retinal development of characterisation of organoids derived from patients with inherited and age related retinal dystrophies. We have also combined our fluorescent tagging approach with bulk and single cell RNA-Seq as well as flow cytometric based cell surface profiling and we identified cell surface markers for photoreceptors, which are currently under validation.
Given the success of our differentiation protocols for generating 3D retinal organoids as well as RPE monolayers, we went on to perform disease modelling studies for the following: (1) Retinitis Pigmentosa caused by mutation in the pre-mRNA splicing factor, PRPF31; (2) Age related macular degeneration caused by polymorphism in the CFH gene; (3) patients with mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes caused by m.3243A > G mitochondrial DNA mutation; (4) X-linked Retinitis Pigmentosa caused by mutations in the RPGR gene. These disease modelling studies which are under different stages of publication process have yielded novel molecular insights into molecular pathology of the disease as well as have created in vitro disease models which can be used to test new therapeutic approaches before large and expensive clinical trials.