A spontaneous nonhuman primate model of inherited retinal degeneration

Abstract

Inherited retinal degenerations (IRDs) are important causes of progressive, irreversible blindness. Hereditary macular diseases, in particular, are significant in their effect on the specialized, central cone photoreceptor-rich macula responsible for high resolution vision. Autosomal dominant Best vitelliform macular dystrophy (BVMD), caused by variants in the BEST1 gene, is one of the most common inherited macular dystrophies. Gene therapies have emerged as promising treatments for IRDs, but a lack of suitable animal models has hindered progress both in treatments and in understanding the mechanisms underlying macular diseases. Here, we report a Macaca fascicularis carrying a heterozygous potential pathogenic BEST1p.Q327E variant that disrupts the BEST1 ion channel by destabilizing the A195 helix, mirroring the structural perturbations seen in certain human pathological mutants. Longitudinal imaging over 2 years revealed progressive macular changes, including subfoveal cleft enlargement, lipid-rich deposit accumulation, retinal pigment epithelium (RPE) disruption, and central-to-peripheral photoreceptor degeneration, recapitulating early human BVMD pathology. Histopathology demonstrated diminished BEST1 expression, attenuation of the RPE-photoreceptor interface, and 2 distinct types of lipid deposits, including heretofore unappreciated cone mitochondrial-enriched lesions, highlighting selective cone mitochondria vulnerability. This is, to our knowledge, the first nonhuman primate model of inherited macular dystrophy, and it links BEST1 mutations, mitochondrial dysfunction, and progressive macular degeneration, offering new insights into BVMD pathophysiology and highlighting its utility for studying disease progression and potential therapeutic interventions.

Keywords: Genetic diseases; Genetics; Mitochondria; Ophthalmology; Retinopathy.

Figure 1. Identification of a macaque with a heterozygous deleterious BEST1p.Q327E variant.

(A) SD-OCT imaging reveals a small subfoveal cleft (as indicated by the yellow arrow) in the abnormal animal (6.8 years old), compared with the age-matched normal macaque (6.8 years old). (B) Fundus images from the mutant macaque did not show significant pathologic changes, including vitelliform lesion. (C) Sanger sequencing showed that the abnormal Macaca fascicularis carries a heterozygous BEST1p.Q327E variant. (D) The allele frequency of the BEST1:c.C979G: p.Q327E variant in Homo sapiens, Macaca mulatta, and Macaca fascicularis. (E) Structural modeling of the human Bestrophin-1 (hBest1) pentamer structure (PDB: 8D1K) showed that substitution of the glutamine (Q) side chain to glutamic acid (E) increases negative charge and enhances its electrostatic interactions with the main chain of K194 in the adjacent protomer. Q327/E327 in protomer 1 and adjacent protomer 5 are showcased with a protein electrostatic potential rendering, ranging from red (–50) to blue (50). (F) Current-voltage relationship of HEK293T cells transiently expressing WT or BEST1 variants, measured by whole-cell patch clamping for calcium-activated chloride ion conductance. The Q327E and A195V exhibited smaller currents compared with the WT. The mock group transfected with empty vectors showed the lowest current signal. Data are presented as mean ± SEM. n = 6–9. P < 0.001 for both Q327E and A195V, compared with WT mfBest1. P values were calculated using 2-way ANOVA. (G) Cotransfection of Q327E and WT BEST1 plasmids in a 1:1 ratio significantly reduced the current from WT BEST1. Data are presented as mean ± SEM. n = 6–10. P < 0.001 for Q327E+WT compared with WT mfBest1. P values were calculated using 2-way ANOVA.

Figure 2. The Q327-mutant macaque exhibited clinical manifestations of Stage 1 Best vitelliform macular dystrophy.

(A) SD-OCT scans of the mutant macaque and an age-matched control over 26 months of follow up. The mutant animal exhibited a progressively enlarging subfoveal cleft (area outlined in yellow) in both eyes. Deposits in the subfoveal cleft were also observed (white arrows). The first month SD-OCT image in A is the same as the mutant SD-OCT image shown in Figure 1A. (B) The subfoveal cleft areas from both eyes of the mutant macaque were enlarged during the 26 months of follow up. (C) Total retinal thickness of the mutant macaque decreased over time in both eyes, while that of the control remained unaltered. (D) The thickness of the parapapillary retinal nerve fiber layer remained unchanged in both the mutant and age-matched control macaque. (E) Blue light fundus autofluorescence (B-AF) examinations disclosed a slight increase in the size of the autofluorescence lesion in the right eye over 19 months from the seventh-month follow-up visit to the 26th-month follow-up visit. The autofluorescence lesion in the left eye was not as prominent.The red box indicates the enlarged area in the adjacent image. (F) Representative ERG waveforms of mutant animal and control, including scotopic 0.01, photopic 10.0, and the photopic negative response (PhNR). ONL, outer nuclear layer; B-AF, blue light fundus autofluorescence; ERG, electrophysiology.

Figure 3. Change in outer nuclear layer and RPE thickness over time in the Q327-mutant macaque.

(A) Schematic diagram showing the four directions and anatomical subdivisions of the retina (left panel) and associated segmentation (right panel) for analysis. The retina was divided into nasal (N), inferior (I), temporal (T), and superior (S), relative to the fovea (no. 1), parafoveal region (no. 2), and perifoveal region (no. 3). The OCT boundaries used to measure total retinal thickness, thickness of the ONL, and thickness of the RPE-IZ region, respectively, are denoted (right panel). In this study, the fovea was defined as a 1.0 mm diameter ring centered on the foveola, the parafovea was defined as an annulus centered on the fovea between circles with diameters of 1.0 mm and 2.22 mm, and the perifovea was defined as an annulus centered on the fovea between the circles with diameters of 2.22 mm and 3.45 mm. The position of the foveola was determined by manually determining the central fovea on the horizontal scans. (B and C) The baseline thickness of RPE-IZ (B) and ONL (C) in the first OCT examination from the mutant macaque and the age-matched control macaque. (D–G) The thickness change of RPE-IZ and ONL during the follow-up visits compared to the initial examination are displayed. RPE-IZ, the RPE-Bruch’s membrane (BM) complex and the RPE-photoreceptor interdigitation zone; ONL, the outer nuclear layer.

Figure 4. Histopathologic analysis reveals typical pathogenic features of BVMD.

(A) Bright field and immunostaining imaging showing neuroretina detachment from the RPE in the mutant fovea, forming a cleft with the accumulation of autofluorescent deposits within it. Autofluorescence was detected at 555 nm (red), and nuclei were stained with DAPI (blue). (B) Immunostaining for BEST1 protein (red) and RPE cell marker RPE65 (green), with nuclei stained with DAPI (blue). (C) DAPI staining highlights fragmented RPE nuclei in the mutant animal. Fragmented nuclei are indicated by red arrowheads, while normal nuclei are indicated by yellow arrowheads. Costaining with RPE65 was performed to identify the nuclei of RPE cells. (D) The density of RPE cells was significantly reduced in the fovea and parafoveal/perifoveal regions and unchanged in the periphery in the mutant macaque. Each data point represents one fluorescent image. (E) Immunostaining for EZRIN (pink) indicated the absence of RPE apical microvilli in the macular region of the mutant retina. (F) Immunostaining for MCT1 (green) showed decreased signals in the apical surface and processes of the RPE and photoreceptor outer segments in the mutant macaque macula. Statistical significance was determined by 2-way ANOVA and Šidák’s post hoc comparisons. (***P < 0.001).

Figure 5. The Q327 mutant macaque exhibits mitochondrial-negative and -positive lipid deposits in the macula.

(A–C) Representative images from immunostaining of BODIPY (green), TOM20 (red), and DAPI (blue) in different retinal regions, including fovea, parafovea/perifovea, and periphery. The bottom row shows zoomed-in views of the areas indicated by boxes in the top row. (A and B) In the fovea (A) and parafovea/perifovea (B), there was the accumulation of 2 types of BODIPY high deposits: deposits beneath the outer segments were TOM20 negative (indicated by yellow circles), while deposits above the outer segments were TOM20 positive (indicated by the white arrowheads). Mitochondria in the inner-segment region of photoreceptors, with low BODIPY signal, are outlined in white. A wavy distribution pattern of mitochondria was observed in the mutant animal. (C) The peripheral retina of the mutant animal showed a normally organized ellipsoid zone band with mitochondria, and deposits were rarely detected. TOM20, Translocase of outer mitochondrial membrane 20; EZ, ellipsoid zone; OS, outer segments; IS, inner segments.

Figure 6. Alteration of outer retinal bands of OCT scans during disease progression in the Q327 mutant macaque.

(A) Schematic representation of the structure of photoreceptor cells and RPE cells and their corresponding bands in the OCT image. ONL, outer nuclear layer; ELM, external limiting membrane; MZ, myoid zone of the photoreceptors; EZ, ellipsoid zone of the photoreceptors; OS, outer-segments; RPE-IZ, the RPE-Bruch’s membrane (BM) complex and the RPE-photoreceptor interface (interdigitation zone, IZ). (B and C) Longitudinal OCT scans show changes in the outer retinal bands (scanning locations indicated as red lines in the B-AF images on top) during follow-up imaging of the left (B) and right (C) eyes. The RPE-IZ shows early attenuation with time and the EZ showed a wave-like pattern following the RPE-IZ attenuation, with hyperreflective deposits appearing above the RPE-IZ band beneath the waves at some locations (yellow arrows). The schematic B-AF images are the same as the 26th month B-AF image shown in the Figure 2E. (D) Hypothesis for the early progression of Best disease. The RPE-photoreceptor interface is the first to be damaged, leading to cellular dysfunction. The accumulation of membrane discs beneath the OS and damaged cone mitochondria in the macula further exacerbates the damage. This results in the death of RPE and photoreceptor cells in the macula. Corresponding changes in OCT imaging include initial thinning of RPE-IZ, followed by distortion of EZ band, accumulation of deposits, and thinning of the ONL, progressing in a center-to-peripheral pattern in the macula.