The Neuroprotective Effect of Activation of Sigma-1 Receptor on Neural Injury by Optic Nerve Crush

Abstract

Purpose: This study aimed to explore the neuroprotective effects of sigma-1 receptor (S1R) on optic nerve crush (ONC) mice by upregulating its expression through intravitreal injection of adeno-associated virus (AAV).

Methods: The animals were divided into four groups. Mice that underwent ONC were administered an intravitreal injection with blank vector (ONC group), with AAV targeting downregulation of S1R (S1R-sh group), or with AAV targeting overexpression of S1R (S1R-AAV group). Mice in the control group underwent intravitreal injection with blank vector. The thickness of each layer of the retina was measured through optical coherence tomography, and the apoptotic rate of retinal neurons was determined using the TUNEL assay. The expression levels of brain-derived neurotrophic factor (BDNF) and S1R were quantified through western blot. Electroretinogram (ERG) was performed to evaluate the visual function.

Results: The thickness of the total retina (P = 0.001), ganglion cell layer (P = 0.017), and inner nuclear layer (P = 0.002) in S1R-AAV group was significantly thicker than that of the ONC group. The number of retinal apoptotic cells in the S1R-AAV group was 23% lower than that in the ONC group (P = 0.002). ERG results showed that, compared to the ONC group, the amplitudes of the a- and b-waves were higher in the S1R-AAV group (a-wave, P < 0.001; b-wave, P = 0.007). Western blot showed that the expression of BDNF in the S1R-AAV group was higher than that in the ONC group (P < 0.001).

Conclusions: Activation of S1R in the retina through intravitreal injection of AAV can effectively maintain the retina structure, promote neuronal cell survival, and protect visual function.

The retina is derived from the central nervous system during the embryonic stage¹ and cannot regenerate after injury.^2,3 Retinal ganglion cells (RGCs) transmit light from the retina to the superior colliculus and lateral geniculate body through axons.⁴ They constitute the only nerve conduction pathway and are a potential target for therapies for optic neuropathies.⁵ Studies have shown that optic nerve crush (ONC) can cause acute optic nerve axon loss, declines in retinal function, and structural changes, particularly damage in the ganglion cell layer and RGC death.^6,7 

Sigma-1 receptor (S1R) is abundantly expressed in the central nervous system.⁸ The precise role of S1R in diverse ocular tissues remains incompletely understood; however, the literature indicates that S1R may serve a protective function for ganglion and photoreceptor cells, as evidenced by both in vitro and in vivo experiments. Specifically, numerous reports have demonstrated that these receptors mitigate neuron loss in vitro and contribute to the repair of acute neurodegeneration in vivo.⁹ S1R expression is enriched in the eye and may inhibit degenerative changes in the eye to promote the survival of lens cells, regulation of intraocular pressure, and retinal neuroprotection of the retina.¹⁰ S1R knockout mice showed decreased electroretinogram (ERG) b-wave amplitude, reduced negative scotopic threshold responses, dysfunction of the inner retina (denoted by increased TUNEL-positive cells and active caspase-3–positive cells), and significant loss of the ganglion cells (especially in the central retina) at 12 months.¹¹ The mice also showed damage to the microstructure of optic nerve axons at 6 months of age.¹² 

In Müller cells and ganglion cells, S1R is expressed in the endoplasmic reticulum and nuclear membrane.¹⁰ It prevents oxidative stress-induced RGC death in vivo by regulating intracellular calcium ion levels and reducing gene activation before apoptosis. Many studies have confirmed that S1R regulates the response of retinal Müller cells to stress.¹³ The activation of S1R not only inhibits glutamate excitotoxin-induced development of edema in Müller cells¹⁴ but also prevents lipopolysaccharide-induced release of inflammatory factors.¹⁵ Wang et al.¹⁶ cultured Müller cells from S1R knockout mice and found that S1R regulated cellular reactive oxygen and antioxidant factors. Li et al.¹⁷ conducted a study wherein they discovered that the overexpression of the S1R through intravitreal injection of adeno-associated virus 2 (AAV2) facilitated the restoration of RGC function. Additionally, the activation of S1R by pentazocine was observed to decelerate the degeneration of RGC cells and enhance RGC activity subsequent to ONC, thereby providing further evidence of the protective properties associated with S1R. In vivo and in vitro studies have confirmed that activated S1R has protective effects on ganglion cells and photoreceptor cells.¹⁸ S1R induces protective effects by reducing peroxide product levels, inhibiting the release of inflammatory factors, and regulating cellular apoptosis.¹⁹ 

Neuronal survival benefits from the persistent inhibition of apoptotic pathways by trophic factors in the microenvironment, and brain-derived neurotrophic factor (BDNF) is crucial in endogenous repair.²⁰ BDNF in the retina is produced endogenously by ganglion cells and glial cells or retrogradely transported from the brain by ganglion axons.²¹ Studies have found that in patients with glaucoma, the transport of BDNF is blocked, and apoptosis of ganglion cells increases.²² Increased BDNF expression through exogenous supplementation or genetic modification increases ganglion cell survival in ocular hypertension and optic nerve crush models.^23,24 At present, S1R agonists have been shown to enhance BDNF expression and downstream signal transduction in retinal astrocytes,²⁵ whereas overexpression of S1R promotes precursor BDNF maturation and extracellular secretion.²⁶ 

In this study, a model of acute optic nerve injury was developed and an intravitreal injection of AAV targeting S1R was administered to investigate the impact of S1R expression on retinal structure, retinal cell apoptosis, and visual function in mice subjected to optic nerve crush. The findings revealed the involvement of BNDF in S1R-mediated neuroprotection and established a novel foundation for the promotion of retinal cell survival following optic nerve injury by S1R. 

Materials and Methods

Animals and Ethics

A total of 48 3-month-old C57BI/6J mice were provided by the Laboratory Animal Department of Central South University. The experiments adhered to the ethical principles stated in the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. This study was approved by the Medical Ethics Committee of Xiangya Hospital affiliated with Central South University (approval number: 2019030405). These mice were randomly divided into four groups, with 12 mice in each group. In the ONC group, the mice underwent ONC followed by intravitreal injection with blank vector. Mice in the S1R-sh group underwent ONC followed by intravitreal injection with AAV targeting downregulation of S1R. Mice in the S1R-AAV group underwent ONC followed by intravitreal injection with AAV targeting overexpression of S1R. The control group mice were administered the blank vector without ONC. 

Optic Nerve Crush and Intravitreal Injections

Surgeries were performed after the animals were anesthetized using an intraperitoneal injection of 1% sodium pentobarbital solution at 60 mg/kg. Curved fine-tip forceps were used to induce focal crush injury to the optic nerve approximately 1 to 2 mm behind the optic globe. AAV (2 µL) containing 1 × 10¹⁰ vector genomes was injected into the vitreous using a 30-gauge needle through the limbus (syringe: 5 µL, #65RN; needle: 33-gauge, 8 mm, point style [pst] 2; Hamilton Company, Reno, NV, USA). Care was taken to avoid penetration of the lens or damage to the vortex veins during the intravitreal injection. Injections were given slowly over 20 seconds to allow diffusion of the vector suspension and to allow intraocular pressure equilibration before needle removal. 

Construction of AAV

Specific primers for mouse S1R were designed to contain the XbaI and NcoI restriction sites (pAAV) and were synthesized by Obio Company (Shanghai, China). The vector maps are shown in Figures 1A, 1B, and 1C. The sequences of primers used for polymerase chain reaction (PCR) were as follows: upstream 5′-CCGGTCTAGAATGCAGTGGGCCGTGGGCCGGCG-3′; downstream 5′-GGAACCATGG TCAAGGGTCTTGGCCGAAGAG-3′. The length of the amplified fragment was 692bp. Pfu DNA polymerase was used to amplify the S1R gene using mouse retina cDNA as a template. The shuttle plasmid and amplified products for the S1R gene were digested with XbaI and NcoI. The digestion products were then recovered, ligated, and transformed into competent DH5a. The clones were extracted from plasmids and identified using restriction enzyme digestion with XbaI and NcoI. The clones were sequenced by the Obio Company. Subsequently, the recombinant adeno-associated virus plasmid, pAAV-CMV-EGFP-P2A-Sigmar1-3xFLAG-WPRE, with the correct sequence was amplified using BJ5183 bacteria. Empty pAAV-CMV-EGFP-P2A-3xFLAG-WPRE vector was used as a control. 

The recombinant AAV plasmid (5 µg) and packaging plasmid (5 µg) were co-transfected with Lipofectamine 2000 (30 µL) into 293T cells in a 10-cm culture dish. After transfection for 24 hours, we observed the transfection. Ten days after transfection, the cells were collected, taken through repeated freeze-and-thaw cycles in dry ice/methanol and a 37°C water bath, and the obtained virus supernatant (1000 g) was centrifuged to remove cell debris at 4°C. The virus supernatant was desalted by iodixanol gradient ultracentrifugation and suspended in phosphate-buffered saline (PBS) buffer. The supernatant was concentrated by centrifugation for 3 hours, and the virus was collected and stored at −80°C. The concentrated virus (20 µL) was diluted six or seven times through serial dilution at a gradient of 1:10, and then quantitative PCR was used to determine the virus titer. 

Hematoxylin and Eosin Staining of Mice Retina

After enucleation, the cornea and the lens were removed, and the optic cup was fixed overnight in freshly prepared 40-g/L paraformaldehyde. The optic cup was then dehydrated with conventional gradient ethanol, rendered transparent using xylene, waxed, embedded, and sectioned into 6-µm-thick paraffin sections. The sections were heated, and conventional hematoxylin and eosin (H&E) staining was conducted. The tissue sections were sealed with a neutral resin seal sheet. Retinal thickness was determined using a microscope by taking three random slices from each group, measuring three times, and taking the average value. 

Immunofluorescence

The mice eyeballs were rinsed with PBS, fixed in O.C.T. Compound (Fisher Scientific, Waltham, MA, USA) in liquid nitrogen, and stored at −80°C for at least 4 hours; they were then sectioned and stored again at −80°C. The following primary antibodies were used for immunofluorescence: S1R, Recombinant Anti-Sigma1-receptor antibody [EPR25145-11] (1:200, ab307548; Abcam, Cambridge, UK); BDNF, Invitrogen BDNF Polyclonal Antibody (1:100, PA5-85730; Thermo Fisher Scientific, Waltham, MA, USA), Invitrogen Goat anti-Rabbit IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 488 (1:500, A-11008; Thermo Fisher Scientific), or Invitrogen Goat anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 568 (1:500, A-11036; Thermo Fisher Scientific). The slides were observed under a ZEISS fluorescence microscope (ZEISS Microscopy, Jena, Germany) after Permount anti-fading medium (Fisher Scientific) was used to mount the cover glass on the slides. 

Protein Extraction and Western Blot Analysis

The neuroepithelial layer of each mouse's retina was weighed and cut into small pieces. The tissue was then lysed in lysis buffer (50-mM Tris/HCl, pH 7.3; 150-mM KCl; 50-mM EDTA; 0.2% Triton X-100; 0.2% NP-40) at a volume of 1:10 on ice for 20 minutes, followed by 15-minute centrifugation at 9000g at 4°C. The supernatant was collected immediately for western blot analysis. Crude protein lysate (25 µg) was mixed with 1 µL of loading dye buffer, heated at 95°C for 1 minute, and directly loaded onto 4% to 21% gradient SDS-PAGE gel together with a prestained standard protein molecular weight marker. Electrophoresis was run at a constant voltage of 120 V for <60 minutes until the front line was close to the edge of the gel. The proteins were then transferred to a polyvinylidene chloride (PVDC) membrane at 4°C overnight and probed with S1R (1:500, 4912; Cell Signaling, Danvers, MA, USA) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; 1:1000, ab8245; Abcam) antibodies. The target bands were analyzed using Bandscan 5.0 image analysis software. GAPDH was used as an internal control to calculate the relative expression of the protein. 

TUNEL Staining and The Quantification of Retinal Cell Number

Mouse eyeball paraffin sections (5 µm) were dewaxed, incubated in ultrapure water containing 2% protein kinase K for 15 minutes at room temperature, and then washed twice in ultrapure water. The TUNEL reaction solution consists of terminal deoxynucleotidyl transferase (TdT) and deoxynucleotide triphosphate (dUTP) with a volume ratio of 1:9. The sections were then stained with TUNEL reaction solution at 37°C for 1 hour in the dark, rinsed twice with PBS, stained with 4′,6-diamidino-2-phenylindole (DAPI), and mounted using 20% glycerol buffer. The slices were observed under a fluorescence microscope at an excitation wavelength of 552 nm and an emission wavelength of 535 nm. Apoptotic cells emitted green fluorescence. Five fields of view were randomly selected for each section, and the apoptosis index was calculated as Apoptosis index = Number of positive cells/Total number of cells × 100%. Photographs were acquired using a Zeiss fluorescence microscope with a 20× objective lens on microscopic fields 200 µm from the optic nerve head. The number of TUNEL-positive cells was quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA). 

Optical Coherence Tomography

The mice were anesthetized as earlier described. Successful anesthetization was indicated by unresponsive limbs and relaxed back. The mouse was placed on a fixed frame and administered mydriatic medicine afterward. The right eyeball was then closely attached to the mirror and positioned for observation. The Phoenix MICRON small animal retina microscopic imaging system (Phoenix-Micron, Bend, OR, USA) was used for OCT examination of the retina and to obtain photographs of the fundus. We measured the thickness of the total retina, inner nuclear layer (INL), and ganglion cell layer (GCL) thickness using manual calipers. After the experiment, antibiotic eye drops were applied to avoid prolonged exposure to dehydration and prevent infection. 

Electroretinogram

Electroretinograms were recorded 4 weeks after intravitreal injection, in accordance with the International Society for Clinical Electrophysiology of Vision. The recording electrode was a ring-shaped corneal contact silver-plated copper electrode with a diameter of 3 mm, and the reference electrode and ground electrode were self-made needle electrodes. Mice were placed in absolute darkness for >12 hours for dark adaptation. The mice were anesthetized and administered compound tropicamide eye drops (Santen Pharmaceutical, Chongqing, China). The head and limbs of the mouse were fixed on the detection platform in the dark room, the recording electrode was placed on the cornea, the reference electrode pierced under the skin of the mouse's front cheek, and the ground electrode pierced under the skin of the mastoid. 

This experiment used the Roland visual electrophysiological system (Roland Consult, Brandenburg, Germany), and the stimulator was a Roland Ganzfeld Q 450. The maximum mixed-response stimulation frequency was 0.095 Hz, and the recording duration was 150 ms. Maximum mixed-reaction a- and b-wave amplitudes were recorded and analyzed. The a-wave amplitude refers to the peak from baseline to the a-wave bottom, and the b-wave amplitude refers to the peak from the a-wave bottom to the b-wave. The entire recording process was carried out under a weak red light. 

Statistical Analysis

Statistical analysis was carried out using SPSS Statistics 22.0 (IBM, Chicago, IL, USA). Data are expressed as mean ± SD with ANOVA or Student's t-tests. Pearson correlation analysis was conducted to compare the expression of S1R and TUNEL-positive cells and the expression of S1R and BDNF. P < 0.05 was considered statistically significant. 

Results

Construction and Evaluation of the AAV Vector

The AAV interference vector (pAAV-U6-shRNA-CMV-EGFP-WPR), the AAV overexpression vector (pAAV-CMV-EGFP-P2A-SigmaR1-3xFLAG-WPRE), and the blank vector (pAAV-CMV-EGFP-P2A-3xFLAG-WPRE) that were used in this study are shown in Figures 1A, 1B, and 1C, respectively. The results of western blot experiments showed that there was a significant difference in the expression of S1R in the retina of mice after virus injection. Compared with the ONC group, the expression of S1R in the S1R-AAV group was upregulated, whereas in the S1R-sh group the expression of S1R was downregulated (Fig. 1D). 

Figure 1.

 

The construction of AAV vectors. (A) AAV interference vector pAAV-U6-shRNA-CMV-EGFP-WPR. (B) AAV overexpression vector pAAV-CMV-EGFP-P2A-SigmaR1-3xFLAG-WPRE. (C) Blank vector pAAV-CMV-EGFP-P2A-3xFLAG-WPRE. (D) Western blot for the expression of S1R in the control group, ONC group, S1R-AAV group, and S1R-sh group.

The vectors carry enhanced GFP so the transfection can be indirectly observed through the degree and location of fluorescence expression. Four weeks after intravitreal injection of AAV, the green fluorescence associated with AAV was distributed in the retina, mainly in the GCL and Müller cells (Fig. 2). The intensity of S1R fluorescence staining was higher in the S1R-AAV group than in the ONC group, and the staining was enhanced in cells located in the GCL and INL. The retina in the S1R-sh group only showed faint red fluorescence (Fig. 2). The above results revealed the effectiveness of the constructed AAV vectors and their capability to integrate into the retina. 

Figure 2.

 

Distribution of AAV and S1R expression in mouse retina 4 weeks after AAV transfection. AAV is shown in green, DAPI in blue, and S1R in red.

Expression of S1R in the Eye Maintained the Structure of the Retina

OCT was performed for retinal morphological structure observation and thickness measurement. The morphology of the retina in the control group reveals distinct layers that are arranged in an orderly fashion (Fig. 3A). In the ONC group (Fig. 3B) and S1R-sh group (Fig. 3D), the retinal layers appear smudged and unclear, with distinct retinal thinning in the S1R-sh group. Compared with the ONC and S1R-sh group, the retina in the S1R-AAV group is thicker, and the structure of each layer is relatively ordered and clear (Fig. 3C). Comparing retinal thickness (mean ± SD), the total retinal thickness for the control group was 185.0 ± 3.5 µm; for the ONC group, 145.5 ± 16.7 µm; for the S1R-AAV group, 164.1 ± 10.1 µm; and for the S1R-sh group, 137.6 ± 9.3 µm. The GCL thickness for the control group was 12.8 ± 2.8 µm; for the ONC group, 8.4 ± 1.6 µm; for the S1R-AAV group, 11.2 ± 2.2 µm; and for the S1R-sh group, 6.2 ± 1.9 µm. The INL thickness for the control group was 42.38 ± 3.8 µm; fo the ONC group, 28.8 ± 6.8 µm; for the S1R-AAV group, 35.8 ± 2.6 µm; and for the S1R-sh group, 33.1 ± 3.1 µm. Measurements of retinal thickness in the control group were consistently thicker than those in the ONC group (total retina, P < 0.0001; GCL, P < 0.0001; INL, P < 0.0001), and the S1R-AAV group also showed higher retinal thickness than the ONC group (total retina, P = 0.001; GCL, P = 0.017; INL, P = 0.002). There was no significant difference in retinal thickness between the ONC group and the S1R-sh group. The thickness of the total retina and INL was higher in the control group than in the S1R-AAV group (total retina, P = 0.0002; INL, P = 0.002). 

Figure 3.

 

The thicknesses of the retinal layers. (A–D) OCT revealed the retinal thickness of the control group (A), the ONC group (B), the S1R group (C), and the S1R-sh group (D). (E–G) Box plots compare the thickness of retinal layers between groups: (E) total retinal thickness, (F) GCL thickness, and (G) INL thickness. Statistically significant at *P < 0.05.

Activation of S1R Improved Survival of Retinal Cells

Because upregulation of S1R preserved the retinal structure and thickness, we performed H&E staining and TUNEL assays in the control group, ONC group, and S1R-AAV group. H&E staining indicated that the retinal cells in the control group were orderly and tightly arranged (Fig. 4A). In the ONC group, the cells in the INL were relatively loosely arranged, whereas the ganglion cells decreased in number and reduced in size with cytoplasmic vacuolation and pyknotic nuclei (Fig. 4B). Compared with the ONC group, the S1R-AAV group presented improved retinal structure, increased density of ganglion cells and INL nuclei, and alleviated cell degeneration (Fig. 4C). In the TUNEL assay, positive TUNEL staining showed green fluorescence, with blue fluorescence of DAPI showing nuclei. The nuclei with green and blue fluorescence were judged to be apoptotic cells. The positive rate (mean ± SD) in the TUNEL assay in the control group was 0.043 ± 0.024; in the ONC group, 0.284 ± 0.113; and in the S1R-AAV group, 0.056 ± 0.022. The number of apoptotic cells in the ONC group was significantly higher than in the control group (P < 0.001) (Fig. 4D), and the cells were mainly distributed in the GCL, INL, and outer nuclear layer (ONL) (Fig. 4E). Compared with the ONC group, S1R-AAV showed reduced apoptotic cells (P = 0.002) with the TUNEL-positive rate decreased by approximately 23% (Figs. 4F, 5A). 

Figure 4.

 

H&E staining and TUNEL assay of the retina. (A–C) H&E staining of retinas in the control group (A), the ONC group (B), and the S1R-AAV group (C). (D–F) TUNEL staining of retinas in the control group (D), the ONC group (E), and the S1R-AAV group (F).

Figure 5.

 

TUNEL-positive rate and expression of S1R. (A) TUNEL-positive rates in the control group, ONC group, and S1R-AAV group. (B) Western blot analyses of S1R of the control group, ONC group, and S1R-AAV group. (C) Relative expression of S1R in the control group, ONC group, and S1R-AAV group. (D) Correlation analysis between the TUNEL-positive rate and S1R expression. Statistically significant at *P < 0.05.

We subsequently performed western blot to understand the effect of S1R on the TUNEL-positive rate (Fig. 5B). Compared with the control group, the relative expression of S1R (the expression ratio of S1R/GAPDH) significantly decreased in the ONC group (P < 0.0001). The relative expression of S1R was elevated in the S1R-AAV group compared to the ONC group (P < 0.0001) (Fig. 5C). Pearson correlation analysis was performed on the expression level of retinal S1R and the TUNEL-positive rate of retinal cells among these three groups, and it showed a negative correlation between S1R and TUNEL-positive rate (r = −0.804, P = 0.002) (Fig. 5D). 

S1R Promotes the Expression of BDNF

The results above showed that the activation of S1R preserved the retinal structure and promoted the survival of neural retinal cells. By immunofluorescence, the fluorescent intensity of BDNF was weaker in the ONC group (Fig. 6B) compared with the control group (Fig. 6A) and the S1R-AAV group (Fig. 6C). We verified that the expression of S1R in the S1R-AAV group was higher compared with the control and ONC groups through western blot analysis of the retina (Fig. 6D). The relative expression of BDNF (the expression ratio of BDNF/GAPDH) of the ONC group was lower than that of the control group (P = 0.002) and the S1R-AAV group (P < 0.0001) (Fig. 6E). There was a positive correlation between BDNF and S1R content (r = 0.979, P < 0.001) (Fig. 6F). 

Figure 6.

 

The relation of BDNF and S1R expression. (A–D) Immunofluorescence staining of BDNF (red) of the retina in the control group (A), the ONC group (B), and the S1R-AAV group (C). (D) Western blot of retinal BDNF in the control, ONC, and S1R-AAV groups. (E) Relative expression of S1R in the control, ONC, and S1R-AAV groups. (F) Correlation analysis between BDNF and S1R expression. Statistically significant at *P < 0.05.

Activation of S1R Preserved the Function of the Eye

We performed ERG on mice to evaluate the visual function (Figs. 7A–7C). In the control group, the amplitude (mean ± SD) of the a-wave was 199.80 ± 46.19 µV, and the amplitude of the b-wave was 459.80 ± 128.98 µV. In the ONC group, the amplitude of the a-wave was 42.48 ± 19.74 µV, and the amplitude of the b-wave was 121.90 ± 67.42 µV. In the S1R-AAV group, the amplitude of the a-wave was 116.47 ± 13.57uV, and the amplitude of the b-wave was 280.43 ± 66.95 µV (Figs. 7D, 7E). ERG analysis showed that the amplitudes of the a- and b-waves were significantly higher in the S1R group than in the ONC group (a-wave, P = 0.001; b-wave, P = 0.02), but significantly lower than that of the control group (a-wave, P < 0.0001; b-wave, P = 0.002), suggesting that S1R preserved mice visual function in part. 

Figure 7.

 

ERG responses and maximum a- and b-wave amplitudes. (A–C) ERG responses of mice in the control group (A), the ONC group (B), and the S1R-AAV group (C). (D) Comparison of ERG maximum a-wave amplitudes among the three groups. (E) Comparison of ERG maximum b-wave amplitudes among the three groups.

Discussion

S1R is widely distributed in the neural retina and is mainly expressed in the endoplasmic reticulum and nuclear membrane of Müller cells and ganglion cells, as well as the nuclear membrane of photoreceptors, bipolar cells, and amacrine cells.¹² In this experiment, the AAV was mainly distributed in the GCL and Müller cells after intravitreal injection of AAV. Optic nerve crush–induced retinal damage is mainly manifested in changes in the retinal structure, thinning of the inner neural retina, and apoptosis of GCL and inner nuclear cells.²⁷ The observed reduction in S1R expression within the ocular tissue of mice afflicted with ONC led to significant structural damage to the retina, including disorganized stratification and an unidentifiable GCL. These findings strongly suggest that S1R plays a critical role in mitigating the retinal damage caused by ONC. This is consistent with the results of Ha et al.,¹¹ who observed a significant loss of GCL in S1R^–/– mice and increased TUNEL-positive cells. In addition, overexpression of S1R resulted in preserved thickness of the total retina, GCL, and INL and reduced the death of retinal cells by 23%. These findings suggest that the protective effect of S1R was exerted by maintaining retinal structure and the survival of retinal cells. Studies have shown that S1R protects RGCs against excitotoxic damage by regulating oxidative stress in Müller cells and the secretion of neuroprotective factors through the Nrf2 signaling pathway.¹⁶ The study conducted by Zhao and colleagues²⁸ demonstrated that the activation of S1R can provide neuroprotection to RGCs in vivo by augmenting the phosphorylation of extracellular signal-regulated kinase 1/2–mitogen-activated protein kinase (ERK1/2/MAPK) under the influence of excitotoxic stress. Similarly, Geva and co-authors²⁹ reported that the administration of pridopidine, a selective S1R agonist, resulted in an increase in both basal and maximal mitochondrial respiration, as well as the stabilization of mitochondrial membrane potential in two distinct rat models of glaucoma, thereby exhibiting neuroprotective properties toward RGCs. Ellis et al.³⁰ demonstrated that the overexpression of S1R mediated by AAV2 in a RGC glucose–oxygen deprivation model significantly augmented the mitochondrial membrane potential while reducing the activity of cytochrome c oxidase and caspase-3/7, thereby enhancing mitochondrial function. Li et al.¹⁷ demonstrated that the activation of S1R via intravitreal injection of AAV2 exhibited a mitigating effect on the apoptotic pathway in RGC losses, leading to the restoration of both RGC quantity and functionality. Our experiment revealed that, in addition to preserving RGC, the upregulation of S1R expression in the presence of ONC can also effectively maintain the structural integrity of other retinal layers, including total retinal thickness and INL thickness. The rationale behind this is that S1R has the ability to enhance the viability of retinal cells following damage, as demonstrated by the TUNEL assay. Moreover, the inverse relationship observed between the positive rate of TUNEL and S1R expression provides additional evidence to support this claim. However, future studies are required to determine if direct activation of S1R on Müller cells can promote the repair of neural retina. 

To investigate the neuroprotective mechanism of S1R in the context of retinal injury caused by ONC in mice, we conducted an analysis of the expression levels of S1R and BDNF proteins within the retina. BDNF, a potent neuroprotective agent, primarily facilitates the activation of survival effector pathways through its interaction with TrK receptors.³¹ This interaction enhances downstream signaling, including activation of the PI3K/Akt pathway to inhibit pre-apoptotic targets and activation of the extracellular signal-regulated protein kinase (ERK)/cAMP response element binding protein (CREB) pathway to induce the transcription and expression of genes associated with neuronal survival.³² BDNF exhibits a dual mechanism of action, whereby it not only exerts a direct neuroprotective effect on ganglion cells but also indirectly influences glial cells to upregulate the expression of neurotrophic factors, thereby exerting its effects indirectly.³³ Our findings revealed a significant increase in BDNF expression within the S1R-AAV group compared to the ONC group, effectively compensating for the downregulation of BDNF induced by ONC. This suggests a correlation between the upregulation of S1R expression and the observed elevation in BDNF content. In conjunction with the aforementioned experimental findings, it is hypothesized that the S1R potentially facilitates cellular survival by directly inducing the expression of BDNF and that this protective role encompasses the maintenance of retinal structural integrity, promotion of neuronal survival, and safeguarding of visual function in mice. 

ERG analysis indicated that S1R activation improved the visual function of mice based on the amplitude of a- and b-waves. It has been reported that S1R^–/– mice show decreased ERG b-wave amplitude, decreased negative scotopic threshold responses, and inner retinal dysfunction at 12 months, which is consistent with our results.¹¹ The amplitudes of a- and b-waves were significantly higher in the S1R-AAV group compared with the ONC group. As such, the waveform suggests ameliorated function of rod and cone photoreceptors and bipolar cells. Taken together, the increased expression of S1R improved the visual function of mice after ONC. 

To assess the neuroprotective function of S1R, there have been in vivo studies evaluating the effect of activating S1R in alleviating retinal degeneration. Smith et al.³⁴ found that intraperitoneal injection of (+)-pentazocine, a highly specific S1R ligand, in diabetic mice reduced the apoptosis of retinal neurons and effectively alleviated the death of RGC caused by excitotoxins (glutamate and homocysteine) in vitro. In this study, the expression of S1R was upregulated through intravitreal injection of AAV, which is more targeted. AAV has emerged as the most secure viral vector at present due to its minimal immunogenicity and pathogenicity. Furthermore, the augmentation of tissue targeting has enhanced the precision and effectiveness of transfection, rendering it a popular choice for gene transfer to regulate gene expression. In this study, intravitreal viral vector injection was employed to enhance the local tissue virus content and transfection efficiency, resulting in a noteworthy increase in S1R production and endogenous BDNF expression. Through this mode of administration, AAV is more likely to be distributed on the easily damaged GCL and INL, which is more suitable for retinal damages caused by neural degeneration or optic nerve compression. 

In conclusion, upregulation of S1R through intravitreal injection of AAV in the retina can effectively maintain the morphology of the retina and protect visual function in mice after ONC. These findings provide deeper insights into the neuroprotective role of S1R and demonstrate the potential of S1R overexpression in the treatment of optic nerve diseases. 

Acknowledgments

Supported by a grants from the Horizontal Science and Technology Project of Central South University (W2017QRX07) and National Natural Science Foundation of China (81900888). 

Disclosure: Y. Chen, None; X. Yang, None; J. Mao, None