Stem cell culture and maintenance

H9, CRX-GFP H9 (gift from Dr. Majlinda Lako, NewCastle University, United Kingdom), and Nrl-eGFP H9 and CRX-tdTomato H9 (gift from Dr. David Gamm, University of Wisconsin-Madison, WI, United States of America) cell lines were maintained on Geltrex coated plates in Essential 8 (Thermofisher Scientific, Mississauga, ON, Canada), or mTESR+ (STEMCELL Technologies, Vancouver, ON, Canada) medium and passaged at 75–80% confluence using ReLeSR (STEMCELL Technologies, Vancouver, BC, Canada). Routine mycoplasma testing was performed using the MycoAlert Mycoplasma Detection kit (Lonza, Basel, Switzerland).

Retinal organoid differentiation

Retinal organoids were differentiated according to previously established protocols, with some modifications [32, 33, 37, 38]. Media formulations are listed in Additional file 1: Table S1. After reaching 90–95% confluence in Essential 8, Essential 6 (Thermofisher Scientific, Mississauga, ON, Canada) was used to initiate the differentiation period for 2 days. The cells were then fed with proneural induction medium (pNIM) for 3 weeks, during which laminated optic vesicle structures would appear. A P200 pipette tip was used to scrape an approximate 4 mm grid pattern before using a cell scraper to detach the retinal organoids after 21–28 days. The detached organoids were allowed to settle to the bottom of the falcon tube by gravity (~ 3–5 min) and washed 3 times in DMEM to remove single cells and debris. The organoids were then cultured in retinal initiation media (RIM) in poly-HEMA (Sigma Aldrich, Mississauga, ON, Canada) coated plates. The organoids were separated from clumps or nonretinal tissues using 2 sterile needles. Organoids were denoted as 4 weeks old after transitioning to suspension culture. Retinal maturation media (RMM) was supplemented with 10% FBS and taurine at week 6 (RMM-RA), with retinoic acid (1 μM) added to the media at week 10 (RMM + RA). The concentration of retinoic acid was reduced to 0.5 μM at week 12 and the media was further supplemented with 1% N-2 supplement (Final Retinal Maturation Media). Media changes were performed 2–3 times per week.

Immunohistochemistry

Organoids were fixed in 4% paraformaldehyde (PFA) for 20 min, washed in PBS, and immersed in 30% sucrose overnight. Samples were embedded in OCT and frozen in isopentane cooled with dry ice. Cryosections of 20 μm were collected for immunohistochemistry. Sections were permeabilized in PBS with 0.3% Triton-X for 15 min and blocked in 10% donkey serum diluted in PBS with 0.1% Triton-X for 1 h at room temperature. Primary antibodies (listed in Additional file 1: Table S2) were diluted in 1% donkey serum with 0.1% Triton-X overnight at 4 °C. Slides were washed with PBS + 0.1% Triton-X (PBST) 3 times to remove unbound primary antibody. Secondary antibodies were diluted in 1% donkey serum + 0.1% Triton-X in PBS and incubated for 1 h at room temperature in the dark. The sections were washed 3 times in PBST, counter-stained with Hoechst 3342 at 1:1000 dilution in PBS (Cell Signaling Technology, Danvers, MA, USA) for 11 min, washed an additional three times in PBS, and mounted with ProLong Gold Antifade (Thermofisher Scientific, Mississauga, ON, Canada).

RNA extraction, cDNA synthesis, and qRT-PCR

Organoids were washed once in PBS and lysed in RA1 buffer from the Nucleospin RNA II Kit (Macherey–Nagel, Düren, Nordrhein-Westfalen, Germany) according to the manufacturer’s instructions. Samples were either frozen at − 80 °C immediately or processed for RNA extraction. Lysates were sonicated for 3 min at room temperature and then further processed according to kit instructions. An additional DNase removal step was performed for 15 min at RT. Purified RNA was stored at − 80 °C. For cDNA synthesis, 250 ng of RNA was used per 20 μl reaction using the Superscript VILO synthesis cDNA kit (Invitrogen, Waltham, Massachusetts, USA). RNA used for cDNA synthesis had minimum 260/280 values ranging from 1.8 to 2.0.

For 10 μl qRT-PCR reactions using Sso Advanced SYBR Green (Bio-Rad Laboratories, Hercules, California, USA), 2 μl of cDNA (diluted 1:5) was added to 1 μM of forward and reverse primer mix. qRT-PCR was performed using the QuantStudio 6 Flex Real Time-PCR System (Thermofisher Scientific, Mississauga, ON, Canada) for 40 cycles using an annealing temperature of 55 °C. Primers sequences (Additional file 1: Table S3) were optimized in silico and validated using a 4-point standard curve to ensure that efficiencies were between 85 and 115%. Primer specificity was determined by the presence of a single melt curve or single PCR product resolved on an agarose gel. Cycle threshold (Ct) values were normalized to GAPDH as a reference gene (ΔCt) and normalized undifferentiated H9 cells as a reference sample (ΔΔCt). Fold change was calculated by the 2−(ΔΔCt) method.

Organoid dissociation

Organoids were dissociated using the Worthington Papain Dissociation Kit (Worthington Industries, Columbus, OH, USA) as per the manufacturer’s instructions. After 40–50 min of incubation in papain, organoids were triturated into a single cell suspension and neutralized with media. Cells were pelleted at 300g for 10 min and resuspended in a solution of Earle’s Balanced Salt Solution with ovomucoid inhibitor and DNase I. The cell suspension was gently layered on top of 5 ml of ovomucoid inhibitor, and the ovomucoid gradient was centrifuged at 60g for 10 min, with the acceleration/deceleration switched off, to remove debris and dead cells. The cell pellet was resuspended in PBS and passed through a 40 μm cell strainer. Viability and cell counts were assessed by trypan blue. For mouse in vitro cultures, retinas were dissected from postnatal day 3–5 mice in cold CO2-independent media and dissociated using the Worthington Papain Dissociation Kit using the same protocol.

Cell sorting

Organoids were dissociated as described above. Cells were resuspended into FACS buffer (2% BSA, 25 mM HEPES, 300 U/ml DNase I in PBS) and strained through a 40 μm cell strainer. 7-AAD (1:50 dilution) was used for live/dead discrimination. After gating for single live cells on the BD Aria III, Aria IIIu, and Melody, photoreceptors were gated using GFP. A blank BV421 or APC channel was used to gate out weak auto-fluorescent cells. Cells were sorted into cold collection buffer consisting of 50% DMEM and 50% FBS. After FACS, cells were pooled, washed with PBS, and centrifuged at 300g for 15 min. Viability and cell counts were confirmed using trypan blue and a hemocytometer prior to cell transplantation.

Cell transplantation

Animal work was approved by the University Health Network Animal Care Committee (protocol 3499), the Canadian Council on Animal Care guidelines and the guidelines set by the Association for Research in Vision and Ophthalmology (ARVO). All animal experiments adhered to the ARRIVE guidelines. Male and female Nrl−/−, C57BL/6J, and NSG mice (6–16 weeks) were used as transplant recipients. Immune competent recipient animals, Nrl−/− and C57BL/6J, (Additional file 1: Table S4) were given an intraperitoneal (IP) injection of cyclosporine at 30 μg/g bodyweight diluted in sterile 0.9% NaCl (Bioshop, Burlington, ON, Canada) for immunosuppression every day for 2 days prior to the transplantation day and continuing for 7 days more, after which cyclosporine added to the drinking water for the remainder of the study.

Animals were anesthetized by an IP injection of 50 mg/kg ketamine (Ketalean, 8KET004D, Bimeda MTC Animal Health Inc. Cambridge, ON, Canada) and 1 mg/kg medetomidine (Cepetor, 236 1506 0, Modern Veterinary Therapeutics LLC, Miami, FL, USA) prepared in sterile 0.9% NaCl. The pupils were dilated with 1% tropicamide (Mydriacyl, 0065-0355-03 Alcon, Mississauga, ON, Canada) and eyes were kept lubricated with Systane Gel (Alcon). The eye was immobilized with a custom latex dam and a plastic coverslip was used to view the fundus under a microscope. A sharp 30 G needle was used to make an incision in the sclera to allow for entry with a 33 G blunt needle. Once the blunt needle was positioned correctly in the subretinal space, a puncture in the cornea was made to relieve intraocular pressure. A total of 1 μl was delivered to the subretinal space (1. 75 × 105 cells/eye) and the needle remained in position for an additional 30 s post-delivery to reduce the chance of cell reflux. For recovery, animals were returned to their cages that were warmed on a heating pad and given an IP injection of atipamezole of 1 mg/kg (Revertor, 236 1504 0, Modern Veterinary Therapeutics LLC, Miami, FL, USA).

Immune competent animals (Nrl−/− and C57BL/6J) received a 1 μl intravitreal injection of triamcinolone acetonide (40 mg/ml) after cell delivery. Systemic cyclosporine injections (IP) were administered daily for 7 days after transplantation (30 μg/g bodyweight), followed by oral cyclosporine administered at 200 μg/ml in the drinking water for the remainder of the study, which was replaced weekly until tissue harvest (Neoral, Novartis, Cambridge, MA, United States).

Tissue harvest

Animals were euthanized with an overdose of sodium pentobarbital and perfused transcardially with PBS and then 4% PFA. After enucleation and corneal puncture, eyes were further post-fixed in 4% PFA for 2 h at RT, washed 3 times in PBS, and left in 30% sucrose dissolved in PBS at 4 °C overnight. The eyes were equilibrated in a 1:1 solution of 30% sucrose: OCT for 1 h before embedding. OCT blocks were flash frozen and stored at − 80 °C until sectioning. Retinal sections were cut at 20 μm thickness and processed with the same immunohistochemistry protocol as stated above. Quantifications of GFP+ donor cells were performed in serial sections, where GFP + donor cells in the subretinal space were quantified in every 5th slide. Integration was calculated by normalizing the number of GFP+ cells that were quantified in the ONL with the number of GFP + cells in the SRS. Transplant recipients with less than 50 donor cells in the SRS were excluded from analysis.

In vitro co-culture

Human retinal organoids (Nrl-eGFP H9 and CRX-tdTomato H9) and retinas isolated from postnatal day 3 to 5 mice (Nrl::GFP) were dissociated as described earlier using papain and resuspended at 1 × 106 cells/ml in final retinal maturation media. For co-culture, 1 × 105 cells/well of each population were seeded into a 96-well plate. Media changes were performed every 3 days on the in vitro retinal dissociates.

To harvest the cells for flow cytometry, papain was used to lift the cells (40–50 min at 37 °C with trituration). Media was added to neutralize the dissociation and the cells were then centrifuged at 300g for 10 min in a 96-well V-bottom plate to pellet the cells. After washing with flow staining buffer (0.5% BSA, 0.05% sodium azide in PBS), UV-Zombie fixable viability dye (1:800, diluted in PBS) was used to stain dead cells at room temperature for 10 min. The cells were washed twice, resuspended in flow buffer, and kept in the dark until ready for analysis. The samples were analyzed on the day of harvest at each time point using Nrl::GFP, Nrl-eGFP H9 and CRX-tdTomato H9 cells that were cultured separately and mixed 1:1 right before data acquisition as a baseline control.

Mitotracker staining and co-culture

To stain for Mitotracker, cells were incubated in 100 nM MitoTracker™ Red (Thermofisher Scientific), diluted in DMEM, at 37 °C for 30 min, after which the cells were pelleted by centrifuging at 300g for 15 min. The stained cells were washed in PBS and spun down at 300g for 15 min, 3 times. For the final wash, the cells were washed in DMEM and after centrifugation, the supernatant was collected to add to unstained cells as a control to ensure that there was no free MTR dye that could incorporate into the cells during the in vitro culture (known as the wash control). The cells were harvested on day 3 and stained with UV-zombie fixable viability dye. After staining, the samples were washed in FACS buffer twice and fixed in 4% PFA at room temperature, for 10 min. The cells were washed 3 times in FACS buffer and stored at 4 °C in the dark until flow acquisition.

Flow cytometric analysis

Data was acquired on the BD LSR Fortessa X-20 and analyzed using FlowJo 10.7.1. The cell population of interest was identified through forward area (FSC-A) and side area (SSC-A) scatter. Doublets were excluded by forward scatter height (FSC-H) and forward scatter width (FSC-W) followed by side scatter height (SSC-H) vs side scatter width (SSC-W). Live cells were gated by the UV-zombie fixable viability dye, and biaxial plots were generated with MTR versus CRX-GFP. For cocultures with photoreceptor-specific fluorescent reporters, one recipient population reporter line (Nrl-eGFP+) was gated first from the bulk population of live cells, followed by the donor population reporter (tdTomato+) to quantify the number of double positive photoreceptors. Gating was set using the FMO (fluorescence-minus one) controls and the day 0 stained samples to establish baseline fluorescence.

Imaging

Images were acquired on an Olympus FV1000, Zeiss LSM 780 or 880 point scanning confocal microscope using 20X air, 40X air and oil, and 63X oil immersion lenses. For quantification of donor cells after transplantation, whole sections were scanned on the Axioscan slide scanner. Live imaging of dissociated retinal organoids was performed on the Zeiss AxioObserver Z1 spinning disk confocal equipped with an incubator to maintain a constant temperature of 37 °C, 5% CO2, and humidity.

Statistical analysis

Data is presented as mean ± SEM. Statistical tests were performed in Graphpad Prism 9.1.1. For comparisons made between two groups, a t test was performed. For a comparison made between two or more groups, a one-way ANOVA was performed with Tukey’s post-hoc. Comparisons with more than one independent variable were analyzed using a two-way ANOVA with Sidak’s post-hoc.