Using a self-organizing method, we develop a protocol with the addition of COCO that could significantly increase the generation of photoreceptors.
Retinal cell transplantation is a promising therapeutic approach, which could restore the retinal architecture and stabilize or even improve the visual capabilities to the degenerated retina. Nevertheless, progress in cell replacement therapy presently faces the challenges of requiring an off-the-shelf source of high quality and standardized human retinas. Therefore, an easy and stable protocol is needed for the experiments. Here, we develop an optimized protocol, based on a self-organizing method with the use of exogenous molecules and reagent A as well as manual excision to generate the three-dimensional human retina organoids (RO). The human Pluripotent Stem Cells (PSCs)-derived RO expresses specific markers for photoreceptors. With the addition of COCO, a multifunctional antagonist, the differentiation efficiency of photoreceptor precursors and cones is significantly increased. The efficient use of this system, which has the benefits of cell lines and primary cells, and without the sourcing issues associated with the latter, could produce confluent retinal cells, especially photoreceptors. Thus, the differentiation of PSCs to RO provides an optimal and biorelevant platform for disease modelling, drug screening and cell transplantation.
Pluripotent stem cells (PSCs) are characterized by their self-renewal and ability to differentiate into all kinds of somatic cells. Thus, organoids derived from PSCs have become an important resource in regenerative medicine research. Retinal degeneration is characterized by the loss of photoreceptors (rods and cones) and retinal pigment epithelium. Retinal cell replacement could be an encouraging treatment for this disease. However, it is not feasible to obtain human retinas for disease research and therapy. Therefore, retinal organoids (ROs) derived from PSCs, which effectively and successfully recapitulate multi-layered native retinal cells, are beneficial for basic and translational research1,2,3. Our research focuses on RO differentiation to provide sufficient and quality cells for studying retinal degeneration4.
Methods for differentiating ROs are continuously emerging, with three-dimensional (3D) suspension differentiation pioneered by the Sasai laboratory in 20125. We introduced the CRX-tdTomato tag in the human embryonic stem cells (hESCs) to specifically track the photoreceptor precursor cells and modified the method with the addition of COCO, a multifunctional antagonist of the Wnt, TGF-β, and BMP pathways6. COCO has been shown to efficiently improve the differentiation efficiency of photoreceptor precursors and cones6,7.
Altogether, by modifying the classical differentiation method, we have developed an accessible protocol to harvest abundant photoreceptor precursors and cones from human ROs for analyzing the retinal disease associated with the photoreceptors through laboratory investigations and for further clinical application/transplantation.
This study was approved by the institutional Ethics Committee of Beijing Tongren Hospital, Capital Medical University. H9 hESCs were obtained from the WiCell Research Institute and genetically engineered to tdTomato-tagged cell line.
1. Generation of human ROs
2. Analyzing human ROs
The schematic illustration depicts the differentiation protocol to improve precursor cells with COCO (Figure 1). From PSC to ROs, numerous details could cause result variations. It is recommended to record every step and even the catalog number and lot number of every medium to track the entire procedure.
Herein, we provide bright field images for days 6, 12, 18, and 45 (Figure 2). On day 6, the organoids are usually around 600 µm in diameter in a 96-well plate, with dense connections inside and bright rim (Figure 2A). On day 12, the optic vesicle-like structures initially generate (Figure 2B). From day 12 to day 18, the presence of optic vesicle structures is clear, and they continued to grow after day 18. The organoids without the vesicle-like architecture are discarded (Figure 2C). By day 30, the vesicle-like architecture is more obvious, and it is easier to distinguish the superior ROs from the inferior ones (Figure 2D-E). The organoids that lose the translucent structure (asterisks in Figure 2D-E), should be removed in the following days.
The organoids express the CRX, which is a marker of photoreceptor precursors, from day 45 onwards (Figure 2F,2I). Other photoreceptor precursor markers, such as RCVRN and OTX2, were also positively detected at day 45 (Figure 2G-H). The addition of COCO promotes the generation of photoreceptor precursors.
Figure 1. Timetable for stepwise treatment for RO differentiation from hESC. Please click here to view a larger version of this figure.
Figure 2. Human retinal organoid generation. (A-B) Early-stage optic vesicle-like structure formed in a 96-well plate. The black arrows indicate the optic vesicle-like structures. (C) The first day of suspension culture in a Petri dish on day 18. (D-E) Optic cup structures are observed at this stage. The stars show the inferior organoids (F) The bright field image on day 45. Scale bars = 400 µm. (G-H) Immunostaining results of RCVRN (G) and OTX2 (H) on day 45. Scale bars = 50 µm. (I) TdTomato-positive signals indicate the expression of CRX on day 45 in (F). Scale bars = 400 µm. Please click here to view a larger version of this figure.
Medium I (50 mL) | ||||||||
KSR | G-MEM | NEAA | Pyruvate | b-ME | IWR1e | COCO | PS | |
Percentage % OR final concentration | 20 | 78 | 0.1 mM | 1 mM | 0.1 mM | 3 µM | 30 ng/mL | 1 |
Volume | 10 mL | 39 mL | 0.5 mL | 0.5 mL | 90.9 µL | 5 µL | 10 µL | 0.5 mL |
The store concentration of IWR1e is 30 mM. COCO is 150 µg/mL. | ||||||||
Medium II (50 mL) | ||||||||
FBS | G-MEM | NEAA | Pyruvate | b-ME | SAG | PS | ||
Percentage % OR final concentration | 10 | 88 | 0.1 mM | 1 mM | 0.1 mM | 100 nM | 1 | |
Volume | 5 mL | 44 mL | 0.5 mL | 0.5 mL | 90.9 µL | 2.5 µL | 0.5 mL | |
The store concentration of SAG is 2 mM. | ||||||||
Medium III (50 mL) | ||||||||
FBS | DMEM/F12- Glutamax | Supplement 1 | RA | Taurine | PS | |||
Percentage % OR final concentration | 10 | 88 | 1 | 0.5 µM | 100 µM | 1 | ||
Volume | 5 mL | 44 mL | 0.5 mL | 5 µL | 50 µL | 0.5 mL | ||
The store concentration of RA is 5 mM, and Taurine is 100 mM. |
Table 1: Medium I, II, and II recipes
Retinal organoid differentiation is a desirable method for the generation of ample functional retinal cells. The RO is a composite of different retinal cells, such as ganglion cells, bipolar cells, and photoreceptors, generated by pluripotent stems cells toward the neural retina4,5,8,9. Although confluent ROs could be harvested, it is time-consuming, which may require long culturing periods (up to 180 days). However, for photoreceptor transplantation, or studying cone-rod or rod-cone dystrophy, it is advantageous to obtain a relatively high percentage of photoreceptors in the 3D culturing system10.
It is also challenging to monitor the organoids’ development without interrupting the normal developmental processes. Therefore, we used CRX, a cone-rod homeobox protein, predominantly expressed in photoreceptor precursors, as a target gene to trace photoreceptor precursor cells during their 3D differentiation. With the tdTomato system, CRX-expressing cells can be spatio-temporally tracked by the red fluorescence without interrupting retinalization during their 3D differentiation. The utilization of CRX-tdTomato system could accelerate the process of drug screening for photoreceptor precursor differentiation in the ROs.
Using our method, around 70% organoids could develop into retinal organoids, which display the vesicle-like structures. Importantly, with the incision of superior organoids, we usually could harvest around 100 retinal organoids from a 96-well plate. Additionally, abundant photoreceptor precursors are generated in the early stage of RO maturation with the COCO culture, which helps progression toward direct differentiation of certain cells through pathway regulation11,12. The protein concentration of reagent A is crucial for the differentiation. Altogether, sufficient reagent A with the aggregates in the early days as well as the cutting of the organoids on day 18 are important to harvest abundant ROs with high quality. This method also promotes the development of directional differentiation of photoreceptor cells in 3D organoids and contributes to the transplantation of photoreceptor cells.
We thank members of 502 laboratory for their technical supports and helpful comments regarding the manuscript. This work was partly supported by the Beijing Municipal Natural Science Foundation (Z200014) and National Key R&D Program of China (2017YFA0105300).
2-mercaptoethanol | Life Technologies | 21985-023 | |
COCO | R&D Systems | 3047-CC-050 | DAN Domain family of BMP antagonists |
DMEM/F-12 | Gibco | 10565-042 | |
DMSO | Sigma | D2650 | |
DPBS | Gibco | C141905005BT | |
EDTA | Thermo | 15575020 | |
Fetal Bovine Serum (FBS), Qualified for Human Embryonic Stem Cells | Biological Industry | 04-002-1A | |
GMEM | Gibco | 11710-035 | |
KnockOut Serum Replacement-Multi-Species | Gibco | A3181502 | |
MEM Non-essential Amino Acid Solution (100X) | sigma | M7145 | |
Pen Strep | Gibco | 15140-122 | |
Primesurface 96 V-plate | Sbio | MS9096SZ | Cell aggregation in 1.2.7 |
Pyruvate | Sigma | S8636 | |
Reagent A | BD | 356231 | Matrigel in 1.1.1 |
Reagent B | StemCell | 5990 | mTeSR- E8 , PSCs basal medium in 1.1.2 |
Reagent C | Gibco | 12563-011 | TrypLE Express in 1.2 |
Reagent D | Roche | 11284932001 | DNase I , in 1.2 |
Retinoic acid | Sigma | R2625-100MG | |
SAG | Enzo Life Science | ALX-270-426-M001 | |
Supplement 1 | Life Technologies | 17502-048 | N-2 Supplement (100X), Liquid, supplemet in medum III |
Taurine | Sigma | T-8691-25G | |
Trypsin-EDTA (0.25%), phenol red | Gibco | 25200056 | organoids dissociation in 2.1.3 |
Wnt Antagonist I, IWR-1-endo – Calbiochem | Sigma | 681669 | Wnt inhibitor |
Y-27632 2HCl | Selleck | S1049 |