Transient functional preservation precedes retinal ganglion cell degeneration following optic nerve crush

Abstract

Aims

Optic nerve crush (ONC) induces progressive retinal ganglion cell (RGC) degeneration; however, the temporal relationship between functional impairment and structural loss remains incompletely understood. This study investigated early changes in retinal function and subtype-specific RGC responses following ONC to determine whether retinal activity is preserved before substantial RGC degeneration.

Methods

Adult Sprague-Dawley rats underwent ONC and were examined at 1, 3, 7, and 14 days after injury. Retinal function was assessed using electroretinography (ERG), photopic negative response (PhNR), and multi-electrode array (MEA) recordings. RGC survival was quantified by Fluoro-Gold labeling, while microglial activation and oxidative stress-related markers were evaluated by immunofluorescence.

Results

RGC survival remained largely preserved during the first 3 days following ONC but declined markedly at days 7 and 14. MEA recordings showed relative preservation of light-evoked responses at day 3, followed by progressive reductions in responsiveness and spontaneous firing at later stages. PhNR amplitudes showed a similar trend, although the changes did not reach statistical significance. Subtype analysis revealed differential vulnerability, with ONOFF-RGCs exhibiting greater reductions in proportion and functional responses, whereas ON-RGCs accounted for an increasing proportion of surviving cells over time. Microglial activation increased during the degenerative phase, while the oxidative damage marker 8-OHG immunoreactivity appeared elevated early after injury. No obvious changes were detected in NQO-1 or SOD-2 expression.

Conclusions

Retinal function remained relatively preserved during the first 3 days following ONC despite ongoing injury and before substantial RGC loss occurred. ONOFF-RGCs exhibited greater vulnerability than other functional subtypes. These findings identify an early post-injury period preceding substantial RGC degeneration and provide insight into the temporal dynamics of RGC dysfunction following optic nerve injury.

Highlights

• •Retinal ganglion cells (RGCs) functional degeneration after optic nerve crush (ONC) is subtype-dependent.
• •ONOFF-RGCs exhibited early vulnerability and ON-RGCs demonstrated greater resilience after ONC.
• •RGCs could preserve their functional activity before significant cell death.
• •There may be a short therapeutic window after ONC.

1. Introduction

Traumatic optic neuropathy (TON), resulting from direct or indirect injury to the optic nerve, frequently leads to irreversible visual impairment (Blanch et al., 2024). Despite advances in clinical management, effective treatment options remain limited and visual outcomes are often unpredictable. Recent systematic reviews and meta-analyses have continued to question the efficacy of corticosteroid therapy and optic canal decompression, highlighting the need for novel neuroprotective strategies (Chen et al., 2022, Pandey et al., 2023, Fallahzadeh et al., 2024, Cheng et al., 2025). The rodent optic nerve crush (ONC) model is widely used to investigate the pathophysiology of optic nerve injury and to evaluate potential therapeutic approaches relevant to TON, glaucoma, and other optic neuropathies (Ahmed, 2023, Cen et al., 2023).

ONC disrupts axonal integrity and initiates a cascade of secondary pathological events, including neuroinflammation, oxidative stress, mitochondrial dysfunction, and progressive retinal ganglion cell (RGC) degeneration (Liu et al., 2014, Meng et al., 2023, Zapadka et al., 2025, Chen et al., 2026). Although substantial RGC loss develops following injury, accumulating evidence suggests that neuronal dysfunction may precede overt structural degeneration. Identifying these early functional alterations may therefore reveal a critical window for neuroprotective intervention (Cen et al., 2023, Cheng et al., 2025).

RGCs comprise a heterogeneous neuronal population with distinct morphological, molecular, and functional characteristics, resulting in varied responses to injury (Yang et al., 2020, Chen et al., 2026). Increasing evidence indicates that susceptibility to optic nerve damage differs among RGC subtypes. Subtype-specific vulnerability has been reported in experimental models of glaucoma, ocular hypertension, excitotoxicity, and optic nerve injury (Feng et al., 2013, Puyang et al., 2015, Christensen et al., 2019, VanderWall et al., 2020). These observations suggest that analyses based solely on total RGC counts may overlook important subtype-dependent differences in degeneration and functional resilience.

Multi-electrode array (MEA) recordings enable functional classification of RGCs into ON-, OFF-, and ONOFF-RGCs according to their responses to light stimulation. Previous studies have demonstrated that RGC subtypes differ in their susceptibility to optic nerve injury, although the relative vulnerability of individual subtypes remains incompletely understood. OFF-RGCs have frequently been reported to exhibit earlier or more severe functional impairment than ON-RGCs, whereas other studies have identified distinct patterns of vulnerability that vary according to the injury model and disease stage (Feng et al., 2013, Puyang et al., 2015, Puyang et al., 2017, Chen et al., 2015, Dunn et al., 2023). However, most investigations have focused on established degeneration rather than the early post-injury period. Clarifying the temporal dynamics of subtype-specific functional responses during the early stages of ONC may provide insight into mechanisms of RGC degeneration and help identify targets for neuroprotective intervention.

To address this gap, we investigated the temporal changes in retinal function and subtype-specific RGC responses during the first two weeks following ONC in rats. Retinal function was evaluated using electroretinography (ERG), photopic negative response (PhNR) recordings, and MEA analysis, while RGC survival, glial activation, and oxidative stress-related markers were assessed histologically. We aimed to determine whether retinal function remains preserved during the early post-injury period preceding substantial RGC loss and to characterize the differential vulnerability of ON-, OFF-, and ONOFF-RGC populations following optic nerve injury.

2. Material and methods

All procedures adhered to the National Institute of Health Guide for the Care and Use of Laboratory Animals, the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Committee for the Use of Live Animals in Teaching and Research (CULATR: 4022–16), University of Hong Kong.

2.1. Animal preparation

Female Sprague–Dawley (SD) rats, weighing approximately 250 g ± 20 g, were obtained from the Laboratory Animal Unit (LAU) of the LKS Faculty of Medicine, University of Hong Kong. Animals were housed under standard laboratory conditions with a 12-hour light/dark cycle and ad libitum access to food and water. Rats were randomly assigned to five groups: control (no injury), and post-ONC groups at 1, 3, 7, and 14 days. At the designated experimental endpoints, animals were euthanized with an overdose of pentobarbitone sodium (150 mg/kg, Alfasan International B.V., Netherlands). Retinas were then harvested for MEA recordings and histological analyses.

2.2. RGC labeling and optic nerve crush (ONC) model

Animals were anesthetized with intraperitoneal injections of ketamine (80 mg/kg) and xylazine (8 mg/kg) (Alfasan International B.V., Netherlands). RGCs were retrogradely labeled by placing a gelatin sponge soaked in 6% Fluoro-Gold (FG; Fluorochrome, USA) onto the surface of the superior colliculus (SC) (Li et al., 2019). Rats recovered for seven days post-labeling before undergoing ONC surgery.

The ONC procedure involved exposing the optic nerve via blunt dissection of the surrounding muscle and fat. The optic nerve sheath was carefully incised with a 30 G needle, avoiding blood vessels. The nerve was crushed 1.5–2 mm from the optic nerve head using ultrafine forceps (#5, Fine Science Tools, Germany) for 5 s. Postoperative application of erythromycin eye ointment facilitated recovery.

2.3. Electroretinography (ERG) and photopic negative response (PhNR) recording

Following overnight dark adaptation, ERGs were recorded using the Espion 2 system (Diagnosys, USA). Under dim red illumination, rats were anesthetized, and pupils dilated with 0.5% tropicamide (Alcon, USA). The animals were positioned on a heated platform (37°C). Stainless-steel needle electrodes served as ground (tail) and reference (neck skin). Gold ring electrodes were placed on the cornea for signal acquisition.

• •Scotopic ERGs: Recorded after dark adaptation with white flashes (3.0 cd·s/m²).
• •Photopic ERGs: Conducted after 5-minute light adaptation under a green background (40 cd/m²), followed by white flashes (16.6 cd/m²).

ERG signals were filtered (50 Hz low-pass), and amplitudes and latencies of the a-wave, b-wave, and PhNR were analyzed at baseline and at 1, 3, 7, and 14 days post-ONC.

2.4. MEA recording and data analysis

Functional assessment of RGCs following ONC was performed using whole-mounted retinal preparations, as previously described with minor modifications (Bao et al., 2019, Liu et al., 2018). After 3 h of dark adaptation, animals were euthanized, and 2 × 2 mm retinal segments were dissected and mounted on an 8 × 8 electrode MEA grid (electrode diameter, 20 μm; interelectrode spacing, 100 μm; P210A, Alpha MED Scientific, Inc.). The retina was positioned with the ganglion cell layer facing the electrodes. Recordings were performed in oxygenated Ames’ medium (Sigma, USA) at 31–33°C, superfused at 4 ml/min. Extracellular spikes were bandpass filtered (100–5000 Hz), digitized at 20 kHz, and analyzed offline using Mobius software (Alpha MED Scientific, Inc.).

Visual stimuli were delivered by a white LED light source (LEDWE-15; Thorlabs, USA) controlled by a stimulus generator. Each stimulus consisted of 2 s of light (intensity: 3.36 ×10⁷ photons/μm²/s), followed by 8 s of darkness, repeated 30 times. Spike sorting was performed with Offline Sorter (Plexon, USA). Peristimulus time histograms (PSTHs) and raster plots were generated using 10-ms bins.

RGCs were classified into ON-, OFF-, or ONOFF-types based on their light response characteristics, using the response dominant index (RDI):

$$\mathbf{RDI}=\frac{({\mathit{R}}_{\mathit{ON}}-{\mathit{R}}_{\mathit{OFF}})}{({\mathit{R}}_{\mathit{ON}}+{\mathit{R}}_{\mathit{OFF}})}$$

where R_ON and R_OFF represent the average spike rates during the first 2 s of the light onset (ON stimulus) and light offset (OFF stimulus), respectively, with spontaneous activity subtracted.

• •RDI > 0.6: ON-RGCs
• •RDI ≤ -0.6: OFF-RGCs
• •-0.6 <RDI < 0.6: ONOFF-RGCs (Bao et al., 2019, Liu et al., 2018)

Response parameters such as peak response, average response, latency to peak, spontaneous firing rate, and signal-to-noise ratio (SNR) were calculated. The SNR was defined as the ratio of the average light-evoked spike rate to the standard deviation of spontaneous firing (Liu et al., 2022).

2.5. Immunofluorescence staining and imaging

Retinal sections were deparaffinized, subjected to antigen retrieval in boiling citric acid buffer (0.01 M, pH 6.0) for 15 min, then blocked with 0.3% PBST containing 3% donkey serum and 1% BSA for 1 h. Primary antibodies were incubated overnight at 4°C, including Ionized calcium-binding adaptor molecule 1(Iba-1) (1:500, rabbit; Wako, Japan), glutamine synthetase (GS) (1:500, rabbit; Thermo Fisher, USA), quinone oxidoreductase 1(NQO-1) (1:1000, rabbit; Life Technologies, USA), 8-hydroxyguanine (8-OHG) (1:1000, mouse; Abcam, UK), and superoxide dismutase-2 (SOD-2) (1:500, rabbit; Thermo Fisher, USA). Secondary antibodies conjugated to Alexa Fluor 488 or 568 (Life Technology, USA) were applied at 1:500 for 1 h at room temperature. Nuclei were counterstained with DAPI (1:1000). Fluorescent images were captured using a fluorescence microscope, and analyses were performed with ImageJ (NIH, Bethesda, MD). For quantitative analysis, three retinal sections from each eye were evaluated. Mean fluorescence intensity was quantified in the GCL and INL following background subtraction. Fluorescence intensities were normalized to the mean value of the control group. Image analysis was performed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

2.6. RGC and microglia cell counting

Following euthanasia with an overdose of pentobarbitone sodium and fixation in 4% paraformaldehyde for 1 h, retinas were dissected, flat-mounted, and divided into four quadrants: superior nasal, superior temporal, inferior nasal, and inferior temporal. FG-labeled RGCs were imaged along the median line from the optic disc to the periphery at 500 μm intervals using a fluorescence microscope (400 × magnification, UV filter). Approximately 7–8 fields per quadrant (200 ×200 μm) were imaged, and RGCs and microglia were quantified by an investigator masked to experimental group assignment. Results are expressed as cells/mm² (mean ± SD). Activated microglia can phagocytose degenerating FG-labeled RGCs and thereby become fluorescently labeled through the ingestion of cellular debris. These cells were identified by their distinct amoeboid morphology, which is readily distinguishable from RGC somas in flat-mounted retinas.

2.7. Statistical analysis

Data normality was assessed with the Kolmogorov–Smirnov test. Data are presented as mean ± standard deviation (SD). For comparisons across multiple groups, One-way ANOVA was used, followed by Tukey’s or Student–Newman–Keuls post hoc tests. For non-normal data, the Kruskal–Wallis test with Dunn’s correction was applied. A p-value < 0.05 was considered statistically significant (Prism 9.5, GraphPad Software).

3. Results

3.1. Reduction in retinal light response following ONC

Retinal function was evaluated through scotopic and photopic responses using ERG and PhNR recordings. Representative images are shown in Fig. 1 A and 1D. Significant changes were observed in the amplitudes of the scotopic a-wave and b-wave following ONC (One-way ANOVA: a-wave: p = 0.0001; b-wave: p = 0.0002). Specifically, the scotopic a-wave, indicative of photoreceptor activity, decreased by 18% at 1 day post-ONC, returned to near-baseline levels at day 3, and then declined to 72% and 50% of control levels at days 7 and 14, respectively (Fig. 1B). The scotopic b-wave, representing bipolar cell responses, remained stable on days 1 and 3 but showed a significant reduction by day 14 (Fig. 1 C.).

Fig. 1.

Decline in F-ERG and PhNR responses following ONC.

Under photopic conditions, similar trends were observed, although the changes were less pronounced. The b-wave and PhNR did not show statistically significant differences (One-way ANOVA: b-wave: p = 0.67; PhNR: p = 0.07; Figs. 1E and 1 F.). The PhNR amplitude decreased slightly to 92% at day 1, remained relatively preserved at day 3, and then declined to 56% and 64% of control levels at days 7 and 14, respectively. A latency delay was noted in both scotopic and photopic responses post-ONC, but these differences were not statistically significant.

3.2. RGC loss and glial cell response following ONC

To assess RGC survival and microglial responses, Fluoro-Gold (FG) labelling was performed. Representative images of FG-labeled RGCs (red arrows) and FG-positive amoeboid microglia (yellow arrows) in flat-mounted retinas are shown in Fig. 2 A. In control retinas, the average RGC density was 2590 ± 45 cells/mm². RGC survival remained relatively stable during the first 3 days following ONC but declined significantly to 38% and 15% of control levels at days 7 and 14, respectively (One-way ANOVA with Tukey’s post hoc test, p < 0.001; Fig. 2B). The density of FG-positive amoeboid microglia increased following ONC, reaching 272 ± 45 cells/mm² at day 7 and 464 ± 75 cells/mm² at day 14 (unpaired t-test, p = 0.0012; Fig. 2 C).

Fig. 2.

RGC survival and microglia activation following ONC.

Immunofluorescence analysis further demonstrated increased Iba-1 immunoreactivity at days 3 and 7 following ONC (Fig. 3 A, B). In addition, Iba-1-positive cells exhibited a more amoeboid morphology and were observed within the ganglion cell layer (GCL), inner plexiform layer (IPL), and outer plexiform layer (OPL). In contrast, glutamine synthetase (GS) immunoreactivity showed no apparent changes across the experimental time points (Fig. 3 C).

Fig. 3.

Expression of Iba-1 and GS in retina following ONC.

3.3. Functional alterations in RGC light-evoked responses following ONC

MEA recordings revealed progressive functional changes in individual RGCs following ONC. Representative spike responses to a 2-s full-field light stimulus are shown in Fig. 4 A. Peak light-evoked responses were significantly reduced at days 7 and 14 compared with the control group (Kruskal-Wallis test, p < 0.001; Fig. 4B). Light-evoked responses showed relatively little change at day 3 but declined markedly at later time points. Average response amplitudes exhibited a similar pattern (Fig. 4 C). Latency to peak response was significantly shortened at day 3 and subsequently returned toward baseline values at days 7 and 14 (p < 0.05; Fig. 4D). Spontaneous firing rates declined significantly from day 3 onward (p < 0.001; Fig. 4E). The signal-to-noise ratio (SNR), defined as the ratio of the average light-evoked response to the standard deviation of spontaneous firing, increased from day 3 and remained elevated at later time points (p < 0.001; Fig. 4 F).

Fig. 4.

Decline in light-evoked RGC responses following ONC.

3.4. Differential changes in ON and OFF pathways following ONC

RGCs were classified into ON-, OFF-, and ONOFF-types based on their response dominant index (RDI). The proportional distribution of these subtypes was analyzed at different time points following ONC. In control retinas, ONOFF-RGCs represented the largest proportion of recorded cells, followed by ON- and OFF-RGCs. By day 7, the proportion of ON-RGCs increased and reached 79.5% at day 14, whereas ONOFF-RGCs progressively decreased to 16.3%. OFF-RGCs showed fluctuations over time but no significant change, accounting for 4.2% of the recorded population at day 14. Statistical analysis confirmed significant changes in the proportions of ON-RGCs and ONOFF-RGCs, but not OFF-RGCs (One-way ANOVA with Tukey’s post hoc test: ON-RGCs, p = 0.0001; ONOFF-RGCs, p = 0.0001; OFF-RGCs, p = 0.56; Fig. 5 A).

Fig. 5.

Light response of RGC subtypes changed differently after ONC.

Analysis of subtype-specific light responses revealed distinct patterns following ONC. Peak light responses of ON-RGCs remained relatively preserved at day 3 before declining at days 7 and 14. ONOFF-RGCs showed a similar pattern, with relatively preserved responses at day 3 followed by progressive reductions at later time points. OFF-RGC responses varied across time points and were reduced at day 7 (Kruskal-Wallis test with Dunn’s multiple comparisons: ON-RGCs, p < 0.0001; OFF-RGCs, p = 0.0006; ONOFF-RGCs, p < 0.0001; Fig. 5B).

Average light-evoked responses showed trends similar to those observed for peak responses (Fig. 5 C). ON-RGC responses remained relatively preserved at day 3 but were significantly reduced at days 7 and 14. ONOFF-RGCs exhibited a progressive decline over time, whereas OFF-RGC responses remained comparatively stable despite temporal fluctuations. Spontaneous firing rates generally decreased following ONC across all subtypes, although ON-RGCs showed higher spontaneous activity at day 3 than at later time points (Fig. 5D).

3.5. Expression changes of NQO-1, SOD-2, and 8-OHG in retina

To investigate oxidative stress, we examined the expression of antioxidant enzymes and oxidative damage markers (Fig. 6 A). Immunofluorescence revealed a trend toward increased NQO-1 and SOD-2 immunoreactivity at day 3 following ONC; however, these observations were not subjected to statistical analysis because of the limited sample size. In contrast, 8-OHG immunoreactivity appeared elevated as early as day 1 after ONC and was detected in both the ganglion cell layer (GCL) and inner nuclear layer (INL) (Fig. 6B-D).

Fig. 6.

Oxidative stress markers expression in retina following ONC.

4. Discussion

In the present study, retinal function and RGC survival remained largely preserved during the first 3 days following optic nerve crush (ONC), whereas progressive functional impairment and substantial RGC loss became evident at later stages. In parallel, ONOFF-RGCs exhibited greater vulnerability than other functional RGC subtypes. These findings indicate that retinal activity is maintained during the early post-injury period despite ongoing axonal damage, preceding the onset of marked neurodegeneration.

Quantitative analysis revealed a marked reduction in RGC survival, declining from 98.2% at day 3–15% by day 14. Consistent with previous ONC studies, substantial RGC loss occurred only after the initial phase of injury (Liu et al., 2014, Tran et al., 2019, Yang et al., 2020). In contrast, PhNR amplitudes showed only non-significant reductions despite progressive RGC loss. Because PhNR reflects the functional activity of surviving RGCs and inner retinal circuitry rather than cell number alone, this apparent dissociation is not unexpected. Similar discrepancies between PhNR amplitude and RGC survival have been reported following optic nerve injury (Li et al., 2005, Liu et al., 2014). Together, these findings suggest that functional and structural changes do not necessarily occur in parallel during ONC progression.

While PhNR primarily reflects inner retinal and RGC function, full-field ERG provides additional information regarding broader retinal circuitry. In addition to changes in PhNR, scotopic a-wave and b-wave amplitudes were reduced at later stages following ONC. Since these components primarily reflect photoreceptor and bipolar cell function, respectively, the findings suggest that the impact of optic nerve injury extends beyond RGCs and may involve broader retinal dysfunction as degeneration progresses. Similar ERG alterations have been reported following optic nerve injury in both rat and mouse models (Alarcón-Martínez et al., 2009, Smith et al., 2014), supporting the concept of progressive retinal remodelling after axonal injury.

To further characterize these functional changes at the cellular level, we performed MEA recordings of individual RGC responses. Light-evoked responses remained relatively preserved at day 3 but declined substantially by days 7 and 14. These observations are consistent with previous studies reporting early electrophysiological abnormalities following optic nerve injury (Tran et al., 2019, Zapadka et al., 2025). Together with the PhNR findings, the MEA data suggest that functional measures may detect injury-related changes before substantial structural degeneration becomes apparent.

An additional insight provided by the MEA analysis was the progressive reduction in spontaneous firing activity. Spontaneous firing decreased from day 3 onward despite relatively preserved RGC survival, indicating impaired neuronal excitability during the early post-injury period. Similar reductions in spontaneous activity have been reported in experimental glaucoma and ocular hypertension models (Della et al., 2013, Ou et al., 2016). Notably, these changes preceded substantial RGC loss, suggesting that altered intrinsic neuronal activity may represent one of the earliest functional consequences of optic nerve injury. However, the apparent preservation of light-evoked responses should be interpreted cautiously, as reduced spontaneous firing may have contributed to the increased signal-to-noise ratio observed at day 3. In addition, selective recording of more resilient RGCs cannot be excluded. Therefore, the observed preservation of early functional responses should be viewed as a hypothesis-generating observation rather than evidence of functional recovery.

In addition to these population-level functional changes, ONC produced marked differences among RGC subtypes. ONOFF-RGCs exhibited substantial reductions in both relative proportion and functional responsiveness at days 7 and 14, indicating greater sensitivity to injury. In contrast, ON-RGCs accounted for an increasing proportion of the surviving population over time, suggesting greater resistance to axonal damage. These findings are consistent with previous studies demonstrating differential susceptibility among RGC subtypes following optic nerve injury and other neurodegenerative conditions (Puyang et al., 2017, Tran et al., 2019, VanderWall et al., 2020, Yang et al., 2020). The mechanisms underlying these differences remain incompletely understood but may involve variations in dendritic organization, intrinsic excitability, and metabolic demand (El-Danaf and Huberman, 2015, Margolis et al., 2010, Zhang et al., 2022). Further studies integrating functional, molecular, and morphological analyses will be required to clarify the biological basis of subtype-specific vulnerability.

To explore biological processes associated with degeneration, we also examined glial and oxidative stress-related changes following ONC. Microglial activation appeared to become more prominent during the later stages of degeneration. Increased Iba-1 immunoreactivity and the presence of FG-positive amoeboid cells in retinal flat mounts were consistent with previous reports describing microglial activation and phagocytic engagement following optic nerve injury (Wang et al., 2021, Huang et al., 2025, Venanzi et al., 2025, Bodeutsch and Thanos, 2000). However, as these observations were primarily descriptive, the contribution of microglial responses to RGC degeneration cannot be determined from the present study.

In addition to neuroinflammatory responses, oxidative stress has been implicated in optic nerve degeneration and may contribute to injury progression (Buonfiglio et al., 2023, Pan et al., 2024). In the present study, the oxidative damage marker 8-OHG showed a tendency to increase during the early post-injury period, whereas no obvious changes were detected in SOD-2 or NQO-1 expression. Given the exploratory nature of these analyses and the limited changes observed in antioxidant markers, the findings should be interpreted cautiously. Although they are consistent with the possibility that oxidative stress participates in the early response to ONC, further studies will be required to determine its temporal progression and functional significance.

Several limitations of this study should be acknowledged. First, the number of animals included in the MEA analyses was limited at certain time points, particularly at days 1 and 14 following ONC, which reduced statistical power and may have limited the detection of subtle functional changes. Second, microglial activation in retinal flat mounts was identified based on FG accumulation and amoeboid morphology. Although these features are consistent with phagocytic microglia, they provide indirect rather than definitive evidence of microglial activation. Third, the oxidative stress analyses were exploratory and restricted to a limited number of markers. Consequently, the present findings do not allow firm conclusions regarding the contribution of oxidative stress to RGC degeneration following ONC. Finally, RGC subtypes were classified according to electrophysiological response characteristics without molecular validation. While this approach enabled functional assessment of subtype-specific responses, the cellular mechanisms underlying the observed differential vulnerability remain to be determined.

Despite these limitations, the present study provides new insight into the temporal relationship between retinal dysfunction and degeneration following ONC. In conclusion, retinal function remained relatively preserved during the first 3 days following ONC despite ongoing injury and before substantial RGC loss became evident. ONOFF-RGCs exhibited greater vulnerability than other functional RGC subtypes, highlighting the heterogeneous nature of RGC degeneration following optic nerve injury. Together, these findings identify an early post-injury period preceding marked neurodegeneration and provide insight into the temporal dynamics of RGC dysfunction after ONC.

CRediT authorship contribution statement

Shasha Yu: Writing – original draft, Methodology, Investigation, Conceptualization. Feng Liu: Writing – review & editing, Software, Methodology, Investigation. Yun Zhao: Writing – review & editing, Validation, Data curation. Ying Xu: Writing – review & editing, Visualization, Supervision, Data curation. Kin Chiu: Writing – review & editing, Visualization, Supervision, Project administration.

Funding

1.This research was funded by the Science and Technology Foundation of Tianjin Eye Hospital (No.YKQN2003 to S.S.YU). 2. Natural Science Foundation of China grants (No.32300967 to F.L.); 3. Guangdong Basic and Applied Basic Research Foundation grants (No. 2024A1515010181 to F.L.).

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper

Data availability

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.