P23H, the first mutation reported in ADRP (Dryja et al., 1991), is the most common mutation found in rhodopsin in the United States (Wu et al., 2019; Woods and Pfeffer, 2020). P23H rhodopsin is a typical example of a class II mutation.

The pathogenesis of the P23H mutation has been extensively studied in a variety of animal models and in vitro cultured cells. The P23H mutation affects the adjacent H-bonding network in the chromophore binding region critical for the activity of the chromophore in rhodopsin expressed cultured HEK293 cells (Woods and Pfeffer, 2020). Additionally, it also alters the overall structure and activity of rhodopsin (Woods and Pfeffer, 2020). The monomer of mutant P23H is not functional, and it is unstable by itself and tends to adopt a specific homodimer arrangement (Woods and Pfeffer, 2020). P23H also exerts a destructive effect on disk membranes—likely through a homodimerization process—even at very low concentrations (Woods and Pfeffer, 2020). Another proposed possible pathogenic mechanism is related to activation of the UPR resulting from misfolding of the mutant protein as demonstrated in culture cells and transgenic mice (Frederick et al., 2001; Lin et al., 2007). UPR-mediated endoplasmic reticulum stress (ERS) triggers Ca²⁺ release from the ER, leading to activation of calpains and caspase-12 (Kitamura, 2008; Choudhury et al., 2013). Activated calpain can cleave the mitochondrial protein AIF (apoptosis-inducing factor), promoting AIF exit from the mitochondria through a pore formed by BAX and translocation into the nucleus, where AIF recruits cyclophilin A for chromatin condensation and fragmentation (Candé et al., 2004; Moubarak et al., 2007). AIF has been demonstrated to be present in the nuclei of most dying photoreceptor cells in P23H transgenic and knock-in mouse models (Comitato et al., 2016, 2020). Blocking calpain activity effectively protects the retina from degeneration in P23H knock-in mice (Comitato et al., 2020), suggesting calpain activation as one of the major causes of P23H mutant rhodopsin-induced photoreceptor degeneration.

The T17M mutation in RHO results in methionine replacing threonine at position 17 (Choudhury et al., 2013). This mutation is another class II mutantion that causes rhodopsin protein msifolding (Krebs et al., 2010), as described above. The mutation affects the binding of opsin proteins to 11-cis-retinaldehyde, resulting in ADRP (Li et al., 1998; Mendes et al., 2005; Krebs et al., 2010; Choudhury et al., 2013). The T17M rhodopsin mutant expressed in cultured cells is abnormally mislocalized in the endoplasmic reticulum (Li et al., 1998; Krebs et al., 2010; Jiang et al., 2014), with no colocalization with the Golgi apparatus as normal rhodopsin (Deretic and Papermaster, 1991), which may activate UPR and upregulating ERS-related proteins, such as BIP, GRP94, CHOP, peIF-2a/eIF-2a, and activating ATF-6a (Baumeister et al., 2005; Wang et al., 2009). Upregulation of ERS-related genes has been documented in the retina of T17M transgenic mice (Kunte et al., 2012). In vitro evidence suggests the mutant protein is unstable and susceptible to degradation by the proteasome system (Jiang et al., 2014). The demise of photoreceptors caused by T17M is partially attributed to the activation of caspase-7. Ablation of the gene encoding caspase-7 protects photoreceptors in T17M transgenic mice, likely through UPR reprogramming and inhibition of TRAF2-JNK apoptosis (Choudhury et al., 2013).

ROS are also suggested to play an important role in T17M-related retinal degeneration. ROS are the byproducts of aerobic metabolism, including oxygen ions, peroxides, and oxygen-containing free radicals (Jiang et al., 2014). High levels of ROS may damage lipids, proteins, and nucleic acids and affect the functions of organelles (Halliwell, 2011). Mitochondria are the main source of ROS. Increased ROS levels have been observed in cells expressing mutant T17M rhodopsin (Jiang et al., 2014), as ERS promotes chaperone activities that require more energy. Treatment of cells expressing T17M rhodopsin with ROS scavengers reduces cell death.

The G90D mutation, which is due to the substitution of aspartic acid for glycine at position 90 in rhodopsin, destabilizes a crucial ionic bond between E113 and K296 (Singhal et al., 2013; Silverman et al., 2020). In vitro data suggest that the G90D mutant belongs to a group of constitutively active mutants (including K29E) that can activate transducin in the dark (Rao et al., 1994; Robinson et al., 1994; Toledo et al., 2011; Park, 2014), resulting in a light-adapted state and the desensitization of rod photoreceptor cells in the dark (Sieving et al., 2001; Jin et al., 2003; Naash et al., 2004; Dizhoor et al., 2008). The mutation of neutral G to charged D alters the water-mediated H-bond network at the Schiff base region of the chromophore and the central transmembrane region, which may cause slow binding of the chromophore during pigment regeneration and constitutive activation of transducin (Rao et al., 1994; Gross et al., 2003). Persistent rhodopsin activation can cause retinal degeneration in both a transducin-dependent manner and a transducin-independent manner (Hao et al., 2002). Earlier studies showed that mutation of this gene causes congenital night blindness (Dizhoor et al., 2008). Patients with the G90D mutation have shown decreased light sensitivity of rod cells and night vision dysfunction. Most patients with the G90D mutation express normal amounts of rhodopsin, and the structure of the rods is well preserved (Sieving et al., 1995), while others exhibit the typical RP manifestation, in which the loss of rods is accompanied by the subsequent death of the cone cells and blindness (Berson, 1993). A cohort study comprised of 15 patients showed that 20 and 53.3% of patients with the G90D mutation displayed CSNB and classic RP, respectively, with no obvious sex differences (Kobal et al., 2021). Recently, slow retinal degeneration was also found in homozygous G90D transgenic mice (Colozo et al., 2020). From this point of view, the human G90D mutation may be right at the boundary between dysfunction and degenerative disease, providing an opportunity to further explore the mechanism by which the spontaneous activation of the visual transduction cascade leads to rod structure damage and cell death (Dizhoor et al., 2008).

The Q344ter mutation in rhodopsin causes the glutamine-encoding codon 344 to be replaced by a stop codon, resulting in early termination of the polypeptide and loss of the signature C-terminal motif sequence QVAPA. Rhodopsin is synthesized in the rough ER of the inner segment, processed by the Golgi apparatus, and then transported across the connecting cilium to the outer segment where phototransduction occurs. The C-terminal motif of rhodopsin is sufficient and necessary for rhodopsin to be correctly transported to outer segments. It has been shown that the carboxy-terminal cytoplasmic tail of the rhodopsin protein is involved in the post-Golgi transport of rhodopsin (Sung and Tai, 1999). The Q334Ter mutant expressed in transgenic animal models is mislocalized to the plasma membrane of the inner segments and subsequently translocated to the lysosome for degradation (Ropelewski and Imanishi, 2019). Degradation of mislocalized mutant rhodopsin causes disruption of plasma membrane protein homeostasis and downregulation of the sodium-potassium ATPase α-subunit (NKA). Compromised NKAα function is sufficient to cause shortening and loss of rod outer segments, which may underpin the mechanism of retinal degeneration related to the Q334Ter mutation (Ropelewski and Imanishi, 2019). A recent transcriptomic analysis of retinas from Q334Ter knock-in mice revealed alterations in the expression of chromatin complex genes such as histone genes (Bales et al., 2018). A more recent study showed that the upregulation of proinflammatory cytokines and pathways is involved the pathogenesis of retinal degeneration in this knock-in mouse model (Hollingsworth et al., 2021).

These mechanisms represent a summary of representative pathogenesis related to rhodopsin mutations (Figure 1). Information for the mechanisms associated with other types of rhodopsin mutation can be found in several great earlier reviews (Athanasiou et al., 2018). Given the complexity of the interaction between all physiological processes and the wide spectrum of rhodopsin mutations, as well as the unique genetic background of each individual, we are still far from a complete understanding of these pathogenic mechanisms.

Electrical stimulation therapy (EST), as a classic physical therapy, can improve muscle and nerve function. As early as the 19th century, the rehabilitative effects of electrical stimulation on the eyes were observed (Dor, 1873). The administration of low-level electric current by various approaches has been demonstrated to improve visual function, validating the feasibility of this therapeutic strategy in the treatment of eye diseases (Hanif et al., 2016). Using methods such as subretinal implants or transcorneal electrical stimulation (TES), low-level electrical stimulation of the eye has been shown to have neuroprotective effects on retinal degeneration in humans and animal subjects (Ciavatta et al., 2013; Hanif et al., 2016; Yu et al., 2020).

RHO P23H transgenic rats receiving 30 min of low-level electrical stimulation at a frequency of twice a week (4 μA at 5 Hz; n = 10) from 4 to 24 weeks of age exhibited significant improvements in visual function, as exemplified by better responses in the electroretinography (ERG) test (Hanif et al., 2016). The underlying mechanism could be due to increased expression of neuroprotective factors, such as ciliary nerve trophic factor (CNTF) and brain-derived neurotrophic factor (BDNF), in Müller cells (MCs; Ferrari et al., 2011; Gall et al., 2011; Hanif et al., 2016). Transcorneal electrical stimulation therapy in 21 patients with RP showed that TES might cause trivial symptoms such as foreign body sensation, burning, and itching, which, however, were mild and controllable (Demir et al., 2022). In the future, EST can be combined with other treatment methods to increase its efficacy.

Pharmacological compounds are often used to alleviate or treat various types of RP due to their various beneficial factors, such as their wide variety and easy availability. Several types of pharmacological compounds have been experimentally and clinically tested to treat retinal dystrophies caused by rhodopsin mutations.

Antisense-mediated gene suppression was first reported in 1978 (Zamecnik and Stephenson, 1978). In recent years, antisense oligonucleotides (ASOs) have emerged as a potential strategy for the treatment of inherited retinal diseases (Collin and Garanto, 2017). ASOs are small DNA or RNA molecules that are complementary to their target mRNAs. Therapeutic ASOs can be chemically synthesized oligonucleotides 18–30 nucleotides in length. The binding of ASOs to targeted RNAs promotes RNA fragmentation and degradation. They may also inhibit the expression of target RNA by blocking the translation machinery. One study showed that intravitreal administration of second-generation ASOs effectively and specifically reduces the level of allele-specific mutant rhodopsin in a transgenic rat that expresses a murine P23H rhodopsin gene (Murray et al., 2015). In this study, the treated eyes also exhibited improved photoreceptor morphologies, function, and cell survival. ASOs have several apparent advantages over other gene-silencing agents (Murray et al., 2015), such as high selectivity for alleles with single-base mutations and simple delivery in a water-based formulation (Østergaard et al., 2013). A first-generation ASO for the treatment of cytomegalovirus (CMV) retinitis has been approved by the FDA (Group, 2002; Jabs and Griffiths, 2002). The half-life of second-generation ASOs is notably longer than that of first-generation ASOs. Thus, the treatment cycle will also be longer without the need for frequent injections (Murray et al., 2015). These advantages make ASOs great candidates for the treatment of retinal degeneration caused by rhodopsin mutations.

Gene therapy enables the addition of exogenous genes to correct pathogenic symptoms. Depending on the mutation type, several strategies have been developed to treat RP. Gene augmentation is a simple and straightforward strategy featuring the transfer and expression of wild-type exogenous genes into host cells. This approach has proven to be successful in the treatment of autosomal recessive RP both experimentally and clinically (Kumar et al., 2016; Bennett, 2017). However, this strategy had limited success in treating ADRP. Although gene augmentation may improve pathogenic symptoms by diluting mutant proteins, the gain-of-function or dominant-negative mutant proteins that remain within the cells will still exert their toxic effects. Therefore, researchers are striving to develop alternative strategies to treat ADRP, particularly rhodopsin-associated RP, which accounts for over 25% of all ADRP cases (Dryja et al., 1991; Sullivan et al., 2013).

Trophic factors are secreted small proteins that regulate cell proliferation, maturation, and viability (Snider and Johnson, 1989; von Bartheld, 1998). Some trophic factors have been documented to effectively delay retinal degeneration in various animal models (Faktorovich et al., 1990; LaVail et al., 1992; Unoki and LaVail, 1994; Green et al., 2001; Miyazaki et al., 2003; Buch et al., 2006). Basic fibroblast factor (bFGF) was the first factor to demonstrate neuroprotective effects on degenerative photoreceptors in light-damaged rats, and RCS rats suffer from a mutation in the MERKT gene (Faktorovich et al., 1990). Since then, more neurotrophic factors, including brain-derived neurotrophic factor (BDNF; LaVail et al., 1992), ciliary neurotrophic factor (CNTF; Unoki and LaVail, 1994), glial cell line-derived neurotrophic factor (GDNF; Buch et al., 2006), and pigment epithelium-derived factor (PEDF; Miyazaki et al., 2003), have been found to effectively counter retinal degeneration.

As neurotrophic factors are proteins, they are susceptible to degradation by proteases. Consequently, they have a relatively short half-life and can only provide short-term protection. Viral vector-mediated delivery of genes encoding neurotrophic factors has been shown to achieve sustained expression of these factors and offer long-term protection. One study showed that FGF-15 and FGF-18 expressed from the recombinant AAV virus notably delayed retinal degeneration caused by transgenic rats expressing a P23H or Q334ter rhodopsin mutation (Green et al., 2001). FGF-15 and FGF-18 are two members of the fibroblast factor family. In addition, AAV-mediated expression of CNTF prolongs photoreceptor survival in mutant rhodopsin mice (Liang et al., 2001). A recent study showed that treatment of P23H rats using neurotrophic factors in combination with suppression of microglia by minocycline achieved better results than using neurotrophic factors alone. Thus, neurotrophic factors are promising candidates for the treatment of retinal degeneration related to rhodopsin mutations.

Although CNTF and PEDF have been shown to rescue photoreceptor morphologies and prolong photoreceptor survival in rodent models of retinal degeneration, they may suppress retinal function, as determined by ERG recoding (Liang et al., 2001; Miyazaki et al., 2003; Buch et al., 2006). Additionally, translation of this approach into clinical applications encountered a major setback, based on one clinical study showing that RP patients treated with CNTF released continuously from an intravitreal implant had a greater loss of total visual field sensitivity in the treated eyes than in the sham-treated eyes (Birch et al., 2016). The reason for the contradictory results between animal models and human patients is unknown. Therefore, the use of neurotrophic factors for treating RP requires further evaluation.

During the process of RP, the progressive death of rod photoreceptors is followed by cone cell death. After photoreceptor cells die in advanced RP, the structure and function of the remaining cells, such as retinal ganglion cells and bipolar cells, remain intact. Therefore, it is possible to convert these remaining light-insensitive cells into photosensitive cells using an optogenetic therapy approach, thereby partially restoring vision (Pan et al., 2015). Channelrhodopsin 2 (ChR2) is a widely used optogenetic protein for this purpose. ChR2, originally cloned from Chlamydomonas reinhardtii (Nagel et al., 2003), is a direct light-gated cation ion channel that opens rapidly upon the absorption of photons and depolarizes the cell membrane (Nagel et al., 2002; Busskamp et al., 2010; Mutter et al., 2014). Restoration of vision using an optogenetic strategy was achieved in preclinical mouse and rat models of RP (Bi et al., 2006; Tomita et al., 2007, 2010). In addition to optogenetic proteins, chemical photoswitchs, such as DENAQ, AAQ and diethylamino-azo-diethylamino (DAD), have been shown to restore retinal responses to light in mice with degenerated photoreceptors (Polosukhina et al., 2012; Tochitsky et al., 2014; Laprell et al., 2017). A recent clinical study showed that a blind patient partially recovered visual function after optogenetic therapy using the channelrhodopsin protein ChrimsonR fused to the red fluorescent protein tdTomato (Sahel et al., 2021). The patient recovered to the extent that he could recognize the location of objects and reach out to locate them with the help of goggle-assisted-light stimulation. As optogenetic therapy is a gene and mutation-independent approach, it is conceivable that optogenetic therapy is quite promising to restore partial vision for advanced RP patients with rhodopsin mutations.

With the improvement of our understanding of stem cells, stem cell therapy has emerged as a promising treatment method for RP, particularly RP at an advanced stage when most photoreceptors are lost and gene therapy is difficult. There are two major strategies for treating RP using stem cell therapy: (1) the transplantation of exogenous stem cells into the retina and the induction of the differentiation of the cells into desired cell types and (2) the induction of transdifferentiation of MCs into other types of neural cells.