Survival and Functional Integration of Human Embryonic Stem Cell–Derived Retinal Organoids After Shipping and Transplantation into Retinal Degeneration Rats

Abstract

Because derivation of retinal organoids (ROs) and transplantation are frequently split between geographically distant locations, we developed a special shipping device and protocol capable of the organoids' delivery to any location. Human embryonic stem cell (hESC)–derived ROs were differentiated from the hESC line H1 (WA01), shipped overnight to another location, and then transplanted into the subretinal space of blind immunodeficient retinal degeneration (RD) rats. Development of transplants was monitored by spectral-domain optical coherence tomography. Visual function was accessed by optokinetic tests and superior colliculus (SC) electrophysiology. Cryostat sections through transplants were stained with hematoxylin and eosin; or processed for immunohistochemistry to label human donor cells, retinal cell types, and synaptic markers. After transplantation, ROs integrated into the host RD retina, formed functional photoreceptors, and improved vision in rats with advanced RD. The survival and vision improvement are comparable with our previous results of hESC-ROs without a long-distance delivery. Furthermore, for the first time in the stem cell transplantation field, we demonstrated that the response heatmap on the SC showed a similar shape to the location of the transplant in the host retina, which suggested the point-to-point projection of the transplant from the retina to SC. In conclusion, our results showed that using our special device and protocol, the hESC-derived ROs can be shipped over long distance and are capable of survival and visual improvement after transplantation into the RD rats. Our data provide a proof-of-concept for stem cell replacement as a therapy for RD patients.

Introduction

Blindness causes major emotional burden to patients and economic burden to society [1,2]. Blindness resulting from photoreceptor (PR) cell death is currently incurable, an urgent unmet medical need [2–10]. Retinal degeneration (RD) diseases that lead to PR degeneration are the third leading cause of worldwide blindness. Age-related macular degeneration (AMD), a leading cause of RD in people older than 60 years in developed countries, affects ∼15 million people in the United States, with the number projected to double by 2030 [5,6,10]. Retinitis pigmentosa (RP) is the most frequent cause of inherited visual impairment, affecting ∼100,000 people in the United States and 1.5 million people worldwide [9,11–14].

Retinal repair is a major challenge with no currently available solutions, especially in cases of advanced RD. Apart from anti-vascular endothelial growth factor antibody treatment, there are no drugs that can substantially repair retinal damage. Preservation of retinal layered neuroanatomical structure and synaptic networks are needed to maintain visual function.

Restoring PRs is a major therapeutic goal to alleviate blindness. Among all cell replacement therapies, retinal stem cell therapy stands out because it is one of the most technically feasible cell therapies. The ocular space is accessible through transvitreal [4,15] or trans-scleral [16] injection. Furthermore, subretinal grafts can be easily visualized using noninvasive methods [17–19]. Vitreoretinal surgery in large-eye animal models, although technically demanding, is a routine and established procedure, with many ongoing innovations and new tools and equipment described each year [17,18,20–22].

The relative retinal immune privilege enables long-term maintenance of allogeneic (nonautologous) cells and tissues (eg, PRs [4,23,24] and retinal tissue grafts [25,26]). Advances in regenerative medicine enabled generation of three-dimensional (3D) retinal organoids (ROs) from pluripotent stem cells, closely recapitulating the biological complexity and physiology of human retina [27–33]. This created new opportunities for cell replacement therapies and provides us with unlimited supply of clinical-grade retinal cells and tissue for 3D subretinal implants.

Research in the past 20 years indicates that retina can be replaced by fetal tissue (8–17 weeks of human gestation) [25,34]. We have demonstrated that human fetal tissue at this age is almost identical to human embryonic stem cell (hESC)-retinal tissue based on cellular composition and lamination (critical for biological function) [31,35]. A major advantage of retinal replacement technology is that semi-differentiated retinal tissue maintains the original lamination and cell fate commitment when transplanted as a 3D sheet. This happens even in cases when retinal tissues are grafted in the eye with advanced RD [34,36,37] and allows preservation of the developing neuroanatomical circuitry of hESC-retina, which is not the case with PR cells delivered as a suspension of cells into the subretinal space [4,38]. The immature retinal tissue continues differentiation and establishes functional connectivity with the recipient's retinal circuitry and the visual cortex after grafting [39–42].

In vivo validation and refinement of any vision restoration technique in RD animal models is mandatory before clinical trials. We earlier reported vision improvements in two established RD rat models with retinal tissue implants from hESC line CSC-14 [19,43]. Here we expand our work by testing hESC-retina derived from the commonly used hESC line H1 (WA01) [33] and also report the improvement of vision in rats with advanced RD, immunodeficient RhoS334ter-3 rats [44] that have lost most rod PRs at the age of 4 weeks and never developed outer segments [45,46]. Our results validate the safe use of hESC-derived retinal tissue as alternative to human fetal retina to restore vision in eyes with complete PR loss, irrespective of the stem cell source. This work brings us closer to initiating testing of this promising technology in patients with advanced RD.

Materials and Methods

Cell culture

ROs were generated from hESC line H1 (WA01) [31] as we earlier reported [33] and closely resembled ROs previously used for transplantation [18]. hESC colonies were allowed to grow for 6–8 days in mTeSR1, then (when they reached ∼40%–50% density) neuralized using human noggin (100 ng/mL) and neurobasal complete medium (with N2 and B27, as we described [33]). Well-defined retinal structures were dissected out from the differentiating monolayer (at ∼weeks 7–8) using a sharp sterile glass rod pulled from a Pasteur pipette, and further maintained in suspension in six-well plates in an incubator with slow agitation (linear shaker) until reaching the age of 70 days.

Shipping ROs with portable device

Organoids were collected from suspension cultures and placed into a 15-mL culture tube with media that were placed and secured horizontally into the shipping device. We used a battery-driven portable shipper with 0.028 m³ capacity (Uline, Part No. S-20589, Model No. 20589) [47]. Two types of tracking devices were used to track the temperature, G-forces and the GPS location [47]. The device was placed into a FEDEX-approved shipping crate labeled with “UN3373” (Category B biological substance), “do not X-ray” stickers and “up” arrows.

The ROs were shipped at 37°C by Fedex overnight to arrive the next morning (FEDEX Priority overnight) from a stem cell facility in Alameda, CA to UC Irvine without loss of viability (as we previously reported [47]) and prepared for transplantation. The organoids were immediately placed into an incubator upon arrival, followed by examination under a cell culture microscope. Surgery was usually 1–2 days later. Organoids not used for transplantation were fixed and analyzed later. As a batch control for each RO lot used for retinal surgeries, we differentiated a few randomly selected ROs for further characterization at 3–4 months. Immunohistochemistry delineated the distribution of early retinal markers in ROs [33].

Experimental animals

Animals were treated in accordance with NIH guidelines for the care and use of laboratory animals, the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and under a protocol approved by the Institutional Animal Care and Use Committee of UC Irvine. SD-Foxn1 Tg(S334ter)3Lav (RD nude rat) transplant recipients were generated by crossing SD-Tg(S334ter)3Lav rats and NTac:NIH-Whn rats [44].

Retina sheet preparation

ROs (day 70 of differentiation) were selected based on transparency, and morphology of a hollow spherical shape with a laminated structure seen under phase contrast and stereoscope. Structures often contained adherent retinal pigment epithelium aggregates that were removed during the preparation of retinal rectangular sheets (1.0–1.7 × 0.6 mm) for transplantation. The rims of ROs (carrying a layer of human developing PRs) were then dissected for transplantation (Supplementary Fig. S1).

Transplantation

Recipient rats (P27–45, either sex) were randomized into age-matched nonsurgery (AMC) (n = 11), sham (n = 7), and transplant (n = 56) experimental groups.

Rats received a subcutaneous injection of ketoprofen (4 mg/kg) (Parsippany-Troy Hills, NJ) and dexamethasone eye drops (Bausch & Lomb, Inc., Rancho Cucamonga, CA) before anesthesia to prevent eyelid swelling. After ketamine/xylazine anesthesia (40–55 mg/kg Ket, 6.0–7.5 mg/kg Xyl), pupils were dilated by 1% atropine eye drops (Akorn, Lake Forest, IL). Eyes were disinfected with ophthalmic betadine (Alcon, Fort Worth, TX). Nonsurgical eyes were kept moist by artificial tears (Akorn). During surgery, the eye was frequently treated with 0.5% tetracaine (Bausch & Lomb, Inc.) and 0.1% dexamethasone eye drops (Bausch & Lomb, Inc.).

Transplantation of retinal sheets has been previously described [39,43]. In brief, a small incision (∼1 mm) was made posterior to pars plana, parallel to the limbus, followed by local retinal detachment. Using a custom implantation instrument, donor tissue was gently placed into the subretinal space of the left eye (LE). Media alone was injected into the eyes of sham surgery rats. The incision was closed with 10-0 sutures. Eyes were examined by fundoscopy immediately after surgery. For recovery, gentamicin/polymyxin/bacitracin ointment (Bausch & Lomb, Inc.) was applied to the surgical eyes.

Spectral-domain optical coherence tomography imaging

Spectral-domain optical coherence tomography (OCT) imaging (Bioptigen Envisu R2200; Bioptigen, Research Triangle Park, NC) was used to document and monitor transplant development, as described previously [34,39]. Transplanted rats (n = 56) were imaged every 1–2 months, starting 2 weeks after surgery, up to 8.9 months of age [7.5 months postsurgery (MPS)]. Rats with transplant misplacement into the vitreous or excessive surgical trauma were excluded from further analysis. The last scan was scheduled as close as possible to the terminal superior colliculus (SC) recording.

The transplant area was determined by using sequential cross-sectional B-scans to define the host/donor boundary. These positions were then compared with the corresponding locations in the fundus image to outline the transplant (Fig. 2c, f, i). The InVivoVue program (Bioptigen) contains calipers that were used to create several geometric figures to overlay the transplant border.

FIG. 2.

Optical coherence tomography of transplant 1. (a, b) 13 dps. (a) B-scan showing transplant nasal to optic disc. (b) B-scan turned 90° through the whole extent of subretinal transplant. Some blood on retinal surface. Boxes indicate area of enlargement in inserts. Scales = 200 μm. (c) Fundus image at 33 dps with transplant outline (in green). The outline was deducted from corresponding B-scans. The optic disc is indicated by a green dashed circle. (d–f) 116 dps. (d) B-scan showing transplant at optic disc which has also grown onto the retinal surface. (e) B-scan turned by 90°. Whitish area in center of enlargement indicates PR rosette. (f) Fundus image at 116 dps with outline of subretinal transplant (green). The extent of the epiretinal transplant is indicated in pink. (g–i) 157 dps. (g) B-scan showing transplant at optic disc which has also grown onto the retinal surface. (h) B-scan turned by 90°. The yellow arrows indicate PR rosettes. (i) Fundus image at 157 dps with outline of subretinal transplant (in green). The extent of the epiretinal transplant is indicated in pink. dps, days postsurgery.

Optokinetic response testing

After at least 1 h of dark adaption, visual acuity of transplanted, sham surgery, and nonsurgery AMC control rats was measured by recording optomotor responses to a virtual cylinder with alternating black and white vertical stripes at six different spatial frequencies (OptoMotry; Cerebral Mechanics, Alberta, Canada) at 1, 2, and 4 MPS, as previously described [39,43]. Tests were videotaped and evaluated off-line by two independent observers blinded to the experimental condition. The best visual acuity of the two same-day tests was used for analysis. If there was a discrepancy between the two observers, videos were reanalyzed by a third observer. All testers and video watchers were blinded to the experimental group assignment of rats.

SC electrophysiological recording

Visual responses from the SC were recorded as previously described [34,39] after overnight dark adaption, at the age of 6.4–8.9 months (5.1–7.4 months after surgery): multiunit electrical responses from ∼50 locations on the SC surface, 200–400 μm apart, using a tungsten microelectrode (0.5 MΩ impedance; MicroProbe, Inc., Carlsbad, CA). At each location, light stimuli (20 ms, +0.58 to −6.13 log cd/m²) were delivered 10 times at 10-s intervals. When responses were found, intensity of light stimuli was reduced to determine the response threshold. Responses to the strongest light stimuli (stimulus level 0.58 log cd/m²) were quantified into a map over the area of the SC. All spikes occurring 30–210 ms after the onset of the photic stimulus were counted. Spike counts and locations of responses were analyzed using a custom MATLAB program (MathWorks, Natick, MA).

Histology and immunofluorescence

After killing with injection of anesthetic overdose, rats were perfusion fixed with cold 4% paraformaldehyde in 0.1 M sodium–phosphate buffer. After opening the cornea, eyes were postfixed overnight at 4°C and then washed. Eye cups were dissected along the dorsoventral axis, cryoprotected (30% sucrose), and frozen in optimal cutting temperature embedding medium (O.C.T.). compound (Fisher Scientific). ROs were frozen using a similar procedure. Serial 10 μm cryostat sections were stored at −20°C. Every fifth slide was stained using hematoxylin and eosin (H&E) and imaged on an Olympus BXH10 using an Infinity 3-1U camera.

For immunofluorescence and 3′,3′-diaminobenzidine (DAB) analysis, cryostat sections underwent antigen retrieval at 70°C with Histo-VT One (Nacalai USA, Inc., San Diego, CA), followed by phosphate-buffered saline washing, blocking with 20% goat serum, and primary antibodies overnight at 4°C. Slides were incubated for 30–60 min at room temperature in fluorescent or biotinylated secondary antibodies. Primary and secondary antibodies are listed in Supplementary Table S1. Fluorescent sections were coverslipped using VECTASHIELD mounting media (Vector Labs, Burlingame, CA) with 5 μg/mL DAPI (4,6-diamidino-2-phenylindole).

For DAB (3′,3′-diaminobenzidine) staining, sections were incubated with ABC kit (Vector Labs) after the biotinylated secondary antibody and developed with DAB for up to 4 min.

A Zeiss LSM700 or LSM900 confocal microscope (Zeiss, Oberkochen, Germany) was used for imaging fluorescent images (tiled stacks of 5–8 μm thickness at 20 × and 40 × ). Confocal images were extracted in Zen 2012 and 2.3 software (Zeiss). 3D images were extracted separately for each channel and combined in Adobe Photoshop CS6 to Photoshop 2023/2024 software (San Jose, CA). Imaris software (Oxford Instruments) was also used for 3D rendering and to analyze colocalization (closeness defined as a distance of 1 μm or less) (Fig. 7).

FIG. 7.

Evaluating interaction of human synaptophysin processes in the host IPL (Imaris software). Colocalization with Calretinin and rat-specific α-synuclein at 5.3 months after grafting (transplant 1). This rat had a response in the SC. (a) Overview image showing transplant with rosettes; Calretinin = red, human synaptophysin = green, rat-specific α-synuclein = white. (b) Area of enlargement of rat host IPL. (c, d) combination of human synaptophysin (surface) with α-synuclein (spots) and calretinin (surface). (c) Selection of rat host IPL area for analysis. (d) Enlargement, tilted. (e) Human synaptophysin volume (green), with α-synuclein (spots) (close to green [1 μm or less] = magenta; far from green = yellow). (f) Combination of human synaptophysin spots (green) with α-synuclein spots (gray) and calretinin spots (red). (g) Human synaptophysin filaments (green) in combination with α-synuclein spots (magenta) and calretinin spots (red). Colors of spots are differentiated for their location close to or far away from the human synaptophysin green surface. (f, g) Taken from same section as shown in Fig. 6b. Bars = 50 μm.

Statistical analysis

For all statistical analyses, significance level was calculated in GraphPad Prism software (GraphPad Software LLC, La Jolla, CA) with paired and unpaired two-tailed t-tests using mean ± standard error of the mean. Level of significance was set at 0.05.

Results

Some transplanted animals were used only for histology and not for functional tests to investigate transplant development. Some rats were eliminated from analysis after the first or second OCT examination because of faulty surgeries (eg, optic nerve damage, choroid damage, or epiretinal placement of the transplant) or corneal ulcers. Four transplant rats could not be used for the final analysis because they died before recording.

Development and characterization of ROs

An example of organoid development at the time of transplantation is given in Fig. 1. Organoids developed some level of laminated structure. As previously reported, at the time of transplantation, the donor tissue had no mature PRs, as determined by quantitative polymerase chain reaction and histology (data not shown and Fig. 1). The ROs also had no mature cones (Opsin+) or rods (Rhodopsin+) at this stage but contained PR precursors (Crx+) (Fig. 1d).

FIG. 1.

Selected images of young ROs (days ∼65–70) before surgeries (brightfield image (a) and immunohistochemical characterization of selected ROs at day 70). DAPI was used for nuclear counterstaining. (a) Phase contrast image of organoid differentiated for 70 days. Scale = 100 μm. (b) Outer organoid layer expresses transcription factors pax 6 (important for early eye development). (c) Staining for early retinal transcription factor RAX (retinal homeobox protein Rx). (d) Staining for CRX (cone-rod homeobox protein), a transcription factor for PR development. DAPI, 4,6-diamidino-2-phenylindole; PR, photoreceptor; ROs, retinal organoids.

In vivo development of RO transplants monitored by OCT

Rat eyes were imaged by OCT imaging to check initially for the presence of the graft at 2 and 4 weeks post-transplantation, and then for graft expansion and lamination every 2 months up to 6 MPS. Initial transplant size was 0.6–0.84 mm². Transplants matured and developed into rosettes several months post-transplantation (Fig. 2g, h). Examples of OCT images are given in Fig. 2 with arrows pointing out the hyper-reflective PR rosettes within transplants (Fig. 2g, h). Transplant morphology in OCT scans corresponded to that seen in H&E stained sections (Supplementary Fig. S2).

Visual function improvement evaluated by optokinetic testing

Visual acuity significantly decreased with age in nonsurgery AMC RD rats (n = 11) and sham surgery RD rats (n = 6) (Fig. 3). There was no significant difference between LE and right eye (RE) in sham surgery and AMC groups, and no difference between AMC and sham control rats. In addition, the right nonsurgery eyes of the transplanted rats (n = 14) showed the same degree of visual acuity loss as nonsurgery AMC and sham surgery rats. At 1 MPS, the transplanted eyes showed improvement compared with the nonsurgery REs of the same rats but not significant because of low N, whereas the transplanted eyes showed significant improvement compared with nonsurgery AMC and sham surgery rats. This difference became larger at later time points (Fig. 3). The improvement was also significant compared with nonsurgery AMC (P < 0.05; Fig. 3).

FIG. 3.

OKT–visual acuity testing over time (up to 4 months postsurgery) shows significant better visual acuity in the eyes with transplants (red, dotted), but not in their nonsurgery fellow eyes (red, striped), and sham-transplanted (green) or/and age-matched control (blue) eyes. *P < 0.05; **P < 0.01; ***P < 0.001. ns, nonsignificant. OKT, optokinetic testing.

Visual function improvement evaluated by SC electrophysiological recording

After transplantation, at the age of 6.4–8.9 months, 4 of 5 rats showed responses to flashes of light in the SC (Fig. 4 and Supplementary Table S2). Four transplanted rats showed responses to dimmer light (the best light threshold was −0.16 log cd/m² (Supplementary Table S2 and Fig. 4c–f). No responses could be evoked in any SC area of control nonsurgery AMC rats (n = 9) (Fig. 4a) and sham surgery rats (n = 7) (Fig. 4b) at the maximum light level (+0.58 log cd/m²). The vision improvement demonstrated in the transplanted group was significant compared with AMC and sham surgery groups (P < 0.05). Of interest, in one of the transplants, the SC response heatmap showed similar shape as the transplant location in the eye, which suggested that the transplant restore the basic point-to-point projection function from eye to the SC (Fig. 4c).

FIG. 4.

Heat map of visual responses in SC at 5.3–7.5 months after subretinal grafting (Supplementary Table S2): (a) Age-matched nonsurgery control (AMC). (b) Sham surgery shows no responses. (c–f) Four rats with hESC-RO grafts: top row visual response heat maps; bottom row: images of dissected eye cups indicating the position of the transplants (dotted line). hESC, human embryonic stem cell. o.d., optic disc; SC, superior colliculus.

Immunohistochemical analysis of transplant cell type development

After transplantation, the human nuclear marker Ku80 (Fig. 5a–d) and human-specific synaptophysin (Fig. 6) confirmed the distribution of transplanted tissue within the host subretinal space. Some donor cells migrated into the host retina (Fig. 5a–d). We further characterized the 5.1- to 7.4-month hESC-RO grafts in rats with SC responses by an array of immunohistochemical markers to developing and mature PRs (Figs. 5 and 6): ROs produced PRs immunoreactive for recoverin (Figs. 5g, j, k and 6a), Red-Green Opsin+ cones (Fig. 5b, c), Rhodopsin+ rods (Fig. 5d–f, h, i), and organized into rosettes with putative outer segments elongating and maturing over time.

FIG. 5.

PR development in grafts and interaction with RPE. (a–c). Abundance of mature human PRs in grafts. Immunohistochemistry with markers for HNu, in combination with PR marker AIPL1 or cone opsin. (d, e) Abundance of human PRs in grafts (marker rhodopsin). (g–l) Interaction with host RPE (Ezrin). (i) Graft PRs express marker peripherin (PRPH2). *indicate position of host RPE. Bars = 20 μm (a, b, d, e, i, l); = 10 μm (c, f, j). AIPL1, aryl hydrocarbon receptor interacting protein like 1; HNu, human nuclei.

FIG. 6.

Synaptic connectivity graft—host (transplant 1, 160 dps). (a, b) Combination of hSyn (green) with host markers Calretinin (red) and recoverin (RCVRN) [blue, (a)] or rat-specific α-synuclein (blue) (c, d) shows processes in the IPL of the host. Bars = 50 μm. hSyn, human-specific synaptophysin; IPL, inner plexiform layer.

Synaptic connectivity

Transplant and host cells formed synaptic connections as shown by antibodies specific for calretinin, human synaptophysin (synaptic vesicles), in combination with rodent-specific α-synuclein (rodent IPL Fig. 6a, b and Fig. 7). A rodent-specific α-synuclein antibody (Fig. 6) demonstrated intermingling of transplant (human synaptophysin) and host (α-synuclein) processes in the host inner plexiform layer (IPL) (Fig. 7). Imaris image analysis of the host IPL with triple staining for calretinin, human-specific synaptophysin, and rodent α-synuclein (Fig. 7) demonstrated colocalization (“closeness”) of all three markers in the host IPL. Overall, in transplant Nos. 1–4, ∼2.11% ± 0.34% of rodent α-synuclein dots were colocalized with human synaptophysin immunoreactive spots, and 3.87% ± 0.59% of synaptophysin immunoreactive spots in the host IPL were colocalized with rodent α-synuclein.

Glial and microglial markers in host and graft

Analysis of glial fibrillary acidic protein (GFAP) in the transplant sections revealed that Müller glial cells had lost their normal radial orientation in the RD host retina adjacent to the transplant (Figure 8 a–d). GFAP expression was also upregulated in the RD host retina (Figure 8 e,f). A low number of Iba-1-immunoreactive microglial cells was found in the graft (Figure 8 g,h).

FIG. 8.

Analysis of host and graft glial and microglial cells (GFAP and Iba-1 staining). (a, b) Transplant No. 9, 211 dps, age 239 days. This rat was not recorded in the SC because of cornea issues. Combination of GFAP (green) and human nuclei (red). (b) Enlargement of (a). Dense network of glial processes. Note that there is almost no GFAP staining in the body of the transplant. (c, d) Transplant No. 4, 211 dps, age 266 days. This rat had a response in the SC. Combination of GFAP (red) and Iba-1 (microglia, red). Low concentration of microglial cells. Note that there are some microglial cells in the lumen of a PR rosette in (h). Bars = 50 μm in (a–f, h); 200 μm in (g). GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer.

Discussion

We developed a successful overnight RO shipping protocol, which provided reliable temperature control and live monitoring of the shipment conditions. This allows for differentiating the best (transplantation competent) hPSC-derived ROs in a special facility, and then shipping them over large distances to any hospital over the world for transplantation on patients there. Our data showed the efficient and tumor-free engraftment of hESC-derived human retinal tissue (from ROs) after long distance shipping, which is a good starting point for developing clinical-grade retinal tissue implants for restoring vision in profoundly blind patients with significant loss of PRs. Our grafts survived in subretinal space for >6 months, developed mature PRs with inner and outer segments, established dense array of graft-with-host synapses in the host inner nuclear layer and retinal ganglion cell layers, and critically, reproducibly improved vision in the blind recipient animals with completely degenerated PRs. These results corroborated our earlier data in the same rat model with a different hESC line as source of retinal grafts [43].

Correlation of vision improvement with transplant location in the retina

The hESC-retinal tissue transplanted to rats with RD induced vision improvement, evident on optokinetic test (OKT; visual acuity test) and activation of contralateral SC, but not in retina with sham surgery or no surgery. Of interest, in one of the transplants, the SC response heatmap showed the similar shape as the transplant location in the eye, which suggested that the transplant restored the basic point-to-point projection function from the eye to the SC (Fig. 4).

Importance of rat model

Recipient rats were immunodeficient (RD nude) animals without T cells (not capable of rejecting xenografts) and the S334ter rhodopsin mutation which causes complete PR degeneration by 3–4 months after birth [44]. Therefore, no immunosuppressants were needed to enable graft survival. Immunodeficiency and lack of PRs in the recipient eyes eliminated the always present concerns about the survival of xenogeneic human grafts in animal models and the presence of some functional host PRs. Such concerns could lead to skewing the in vivo results and confounding the overall readout of the efficacy. We have previously shown that optokinetic responses in this rat are already severely reduced at the age of 23–27 days [43]. The RD nude rat model, therefore, is a superb model for demonstrating the preclinical efficacy of our vision restoration technology.

Challenges of surgery

The small eyes and the large lens of RD nude rats present a challenge for surgical grafting. Because of this, we had a number of unsuccessful surgeries, where the animals had to be terminated owing to (mostly) developing cataracts resulting from the surgical instrument inadvertently damaging the lens (most frequent case) or the degenerating retina. The surgery had to be carried out without the ability to clearly see the intravitreal space and the retina, thus complicating the overall precision and placement of a slice of hESC-retina into a subretinal space. Critically, all the above issues are not applicable in a much larger eye of a human patient where complex vitreoretinal procedures like subretinal surgeries have been worked out [20,48].

Source and maturation of retinal tissue

Here we used H1 (WA-01) hESC line (WiCell) as a source of human retinal tissue for subretinal grafting. We recently reported the derivation of hESC-ROs from hESC line H1 [31,33].

Our approach demonstrates efficient derivation of retinal tissue from hESCs carrying all the essential markers of developing human fetal retina including rod and cone PRs (Fig. 5), resembling developing human retina of developmental age 10–13 weeks [33]. This is important because human fetal retina of similar age has been used in a successful phase 2 clinical trial of RP and AMD, which reported vision improvements in several patients [25]. Upon maturation, hESC-derived retinal tissue (ROs) revealed the presence of a dense PR layer with mitochondria-rich inner segments, connecting cilia, and developing outer segments [33]. Collectively, this indicates that the grafts placed in the subretinal space of RD nude animals could similarly develop mature PRs, one of the critical mandatory requirements to restore vision.

Immunohistochemical analysis of subretinal grafts in rats with vision improvement (OKT improvement, SC activation) after transplantation reveals presence of RHO [+], RCVRN [+] outer segment-like protrusions from PRs in the graft. Of interest, in one of the transplants, the SC response heatmap showed a similar shape as the transplant in the eye, which suggested that the transplant restored the basic point-to-point projection function from eye to the SC (Fig. 3).

Synaptic connectivity

Using donor and host-specific markers in combination with synaptic markers also showed that transplants extend neuronal processes into the IPL of the host retina. Imaris software analysis showed colocalization of these markers, indicating synaptic connectivity between transplant and host. In addition, some donor cells migrated into the host retina. These results are similar to previous results using ROs from a different cell line [19,43]. However it is still unknown which specific cell types are involved.

Limitations of the study

Our study showed that transplanted ROs can have similar effects on the vision of a RD recipient when shipped long distance as tissue received from a close location. However, it was not possible to have a parallel control group on nonshipped RO transplants as the company producing the organoids has no facilities for rat housing and surgery.

Grafted sheets of PRs from hESC-derived retinal tissue may improve vision. Vision improvements remain so far small, but with additional improvements of grafts, this approach has the potential to restore vision to RD retina with no remaining PRs.

Future work should be focused on generating grafts in large RD models, improving functional integration to achieve better SC responses, studying vision restoration in the translational large eye models with RD, and developing additional functional tests, such as cortical recordings, visual evoked potentials, and so on, enabling translation of preclinical efficacy results to clinical trials in patients with profound/total vision loss.

Conclusions

The data show that hESC-derived retinal tissue has a promise in restoring vision in the eyes with total loss of vision and PRs. These results, together with results from other publications [19,43,49–51] support the development of the stem cell replacement therapy in vision restoration and testing this technology in future clinical trials.

Ethics Approval

IACUC protocol AUP18-145 (UC Irvine); hSCRO protocol: 2006-5316 UC Irvine; no human subjects were involved.

Code Availability

MATLAB code for analysis of SC responses is available upon request.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Supported by NIH 1R44EY027654, NEI 3 R44EY 027654–02S1 (PI: I.O.N., Multi-PI: M.J.S.); NIH R01 EY031834 (PI: M.J.S.); CIRM TR1-10995 (PI: M.J.S.). This study was made possible in part through access to the Optical Biology Core Facility of the Developmental Biology Center, a shared resource supported by the Cancer Center Support Grant (CA-62203) and Center for Complex Biological Systems Support Grant (GM-076516) at the University of California, Irvine. The authors acknowledge support to the Gavin Herbert Eye Institute at the University of California, Irvine from an unrestricted grant from Research to Prevent Blindness and from NIH grant P30 EY034070.