Molecular Vision 2024; 30:219-233
<http://www.molvis.org/molvis/v30/219>
Received 05 July 2023 | Accepted 13 November 2023 | Published 02 April 2024
Molecular Vision 2024; 30:219-233
<http://www.molvis.org/molvis/v30/219>
Received 05 July 2023 |
Accepted 13 November 2023 |
Published 02 April 2024
Tomoaki Araki,1 Masamitsu Shimazawa,2,3 Shinsuke Nakamura,2 Wataru Otsu,3 Yosuke Numata,1 Megumi Sakata,1 Koji Kabayama,1 Hideshi Tsusaki,1,3 Hideaki Hara2,3
1Shin Nippon Biomedical Laboratories Ltd. Drug Safety Research Laboratories (SNBL DSR), Kagoshima, Japan; 2Molecular Pharmacology, Department of Biofunctional Evaluation, Gifu Pharmaceutical University, Gifu, Japan; 3Biomedical Research, Gifu Pharmaceutical University, Gifu, Japan
Correspondence to: Masamitsu Shimazawa, Gifu Pharmaceutical University Molecular Pharmacology, Department of Biofunctional Evaluation 1-25-4 Daigaku-nishi, Gifu, 501-1196, Japan; FAX: +81-58-230-8126; Phone: +81-58-230-8126; email: shimazawa@gifu-pu.ac.jp
Purpose: We screened 28 female cynomolgus monkeys (CMs) and 25 female rhesus monkeys (RMs) for white dots (WDs) in the macula and detected several animals with WDs in colonies at the Shin Nippon Biomedical Laboratories, Ltd., Drug Safety Research Laboratories (SNBL) facility. To determine the functional and morphological characteristics of WDs, we conducted ophthalmological and pathological examinations on these animals.
Methods: Fundus examination, optical coherence tomography (OCT), and focal electroretinogram (f-ERG) were conducted for all animals. Histopathology and transmission electron microscopy were conducted for one representative adult CM with WDs.
Results: In both CMs and RMs, individual differences were observed in the number of WDs in the macula (ranging from approximately 10 to 500 per eye). Hyperreflective granules were observed between the ellipsoid zone (EZ) and the retinal pigment epithelium (RPE) in OCT. Both CMs and RMs exhibited a significant increase in the thicknesses of the RPE and choroid in WD animals compared to their normal counterparts. In the f-ERG, significant decreases and/or tendencies toward decreases in amplitudes and increases in implicit times of both a- and b-waves were observed in animals with WDs. In pathology, diffuse vacuolization of the RPE cells with tiny granules was observed in the macula.
Conclusions: Based on the results of OCT and pathological examinations, it was suggested that animals with WDs can develop macular degeneration in the future. To assess their suitability as a model for precursor lesions of age-related macular degeneration, it is imperative to continue monitoring the animals used in the present study until they reach a more advanced age of approximately another 5–10 years.
Cynomolgus monkeys (CMs) and rhesus monkeys (RMs) have been commonly used in the studies conducted on nonclinical safety during the development of new medicines [1-4]. Nonhuman primates (NHPs) have also been used in the research on various ocular diseases, such as inherited retinal disease [5], age-related macular degeneration [6], glaucoma [7], and presbyopia [8]. Over the years, the Shin Nippon Biomedical Laboratories, Ltd. (SNBL) facility has handled thousands of NHPs annually for contracted nonclinical safety studies and eye research. Health checkups on the NHPs before being assigned to contracted studies include ophthalmological prescreening using an indirect ophthalmoscope. Macular white dot (WD) lesions are occasionally found in monkeys in these colonies; however, accurate incidence rates of these lesions have not yet been determined. The WDs observed through an indirect ophthalmoscope resemble drusen, which are also seen in the precursor lesions of age-related macular degeneration (AMD) in humans.
In the present study, we selected CMs and RMs aged between 3 and 17 years, including those suspected of having WD lesions during prescreening. The aim was to elucidate the functional and morphological characteristics of WDs using ophthalmological and pathological examinations. Through this analysis, we sought to assess the potential of these animals, with WDs, being used as models for the precursor lesions of AMD. The eyes of the monkeys with or without WDs were examined using fundus examination, optical coherence tomography (OCT), and focal electroretinogram (f-ERG), and the eyes of one representative CM with WDs were examined histopathologically as well as using an electron microscope.
The CMs and RMs in this study were provided by and housed at Shin Nippon Biomedical Laboratories, Ltd., Drug Safety Research Laboratories (SNBL DSR). We prescreened 30 CMs and 25 RMs and selected those with no abnormal clinical signs, no corneal or lens opacities, and clearly visible retinas. This resulted in 28 CMs (female, 3–17 years old, China) and 25 RMs (female, 3–14 years old, China), including animals with suspected WD lesions in the macula, being selected for use in the study.
The study was approved by the Institutional Animal Care and Use Committee of SNBL DSR (Approval No. IACUC999–891). All animals were kept indoors in accordance with the animal welfare bylaws of SNBL DSR, which is accredited by AAALAC International. The following environmental conditions were maintained: cages: stainless steel [680 mm (D) × 620 mm (W) × 770 mm (H)], temperature: 23 °C to 29 °C, humidity: 30%–70%, ventilation: 15 times/h, and illumination: 12 h/day of artificial light (07:00 to 19:00). The rooms and cages were washed daily with water. Nine food pellets (approximately 12 g/pellet, HF Primate J 12G 5K9J, Purina Mills, LLC) were provided to each animal daily. Water conforming to the Japanese Waterworks Law quality standards was available ad libitum from an automatic supply system. The enrichment programs were as follows: toys: available 24 h per day; audio: 12 h/day of music or radio (07:00 to 19:00); and treats: pieces of apple, bread, raisins, and sweet potatoes were supplied on the day of the ophthalmologic examinations. All animals were observed daily for clinical signs (external appearance, behavior, fecal conditions, and feeding condition). To alleviate suffering, ophthalmologic examinations were performed under ketamine anesthesia in accordance with standard operating procedures. A veterinarian was available throughout the study for medical treatment if required. A representative WD monkey (No. 19, CM, age 11 years) was euthanized for conducting histopathological examinations and transmission electron microscopy (TEM). All other animals were returned to the laboratory stock colony after completion of the study. No other animals were euthanized in the study.
Fundus examination, OCT examination, and f-ERG were performed on all animals. For each test, a mixture of 0.5% phenylephrine and 0.5% tropicamide (Mydrin-P Ophthalmic Solution, Santen Pharmaceutical, Co., Ltd., Osaka, Japan) was instilled, and the animals were anesthetized with an intramuscular injection (0.2 ml/kg, 10 mg/kg) of ketamine hydrochloride (Ketalar for Intramuscular Injection 500 mg, Daiichi Sankyo Propharma Co., Ltd., Tokyo, Japan, 50 mg/ml).
Photographs (3,216 × 2,136 pixels) of the ocular fundi were taken with a digital fundus camera (VX-10α, Kowa, Ltd., Aichi, Japan) and used to count the number of WDs in the macula. The counting was performed by dividing the macula into nine regions (Figure 1A,B).
The ocular fundi were examined using OCT (Heidelberg Spectralis OCT2, Heidelberg Engineering GmbH, Germany). A single 7.7 mm horizontal scan covering the optic papilla and fovea and a 2.0-mm-diameter 360° radial scan at 7.5° intervals centered on the fovea were performed. The retina and choroid were observed using tomographic imaging, and the thicknesses of the retinal pigment epithelium (RPE) and choroid at the macula were measured.
After the fundus and OCT examinations, a contact lens-type electrode was placed directly on the cornea. The f-ERG on the fovea was measured with photic stimulation (background light: 1.5 cd/m2, flash intensity: 15 cds/m2, duration: 100 ms, stimulation light size: 15°) using Kowa ER-80 (Kowa Co., Ltd., Aichi, Japan) and PuREC (Mayo Co., Ltd., Aichi, Japan). The f-ERG was recorded using a 5 Hz high-pass filter, 500 Hz low-pass filter, and 10 × 1000 gain, and 200 responses were averaged. While checking the fundus image of the animal on the monitor via the infrared imaging built into the f-ERG, the center of the stimulating light was adjusted to always hit the fovea.
'A representative WD monkey (No. 19, CM, age 11 years) was anesthetized for euthanasia with intramuscular injections of ketamine hydrochloride (Ketalar for Intramuscular Injection 500 mg, Daiichi Sankyo Propharma Co., Ltd., Tokyo, Japan, 50 mg/ml, 0.3 ml/kg) and medetomidine hydrochloride (Domitor, Orion Corporation, Finland, 1 mg/ml, 0.08 ml/kg) and euthanized by exsanguination. The eyeballs were fixed in a 3% glutaraldehyde and 2.5% formalin mixture. A portion was embedded in paraffin, and the remainder was washed with 0.2 mol/l sucrose-added 0.1 mol/l phosphate buffer and refixed in 3% glutaraldehyde. The paraffin-embedded tissue was sectioned, stained with hematoxylin-eosin (HE), and examined using an optical microscope (Ni-U, Nikon Co., Ltd., Japan). The 3% glutaraldehyde-fixed tissue was postfixed in 1% osmium tetroxide, embedded in epoxy resin, and examined using TEM (JEM-1400Plus, JEOL Co., Ltd., Japan).
Fundus examination-- The fundi of each animal were divided into nine regions in the macula, as depicted in Figure 1A,B. Based on the number of WDs counted in these regions, the animals were scored according to the following proprietary criteria. Animals with a score of 0 or 1 were classified as normal animals, and those with a score of 2 or 3 were classified as WD animals.
Criteria-- Score 0: No WD
Score 1: The total number of WDs across the nine regions was fewer than 10, and WDs were observed in fewer than five regions.
Score 2: The total number of WDs across the nine regions was 10 or more, and WDs were observed in fewer than five regions.
Score 3: The total number of WDs across the nine regions was 10 or more, and WDs were observed in five or more regions.
The number of hyperreflective granules between the ellipsoid zone (EZ) and RPE was visually measured using radial images of the macula. Starting with the sagittal axis, four axes at 45° intervals were selected (Figure 1C), and the total number of hyperreflective granules in the four images was used for the evaluation. The RPE and choroidal thicknesses were measured at the fovea and at distances of 500, 1000, and 1500 μm from the fovea in both nasal and temporal directions, using horizontal images spanning from the optic disc to the fovea. These measurements were subsequently averaged. The RPE and choroidal thicknesses were measured using the software Eye Explorer, version 6.12.3.0, included with the OCT device. The amplitudes and implicit times of the a-and b-waves were calculated from the recorded f-ERG using the analysis software included with the PuREC recording device.
To assess differences in WD count between the foveal and other regions, we used the Student’s t-test. The relationship between age and WD count scores was determined using a chi-square test. To detect significant differences between normal and WD animals in terms of the number of hyperreflective granules, as well as the amplitudes and implicit times of a-and b-waves in the f-ERG, we employed the Student’s t-test. Normal animals were further divided into young adults (under 10 years of age) and adults (above 10 years of age), and the Student’s t-test was used to detect any significant differences between them in terms of RPE and choroidal thicknesses. Using a two-way analysis of variance (ANOVA), we analyzed variations in RPE and choroidal thicknesses between normal and WD animals, taking into account both animal species (CM and RM) and their categorizations as either normal or WD. Finally, we employed Spearman’s rank correlation coefficient to study the link between hyperreflective granules and WD, while the Pearson correlation coefficient was used to assess their relation with RPE and choroid thickness. Data are shown as the mean ± standard error (mean ± SE).
The animals were classified into two groups (normal animals and WD animals) based on the criteria shown in the Methods section (Table 1 and Table 2). According to the literature, female CMs reach sexual maturity between 3.3 and 4.1 years of age [9] and experience menopause between 22 and 25 years [10,11], while female RMs mature at around 3 years [12] and undergo menopause at around 25 years [13,14]. Based on these findings, we categorized the monkeys aged 3–9 years as “young adults,” those aged 10–17 years as “adults,” and those aged 18–25 years as “mature.” However, as none of the animals used in our study were over 18 years old, we divided them into just two groups: “young adults” and “adults.” Of the 28 CMs and 25 RMs examined, 15 CMs and 14 RMs were found to be normal and 13 CMs and 11 RMs had multiple WDs (Figure 2). The WDs were present inside the retinal vascular arcade and were most dense in and around the fovea (Figure 3). The number of WDs per eye varied among the individuals, ranging from approximately 10 to 500. Despite this variability, no differences were observed in the shape, size, or color of the WDs. Using the chi-square test, we analyzed the association between age and WD count score. For CMs, there was no significant association (χ^2 = 17.13504, df = 11, p-value = 0.10394). In contrast, a significant relationship was found for RMs (χ^2 = 18.09614, df = 3, p-value = 0.00042).
In the OCT images, slightly hyperreflective lesions were observed in the RPE at the positions where WDs were observed in fundus examination (Figure 4A–D). In the HE-stained fundus samples of the representative WD CM (animal number 19, age 11 years), a small amount of vacuolizations were observed in the RPE (Figure 4E,H), which corresponded to the vacuolizations with microgranules observed using TEM (Figure 4I,J). The TEM revealed the presence of structures within the RPE that resembled lysosomes. These structures were enclosed by membranes, and inside the vesicles, both membrane-bound and nonmembrane-bound organelles, including mitochondria, were evident. In both the OCT images and histopathological sections, no drusen were observed.
In the radial OCT images, hyperreflective granules approximately 5–10 µm in diameter, were observed between the EZ and RPE in all eyes diagnosed with WDs (Figure 5A–D). The mean total numbers of hyperreflective granules between the EZ and RPE for both young adult and adult CMs and RMs, considering both normal and WD animals, are depicted in Figure 5E,F, respectively. The number of hyperreflective granules in the RMs was slightly higher than that in the CMs. The WD animals exhibited significantly more hyperreflective granules than the normal animals; however, the positions of the hyperreflective granules and WDs were not aligned. This observation was consistent in both CMs and RMs. Similarly, in the fundus images, the position of the hyperreflective granules were not aligned with that of the WDs. A positive correlation between the hyperreflective granules and the number of WDs was observed for both CMs and RMs (Figure 5G,H). However, the scatter plot showed considerable variability.
Based on the thickness measurements of the RPE and choroid derived from OCT images, the RPE thickness in CMs was 16.43 ± 0.31 µm for normal and 20.52 ± 0.25 µm for WD animals, whereas in RMs, it was 18.36 ± 0.35 µm for normal and 21.69 ± 0.35 µm for WD animals. The choroidal thickness in CMs was 166.65 ± 2.55 µm for normal and 178.16 ± 3.71 µm for WD animals, and in RMs, it was 165.64 ± 3.92 µm for normal and 178.76 ± 5.23 µm for WD animals (each value represents the mean of measurements from seven locations ± SE). From these results, a two-way ANOVA was conducted, evaluating both condition (normal versus WD) and species (CM versus RM). According to the analysis results, WD animals consistently exhibited a thicker RPE and choroid compared to their normal counterparts, with a mean difference of 3.708 (p < 0.001). When assessing differences based on species, RMs demonstrated greater RPE and choroidal thicknesses compared to CMs, with a mean difference of 1.554 (p < 0.001). This trend persisted within specific categories: among normal animals, RMs exhibited thicker RPE and choroid than CMs, with a mean difference of 1.928 (p < 0.001). Likewise, within the WD category, RMs had pronounced thicknesses of the RPE and choroid relative to CMs, with a mean difference of 1.180 (p = 0.015). An examination of the condition within each species revealed that CMs with WDs had markedly thicker RPE and choroid compared to their normal peers (mean difference of 4.082, p < 0.001). Similarly, RMs with WDs also presented a thicker layer than normal RMs, with a mean difference of 3.334 (p < 0.001). The RPE and choroid tended to be thicker in WD animals than in normal animals, as illustrated in Figure 6A,B. When comparing the RPE and choroid in normal animals between the “young adults” and “adults” categories, both CMs and RMs distinctly exhibited a thicker RPE and choroid in adults than in young adults (Figure 6C–F). A positive correlation was observed between the number of hyperreflective granules and RPE thickness in both CMs and RMs (Figure 6G,H). However, there was a negligible correlation between the number of hyperreflective granules and choroidal thickness within the RPE (Figure 6I,J).
Table 3 shows the a-and b-wave amplitudes and implicit times in normal and WD CMs and RMs. Significant decreases in the amplitudes of a-and b-waves were observed in WD animals when compared to their normal peers. In contrast, significant increases in implicit times were observed for a-and b-waves in WD CMs and b-waves in WD RMs (Figure 7).
In the present study, numerous WDs were observed in the macula of 13 out of 28 female CMs and 11 out of 25 female RMs. The numbers of WDs varied from 10 to 500 per eye in all WD-positive animals, and although the WDs were spread across the retinal vascular arcade, they were concentrated mainly near the fovea centralis. While CMs showed no significant relationship between age and WD count, RMs exhibited a significant relationship. One possible reason for this discrepancy in the results could be that CMs encompassed a broader age range, whereas RMs were more skewed toward a restricted age bracket. In the OCT images of eyes with WDs, slightly hyperreflective lesions were scattered throughout the RPE. Furthermore, a large quantity of hyperreflective granules, approximately 10 µm in size, was noted between the EZ and RPE. OCT uses low-coherence near-infrared light to calculate wave intensity and positional information to produce a tomographic image. In OCT images of the retina, layers of the retina that reflect more near-infrared light appear more hyperreflective [15]. Hyperreflection in OCT images can be associated with various causes, including migrating RPE cells, lipid-laden macrophages, microglial cells, and extravasated proteinaceous or lipid material [16]. Furthermore, anatomical structures in the outer retina, such as specific bands and the movement of organelles like melanosomes within the RPE, can contribute to such hyperreflective signals [17-19]. Therefore, the hyperreflective regions and dots observed in the present study may be attributed to such changes. The position of the hyperreflective granules between the EZ and RPE in the OCT images did not align with that of WDs noted in the ocular fundus images in both normal and WD animals. The scatter plot comparing hyperreflective granules with WDs indicated significant variability. Nevertheless, the Spearman coefficient indicated a positive correlation for both CMs and RMs. While this points to a general trend, the variation in individual data points suggests the influence of diverse factors, meriting further study. In our study, histopathological findings from a single specimen offered valuable insights into the potential degradation processes of RPE cells associated with WDs. Although based on just one sample, the observed vacuolization in the RPE, which aligns with the hyperreflective lesions noted in OCT images, is of particular interest. The TEM analysis further identified structures reminiscent of lysosomes within the RPE. The existence of both membrane-and nonmembrane-bound organelles, including mitochondria inside these vesicles, indicates a potential degenerative process within the RPE cells. Such phagolysosomal activities might represent the RPE cells’ efforts to clear cellular debris or damaged organelles.
Many previous studies have reported WDs in the ocular fundus of elderly CMs and RMs [6,20-26], such as reports of lipid degradation in RPE cells [27,28], where numerous ocular fundus lesions similar to the WDs noted in the present study were observed. However, since each WD is extremely small and details are difficult to identify in OCT images [25], it is not clear whether WDs are actually a sign of degradation in RPE cells; the pathological significance is also unclear. These results suggest that, in the present study, the WDs observed in fundus examination are evidence of degradation in RPE cells and that these changes can be identified in OCT images by looking for slight contrast changes in the RPE. However, while the hyperreflective granules between the EZ and RPE were observed in many OCT images of WD animals, pathological examination failed to reveal any details. In this study, we limited our pathological examination to a representative case, primarily due to ethical considerations aimed at minimizing animal euthanasia. This limitation indicates the potential constraints on the depth of our findings. Although our dataset may not allow for absolute conclusions, the observable trend indicates that hyperreflective granules are markedly more prevalent in WD animals than in their normal counterparts. This underscores the potential of hyperreflective granules being used as an indicative marker for assessing the RPE condition.
The aging of RPE cells—influenced by factors such as the accumulation of amyloid beta, oxidative stress, and inflammation, among other multifactorial triggers—has been proposed as a potential contributor to the progression of AMD [29,30]. In the CCR2 knockout mouse, which is a recognized model for AMD, accelerated aging of RPE cells and pronounced vacuolization of the RPE have been observed [31,32]. This lends credence to the idea that vacuolization in RPE cells signifies their senescence.
In this study, in normal CMs and RMs, the RPE and choroid were significantly thicker in adults than in young adults. Furthermore, the RPE and choroid were consistently thicker in all WD animals than in their normal counterparts. These observations suggest that RPE and choroidal thickening is associated with aging or WDs. However, little correlation was observed between RPE or choroidal thickness and the quantity of hyperreflective granules identified during the OCT examinations. The increased RPE thickness might be linked to the vacuolization observed in the RPE cells of typical WD animals, although the precise cause could not be determined in the present study. While the hyperreflective granules detected through OCT could not be identified during histopathological evaluations, their existence could indirectly signify alterations in the surrounding milieu that might affect RPE thickness. The precise dynamics of this relationship are yet to be elucidated. In conclusion, the present study posits that both aging and the presence of WDs could contribute to an increase in RPE thickness within the macula.
In the f-ERG measurements, significant decreases in a-and b-wave amplitudes were observed in both WD CMs and RMs. Meanwhile, implicit times of the a-and b-waves tended to increase in both WD CMs and RMs, although the increase in the a-wave in RMs was not statistically significant. Considering that photoreceptors provide input to bipolar cells, the observed changes are likely attributable to alterations in the photoreceptor function. Moreover, WDs in the macula have been shown to be associated with photoreceptor cells. The amplitude is directly related to the overall activity of these cells. A decrease in amplitude typically indicates a reduction in cell activity or count, yet no discernible change was noted in the OCT evaluations of the WD animals or in the histopathological examinations. Although the mechanisms behind decreased amplitude and increased implicit time remain unclear, an impact on the visual function is possible, even if there are no severe changes leading to visual defects.
In NHPs, there are no reported instances of dry AMD. However, both humans and NHPs are known to develop drusen and drusenoid lesions characterized by an ultrastructure of accumulated lipid particles [20,23]. An increase in the prevalence of these lesions with age has been documented [22]. Subretinal drusenoid deposits (SDDs), originally identified as reticular pseudodrusen, are considered precursors of AMD. They are more prevalent in the elderly, especially among those with thinner choroids, and are more common in women than in men [33]. In OCT, early-stage SDDs primarily manifest in the retinal layers between the RPE and EZ, appearing as granular hyperreflective granules. In later stages, these granules forms small mounds [34,35]. While the WDs observed in the present study did not resemble the SDDs reported in humans and no drusen were detected, the identification of hyperreflective granules between the EZ and RPE in OCT images, coupled with the pathological vacuolization of the RPE cells in the macula, suggests that animals with WDs are at risk of developing macular degeneration in the future. Thus, these animals could serve as models for AMD precursor lesions. Prior research has shown an increased occurrence of drusen-like fundus images in RMs aged 17 years and above compared to those between the ages of 10 and 16, which indicates an age-associated progression of these features [22]. In our study, the oldest RM was 14 years old, and there were no animals older than 17 years. For CMs, our examination was limited to those aged up to 17 years. Continued longitudinal studies in both species are expected to shed more light on the evolution of WDs and drusen. To determine the relationship between WDs and drusen formation and to evaluate their potential as a model for AMD precursor lesions, it is crucial to continue monitoring the animals from the present study for an additional 5–10 years.
We are grateful to N. Yotsumoto (SNBL DSR) for assistance with the TEM. Financial support: This work was supported by Shin Nippon Biomedical Laboratories Ltd. (SNBL).