Bim expression indicates the pathway to retinal cell death in development and degeneration

Abstract

Programmed cell death (PCD) during development of the mouse retina involves activation of the mitochondrial pathway. Previous work has shown that the multidomain Bcl-2 family proteins Bax and Bak are fundamentally involved in this process. To induce mitochondrial membrane permeabilization, Bax and Bak require that prosurvival members of the family be inactivated by binding of "BH3-only" members. We showed previously that the BH3-only protein BimEL is highly expressed during postnatal retinal development but decreases dramatically thereafter. The purpose of this study was to investigate a possible role for Bim, in retinal development and degeneration, upstream of Bax and Bak. Bim-/- mice analyzed for defective retinal development exhibit an increase in retinal thickness and a delay in PCD, thereby confirming a role for Bim. We also demonstrate that in response to certain death stimuli, bim+/+ retinal explants upregulate BimEL leading to caspase activation and cell death, whereas bim-/- explants are resistant to apoptosis. Finally, we analyzed Bim expression in the retinal degeneration (rd) mouse, an in vivo model of retinal degeneration. Bim isoforms, which decrease during development, are not reexpressed during retinal degeneration and ultimately photoreceptor cells die by a caspase-independent mechanism. Thus, we conclude that in cases in which BimEL is reexpressed during pathological cell death, developmental cell death pathways are reactivated. However, the absence of BimEL expression correlates with caspase-independent death in the rd model.

Figure 1.

Developmental apoptosis in the retina is delayed in bim^−/− mice. A, TUNEL staining in retinal sections from P2, P5, P8, and P15 mice demonstrates the pattern of cell death in wt versus bim^−/− mice. NBL, Neuroblastic layer. B, The number of TUNEL-positive cells in each field was counted and calculated as a percentage over the total number of cells in the same field. To determine the total number of cells, a gridline was established. Twenty percent of the squares in each retinal layer were selected at random, and the cells in each grid were counted. This value was extrapolated to the number of squares covering each layer in the entire field. Error bars represent ± SD (*p < 0.05; **p < 0.01). The ONL is not represented until P5 when the NBL separates, yielding the ONL and INL.

Figure 2.

bim^−/− retinas are thicker than their wt counterparts. A, Retinal sections from wt and bim^−/− mice at P2, P5, P13, and P60 were stained with an antibody reactive to cleaved caspase-9. Caspase-9 processing is detectable in wt and bim^−/− retinas, as indicated by the arrows. B, Thickness of the ONL, INL, and GCL in micrometers measured with a calibrated scale of the objective of the light microscope. NBL, Neuroblastic layer. Comparisons between bim^−/− and wt were performed using unpaired, two-tailed t test analysis. Error bars represent ± SD (*p < 0.05; **p < 0.01; ***p < 0.001).

Figure 3.

Bim-deficient retinal explants are resistant to calcium overload. A, TUNEL staining and immunohistochemistry for cleaved caspase-9 in retinal sections from untreated (unt) and calcium ionophore-treated (A23187; 5 μm) P10 and P15 explants after 24 h. P10 explants are more susceptible to cell death and caspase activation than P15 explants. B, Western blot analysis of Bim expression in untreated and treated (A23187; 5 μm) P10 and P15 retinal explants after 24 h. Blots were reprobed with GAPDH to demonstrate equal protein loading. C, D, TUNEL staining of retinal sections and Western blot analysis of Bim expression from untreated wt and A23187-treated wt, bim^+/− (+/−), and bim^−/− (−/−) explants after 24 h. E, Immunohistochemistry for cleaved caspase-9 in a retinal section from A23187-treated bim^−/− (−/−) explants.

Figure 4.

Bim-deficient retinal explants are also resistant to H₂O₂ and STS. A, TUNEL staining of retinal sections from untreated (unt) wt and H₂O₂-treated wt, bim^+/− (+/−), and bim^−/− (−/−) P15 explants after 24 h. B, TUNEL staining of retinal sections from untreated wt and STS-treated wt, bim^+/− (+/−), and bim^−/− (−/−) P15 explants after 24 h. C, For each stimulus, at least three animals were used, and the number of TUNEL-positive cells in three fields (40× magnification) per section of at least three different sections was counted. Error bars represent ± SD.

Figure 5.

Caspase-independent cell death in the rd mouse correlates with Bim downregulation. A, TUNEL staining (arrowheads) and immunohistochemistry for cleaved caspase-9 (arrows) in the rd mouse retina during development (P7 and P8) and degeneration (P12 and P14). B, Western blot analysis of Bim expression in the rd retina from P8 to P15. Blots were reprobed with GAPDH to demonstrate equal protein loading. C, Semiquantitative RT-PCR of bimEL in the rd retina with gapdh as a loading control. D, TUNEL staining and immunohistochemistry for cleaved caspase-9 in retinal sections from untreated (24 h) and A23187-treated (1 μm for 24 h) rd P12 explants. E, Western blot analysis of Bim expression in untreated and A23187-treated (1 μm for 24 h) rd retinas at P12.