Introduction

The world’s population is aging. More and more countries are facing age-related health problems, among them is vision deterioration. Declining vision has been shown to have a detrimental impact on an individual’s quality of life by limiting mobility, visually intensive tasks and independence1,2,3. The absence of mobility and self-care reduces social activities and activities of daily living and therefore can lead to social isolation4,5. In addition, visual impairment is associated with an increased risk of accidents, injuries and even mortality6,7. Hence, there is an urgent need to understand age-related visual declination.

Rodents are the commonly used animal species to investigate the physiological aging process or to identify possible therapeutic targets. We and others have utilized several mouse models of glaucoma, an eye disease characterized by progressive degeneration of retinal ganglion cells (RGCs) and their axons, such as DBA/2J mice8 and glutamate transporter knockout mice (GLAST and EAAC1 knockout mice)9. These animal models have been useful in examining potential therapeutic targets10,11. However, due to anatomical differences and their genetic distance to humans, translation of findings is sometimes complicated.

The common marmoset (Callithrix jacchus), a non-human primate with structurally well-developed brain and eyes, is becoming increasingly attractive as an experimental animal model, particularly in neuroscience research12. Their compact lifespan (10–15 years, with a maximum lifespan of approximately 20 years) allows monitoring of the effects of aging or progressive disease over a relatively short period of time, from 8 years of age onward13, making them ideal for aging research. We previously examined the eyes of 36 aged marmosets and found that 11% of them presented with spontaneous normal tension glaucoma (NTG)14. As a step toward aging study in vision using non-human primate marmosets, in the present study, we studied the eyes of aged marmosets without spontaneous NTG non-invasively and compared them with young marmosets. We found that mild retinal degeneration occurred in aged marmosets.

Results

Deteriorated retinal response in aged marmosets

Marmosets used in Figs. 1, 2, 3, 4 and 5 were summarized in Table 1. Fundus photographs of all marmosets were taken and there was no difference between young (Y1–Y5) and aged marmosets (A1–A4) (Fig. 1). Hallmarks of glaucoma such as increased excavation and thinning of the neuroretinal rim were not observed in all the marmosets, indicating that these marmosets are not glaucomatous marmosets. We then examined visual function using multifocal electroretinogram (mfERG). We analyzed the second-order kernel (2 K) component, which appears to be a sensitive indicator of inner retinal dysfunction and is impaired in glaucoma patients15 and marmosets14. The response topography demonstrating the 2 K component revealed that the average visual responses were declined in aged marmosets compared with young marmosets (Fig. 2a). Quantitative analysis revealed that retinal response was 3.3 ± 0.2 in young marmosets, and 2.8 ± 0.1 in aged marmosets (Fig. 2c). We concomitantly assessed the first-order kernel (1 K), which primarily reflects the outer retinal activity (Fig. 2b)16, and found that the visual response was reduced from 7.2 ± 0.2 in young marmosets to 6.2 ± 0.4 in aged marmosets (Fig. 2d).

Fig. 1
figure 1

Evaluation of ocular fundus images of young and aged marmosets. The optic cup/disc ratios of all young (Y1-5) and aged (A1-4) marmosets were < 0.5. No glaucoma-like characteristics were found in all the marmosets.

Full size image
Fig. 2
figure 2

Reduced retinal response in aged marmosets. (a, b) Representative images of three-dimensional plots depicting averaged visual responses of the second-order kernel (a) and first-order kernel (b) examined by multifocal electroretinogram (mfERG). A higher score (white) indicates highly sensitive visual function, and a lower score (green) indicates retinal dysfunction. Values are given in nV per square degree (nV/deg2). (c, d) Quantitative analysis of the retinal responses of second-order kernel (c) and first-order kernel (d). n = 10 from both eyes of 5 young marmosets (Y) and 8 from both eyes of 4 aged marmosets (A). *P < 0.05 (Mann–Whitney U test). Data are expressed as the mean ± S.E.M.

Full size image
Fig. 3
figure 3

Impaired implicit times and amplitudes of retinal responses in aged marmosets. (a) Representative waveforms of mfERG recordings showing the second-order kernel (2K) components from marmosets. Arrows indicate the P1, N1, and P2 waves. (b) Quantitative analysis of implicit times for the P1, N1, and P2 waves in young and aged marmosets. (c) The visual stimuli consisted of 61 hexagonal regions, and the resulting responses were grouped into five concentric rings for analysis. (d) Quantitative analysis of the N1 wave amplitudes in each concentric ring (rings 1 to 5). n = 10 from both eyes of 5 young marmosets and 8 from both eyes of 4 aged marmosets. *P < 0.05, ***P < 0.001 (Mann–Whitney U test). Data are expressed as the mean ± S.E.M.

Full size image
Fig. 4
figure 4

Impaired implicit times and amplitudes of the first-order kernel components in aged marmosets. (a) Representative waveforms of mfERG recordings showing the first-order kernel (1K) components from marmosets. Arrows indicate the N1 and P1 waves. (b) Quantitative analysis of implicit times for the N1 and P1 waves in young and aged marmosets. (c) The visual stimuli consisted of 61 hexagonal regions, and the resulting responses were grouped into five concentric rings for analysis. (d) Quantitative analysis of the P1 wave amplitudes in each concentric ring (rings 1 to 5). n = 10 from both eyes of 5 young marmosets and 8 from both eyes of 4 aged marmosets. *P < 0.05 (Mann–Whitney U test). Data are expressed as the mean ± S.E.M.

Full size image
Fig. 5
figure 5

Retinal degeneration in aged marmosets. (a) Fundus photo of a marmoset. The fundus was divided into the superior (S), inferior (I), temporal (T), and nasal (N) quadrants, with the optic disc at the center. (b) In vivo imaging of the retina. GCC: ganglion cell complex. (c) Quantification of the thickness of GCC in the temporal (T), superior (S), nasal (N), inferior (I) and temporal (T) retina measured by OCT. (d) Quantification of the averaged thickness of GCC in the whole measured retina region. (e) Quantification of the averaged thickness of the outer retina measured by OCT. n = 10 from both eyes of 5 young marmosets in (c), (d) and (e). n = 8 from both eyes of 4 aged marmosets in (c) and (d), and 6 from both eyes of 3 aged marmosets in (e). *P < 0.05, **P < 0.01 (Mann–Whitney U test). Data are expressed as the mean ± S.E.M.

Full size image
Table 1 Summary of marmosets used in Figs. 1, 2, 3 and 4.
Full size table

We also analyzed the implicit times and amplitudes of the 2 K (Fig. 3a). The implicit times of the first positive wave (P1), the first negative wave (N1), and the second positive wave (P2) were significantly prolonged in aged marmosets (Fig. 3b). As for the concentric analysis from the fovea to the peripheral (Fig. 3c), the N1 amplitude was notably reduced in the peripheral retina (ring 4 and 5 in Fig. 3d), but not in the center region (ring1 in Fig. 3d). Consistently, the implicit times of N1 and P1 of the 1 K were significantly prolonged in aged marmosets (Fig. 4a,b). And the P1 amplitude derived from the concentric analysis was notably reduced in the peripheral retina (Fig. 4c,d). These results suggest that the retinal peripheral region is more susceptible to aging.

Retinal degeneration in aged marmosets

To determine how the deterioration in retinal response found in aged marmosets is related to the retinal structure, we examined the retinal structure by using OCT, an invasive in vivo imaging method for the retina. For quantitative analysis of the inner retina, the ganglion cell complex [GCC; between the internal limiting membrane and the interface of the inner plexiform layer and the inner nuclear layer] was measured by scanning the retina in a circle centered around the optic nerve disc (Fig. 5a), and the average GCC thickness was determined from the acquired images (Fig. 5b). Quantitative analysis of the GCC thickness showed that the thickness of the GCC layer was deteriorated significantly with age in most quadrants, excluding the inferior region (Fig. 5c). And the averaged thickness of the GCC layer for the entire retinal region was significantly reduced with age (Fig. 5d). However, no significant age-related changes were found for the thickness of the outer retinal layer (Fig. 5b,e).

Increased oxidative stress in aged marmosets

It has been well known that oxidative stress plays an important role in the development of age-related diseases17,18. To elucidate the possible mechanism of retinal degeneration observed in aged marmosets, we investigated the expression of 4-hydroxy-2-nonenal (4-HNE), one of the major mediators of oxidative stress in cells and tissues, in aged marmosets. Serums from young (Y6-Y10) or aged (A5-A8) marmosets were prepared (Fig. 6a) and the 4-HNE expression was investigated by Western blot analysis (Fig. 6b). Quantitative analysis revealed that 4-HNE expression was significantly increased in the serum of aged marmosets compared with young marmosets (~ 2.6 times, Fig. 6c). We also measured the GCC thickness in young and aged marmosets. We found that the GCC thickness was significantly reduced in aged marmosets compared to young marmosets (Fig. 6d). We then tried to clarify the relationship between 4-HNE levels and the GCC thickness. However, we found that there was no significant correlation between the two variables.

Fig. 6
figure 6

Increased oxidative stress in the blood of aged marmosets. (a) Summary of young (Y6–Y10) and aged (A5–A8) marmosets whose serum were used for 4-HNE detection by immunoblot analyses. (b) Immunoblot images of 4-HNE expression in the blood of marmosets summarized in (a). (c) Quantification of 4-HNE expression in (b). n = 5 (Y) and 4 (A). (d) Quantification of the thickness of GCC. n = 10 from both eyes of 5 young marmosets (Y) and 8 from both eyes of 4 aged marmosets (A). *P < 0.05, ***P < 0.001 (Mann–Whitney U test). Data are expressed as the mean ± S.E.M.

Full size image

Discussion

In the present study, we examined the eyes of young and aged marmosets non-invasively. We found that the retinal response along with the retinal structure deteriorated in aged marmosets. In addition, an age-related decline in the 1 K component and an increase in the implicit times of N1 and P1 were found in aged marmosets, which are consistent with previous studies19,20. The 2 K component, a sensitive indicator of RGC loss, was also declined in aged marmosets (Fig. 2). In agreement with our results, several studies have reported RGC loss in aged humans and rodents21,22,23. For example, Harmon et al.21 demonstrated that the average density of RGCs per retina decreased significantly with age although values did not fall significantly in the macula. These results seem to resemble with our concentric analysis results that the amplitudes of N1 declined significantly in the peripheral regions but not in the center regions (Fig. 3d). However, in contrast to these studies, several groups have reported that the ganglion cell population in rodents or primates does not change with age24,25,26. In addition, Nadal-Nicolás et al.27 recently showed that various cell populations in the marmoset retina such as microglia, photoreceptors, bipolar cells, amacrine cells and RGCs do not alter by aging. The discrepancy between this literature and ours may reflect the different detection methods used. We used OCT for in vivo analysis of the GCC thickness and mfERG for detecting retinal responses. Our approach might be more sensitive than histological methods. It will be interesting to investigate whether RGC loss occurs in the retina of aged marmosets after the thinning of the GCC.

Age-related thinning of the peripapillary inner retinal layers is well-documented in humans28, with more pronounced thinning in the superior region compared with the inferior region. In this study, we non-invasively identified a similar pattern in marmosets (Fig. 5c), suggesting a conserved age-related change across species. These changes were more sensitive than full-thickness measures, aligning with mouse and human data29,30. The superior-inferior asymmetry in age-related thinning has not been reported in mice, suggesting it may be unique to primates with a macula28.

Another finding of the present study is that 4-HNE was significantly increased in the serum of aged marmosets accompanied with reduced GCC thickness. However, the correlation between 4-HNE levels and the GCC thickness was not significant, which might be due to the small sample size in the present study. It is worth investigating whether there is a significant correlation between 4-HNE levels and the GCC thickness using large sample sizes in the future study.

4-HNE with its ability to form protein adducts and to propagate oxidative stress, has been found to be elevated in the retina, brain tissue and body fluids of various neurodegenerative diseases such as NTG, Alzheimer’s disease and amyotrophic lateral sclerosis14,31. Recently, Sato et al. developed the PSEN 1 mutant marmosets, which has been reported as the causative gene of familial Alzheimer’s disease32. Our current data would be available as control data for these PSEN 1 mutant marmosets. Therefore, oxidative stress is a good therapeutic target for age-related retinal degeneration. We previously reported that edaravone, a free radical scavenger, prevented retinal degeneration in mice following optic nerve injury and in EAAC1 knockout mice, a mouse model of NTG33,34. It will be interesting to investigate whether edaravone can prevent age-related vision loss in marmosets.

A proposed cause of aging is the accumulation of epigenetic noise that disrupts gene expression patterns, which leads to decreases in tissue function and regenerative capacity35,36. A recent study demonstrated that AAV2-medicated ectopic expression of Oct4, Sox2 and Klf4 genes (OSK) in mouse RGCs reversed vision loss in aged mice through restoring youthful DNA methylation patterns and transcriptomes37. Moreover, we recently developed a system that forces membrane localization of the intracellular domain of tropomyosin receptor kinase B (TrkB) by farnesylation (F-iTrkB)38. Overexpression of F-iTrkB results in constitutive activation of downstream signaling pathways in the absence of its ligand BDNF. Using AAV2-mediated gene therapy in the eyes, we demonstrated that F-iTrkB expression enhances neuroprotection in mouse models of glaucoma. It will be interesting to examine whether these gene therapies could restore vision in aged marmosets in the future.

In conclusion, our study demonstrated that marmosets offer a promising translational model for the research of age-related vision declination.

Methods

Animals

Experiments were performed in common marmosets (Callithrix jacchus), which were derived from breeding colonies at the Tokyo Metropolitan Institute of Medical Science (TMiMS) and Central Institute for Experimental Medicine and Life Science (CIEM). Animal experiments were approved by the Institutional Animal Care and Use Committee of TMiMS (Approval number: 15032) and CIEM (Approval number: AIA240014). All animals were handled with care and all animal experiments adhere to the ARRIVE guidelines and were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the guidelines for the care and use of animals at both institutes. Experiments were performed under anaesthesia as previously reported14,39. Ketamine hydrochloride plus xylazine (30 mg/kg and 1.2 mg/kg respectively, i.m.) were used for sedation followed by sodium thiopental (15–30 mg/kg, i.p.) for anaesthesia. Sodium thiopental is a rapid-onset short-acting barbiturate general anaesthetic.

Fundus camera investigation

Marmosets were anaesthetized and the pupils were dilated with 0.5% phenylephrine hydrochloride and 0.5% tropicamide. Ocular fundus photographs were obtained using a small animal fundus camera (Genesis-D, Kowa, Tokyo, Japan).

mfERG

Marmosets were anaesthetized and the pupils were dilated with 0.5% phenylephrine hydrochloride and 0.5% tropicamide. mfERGs were recorded using a VERIS 6.0 system (Electro-Diagnostic Imaging, Redwood City, CA, USA). The visual stimulus consisted of 61 hexagonal areas scaled with eccentricity and the stimulus array was displayed on a high-resolution black and white monitor driven at a frame rate of 75 Hz. The resulting responses were analyzed by grouping them into five concentric rings. Furthermore, the 2 K, which are impaired in human and marmoset glaucoma14,40, were evaluated by analyzing the implicit times and amplitudes of the first and second positive waves and the first negative wave. Similarly, the 1 K parameters were analyzed using the same measures.

Imaging acquisition of SD-OCT

Marmosets were anaesthetized and the pupils were dilated with 0.5% phenylephrine hydrochloride and 0.5% tropicamide. SD-OCT (RS-3000, Nidek, Aichi, Japan) examinations were performed to monitor retinal degeneration in live marmosets14,41. A 40-D adaptor lens was placed on the objective lens of the Multiline OCT to focus on the marmoset retina. For imaging of the retinal layers, line scans and circular scans around the optic disc were performed, and the thicknesses of the GCC and outer retina [between the interface of the inner nuclear layer and the outer plexiform layer and the interface of the photoreceptor layer and the retinal pigment epithelium] were measured. In this study, the maximum number of B-scans set by the manufacturer (50 for line scans) was used for averaging.

Western blotting analysis

For immunoblot analysis, marmoset serums were prepared and added with SDS-PAGE loading buffer. Samples were subjected to immunoblot analysis using an antibody against 4-HNE (MHN-100P, JaICA, Shizuoka, Japan). Quantitative analysis was carried out using ImageJ version 2.14.042 and results were normalized to the total in-gel protein detected by Coomassie brilliant blue staining.

Statistics

For statistical comparison, Mann–Whitney U test was used. The error bars in all figures represent the mean ± S.E.M. P < 0.05 was regarded as statistically significant.