C57BL/6J mice were bred in the central animal facility at Pennsylvania State University College of Medicine in Hershey, Pennsylvania. This study used 3-month-old male mice for all experiments. A small cohort of female mice was used to determine that there were no sex-based differences in the response to NaIO₃, at least detected by the imaging modalities used in this study. All animals had unlimited access to food and water and were maintained on a 12-hour light/dark cycle. All animal husbandry and experimental procedures were performed in accordance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals, and were approved by the Pennsylvania State University College of Medicine Institutional Animal Care and Use Committee. Experiments were also in compliance with the Association for Research in Vision and Ophthalmology statement for the use of animals in ophthalmic and vision research.

NaIO₃ solution (5 mg/mL, Sigma-Aldrich #S4007, St. Louis, MO, USA) was freshly prepared in sterile phosphate-buffered saline (PBS), pH 7.4, on the day of injection. Mice received a single intraperitoneal (i.p.) injection of 0 (PBS vehicle control), 15, 30, 45, or 60 mg/kg (n = 4 per dose) and were housed for seven days post-injection. This range was chosen to mimic intermediate AMD, based on other studies showing moderate to severe damage to the retina after NaIO₃ doses 30 mg/kg or higher with a more rapid decline after doses $>$60 mg/kg [16, 17, 18], while lower concentrations produce less rapid degeneration.

Mice were anesthetized using a 100 mg ketamine (Vedco, Inc. #50989-996-06, St. Joseph, MO, USA)/10 mg xylazine (Akorn #59399-110-20, Decatur, IL, USA)/kg i.p. injection. A drop of 2.5% proparacaine-HCl (Bausch & Lomb #17478-263-12, Lakefront, IL, USA) solution and 1% phenylephrine (Paragon BioTeck, Inc. #42702-102-15, Portland, OR, USA) was placed on each eye for topical anesthesia and pupil dilation, respectively. Artificial tears and an eye cover (custom 3D-printed) were used to protect the cornea from dehydration before image acquisition. OCT images were collected using Envisu R2210 (Bioptigen, Durham, NC, USA) fitted with mouse optics pre-NaIO₃, and 1-, 3-, 5- and 7 days post-NaIO₃ injection. Images were centered on the optic nerve head, and 1.4 mm $\times$ 1.4 mm B-scans and radial scans of the retina were acquired with a resolution of 8108 DPI.

The greyscale of the OCT images was inverted to assist with easier quantification. The number of HRF was quantified manually. Before quantification, the images were deidentified and the concentration of NaIO₃ was removed from labels. Intraretinal hyperreflective spots were characterized as being distinct from the RPE layer but with similar reflectivity, and their size ranged from 10–45 µm. Other artifacts were not included due to their lack of hyperreflectivity and/or lack of distinction from other structures in the retina. HRF in the vitreous were not counted.

At day seven after NaIO₃ treatment, mice were euthanized by CO₂ asphyxiation and the eyes were removed and placed in ice-cold PBS (pH 7.4). Eyecups were prepared as previously described [19, 20, 21]. Briefly, the eye was placed in ice-cold PBS and the cornea, lens and vitreous were removed under a dissection microscope. Eyecups were then fixed with 4% paraformaldehyde (PFA) (Fisher Scientific #T353-500, Hampton, NH, USA) for 15 min at room temperature (RT) and washed three times for 5 min in PBS. The eyecups were then placed in 30% Sucrose (MP Biomedical #194747, Santa Ana, CA, USA)/0.05% sodium azide (Sigma Aldrich #S-8032, St. Louis, MO, USA) solution and stored at 4 °C until processing. The eyes were frozen in 7.5% gelatin (Sigma Aldrich #G2500-500G)/15% sucrose (MP Biomedical #194747) and sectioned using a cryostat (12 µm).

Six sections per mouse were examined. The sections were blocked using 5% normal donkey serum (NDS) (Jackson ImmunoResearch Labs #017-000-121, West Grove, PA, USA) for 1 hour and incubated in primary antibodies overnight at 4 °C. The sections were then rinsed using 1$\times$ PBS and incubated with the secondary antibodies and/or conjugated antibodies for 1 hour at room temperature. After the incubation, the sections were rinsed once more with 1$\times$ PBS, mounted using aqua-polymount (Polysciences Inc. #18606-20, Warrington, PA, USA), and imaged using a Leica SP8 confocal microscope (Department of Cellular and Molecular Physiology core facility, Deerfield, IL, USA). The antibodies used were as follows: monoclonal mouse anti-RPE65 (1:500, Abcam #401.8B11.3D9, Cambridge, UK), polyclonal rabbit anti-Vimentin [VIM] (1:250, Atlas #HPA001762, Solna, Sweden), monoclonal rat anti-F4/80 conjugated to Phycoerythrin (1:100, Invitrogen #12-4801-82, Carlsbad, CA, USA), polyclonal donkey anti-mouse Immunoglobulin G (IgG)-Alexa Fluor Plus 488 (1:1000, Invitrogen #A32766), polyclonal donkey anti-rabbit IgG-Alexa Fluor Plus 594 (1:1000, Invitrogen #A32754), and Hoechst 33342 (1:200, ThermoFisher #62249, Waltham, MA, USA). The images were processed using Image J (Version 2.14.0/1.54f, NIH, Bethesda, MD, USA) by applying maximum projections to Z-stacks.

Six sections per mouse were fixed again in 4% PFA overnight at 4 °C and rehydrated through graded ethanol (100%, 95%, 70%) before 10 min. incubation in hematoxylin. Then they were rinsed using tap water, dipped in EtOH/HCl twice, and then ammonia water for 30 s. The sections were then submerged in Eosin-R for 2 min. Sections were dehydrated through graded ethanol, and cleared in xylene. The slides were mounted using Cytoseal 60 mounting media (Thermo Scientific #8310-16, Waltham, MA, USA). The sections were imaged using a Zeiss AxioObserver v7.0 microscope fitted with an Apotome (Zeiss, Oberkochen, Germany). Images were acquired using Zen software (version Lite, Zeiss, Wetzlar, Germany). The thickness of retinal layers and the number of pigmented lesions were quantified manually. The central and peripheral regions were determined by dividing the length of the retina into four parts. The inner portions were considered to be central and the outer portions were considered to be peripheral.

Mice were euthanized as described above. Eyes were removed and placed in transmission electron microscopy (TEM) fixative (0.16 M cacodylate buffer (Electron Microscopic Sciences Cat #1165, Fort Washington, PA, USA), 16% paraformaldehyde (Electron Microscopic Sciences #15700), 8% glutaraldehyde (Electron Microscopic Sciences #16000), 30% sucrose, and 1 M CaCl₂) for 1 hour at 25 °C. The iris and lens were removed and the eyecup was placed in TEM fixative for 24 hours at 4 °C. After fixation, the eye cup was cut in half through the optic nerve. The central portion of each half was removed and placed in TEM fixative for processing in the Pennsylvania State College of Medicine TEM core facility. Samples were fixed using 2.5% glutaraldehyde/2% paraformaldehyde in 0.1 M cacodylate buffer (pH 7.4) and further fixed in 1% osmium tetroxide (Electron Microscopic Sciences #19100) in 0.1 M cacodylate buffer (pH 7.4) for 1 hour. Samples were dehydrated in a graduated ethanol series and embedded in Epon 812 (Electron Microscopic Sciences #14121). Thin sections (70 nm) were stained with uranyl acetate (Electron Microscopic Sciences #22400) and lead citrate (Electron Microscopic Sciences #17800). 9 sections per eye were examined for qualitative differences using a JEOL JEM1400 Transmission Electron Microscope (JEOL USA Inc., Peabody, MA, USA). The TEM Core (RRID:SCR_021200) services and instruments used in this project were funded, in part, by the Pennsylvania State University College of Medicine via the Office of the Vice Dean of Research and Graduate Students and the Pennsylvania Department of Health using Tobacco Settlement Funds (CURE). The content is solely the responsibility of the authors and does not necessarily represent the official views of the University or College of Medicine. The Pennsylvania Department of Health specifically disclaims responsibility for any analyses, interpretations, or conclusions.

Data are presented as mean $\pm$ standard error of the mean (SEM). Statistical analyses were performed using Prism 8 (GraphPad, Inc., La Jolla, CA, USA). One-way and two-way analyses of variance (ANOVA) were used to determine overall significance, where individual group comparisons were made by post-hoc Tukey test. For correlation analyses, the Pearson product-moment correlation coefficient R was determined [20, 22], and correlational strength was defined as: no correlation for 0.0 $\le$ r $\le$ 0.2, weak correlation for 0.2 $\le$ r $\le$ 0.4, moderate correlation for 0.4 $\le$ r $\le$ 0.6, and strong correlation for 0.6 $\le$ r $\le$ 1.0, as described previously [20, 22]. Significant correlations (p 0.05) are labeled with both the Pearson product-moment correlation coefficient r and p-values.

Retinas were imaged by OCT on Day 0, before injection with NaIO₃ or the vehicle, and also 1, 3, 5, and 7 days after NaIO₃. On day 0 there were no observable differences in the morphology of the retinas between groups. No HRF or other anomalies were noted on OCT images No HRF were present in retina of mice given vehicle or 15 mg/kg NaIO₃ after 7 days; however, HRF were observed in mice given 30 mg/kg, 45 mg/kg, and 60 mg/kg by 5 days post-injection (Fig. 1). Quantification of OCT images revealed that the number of HRF significantly increased in retinas of mice 5 days after $\ge$30 mg/kg NaIO₃, compared to the control and 15 mg/kg groups (Fig. 2A–D. 2-way ANOVA with Tukey multiple comparisons, p 0.001). Retinal thickness was also significantly less in the $\ge$30 mg/kg NaIO₃, compared to the control and 15 mg/kg groups, 5 and 7 days after NaIO₃ (Fig. 3A–E. 2-way ANOVA with Tukey multiple comparisons, p 0.001). There was also a significant inverse correlation between retinal thickness and HRF number when the data was pooled regardless of concentration or time passed (Fig. 3F, Pearson r = 0.6601; p 0.0001). Another interesting observation from this data is that the amount of HRF were present by day 1 and 3 post-NaIO₃ injection, but significant retinal tissue loss was not seen until day 5 and 7 post-NaIO₃ injection. This suggests that the HRF in this model may be predictive of photoreceptor cell death. There were no differences in HRF distribution on OCT (Supplementary Fig. 1).

Fig. 1.

Optical Coherence Tomography Images of Retinas After Sodium Iodate Contain Hyperreflective Foci. Optical coherence tomography (OCT) images were acquired from mice at baseline (pre-sodium iodate [NaIO₃]), and 1-, 3-, 5-, and 7-days post-injection (n = 4 per group). (A) Representative OCT images of mice given varying concentrations of NaIO₃ at different time points are shown. Hyperreflective foci (HRF) (orange arrowheads) and thinning of the retina were induced at NaIO₃ doses $\ge$30 mg/kg. (B) Magnification of the representative OCT images (orange arrows). HRF (orange arrowheads) and RPE abnormalities (blue asterisk) are highlighted. Scale bar = 100 µm. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; ELM, external limiting membrane; IS/OS, inner segments/outer segments; RPE, retinal pigment epithelium.

Fig. 2.

Sodium Iodate Increases the Number of Hyperreflective Foci over Time. The retinal thickness and number of hyperreflective foci (HRF) were manually quantified in OCT images, n = 4 per group. (A–D) Number of HRF 1, 3, 5, and 7 days after NaIO₃. There was a significant increase in the number of HRF 5 and 7 days after NaIO₃ with doses $\ge$30 mg/kg compared to the control and 15 mg/kg groups, *p 0.05 (One-way ANOVA, with Tukey’s multiple comparisons test. The data expressed as mean $\pm$ SEM).

Fig. 3.

Sodium Iodate Decreases Retinal Thickness over Time and Correlation between the number of Hyperreflective Foci and Retinal Thickness. (A–E) Total retinal thickness (µm) was measured 0, 1, 3, 5, and 7 days after NaIO₃. The total retinal thickness was significantly less 5 and 7 days after NaIO₃ with doses $\ge$30 mg/kg compared to the control and 15 mg/kg groups, *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001 (2-way ANOVA with Tukey multiple comparisons test. Data expressed as mean $\pm$ SEM). (F) Correlation between retinal thickness (µm) and HRF number regardless of time or concentration. Pearson correlation, R² = 0.4357 r = 0.6601; p 0.0001.

Eyes from mice 7 days after NaIO₃ injection were sectioned for Hematoxylin & Eosin (H&E) staining. Sections from mice given the vehicle contained no pigmented abnormalities and the RPE layer was smooth and continuous, while sections from groups given $\ge$15 mg/kg NaIO₃ contained pigmented abnormalities within the outer retina as well as extensive morphological abnormalities such as folding of the outer and inner nuclear layers (Fig. 4A). Sections from the 15 mg/kg group contained subtle morphological abnormalities but no pigment was noted outside the RPE layer, while some areas of the RPE were irregular (indicated by yellow arrowhead). There was also extensive distortion of the RPE layer. These pigmentary changes included breaks in the RPE layer as well as areas of aggregated pigment. Some of the pigment in mice given $\ge$30 mg/kg of NaIO₃ was detached from the RPE layer and appeared in the neural retina. This pigment was found in clusters in the outer plexiform layer (OPL). These retinas also appeared thinner and distorted compared to controls.

Fig. 4.

H&E Sections Revealed Pigmented Anomalies and Distorted Retinal Morphology Seven Days after Sodium Iodate Injection. Cross-sections of mouse retinas were collected and stained using H&E, 7 days after NaIO₃ injection (n = 4–5 per group). (A) Subtle changes in RPE morphology were noted in the 15 mg/kg NaIO₃ group (yellow arrowhead). The RPE was discontinuous in mice given 30, 45, and 60 mg/kg NaIO₃ (yellow asterisk), and abnormal retinal pigment epithelium (RPE) layer structure is highlighted, and pigment abnormalities appeared within the outer retina (red arrowheads). Other abnormalities included folding of the nuclear layers and loss of photoreceptor outer segments. (B–F) The thickness of the IS/OS, ONL, and OPL layers was significantly less in mice given NaIO₃ compared to the control group, whereas the IPL and INL experienced little or no significant changes (Data expressed as mean $\pm$ SEM, *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001, this data had a significant F-test (p 0.001) so ANOVA could not be used. Mixed-effects analysis was used to determine significance). IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; IS/OS, inner segments/outer segments; RPE, retinal pigment epithelium; Ch, choroid. Scale bar (top panels) = 100 µm, scale bar (bottom panels) = 50 µm.

There was a significant reduction in the thickness of the inner segments/outer segments (IS/OS) in groups given $\ge$30 mg/kg of NaIO₃ compared to controls (Fig. 4B. p 0.0001). There was also significant thinning of the outer nuclear layer (ONL) and outer plexiform layer (OPL) in these groups compared to the control (Fig. 4C,D, p 0.0001 and p 0.001 respectively). The 15 mg/kg NaIO₃ group also had significant thinning of the ONL, OPL, and inner nuclear layer (INL) compared to the control (Fig. 4E, p 0.05). The inner plexiform layer (IPL) thickness remained unchanged in all NaIO₃ groups compared to the control (Fig. 4F). There were no significant differences between the amount of pigment in central and peripheral parts of the neural retina (Supplementary Fig. 1B).

TEM images of mouse retinas were collected 7 days post-NaIO₃ injection. The RPE cells of the NaIO₃ mice had abnormal morphology; specifically, the microvilli of the RPE cells were absent and the basal infoldings had deteriorated in the NaIO₃ retinas. The basal infoldings were also missing from some areas, and, when present, appeared smaller than in controls (Fig. 5).

Fig. 5.

Sodium Iodate Alters Retinal Pigment Epithelium Anatomy and Organization. Representative transmission electron microscope (TEM) image of the RPE layer of mice 7 days after 60 mg/kg of sodium iodate (NaIO₃) (n = 3 per group). The basal infoldings of the retinal pigment epithelium (RPE) are absent, and the microvilli of the RPE are not seen. The RPE layer is disorganized and generally unrecognizable. Bm, Bruch’s membrane; IS, inner segments; M, Melanin; OS, outer segments; RPE, Retinal pigment epithelium. Scale bar = 2 µm.

Pigmented abnormalities were observed in the ONL and the IS/OS layers of the retina of NaIO₃-treated mice (Fig. 6A). The control retinas had intact photoreceptor outer segments and the RPE cells had basal infoldings along Bruch’s membrane and microvilli that extended into the OS layer. In contrast, the OS layer of the NaIO₃-treated mice was degraded or missing in several regions. The OS and IS layers contained displaced electron-dense structures that resembled pigment, which was not seen in the control retinas. The size and abundance of the displaced electron-dense pigment varied and were noted in multiple layers of the outer retina, such as the photoreceptor OS and ONL layers. Along with the displaced electron-dense pigment, outer segment fragments were observed in the ONL. TEM of the NaIO₃ retinas also revealed that the electron-dense pigment was contained intracellularly, based on its presence within the same plasma membrane shared with a nucleus (Fig. 6B). The image shows a displaced pigment-containing cell within the ONL (indicated by surrounding plasma membrane and photoreceptor nucleus). The morphology of the cell appeared abnormal compared to typical photoreceptor cells, suggesting that it originated outside the ONL. The cell also contained electron-dense pigment and partially digested photoreceptors.

Fig. 6.

Transmission Electron Microscopy Images Show Abnormal Electron-Dense Structures in the Outer Retina. Representative transmission electron microscope (TEM) of control mice and mice given sodium iodate (NaIO₃) are shown (n = 3 per group). (A) Electron-dense masses were exclusively found in the RPE of the control mice. After concentrations of NaIO₃ $\ge$30 mg/kg, electron-dense masses were found in multiple layers of the ONL of the neural retina. ONL scale bar = 5 µm. Magnified ONL scale bar = 1 µm. RPE scale bar = 2 µm. Magnified RPE scale bar = 2 µm. (B) The electron-dense masses were contained intracellularly in the NaIO₃-treated mice, indicated by the presence of a plasma membrane and nucleus (highlighted in blue). The electron-dense masses are marked with red asterisks and partially digested photoreceptors with yellow asterisks. Bruchs, Bruch’s membrane; IS, Inner Segments; ONL, Outer Nuclear Layer; OS, Outer Segments; RPE, Retinal Pigment Epithelium. Scale bar = 5 µm. Magnified scale bar = 1 µm.

Retinal tissue was collected 7 days after NaIO₃ injection and processed for immunofluorescent labeling. Confocal imaging revealed that cells in the RPE layer were positive for the RPE cell marker, RPE65, and the EMT marker, VIM, in the retinas of mice treated with NaIO₃ $\ge$30 mg/kg (Fig. 7). The RPE⁺ VIM⁺ immunoreactive cells appeared to be more abundant in mice treated with higher concentrations of NaIO₃.

Fig. 7.

Sodium Iodate Induced Vimentin Expression in the Retinal Pigment Epithelium Layer. Representative confocal images of retinal sections from mice 7 days after NaIO₃ injection, immunolabelled for RPE65 (green) and vimentin (VIM, red) and counterstained for nuclei with Hoechst (blue) (n = 4 per group). The overlay column (Comp) shows the merged image of all three fluorescent channels. The transmitted light channel (TL) is shown in the righthand column. First row: Control conditions. There was no VIM⁺ immunoreactivity in the RPE layer. Second row: NaIO₃ 15 mg/kg. No VIM⁺ immunoreactivity was detected in the RPE layer. Third row: NaIO₃, 30 mg/kg. VIM⁺ and RPE65⁺ immunoreactivity colocalized (white arrows) and in the pigmented RPE layer (black arrows). Fourth row: NaIO₃, 45 mg/kg. VIM⁺ and RPE65⁺ immunoreactivity colocalized (white arrows) in the pigmented RPE layer (black arrows). Fifth Row: NaIO₃, 60 mg/kg. VIM⁺ and RPE65⁺ immunoreactivity colocalized abundantly (white arrows) in the pigmented RPE layer (black arrows). ONL, outer nuclear layer. Scale bar = 50 µm.

Retinal sections were also labeled for F4/80, a murine macrophage marker (Fig. 8). In the control retinas, F4/80 immunoreactive cells were detected in the choroid and sclera but not in the RPE and neural retina. Similarly, the retinas of mice given 15 mg/kg NaIO₃ contained only sparse F4/80 immunoreactive cells in the neural retina. In mice given 30 mg/kg NaIO₃, there was a greater abundance of F4/80 immunoreactivity in the neural retina, but only sparse F4/80⁺ immunoreactivity in the RPE layer. In the 45 mg/kg NaIO_3- group, there was limited F4/80⁺ immunoreactivity in the neural retina and more in the RPE layer. Similar results were observed in the retinas of mice given 60 mg/kg NaIO₃.

Fig. 8.

Sodium Iodate Increased F4/80⁺ Immunoreactivity in the Retinal Pigment Epithelial Layer and Neural Retina. Representative confocal images of retinal sections from mice 7 days after NaIO₃ injection, immunolabelled for RPE65 (green) and F4/80 (red), and counterstained for nuclei with Hoechst (blue) (n = 4 per group). The overlay column (Comp) shows the merged image of all three fluorescent channels. The transmitted light (TL) channel is shown in the righthand column. F4/80⁺ immunoreactivity was observed in the choroid in all treatment groups. First row: Control. F4/80⁺ immunoreactivity was restricted to the choroid. Second row: NaIO₃ 15 mg/kg. F4/80⁺ immunoreactivity colocalized with RPE65 infrequently in the neural retina in addition to the choroid (white arrowheads) and corresponded to pigmented puncta (black arrowheads). Third row: NaIO₃, 30 mg/kg. abundant F4/80⁺ immunoreactivity partially colocalized with RPE65 in the neural retina and RPE layer, in addition to the choroid (white arrowheads), and corresponded to pigmented puncta (black arrowheads). Fourth row: NaIO₃, 45 mg/kg. F4/80⁺ immunoreactivity partially colocalized with RPE65 in the neural retina and RPE layer, in addition to the choroid (white arrowheads), and corresponded to pigmented puncta (black arrowheads). Fifth Row: NaIO₃, 60 mg/kg. Occasional F4/80⁺ immunoreactivity partially colocalized with RPE65 in the neural retina and RPE layer, in addition to the choroid (white arrowheads), and corresponded to pigmented puncta (black arrowheads). ONL, outer nuclear layer. Scale bar = 50 µm.

Pigment localized within the neural retina was observed by transmitted light imaging and corresponding confocal imaging of VIM, RPE65, and F4/80 immunofluorescence (Fig. 9). Pigment, VIM immunoreactivity, and RPE65 immunoreactivity colocalized in some places (Fig. 9A), but this was not uniform (Fig. 9B). Pigmented anomalies within the neural retina were also occasionally positive for F4/80⁺ immunoreactivity (Fig. 9C), but in most cases the pigment did not colocalize with F4/80 (Fig. 9D).

Fig. 9.

Heterogeneous Vimentin and F4/80 Immunoreactivity in Pigmented Anomalies. Pigmented anomalies were identified by transmitted light and compared to corresponding confocal images in retinal sections from mice 7 days after NaIO₃ injection. Sections were immunolabelled for RPE65 (green) and F4/80 (red) or VIM (red), and nuclear counterstained with Hoechst (blue) (n = 4 per group). (A) Pigmented lesions identified in transmitted light (TL) images corresponded to Hoechst stain, some RPE65 immunoreactivity, and VIM immunoreactivity. (B) In other cases, pigmented lesions had Hoechst staining but no VIM or RPE65 immunoreactivity. (C) Pigmented lesions also partially corresponded to RPE65 and F4/80 immunoreactivity. (D) In other areas, pigmented lesions had Hoechst staining but no RPE65 or F4/80 immunoreactivity. Scale bar = 50 µm. Magnified scale bar = 25 µm.