Synapses and dendritic spines are eliminated in the primary visual cortex of mice subjected to chronic intraocular pressure elevation

Abstract

Synaptic plasticity is essential for maintaining neuronal function in the central nervous system and serves as a critical indicator of the effects of neurodegenerative disease. Glaucoma directly impairs retinal ganglion cells and their axons, leading to axonal transport dysfuntion, subsequently causing secondary damage to anterior or posterior ends of the visual system. Accordingly, recent evidence indicates that glaucoma is a degenerative disease of the central nervous system that causes damage throughout the visual pathway. However, the effects of glaucoma on synaptic plasticity in the primary visual cortex remain unclear. In this study, we established a mouse model of unilateral chronic ocular hypertension by injecting magnetic microbeads into the anterior chamber of one eye. We found that, after 4 weeks of chronic ocular hypertension, the neuronal somas were smaller in the superior colliculus and lateral geniculate body regions of the brain contralateral to the affected eye. This was accompanied by glial cell activation and increased expression of inflammatory factors. After 8 weeks of ocular hypertension, we observed a reduction in the number of excitatory and inhibitory synapses, dendritic spines, and activation of glial cells in the primary visual cortex contralateral to the affected eye. These findings suggest that glaucoma not only directly damages the retina but also induces alterations in synapses and dendritic spines in the primary visual cortex, providing new insights into the pathogenesis of glaucoma.

Introduction

Synaptic plasticity, a fundamental mechanism underlying learning and memory in the central nervous system (CNS), refers to the ability of synaptic structure, function, and strength to adapt to neuronal activity (Magee and Grienberger, 2020; Bellingacci et al., 2023). Throughout the various stages of synaptic establishment, growth, and maturation, regulation of synaptic plasticity within the CNS involves diverse pathways, with significant involvement from glial cells (Morris et al., 2013; Wang et al., 2016; Batool et al., 2019). Dendritic spine formation and elimination are also essential processes influencing synaptic plasticity. Under pathological conditions, such as in some neurodegenerative diseases, synaptic plasticity may be disrupted, thereby impacting the neural circuitry (Bae and Kim, 2017; Skaper et al., 2017; Batool et al., 2019; Chu, 2020; Dorszewska et al., 2020; Niraula et al., 2023, 2024). Consequently, exploring pathological alterations in synaptic plasticity and understanding the underlying mechanisms are crucial for elucidating the pathogenesis of CNS disease.

Glaucoma, a leading cause of irreversible blindness worldwide, is a multifactorial optic neuropathy characterized by the progressive loss of retinal ganglion cells (RGCs) and the presence of visual field defects (Sulak et al., 2024). Previously, glaucoma was believed to mainly affect RGCs and their axons, leading to impairment of retinal structure and function. However, injury to RGC and their axons can result in the blockade of retrograde and anterograde transport, deficiency in secondary growth factors and nutrients, ischemia, hypoxia, and the accumulation of inflammatory cytokines due to oxidative stress. Additionally, immune abnormalities and glutamate toxicity can further exacerbate damage to neurons and nerve fibers within the visual pathway, ultimately resulting in trans-synaptic degeneration.

Therefore, recent research has revealed that the impact of glaucoma extends beyond the retina, inducing changes in both the structure and function of the entire visual pathway (Liu et al., 2014; Sapienza et al., 2016; Hvozda Arana et al., 2020, 2021; Van Hook et al., 2020; Tribble et al., 2021; Fujishiro et al., 2022; Yan et al., 2022; Yang et al., 2024). The available evidence indicates that chronic ocular hypertension (OHT) can result in dendritic reconstruction of neurons, decreased neuronal excitability, and modifications in synaptic plasticity within the lateral geniculate nucleus (LGN) (Liu et al., 2014; Van Hook et al., 2020; Yu et al., 2023). In a primate model of OHT, Lam et al. (2003) observed changes in the expression levels of multiple proteins associated with synaptic plasticity in the visual cortex. These findings collectively suggest that glaucoma has the potential to modulate synaptic plasticity in the primary visual cortex (V1). Using two-photon in vivo imaging, we previously reported a decrease in dendritic spines in the contralateral V1 after optic nerve crush (ONC) surgery (Zhan et al., 2020). We have also confirmed alterations in neural projections in the retinorecipient regions connected to RGCs post-ONC via light sheet microscopy (Zhan et al., 2023). Nonetheless, there are important differences between ONC animal models and the clinical pathogenesis of glaucoma. Moreover, most previous research on OHT has centered predominantly on synaptic dysfunction in the LGN or superior colliculus (SC), with few animal studies investigating neuronal and synaptic alterations in the V1. Thus, our understanding of glaucoma-related synaptic alterations in the V1 region and the underlying mechanisms remains incomplete.

The aim of the current study was to characterize longitudinal changes in neurons, synapses, dendritic spines, and glial cells in brain tissue over time in a mouse model of OHT. Furthermore, we used RNA sequencing (RNA-seq) to comprehensively investigate changes in the V1 region and the associated genes/pathways.

Methods

Animals and anesthesia

Wild-type C57BL/6J mice of both sexes, aged 6–8 weeks (20–26 g), were procured from Jiangsu Jicui Yaokang Biotechnology Co., Ltd. (Nanjing, Jiangsu Province, China; license number: SCXK (Su) 2023-0009). To facilitate direct observation of dendritic spines on pyramidal neurons, our study utilized the Thy1-YFP transgenic mouse strain, which expresses yellow fluorescent protein specifically in its pyramidal neurons (Feng et al., 2000). Male and female Thy1-YFP transgenic mice (YFP-H-Line), aged 6–8 weeks (20–26 g), expressing yellow fluorescent protein in their pyramidal neurons were acquired from Zhongshan School of Medicine, Sun Yat-sen University in China for two-photon microscopy (TPM) imaging. The mice used in the study were housed under a 12/12-hour light/dark cycle and supplied with standard food and water (room temperature: 23 ± 2°C, air humidity: 50%–60%). All the experimental procedures complied with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research (Association for Research in Vision and Ophthalmology, 2021) and were approved by the Institutional Animal Care and Use Committee of Zhongshan Ophthalmic Center (approval No. Z2021004; approval date: June 8, 2021). Both wild-type C57BL/6J mice and Thy1-YFP transgenic mice were randomly divided into an untreated control group (NC group) and an OHT group. Notably, 90% of optic nerves in mice project to the opposite side of the brain (Kondo et al., 1993; Liu et al., 2011). In our study, we thus investigated alterations in brain regions both ipsilateral and contralateral to the OHT eye and compared them with the same brain regions in the NC group. The brain regions on the ipsilateral side of the OHT eye were classified as the OHT-ipsi group, while those on the contralateral side of the OHT eye were categorized as the OHT-contra group. For all experimental procedures, the mice were anesthetized via intraperitoneal injection of Avertin (250 mg/kg, Sigma-Aldrich, Darmstadt, Germany). A total of 138 C57BL/6J mice were used in the experiment, including 36 mice used for intraocular pressure (IOP) measurement and RGC counting, 30 for immunohistochemistry, 36 for immunofluorescence, 30 for enzyme-linked immunosorbent assay, 6 for RNA-seq (two groups, three time points); and 15 Thy1-YFP mice were used for two-photon continuous in vivo imaging.

Chronic ocular hypertension model and intraocular pressure measurement

We used a magnetic microbead occlusion approach, which was described previously (Ito et al., 2016; Tan et al., 2022), to induce unilateral OHT. Briefly, wild-type C57BL/6J mice or Thy1-YFP transgenic mice were anesthetized by intraperitoneal injection of Avertin, and topicalamide (5 mg/mL, Shenyang Xingqi Pharmaceutical Company, Shenyang, China) was administered for pupil dilation. Then, procaine hydrochloride (5 mg/mL, Alcon, Shenyang Xingqi Pharmaceutical Company) was applied topically to anesthetize the corneal surface. Next, a disposable 32G injection needle was used to puncture the cornea, and approximately 2 µL of a suspension containing magnetic microbeads (10 µm in diameter, BM547, Bangs Laboratories, Fishers, IN, USA) was injected into the anterior chamber of the left eye using a Hamilton Microlitre syringe (Hamilton Company, Reno, NV, USA). Finally, a handheld magnet was used to evenly distribute the microbeads within the anterior chamber, effectively obstructing drainage of aqueous humor, thereby increasing the IOP.

The OHT model was evaluated based on changes in IOP and RGC counts. In both the NC group and OHT group, the IOP was measured in both eyes (Additional Figure 1 ^{(573.7KB, tif)} ). The IOP was first measured on postoperative day 4 and subsequently measured every week up to 8 weeks post-injection using a TonoLab tonometer (Icare, Vantaa, Finland) following the manufacturer’s guidelines. The TonoLab tonometer takes five measurements (excluding outliers) and calculates the average, which is considered one reading. Three machine-generated readings were acquired per eye, and the mean value served as the final IOP. The IOP measurements were consistently taken at 10 a.m. to minimize variations due to circadian rhythms. A detailed experimental flowchart is shown in Figure 1.

Figure 1.

Detailed experimental flowchart.

(A) Flowchart of experiments involving C57BL/6J mice. Mice in both groups underwent IOP measurement on days 0, 4, 7, 14, 21, 28, 35, 42, 49, and 56. Flat-mounts of retinas, IHC/IF of brain tissues, and ELISA assays were carried out on days 28, 42, and 56 for both groups. RNA-seq was performed on day 56 in both groups. (B) Flowchart of experiments involving Thy1-YFP mice. Mice in both groups underwent IOP measurement on days 0, 4, 7, 14, 21, 28, 35, 42, 49, and 56. Mice in both groups were observed by TPM on days 0, 14, 28, 42, and 56. On day 0, the OHT group underwent unilateral OHT surgery immediately after two-photon imaging. IOP: intraocular pressure; ELISA: Enzyme linked immunosorbent assay; IF: immunofluorescence; IHC: immunochemistry; NC: no-treatment control; OHT: ocular hypertension; RNA-seq: RNA sequencing; w: week.

Immunofluorescence staining of whole-mount retinas and retinal ganglion cell counting

Briefly, mice were deeply anesthetized and transcardially perfused with 0.9% sodium chloride solution, followed by 4% paraformaldehyde, at weeks 4, 6, or 8. Subsequently, the eyes were excised and fixed in 4% paraformaldehyde for 2 hours at room temperature. The intact retinas were dissected and subjected to four radial incisions, creating a petal-like shape. The retinas were permeabilized using 2% Triton X-100, followed by a 2-hour blocking step with 5% goat serum (Boster, Wuhan, China), and subsequently incubated overnight at 4°C with an anti-RNA binding protein with multiple splicing (RBPMS) antibody (rabbit, 1:100, ProteinTech, Rosemont, IL, USA, Cat# 15187-1-AP, RRID: AB_2238431). Following three washes, the retinas were subjected to a 2-hour incubation at room temperature with a goat anti-rabbit IgG secondary antibody labeled with Alexa Fluor 488 (1:500, Cell Signaling Technology, Danvers, MA, USA, Cat# 4412, RRID: AB_1904025). Finally, the retinas were flattened and mounted on glass slides. Images were obtained with a panoramic scanning microscope (Tissuefaxs, TissueGnostics, Vienna, Austria) using a 20× objective. RGCs were quantified following established protocols (Wu et al., 2020; Chen et al., 2022). In all quadrants, 12 regions (450 µm × 320 µm) were evaluated at equidistant points (1/6, 3/6, and 5/6 of the retinal radius) from the optic disc (each region was assessed three times at these distances) (Figure 2A).

Figure 2.

Induction of OHT via magnetic microbead injection, and RGC loss.

(A) Defined regions (450 µm × 320 µm) in the central, middle, and peripheral regions of the retina are indicated by boxes. Scale bar: 500 µm. (B) The IOP curve following anterior chamber injection of magnetic microbeads (n = 10/group, mean ± SD, ***P < 0.001, vs. control; two-way analysis of variance followed by Sidak’s multiple comparisons test). (C) Representative images of RBPMS staining (Alexa Flour 488, green) in the central, middle, and peripheral regions of the retina in the NC group and on the OHT side at 4, 6, and 8 weeks. After 4, 6, and 8 weeks of OHT, the RGCs in the central, middle, and peripheral regions of the retina on the OHT side were significantly reduced compared with the NC group. Scale bars: 50 µm. (D–F) Quantification of RBPMS-positive RGCs at 4, 6, and 8 weeks post‐OHT in the central (D), middle (E), and peripheral (F) regions of the retina in the NC group and on the OHT side (n = 6/group, mean ± SD, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). IOP: Intraocular pressure; NC: no-treatment control; ns: not significant; OHT: ocular hypertension; RBPMS: RNA binding protein with multiple splicing; RGC: retinal ganglion cell; w: week.

Immunohistochemistry analysis of brain tissues

As described above, the mice were euthanized at 4, 6, or 8 weeks post-surgery. After transcardiac perfusion, the brain tissues were excised and placed in 4% paraformaldehyde overnight, followed by dehydration and paraffin embedding. Coronal sections 5 µm in thickness were prepared for immunohistochemistry. Prior to staining, the sections were subjected to deparaffinization and hydration, followed by antigen retrieval using a sodium citrate antigen repair solution (P0081, Beyotime, Shanghai, China). Next, the sections were treated with 3% hydrogen peroxide for 10 minutes and subsequently rinsed three times with distilled water. Identification of the target proteins was accomplished using a DAB Detection Immunohistochemistry Kit (Abcam, Waltham, MA, USA, Cat# ab64264). Each section was first blocked using a blocking solution and subsequently incubated overnight at 4°C with primary antibodies targeting neuronal nuclei (NeuN, rabbit, 1:200, Cell Signaling Technology, Cat# 24307, RRID: AB_2651140), glial fibrillary acidic protein (GFAP, rabbit, 1:200; Cell Signaling Technology, Cat# 12389, RRID: AB_2631098), and ionized calcium binding adaptor molecule-1 (Iba1, rabbit, 1:800; Cell Signaling Technology, Cat# 17198, RRID: AB_2820254). After three rinses in phosphate-buffered saline, a biotinylated goat anti-polyvalent secondary antibody from the DAB Detection Immunohistochemistry Kit was applied, and the sections were incubated for 10 minutes at room temperature. Following three additional washes, streptavidin peroxidase was applied for 10 minutes at room temperature. After three washes, DAB staining solution was added, and the sections were incubated for 10 minutes at room temperature according to the instruction manual. The sections were then counterstained with hematoxylin and covered with coverslips. Imaging was performed using a panoramic scanning microscope (Tissuefaxs, TissueGnostics, Vienna, Austria) with a 20× objective. The approximate locations of the LGN, SC, and V1 were identified using a brain atlas, and we focused primarily on layers 2 to 6 within V1 for examination. Thereafter, three distinct areas (400 µm × 300 µm) in each brain region were selected to calculate the average optical density value for statistical analysis. To analyze neurons, GFAP-positive areas and Iba1-positive areas were quantified using ImageJ software (Version1.52g, National Institutes of Health, Bethesda, MD, USA) (Schneider et al., 2012), along with an immunohistochemistry toolbox plug-in.

Immunofluorescence staining of brain tissues

Mice were sacrificed and perfused at postoperative weeks 4, 6, or 8. The brains were extracted and immersed in 4% paraformaldehyde overnight, followed by dehydration and embedding in Tissue-Tek OCT compound (SAKURA, Torrance, CA, USA). Then, 15-µm coronal frozen sections were prepared for immunofluorescence analysis. The sections were permeabilized with 0.5% Triton X-100 for 20 minutes, followed by a 1-hour blocking step using 5% goat serum. Subsequently, they were incubated overnight at 4°C with primary antibodies against vesicular glutamate transporter (VGlut1, rabbit, 1:200, Abcam, Cat# ab227805, RRID: AB_2868428), postsynaptic density protein 95 (PSD95, rabbit, 1:100, Cell Signaling Technology, Cat# 3450, RRID: AB_2292883), vesicular γ-aminobutyric acid transporter (VGAT, rabbit, 1:100, Novus, Cat# NBP2-20857, RRID: AB_3101899), and gephyrin (rabbit, 1:100, Abcam, Cat# ab177154, RRID: AB_3101900). Following three washes, the sections were treated with a secondary goat anti-rabbit antibody conjugated to Alexa Fluor 488 and Alexa Fluor 555 (1:500, Cell Signaling Technology, Cat# 4412, RRID: AB_1904025 and Cat# 4413, RRID: AB_10694110) for 1 hour in the dark at room temperature. Finally, coverslips were mounted with the antifade reagent 4′,6-diamidino-2-phenylindole (Cell Signaling Technology, Cat# 8961). Imaging of glutamatergic and GABAergic synapse puncta was performed using a confocal laser scanning microscope (LSM980, Carl Zeiss, Oberkochen, Germany) with a 40× oil-immersion objective. Images were captured at a resolution of 2048 × 2048 pixels, with an average of three images taken from each side of the brain (left/right), mainly observing layers 2 to 6 of the V1. The Synapse Counter plug-in in ImageJ was used to identify and analyze the synapse puncta (marked by VGlut1, PSD95, VGAT, and gephyrin), as previously described (Dzyubenko et al., 2016; Van Hook et al., 2020). Only positive puncta within the following size ranges were counted. The size thresholds for vGlut1, PSD95, VGAT, and gephyrin were as follows: vGlut1 0.25–2.2 μm², PSD95 0.35–1.8 μm², VGAT 0.18–2.0 μm², and gephyrin 0.13–1.8 μm².

Enzyme-linked immunosorbent assay

Mice were anesthetized and euthanized at postoperative weeks 4, 6, or 8 following a previously described protocol. Brain samples were collected, homogenized, and then centrifuged at 12,000 × g for 10 minutes at 4°C, and the resulting supernatant was then isolated and stored at −80°C for further use. Quantitative evaluation of tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β) levels was conducted using commercially available high-sensitivity kits following the manufacturer’s instructions (TNF-α: Cat# abs520010, Absin, Shanghai, China; IL-1β: Cat# abs520001, Absin). Total protein concentrations were assessed using a BCA protein assay kit (Cat# GK10009, GLPBIO, Montclair, CA, USA) to normalize the results based on sample size. The cytokine levels were quantified in picograms per milligram (pg/mg).

In vivo long-term transcranial two-photon imaging and image analysis

The surgical procedure outlined here closely followed that described in previously published articles (Yang et al., 2010; Zhan et al., 2020; Huang et al., 2021). Thy1-YFP transgenic mice were deeply anesthetized with Avertin. Each mouse’s head was shaved, followed by making a midline incision on the scalp to expose the skull. The periosteum tissue overlying the skull was carefully removed. A specific point was designated as the center of the V1 region (3 mm posterior and 3 mm to the right/left of the bregma) and marked accordingly (Allen Institute for Brain Science, 2024). Next, a head fixture consisting of a metal ring and two 1.2-g bars was affixed to the skull using adhesive. The central axis of the head fixture was aligned with the designated point to stabilize the head. Once the adhesive had set, a thinned-skull window measuring 3 mm in diameter was generated employing a high-speed drill, followed by delicate extraction of the thinned portion of the skull using microforceps. The dura was then gently removed with microforceps to enable imaging. Ultimately, a glass coverslip, larger than the exposed cortex, was positioned over the cranial window and affixed with adhesive. Mice were injected intraperitoneally with dexamethasone (5 mg/kg; Cat# GC31658, GLPBIO) every 2 days to prevent inflammation until the end of the observation period (which lasted up to 8 weeks).

The methods used for imaging and analyzing dendritic spines were described previously (Zhan et al., 2020; Huang et al., 2021). The dendritic spines of pyramidal neurons in V1 were observed using a two-photon microscope (FVMPE-RS multiphoton laser scanning microscope; Olympus, Tokyo, Japan) configured at the ideal 920-nm wavelength, using a 25× water-immersion objective with a 1.05 numerical aperture. Employing 2× digital zoom, detailed images were captured (2048 × 2048 pixels, 0.75-µm Z step size). Following the initial imaging, the same neuron in the same brain region of the same mouse was imaged repeatedly at weeks 2, 4, 6, and 8. The images were analyzed using ImageJ software. Elongated, thin protrusions without enlarged heads were considered to be filopodia, whereas other protrusions were considered to be spines. A spine was classified as a different or new spine if it deviated by more than 0.7 µm from the anticipated position established in the previous imaging session. The rate of spine formation or elimination was calculated by dividing the number of newly formed or eliminated spines by the total number of pre-existing spines. Image processing and quantification of dendritic spines were performed using ImageJ software.

RNA sequencing analysis

After 8 weeks of OHT, the mice were euthanized, and the bilateral V1 tissues were collected and immediately sent to Personal Biotechnology Co., Ltd. (Shanghai, China) for complementary DNA library construction and sequencing on an Illumina NovaSeq 6000 platform (Illumina, San Diego, CA, USA). Briefly, the library fragments underwent polymerase chain reaction enrichment, quality control assessment, and subsequent sequencing on the Illumina platform. Raw reads in FASTQ format were generated, and, after filtering out low-quality and adapter-containing data, the resulting clean reads were aligned to the reference genome (http://www.ensembl.org/Mus_musculus/Info/Annotation) using HISAT2 software (version2.2.0, http://ccb.jhu.edu/software/hisat2/index.shtml). Transcript levels were evaluated using fragments per kilobase per million reads (FPKM), and differential gene expression was analyzed using DESeq (version1.38, http://bioconductor.org/packages/release/bioc/html/DESeq2.html). Volcano maps of the differentially expressed genes (DEGs) were created with the ggplots2 R package. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using the Kyoto Encyclopedia of Genes and Genomes database (http://www.kegg.jp/). Data analysis was carried out on the web-based platform Personalbio GenesCloud (https://www.genescloud.cn).

Statistical analysis

All data are presented as the mean ± standard deviation (SD). Statistical analysis was conducted using Prism software (version 9.0 for Windows, GraphPad Software, Boston, MA, USA, www.graphpad.com). Differences among groups were evaluated using one-way or two-way analysis of variance, followed by Sidak’s multiple comparisons test. A two-tailed P value less than 0.05 was considered to be statistically significance.

Results

Magnetic microbead-induced chronic ocular hypertension and retinal ganglion cell loss

Magnetic microbeads were unilaterally injected into the left anterior chamber of the eye of adult C57BL/6 mice to induce OHT, while the untreated right eye served as a control. There was no significant difference in baseline IOP between the eye with OHT and the NC eye. A notable increase in IOP was observed on the fourth day following magnetic microbead injection, with the OHT eyes exhibiting a significantly higher IOPs than the eyes in the NC group (NC vs. OHT: P < 0.0001). Weekly IOP measurements were taken thereafter. In our glaucoma model, IOP peaked at 4 weeks post-OHT (NC vs. OHT: P < 0.0001). Subsequently, the IOP gradually decreased, although a statistically significant difference persisted between the two groups until the 7^th week post-injection (NC vs. OHT: P < 0.0001). By the 8^th week, no significant difference in mean IOP was observed between the OHT eyes and eyes in the NC group (Figure 2B).

Next, the loss of RGCs resulting from OHT was quantified. Immunostaining using an antibody against RBPMS allowed for RGC quantification across different regions of the retina. After 4 weeks of elevated IOP, a significant decrease in the number of RGCs was observed in the central, middle, and peripheral areas of the retina compared with the same regions in the NC group (NC vs. OHT: P < 0.0001). As the duration of elevated IOP increased to 6 and 8 weeks, RGC loss worsened across all retinal regions (Figure 2C–F). These results underscore the effectiveness of anterior chamber injection of magnetic microbeads in inducing chronic IOP elevation because it resulted in a 20%–33% loss of RGCs in mice, confirming its value as an OHT model.

Decreased neuronal soma size in the lateral geniculate nucleus and superior colliculus contralateral to the ocular hypertension eye

Neurons serve as the foundation for creating intricate circuits and processing/transmitting information within the CNS (Koch and Segev, 2000). Therefore, we initially investigated whether OHT had an impact on total neuron counts and neuronal soma size. The neuron-specific marker NeuN was used to stain neurons (Figure 3A–C). Our findings indicated that, irrespective of the duration of elevated IOP (4, 6, or 8 weeks), the number of neurons in the LGN, SC, and V1 regions did not significantly differ among the three groups (OHT-contra/ipsi and NC) (Figure 3D–F). Interestingly, following a 4-week increase in IOP, we observed a decrease in neuronal soma size within the LGN and SC regions contralateral to the OHT eye compared with the ipsilateral side and the NC group (NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001; Figure 3G and H). This reduction in soma size persisted at 6 and 8 weeks post-OHT surgery in the contralateral LGN and SC regions (Figure 3G and H). Notably, there were no significant differences in soma size across the three time points (4, 6, and 8 weeks) in the contralateral LGN and SC regions. Conversely, within the V1 region, no significant changes in neuron soma size were observed contralateral to the OHT eye, even after 8 weeks of sustained IOP elevation, compared with the ipsilateral side and the NC group (Figure 3I). Thus, chronic OHT persisting for 4 weeks or more could result in a decrease in neuronal soma size in the LGN and SC regions contralateral to the OHT eye.

Figure 3.

Neuron numbers and neuronal soma size in the LGN, SC, and V1 regions after OHT.

(A–C) Representative IHC images of NeuN staining of the LGN (A), SC (B) and V1 (C) regions of the NC group and the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks after OHT. OHT did not affect the number of neurons in LGN, SC, and V1. However, following 4 weeks of OHT, a reduction in neuronal soma size was noted in the contralateral LGN and SC regions, whereas the neuronal soma size in V1 exhibited no changes after 8 weeks. Scale bars: 50 µm. (D–F) Changes in neuron counts following OHT (n = 5/group, means ± SD, two-way analysis of variance followed by Sidak’s multiple comparisons test). (G–I) Changes in neuronal soma size following OHT (n = 5/group, mean ± SD, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). Contra: Contralateral; IHC: immunochemistry; Ipsi: ipsilateral; LGN: lateral geniculate nucleus; NC: no-treatment control; NeuN: neuronal nuclei; OHT: ocular hypertension; SC: superior colliculus; V1: primary visual cortex; w: week.

Loss of glutamatergic and GABAergic synapses in the V1 region contralateral to the ocular hypertension eye

Despite 8 weeks of OHT, no significant alterations in neuron counts or neuronal soma size were observed within the V1 region. However, neuronal atrophy in the LGN and SC regions may modify the signal input to the V1 region. Therefore, we further investigated whether there were any changes in the number of excitatory/inhibitory synapses within the visual cortex. Glutamatergic and GABAergic synapses were identified using the presynaptic and postsynaptic markers VGlut1/PSD95 and VGAT/gephyrin, respectively. Synaptic puncta, comprising clusters of presynaptic and postsynaptic proteins that indicate synapse formation and maturation, were compared among the three groups (Dzyubenko et al., 2016).

Initially, we asked whether the size or density of VGlut1 puncta changed in response to elevated IOP (Figure 4A–D). We detected no signs of atrophy or swelling in glutamatergic presynaptic terminals, as the punctum size remained consistent in the V1 region across the three groups at varying time intervals. However, a notable decrease in VGlut1 punctum density was observed in the V1 region contralateral to the OHT eye compared with the other two groups in postoperative week 8 (NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001). Neither the size nor the density of PSD95-positive puncta exhibited significant changes, even after sustaining high IOP for 8 weeks. Notably, there was a marginal decrease in PSD95-positive punctum density in the OHT-contra V1 region after 6 and 8 weeks of sustained high IOP in comparison with that in the other groups (although this difference was not statistically significant) (Figure 4E–H).

Figure 4.

Excitatory/inhibitory pre/postsynaptic protein marker VGlut1/PSD95/VGAT/gephyrin staining in the V1 region after unilateral OHT.

(A) Representative merged IF images (VGlut1 staining (Alexa Flour 555, red), DAPI staining (blue)) in the V1 region in the NC group and on the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks after OHT. Eight weeks of OHT led to a reduction in the number of Vglut1-positive puncta in the contralateral V1 region, with no significant alteration in puncta size at different time points. Scale bars: 20 µm. (B) Representative images of VGlut1 staining after magnification. Each arrow (white) indicates a VGlut1-positive punctum. Scale bar: 10 µm. (C) Quantitation of VGlut1-positive puncta in the V1 region following OHT (n = 5/group, means ± SD, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). (D) VGlut1-positive puncta size in the V1 region following OHT (n = 5/group, means ± SD, two-way analysis of variance followed by Sidak’s multiple comparisons test). (E) Representative merged IF images (PSD95 staining (Alexa Flour 488, green), DAPI staining (blue)) in the V1 region of the NC group and the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks after OHT. Four, six, and eight weeks of OHT did not significantly affect the number and size of PSD95-positive puncta in the contralateral V1 region. Scale bars: 20 µm. (F) Representative images of PSD95 staining after magnification. Each arrow (white) indicates a PSD95-positive punctum. Scale bar: 10 µm. (G) Quantitation of PSD95-positive puncta in the V1 region following OHT (n = 5/group, means ± SD, two-way analysis of variance followed by Sidak’s multiple comparisons test). (H) PSD95-positive puncta size in the V1 region following OHT (n = 5/group, means ± SD, two-way analysis of variance followed by Sidak’s multiple comparisons test). (I) Representative merged IF images (VGAT staining (Alexa Flour 555, red), DAPI staining (blue)) in the V1 region of the NC group and the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks after OHT. Eight weeks of OHT led to a reduction in the number of VGAT-positive puncta in the contralateral V1 region, with no significant alteration in puncta size at different time points. Scale bars: 20 µm. (J) Representative images of VGAT staining after magnification. Each arrow (white) indicates a VGAT-positive punctum. Scale bar: 10 µm. (K) Quantitation of VGAT-positive puncta in the V1 region following OHT (n = 6/group, means ± SD, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). (L) VGAT-positive puncta size in the V1 region following OHT (n = 6/group, means ± SD, two-way analysis of variance followed by Sidak’s multiple comparisons test). VGAT: vesicular γ-aminobutyric acid transporter. (M) Representative merged IF images (gephyrin staining (Alexa Flour 488, green), DAPI staining (blue)) in the V1 region of the NC group and the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks after OHT. Six and eight weeks of OHT led to a reduction in the number of gephyrin-positive puncta in the contralateral V1, with no significant alteration in puncta size at different time points. Scale bars: 20 µm. (N) Representative images of gephyrin staining after magnification. Each arrow (white) indicates a gephyrin-positive punctum. Scale bar: 10 µm. (O) Quantitation of gephyrin-positive puncta in the V1 region following OHT (n = 6/group, means ± SD, *P < 0.05, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). (P) Gephyrin-positive puncta size in the V1 region following OHT (n = 6/group, means ± SD, two-way analysis of variance followed by Sidak’s multiple comparisons test). Contra: Contralateral; DAPI: 4′,6-diamidino-2-phenylindole; IF: immunofluorescene; Ipsi: ipsilateral; NC: no-treatment control; OHT: ocular hypertension; V1: primary visual cortex; VGlut1: vesicular glutamate transporter 1; PSD95: postsynaptic density protein 95; VGAT: vesicular γ-aminobutyric acid transporter; w: week.

Subsequently, we analyzed variations in GABAergic synapses. We found no significant differences in VGAT or gephyrin puncta size among the three groups at postoperative weeks 4, 6, and 8. However, following 8 weeks of OHT, a notable decrease in VGAT punctum density was noted in the contralateral V1 region (NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001; Figure 4I–L). Additionally, we observed a decrease in the density of the inhibitory postsynaptic terminal marker gephyrin at 6 weeks (NC vs. OHT-contra: P = 0.0223, OHT-ipsi vs. OHT-contra: P = 0.0416) and 8 weeks (NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001) post-OHT (Figure 4M–P).

Increased elimination and reduced formation of dendritic spines in the V1 region contralateral to the ocular hypertension eye

The observed decrease in the number of synapses in V1 described in the preceding section indicated a potential alteration in synaptic plasticity. Synaptic plasticity encompasses not only modifications in synaptic structure and quantity, but also, crucially, formation and elimination of dendritic spines (Chidambaram et al., 2019). To determine whether OHT triggers changes in dendritic spine plasticity, dendritic spines on L2/3 pyramidal neurons in the visual cortex were repetitively imaged using transcranial TPM over a 2-week period extending up to 8 weeks post-OHT surgery (Figure 5A and B). In the contralateral V1 region, 10.64% ± 2.57% of dendritic spines were newly formed and 10.82% ± 3.32% had been eliminated 2 weeks after OHT. Comparable rates of formation and elimination were observed in the OHT-ipsi VI region and NC group during this period (OHT-ipsi: formation: 10.43% ± 2.49%, elimination: 11.50% ± 3.74% vs. NC: formation: 11.04% ± 3.31%, elimination: 11.29% ± 4.12%). From 2 to 4 weeks, 4 to 6 weeks, and 6 to 8 weeks post-OHT, the contralateral V1 region exhibited a significantly greater dendritic spine elimination rate than the ipsilateral and control V1 regions (2–4 weeks: NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001; 4–6 weeks: NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001; 6–8 weeks: NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001). Unexpectedly, from 6 to 8 weeks, the rate of dendritic spine formation decreased, with the contralateral V1 region exhibiting a significantly lower rate than that seen in the ipsilateral and control regions (NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001). Collectively, these findings suggested a loss of dendritic spines in the V1 contralateral to the OHT eye (Figure 5C–H).

Figure 5.

Changes in dendritic spines in the V1 region after unilateral OHT.

(A) Schematic diagram of in vivo two-photon imaging of mice. Created with BioRender.com. (B) Imaging timeline. Mice in both groups were observed by TPM at days 0, 14, 28, 42, and 56. On day 0, the OHT group underwent unilateral OHT surgery immediately after two-photon imaging. (C) Representative continuous imaging of dendritic spines in the V1 region of the NC group and the OHT-ipsi and OHT-contra sides after OHT. From 2 to 4 weeks, 4 to 6 weeks, and 6 to 8 weeks post-OHT, the contralateral V1 region exhibited a significantly greater elimination rate, while from 6 to 8 weeks post-OHT, the contralateral V1 region also exhibited a significantly lower formation rate. Scale bar: 5 µm. Red arrows indicate eliminated dendritic spines, and blue arrows indicate newly formed dendritic spines. (D) Schematic diagram of the area selected for data analysis. Highlighted red boxes indicate the chosen dendritic spines. Scale bar: 50 µm. (E–H) Percentages of dendritic spine elimination and formation over 2 weeks in the V1 region of the three groups from 0–2 weeks (E), 2–4 weeks (F), 4–6 weeks (G), and 6–8 weeks (H) (n = 5/group, means ± SD, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). Contra: Contralateral; Ipsi: ipsilateral; NC: no-treatment control; OHT: ocular hypertension; TPM: two-photon imaging; V1: primary visual cortex; w: week.

Enhanced glial reactions and inflammatory responses in the lateral geniculate nucleus, superior colliculus, and V1 regions contralateral to the ocular hypertension eye

Given the crucial involvement of glial cells in synaptic formation and development, we used specific glial cell markers (GFAP and Iba1) to evaluate the total area occupied by reactive astrocytes (Figure 6A–F) and activated microglia (Figure 6G–L). We found that there was a significant increase in the area occupied by reactive astrocytes and microglia in the contralateral V1 region compared with the ipsilateral V1 region and the V1 region in the NC group only after 8 weeks of OHT (NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001; Figure 6F and L).

Figure 6.

Reactive astrocytes/microglia and TNF-α/IL-1β expression levels in the LGN, SC, and V1 regions after unilateral OHT.

(A–C) Representative IHC images of GFAP staining in the LGN (A), SC (B), and V1 (C) regions in the NC group and on the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks after OHT. After 4 weeks of OHT, astrocyte activation was observed in the contralateral LGN and SC regions, while astrocytes were activated in the contralateral V1 at 8 weeks post-OHT. Scale bars: 100 µm. (D–F) GFAP-positive areas in the LGN, SC, and V1 regions following OHT (n = 5/group, means ± SD, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). (G–I) Representative IHC images of Iba1 staining in the LGN (G), SC (H), and V1 (I) regions of the NC group and on the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks after OHT. After 4 weeks of OHT, activation of microglia was observed in the contralateral LGN and SC regions, while microglia were activated in the contralateral V1 at 8 weeks post-OHT. Scale bars: 100 µm. (J–L) Iba1-positive areas in the LGN, SC, and V1 regions following OHT (n = 5/group, means ± SD, *P < 0.05, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). (M–O) TNF-α levels in the LGN (M), SC (N), and V1 (O) regions of the NC group and on the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks (n = 5/group, means ± SD, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). After 4 weeks of OHT, TNF-α was expressed at a higher level in the contralateral LGN and SC regions, while remaining stable in the V1. (P–R) IL-1β expression levels in the LGN (P), SC (Q), and V1 (R) regions of the NC group and on the OHT-ipsi and OHT-contra sides at 4, 6, and 8 weeks (n = 5/group, means ± SD, *P < 0.05, ***P < 0.001, two-way analysis of variance followed by Sidak’s multiple comparisons test). After 4 weeks of OHT, IL-1β was expressed at a higher level in the contralateral LGN and SC regions, while remaining stable in the V1. Contra: Contralateral; GFAP: glial fibrillary acidic protein; Iba1: ionized calcium binding adaptor molecule 1; IL-1β: interleukin-1 beta; TNF-α: tumor necrosis factor-αlpha; IHC: immunochemistry; Ipsi: ipsilateral; LGN: lateral geniculate nucleus; NC: no-treatment control; OHT: ocular hypertension; SC: superior colliculus; V1: primary visual cortex; w: week.

Consistent with neuronal alterations, glial cell responses were enhanced in the SC and LGN regions as early as 4 weeks following OHT. After 4 weeks of sustained IOP elevation, we observed significantly larger GFAP- and Iba1-positive areas in the contralateral LGN and SC regions than in the OHT-ipsi region and NC group (4 weeks-GFAP: NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001; 4 weeks-Iba1: NC vs. OHT-contra: P < 0.0001, OHT-ipsi vs. OHT-contra: P < 0.0001; Figure 6D, E, J, and K). Nevertheless, the percentages of reactive astrocytes and microglia in the contralateral LGN and SC were similar at all time points tested (4, 6, and 8 weeks); no significance test was performed.

To investigate the potential link between synaptic alterations and neuroinflammatory responses mediated by activated glial cells, we analyzed the levels of the inflammatory cytokines IL-1β and TNF-α. Unexpectedly, when elevated IOP was sustained for nearly 8 weeks, there was no significant increase in the expression levels of IL-1β and TNF-α in the V1 region. However, in the subcortical region, at 4 weeks post-OHT surgery, elevated levels of TNF-α and IL-1β were observed in the contralateral LGN and SC regions compared with those on the ipsilateral side and in the control group (4 weeks-TNF-α: NC vs. OHT-contra: P = 0.0085, OHT-ipsi vs. OHT-contra: P = 0.0037, SC: NC vs. OHT-contra: P = 0.0004, OHT-ipsi vs. OHT-contra: P = 0.0002; 4 weeks-IL-1β: NC vs. OHT-contra: P = 0.0102, OHT-ipsi vs. OHT-contra: P = 0.0104, SC: NC vs. OHT-contra: P = 0.0002, OHT-ipsi vs. OHT-contra: P = 0.0007). The expression levels of inflammatory cytokines in the contralateral LGN and SC regions remained stable from 4 to 8 weeks (Figure 6M–R).

Identification of differentially expressed genes and involved pathways in the V1 region after ocular hypertension by RNA sequencing analysis

Next, we performed RNA-seq analysis to identify genes and pathways associated with synaptic changes in the V1 region following OHT. On the basis of the results described above, we designated 8 weeks as the optimal time point for sacrificing the mice. We quantified and compared gene expression levels in brain tissues from the NC group and the OHT-ipsi and OHT-contra sides (Figure 7A). Compared with the NC group, 101 genes were upregulated and 12 genes were downregulated on the OHT-contra side, whereas 19 genes were upregulated and 29 genes were downregulated on the OHT-ipsi side. Furthermore, 208 genes were upregulated and 16 genes were downregulated genes on the OHT-contra side compared with the OHT-ipsi side (Figure 7B). Volcano plots were generated to highlight DEGs among the three groups (Figure 7C–E). To detect genes whose expression was significantly altered in V1 because of OHT, we focused primarily on changes on the contralateral side. Differential gene expression analysis identified five significantly upregulated genes (Mst1, Tmem30c, Pde6g, Zfp82, and Fpr2) and five significantly downregulated genes (Gprc5d, Otx2, Lhx8, Lpar5, and Krt8); the variability in the expression of these genes was plotted on a heatmap (Figure 7F). KEGG enrichment analysis indicated that the DEGs between the NC group and OHT-contra side mainly corresponded to pathways involving phototransduction, ether lipid metabolism, and neuroactive ligand‒receptor interactions. In contrast, the significantly enriched pathways for DEGs between the OHT-ipsi and OHT-contra sides included those involved in protein digestion and absorption, complement and coagulation cascades, and glutathione metabolism. Furthermore, the DEGs between the NC group and OHT-ipsi side were predominantly associated with the IL-17 and TGF-beta signaling pathways (Figure 7G–I).

Figure 7.

Differential gene expression and related pathways in the V1 region after unilateral OHT.

(A) PCA analysis of the three groups (n = 3/group). (B) Number of total differentially expressed genes (DEGs) and up/down-regulated DEGs among the three groups. (C) Volcano plot of DEGs between the OHT-ipsi and OHT-contra sides. (D) Volcano plot of DEGs between the NC group and the OHT-contra side. (E) Volcano plot of DEGs between the NC group and the OHT-ipsi side. (F) Heatmap of characteristic DEGs between the NC group and OHT-contra side. (G) KEGG enrichment analysis of DEGs between the NC group and OHT-ipsi side. (H) KEGG enrichment analysis of DEGs between the NC group and OHT-contra side. (I) KEGG enrichment analysis of DEGs between the OHT-ipsi and OHT-contra sides. Contra: Contralateral; DEG: differentially expressed genes; Ipsi: ipsilateral; KEGG: Kyoto encyclopedia of genes and genomes; NC: no-treatment control; OHT: ocular hypertension.

Discussion

Previous animal studies and clinical trials have confirmed the existence of changes in both structure and function throughout the visual pathway, from the retina to the visual cortex, in glaucoma. Investigations of injury mechanisms and repair strategies in glaucoma should include not only the retina but also the CNS (Sapienza et al., 2016; Tribble et al., 2021; You et al., 2021; Yan et al., 2022; Yang et al., 2024). Given the critical role of synaptic plasticity in maintaining CNS function, our study used an OHT model to investigate neuronal and synaptic changes in the brain during glaucoma development.

In the CNS, the process whereby primary neuronal injury impacts distal neurons is referred to as transsynaptic or transneuronal degeneration (Lawlor et al., 2018; Fu et al., 2021; You et al., 2021). This theory suggests that glaucoma-induced RGC apoptosis and optic nerve atrophy can subsequently lead to the degradation and atrophy of distal neurons within the SC, LGN, and even the V1 regions. After as few as 4 weeks of elevated IOP, a reduction in neuronal soma size was observed in the SC and LGN regions contralateral to the OHT eye, akin to observations of neuronal atrophy in the LGN region in previous studies (Weber et al., 2000; Van Hook et al., 2020). The lack of observable changes in neuronal density may be attributed to the fact that the neurons were in the early stages of injury and exhibited no notable signs of apoptosis or necrosis. Notably, this alteration in soma size was specific to the contralateral brain region, as rodents primarily exhibit contralateral projection of RGC axons, with only 5%–10% of optic nerves projecting ipsilaterally (Liu et al., 2011). The subcortical neuron atrophy stabilized by 4 weeks, as we found no significant variation in neuronal soma size across the three time points tested. Moreover, no significant changes were noted in either neuron number or neuronal soma size in the V1 region, consistent with observations from a macaque glaucoma model (Yan et al., 2022). However, recent research has revealed a reduction in neuron density within the V1 region in DBA/2J mice. This apparent discrepancy may be ascribed to differences in the glaucoma models used. Given the limitations of our study, in which we were able to maintain elevated IOP for only 8 weeks following a single magnetic bead injection, after which we observed a decrease in IOP, a more sustainable and stable OHT model is needed to investigate neuronal alterations in V1.

Synapses are the fundamental structure forming neural connections, facilitating rapid and precise information transmission between neurons (Dai et al., 2024). Synaptic plasticity in the CNS relies on synapse formation and elimination, and synaptic loss and damage are main characteristics of most CNS disorders. Synapses can be classified as excitatory or inhibitory, depending on the neurotransmitters that they secrete. Presynaptic and postsynaptic components aggregate into clusters, which play an important role in regulating synaptic transmission and plasticity (Heine and Holcman, 2020). VGluts facilitate glutamate transport, while VGATs are responsible for GABA transport; VGlut1 is predominant in the cortical region, and VGlut2 is prevalent in subcortical regions (Pietrancosta et al., 2020; Bolneo et al., 2022). PSD95 acts as an anchor for GluA and GluN receptors, while gephyrin serves as an anchor for GABA_A receptors, making them primary markers for excitatory and inhibitory synapses (Pinto et al., 2013; Groeneweg et al., 2018). Recent studies have demonstrated that glaucoma can lead to disruptions in both glutamatergic and GABAergic synapses in the retina (Berry et al., 2015; Williams et al., 2016; Gramlich et al., 2021). Along the visual pathway to the LGN, scholars observed a reduction in the density of VGlut2 puncta and altered synaptic plasticity in the LGN of DBA/2J mice (Van Hook et al., 2020; Van Hook, 2022). In our study, we found that 8 weeks of OHT resulted in excitatory and inhibitory synapse loss. Notably, we observed a decrease in the number of gephyrin-positive puncta, without a statistically significant alteration in PSD95 levels. This variation could potentially impact the excitation/inhibition ratio in V1, which is crucial for normal function. In the future, we aim to further explore changes in the excitation/inhibition balance within V1 using patch clamp electrophysiology and other techniques. The size of the synaptic terminals did not appear to change significantly, possibly due to constraints related to the imaging technology. Electron microscopy may offer more detailed insight into ultrastructural alterations. Because we did not see a significant reduction in the number of PSD95-positive puncta, we further used TPM to study changes in dendritic spines. Dendritic spine structure and number can affect synaptic function, and dendritic spine morphology can be regulated by synaptic activity. After a critical developmental period, dendritic spine elimination and formation rates are generally balanced, but under certain physiological or pathological conditions the formation/elimination rates of dendritic spines are altered, and the spine structure undergoes reorganization (Rochefort and Konnerth, 2012; Chidambaram et al., 2019). Research has shown that monocular deprivation can increase the rate of dendritic spine formation in the visual cortex, thereby affecting synaptic plasticity (Hofer et al., 2009). However, few studies have explored changes of dendritic spines in the visual cortex induced by glaucoma. Consistent with our earlier study using an ONC model, we also observed a decrease in the number of dendritic spines on pyramidal neurons within layers 2/3 of the V1 region after OHT (Zhan et al., 2020). However, most excitatory synapses are located on dendritic spines, and there are several possible explanations for our failure to observe a decrease in the number of PSD95-positive puncta by immunofluorescence. On the one hand, we observed a decreasing trend in PSD95 expression via immunofluorescence, and TPM is a much higher resolution imaging technique that can be used to continuously image individual neurons. On the other hand, due to the limitations of TPM, we mainly observed neurons in layers 2/3, while PSD95 puncta were observed throughout layers 2 to 6. Finally, further analysis is needed to assess whether the missing dendritic spines contain smaller or more weakly fluorescent PSD95 puncta, making them difficult to detect by immunofluorescence.

The mechanisms underlying neuronal degeneration and synapse/dendritic spine loss are diverse. The predominant cause of this phenomenon is activation of glial cells, which have been demonstrated to play a role in numerous neurodegenerative diseases (Kwon and Koh, 2020; Kölliker-Frers et al., 2021; González et al., 2023; Liu et al., 2023). Consistent with previous studies (Lam et al., 2009; Fujishiro et al., 2020), our study revealed an increase in the proportion of activated astrocytes and microglia in the brain contralateral to the OHT eye. Overactivation of glial cells triggers the release of proinflammatory cytokines, thereby intensifying neuroinflammatory responses and ultimately resulting in neuronal injury (Ramesh et al., 2013). Additionally, glial cells have the capacity to regulate the generation or clearance of synapses and dendritic spines through the release of synapse-promoting molecules or phagocytic receptors. In various neurodegenerative conditions, glial cell dysfunction often coincides with synaptic dysfunction and excessive pruning of dendritic spines, which may be linked to glial cell–mediated complement-dependent pathways and inflammatory cascades (Stevens et al., 2007; Wake et al., 2013; Huang et al., 2018; Dvorzhak and Grantyn, 2020; Dejanovic et al., 2022; Xie et al., 2024). To clarify the specific molecules and pathways involved in glial cell–mediated degradation of neurons, synapses, and dendritic spines, further investigation is necessary. Moreover, a recent study indicates that astrocytes and microglia can exhibit anti-inflammatory or proinflammatory phenotypes (Liddelow et al., 2017). Further studies are needed to determine the predominant phenotype of these activated glial cells and investigate how the proportions of different phenotypes change at varying intervals post-OHT. Interestingly, in some rodent OHT models, researchers have observed glial cell activation in the bilateral SC and LGN regions, in contrast to our findings (Sapienza et al., 2016; Tribble et al., 2021). Through meticulous morphological examination of individual microglia, Tribble et al. (2021) confirmed activation of microglia in the bilateral visual pathways. Future research may require unconventional imaging or analytical methods to examine glial cells in the visual pathways on both sides of the brain.

Differential gene analysis revealed minimal alterations in gene expression within the ipsilateral V1 region compared with the NC group. This finding suggests that, despite the fact that only approximately 5% of optic nerves project into the ipsilateral brain region, changes do occur on the ipsilateral side (Tribble et al., 2021). However, the level of injury was relatively low and necessitated a prolonged duration of OHT. Differential gene analysis identified two significant genes linked to neurons and synapses: Mst1, which was upregulated, and Otx2, which was downregulated. Previous research has identified Mst1 as a key player in the synaptic loss and cognitive decline characteristic of Alzheimer’s disease (Wang et al., 2022). Through p53 activation, Mst1 can trigger neuronal apoptosis and elimination of synapses. Otx2 encodes a homologous protein transcription factor, and most neurons in the visual cortex that express Otx2 are PV interneurons (Sugiyama et al., 2008). PV interneurons are pivotal in synaptic plasticity, with one study indicating that Otx2 accumulation fosters maturation of PV neurons and the perineuronal net, consequently stabilizing synaptic structures and impeding plasticity (Dong et al., 2023). Our RNA-seq findings indicated Otx2 upregulation in the contralateral V1 region, implying a potential self-regulatory process in V1 after synaptic and dendritic spine loss. The changes in synaptic plasticity that occur during this process should be verified with additional electrophysiological data. Additionally, Otx2 has been noted to exert a specific protective effect on retinal neurons (Moya and Ibad, 2021). Currently, there is no evidence in the literature indicating an association between these genes and glaucoma-related alterations in V1. Furthermore, the neuroactive ligand‒receptor interaction signaling pathway identified via KECG analysis plays a critical role in modulating transcription factors and gene expression to regulate neuronal function, and this pathway has been shown to be associated with Alzheimer’s disease. Future studies should investigate these two genes and this pathway.

Our study had some limitations. First, we focused mainly on structural changes, and did not fully investigate neuronal function. In future investigations, employing functional magnetic resonance imaging of whole animal brains could facilitate a deeper examination of functional alterations, alongside tracking changes in electrical activity via electrophysiological techniques. Additionally, numerous recent studies have posited that glaucoma shares neurodegenerative characteristics with Alzheimer’s disease and Parkinson’s disease (Lawlor et al., 2018; Artero-Castro et al., 2020; Prokosch et al., 2023). It is hypothesized that alterations in the brain may precede RGC loss or damage to the optic nerve (Lawlor et al., 2018; Sharma et al., 2022). However, we could not confirm this hypothesis due to limitations related to our experimental design. Future longitudinal studies using neurodegenerative animal models are necessary to monitor the spatiotemporal progression of lesions within the visual pathway.

In conclusion, previous research has documented numerous glaucoma-induced changes in the brain, particularly in the SC and LGN, at various time points. Our study highlights alterations in neurons, synapses, and dendritic spines in the V1 region. Consistent with previous findings, our study revealed reduced neuronal soma size and glial cell activation in the contralateral SC and LGN. Specifically, we noted a loss of synapses and dendritic spines, along with glial cell activation, in the contralateral V1 region. These results offer fresh perspectives on the neuropathology of glaucoma in the brain. Further investigations are necessary to elucidate the underlying mechanisms of retinal and brain impairments in glaucoma.

Additional file:

Additional Figure 1 ^{(573.7KB, tif)} : IOP curves of both eyes in the OHT and NC groups.

Additional Figure 1

IOP curves for both eyes in the OHT and NC groups.

The IOP of the OHT-left eyes was significantly higher than that of the OHT-right eyes and NC eyes in the first 7 weeks. In the eighth week after OHT, there was no significant difference in IOP between the OHT-left eye and the remaining three groups (n = 10/group, mean ± SD, ^***P < 0.001, OHT-left vs. OHT-right/NC-left/NC-right group, two-way analysis of variance followed by Sidak’s multiple comparisons test). IOP: Intraocular pressure; NC: no-treatment control; ns: not significant; OHT: ocular hypertension.

Funding Statement

Funding: This study was supported by the National Natural Science Foundation of China, No. 82271115 (to MY).

Data availability statement:

All data relevant to the study are included in the article or uploaded as Additional file.