Lamp2 Deficiency Enhances Susceptibility to Oxidative Stress–Induced RPE Degeneration

Abstract

Purpose: Autophagy and lysosomal degradation are vital processes that protect cells from oxidative stress. This study investigated the role of lysosome-associated membrane protein 2 (Lamp2), a lysosomal protein essential for autophagosome maturation and lysosome biogenesis, in maintaining retinal health under oxidative stress.

Methods: To induce oxidative stress, young Lamp2 knockout (KO) and wild-type mice received an intravenous injection of a low dose (10 mg/kg) of sodium iodate (NaIO₃). We examined retinal histopathology and morphological changes in the RPE. The involvement of resident microglia or infiltrating macrophages was assessed using immunostaining, flow cytometry, and real-time PCR for chemokines and cytokines.

Results: After administering a low-dose NaIO₃, Lamp2 KO mice showed significant RPE degeneration, whereas wild-type mice had minimal damage. Histological analysis and electron microscopy revealed significant thinning of the outer nuclear layer and loss of RPE epithelial polarity in Lamp2 KO mice. Additionally, there was a significant increase in ionized calcium-binding adaptor molecule 1–positive microglia and macrophages in the outer retina. Early proliferation of CD45^lowMHC-II^low resident microglia was followed by the infiltration of CD45^highLy6C^high monocytes and the engraftment of CD11b⁺CD45^high monocyte-derived macrophages. Transcript levels of monocyte chemoattractant protein 1, macrophage inflammatory protein 1β, Il- 1β, and Il-6 also increased in the retinas of Lamp2 KO mice. Furthermore, pretreatment with the macrophage-depleting agent clodronate prevented NaIO₃-induced RPE degeneration and macrophage infiltration in Lamp2 KO mice.

Conclusions: Lamp2 deficiency, when combined with oxidative stress, leads to RPE degeneration in vivo. Lysosomal dysfunction also promotes macrophage engraftment and triggers neurotoxic inflammation.

Lysosomal degradation is crucial in the process of autophagy, where lysosomes fuse with double-membrane structures that engulf the cell's own components for degradation,^1,2 as well as in the phagocytosis necessary for degrading the distal tips of photoreceptor outer segments that are daily shed into the subretinal space.^3,4 Dysfunctions of lysosome machinery may be involved in the pathology of various retinal disorders, such as retinal detachment, retinitis pigmentosa, and AMD.^4,5 Lysosome-associated membrane protein 2 (Lamp2) is a highly glycosylated protein in lysosomal membranes that plays a critical role in lysosomal biogenesis and autophagosome and phagosome maturation.⁶ Lamp2 deficiency in humans leads to Danon disease, a rare lysosomal storage disorder characterized by cardiomyopathy and skeletal myopathy, intellectual disability, and retinopathy.^7–9 Despite the critical role of Lamp2 in autophagy and phagocytosis, Lamp2 knockout (KO) mice are fertile and have an almost normal lifespan, being accompanied by the accumulation of numerous autophagic vacuoles in several tissues.¹⁰ Because Lamp-1/2 double KO mice showed embryonic lethality, Lamp1 was considered to compensate for lysosomal functions in Lamp2-deficient cells.¹¹ Therefore, Lamp2 KO mice may serve as a valuable tool for studying the impact of defective lysosomal functions in vivo. 

We have previously shown that Lamp2 deficiency in mice led to late-onset retinal degeneration.¹² Although the retina of young Lamp2 KO mice seemed to be subnormal, aged Lamp2 KO mice exhibited a delayed digestion of photoreceptor outer segments and a dysregulated autophagic flux in the RPE. Moreover, Lamp2 KO mice older than 6 months of age exhibited basal laminar deposit formation, an important histopathological hallmark of retinal aging and early AMD. Additionally, Lamp2 was highly expressed in human RPE, although its expression significantly decreased in the eyes of patients with AMD compared with healthy subjects.¹² 

AMD is a multifactorial disease that leads to irreversible vision loss among individuals over 55 years old in developed countries,^13–16 in which oxidative stress is known to be an important risk factor.¹⁷ Oxidized biomolecules and damaged organelles, such as mitochondria, are typically removed by autophagic processes, which contributes to maintaining retinal homeostasis during aging.^18–21 Indeed, in vitro studies have demonstrated that decreasing autophagy increases the susceptibility of RPE cells to oxidative stress.^21,22 However, there is limited evidence about in the vivo retinal phenotypes associated with oxidative stress combined with lysosomal dysfunctions. Although Lamp2 KO mice older than 6 months showed retinal degeneration, their retinal structure was preserved at a younger age.¹² Hence, we aimed to investigate the role of Lamp2 in a model of oxidative stress–induced RPE degeneration. 

Sodium iodate (NaIO₃) is known for its selective toxicity to RPE cells and is widely used in pathological studies related to oxidative stress in the RPE.^23–27 Because NaIO₃ induces dose- and time-dependent RPE degeneration in vivo, previous studies have often used high doses (>40 mg/kg).^28,29 In this study, we tested the effects of a low dose of 10 mg/kg NaIO₃, which reportedly did not induce RPE degeneration in wild-type (WT) mice.³⁰ Thus, this study aimed to determine whether Lamp2 deficiency affects tolerance to retinal degeneration under oxidative stress conditions. 

Material and Methods

Animals

All experiments were approved by the Animal Care Committee of the Institute of Medical Science, Kyushu University, and conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Lamp2 KO mice, originally generated in Prof. Saftig's lab (Kiel University), were backcrossed with C57BL6J mice for at least five generations and maintained as heterozygous colonies in our laboratory. Male Lamp2 y/− mice and their male WT littermates (Lamp2 y/+) were used at 3 months old. The mice were placed in a 12-h light/dark environment (8:00 am to 8:00 pm) with ad libitum access to normal chow and water in the animal facility of Kyushu University. 

NaIO₃-Induced Retinal Degeneration

NaIO₃ (S4007, Sigma-Aldrich, St. Louis, MO, USA) was dissolved in PBS. Mice were injected with 10 mg/kg body weight of NaIO₃ or the same volume of PBS as control via tail veins. For macrophage depletion, 5 µL/g body weight of clodronate liposomes (16001003; Hygieia Bioscience, Osaka, Japan) or the same volume of PBS as a control was injected intraperitoneally 24 h before NaIO₃ administration. 

Immunofluorescence

The mice were euthanized, and their eyes were enucleated. The cornea, lens, and vitreous were removed. After peeling the retina off from the RPE/choroid, radial cuts were made to create flat-mounted specimens. Retina and RPE/choroid specimens were fixed with 4% paraformaldehyde for 1 h and washed with PBS. For immunohistochemistry, the retina or RPE/choroid flat-mounted specimens were blocked with 1% BSA in 0.5% Triton X-100 in PBS for 1 h at room temperature and incubated with rabbit anti-ionized calcium-binding adaptor molecule 1 (Iba-1) antibody (1:1000; 019-19741, Wako, Osaka, Japan) or rat monoclonal zonula occludens-1 (ZO-1) antibody (1:200; sc-33725; Santa Cruz, Biotechnology, CA, USA) at 4 °C overnight and with secondary antibodies conjugated to Alexa Fluor dyes (1:200; Alexa Fluor 488 goat anti-rabbit IgG and Alexa Fluor 647 goat anti-rat IgG, Invitrogen, Carlsbad, CA, USA) at room temperature for 1 h. After washing with PBS, the flat-mounted specimens were mounted on an aqueous mounting medium (PermaFluor, EPREDIA) with a cover glass. Fluorescent images were obtained using a laser confocal microscope (LSM 700; Carl Zeiss, Oberkochen, Germany) or a fluorescent microscope (BZ-X710; Keyence, Osaka, Japan). 

Quantification of Microglia and Macrophages and RPE Degeneration

For quantification of Iba-1–positive cells, Z-stack images under 40 × magnification with 1-µm intervals were taken at the regions 250 and 750 µm from the optic disc (four images for each) in the retina or RPE/choroid flat-mount (n = 8 eyes). The retinal layers were divided into the ganglion cell layer, inner plexiform layer, outer plexiform layer, outer nuclear layer (ONL), photoreceptor layer, and subretinal space. The number of cells was quantified using FIJI software (National Institutes of Health, Bethesda, MD, USA) and normalized per square millimeter area. RPE degeneration is characterized by either the loss of the normal hexagonal morphology of RPE cells observed in flatmount specimens^30–32 or the thinning or complete absence of the RPE layer in retinal sections.^32,33 Degenerated RPE areas were quantified using ZO-1 staining on whole-mount RPE/choroid preparations. The freehand line tool in ImageJ 2.0 software was used to delineate both the total RPE area and the degenerated regions³⁴ (n = 8 eyes). To analyze the relationship between the number of Iba-1⁺ cells and RPE morphology, the circularity index (CI) of RPE cells in five randomly selected images was evaluated using the MorphoLibJ plugin in FIJI combined with Cellpose software³⁵ with the formula: CI = 4π × Area / Perimeter² (a normal RPE cell has a CI of approximately 0.9).^36,37 

Histology

For hematoxylin and eosin staining, the enucleated mice eyes were fixed with a fixative (SUPER FIX, Kurabo, Osaka, Japan) for 24 h at 4 °C and paraffin-embedded. Retinal sections crossing the optic nerve were also prepared. Images of eight areas of the stained sections were taken every 250 µm up to 2000 µm from the optic nerve head on the nasal and temporal sides (n = 8 eyes). Images were captured using a fluorescence microscope (BZ-X710; Keyence). 

Transmission Electron Microscopy

The posterior segments of mice eyes were fixed with 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M cacodylate buffer with 0.08 M CaCl₂ at 4 °C overnight, post-fixed for 1.5 h in 2% aqueous OsO₄, dehydrated in ethanol and water, and embedded in epoxy resin. Ultrathin sections were cut from the resin blocks, mounted on copper grids, and observed under a transmission electron microscope (H-7770; Hitachi, Tokyo, Japan). 

Real-Time PCR

The retinas and RPE/choroid were quickly isolated from mouse eyecups and frozen in liquid nitrogen. RNA was extracted using an RNA Isolation kit (740955, Nucleospin RNA, Takara, Shiga, Japan), and the RNA concentration was determined spectrophotometrically at 260 nm. First-strand cDNA was synthesized using a PrimeScript RT Reagent Kit (Perfect Real Time; Takara). Gene expression was quantified using real-time PCR using a TaqMan probe (Thermo Fisher Scientific, Waltham, MA, USA) and a LightCycler (Roche, Indianapolis, IN, USA). The relative expression of each gene was normalized to an internal housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (Gapdh) and calculated as ΔΔCt. The catalog numbers for Taqman probes are as follows: Mm00520345_m1 for rhodopsin (Rho), Mm00504133_m1for RPE 65 (Rpe65), Mm04411498_m1 for glutathione peroxidase 4 (Gpx4), Mm00477784_m1 for nuclear factor erythroid 2-related factor 2 (Nrf2), Mm01344233_g1 for superoxide dismutase 1 (Sod1), Mm00441242_m1 for monocyte chemoattractant protein 1 (Mcp-1), Mm00443111_m1 for macrophage inflammatory protein-1 beta (Mip-1β), Mm00434228_m1 for Il-1β, and Mm01211445_m1 for Il-6, and Mm99999915_g1 for Gapdh. 

Flow Cytometry

Mice retinas were digested in 1 mL of RPMI-1640 medium (R8758, Sigma-Aldrich) with 1.2 mg/mL of Collagenase D (11088858001, Roche) and 40 µg/mL DNase I (10104159001, Sigma-Aldrich). The tissue was digested at 37 °C for 30 min and filtered with fluorescence-activated cell sorting buffer (PBS with 2% fetal bovine serum) through a 40-µm cell strainer, blocked with 10 µL/sample of Fc blocker (14-0161-86, eBioscience) and incubated for 5 min on ice, and incubated 20 min on ice with anti-mouse antibodies CD11b (FITC-conjugated, clone; BM8, Biolegend, San Diego, CA, USA), CD45 (APC-conjugated, clone 30-F11, Biolegend), CX3C motif chemokine receptor 1 (CX3CR1; PE-conjugated, clone M1/70, Biolegend), lymphocyte antigen 6 family member C (Ly6C; APC-cy7-conjugated, clone 1A8, Biolegend), I-A/I-E (BV421-conjugated, clone M5/114.15.2, Biolegend). We used 7-AAD (BD Pharmingen, BD Biosciences, Franklin Lakes, NJ, USA) to exclude dead cells. Fluorescence minus one control was used to define the positively stained cells. For each flow cytometry panel, a fluorescence minus one control was prepared by omitting one of the fluorescent markers while keeping all other markers and reagents at the same concentrations (Supplementary Fig. S1). The method for defining microglia/macrophages was modified from the protocol described by O'Koren et al.³⁸ Data were analyzed using the BD FACSVerse system (BD Biosciences) and FlowJo software. 6 retinas from each group were collected at each time point, and the numbers of microglia/macrophages in 500, 000 live cells were calculated. 

Statistical Analysis

Statistical analyses were conducted using GraphPad Prism v10.0 (GraphPad Software, La Jolla, CA, USA) and SPSS Statistics v27 (IBM, Armonk, NY, USA). Statistical differences between the two groups were analyzed using the Student t test. For multiple factors, a one-way ANOVA followed by the post hoc Tukey honest significant difference test was used. N represents the number of eyes, with both eyes from each animal being used for the analysis. Results are presented as mean ± SD, and statistical significance was set as a P value of less than 0.05. The relationship between the number of Iba-1–positive cells and the CI of the RPE was evaluated using linear regression analysis. 

Results

Low-Dose NaIO₃ Induced the Expression of Antioxidant Genes in Lamp2 KO Mice

To investigate the impact of Lamp2 deficiency under oxidative stress conditions, we assessed the expression of antioxidant genes in the RPE of 3-month-old Lamp2 KO mice and WT littermate controls after low-dose NaIO₃ administration. Relatively young, 3-month-old Lamp2 KO mice were used in this study because they exhibit limited retinal degeneration at this age.¹² To ensure the isolation of uncontaminated RNA from each tissue, the relative enrichment of RPE-specific gene Rpe65 and photoreceptor-specific gene Rho was analyzed in the retina and RPE. The RNA isolated from the RPE showed a more than 85-fold higher expression of Rpe65 compared with that from the retina. Similarly, Rho was found to be enriched by more than 10-fold in retinal RNA isolates compared with RPE RNA isolates (Supplementary Fig. S2). 

Next, quantitative PCR analysis revealed significant increases in the transcript levels of three key antioxidant enzymes: Gpx4, Sod1, and Nrf2 (Fig. 1). In the retina of Lamp2 KO mice, exposure to 10 mg/kg NaIO₃ led to the upregulation of Gpx4 and Nrf2 on day 1 and Sod1 on day 3, whereas no changes were observed in the WT controls (Fig. 1A). In the RPE of Lamp2 KO mice, Nrf2 and Sod1 were upregulated on days 1, 3, and 7, and Gpx4 expression increased on day 1, with no corresponding changes in the WT mice (Fig. 1B). Overall, these results indicate a significant upregulation of antioxidant genes, such as Gpx4, Sod1, and Nrf2, in both the retina and RPE of Lamp2 KO mice after low-dose NaIO₃ exposure. 

Figure 1.

 

Transcriptional changes in antioxidant genes in the retina and RPE from Lamp2 KO and WT mice following low-dose NaIO₃ injections. Quantitative PCR analysis of Gpx4, Nrf2, and Sod1 in the retina (A) and RPE (B) of WT and Lamp2 KO mice with or without NaIO₃ treatment. Data are represented as mean ± SD; n = 6 eyes per time point. One-way ANOVA followed by Tukey's post hoc test. *P < 0.05, **P < 0.01, ***P < 0.001.

Lamp2 KO Mice Exhibited Retinal Degeneration in Response to Systemic Administration of Low-Dose NaIO₃

Next, we investigated whether RPE degeneration was induced by low-dose NaIO₃ in Lamp2 KO mice. To assess the extent of degeneration, we performed ZO-1 immunostaining on RPE flat-mounts from 3-month-old Lamp2 KO and WT mice after the administration of 10 mg/kg NaIO₃. In the absence of NaIO₃, both WT and Lamp2 KO mice exhibited normal RPE morphology (Figs. 2A, 2E). After NaIO₃ administration, WT mice retained normal RPE morphology (Figs. 2F, 2G, 2H) whereas Lamp2 KO mice exhibited disrupted RPE morphology on day 3, which further worsened on day 7 (Figs. 2B, 2C, 2D). On day 7, the mean area of RPE degeneration in Lamp2 KO mice was quantified as 34% of the total fundus area (Fig. 2I). 

Figure 2.

 

Effect of low-dose NaIO₃ on RPE morphology in WT and Lamp2 KO mice. (A–H) Representative images of the RPE/choroid flat-mount labeled with ZO-1 (red) and the RPE degeneration area over time after injection of NaIO₃, n > 4 for each group. Edges of the degeneration area are highlighted by yellow arrows. Scale bars, 200 µm. (A–H) Higher magnification images of RPE degeneration are shown in the lower panels. Scale bars, 50 µm. (A, E) RPE/choroid flat-mount from WT and Lamp2 KO mice with PBS control showed no apparent morphological changes in the RPE. Lamp2 KO mice with systemic injection of 10 mg/kg NaIO₃ showed RPE degeneration from day 3 (C) and aggravated on day 7 (D), whereas WT mice with the same injection did not show any RPE degeneration (F–H). (I) Quantification of the RPE degeneration area indicated that 34% of the RPE was degenerated in Lamp2 KO mice. Proportions of RPE degeneration area are expressed as mean ± SD (n = 4–8 eyes per group). Student t tests. *P < 0.05, ** P < 0.01, ***P < 0.001.

Transmission electron microscopy confirmed ultrastructural changes in the RPE of NaIO₃-treated Lamp2 KO mice. In Lamp2 KO mice, RPE cells lost their normal basal infoldings and apical microvilli, and the cytoplasm showed numerous vacuolizations and stratification. Additionally, significant mitochondrial cristae disruption was observed in the RPE of NaIO₃-treated Lamp2 KO mice (Fig. 3A). In contrast, WT mice displayed normal RPE ultrastructure under the same treatment. In addition, we observed photoreceptor degeneration in Lamp2 KO mice 7 days after NaIO₃ administration. Hematoxylin and eosin–stained sections revealed significant thinning of the ONL (Fig. 3B). In contrast, no changes in ONL thickness were observed in 10 mg/kg NaIO₃-treated WT mice (Fig. 3C). 

Figure 3.

 

Ultrastructural changes and histopathology of the retinas from Lamp2 KO and WT mice treated with low-dose NaIO₃. (A) Representative transmission electron microscopy images of the RPE from Lamp2 KO and WT mice following 7 days of NaIO₃ injection. The RPE cells from Lamp2 KO mice did not have normal structures of basal infoldings and apical microvilli and showed numerous vacuolizations in the cytoplasm. With a loss of epithelial polarity, a stratification was observed; see the nuclei of overlying RPE (asterisk). Disruption of mitochondrial cristae structures was observed in NaIO₃-treated Lamp2 KO mice (arrows), whereas those of WT mice were preserved (arrowheads). Scale bars in the left panel are 5 µm and the right panel are 2 µm. (B) hematoxylin and eosin staining of cross-sections in Lamp2 KO mice showed a slight decrease in ONL thickness compared with WT under PBS treatment. After 7 days of NaIO₃ injection, the RPE and photoreceptors were severely damaged in Lamp2 KO mice, while no changes were observed in WT. Scale bars, 50 µm. (C) Quantifications for the ONL thickness on day 7 after NaIO₃ injections in Lamp2 KO and WT mice. Data are shown as mean ± SD, n = 4 eyes. Student t tests. **P < 0.01. GCL, ganglion cell layer; INL, inner nuclear layer; IPL, inner plexiform layer; OPL, outer plexiform layer.

Increased Numbers of Iba-1–Positive Cells in the Retina in Lamp2 KO Mice After NaIO₃ Administration

Given the association between macrophages and RPE degeneration,³² we investigated whether macrophage infiltration occurred in the retina of Lamp2 KO mice after low-dose NaIO₃ injection. Even without NaIO₃ administration, the retina of Lamp2 KO mice exhibited a higher number of Iba-1⁺ cells compared with WT mice (Figs. 4A, 4B). After NaIO₃ injection, the number of Iba-1⁺ cells significantly increased in Lamp2 KO retinas compared with WT mice on day 7. A significant fraction of these ameboid Iba-1⁺ cells migrated to the ONL and subretinal space in Lamp2 KO mice (Fig. 4D), contrasting their normal locations in the inner retinal layers in WT mice (Fig. 4C). Quantification of Iba-1⁺ cells revealed significant increases across all retinal layers in Lamp2 KO mice compared with WT (Figs. 4E–I). Furthermore, we observed that the infiltration of Iba-1⁺ cells, which spread from the optic nerve, consistently correlated with the areas of RPE degeneration (Fig. 4J). There was a significant correlation between the number of Iba-1⁺ cells and the CI of RPE in Lamp2 KO mice (Fig. 4K). 

Figure 4.

 

Infiltration of Iba-1–positive cells and associated RPE injury after NaIO₃ injections. (A–D) Representative images of Iba-1 (green)-positive cells and RPE morphology stained by ZO-1 (red) in the retinal and RPE/choroid flat-mounts from 3-month-old Lamp2 KO and WT mice treated with NaIO₃ or PBS on day 7. RPE/choroid flat-mounts were also stained by ZO-1 (red). (E–H) Quantifications of Iba-1⁺ cells in different retinal layers. (I) The total number of Iba-1⁺ cells in all retinal layers. Mean ± SD, n = 4 for the PBS group and 8 for the NaIO₃ group. One-way ANOVA followed by Tukey's post hoc test, ** P < 0.01, ***P < 0.001, ****P < 0.0001. GCL, ganglion cell layer; IPL, inner plexiform layer; OPL, outer plexiform layer, PRL, photoreceptor segment layer; SRS, subretinal space. (J) Representative images showing areas of variable RPE degeneration and associated Iba-1⁺ cells. The outlines of individual RPE cells were sketched using FIJI (lower panels). (K) Correlation between CI of the RPE and the number of Iba-1⁺ cells was evaluated using linear regression analysis. Scale bars, 50 µm.

Early Response of Resident Microglia and Late Infiltration of Monocyte-derived Macrophage in the Retina of NaIO₃-Treated Lamp2 KO Mice

To better understand the source of Iba-1⁺ cells, we performed flow cytometry using retinal specimens from Lamp2 KO and WT mice treated with NaIO₃. We found that the number of CD11b⁺CD45^low cells, rapidly increased in the retina within 24 hours after treatment. To further identify subpopulations within these cells, we gated them based on Ly6C, CX3CR1, and major histocompatibility complex (MHC) II expression. Our analysis revealed that most CD11b⁺CD45^low cells were Ly6C⁻CX3CR1⁺MHC-II^low resident microglia, which increased drastically within 24 h and continued to rise until day 7 (Fig. 5). In the CD11b⁺CD45^high population, CX3CR1⁺Ly6C^high monocytes showed a significant increase on days 3 and 7 in Lamp2 KO mice retinas compared with baseline. Additionally, CX3CR1⁺Ly6C^low monocyte-derived macrophages increased significantly on days 3 and 7 in Lamp2 KO mice retinas. In contrast, there were no detectable increases in these populations in NaIO₃-treated WT mice retinas (Fig. 6). 

Figure 5.

 

Flow cytometry analysis of the microglial populations in WT and Lamp2 KO mice after NaIO₃ administration. (A) CD45^low populations from retinas of WT or Lamp2 KO mice were gated by Ly6C and CX3CR1 expression; MHC-II was used to distinguish microglia from resident macrophages. (B) Among the Ly6C⁻ CX3CR1⁺ population, the number of MHC-II^low microglia in the retina increased on day 1 or later after NaIO₃ injection. Mean ± SD, n = 6 eyes. One-way ANOVA followed by Tukey's post hoc test. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6.

 

Flow cytometry analysis for the monocyte-derived macrophage populations in WT and Lamp2 KO mice after NaIO₃ administration. (A) CD45^high populations from retinas of WT or Lamp2 KO mice were gated by Ly6C and CX3CR1 expression; Ly6C^high population was considered to be monocytes while Ly6C^low population to be monocyte-derived macrophages. (B) Ly6C^high monocytes and Ly6C^low macrophages increased on day 3 or later. Data are shown as mean ± SD, n = 6 eyes. One-way ANOVA followed by Tukey's post hoc test. *P < 0.05, **P < 0.01.

NaIO₃ Increased the Retinal Transcription of Chemokines and Cytokines in Lamp2 KO Mice

To further investigate the inflammatory response in Lamp2 KO and WT mice, we performed real-time PCR to analyze the expression of chemokines/cytokines following NaIO₃ injection. In Lamp2 KO mice, the mRNA levels of chemokines Mcp-1 and Mip-1β were significantly upregulated just 1-day after low-dose NaIO₃ administration (Figs. 7C, 7D), indicating an early chemokine response. On day 3, there was a notable increase in the mRNA transcript levels of cytokines, Il-1β and Il-6 (Figs. 7A, 7B). In contrast, no significant changes in chemokine or cytokine expressions were observed in WT controls throughout the experiment. 

Figure 7.

 

Transcriptional levels of the chemokines and cytokines in the retina of WT and Lamp2 KO mice following NaIO₃ treatment. The mRNA levels of Mcp-1, Mip-1β, Il-1β, and Il-6 were evaluated 1 and 3 days after the NaIO₃ injection. The expression of Mcp-1 and Mip-1β transcripts in the retina of Lamp2 KO mice increased significantly on day 1, and Il-1β and Il-6 increased on day 3, whereas there were no changes in WT mice. Data are shown as mean ± SD, n = 6 eyes. One-way ANOVA followed by Tukey's post hoc test. *P < 0.05, **P < 0.01.

Effects of Clodronate Administration on NaIO₃-Induced Retinal Degeneration in Lamp2 KO Mice

To investigate the functional role of engrafted macrophages in retinal degeneration, we administered clodronate, a chemical agent known to selectively deplete macrophages.^39,40 Treatment with clodronate before 10 mg/kg NaIO₃ injection resulted in a significant reduction of Iba-1⁺ macrophages in the retinas of Lamp2 KO mice (Figs. 8A–F). Additionally, clodronate treatment significantly attenuated NaIO₃-induced RPE degeneration in Lamp2 KO mice compared with the control group (Figs. 8G, 8H). Histological analysis revealed that clodronate ameliorated NaIO₃-induced thinning of the ONL in Lamp2 KO mice compared with the control group, suggesting that macrophage depletion ameliorated NaIO₃-induced photoreceptor damage (Figs. 8I, 8J). 

Figure 8.

 

The effect of the macrophage-depleting agent clodronate in the retinal degeneration induced by NaIO₃ in Lamp2 KO mice. Representative images of retina and RPE/choroid flat-mounts stained with Iba-1 (green) and ZO-1 (red), along with hematoxylin and eosin–stained cross-sections, from PBS control and clodronate injection groups in Lamp2 KO mice 7 days after NaIO₃ treatment. (A–F) Clodronate treatment reduced Iba-1⁺ cells in the retinal layers of Lamp2 KO mice after NaIO₃ injection. Scale bars, 50 µm. (G) Moreover, The RPE degeneration area was significantly reduced by clodronate treatment. Scale bars, 200 µm/50 µm. (H) Quantification of the RPE degeneration resulted in an 83% decrease by clodronate treatment. Data are shown as mean ± SD, n = 4 eyes. Student t test. (I) Hematoxylin and eosin staining revealed depletion of macrophages using clodronate mitigated NaIO₃-induced ONL thinning. Scale bars, 50 µm. (J) Thickness of the ONL on day 7 after NaIO₃ injections in Lamp2 KO mice. Data are shown as mean ± SD, n = 6 eyes. Student t test. **P < 0.01, ***P < 0.001. GCL, ganglion cell layer; IPL, inner plexiform layer; OPL, outer plexiform layer, PRL, photoreceptor segment layer; SRS, subretinal space.

To evaluate the impact of clodronate on retinal oxidative stress levels following NaIO₃ treatment, qPCR was used to analyze the expression of antioxidant genes and tissue-specific markers in the retina and RPE of Lamp2 KO mice. In the retina, clodronate treatment reduced the expression of Gpx4 and Sod1 after 3 days of NaIO₃ exposure (Supplementary Fig. S3A). In the RPE, clodronate induced a reduction in Gpx4 expression on day 1 and a reduction in Nrf2 and Sod1 expression on day 3 (Supplementary Fig. S3B). These results illustrate that the inhibition of macrophage-recruitment may significantly contribute to oxidative stress damage induced by NaIO₃ in Lamp2 KO mice. Furthermore, clodronate treatment also reversed the NaIO₃-induced decrease in Rpe65 expression in the RPE, highlighting a protective effect on this tissue (Supplementary Fig. S2A). In contrast, Rho expression in the retina remained unchanged (Supplementary Fig. S2B), probably due to the limited extent and severity of photoreceptor damage compared with that of the RPE. 

Discussion

In this study, we elucidated the vulnerability to oxidative stress in Lamp2-deficient mice using an NaIO₃-induced RPE degeneration model. Although a 10-mg/kg dose of NaIO₃ did not cause retinal degeneration in WT mice, it led to significant RPE degeneration in Lamp2 KO mice. The RPE degeneration was associated with macrophage infiltration and elevated levels of inflammatory cytokines and chemokines. Furthermore, pretreatment with clodronate, which blocks macrophage infiltration, suppressed NaIO₃-induced RPE degeneration in Lamp2 KO mice. These results suggest that Lamp2 deficiency triggers a detrimental inflammatory response by microglia and macrophages under an oxidative stress condition. Thus, this study identifies a novel retinal pathology that occurs when two aging-related factors, oxidative stress and lysosomal dysfunction, combine in vivo. 

In our previous study, Lamp2-deficient mice displayed slowly progressive retinal degeneration characterized by lipid and basal laminar deposits.^12,41 In contrast, younger Lamp2 KO mice showed minimal RPE cell loss under untreated conditions (Fig. 2A). However, after the administration of 10 mg/kg NaIO₃, these younger mice experienced significant RPE damage. Therefore, Lamp2 deficiency contributes to vulnerability to oxidative stress in the retina. Only in Lamp2 KO mice, not in WT, did the expression of several key antioxidant genes, such as Gpx4, Nrf2, and Sod1,^42,42,44 increase in the retina and RPE after a low-dose NaIO₃ administration (Fig. 1). This upregulation of antioxidant genes likely represents a compensatory response to the oxidative stress damage observed in Lamp2-deficient retinas. Consistent with previous studies that highlight the important role of autophagy in eliminating reactive oxygen species in RPE cells,^24,45,46 oxidative stress damage has been accelerated in the retina of Lamp2 KO mice. Lee et al.⁴⁷. investigated reactive oxygen species–induced damage in Lamp2-deficient ARPE-19 cells in vitro. They showed that Lamp2 knockdown resulted in the upregulation of GPx4, similar to our results in Lamp2 KO mice (Fig. 1). According to their results, Lamp2 deficiency reduced cytosolic cysteine and glutathione that protect cells from oxidative stress. This finding may explain why RPE damage can worsen despite the upregulations of antioxidant genes in Lamp2 deficient cells. 

Similar to the retinal degeneration induced by higher doses of NaIO₃ in WT mice,^28,29 the retinal histology in Lamp2 KO mice treated with a low dose of NaIO₃ mainly involved the outer retina, such as ONL thinning and RPE degeneration (Fig. 3B). Moriguchi et al.³² have shown that Iba-1⁺ cells infiltrate the subretinal space following NaIO₃ administration in WT mice. Similarly, we found that Iba-1⁺ cells accumulated in the retina of Lamp2 KO mice treated with low-dose NaIO₃ (Fig. 4D). Moreover, we determined the population of these infiltrating immune cells in the retinas of Lamp2 KO mice using flow cytometry. CD45^lowMHC-II^low microglia increased at an early phase within 24 h, which remained elevated for up to 7 days (Fig. 5B), indicating that resident microglia are the first cell population to proliferate and migrate to the outer retina and subretinal space. Thereafter, CD45^highLy6C^high monocytes and CD45^highLy6C^low macrophages began to increase on day 3 (Fig. 6B). Consistent with a previous report,⁴⁸ our results suggested that the increase in Ly6C^low macrophages originates from circulating monocytes. Upon injury, inflammatory monocytes are recruited to the site of inflammation.^49,50 These monocytes initially express high levels of Ly6C, which gradually decrease as they differentiate into macrophages.^51,52 Thus, NaIO₃ administration evoked an early response of resident microglia, followed by a late engraftment of monocyte-derived macrophages in Lamp2 KO mice. Taken together with the upregulated transcription of chemokines and cytokines and the effects of macrophage depletion by clodronate, monocytes infiltrating into the retina and the engrafted monocyte-derived macrophages may have evoked neuroinflammation under oxidative stress in Lamp2 KO mice. 

This study had several limitations. Because the Lamp2 KO mice used in this study were systemically deficient in Lamp2, it remains unclear to what extent the observed phenotype is due to Lamp2 deficiency specifically in the retina or in the infiltrating immune cells. Further studies in animals harboring impaired lysosomal functions, specifically in the retina or immune cells, may clarify this. Additionally, there were technical limitations in in vivo settings for detecting reactive oxygen species or damaged organelles, whereas in vitro analyses related to autophagy/lysosomes and oxidative stress have been well-studied.^21,22 

In conclusion, Lamp2-deficient mice exhibited increased susceptibility to oxidative stress-induced RPE degeneration, as evidenced by significant degeneration even at a low dose of NaIO₃. Inhibition of macrophage infiltration mitigated RPE degeneration, highlighting the crucial role of inflammatory immune cells. These findings suggest a synergistic effect of oxidative stress and lysosomal dysfunction on retinal health. 

Acknowledgments

The authors thank Fumiyo Morikawa for the special assistance of transmission electron microscopy imaging. 

Funded by Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research: JP21K09702. 

Disclosure: G. Wu, None; S. Notomi, None; Z. Xu, None; Y. Fukuda, None; Y. Maehara, None; Y. Tao, None; H. Zhao, None; K. Ishikawa, None; Y. Murakami, None; T. Hisatomi, K.-H. Sonoda, None