In 2020, 43 million people were estimated to be blind and 295 million had moderate to severe visual impairment worldwide (GBD 2019 Blindness and Vision Impairment Collaborators, 2021). Visual impairment may lead to depression, unemployment and financial stress, hip fracture, and early mortality (Brody et al., 2001; Wong et al., 2008; Loriaut et al., 2014; Wang et al., 2021). In young children with early-onset ocular disease, vision impairment may result in delayed motor, language, social and cognitive development with lifelong consequences, and a lower level of educational achievement (Gilbert and Awan, 2003).

The vertebrate retina develops from an optic vesicle which is formed from the evagination of an anterior neural tube, the diencephalon, therefore, the eye is considered a part of the central nervous system (CNS) (Wawersik and Maas, 2000). Similar to the mammalian brain, the neurons in the mammalian eye have limited ability to regenerate once the neurons degenerate. There are seven major cell types in the wild-type mammalian retina as shown in Fig. 1A. Degeneration of the inner retina mainly involves ganglion cells (GC) in the ganglion cell layer (GCL) and degeneration of the outer retina involves photoreceptors and/or retina pigment epithelium (RPE) in the outer nuclear layer (ONL) or RPE layer as shown in Fig. 1B to 1D respectively. Both are linked to blindness-causing diseases (Marchesi et al., 2021; Zhang et al., 2021). Degeneration in the outer retina involves two stages, the first stage has degeneration of the rod photoreceptor, and the second stage has degeneration of the cone photoreceptor. Retina degeneration can be caused by hereditary retinitis pigmentosa, macular degeneration, glaucoma, cancer, trauma, and diabetes (Hartong et al. 2006; Almasieh et al., 2012; van Lookeren Campagne et al., 2014; Wong et al., 2016; Maheshwari and Finger, 2018).

Figure 1.Inner and outer retina degeneration. (A) Adult wild-type retina. (B) Inner retina degeneration, ganglion cell loss in GCL. (C) Outer retina degeneration stage 1: rod photoreceptor loss in ONL. (D) Outer retina degeneration stage 2: cone photoreceptor loss in ONL. GC, ganglion cell; GCL, ganglion cell layer; INL, inner nuclear layer; MG, muller glia; NFL, nerve fiber layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium.

Retinal degeneration can be caused by an inherited mutation in photoreceptors, RPE, or in ganglion cells. The form of mutation can be autosomal recessive, autosomal dominant, and X-linked mutation (Hartong et al., 2006). Inherited retinal diseases can be classified into either inner retinal or outer retinal diseases. Inherited inner retinal diseases are early-onset glaucoma and Leber hereditary optic neuropathy (LHON) that have degeneration of ganglion cells as shown in Fig. 1B and Fig. 2. Early-onset glaucoma has a dominant mutation in the Myocilin gene and LHON has a mutation in the mitochondrial DNA leading to blindness (Ratican et al., 2018). Inherited outer retinal diseases are retinitis pigmentosa (RP), Leber congenital amaurosis (LCA), Usher syndrome (USH), and Stargardt disease (STGD) that have degeneration of photoreceptors and/or RPE in the ONL as shown in Fig. 1B and 1C, Fig. 2 (Botto et al., 2022). Retinitis pigmentosa is a major hereditary ocular disease that involves mutations in over 60% of the genes identified, the rest are still unknown. Initial symptoms include impairment of night and peripheral vision and finally progress to the loss of central vision.

Figure 2.Gene therapy for inner/outer retinal diseases and gene therapy. AAV2, adeno-associated virus 2; GC, ganglion cell; LHON, leber hereditary optic neuropathy; LCA, leber congenital amaurosis; RP, retinitis pigmentosa; RPE, retinal pigment epithelium; STGD, stargardt disease; TM, trabecular meshwork; USH, usher syndrome.

If the mutation can be corrected by gene therapy before the degeneration of photoreceptors, visual impairment can be avoided. Autosomal recessive mutation can be supplemented with wild-type genes, however autosomal dominant mutation or large gene that can’t fit into AAV may be difficult. Therefore, an alternative method is to use gene editors CRISPR/Cas9, base editor, or prime editor (Gaudelli et al., 2017; Anzalone et al., 2019; Botto et al., 2022). Both base editors and prime editors do not involve double-stranded DNA breakage, thus, they are considered to be a more precise and efficient editor. Adeno-associated virus (AAV) vector is a preferred delivery system (vehicle) to be used for gene therapy because it induces less immune response and does not integrate into the genome (Hastie and Samulski, 2015). The efficiency of gene therapy would rely also on the efficiency of the in-vivo delivery system, the delivery system needs to be able to deliver gene editor (cargo) to as many target cells as possible.

Gene therapy will be effective only when the mutations are corrected before the retinal degeneration. When retinal precursor cells from developing mouse retina are transplanted into degenerating mammalian retina, the precursor cells showed differentiation and integration into the retina resulting in an improvement of visual function (MacLaren et al., 2006). Numerous studies have derived retinal progenitors from embryonic stem cells (ESC) or induced pluripotent stem cells (iPSC) and transplanted them into degenerating retina to examine the integration of these cells (Lamba et al., 2006; Meyer et al., 2006; Lamba et al., 2009; Lamba et al., 2010). In addition, subretinal injection of stem cell-derived photoreceptors or RPE and intravitreal injection of GC showed integration into ONL and GCL respectively as shown in Fig. 3A and 3B (Carr et al., 2009; Barber et al., 2013; Venugopalan et al., 2016). However, later studies revealed that the donor cells failed to integrate, rather the studies found that the reporter was expressed due to cytoplasmic material transfer between the donor and host cells (Pearson et al., 2016; Santos-Ferreira et al., 2016; Singh et al., 2016).

Figure 3.Stem cell therapy for inner/outer retinal diseases. (A) Injection of iPSC-derived retinal cells to target integration in GCL or ONL. (B) Reporter (green) expression in GC and rod photoreceptors in GCL and ONL respectively are due to cytoplasmic material transfer rather than integration. GC, ganglion cell; GCL, ganglion cell layer; INL, inner nuclear layer; iPSC, induced pluripotent stem cell; MG, muller glia; NFL, nerve fiber layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium.

In contrast to the mammalian retina, zebrafish photoreceptors can regenerate via transdifferentiation of Muller glia (Wan and Goldman, 2016). Upon injury, the Muller glia (MG) residing in the INL, dedifferentiate into a progenitor cell, migrate to ONL and differentiate into photoreceptors. Several studies were able to reprogram Muller glia to transdifferentiate into photoreceptor or retinal ganglion cells (RGC) in mice (Yao et al., 2018; Xiao et al., 2021; Xie and Chen, 2022). Yao et al. (2018) induced MG in wild-type mice to re-enter cell-cycle (proliferate/progenitor state) by β-catenin gene transfer via injection of adeno-associated virus (AAV), where the gene was driven by MG-specific GFAP promoter as shown in Fig. 4A. This was followed by a second AAV injection to express rod photoreceptor cell fate genes Otx2, Crx, and Nrl in MG. After a week, reprogrammed Muller glia showed asymmetric cell division, one daughter cell differentiated into a rod photoreceptor and the other into an MG as shown in Fig. 4C. In 4th week, transdifferentiated MG became mature rod photoreceptors in ONL expressing photoreceptor markers and synaptic markers.

Figure 4.In-vivo transdifferentiation of Muller glia to rod photoreceptor or ganglion cell by reprogramming. (A, C) 1st intravitreal injection of AAV2 GFAP β-catenin to induce dedifferentiation of Muller glia to progenitor state. AAV2 GFAP-GFP (green) and AAV2 Rhodopsin-td Tomato (red) reporters were co-injected to mark Muller glia and rod photoreceptors respectively. 2nd intravitreal injection of AAV2 GFAP Otx2-Crx-Nrl to promote Muller glia to adopt rod photoreceptor fate. Rod photoreceptors (red) in ONL are derived from Muller glia. (B, D) Subretinal injection of AAV9 GFAP Math5-Brn3b-GFP to induce Muller glia to adopt ganglion cell fate. Ganglion cells (green) in GCL are derived from Muller glia. GC, ganglion cell; GCL, ganglion cell layer; INL, inner nuclear layer; MG, muller glia; NFL, nerve fiber layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium.

Xiao et al. (2021) directly transdifferentiated MG in the wild-type mice to RGC by expressing GC-specific transcription factors Math5/Brn3b via AAV9 as shown in Fig. 4B. After 5 days of infection, the MG changed morphology to intermediate cell and migrated toward GCL and started to express RGC markers in GCL by 7 days post-infection as shown in Fig. 4D. The axons of RGC projected the entire pathway to the brain. In addition, reprogramming of MG in mutant mouse that lacks 70% of RGC showed improvement in the visual function. It remains to be examined whether the transdifferentiation of MG to RGC can also occur in the degenerating retina.

Another alternative, although may sound to be possible only in the distant future is WET. Between 1930 and 1942, Stone (Stone and Cole, 1943; Stone and Ellison, 1945) carried out numerous successful WET in amphibian. In these studies, salamander eyes were completely exercised and reimplanted back to the same or a different host. Return of the vision based on behavioral response was shown in both the same and different hosts. In mammals, WET was performed since 1885 in rats, rabbits, canines, swine, ovine, and even in humans, but none of the studies showed restoration of vision (Bourne et al., 2017).

Successful WET requires the establishment of nerve connection, revascularization, and the absence of immune rejection. Successful revascularization was achieved following anastomosis in swine and ovine (Sher, 1981; Shi et al., 2009), however, immune rejection and nerve connection remained major hurdles. Immune rejection can be circumvented by transplanting the whole eye derived from the patient’s own cell reprogramming using Yamanaka or OSKM transcription factors (OCT4, SOX2, KLF4, MYC) (Takahashi and Yamanaka, 2006). The iPSC then can be used to grow a patient’s whole eye in livestock using the blastocyst complementation technique.

The blastocyst complementation (BC) technique was developed by Chen et al. (1993), however, BC is based on the germline chimera techniques initially developed by Evans and Kaufman (1981). BC uses chimera formation between the donor and host pluripotent stem cell in the blastocyst to complement genetically deficient organs in the blastocyst stage embryo (Chen et al., 1993). Thereafter, interspecies BC (IBC) was performed in mouse, rat, pig, and human donor cells (Zheng et al., 2021). Fig. 5 describes how the iPSC-derived human eye can be obtained from interspecies blastocyst complementation for WET. To obtain human organs, the host animal’s organ size needs to be comparable to that of a human (Freedman, 2018). When Kobayashi et al. (2010) generated a rat pancreas in a mouse, the rat pancreas was the size of a mouse pancreas, whereas when a mouse pancreas was generated in a rat, the pancreas was the size of a rat (Yamaguchi et al., 2017).

Figure 5.Whole eye transplantation of human iPSC-derived eye from interspecies blastocyst complementation. (A) Obtain human somatic cells. (B) Obtain iPSC by reprogramming somatic cells using Yamanaka factors. (C) Injection of human iPSC into eye-disabled (Pax6 deactivated) pig blastocyst resulting in chimeric pig blastocyst. (D) Chimeric pig blastocyst is transferred to the oviduct of the recipient female pig. (E) Select chimeric offspring with human iPSC-derived eyes. (F) Enucleate iPSC-derived human eye from the chimeric offspring and transplant the whole eye into a human followed by nerve reconnection and revascularization.

Yamaguchi et al. (2017) showed therapeutic potential of the mouse pancreas generated in rat by transplanting the pancreatic islets to the mice with drug (STZ) induced diabetes. The transplantation did require 5 days of immunosuppression to prevent immune response resulting from donor (rat) supporting tissue contamination. The diabetic mice that received the islets from IBC maintained normal blood glucose level over 370 days without continuous immunosuppression. Thus, WET of human iPSC derived eye from IBC may also induce immune response and require short duration of immunosuppression.

However, studies show difficulties in creating interspecies BC between species that are evolutionary more distant species such as pigs and humans (Ballard and Wu, 2021). Researchers are working to increase the efficiency of distant IBC (Wu et al., 2017; Nishimura et al., 2021; De Los Angeles and Wu, 2022). Wu et al. (2017) show that human intermediate pluripotent stem cell results in more efficient IBC than the naïve or prime pluripotent stem cell in the pig. Nishimura et al. (2021) show that IBC efficiency can be improved by increasing the growth competition of human donor pluripotent cells (Ballard and Wu, 2021).

The last hurdle is the reconnection of the optic nerve, whether the donor eye’s optic nerve can regenerate and connect to the host optic nerve after coaptation needs to be resolved. Studies showed that intraocular inflammation-inducing factors, intrinsic factors, and partial reprogramming factors may promote axon regeneration in ganglion cells after optic nerve axotomy (Yin et al., 2019; Lu et al., 2020). Study shows that unintentional injury to the lens induces ganglion cell axon regeneration of the optic nerve in rat eye and found that this effect could be mimicked by inducing intraocular inflammation by Zymosan (ligand found on the surface of fungi) (Leon et al., 2000). High concentration of ciliary neurotrophic factor (CNTF) also shows the effect of intraocular inflammation on axon regeneration (Müller et al., 2007). Upon completion of axonal growth of retinal ganglion cell (RGC) in early development, the axon growth is repressed by cell-intrinsic transcription factor Klf-4. Therefore, the deletion of Klf-4 in RGC promoted optic nerve regeneration (Moore et al., 2009). Lu et al. (2020) showed that partial reprogramming of ganglion cells using OCT4, SOX2, and Klf4 genes (OSK) can promote axon regeneration and restore vision after optic nerve axotomy.

Although stem cell therapy had integration issues, gene therapy, and in-vivo transdifferentiation via reprogramming hold promising outlooks to restore vision. For stem cell therapy to work, more studies are needed to understand how progenitor or differentiated cells can be induced to migrate to the injury site and replace the damaged cells. The success of gene therapy relies on the efficiency of the delivery system, the precision of the gene editor, and minimal immune response. Improvement in all three criteria would be able to restore/prevent vision in hereditary ocular diseases. In-vivo transdifferentiation of Muller glia cells via reprogramming needs to be verified in the degenerating retina. In addition, the improvement of the delivery system will target more cells and thereby increasing reprogramming and transdifferentiation. Last, the idea of transplanting a patient’s own eye obtained using iPSC and interspecies blastocyst complementation method still needs several hurdles to overcome.

None.

Conceptualization, K.D.R.; writing-original draft, K.D.R.; writing-review & editing, K.D.R.

None.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

No potential conflict of interest relevant to this article was reported.