Activation of the caspase 8 pathway mediates seizure-induced cell death in cultured hippocampal neurons

Abstract

In response to harmful stresses, cells induce programmed cell death (PCD) or apoptosis. Seizures can induce neural damage and activate biochemical pathways associated with PCD. Since seizures trigger intracellular calcium overload, it has been presumed that the intrinsic cell death pathway mediated by mitochondrial dysfunction would modulate cell death following seizures. However, previous work suggests that the extrinsic cell death pathway may initiate the damage program. Here we investigate intrinsic versus extrinsic cell death pathway activation using caspase cleavage as a marker for activation of these pathways in a rat in vitro model of seizures. Hippocampal cells, chronically treated with kynurenic acid, had kynurenic acid withdrawn to induce seizure-like activity for 40 min. Subjecting rat hippocampal cultures to seizures increased cell death and apoptosis-like DNA fragmentation using TUNEL staining. Seizure-induced cell death was blocked by both MK801 (10 microM) and CNQX (40 microM), which suggests multiple glutamate receptors regulate seizure-induced cell death. Cleavage of the initiator caspases, caspase 8 and 12 were increased 4h following seizure, and cleavage of the quintessential executioner caspase, caspase 3 was increased 4h following seizure. In contrast, caspase 9 cleavage only increased 24h following seizure. Using an affinity labeling approach to trap activated caspases in situ, we show that caspase 8 is the apical caspase activated following seizures. Finally, we show that the caspase 8 inhibitor Ac-IETD-CHO was more effective at blocking seizure-induced cell death than the caspase 9 inhibitor Ac-LEHD-CHO. Taken together, our data suggests the extrinsic cell death pathway-associated caspase 8 is activated following seizures in vitro.

Figure 1

Kynurenic acid withdrawal-induced seizure elicits time-dependent cell death. a) Withdrawal of kynurenic acid from cultures chronically treated with kynurenic acid and Mg²⁺ results in seizure-like activity. b) Increasing seizure duration increases the amount of LDH released from neurons. Note seizure duration greater that 5 min results in maximal cell death. Media samples were taken 24 hours following seizure for LDH assay. c) Seizure induced cell death is reduced by the NMDA antagonist MK801 (10 μM) and the AMPA antagonist CNQX (20 μM). Cells were incubated with drug during 40 min seizure period. Data shown are mean ± sem (n= 5), * denotes P<0.01 vs. effect of seizure (post hoc Dunnet's test).

Figure 2

Increased DNA damage following seizure. a) Control or seizure-treated cells were recovered for 24 hours and DNA damage assessed by terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL). Cells were counterstained with DAPI to show nuclear morphology. Note the clumped nuclear morphology in seizure treated cells, inset 40 × magnified view of nucleus of seizure treated cell. Scale bar = 20 μm.

Figure 3

Caspase cleavage following seizures. a) Assessment of fraction purity. Fractions were blotted for caspase 12, cytochrome oxidase IV, Bad and lamin-A as specific markers for microsomal (ER), mitochondrial (mito) cytoplasmic (cyto) and nuclear (nuc) fractions. b) Increased cleavage of caspase 8 and its substrate BAP31 4 hours following seizure. Cytosol fractions were prepared and immunoblotted for cleaved caspase 8 and BAP31. c) Increased cleavage of caspase 12 following seizure. Microsomal fractions were prepared and immunoblotted for cleaved 12. d) Caspase 9 cleavage increase 24 hours following seizure. Cytosol fractions were prepared and immunoblotted for cleaved caspase 9. e) Increased nuclear caspase 3 cleavage following seizures. Nuclear fractions were prepared and immunoblotted for cleaved caspase 3. Caspase 7 cleavage in microsomal fractions did not increase following seizure. Images shown are representative blots of duplicate experiments which gave similar results. Numbers at the side of the blots represent molecular weight in kDa.

Figure 4

Immunoprecipitation of active caspases reveals caspase 8 is activated following seizure. Cells were incubated with biotin-zVAD-fmk (1 μM) and subject to seizures and 4 h recovery. Bound caspases were precipitated with avidin-agarose and immunoblotted for caspase 2, 3,8 9 and 12. As a positive control cortical brain lysate from 100 min MCAO treated rats was used (see (Meller, et al. 2005)). Note lack of cleaved caspase 2 9 and 12, and the increase in cleaved caspase 8, 4 hours following seizure. Caspase 3 was also cleaved following seizure. Images shown are representative blots from 3 separate experiments with identical results. Numbers at the side of the blots represent molecular weight in kDa.

Figure 5

Blockade of caspase 8 reduces seizure-induced cell death. Cells were incubated with the caspase 8 inhibitor Ac-IETD-CHO or caspase 9 inhibitor Ac-LEHD-CHO (both 10 μM) for 10 min prior to seizure, and then for 24 hours following 40 min seizure. Cell death was assessed by LDH release assay. Data shown are mean ± sem (n= 4), * denotes P<0.01 vs. effect of seizure (Dunnet's post hoc test), ** denotes P<0.01 vs. non-seized control.