The production of specialized retinal cells from pluripotent stem cells is a turning point in the development of stem cell-based therapy for retinal diseases. The present paper describes a simple method for an efficient generation of retinal organoids and retinal pigmented epithelium for basic, translational, and clinical research.
The production of specialized cells from pluripotent stem cells provides a powerful tool to develop new approaches for regenerative medicine. The use of human-induced pluripotent stem cells (iPSCs) is particularly attractive for neurodegenerative disease studies, including retinal dystrophies, where iPSC-derived retinal cell models mark a major step forward to understand and fight blindness. In this paper, we describe a simple and scalable protocol to generate, mature, and cryopreserve retinal organoids. Based on medium changing, the main advantage of this method is to avoid multiple and time-consuming steps commonly required in a guided differentiation of iPSCs. Mimicking the early phases of retinal development by successive changes of defined media on adherent human iPSC cultures, this protocol allows the simultaneous generation of self-forming neuroretinal structures and retinal pigmented epithelial (RPE) cells in a reproducible and efficient manner in 4 weeks. These structures containing retinal progenitor cells (RPCs) can be easily isolated for further maturation in a floating culture condition enabling the differentiation of RPCs into the seven retinal cell types present in the adult human retina. Additionally, we describe quick methods for the cryopreservation of retinal organoids and RPE cells for long-term storage. Combined together, the methods described here will be useful to produce and bank human iPSC-derived retinal cells or tissues for both basic and clinical research.
The retina is an integral part of the central nervous system (CNS) and has a limited capacity to spontaneously regenerate following a traumatic injury or diseases. Therefore, degenerative pathologies causing definitive retinal cell loss, such as age-related macular degeneration (AMD), retinitis pigmentosa (RP), glaucoma, and diabetic retinopathy, typically lead to irreversible blindness. Rescuing the degenerated retina is a major challenge for which stem cell-based therapies aiming to replace the damaged or lost cells are one of the most promising approaches1,2,3. Pluripotent stem cells as human embryonic stem cells (ESCs) cells or human-induced pluripotent stem cells (iPSCs) have the capacity to be expanded indefinitely in culture, and they have the potential to produce any cell types. Advances in our understanding of retinal development and the improvement of in vitro protocols for human iPSC differentiation have resulted in the generation of retinal organoids7,8,9,10,11,12. All of the major retinal cells, including retinal ganglion cells (RGCs), photoreceptors, and retinal pigmented epithelial (RPE) cells, have been successfully differentiated from human ESCs and iPSCs4,5,6. Based on the SFEB (serum-free culture of embryoid body-like aggregates) method developed by Eiraku et al.13, self-formation of retinal organoids can be obtained from ESC- or iPSC-derived embryoid body-like aggregates in defined extracellular matrix components7,10,14. But these protocols are intricate, requiring a large number of steps not always compatible with the large production of cells for therapeutic approaches or drug screening. Thus, the choice of the method to produce human retinal cells is critical and the method needs to be robust, scalable, and efficient.
Here, based on our previous publication15, we describe each step for a simple and efficient generation of retinal cells through retinal organoid self-formation from adherent human iPSCs cultivated in a feeder-free and xeno-free condition. Starting from routine cultures of adherent human iPSCs, this protocol requires only a simple successive medium changing to allow the generation of both iPS-derived RPE (hiRPE) cells and neuroretinal structures in 4 weeks. After a manual isolation, hiRPE can be expanded and the retinal structures can be cultured as floating organoids where the retinal progenitor cells are able to differentiate into all retinal cell types in a sequential order consistent with the in vivo human retinogenesis. Finally, for research advancement or clinical translation, we describe a cryopreservation method allowing the long-term storage of whole retinal organoids and hiRPE cells without affecting their phenotypic characteristics and functionality.
The protocol described in this paper follows the guidelines of the Institut de la Vision's research ethics committee. The Institut de la Vision has been allowed the manipulation of human specimen according to the current French regulation. Specimen handling follows patient data protection in accordance with the Tenets of Helsinki, and national regulations after the ethical approval of the "Comité de Protection des Personnes (CPP) Ile-de-France V".
1. Preparation of Culture Media and Dishes
2. Maintenance and Expansion of Human iPSCs
3. Generation of Retinal Organoids
4. Maturation of Retinal Organoids
5. Generation and Amplification of Human iPSC-derived RPE (hiRPE) Cells
6. Cryopreservation of Retinal Organoids and hiRPE Cells
7. Thawing of Retinal Organoids and hiRPE Cells
The first step for human iPSC differentiation cultivated in feeder-free conditions16 is to shut down self-renewal machinery using Bi medium to encourage a spontaneous differentiation (Figure 1A). Then, at D2, the Bi medium is complemented with an N2 supplement to guide differentiating iPSCs cells towards the neural and retinal lineages. This process leads to the appearance of neuroretinal buds at around D28 (Figure 1C – 1E). Self-forming neuroretinal structures can be isolated using a needle as illustrated in Figure 1E and transferred to culture plates to allow the maturation of the retinal organoids in floating culture conditions using ProB27 medium (Figure 1F). To favor growth and development of the neural retina, FGF2 is added to the medium for 1 week (Figure 1A).
At D28, the emerging retinal structures contain mainly retinal progenitor cells which co-express the key transcription factor as PAX6, RAX, and VSX215. These progenitors give rise to the seven major classes of retinal cell types in floating culture conditions in an evolutionarily conserved birth order consistent with human retinal development. Based on qRT-PCR and immunohistochemistry previously described in Reichman et al.15, the broad curves in Figure 2 show waves of early- and late-born retinal cell generations during an in vitro maturation process. Thus, the culture time defines the cell types present in the organoids.
The isolation of hiRPE cells for any further amplification can be performed only when pigmentation is perceptible because this cell coloration allows their visualization. In this way, the ProN2 medium is used from D28 to D42 to favor the pigmentation of RPE cells15.
Depending on the human iPSC clone, the pigmented patches can be detected before or after the self-formation of retinal structures; but mostly 1 or 2 weeks after the use of ProN2 medium at D28 (Figure 3A, B). A representative bright field image of hiRPE cells expanded from the isolated patches at passage 0 is shown in Figure 3C and 3D. Then, the hiRPE cells can be expanded until passage 4, retaining their RPE phenotype without an epithelial-mesenchymal transition (EMT). Nevertheless, EMT can be prevented by the use of ROCK inhibitors such as Y-27632, allowing also an increase of the cell number passages17. Long-term cultures of hiRPE cells after thawing can be easily performed in the conditions described here to obtain a mature and functional epithelium15. An example of mature hiRPEp2 cells at week 52 with a classical cuboidal cobblestone morphology is illustrated in Figure 3E.
Figure 1: Generation and maturation of retinal organoids from adherent human iPSCs. (A) Differentiation protocol allowing the generation of retinal organoids. (B) Human iPSCs at D0. (C) Emerging neuroretinal epithelium at D15. (D) Self-forming neural retina-like structures at D22. (E) Here, the neuroretinal bud is isolated using a needle. (F) Representative images of retinal organoids in floating culture condition at D35. Scale bars = 200 µm. Please click here to view a larger version of this figure.
Figure 2: Waves of human iPSC-derived retinal cell generation. Abbreviations: Retinal progenitor cells (RPCs), retinal ganglion cells (RGCs), horizontal cells (Ho), amacrine cells (Am), Müller glial cells (MGCs), bipolar cells (Bp), and photoreceptors (PR). Please click here to view a larger version of this figure.
Figure 3: Generation and amplification of human iPSC-derived RPE. (A) An illustration of the differentiation protocol allowing the generation of human hiRPE cells. (B) Phase-contrast images showing pigmented patches emerging from differentiating adherent human iPSCs at Week 6 (W6). (C) hiRPEp0 cells cultivated one week (W1) after the pigmented patch picking. (D) hiRPEp0 cells at week 6 (W6) after the pigmented patch picking. (E) A representative image of hiRPEp2 cells after thawing, cultivated for 52 weeks (W52) in ProN2 medium. Scale bars = 200 µm. Please click here to view a larger version of this figure.
This protocol describes how to produce RPE cells and retinal organoids, containing retinal RGCs and photoreceptors, from human pluripotent stem cells in xeno-free and feeder-free conditions. Compatible with the Good Manufacturing Practice (GMP) process, the method cultivated presented here allows a large production of iPSC-derived retinal cells as RPE cells, RGCs, and photoreceptors for the development of stem cell-based therapies and drug discovery approaches for the future treatment of retinal degenerative diseases. The cryopreservation of whole retinal organoids or hiRPE cells also provides a major advantage in establishing intermediate cell banks, an important step for future use in stem cell-based therapies.
The production of large stocks of specific retinal cell types at a specific stage of differentiation will be required for future clinical translation. In this regard, the generation in three months of CD73-positive photoreceptor precursors described as a transplantation-compatible cell population18, and the possibility to generate these immature photoreceptors from freeze-thawed retinal organoids15 reinforce the hope to use these cells for therapeutic purposes. Concerning retinal pigmented epithelium, the ability of hiRPE cells to proliferate in vitro allows large cell productions to bank them. Importantly, thawed human iPSC-derived RPE cells retain their RPE phenotype and function, therefore, as trophic factor secretion or photoreceptor outer segment phagocytosis15, validating their ease of use for screening strategies as well as for future therapeutic approaches.
There are a variety of protocols for the generation of iPSC-derived retinal organoids7,8,9,10,11,12,15 that vary in culture methods (embryoid body-like aggregates vs. adherent cells) as well as efficiency and robustness. The method described here starts from adherent human iPSCs and shows a reproducible efficacy, adaptability, and applicability to a wide range of human iPSC lines15. This process, based on the successive change of serum-free media, recapitulates the main steps of retinal development by exploiting the intrinsic cues of the system to guide differentiation. An important advantage of this protocol is the absence of embryonic body formation and the addition of matrix for future GMP-compliant retinal cell manufacturing protocols to produce cell therapy derivatives. In this way, no difference in the efficiency of the retinal organoid generation and maturation were found between xenogeneic and no-xenogeneic culture conditions using Cell Therapy System (CTS) supplements or not, formulated exclusively with recombinant or humanized components.
The success of the retinal differentiation method largely depends on the quality of the human iPSC cultures. The reprogramming method does not influence the differentiation efficiency of human iPSCs to retinal cells8, but their stemness status need to be optimal. Briefly, routinely cultivated human iPSCs should not show any signs of differentiation. The colonies should not overlap and must display their characteristic circular morphology. Although the efficiency of the retinal differentiation is clone-dependent, a minimum of two retinal structures per cm2 can be picked at D28, corresponding to 50 – 60 neuroretinal structures for one 6-cm dish. For the retinal organoid maturation in floating culture conditions, limiting the number of structures per well avoids structure fusion and medium overconsumption. In these culture conditions, retinal organoids can be maturated for an extensive time, required to obtain late retinal cell types.
Looking forward, retinal organoids generated in vitro by this method constitute powerful tools to model retinal diseases. Patient-specific iPSC-derived retinal cell models will be used to better understand complex or genetic diseases by the exploration of their molecular and cellular mechanisms. These models will be particularly suitable for drug discovery through high-throughput screening, cell and gene therapies, or genome editing approaches, to develop innovative treatments for retinal dystrophies.
The authors have nothing to disclose.
The authors would like to thank the members of Goureau's team for their input during the set-up of the methods described here, and G. Gagliardi and M. Garita for their critical reading. This work was supported by grants from the ANR (GPiPS: ANR-2010-RFCS005; SightREPAIR: ANR-16-CE17-008-02), the Retina France Association and the technology transfer company SATT Lutech. It was also performed in the frame of the LABEX LIFESENSES (ANR-10-LABX-65) supported by the ANR within the Investissements d'Avenir program (ANR-11-IDEX-0004-02).
Vitronectin (VTN-N) Recombinant Human Protein, Truncated | ThermoFisher Scientific | A14700 | Coating |
CTS Vitronectin (VTN-N) Recombinant Human Protein, Truncated | ThermoFisher Scientific | A27940 | Coating |
Essential 8 Medium | ThermoFisher Scientific | A1517001 | medium |
Essential 6 Medium | ThermoFisher Scientific | A1516401 | medium |
CTS (Cell Therapy Systems) N-2 Supplement | ThermoFisher Scientific | A1370701 | supplement CTS |
N-2 Supplement (100X) | ThermoFisher Scientific | 17502048 | supplement |
B-27 Supplement (50X), serum free | ThermoFisher Scientific | 17504044 | supplement |
CTS B-27 Supplement, XenoFree | ThermoFisher Scientific | A1486701 | supplement CTS |
DMEM/F-12 | ThermoFisher Scientific | 11320074 | medium |
MEM Non-Essential Amino Acids Solution (100X) | ThermoFisher Scientific | 11140035 | supplement |
Penicillin-Streptomycin (10,000 U/mL) | ThermoFisher Scientific | 15140122 | antibiotic |
CellStart CTS | ThermoFisher Scientific | A1014201 | Matrix CTS |
Geltrex hESC-Qualified, Ready-To-Use, Reduced Growth Factor Basement Membrane Matrix | ThermoFisher Scientific | A1569601 | Matrix |
Gentle Cell Dissociation Reagent | Stemcell Technologies | 7174 | dissociation solution |
Cryostem Freezing Media | clinisciences | 05-710-1D | Cryopreservation medium |
Fibroblast growth factor 2 (FGF2) | Preprotech | 100-18B | FGF2 |
Fibroblast growth factor 2 (FGF2) animal free | Preprotech | AF-100-18B | FGF2 Xeno free |
AGANI needle 23G | Terumo | AN*2332R1 | Needle |
Flask 25 cm² Tissue Culture Treated | Falcon | 353109 | T-25 cm² |
24 well plate Tissue Culture Treated | Costar | 3526 | 24-well plate |
6 well plate Tissue Culture Treated | Costar | 3516 | 6-well plate |