Retinitis pigmentosa (RP) is a group of hereditary diseases that result in the progressive degeneration of the retina's photoreceptor cells, primarily starting with the rods. This gradual loss of vision is the most common form of inherited retinal dystrophy, and it imposes a significant burden on both individuals and society. RP is a leading cause of visual disability and blindness in people under 60, affecting more than 1.5 million individuals worldwide. The most common symptoms of RP include nyctalopia and gradual peripheral vision loss, which may ultimately lead to complete blindness.
Gene therapy has made significant advancements in the treatment of inherited retinal diseases, particularly in the field of retinitis pigmentosa (RP) [1]. Gene therapy approaches for RP are generally based on two strategies, which are influenced by the disease's inheritance pattern. In recessive RP, where there is a loss of function of the affected protein, the gene complementation approach is applied. On the other hand, gene therapy approaches for dominant RP typically involve gene suppression, with or without gene complementation.
To achieve RP management's therapeutic goal, various gene therapy techniques have been developed using viral or non-viral vectors. This section provides an overview of the primary gene therapy methods currently being studied for the development of therapeutic approaches for RP management.
Gene therapy using viral vectors for the treatment of retinitis pigmentosa (RP) has been extensively researched, with several vectors reaching clinical studies [2]. Three primary vector strategies based on viruses have been studied: adenoviruses (Ad), adeno-associated viruses (AAV), and lentiviruses (LV) [3]. Recently, AAV has gained focus due to its potential. AAV is a nonenveloped icosahedral virus that was described in 1996 and is relatively small in size (25 nm) [4] (pp. 1–23). The small size of AAV enables the efficient targeting of retinal layers, which makes it a favored option in gene therapy for the treatment of inherited retinopathies (IR). Additionally, AAV exhibits low immunogenicity, allowing a prolonged expression of the target gene after single-dose treatment and safety in second-dose administration in the subretinal space [5]. Moreover, AAV is non-pathogenic and has not yet been linked to human diseases, although it can be present [3]. On the other hand, It is important to note that pre-immunity towards Ad-based vectors exists due to their high seroprevalence, which can compromise Ad-based gene therapies [6].
AAV vectors have a simple genome that lacks auto-replication and expression [3]. AAV-mediated DNA editing is performed through the clathrin-mediated endocytosis pathway, and once in the nucleus, the DNA strand undergoes second-strand synthesis followed by gene transcription. The AAV genome can integrate into the human genome on chromosome 19 at the AAVS1 position [7]. AAV is highly transduction efficient and provides stable expression, making it an attractive option for gene therapy [8]. However, the use of single AAV vectors has limitations in the context of RP, including a cloning capacity that is limited to non-large genomes (4.7 kb), a relatively slow onset of transgene expression due to the need for second-strand synthesis, and highly invasive procedures requiring subretinal injections [5]. Researchers have developed novel approaches to address these challenges for treating IRD, such as splitting the coding sequence into separate AAV vectors to reconstitute full-length transgene protein expression in transduced cells, increasing the cloning capacity of AAV up to 14 kb using triple vectors [9], and using dual vectors to treat Stargardt disease in Abca4−/− mice, which improved disease phenotype [10]. However, using multiple AAV vectors for treating monogenic diseases is challenging due to the requirement of laborious post-translational modifications such as recombination and splicing [11]. Nonetheless, novel AAV capsids have been developed that allow for intravitreal gene therapy, which could alleviate the burden associated with intraretinal injections [12][13][14].
The novel CRISPR-Cas9 gene editing system is a widely used tool for gene expression and suppression [15][16]. In retinal diseases, it has successfully silenced dominant mutations via the NHEJ pathway [17][18]. However, clinical application faces a major challenge due to lower performance rates, with success highly dependent on sgRNA design, cell target type, delivery method, and host factors. Despite this, even a small percentage of gene modification through CRISPR-Cas9 can improve phenotype and hold significant promise for IRD.
Recent research has shown a growing interest in RNA-guided gene editing methods [19]. The Cas13 tool is an RNAtargeting endonuclease that can implement gene modifications during transcriptional activity, unlike the DNA-editing CRISPR-Cas9 system [20][21]. Cas13-based RNA editing has successfully shown the restoration of protein product functions in monogenic disease therapy [19]. However, the potential for non-specific collateral RNA cleavage is a major drawback associated with the Cas13 tool [22][23][24], along with a recent finding that it could be neurotoxic, impeding neuronal development in vitro [25]. While CRISPR-Cas13 methods represent a novel approach for the treatment of retinal diseases, further studies are required to assess their potential and safety [26].
RNA replacement therapies involve the depletion of endogenous mutant RNAs. Noncoding RNAs (ncRNAs) play a crucial role in regulating gene expression and can be categorized into microRNA (miRNA), small interfering RNA (siRNA), and short hairpin RNA (shRNA) based on their regulatory elements [27]. MiRNA represses mRNA translation and destabilizes mRNA by binding to its target sites [27]. RNA interference (RNAi) regulates protein-coding gene expression [28], and mirtrons, a subtype of miRNA, can be used in gene therapy for retinitis pigmentosa (RP) treatment [65]. SiRNA binds with high specificity, and shRNA regulates DNA delivery instead of RNA effector molecule regulation [27]. Antisense oligonucleotides (ASO) are synthetic single-stranded RNAs that promote RNA cleavage and have therapeutic effects in inherited retinal diseases [27]. SiRNA is vulnerable to degradation by serological enzymes, but shRNA is more potent and has a sustained effect [29].
In preclinical studies, various genetic mutations have been implicated in the development of RP, affecting apoptotic and non-apoptotic cell death pathways, inflammatory responses, and metabolic pathways (Table 1) [30]. Unfortunately, there is no curative treatment available for RP, and current approaches focus on developing targeted gene therapies for cell rescue, as well as molecular targets to facilitate cell replacement and maintenance of artificial retina [31].
Inheritance Mode | Genes | Preclinical Phase Studies |
---|---|---|
Autosomal Dominant Retinitis Pigmentosa |
RHO |
CRISPR-cas9-induced Rho mutation in Xenopus laevis tadpoles inhibits Müller glial cell proliferation [32].T17M rhodopsin expression induces the upregulation of IL-1β, IL-6 and NFκB and the IB1A microglia markers in C57BL6 mice [33].Proline substitution with leucine (P347L) in rhodopsin gene in transgenic rat strain induces rapid destruction of the outer nuclear layer, by CHOP and BiP activation [34].Subretinal injection of fused zinc-finger to the KRAB domain in combination to a RHO cDNA driven by the GNAT1 promotor in pigs restores Rho gene expression and retinal function [35].CRISPR-cas9 gene therapy in S334ter-3 rats increases visual acuity and retinal preservation [36].AAV-miR-2014 delivery in RHO-P247S transgenic mice enhances retinal function [37] |
Nrl |
CRISPR-Cas9 engineered NRL-deficient ESC retinal organoids exhibit an abnormal number of photoreceptors expressing S-Opsin [38].AAV-delivered CRISPR-cas9 Nrl gene knockdown in RHO-P347S Rd10 mice prevents retinal degeneration [39]. |
|
Nr2e3 |
AAV-delivered CRISPR-cas9 Nr2e3 gene knockdown in Rd10 mice prevents retinal degeneration [40]. |
|
PRPF |
AAV-associated CRISPR-cas9 Prpf31 gene augmentation in mice restores retinal function [41]. |
|
RP1 |
Dual Cas9/sgRNA successfully reduces RP1 expression in edited Hap1-EF1a-RP1 cells [42]. |
|
Autosomal Recessive Retinitis Pigmentosa |
RHO |
HEK293 and HT1080 human cells lines expressing E249ter or W161ter mutants exhibit lower rhodopsin mRNA levels [43]. |
RPGR |
AAV-delivered CRISPR-cas9 restores RPGRORF15 reading frame in rd9 mice [44].AAV-delivered CRISPR-cas9 restores photoreceptor preservation in Rpgr−/yCas9+/WT mice [45]. |
|
CNGB1 |
CNGB1 expression augmentation in Cngb1−/− mice using an AAV vector restores vision and delays retinal degeneration [46] (pp. 733–739). |
|
TULP1 |
Single-knockout (tulp1+/−) and double-knockout zebrafish models for the expression of Tulp1 significantly alters ciliogenesis through the downregulation of tektin2 [47].AAV-mediated Tulp1 gene expression editing restores Tulp1 mRNA and protein levels [48]. |
|
FAM161A |
Fam161a knockout mice display retinal degeneration phenotype as well as enhanced molecular generative markers [49].Homozygous Fam161a p.Arg512∗ have altered visual acuity due to complete loss of the outer nuclear layer and photoreceptor cell death [50]. |
Management of RP has been the subject of numerous studies, which have reviewed current treatment options, focusing on clinical trials or FDA-approved treatments [30][31]. Despite the genetic heterogeneity of RP, new molecular findings from preclinical studies may offer potential targets for drug development. In this section, researchers will review these findings, as well as advances in mutation-specific gene therapy.
Retinal dystrophy is influenced by multiple genes. Most genetic mutations associated with RP lead to the dysfunction of photoreceptors or RPE. Among these, various mutations have been identified as playing a key role in RP development [51][52][53], with some mutations currently in clinical trial phases for drug development. Most genetic mutations associated with RP lead to the dysfunction of photoreceptors or RPE [54].
Photoreceptors, including rod and cone cells, are specialized neurons responsible for the initial step of vision: phototransduction. Rod cells are responsible for dim light vision, whereas cone cells allow for color vision. Each photoreceptor is composed of five distinct regions: the outer segment (OS), connecting cilium (CC), inner segment (IS), nuclear region, and synaptic region. The OS of rod cells contains stacked disks, which significantly increase the membrane surface area and promote a high density of visual pigments such as rhodopsin. Rhodopsin is a G-protein-coupled receptor synthesized in the endoplasmic reticulum that allows phototransduction once activated by light. Activation of rhodopsin initiates GTP-mediated signaling, leading to the hydrolysis and reduction of cGMP in the cytosol of photoreceptors. The lower levels of cGMP cause the closure of CNG ion channels, which inhibits calcium and sodium influx, resulting in the hyperpolarization of rod photoreceptors and the subsequent inhibition of glutamate release at the synapse.
Photoreceptor cell death is a critical aspect of the pathogenesis of inherited retinal diseases such as RP, leading to visual loss. Apoptosis pathway activation, which involves three caspase-dependent pathways, has been identified as the most studied cell death mechanism in retinal degeneration. The intrinsic apoptotic pathway is mediated by pro-apoptotic Bcl-2 family proteins, which induce pore formation in the mitochondrial membrane, leading to cytochrome c release. Cytochrome c forms a complex with APAF1 and procaspase9, called the apoptosome, leading to caspase-9 activation, which ultimately induces irreversible DNA fragmentation via activation of caspase-3. However, inflammasome activation by NOD-like receptors (NLRs) was shown to significantly contribute to photoreceptor cell death through immunity activation and executioner caspase cleavage, demonstrating the importance of the immune system in the pathogenesis of retinal degeneration.
Research has demonstrated the significance of unfolded protein response (UPR) signaling through the ER stress pathway in retinal degeneration. When there is an accumulation of unfolded or misfolded proteins in the cytosol, the UPR pathway is activated, mediated by three main pathways involving ER transmembrane proteins: IRE1α, eIF2 kinase-PERK, and ATF6 (Figure 1) [55][56][57]. Uncontrolled activation of the ER stress pathway can lead to cell apoptosis [58], while in normal conditions, inhibitory chaperone molecules such as glucose regulatory protein 78 (GRP78/BiP) maintain ER transmembrane protein homeostasis [59].
RP pathogenesis involves various regulated necrosis pathways, including necroptosis, ferroptosis, and pyroptosis. Necroptosis is activated by TNFα or reactive oxygen species (ROS) stimulation, and through ischemia-induced cell death, leading to RIPK1-RIPK3-MLKL activation and membrane alteration, and eventually inducing inflammatory response [60]. Ferroptosis involves lipid peroxidation and intracellular iron accumulation, leading to ROS level elevation. Pyroptosis leads to caspase activation and causes cell membrane lysis and inflammatory response, similar to the apoptosis pathway [61].
Autosomal dominant disorders are highly studied due to their severity as a single allele mutation can induce a change in phenotype. Mutations associated with adRP make up 50-75% of cases, prompting the need for new therapeutic approaches targeting these genes [62]. The pathogenesis of adRP involves numerous genes, with RHO being the most studied. In this section, researchers will review new advances in other molecular targets.
RP is often caused by mutations in the RHO gene, affecting about 25% of patients [63]. To find a treatment for rhodopsin-associated RP, researchers have focused on understanding the pathogenesis and biomolecular changes caused by RHO mutations using cell lines and animal models, primarily rodents. With valid and robust animal models in place, scientific interest has shifted to developing gene therapy vectors.
In preclinical phases, gene therapy vectors for RP primarily target adRP, utilizing advances in mammalian models with rhodopsin mutations. However, recent in vivo models have revealed important insights into the pathogenesis of RP, including the inhibition of Müller glial cell proliferation in Xenopus laevis tadpoles with a CRISPRcas9-induced rho mutation [32], and the rapid destruction of the outer nuclear layer due to rhodopsin accumulation in a P347L transgenic rat strain [34]. In addition, the ER stress pathway was activated in P347L transgenic rats, as evidenced by significant upregulation of CHOP and BiP mRNA expression [34].
The P347S mutation in rhodopsin has been shown to impair transport to the photoreceptor outer segment in transgenic mice and pigs. Strategies to suppress the expression of the mutant transcript and replace it with a wild-type copy have shown promise in restoring retinal morphology and function. Specifically, transcriptional repression with the KRAB domain and zinc-finger DNA-binding domains has successfully decreased transcript levels of the mutant while preserving endogenous murine RHO gene expression in transgenic mice. RNA replacement therapies using shRNA delivery have also successfully replaced mutant rhodopsin RNA transcripts, leading to increased rhodopsin levels and improved ERG responses. The CRISPR-cas9 system has demonstrated success in allele-specific disruption of the RHOS334 gene in S334ter-3 rats, leading to increased retinal preservation and visual acuity. Additionally, NRL and NR2E3 gene disruptions with dual subretinal injection of AAV-cas9 vector have shown promising results in RHO-P347S/Rd10 and Rd10/Rd1 mice, respectively. Dysregulation of miRNA levels has also been implicated in RP pathogenesis, and AAV-mediated delivery of pre-miR-204 and AAV-mirtron vector in RHO-P247S transgenic mice and Nrl.GFP/+, RhoP23H/+ mice, respectively, have shown success in delaying degeneration and attenuating microglia activation and photoreceptor cell death. Lastly, inhibition of ceramide de novo synthesis with Myriocin has been shown to support cell survival and preserve ERG responses in Tvrm4 mice carrying a Rhodopsin gene mutation.
The T17M rhodopsin mutation activates the unfolded protein response (UPR) pathway, leading to ER stress and photoreceptor cell death [33][64]. In transgenic C57BL6 mice, T17M rhodopsin expression upregulated IL-1β, IL-6, and NFκB and the IB1A microglia markers, indicating UPR pathway activation and loss of photoreceptor function and retinal structure [33]. Knockdown of ATF4, a UPR pathway mediator, delayed retinal degeneration in T17M-induced RP mice, with double knockdowns having a more significant therapeutic effect, likely due to the attenuation of pEIF2α, ATF6, and CHOP expression [65].
Mutations in pre-mRNA processing factors (PRPFs) are linked to up to 20% of adRP cases [66]. PRPFs are involved in pre-mRNA splicing, ciliogenesis, and DNA damage repair pathways. Prpf31 gene augmentation using AAV-associated CRISPR-cas9 vector gene augmentation restored the integrity of retinal structure in mice [79]. Similarly, Rpgr gene-editing using sgRNA (cas9) and AAV vectors induced photoreceptor preservation in Rpgr−/yCas9+/WT mice [45].
Mutations in the RP1 gene, which is involved in adRP, disrupt protein transport within photoreceptors, cilial structure maintenance, and disc membrane stability, leading to photoreceptor cell death. Recently, dual Cas9/sgRNA technology has been shown to successfully reduce RP1 expression in edited Hap1-EF1a-RP1 cells, which could lead to further clinical trials.
RHO mutations are associated with both adRP and arRP, with only a few linked to the latter. The five known variants for rhodopsin-associated arRP are E150K, W161ter, E249ter, and M2531. The missense mutation E150K impairs rhodopsin stabilization, leading to retinal degeneration due to a disorganized photoreceptor disc structure and impaired phagocytosis. Homozygous E150K mice exhibit higher retinal immune activation. Two nonsense mutations (E249ter and W161ter) were also found to be involved in arRP, with lower levels of rhodopsin mRNA detected in human cells transfected with the rhodopsin nonsense mutants. The M2531 rhodopsin mutants were shown to be linked to the asymptomatic arRP form in heterozygous patients from consanguineous Turkish descents. Further studies are required to better understand the functions and impacts of autosomal recessive mutations in RP development.
CNGB1 sequence variants are associated with arRP, which accounts for almost 4% of arRP cases [67]. A recent literature review has linked a total of 62 genetic variants to inherited retinal diseases [67]. Restoring CNGB1 expression in Cngb1−/− mice using an AAV vector has been shown to improve vision. CNGB1-expressing mice had better performance in a rod-dependent vision-guided behavior test [46], and retinal degeneration was significantly delayed, with morphology preserved [46].
TULP1 mutation causes early-onset retinal degeneration, but its pathogenesis remains largely unknown [68]. To investigate the pathogenic mechanism of various retinal diseases, the zebrafish model has gained popularity in recent years [69]. Zebrafish models were generated with single-knockout (Tulp1+/−) and double-knockout for Tulp1, which significantly altered ciliogenesis by downregulating tektin2, a microtubular component [47]. AAV-mediated TULP1 gene expression editing successfully restored TULP1 mRNA and protein levels [48]. However, in Tulp1−/− mice, the gene editing process was not sufficient to improve the thickness of the outer nuclear layer [48].
Generating new animal models is a crucial step in developing gene therapy treatments. Beryozkin and colleagues have recently generated a knockout mouse model for Fam161a deficiency, which displays retinal degeneration and enhanced molecular generative markers such as microglia [49]. Homozygous Fam161a p.Arg512∗ have also been shown to have altered visual acuity due to complete loss of the outer nuclear layer and photoreceptor cell death [50].
The retinitis pigmentosa GTPase regulator (RPGR) gene is the most commonly studied gene and has been shown to cause recessive RP [70]. Promising clinical trials are being conducted for therapeutic approaches using AAV vectors, while preclinical phase studies are ongoing to develop other gene therapy methods. In the rd9 mice model, AAV-guided CRISPR-cas9 vector injection has been used to restore the reading frame of RPGRORF15 [44].
The identification of novel genes involved in RP pathogenesis is essential for future gene therapy development. Gene sequencing studies of large patient cohorts have identified potential gene targets. Genome sequencing studies in the past five years have highlighted the involvement of multiple genes in RP pathogenesis [71][72][73]. For instance, SLC66A1 and SLC39A12 were identified as novel targets in inherited retinal disease cases by analyzing sequence variants of SLC genes in Israeli and Palestinian cohorts [74]. SLC7A14 was also shown to be involved in retinal development as SLC7A14 knockout mice exhibited retinal degeneration and altered visual function [75]. In addition, rare heterozygous variants of AGBL5, p.Arg281Cys and p.Arg487*, were potentially implicated in RP [76]. However, further research is necessary to elucidate the pathophysiology of these novel genes in RP. The Retinal Information Network’s website (https://web.sph.uth.edu/RetNet/ (accessed on 20 January 2023)) provides an updated list of genes and mapped loci causing retinal diseases.