Acute neuronal apoptosis in a rat model of multiple sclerosis

Abstract

Demyelination caused by inflammation of the CNS has been considered to be a major hallmark of multiple sclerosis (MS). Using experimental autoimmune encephalomyelitis, a model of MS, we demonstrate that an immune-mediated attack of the optic nerve is accompanied by an early degeneration of retinal ganglion cells (RGCs). The decrease of neuronal cell density was correlated with functional disabilities as assessed by visual evoked cortical potentials and electroretinogram. Visual acuity was significantly reduced. DNA degradation and activation of caspase-3 in RGCs indicate that cell death of RGCs is apoptotic. These findings show for the first time that an inflammatory attack against myelin components can lead to acute neuronal cell loss by apoptosis.

Fig. 1.

Lesional pathology in EAE (b,d, e, g, h) compared with healthy rats (a, c,f). Serial longitudinal optic nerve sections were stained with Luxol fast blue (a, b), ED1 (immunocytochemistry for activated macrophages and microglia;c–e), and Bielschowsky silver impregnation for axons (f–h). On the left-hand side, the healthy optic nerve stained for Luxol fast blue (a), ED1 (c), and Bielschowsky silver impregnation (f) is depicted. b, d, e, g, and h show a mainly inactive demyelinated plaque with some ongoing demyelinating activity (b,blue spots). At the edge of the plaque (left-hand side of b, d, g), some healthy Luxol fast blue-stained fibers remain unaffected (b), whereas the center of the plaque is completely demyelinated (b). The demyelinated area is heavily infiltrated by macrophages (d, e). Bielschowsky silver impregnation (g) reveals reduced density of apparently normal axons at the plaque edge. In the plaque center, density of axons is markedly reduced, and swollen and distorted axons (h) can be detected. Magnification: a–d, f,g, 300×; e, h, 700×.

Fig. 2.

VECP and ERG responses in MOG-EAE and controls.a, VECP (left panel) and ERG (right panel) recordings in response to a flash-light indicated by arrows. The top traces represent a typical measurement from a control rat, and the bottom traces represent an rrMOG^Igd-immunized BN rat with optic nerve demyelination. In diseased rats with MOG-EAE, no VECP signals could be recorded, but ERG was almost unchanged compared with control.b, Pattern VECP (left panel) and pattern ERG (right panel) traces recorded from a control BN rat (top traces) and from a BN rat with optic nerve demyelination (bottom traces) at varying spatial frequencies as marked by numbers are shown. The stimulating pattern is indicated on top of the recordings. Amplitudes of VECP and ERG decreased with increasing spatial frequencies. From the rat with optic neuritis, no VECP signals (left traces) in synchrony with the visual stimuli were obtained, and only ERG signals (right traces) with smaller amplitudes compared with controls could be detected. Traces at the bottom represent experiments with no visual stimulus to determine the noise level. c, Amplitudes of ERG and VECP recordings were plotted versus the spatial frequency. Open symbols mark data recorded from a control rat, andfilled symbols mark data determined from a rat with optic neuritis. Visual acuities were evaluated by extrapolating amplitudes determined from VECP (right panel) and ERG (left panel) measurements to the noise level. Averaged data from three measurements with SEM are shown.

Fig. 3.

RGC density is reduced in BN rats with manifest optic neuritis. a, b, Representative whole-mount areas at 3/6 retinal radius from a control rat (a) and from an rrMOG^Igd-immunized rat with optic neuritis (b) are shown. The numbers of FG-labeled RGCs are significantly reduced in BN rats with optic neuritis. c,The decrease of RGC density in rats with optic neuritis correlates with a reduction of visual acuity determined from ERG measurements. Visual acuities and RGC densities from rats with optic neuritis (filled circles) are reduced compared with controls (filled triangle). rrMOG^Igd-immunized rats without optic neuritis (open squares) had similar visual acuities as controls, but RGC densities revealed a broader variability. Thecurved line marks the trend of all shown data points.

Fig. 4.

Cell death of RGCs is accompanied with DNA degradation and caspase activation indicating apoptotic neuronal death.a, b, Double staining of a representative retina section from a BN rat with optic neuritis. FG labeling (a) colocalizes with TUNEL staining (b), indicating apoptosis in RGCs. Large arrows mark identical cells. c, d, Detection of activated caspase-3 (d) by immunohistochemistry in RGCs identified by FG labeling (c). Thesmall arrow marks an FG-labeled cell with no caspase-3 staining. Scale bars: a, b, 100 μm; c, d, 70 μm.