Pathomechanisms of Inherited Retinal Degeneration and Perspectives for Neuroprotection

Abstract

The precise processes causing photoreceptor cell death in retinal degeneration (RD) are still largely unknown but are likely to follow a variety of degenerative mechanisms. While different genetic insults can trigger distinct molecular pathways, eventually these may converge into a limited number of common cell death mechanisms. These mechanisms often involve deregulation of cyclic guanosine monophosphate (cGMP)-signaling and proteostasis, which both may increase photoreceptor energy expenditure. Comprehensive information on these mechanisms may allow for targeted interventions to delay or prevent photoreceptor loss. Here, we review the current knowledge on photoreceptor degenerative mechanisms, focusing on processes triggered by aberrant cGMP-signaling, proteostasis, and energy metabolism. Afterward, we discuss how these pathways could potentially be used to treat photoreceptor degeneration, highlighting data from a number of recent studies on inhibitory cGMP analogs, proteostasis blockers, and interventions aimed at fortifying energetic status. Finally, we provide perspectives on how such experimental approaches could be translated into future clinical applications.

PATHOPHYSIOLOGY OF RD: FROM NECROSIS AND APOPTOSIS TO ALTERNATIVE CELL DEATH MECHANISMS

Cell death was traditionally seen as a biological accident triggered by overwhelming physical stress (e.g., trauma, intoxication, oxidation), leading to unordered, chaotic destruction of the cellular machinery. For this phenomenon, the term “necrosis” was coined (Glücksmann 1951). However, in the past decades, especially through the works of Kerr, Wyllie, and Currie, who introduced the term “apoptosis” (Kerr et al. 1972), it has become increasingly clear that in multicellular organisms cell death is often a well-orchestrated and program-driven process, which has little to no resemblance to the original definition of necrosis. Since its introduction, the term apoptosis has in many ways become a synonym for cell death. However, it is important to understand that apoptosis is only one of several possible cell death mechanisms that a cell may employ (Galluzzi et al. 2018). Notably, apoptosis is typically associated with developmental processes (Ghose and Shaham 2020; Montero et al. 2022) while pathological cell death associated with inherited retinal degeneration (RD) is likely to be nonapoptotic (Arango-Gonzalez et al. 2014; Power et al. 2020).

In RD, but also in essentially all other neurodegenerative diseases, the execution of cell death may employ a variety of very different metabolic processes, depending on the type of trigger, developmental stage, energetic status, pathogen load, etc. (Leist and Jäättelä 2001). Consequently, many alternative cell death mechanisms could potentially be exploited for therapeutic purposes (Kepp et al. 2011). However, the targeting of these mechanisms is complicated by our current lack of knowledge and the possibility that this variety of cell death mechanisms may not be clearly delineated pathways but a continuum of processes and metabolic subroutines, with considerable ambiguity and flexibility in between (Galluzzi et al. 2018).

In preclinical RD research on animal models, another confounding factor arises from the fact that early mutation-induced degeneration often coincides with retinal development and may overlap with secondary and tertiary degenerative processes and retinal remodeling (Sancho-Pelluz et al. 2008; Strettoi 2015). Therefore, the cell death mechanisms seen in animal models must be thoroughly analyzed and related to developmental stages to assess to what extent they may reproduce the human disease condition.

As diverse as cell death mechanisms may be, they likely have at least a few things in common. Cell death may be caused either via an uncompensated loss-of-function or via a toxic gain-of-function mechanism. Loss-of-function may include hypomorphic or total loss of specific gene function that affects processes central to cellular integrity, such as intracellular ciliary transport between inner (IS) and outer segments (OS), or recycling of critical visual components like retinal between photoreceptor and retinal pigment epithelium (RPE) cells. Gain-of-function may include overactivity of certain signal transducers, such as cyclic guanosine monophosphate (cGMP), the aggregation and inclusion formation of critical gene products with subsequent activation of the unfolded protein response (UPR), mis-sorting and interference with the trafficking machinery, constitutive or dysregulated activation, or structural destabilization of the OS.

For the purposes of this review, we focus on degenerative mechanisms of photoreceptors that are related to pathologic changes in cGMP signaling, proteostasis, and energy metabolism. It is plausible to think that once the underlying mechanisms are sufficiently well understood, targeted interventions can be designed to prevent or delay photoreceptor degeneration. Accordingly, we provide an overview of recent studies on inhibitory cGMP analogs, proteostasis blockers, and interventions aimed at boosting energy production. Finally, we provide perspectives on how such experimental approaches could be translated into future clinical applications.

cGMP-SIGNALING IN PHOTORECEPTOR PHYSIOLOGY AND PATHOPHYSIOLOGY

The second messenger, cGMP, plays a key role in photoreceptor function. In the absence of light, cGMP is present at high levels in the photoreceptor OS where it maintains the cyclic nucleotide-gated channel (CNGC) in the open state, leading to an influx of Na⁺ and Ca²⁺ ions. Two main exchangers regulate the outflow of these ions: The Na⁺/Ca²⁺/K⁺ exchanger (NCKX) in the OS and the ATP-driven Na⁺/K⁺ exchanger (NKX) located in the IS (Tolone et al. 2019; Yan et al. 2021). The influx of positive ions through CNGC into the OS creates a depolarizing current, which is ultimately driven by the Na⁺ and K⁺ gradients produced by NKX in the IS (Vinberg et al. 2018). The constant flux of ions across the OS and IS membranes during darkness is referred to as the dark current, which likely is the most energy-intensive task in a photoreceptor (Wong-Riley 2010). Notably, NKX activity alone is responsible for at least 50% of the ATP consumption of a photoreceptor (Ames 1992).

cGMP is synthesized by retinal guanylyl cyclases (GCs), which are encoded by the GUCY2D and GUCY2F genes in human (Gucy2e in mouse), the activity of which is controlled indirectly by Ca²⁺ via GC-activating protein (GCAP) (Lim et al. 2018). High intracellular Ca²⁺ levels in the dark inhibit GCAP and limit cGMP production, thereby establishing a negative feedback loop that holds the photoreceptor cGMP level in its physiological range (Pugh and Lamb 1990; Burns et al. 2002; Dell'Orco et al. 2009). In light, cGMP levels are decreased by the activity of phosphodiesterase 6 (PDE6) and, consequently, the CNGC closes. While the influx of Na⁺ through CNGC is drastically reduced, NKX continues to pump Na⁺ out of the IS. This causes photoreceptor hyperpolarization (Vinberg et al. 2018), which spreads along the cell membrane to the synaptic terminal. Here, the change in membrane potential closes the voltage-gated Ca²⁺ channel (VGCC), consequently reducing intracellular Ca²⁺ (Waldner et al. 2018). In addition, hyperpolarization in bright light induces the opening of hyperpolarization-activated cyclic nucleotide-gated channel-1 (HCN1) expressed in the IS. HCN1 opening generates an inward current of Na⁺ and K⁺ ions that depolarize the cell, decreasing the refractory period and increasing the temporal resolution of photoreceptors (Tanimoto et al. 2012). Overall, photoreceptor physiology rests on the balance between cGMP and Ca²⁺, linked via the Ca²⁺-GCAP negative feedback loop and whose disruption appears to be a feature common to several forms of RD (Olshevskaya et al. 2002).

Although, in the past decades, research on RD focused on the presumed damaging role of excessive intracellular Ca²⁺ levels and apoptosis (Orrenius et al. 2003; Zeiss et al. 2004), in recent years, the focus has shifted toward nonapoptotic and Ca^2+--independent mechanisms of cell death. These mechanisms are characterized by an abnormal accumulation of cGMP in photoreceptors (Power et al. 2020). This phenomenon has been observed in mouse models with loss-of-function mutations in the Aipl1, Prph2, Cngb1, Cnga3, Pde6a, Pde6b, Pde6c, and Rho genes and excessive photoreceptor cGMP levels are likely found in additional RD-disease genes (Arango-Gonzalez et al. 2014; Power et al. 2020).

Excessive cGMP signaling may increase the activity of one or more of its three known photoreceptor targets: CNGC, HCN1 (Biel and Michalakis 2009), and protein kinase G (PKG). While CNGC overactivation has been associated with increased Ca²⁺ levels in the cell (Paquet-Durand et al. 2011; Kulkarni et al. 2016), the direct effects of PKG overactivation have yet to be elucidated. Indeed, as a protein kinase, PKG has many potential phosphorylation targets. To date, we know that downstream of excessive cGMP and PKG signaling, there is an increase in the enzymatic activities of histone deacetylase (HDAC), poly(ADP-ribose) polymerase (PARP), and Ca²⁺-dependent calpain-type proteases, even though for HDAC and PARP there is still no evidence linking them directly to PKG or elevated Ca²⁺ levels (Arango-Gonzalez et al. 2014; Das et al. 2021).

Recent studies identified several PKG targets with a potential role in cell death in RD (Roy et al. 2022). These include 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase-3 (PFKFB3) and voltage-dependent potassium channels belonging to the Kv1 family. Both targets were found to be expressed in the retina and were more strongly phosphorylated in the rd1 situation (Roy et al. 2022). Increased phosphorylation of Kv1 channels implies an abnormal increase in potassium leakage from the cell, which would raise energy consumption to bring the equilibrium potential of K⁺ back to a physiological range. PFKFB3 converts fructose-6-phosphate to fructose-2,6-bisP (F2,6BP). F2,6BP is a potent allosteric activator of 6-phosphofructokinase-1 (PFK-1), stimulating glycolysis (Clem et al. 2008; Shi et al. 2017). A PKG-dependent change of PFKFB3 activity thus has the potential to strongly impact retinal energy metabolism, by shifting ATP-production away from the efficient Krebs cycle and oxidative phosphorylation toward the highly inefficient glycolysis (Chen et al. 2022). Thus, excessive activation of PKG by high photoreceptor cGMP-levels could, on the one hand, increase energy expenditure and at the same time reduce the efficiency of ATP-production. Overall, this has the potential of causing a serious “energy crisis” in photoreceptor cells.

PROTEOSTASIS, PROTEASOMAL DEGRADATION, AND AUTOPHAGY

Proteostasis comprises the folding, assembly, and degradation pathways that maintain a balance of functional proteins within cells. The retina can be exposed to a variety of detrimental factors, including light-induced damage, oxidative stress, or inherited mutations, each of which can lead to perturbations of proteostasis (Athanasiou et al. 2013; Zhang et al. 2014). Photoreceptors seem especially vulnerable to disruptions in protein homeostasis, notably due to the massive opsin production and high metabolic activity. This production of large quantities of the visual pigment rhodopsin in photoreceptors places a substantial strain on the protein-folding machinery, resulting in misfolded/unfolded protein accumulation and protein aggregation. Indeed, misfolded protein aggregates are a central pathological phenotype in several neurodegenerative diseases (Moreno-Gonzalez and Soto 2011), including in RD (Griciuc et al. 2010, 2011; Tzekov et al. 2011).

Cellular proteostasis is regulated by a complex network of interacting and competitive biological modules (Jayaraj et al. 2020). An imbalance between these modules will activate integrated response pathways such as the UPR, endoplasmic reticulum–associated degradation (ERAD), and autophagy, likely in an attempt to induce signaling and transcriptional changes to adapt to the stressor(s). However, if the imbalance is prolonged or too excessive, the activation of death pathways will reduce photoreceptor cell viability and function (Athanasiou et al. 2013).

The endoplasmic reticulum (ER) is the site of extensive protein synthesis, folding, maturation, and export, and it has been equipped with highly specialized systems that detect and correct abnormal protein-folding states to prevent pathology. Accumulation of misfolded proteins in the ER membrane and lumen provokes UPR (Walter and Ron 2011). The unfolded proteins are sensed by a signaling pathway that includes inositol requiring enzyme 1 (IRE1), protein kinase RNA-like ER kinase (PERK), and activating transcription factor 6 (ATF6) transmembrane receptors, which transmit signals aiming to (1) increase ER protein-folding capacity, (2) decrease ER protein-folding load, and (3) promote the clearance and degradation of misfolded proteins. To cope with misfolded or unfolded proteins, cells have evolved discrete mechanisms, which comprise two major proteostasis networks: (1) ERAD, and (2) the autophagy pathway (Griciuc et al. 2011; Smith et al. 2011).

ERAD

In this process, the unfolded or misfolded protein in the ER is recognized by molecular chaperones and targeted to the retrotranslocation machinery (retrotranslocon) (Wu and Rapoport 2018). Subsequently, the protein is ubiquitinated by one of two ER membrane–localized ubiquitin ligases and finally extruded through the retrotranslocation channel to the cytosol. This process depends on ATP hydrolysis by the ATPase associated with diverse cellular activities (AAA) known as valosin-containing protein (VCP) or p97 (Ye et al. 2001). Once released into the cytosol, the protein is delivered to the proteasome for degradation (Griciuc et al. 2010, 2011). Failure of this degradation pathway will lead to the accumulation of polyubiquitinated protein in the cytosol.

Autophagy

An alternative and possibly complementary protein degradation route in photoreceptors is autophagy (Mizushima 2007). This pathway might benefit proteins with a high propensity to aggregate within the ER, such as rhodopsin. Autophagy is a tightly regulated process in cell growth and development that creates and maintains a balance between the synthesis, degradation, and recycling of cellular products (Griciuc et al. 2011). There are three main autophagic processes: macroautophagy, microautophagy, and chaperone-mediated autophagy (Kroemer and Levine 2008). They all degrade intracellular components via the lysosome. The most well-known mechanism of autophagy is macroautophagy, which mediates the degradation of damaged organelles or proteins, and involves forming a double-membrane structure (phagophore), which becomes an autophagosome through lipid acquisition (Baba et al. 1994). The autophagosome then fuses with the lysosome, forming autolysosomes, degrading, and recycling the cargo (Lamb et al. 2013). The initiation of macroautophagy is tightly regulated by AMP-activated protein kinase (AMPK) and the mammalian target of rapamycin complex 1 (mTORC1) (Wong et al. 2013). Incidentally, AMPK is also a key regulator of cellular energy metabolism (Xu and Ash 2016).

In addition, an imbalance in proteostasis may also lead to a Ca²⁺ imbalance and mitochondrial dysfunction through excessive ER stress (Cao and Kaufman 2014). Production of reactive oxygen species (ROS) and release of cytochrome c from dysfunctional mitochondria to the cytosol may initiate mitochondria-mediated cell death (Shore et al. 2011). Ultimately, such mitochondrial dysfunction will compromise the energy metabolism to the extent that key cellular functions can no longer be maintained.

THE ROLE OF ENERGY METABOLISM FOR PHOTORECEPTOR CELL DEATH AND SURVIVAL

The retina is one of the most metabolically active tissues in the body (Trick and Berkowitz 2005) and the majority of its large energy demand is likely caused by the photoreceptor dark current (Okawa et al. 2008). Yet even today very little is known about how photoreceptors satisfy their energetic needs, what fuels they use, and how they obtain these. A recent hypothesis proposed that photoreceptors use predominantly glucose and aerobic glycolysis, with the resultant lactate used by the RPE and Müller glial cells for mitochondrial respiration (Kanow et al. 2017; Viegas and Neuhauss 2021). However, this hypothesis is contradicted by the inefficiency of aerobic glycolysis and the high density of mitochondria in photoreceptor IS (Giarmarco et al. 2020).

In principle, a cell may generate energy-containing substrates such as ATP via two main pathways: (1) the very rapid glycolysis in the cytoplasm, and (2) the comparatively slow mitochondrial Krebs cycle coupled to oxidative phosphorylation. Retinal cells that have access to a large supply of lipids, such as RPE cells, can also use β-oxidation for their energy production via oxidative phosphorylation. The energetic advantage of oxidative metabolism in the mitochondria far outweighs that of glycolytic metabolism in the cytosol, as one molecule of glucose yields only two molecules of ATP via the glycolytic pathway but up to 36 molecules of ATP via oxidative phosphorylation. Paradoxically, the retina has long been suggested to prefer aerobic glycolysis (Warburg 1925) (i.e., the conversion of glucose to lactate, even when O₂ is abundant, a phenomenon called the “Warburg effect”) (Diaz-Ruiz et al. 2011). The incomplete glucose oxidation is suggested to promote anabolic activity, such as synthesis of nucleic acids, proteins, and lipids (Lunt and Heiden 2011). Since photoreceptor disk shedding, subsequent OS growth, and the corresponding lipid and protein synthesis follow the dark–light cycle (LaVail 1980), it is tempting to speculate that the switch from oxidative metabolism to aerobic glycolysis could be driven by cGMP- and PKG-dependent regulation of PFKFB3 activity.

AMPK is another key regulator of cell metabolism (Herzig and Shaw 2018). AMPK is triggered through an allosteric mechanism that stimulates its kinase activity in response to variations in energy availability, notably changes in the ATP-to-AMP ratio. Once activated, AMPK redirects metabolism toward increased catabolism and decreased anabolism through the phosphorylation of key proteins in multiple pathways, including mTORC1, lipid homeostasis, and glycolysis (Bando et al. 2005; Gwinn et al. 2008; Ahmadian et al. 2011). In addition, the insulin/mTOR pathway promotes anabolic processes such as protein synthesis and ribosome biogenesis under conditions of high cellular energy (Iadevaia et al. 2012). In RD, high levels of cGMP keep CNGC open, resulting in a continuous ion influx to photoreceptors (Das et al. 2021) and resembling a situation of constant darkness. To counterbalance the resultant constant dark current, ATP must be consumed (Okawa et al. 2008), leading to a net increase in energy expenditure.

Moreover, photoreceptor degeneration in 10 different animal models was found to be connected to the overactivation of PARP (Arango-Gonzalez et al. 2014), an enzyme that consumes nicotinamide dinucleotide (NAD⁺) to mediate DNA repair (Bai 2015). Excessive consumption of NAD⁺ by overactivated PARP-1 results in a reduced redox capacity of cells, decreasing ATP-production. Furthermore, the production of poly(ADP-ribose) polymers may compromise mitochondrial function also directly (Baek et al. 2013). Eventually, an imbalance between energy consumption and production, perhaps aggravated by an unwarranted switch from mitochondrial respiration to aerobic glycolysis, will lead to the breakdown of a cell's membrane potential and subsequently to cell death. Therefore, retinal energy metabolism and its regulation probably have direct consequences for RD disease pathogenesis.

TARGETS FOR THERAPEUTIC INTERVENTIONS AND THERAPY DEVELOPMENT

An improved understanding of the cellular mechanisms governing photoreceptor degeneration will likely enable the development of novel therapies targeting these mechanisms. Here, we detail three possible avenues for exploiting mechanistic knowledge to improve photoreceptor viability. These approaches concern intervening with cGMP-signaling, protein homeostasis, and photoreceptor energy metabolism.

Targeting CNGC and/or PKG

Rising cGMP levels appear to be an early event in photoreceptor cell death (Farber and Lolley 1974) and therefore neuroprotective strategies targeting excessive cGMP signaling may be well suited to prevent or delay the progression of RD. Downstream targets of cGMP signaling include CNGC and PKG, which could hence be addressed for therapeutic purposes.

Knockout studies on double-mutant animals carrying genetic defects that disable both CNGC and PDE6 function (i.e., Cngb1^−/− × rd1 and Cngb1^−/− × rd10) hinted at an important role of CNGC in RD (Paquet-Durand et al. 2011; Waldner et al. 2018). Nevertheless, targeting CNGC for therapeutic purposes faces important challenges. First, cones and rods express different isoforms of CNGC (Das et al. 2021), and the potential inhibitor should be specific for a given isoform. For example, in retinitis pigmentosa (RP), only rod-specific channels should be inhibited to avoid impairing cone-mediated vision. Furthermore, as CNGCs are involved in visual signal transduction, their inhibition could cause detrimental effects on vision in the long run. However, pharmacological inhibition of CNGC is possible via, for example, the L-cis enantiomer of the antihypertensive drug diltiazem (Das et al. 2022). An alternative approach may be the use of antisense oligonucleotides, which would allow selective knockdown of the CNGC isoform of interest (e.g., Cngb1).

In contrast to CNGCs, PKG appears not to be expressed differently in cones and rods, and its inhibition should not affect phototransduction (Ekstrom et al. 2014; Vighi et al. 2018). These two aspects make PKG an attractive candidate for targeting excessive cGMP signaling. There are several types of PKG inhibitors on the market. The first group of inhibitors includes KT5823 and the derivative of the fungal metabolite balanol N46, which block PKG activity by targeting the ATP-binding site (Olivares-González et al. 2016; Qin et al. 2018). Another type of peptide-based PKG inhibitors can bind to the kinasés substrate recognition site (Nickl et al. 2010). Finally, an additional class of PKG inhibitors are cGMP analogs that target the cGMP-binding domains (Lohmann et al. 1997) present in no other kinase type. Such cGMP analogs are particularly effective in slowing photoreceptor degeneration in vitro and in vivo in rd1, rd2, and rd10 mouse models (Vighi et al. 2018), boding well for the development of pharmacological therapies.

Recent studies showed that pharmacological inhibition of CNGCs alone is not sufficient to prevent RD and can, in fact, accelerate photoreceptor loss (Das et al. 2022). On the other hand, cGMP analogs that inhibit PKG and are effective in vitro and in vivo are also mild inhibitors of CNGCs (Tolone et al. 2021). This suggests that a synergistic action on the different events downstream of excessive cGMP signaling might be an appropriate strategy for the development of new treatments for RD.

Rebalancing Photoreceptor Proteostasis

As mentioned above, RD can disturb the folding capacity of cellular proteins and lead to the accumulation of misfolded proteins in the ER, causing ER stress and subsequent photoreceptor cell death. Specifically, in RP, misfolded proteins, as triggered, for instance, by the P23H mutation in the rhodopsin gene, disrupt cellular proteostasis and lead to photoreceptor cell degeneration. Since dysbalanced proteostasis is a common feature of RD, strategies to manipulate its network, like targeting ERAD, autophagy, or UPR, are particularly interesting as a neuroprotective approach.

Regulation of proper protein integrity and folding is crucial for cellular processes to maintain proteostasis (Díaz-Villanueva et al. 2015). In fact, inhibition of the HSP90 protein, responsible for proper protein folding in the ER, protects photoreceptor cells in models for autosomal-dominant RP (Aguilà et al. 2014). Targeting upstream ER stress sensors via pharmacological chaperones can improve protein folding and promote misfolded P23H rod opsin trafficking (Athanasiou et al. 2018). The AMPK activator metformin improved P23H folding and trafficking and reduced cell death in cell culture models; however, metformin accelerated the progression of RD in the P23H-1 transgenic rat model and in P23H knockin (KI) mice (Athanasiou et al. 2017). Chemical chaperones, also known as Kosmotropes (Wiggins 2001), can bind to proteins nonspecifically and improve the stability in their native conformation, reducing aggregation caused by misfolded rhodopsin (Athanasiou et al. 2018). For example, in vitro studies have shown that sodium 4-phenylbutyrate (4-PBA) could decrease P23H rod opsin aggregation (Mendes and Cheetham 2008) and reduce UPR-signaling and ER stress associated with the T17M rod opsin mutation (Mendes and Cheetham 2008). Importantly, administration of 4-PBA in the Rpe65^R91W/R91W mouse model of Leber's congenital amaurosis (LCA) improved cone survival and vision (Li et al. 2016). However, 4-PBA did not protect against photoreceptor degeneration in the P23H rat model (Athanasiou et al. 2018). Another molecule, curcumin, a polyphenolic agent derived from the Curcuma longa plant, was reported to improve photoreceptor function and morphology in P23H rats, reduce P23H aggregation, and enhance rhodopsin localization to the rod OS (Vasireddy et al. 2011). The chemical chaperone tauroursodeoxycholic acid (TUDCA), which inhibits apoptosis by preventing the proapoptotic protein Bax from being transported to the mitochondria (Boatright et al. 2009), preserved photoreceptor structure and function in an LCA animal model (Fu and Zhang 2014), in the rd1 (Lawson et al. 2016) and rd10 mouse models of RP (Oveson et al. 2011; Drack et al. 2012), and preserved photoreceptor function and survival in the P23H-1 rat model (Athanasiou et al. 2018).

Treatment with arimoclomol, which amplifies the natural response to cellular stress by inducing the heat-shock response, also slowed down the progression of RD in P23H rats (Parfitt et al. 2014). Inhibition of the PERK branch of the UPR accelerated RD in P23H rats, suggesting that PERK activation in this model for autosomal-dominant RP was protective (Athanasiou et al. 2018). On the other hand, intravitreal administration of KIRA6, an IRE1-dependent mRNA decay inhibitor, appeared to preserve photoreceptor numbers in P23H-1 rats (Ghosh et al. 2014), suggesting that activation of the UPR was detrimental.

Overexpression of the ER-resident chaperone network between ER degradation–enhancing α-mannosidase-like 1 (EDEM1), the ER-resident protein-containing DNAJ domain (ERdj5; also known as DnaJ homolog subfamily C member 10 [DNAJC10]), and the UPR regulator-binding immunoglobulin-protein (BiP) has been shown to reduce P23H aggregation in vitro (Kosmaoglou et al. 2009; Athanasiou et al. 2012, 2014). BiP overexpression also reduced photoreceptor cell death by reducing ER stress in P23H rats (Gorbatyuk et al. 2010). Additionally, modulation of the proteostasis by VCP or proteasome inhibitors rescued photoreceptor cell degeneration in P23H rats and P23H KI mice (Arango-Gonzalez et al. 2020; Sen et al. 2021a,c). Moreover, inhibition or down-regulation of VCP promoted the correct localization of rhodopsin to the OS upon treatment in vitro as well as in vivo (Arango-Gonzalez et al. 2020; Sen et al. 2021b,c). Further studies have shown that inhibition of VCP can exert neuroprotection in animal models for other retinal pathologies, for example, autosomal recessive RP (Ikeda et al. 2014), glaucoma (Nakano et al. 2016), and retinal ischemia (Hata et al. 2017).

In RP, the autophagy pathway also appears to be involved in the degradation of misfolded rhodopsin (Wen et al. 2019), and the inactivation of autophagy leads to accumulation of rhodopsin and RD (Yao et al. 2016). Treatment with the autophagy-inducing drug rapamycin reduced P23H aggregation in vitro (Mendes and Cheetham 2008), while systemic administration of rapamycin to P23H rats slowed down rod cell degeneration (Sizova et al. 2014). In contrast, normalization of the autophagic flux in the retina of P23H KI mice relative to proteasome activity may support photoreceptor cell homeostasis, resulting in increased photoreceptor cell survival (Qiu et al. 2019).

Improving Energy Metabolism

The crucial role of retinal energy metabolism could potentially make it a target for therapeutic interventions. Such interventions could, for instance, address defects in nutrient supply, alleviate the effects of high oxidative metabolism, or redirect cellular metabolism toward increased energy production. For instance, elevated insulin levels increased glucose uptake in cones, activated cone mTOR signaling, and delayed the secondary loss of cones in four different models for RP (Punzo et al. 2009). Potentially, the regulation of glycolysis via PFKFB3 could be used to similar effect (Shi et al. 2017).

An alternative way of influencing photoreceptor energy metabolism is through the use of both antioxidants and amino acids, some of which are already used in dietary supplements to help prevent RD or age-related macular degeneration (AMD) (Grover and Samson 2014; Gastaldello et al. 2022). Notably, antioxidants may help the detoxification of ROS thought to be generated as byproducts of metabolic processes. Excess generation of ROS may lead to oxidative damage to different targets, including lipids and enzymes involved in ATP synthesis (Quijano et al. 2016). Lutein is an antioxidant belonging to the carotenoid group of compounds with antioxidative and anti-inflammatory properties. Its beneficial effects have been studied in many diseases, such as AMD and diabetic retinopathy (Li et al. 2020). Apart from carotenoids, nutraceuticals such as polyphenols, saponins, and other compounds have revealed protective effects in retinal disorders (Rossino and Casini 2019). Fatty acids, such as omega-3, are also reported for their protective roles in the retina (Querques et al. 2011). An additional (exogenous) supply of fatty acids could be applicable to cells with access to a large supply of lipids, such as RPE cells, which can then use β-oxidation for their energy production via oxidative phosphorylation (Adijanto et al. 2014). Therefore, the use of these compounds may represent a natural alternative method to conventional retina disease treatments (Fig. 1).

Figure 1.

Degenerative mechanisms in retinal degeneration (RD) and therapeutic approaches. The ensemble of cell death mechanisms leading to photoreceptor degeneration likely forms a continuum that may be structured according to the severity of the causative genetic insult. While apoptotic cell death is typically associated with retinal development, RD-causing mutations may initially affect cyclic guanosine monophosphate (cGMP)-signaling and proteostasis. Eventually, different genetic triggers may converge on excessive photoreceptor energy consumption, ultimately resulting in cell death. (PKG) Protein kinase G, (CNGC) cyclic nucleotide-gated channel, (VCP) valosin-containing protein, (PARP) poly(ADP-ribose) polymerase.

Moreover, genetic alterations modifying photoreceptor metabolism so as to improve resistance to stress have been investigated in animal models (Caruso et al. 2020). Such approaches include overexpression of transforming growth factor β (TGF-β), rod-derived cone viability factor (RdCVF), ablation of sirtuin-6, pyruvate kinase M2 (PKM2), and the phosphatase and tensin homolog (PTEN) (Byrne et al. 2015; Venkatesh et al. 2015; Zhang et al. 2016, 2020; Wang et al. 2020).

Taken together, balancing the energy metabolism of photoreceptors, as well as reducing their overall energy consumption caused by high metabolic and proteostatic activity, may constitute a mutation-independent method for the treatment of RD. While relatively coarse systemic approaches, such as the addition of antioxidants in the diet, may show some effectiveness, more sophisticated and targeted approaches designed to improve energy production specifically in photoreceptors or the RPE may have significant potential in the future.

CONCLUSION: PERSPECTIVES FOR FUTURE CLINICAL DEVELOPMENT

Alterations of cGMP-signaling, ion currents, proteostasis, or lack of energy equivalents can cause photoreceptor cell death (Power et al. 2020; Yan et al. 2021; McLaughlin et al. 2022). Different pathological mechanisms may require different treatment approaches, yet there is a demand for mutation-independent treatment given the high number of distinct mutations affecting retinal function. Moreover, there is an advantage in having larger cohorts of well-stratified patients for clinical studies, especially when compared to gene-specific treatment for rare forms of RD, where the number of patients that can be recruited for a study may be very small (Schneider et al. 2022). Given the multitude of causative genetic defects in RD, gene therapy is unlikely to be available for most patients within this decade. Here, the development of mutation-independent interventions in cGMP-signaling, proteostasis, and energy metabolism may provide an alternative option to preserve a patient's vision until gene-specific treatments become available. Nevertheless, it will be of major importance to carefully select only those patients that are likely to benefit from a particular treatment to reach statistically meaningful clinical end points.

Another major concern in any therapy for retinal diseases is the route of administration and the delivery of the treatment to the affected photoreceptors or RPE cells (Himawan et al. 2019). While recent years have seen the development of relatively targeted adeno-associated viruses (AAV) for gene therapy (Botto et al. 2022), the targeted delivery of pharmacological agents still faces important problems. When delivered systemically, drugs need to be able to cross the blood–retinal barrier and achieve a sufficient and long-lasting concentration in the target cell. This also applies when a local administration to the eye (e.g., intravitreal injection, topical administration) is chosen, as the administration intervals will need to be as long as possible to allow for adequate patient compliance. Smart drug delivery systems are needed for long-term sustained release, such as targeted liposomes, lipid nano-capsules, poly(lactic-co-glycolic acid)-based, methoxy-poly (ethylene glycol) 5 kDa-cholane (mPEG 5 kDa-cholane)-based, or cyclodextrin-based (Yang et al. 2020; Christensen et al. 2021; Prajapati et al. 2021; Sen et al. 2021a). Local delivery to the eye will also limit systemic side effects and improve the overall safety profile for a given treatment.

In conclusion, profound insight into the cell death mechanisms that govern photoreceptor loss in RD is needed to advance current strategies toward mutation-independent clinical treatments. As disease mechanisms may be diverse, in-depth knowledge on disease etiology is key to define the right targets as well as to ascertain appropriate patient selection for clinical trials.