Reduced hippocampal damage and epileptic seizures after status epilepticus in mice lacking proapoptotic Puma

Abstract

The functional significance of neuronal death for pathogenesis of epilepsy and the underlying molecular mechanisms thereof remain incompletely understood. The p53 transcription factor has been implicated in seizure damage, but its target genes and the influence of cell death under its control on epilepsy development are unknown. In the present study, we report that status epilepticus (SE) triggered by intra-amygdala kainic acid in mice causes rapid p53 accumulation and subsequent hippocampal damage. Expression of p53-up-regulated mediator of apoptosis (Puma), a proapoptotic Bcl-2 homology domain 3-only protein under p53 control, was increased within a few hours of SE. Induction of Puma was blocked by pharmacologic inhibition of p53, and hippocampal damage was also reduced. Puma induction was also blocked in p53-deficient mice subject to SE. Compared to Puma-expressing mice, Puma-deficient mice had significantly smaller hippocampal lesions after SE. Long-term, continuous telemetric EEG monitoring revealed a approximately 60% reduction in the frequency of epileptic seizures in the Puma-deficient mice compared to Puma-expressing mice. These are the first data showing genetic deletion of a proapoptotic protein acting acutely to influence neuronal death subsequently alters the phenotype of epilepsy in the long-term, supporting the concept that apoptotic pathway activation is a trigger of epileptogenesis.-Engel, T., Murphy, B. M., Hatazaki, S., Jimenez-Mateos, E. M., Concannon, C. G., Woods, I., Prehn, J. H. M., Henshall, D. C. Reduced hippocampal damage and epileptic seizures after status epilepticus in mice lacking proapoptotic Puma.

Figure 1

Up-regulation of p53 in hippocampus during seizure-induced neuronal death. A, B) Photomicrographs show representative FJB staining at 72 h in hippocampus of vehicle-injected control mice (A) and mice that underwent SE (B). Arrows denote cells (white) undergoing degeneration. DG, dentate gyrus; H, hilus; CA1–3, cornu ammonis; CA3a-c, subfields of CA3. C) Photomicrograph (×20 lens) showing representative TUNEL staining in the ipsilateral CA3a/b subfield of a mouse 72 h after SE. D) Representative Western blot (n=4/lane) showing presence of cleaved caspase 3 (cl Casp3) after SE but not in control (Con). Tubulin (α-Tub) is shown as a guide to protein loading. E) Western blot (n=1/lane) showing increased levels of p53 after SE compared to control. Same blot was reprobed for actin as a guide to protein loading. F) Graph showing semiquantification of p53 protein levels. Data are from 10 independent experiments. A.U., arbitrary units. *P < 0.05 vs. control. Scale bars = 250 μm (A, B); 50 μm (C).

Figure 2

Puma is induced shortly after SE. A) Representative Western blot (n=1/lane) of hippocampal whole-cell lysates showing increased expression of Puma after SE. B) Graph showing semiquantitative analysis of Puma levels in hippocampus after SE (n=6/group). C) Marker analysis of hippocampal fractions. Cytoplasm (Cyt) contained only Bad, the nuclear fraction (Nuc) only lamin A/C, and the mitochondrial fraction (Mit) CoxIV. D) Representative Western blot (n=4/lane) showing increasing Puma levels in the mitochondrial fraction of seizure-damaged mouse hippocampus. CoxIV is shown as a loading control. E) Graph showing semiquantitative analysis of Puma levels in pooled mitochondrial fractions of hippocampus after SE (data from 2 independent experiments). F) Lysates from puma-expressing and puma^−/− HCT cells treated with 3 μM thapsigargin were blotted to confirm specificity of the Puma antibody. G) Graph showing noxa mRNA levels, corrected to β-actin, within hippocampus after SE in mice. Graph at right shows noxa expression in mouse cortical neurons treated with the proteasome inhibitor epoximicin (Epox; 50 nM) as a positive control (n=4/group). H) Graph showing a significant increase of the p53 target gene p21^WAF1/CIP1 mRNA, corrected to β-actin (n=4/group). A.U., arbitrary units. *P < 0.05 vs. control.

Figure 3

Puma induction after SE in mice is p53-dependent. A) Representative Western blots (n=1/lane) showing hippocampal Puma and p53 expression 8 h after SE in vehicle (Veh)- and pifithrin-α (PFT)-treated mice. Note that Puma levels are reduced in PFT-treated mice, while p53 levels were similar between groups. B) Graph showing significantly lower Puma levels in PFT-treated mice when compared to vehicle-treated mice after SE (n=3/group). C) Graph showing lower Puma levels in p53^−/− mice when compared to wild-type (wt) mice 8 h after SE (n=3/group). D–G) Graphs (D, F) and representative photomicrographs (×40 lens; E, G) showing hippocampal damage 24 h after SE as assessed by counts of FJB-positive CA3 cells (D, E), and CA3 cells with normal-appearing NeuN immunoreactivity (F, G), in PFT-treated mice compared to vehicle-treated mice (n=4–7/group). Note significantly fewer FJB-positive cells and significantly more surviving neurons in PFT-treated mice. A.U., arbitrary units. *P < 0.05 vs. control. Scale bar = 100 μm.

Figure 4

Reduced hippocampal damage after SE in puma^−/− mice. A) Graph and representative photomicrographs (×40 lens) showing significantly lower FJB counts in the ipsilateral CA3 in puma^−/− mice when compared to puma^+/− and wild-type (wt) mice. B) Graph and representative photomicrographs showing significantly lower TUNEL counts in ipsilateral CA3 from puma^−/− mice when compared to puma^+/− and wild-type mice. C) Graph and representative photomicrographs showing significantly more surviving NeuN-positive cells in ipsilateral CA3 from puma^−/− mice when compared to puma^+/− and wild-type mice. *P < 0.05 vs. wild-type and puma^+/− mice (n=5–7/group). Scale bar = 50 μm.

Figure 5

Milder epileptic phenotype and long-term neuroprotection in Puma-deficient mice. A) Representative EEG traces showing baseline EEG (top lane) and examples of typical epileptic seizures captured by EEG telemetry (between arrows) in puma^+/− and puma^−/− mice. B, C) Graphs depict daily epileptic seizure occurrence for puma^+/− (B) and puma^−/− (C) mice over 12 d of continuous seizure monitoring. D) Graph showing counts of NeuN-positive cells in CA3 72 h after epilepsy monitoring in puma^−/− mice when compared to puma^+/− (n=4–7/group). Note that NeuN counts in puma^−/− mice remained higher. E) Representative photomicrographs (×40 lens) of NeuN staining within the ipsilateral hippocampal CA3 region 15 d after SE in puma^+/− and puma^−/− mice. Arrows mark sites of neuronal loss. *P < 0.05 vs. control and puma^−/− mice. Scale bar = 100 μm.