Removal of clock gene Bmal1 from the retina affects retinal development and accelerates cone photoreceptor degeneration during aging

Abstract

The mammalian retina contains an autonomous circadian clock system that controls many physiological functions within this tissue. Previous studies on young mice have reported that removal of the key circadian clock gene Bmal1 from the retina affects the circadian regulation of visual function, but does not affect photoreceptor viability. Because dysfunction in the circadian system is known to affect cell viability during aging in other systems, we compared the effect of Bmal1 removal from the retina on visual function, inner retinal structure, and photoreceptor viability in young (1 to 3 months) and aged (24 to 26 months) mice. We found that removal of Bmal1 from the retina significantly affects visual information processing in both rod and cone pathways, reduces the thickness of inner retinal nuclear and plexiform layers, accelerates the decline of visual functions during aging, and reduces the viability of cone photoreceptors. Our results thus suggest that circadian clock dysfunction, caused by genetic or other means, may contribute to the decline of visual function during development and aging.

Keywords: aging; circadian rhythm; cone photoreceptor; development; retina.

Fig. 1.

Scotopic and photopic ERGs in Chx10^Cre;Bmal1^fl/fl and Bmal1^fl/fl mice. Scotopic and photopic ERGs were obtained from young (A–C) and old (D–F) mice. The scotopic and photopic b waves, but not a waves, were affected in Chx10^Cre;Bmal1^fl/fl mice compared with littermate Bmal1^fl/fl controls (P < 0.05; n = 5 to 8). The amplitudes of both scotopic and photopic ERGs decline with age (P < 0.01), and there was a trend for this decline to be more pronounced in Chx10^Cre;Bmal1^fl/fl mice compared with Bmal1^fl/fl mice.

Fig. 2.

The dendritic processes of rod bipolar cells are affected in Chx10^Cre;Bmal1^fl/fl mice. Synaptic connections between photoreceptor cells (anti-PSD95, red) and bipolar cells (anti-PKCα, green) in Chx10^Cre;Bmal1^fl/fl and Bmal1^fl/fl mice at the ages of 1, 3, and 26 mo. White asterisks show the presence of dendritic arborization in Bmal1^fl/fl mice; white arrows show the stunted dendritic arborizations in Chx10^Cre;Bmal1^fl/fl mice at all of the ages tested.

Fig. 3.

Quantification of SD-OCT. (A) A small but significant difference in the total thickness of the retina was detected between Bmal1^fl/fl and Chx10^Cre;Bmal1^fl/fl genotypes at both ages investigated (*P < 0.05 at 3 mo and ***P < 0.001 at 24 mo). (B) No difference in the thickness of PRL was observed between young Bmal1^fl/fl and Chx10^Cre;Bmal1^fl/fl mice, whereas a significant reduction in the thickness of the PRL was detected in older Chx10^Cre;Bmal1^fl/fl mice with respect to the Bmal1^fl/fl mice (***P < 0.001 at 24 mo).

Fig. 4.

Morphological evaluation of cone photoreceptors in young and old Bmal1^fl/fl and Chx10^Cre;Bmal1^fl/fl mice. Peanut agglutinin (PNA) labeling in the central retina of young (A) and old (B) Chx10^Cre;Bmal1^fl/fl mice. (C) Quantification of PNA-labeled cones of Bmal1^fl/fl and Chx10^Cre;Bmal1^fl/fl mice retinas at 3 and 26 mo of age. A significant change in the number of PNA-positive cells is observed between Bmal1^fl/fl and Chx10^Cre;Bmal1^fl/fl mice at 26 mo of age (***P < 0.001). Cone arrestin (Arr4) staining in young (D) and old (E) Chx10^Cre;Bmal1^fl/fl mice. (F) Quantification of cone arrestin staining in Bmal1^fl/fl and Chx10^Cre;Bmal1^fl/fl mice retinas at 3 and 26 mo of age. A significant change in the number of arrestin-positive cells is observed between Bmal1^fl/fl and Chx10^Cre;Bmal1^fl/fl mice at 26 mo of age (**P < 0.01). A significant reduction in the number of cones is also observed in both genotypes during aging (Bmal1^fl/fl, **P < 0.01; Chx10^Cre;Bmal1^fl/fl, ***P < 0.001). Each bar represents the mean ± SEM (n = 3 to 4).