Human ESC differentiation to RPE
Tampere University has National Supervisory Authority for Welfare and Health (Dnro 1426/32/300/05) approval to conduct research on human embryos. The institute also has supportive statements of the Ethical Committee of the Pirkanmaa Hospital District to derive, culture, and differentiate hESC lines (Skottman/R05116). No new cell lines were derived for this study.
Human ESC line Regea08/017 was cultured and differentiated to RPE as described previously . Briefly, hESCs were detached with TrypLE™ Select into suspension in xeno-free differentiation medium (XF-Ko-SR) containing KnockOut DMEM supplemented with 15% KnockOut SR XenoFree CTS, 2 mM GlutaMAX, 0.1 mM 2-Mercaptoethanol, 1% MEM Non-Essential Amino Acids Solution, and 50 U/ml Penicillin-Streptomycin (all from Thermo Fisher Scientific). Embryoid body (EB) formation was induced overnight by addition of 10 µM blebbistatin (Sigma-Aldrich, Saint Louis, MO, USA). Following EB formation, a 3-day neuroectodermal induction was performed with 10 µM SB-505124 hydrochloride hydrate (Sigma-Aldrich) and 10 µM IWP-2 (Merck Millipore), after which the EBs were plated down to 0.75 µg/cm2 laminin-521 (Biolamina) and 10 µg/cm2 human placental collagen Type IV (Sigma-Aldrich) in XF-Ko-SR medium. Pigmented foci were selected and replated, and hESC-RPE stocks were cryopreserved at passage 3. For transplantation 250,000 hESC-RPE cells/cm2 were thawed on laminin-521 and collagen Type IV coated PET inserts containing 1 µm pores (Merck Millipore). The cells were cultured in XF-Ko-SR medium for 47 ± 8.6 days (mean ± SD) prior to live cell shipment and transplantation.
Authentication of hESC-RPE
Human ESC-RPE authentication was performed as previously described . Briefly, transepithelial electrical resistance (TEER) was triplicate measured with Millicell volt-ohm meter (Merck Millipore) . Key RPE protein expression and localization was verified with indirect immunofluorescence labeling for zonula occludens-1 (ZO-1), claudin-3, claudin-19, sodium–potassium adenosine triphosphatase (Na+/K+-ATPase), bestrophin, and MER Proto-Oncogene, tyrosine Kinase (MERTK). Enzyme-linked immunoassay (ELISA) for pigment epithelium-derived factor (PEDF) was carried out from apical and basal media collected after overnight incubation and analyzed with the Human PEDF ELISA kit (BioVendor) following manufacturer’s instructions. Phagocytosis assay was conducted with porcine photoreceptor outer segments (POS) by 4 hours apical incubation at 37°C in the presence of 10% fetal bovine serum (Thermo Fisher Scientific), followed by labeling with anti-rhodopsin antibody and tetramethylrhodamine (TRITC). Nuclei were counterstained with DAPI included in ProLong Gold mounting medium (Thermo Fisher Scientific). Images were acquired with an LSM 700–800 Confocal microscope (Carl Zeiss) and processed with the Zen 2.3 SP1 Black software (Carl Zeiss). All primary and secondary antibody details appear in Table S1 (Additional file 1).
Cell shipment and viability testing
Temperature controlled (+ 15 to + 25°C) live shipment of the hESC-RPE grafts was arranged via World Courier. The grafts were shipped as intact PET inserts placed in conical 50 mL tubes (Falcon Centrifuge Tubes, Corning) in Gibco Hibernate A medium (Thermo Fisher Scientific) supplemented with 15% KnockOut SR XenoFree CTS, 2 mM GlutaMAX, and 50 U/ml penicillin-streptomycin (all from Gibco, Thermo Fisher Scientific). The graft-containing tubes were placed in a thermally insulated box with heated (+ 37°C) gel pads and temperature monitoring. RPE morphology, expression/polarization of RPE marker proteins and TEER were monitored before shipment at the manufacturing site (Finland) and after arrival at the transplantation site (Singapore). Morphology was examined with phase contrast microscopy (Nikon Instruments Europe and Carl Zeiss Meditec) and TEER was measured with a Millicell volt-ohmmeter as described above. Identical measurement systems were used at both locations. At their destination, hESC-RPE inserts were placed in a cell culture incubator in XF-Ko-SR medium for at least 2 days recovery prior to surgery. Immunostainings for claudin, Na+K+-ATPase, bestrophin, and MERTK were performed as described in “authentication of hESC-RPE” (before shipment) and “immunolabeling of tissue sections” (after shipment).
Thirteen Cynomolgus monkeys (Macaca fascicularis) (body weight 3.0–6.0 kg, 4–6 years old) were sourced from SingHealth Experimental Medicine Center, Singapore. All animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of SingHealth (Singapore). Bilateral surgery (n = 1) was only allowed in succession and if the previously operated eye was assumed to have regained good visual function (assessed by animal behavior and the absence of significant structural damage on multimodal imaging). In accordance with this IACUC regulation, animals scheduled for both eyes were planned for sham surgery in the first eye and submacular hESC-RPE implantation in the second (unoperated) eye. All animals were handled in accordance with the Association for Research in Vision and Ophthalmology (ARVO) statement for the use of animals in Ophthalmic and Vision Research and performed in American Animal Association LAC (AAALAC) International approved facility.
Human ESC-RPE graft transplantation
A total of 13 NHPs were utilized in this study. Eleven animals had hESC-RPE grafts transplantation, and 2 animals underwent sham surgery alone (native RPE was removed but not followed by RPE graft implantation); one animal had bilateral surgery with a sham procedure in the right eye and hESC-RPE transplantation in the left eye, making a total of 14 eyes (Table 1). Immunosuppression, general anesthesia and transplantation were performed as previously described . In brief, a 25-gauge (G) trans pars plana vitrectomy was performed using a Stellaris PC (Bausch & Lomb, Singapore) or Constellation (Alcon, Singapore) vitrectomy machine and a surgical microscope equipped with microscope integrated intraoperative optical coherence tomography (miOCT, OPMI-Lumera 700, C. Zeiss Meditec, Singapore). The subretinal injection by a 38 G subretinal cannula (Cat #3247, MedOne Surgical Inc., Sarasota/ FL, USA) was performed either by an automated injection system connected to the vitrectomy machine (Constellation, Alcon) or manually. Surgical removal of submacular RPE (ca. 2 x 3 mm) was achieved with a 20 G custom extensible loop instrument at elevated intraocular pressure (IOP) . A custom-built, subretinal implant shooter instrument enabled subretinal implantation of the RPE graft (bullet-shape, 1.1 x 2 mm) . The remaining hESC-RPE material was immediately fixed with 10% formalin similarly serving as quality control for the shipment. Air-fluid exchange was then performed via active extrusion using a brushed silicone soft tip cannula (Cat #3222, MedOne Surgical Inc., Sarasota/ FL, USA), and included gentle subretinal fluid drainage from the bleb retinal detachment and retinotomy edge apposition.
In vivo animal follow-up
Postoperative transplantation site follow-up was monitored non-invasively by spectral domain OCT (SD-OCT) (Spectralis®, Heidelberg Engineering, Heidelberg, Germany). Imaging of the retina via a confocal scanning laser ophthalmoscope (blue fundus autofluorescence/ BAF, infrared reflectance/ IR and fluorescein angiography/ FA) was achieved using the same device. All animals had the above ophthalmic imaging at baseline and postoperative days 5, 14, 28. SD-OCT images were assessed by two masked independent graders (summarized in the Table S2, Additional file 1).
Retinal function was assessed by full-field electroretinography (ERG) using an Espion system (Diagnosis LLC, USA) prior to and 28 days post-surgery using protocols based on those recommended for human patients by the International Society for Clinical Electrophysiology of Vision (ISCEV) , but with a light adapted (LA) flash strength of 5.0 cd.s.m− 2 (LA 5.0) [21, 22], Animals were anesthetized and pupils dilated, followed by 20 minutes dark-adaptation prior to full-field ERG recording.
After 28 days, animals were sacrificed in deep intramuscular anesthesia with an intracardiac injection of the euthanizing agent (Phenobarbital, VALABARB®, Jurox Pty Limited), followed by perfusion fixation via carotid artery with 4% formaldehyde or Davidson’s fixative medium (13% formaldehyde, 15% ethanol, 5% glacial acetic acid) . Six of 13 animals had both eyes enucleated; the remaining animals had only the operated eye removed. Entire globes were immersed in the same fixative medium overnight. This resulted in 11 transplanted eyes, 3 surgical (sham) control eyes and 5 naïve eyes being processed for histological analysis. Following removal of the anterior segments, full thickness samples (3 x 2 mm, retina→sclera) were cut and embedded in paraffin (standard histological processing). Sections were serially cut at 5 µm thickness with a microtome (Leica RM2255) and stained.
The paraffin sections were stained for evaluation of retinal integrity with hematoxylin and eosin (H&E) using standard protocols and imaged with Hamamatsu NanoZoomer S60 WSI scanner (Hamamatsu Photonics). H&E stained sections containing macula were imaged for quantification of the outer nuclear layer (ONL) nuclei number with a Nikon Eclipse TE2000-S phase contrast microscope (Nikon Instruments Europe B.V.) using 10× objective and the nuclei counted from 3–9 sections per animal. Three corresponding measurement sites of a 100 µm in length were chosen from each section and nuclei manually counted with ImageJ cell counter plugin .
Immunolabeling of tissue sections
Samples were deparaffinated using standard protocols. Heat induced epitope retrieval was performed either with 10 mM sodium citrate, 0.05% Tween-20, pH 6 or 10 mM Tris (Table S1, Additional file 1) in preboiled buffer for 30 minutes at room temperature (RT). After blocking with 10% normal donkey serum (Merck Millipore) and 5% bovine serum albumin (Sigma-Aldrich) for 1 hour at 37°C, samples were incubated overnight with primary antibodies at 4°C, and secondary antibodies for 1 hour at RT (Table S1, Additional file 1). Prior to mounting with ProLong Gold containing DAPI (Thermo Fisher Scientific), lipofuscin autofluorescence was quenched with TrueBlack (Biotium) according to manufacturer's instructions. Sections not incubated with primary antibodies acted as controls. Images were taken using Olympus IX51 fluorescence microscope (Olympus).
Data are presented as mean ± SD. Comparisons of non-parametrically distributed data were performed using two-tailed Mann–Whitney U test with GraphPad Prism 5 Software (La Jolla, CA, USA, www.graphpad.com). Differences were considered significant at *P < 0.05, **P < 0.01, and ***P < 0.001.