Pyrroloquinoline Quinone Protects Against Light-Induced Retinal Damage in Association with the Suppression of c-Fos Signalling

TL;DR

Pyrroloquinoline quinone (PQQ) protects the retina from light-induced damage by suppressing c-Fos signaling, suggesting it could be a treatment for age-related macular degeneration.

Contribution

This study is the first to demonstrate PQQ's protective effects in an in vivo model of AMD through c-Fos signaling suppression.

Findings

PQQ pretreatment reduced ATR-induced cytotoxicity in ARPE-19 cells in a dose-dependent manner.

Rats treated with 5 mg/kg PQQ showed significantly better retinal function and preserved photoreceptor layers after light exposure.

PQQ suppressed light-induced c-Fos upregulation in a dose-dependent manner, linking its protective effects to this signaling pathway.

Abstract

Age-related macular degeneration (AMD) is a progressive retinal disorder characterised by oxidative stress and inflammation. Although pyrroloquinoline quinone (PQQ) has been reported to exert neuroprotective effects, its specific efficacy in in vivo models of AMD pathophysiology has not yet been elucidated. In this study, we evaluated the protective effects of PQQ against all-trans-retinal (ATR)-induced cytotoxicity in ARPE-19 cells and light-induced photoreceptor degeneration in rats. Pretreatment of ARPE-19 cells with PQQ dose-dependently mitigated ATR-induced cytotoxicity. In the in vivo model, rats received a single intraperitoneal injection of PQQ (2 or 5 mg/kg) 1 h prior to 1000-lux light exposure. Retinal function and morphology were evaluated by electroretinography and haematoxylin–eosin staining, respectively. The 5 mg/kg PQQ group retained significantly greater retinal function than the vehicle group at 3 days postexposure and demonstrated significant preservation of the outer nuclear layer at 7 days postexposure, indicating the suppression of photoreceptor cell death. Western blot analysis detected the dose-dependent suppression of light-induced c-Fos upregulation following PQQ treatment. These findings suggest that the protective effect of PQQ against phototoxic damage is associated with the suppression of c-Fos signalling, thus lending support to the further investigation of PQQ as a potential therapeutic agent for AMD.

Figures

Full text

1. Introduction

Age-related macular degeneration (AMD), a progressive retinal disease, is a leading cause of irreversible visual impairment and blindness among older adults [1,2]. According to recent consensus on nomenclature, AMD is clinically classified into two subtypes: neovascular and atrophic AMD [3]. Although anti-vascular endothelial growth factor therapies have significantly improved outcomes for neovascular AMD, atrophic AMD (85–90% of all AMD cases) currently has no definitively approved treatment [4,5]. Given the ageing of the global population, AMD prevalence is projected to increase significantly; this situation underscores the urgent need for effective therapeutic strategies, particularly for dry AMD [6,7].

AMD has a complex pathophysiology characterised by the formation of extracellular drusen between the retinal pigment epithelium (RPE) and Bruch’s membrane [8,9], a process that is closely linked to RPE dysfunction, chronic oxidative stress, and subsequent inflammatory responses [9,10,11]. Furthermore, mitochondrial dysfunction in the RPE and photoreceptors has been increasingly recognised as a central pathogenic factor of AMD progression [12,13]. To investigate these mechanisms, various animal models, including chemical degeneration [14] and genetic models [15], have been developed. Among these, models for continuous light-induced photodamage have been widely adopted for atrophic AMD because they effectively replicate the features of oxidative stress and photoreceptor degeneration observed in human diseases [16,17,18,19].

Pyrroloquinoline quinone (PQQ), a quinone cofactor first discovered in bacterial dehydrogenases [20], has recently emerged as a micronutrient of considerable interest. PQQ exhibits potent scavenging activity against reactive species, such as free radicals and singlet oxygen, which are central to the oxidative stress implicated in AMD [21,22]. Furthermore, PQQ promotes mitochondrial biogenesis, enhances mitochondrial function, and exerts broad neuroprotective effects [23,24,25]. These properties make PQQ a compelling candidate for therapeutic intervention in AMD.

Given its potent antioxidant and mitochondrial-protective properties, PQQ holds significant promise as a therapeutic candidate for atrophic AMD. Although PQQ has been reported to confer general neuroprotective effects, its specific efficacy in an in vivo model of AMD pathophysiology, such as visual function changes in an AMD animal model, has not yet been elucidated. This study aimed to identify the effects of PQQ on the pathophysiology of light-induced retinal degeneration and evaluate its protective potential and possible therapeutic applications in atrophic AMD.

2. Results

2.1. PQQ Protects ARPE-19 Cells from All-Trans-Retinal (ATR)-Induced Cytotoxicity

We first investigated the potential of PQQ to protect RPE cells from the cytotoxicity induced by ATR, a toxic byproduct of the visual cycle associated with oxidative stress. Figure 1A shows a summary of the experimental protocol and the results of the cell viability assay. ARPE-19 cells were pretreated with various concentrations of PQQ for 2 h, followed by exposure to cytotoxic doses of ATR for 24 h. Thereafter, cell viability was assessed using a 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay. Exposure to ATR alone significantly reduced cell viability to approximately 50% of that of the untreated control group, which served as a positive control for cytotoxicity. By contrast, pretreatment with PQQ conferred robust protection against ATR-induced cell death. This protective effect was observed in a concentration-dependent manner, with cell viability significantly higher than the 50% baseline in concentrations ranging from 6.25 µM to 25 µM (Figure 1A). These findings indicate that PQQ protects RPE cells against cytotoxic insults.

2.2. Measurement of Retinal Thickness by Optical Coherence Tomography (OCT)

To determine whether PQQ affects the retinal structure in vivo, we assessed retinal thickness using OCT at baseline and 1 day after light exposure. Figure 2A shows the representative OCT images from groups treated with vehicle (phosphate-buffered saline [PBS]), 2 mg/kg PQQ, and 5 mg/kg PQQ. At baseline, outer nuclear layer (ONL) thickness did not significantly differ among the groups. One day after light exposure, all groups exhibited significant thinning of the retina compared to baseline, confirming the successful induction of photodamage. However, there were no statistically significant differences in ONL thickness between the vehicle-treated and PQQ-treated groups (2 and 5 mg/kg) in the superior and inferior retinas (Figure 2B).

2.3. PQQ Preserves Electroretinographic Responses in Light-Induced Photoreceptor Degeneration

To assess retinal function, we performed electroretinography (ERG) to measure the photoreceptor-derived a-waves and inner retinal cell-derived b-waves at baseline and 3 days after light exposure. Figure 3A shows the representative ERG waveforms recorded at baseline and following light exposure. Consistent with structural damage, light exposure induced a functional deficit in the vehicle-treated group. However, PQQ treatment preserved the retinal function, as shown by quantitative analysis. Although no significant difference was detected between the vehicle and 2 mg/kg PQQ groups, the 5 mg/kg PQQ group exhibited significantly higher amplitudes of both the a- and b-waves than the vehicle and 2 mg/kg groups (Figure 3B). These results indicate that PQQ effectively preserved retinal function from light-induced damage.

2.4. PQQ Protects the Photoreceptor Cells from Light-Induced Photoreceptor Degeneration

Retinas were enucleated 7 days following light exposure, fixed, and stained with haematoxylin and eosin (HE). No significant differences were observed in ONL thickness between the vehicle and 2 mg/kg PQQ groups at any measured location. By contrast, the 5 mg/kg PQQ group exhibited a significant preservation of ONL thickness (Figure 4A). Representative photomicrographs further illustrated the retention of ONL in the 5 mg/kg PQQ group (Figure 4B). These results provide histological evidence for the protective effect of PQQ on photoreceptors.

2.5. Western Blot Analysis of c-Fos and Phosphorylated Akt (pAkt) Expression

To elucidate the molecular mechanisms underlying the protective effects of PQQ, we performed Western blot analysis of retinal extracts 1 h after light exposure. Exposure to 1000-lux light induced a rapid and transient increase in the expression of the immediate-early gene Fos, with levels peaking at 1 h postexposure in the time-course analysis (Figure 5A). At this time point, light exposure markedly upregulated c-Fos expression in the vehicle-treated group; however, treatment with PQQ significantly suppressed this response (Figure 5B). Concurrently, we investigated the activation of the pro-survival Akt signalling pathway by measuring pAkt levels. In sharp contrast to the vehicle group, the PQQ-treated group had retinas exhibiting a significant increase in pAkt levels (Figure 5C). These results suggest that the retinal protection conferred by PQQ is associated with the modulation of early signalling events, including the suppression of c-Fos expression and the activation of the pro-survival Akt pathway.

3. Discussion

This study aimed to evaluate the protective effects of PQQ against light-induced photoreceptor degeneration. The data suggest that PQQ modulates the pathways involved in light-induced photoreceptor degeneration, resulting in the partial inhibition of its progression. Moreover, PQQ protected ARPE-19 cells from ATR-induced toxicity, indicating its antioxidant properties. Studies have reported that antioxidant agents suppress light-induced photoreceptor degeneration [26,27,28]. Consistent with these findings, the present study suggests that the antioxidant activity of PQQ contributed to the suppression of light-induced photoreceptor degeneration.

At 24 h post-light exposure, no significant structural differences were observed by OCT between the superior and inferior retinas, and there was no discernible protective effect of PQQ. This observation is likely attributable to the nascent stage of photoreceptor degeneration, where substantial structural changes, such as ONL thinning, have not yet manifested. This finding is consistent with previous reports [29] showing a clear temporal dissociation between functional decline and morphological alterations in light-induced retinopathy models. For instance, severe retinal dysfunction, as measured by ERG, is already present at a time point when morphological changes in the ONL are still minimal or statistically insignificant [30]. Therefore, the absence of OCT-detectable changes in the current study suggests that the structural endpoints at such early time points might not be sufficiently sensitive for capturing the initial wave of photoreceptor distress and its potential mitigation. In contrast, ERG measurements conducted 3 days after light exposure revealed the onset of functional deterioration. ERG responses are markedly reduced by light-induced retinal damage [18,19,27]. In the current study, a significant preservation of ERG responses was observed in the high-dose PQQ group, thereby providing functional evidence for the photoreceptor-protective effect of PQQ.

Light-induced photoreceptor degeneration is a well-established model for drug screening and for investigating the mechanisms underlying photoreceptor cell death. Studies have indicated that light-induced photoreceptor death is more pronounced in the superior retina than in the inferior retina [17,30,31,32]. Our histological findings were consistent with this pattern. Furthermore, the factors contributing to the severe degeneration of the superior retina include the reduced expression of basic fibroblast growth factor, which is a neuroprotective factor that confers retinal protection [28,33], and the differences in the angle of light incidence [34]. However, the complete mechanism of action remains unclear.

c-Fos expression increases after LD [35]. Studies have reported that mice lacking c-Fos do not exhibit light-induced photoreceptor cell death [36,37], indicating an important role of c-Fos in light-induced retinal degeneration. Additionally, c-Fos expression and photoreceptor damage are attenuated by the intraperitoneal administration of the antioxidant phenyl-α-tert-butylnitrone (PBN) [26]. Based on these findings, we hypothesised that PQQ might exert a similar effect on c-Fos expression. Thus, c-Fos levels were assessed by Western blotting. In line with these reports, the current study showed that light exposure markedly upregulated c-Fos expression in the retinas of vehicle-treated animals. By contrast, pretreatment with PQQ significantly suppressed light-induced c-Fos upregulation. This observation is noteworthy for two reasons: First, it aligns the mechanism of action of PQQ with that of other known protective agents, such as the antioxidant PBN, and suggests that PQQ may counteract upstream the oxidative stress signals associated with c-Fos induction. Second, given prior evidence indicating that c-Fos deficiency confers resistance to light-induced degeneration, the suppression of c-Fos by PQQ may contribute to the functional and structural protection observed in our ERG and histological analyses. Taken together, these findings indicate that the modulation of c-Fos-associated stress signalling may play a role in the neuroprotective effects of PQQ against photic injury. c-Fos functions as an immediate-early transcription factor with functions in circadian rhythm, phototransduction, gliotic responses and tissue remodelling. Therefore, the increase in c-Fos expression observed following light exposure may reflect multiple retinal processes and is insufficient to prove a pro-death role. While prior reports have linked AP-1 activity and c-Fos to light-induced photoreceptor apoptosis, the present results are only correlative. Our investigation revealed that PQQ attenuated c-Fos induction and was associated with reduced photoreceptor degeneration. Definitive determination of causality would require targeted Studies of c-Fos (for example, retinal c-Fos knockdown using AAV-delivered shRNA, conditional c-Fos knockout animals, or specific pharmacological inhibitors) combined with detailed profiling of cell-death modalities (e.g., apoptosis markers, parthanatos, and necroptosis). Such experiments are important future directions for research to clarify whether the modulation of c-Fos underlies the protective effects of PQQ.

We further investigated the effects of PQQ on the Akt pathway, a central regulator of cell survival and proliferation [38,39]. Although Akt is a well-established pro-survival kinase that protects against various insults [40], its overactivation has also been implicated in pathological conditions [41,42]. In the context of retinal photodamage, studies have shown that Akt deficiency exacerbates photoreceptor cell death, thus highlighting its essential protective function against this specific pathology [43,44]. In the current study, PQQ treatment significantly increased pAkt levels. This finding suggests that the activation of the Akt signalling pathway is a key component of the protective mechanism of PQQ, a strategy also utilised by other neuroprotective agents [41,45]. By actively promoting this pro-survival signal [46], PQQ enhances the intrinsic defence capacity of retinal cells, thereby counteracting the degenerative cascade initiated by light-induced stress. These findings also raise an important question: do the known antioxidant properties of PQQ stem from direct radical scavenging or from the activation of endogenous cellular defence mechanisms? Our observation that PQQ activates the Akt pathway provides compelling evidence. The Akt pathway is a well-established upstream activator of the transcription factor Nrf2, which is a master regulator of antioxidant response. Activated Nrf2 translocates to the nucleus and promotes the expression of a battery of cytoprotective genes, such as HMOX1. However, one important limitation of the present study is that we did not mechanistically examine whether p-Akt activation directly mediates c-Fos suppression in our light-damage model. While PI3K/Akt signalling can antagonize stress-activated pathways in other systems, a direct Akt–c-Fos regulatory axis has not yet been definitively established in retinal light injury. As such, future studies using specific inhibitors or genetic models are required to clarify this relationship and to determine whether Akt signalling is a necessary factor underlying the suppression of c-Fos observed with PQQ treatment.

To further integrate these findings, we proposed a temporal framework linking molecular, functional, and structural outcomes. The experimental timepoints in this study were chosen to capture the distinct stages of the degenerative cascade following light exposure. At 1 h post-exposure we interrogated the rapid signalling responses which represent early trigger events (e.g., Akt phosphorylation and c-Fos induction). ERG was assessed at Day 3 to evaluate the intermediate functional changes that could precede any overt structural loss. Finally, the ONL thickness was measured at Day 7 to quantify the cumulative structural outcomes following the clearance of dying cells. While these observations are correlative, the convergent pattern—including the early modulation of stress/survival signalling, preservation of electrophysiological function, and reduced long-term photoreceptor loss—provides a coherent temporal narrative linking the molecular modulation by PQQ to functional and structural neuroprotection.

Although a direct scavenging effect cannot be excluded, our data strongly suggest that the protective action of PQQ is associated, at least in part, with the activation of an Akt-involving endogenous antioxidant response rather than solely by passive antioxidant activity. This action represents a more dynamic and potent mechanism of cellular protection and warrants further investigation. Despite these promising findings, the present study has several limitations. First, the light-induced retinal degeneration model represents an acute injury paradigm that is primarily driven by oxidative stress. Although this model reproduces several key pathological features of atrophic age-related macular degeneration (AMD), it does not fully recapitulate the chronic, age-dependent progression or the complex, multifactorial aetiology of the human disease. Second, rodents lack a true macula, which limits the direct anatomical translatability of the findings to human macular degeneration. Third, we did not characterise the specific cell-death pathways activated following light exposure. Distinguishing among apoptosis, parthanatos, necroptosis, autophagy-related death, and phagoptosis requires specific targeted assays (e.g., TUNEL/cleaved-caspase-3; PAR accumulation/AIF translocation; p-RIPK3/p-MLKL; LC3-II/p62 and EM; Iba1/CD68 co-localization). These experiments are important future research directions. Fourth, PQQ was administered via intraperitoneal injection to ensure consistent systemic delivery in this proof-of-concept study; however, this route of administration is not optimal for clinical application. Future studies should therefore investigate the efficacy of clinically relevant administration routes, such as oral delivery or topical instillation, and assess the long-term safety profile of PQQ. In addition, validation of these findings in other chronic models of atrophic AMD (e.g., Abca4-/- or Sod-/- mice) would further strengthen the evidence supporting its therapeutic potential.

4. Materials and Methods

4.1. Reagents

PQQ disodium salt (Mitsubishi Gas Chemical Co., Inc., Tokyo, Japan) was provided by Rohto Pharmaceutical Co., Ltd. (Osaka, Japan).The reagent was stored at 4 °C, protected from light, and prepared immediately before use by dissolving it in PBS. The solution was vortexed thoroughly and shielded from light during preparation and administration. Solutions were prepared at two concentrations for subsequent administration to animals: 1 mg/mL (2 mg/kg) and 2.5 mg/mL (5 mg/kg).

4.2. MTS Assay

ARPE-19 cells were seeded (5 × 10^3^ cells/well) into 96-well plates and cultured overnight. Each reagent and ATR solution were added to corresponding wells, and cell viability was assessed after 24 h through MTS assay by adding 20 µL of CellTiter 96^®^ AQueous One Solution (Promega, Tokyo, Japan). After incubation for 20 min, absorbance was recorded at 490 nm.

4.3. Animals

Eight-week-old male Sprague Dawley rats (CLEA Japan, Inc., Tokyo, Japan) were used in this study. These rats were housed under 12 h dark and 12 h light (5 lux) cycles with available chow and water. All animal experiments were performed in accordance with the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research and the Guidelines of the Iwate University Animal Experimentation Committee on Animals in Research.

4.4. Induction of LD

The rats were adapted to the dark for 24 h prior to the procedure. Thirty minutes before light exposure, pupils were dilated with tropicamide (Mydrin-P; Santen Pharmaceutical Co., Ltd., Osaka, Japan). The animals then received a single intraperitoneal injection of PBS (vehicle) or PQQ. The administration volume was 2 mL/kg of body weight in all groups. The PQQ-treated groups received a final dose of 2 or 5 mg/kg PQQ. Rather than being directly extrapolated from any in vitro experiments the PQQ doses (2 and 5 mg/kg) were selected based on previously published pharmacological studies demonstrating their safety and neuroprotective efficacy in rodent models [23,47]. To minimize the influence of circadian variation, all light-damage experiments were initiated at 08:00 (Zeitgeber time 0, ZT0), and all retinal tissue samples for molecular analyses were collected 1 h after the onset of light exposure (09:00, ZT1). Following the injection, the rats were exposed to 1000-lux white fluorescent light in a light box for 24 h. This light-exposure protocol served as the primary model for inducing photoreceptor degeneration. Subsequent functional, histological, and molecular analyses were performed at specific time points, as described in the respective sections.

4.5. OCT

OCT measurements were performed both at baseline and 1 day after light exposure by using a previously described method [17]. Rats were anaesthetised via intramuscular injection of ketamine (75 mg/kg; Daiichi Sankyo Propharma Co., Ltd., Tokyo, Japan) and Dorbene (0.5 mg/kg; Kyoritsu Seiyaku Corporation, Tokyo, Japan). Pupil dilation was achieved by administering Mydrin-P eye drops. OCT imaging was performed in a dark room by using an RS-3000 OCT system (Nidek Co., Ltd., Aichi, Japan) equipped with a rat-specific lens. Retinal cross-sectional images centred on the optic nerve head were acquired and scanned in both the 12 and 6 o’clock directions over a scan length of 6 mm (approximately 1.1 mm in the actual measurement). The retina was segmented into upper and lower portions, and both the full retinal thickness and photoreceptor layer thickness were measured.

4.6. ERG

ERG measurements and analyses were performed using the PuREC system (Mayo Co., Inazawa, Aichi, Japan) at baseline and 3 days after light exposure, following a previously reported protocol [17]. Briefly, rats were dark-adapted overnight and anaesthetised by the intramuscular injection of a mixture of ketamine (45 mg/kg; Daiichi Sankyo Propharma Co., Ltd.) and xylazine (4.5 mg/kg; Elanco Japan, Tokyo, Japan). Topical anaesthesia was achieved using Benoxil ophthalmic solution 0.4% (Santen Pharmaceutical Co., Ltd.), and the pupils were dilated using a 10-fold diluted solution of tropicamide (Santen Pharmaceutical Co., Ltd.). A small contact lens electrode equipped with a gold wire loop that was lightly coated with hydroxyethyl cellulose eye drops (Scopisol 15; Senju Pharmaceutical Co., Ltd., Osaka, Japan) was placed on the test eye. A reference electrode was positioned in the oral cavity, and a ground electrode clip was attached to the foot of the rat. Stimulus intensities of 0.01, 3.00, and 10.0 cd·s/m^2^ were set in accordance with the recommendations of the International Society for Clinical Electrophysiology of Vision [20].

4.7. Paraffin-Embedded Sections and HE Staining

The rats were euthanised with an overdose of CO_2_ 1 week after LD. The eyes of the rats were enucleated, fixed, and embedded in paraffin. Five-micrometre sections were cut along the vertical meridian and stained with HE to compare all areas of the retina in the superior and inferior hemispheres. The thickness of the retinal layers was measured at 500 μm intervals in the vertical direction, including the optic nerve head.

4.8. Western Blot Analysis

Total retina lysates were prepared using Pierce^®^ RIPA Buffer (Thermo Fisher Scientific, Tokyo, Japan) and the Halt Protease and Phosphatase Inhibitor Single-Use Cocktail, EDTA-free solution (100×; Thermo Fisher Scientific, Tokyo, Japan) and 0.5 M EDTA solution (Thermo Fisher Scientific, Tokyo, Japan). The protein concentration was measured using a bicinchoninic acid assay kit (Thermo Fisher Scientific, Tokyo, Japan). Twenty micrograms of protein per sample were loaded onto a 4–15% mini-protean TGX precast polyacrylamide gel (Bio-Rad Laboratories, Tokyo, Japan) and then transferred to a polyvinylidene fluoride membrane (Bio-Rad Laboratories, Tokyo, Japan). After blocking with Block Ace (KAC Co., Ltd., Kyoto, Japan), the membranes were incubated with the primary antibodies listed in Table 1. After washing, the membranes were incubated with an alkaline phosphatase–conjugated secondary antibody. Band density was measured using ImageQuant software (GE Healthcare, Tokyo, Japan).

4.9. Statistical Analyses

Dunnett’s multiple comparison test, Tukey’s multiple comparison test, and unpaired t-test were used for comparative analyses. Statistical analyses of the in vitro and in vivo experiments were performed using GraphPad Prism 4 software (MDF, Tokyo, Japan).