To investigate the involvement of apoptotic proteins in excitotoxic neuron death, we recently established a model in which a transient, 5-min exposure to the glutamate receptor agonist NMDA (300 μM) induced a delayed cell death in primary cultures of rat hippocampal neurons 78. This excitotoxic cell death is characterized by mitochondrial cyt-C release and ΔΨm depolarization setting in 4 to 8 h after the NMDA exposure, followed by nuclear condensation and cell shrinkage 78. We determined if FL-Bid is able to translocate to mitochondria in this model of excitotoxic apoptosis. To this end, immunoblotting experiments were performed using a rabbit polyclonal antibody (AR-53) that detects p21 FL-Bid, as well as its caspase-8-/caspase-3-generated NH₂-terminal cleavage product 2930. Control experiments demonstrated that the antibody was able to detect FL-Bid and its caspase-generated NH₂-terminal p7 fragment in HeLa D98 cells after an activation of death receptors with TNF-α/Cycloheximide(CHX) (Fig. 1A, see also Additional File 1A). The antibody was also able to detect the p7 fragment in cultured rat hippocampal neurons (Fig. 1B) exposed to the apoptosis-inducing kinase inhibitor staurosporine (STS) or TNF-α/CHX (Fig. 1B see also Additional File 1B). However, neither a decrease in the content of FL-Bid, nor an accumulation of caspase-generated cleavage products could be detected in whole cell lysates of cultured rat hippocampal neurons exposed to NMDA (Fig. 1C, see also Additional File 1C). Activation of proteases after the NMDA exposure was however clearly detectable in the same whole cell lysates. This was evident from the decrease in the calpain I substrate full-length α-spectrin and the accumulation of calpain-specific 150 and 145 kDa α-spectrin cleavage products 4 and 8 h after the NMDA exposure (Fig. 1D).

Next, we performed selective plasma membrane permeabilization and subsequent immunoblotting of the cytosolic fraction and the mitochondria-containing pellet fraction in cultured rat hippocampal neurons exposed to NMDA for 5 min. Immunoblotting with a voltage dependent anion carrier (VDAC) antibody demonstrated that the cytosolic fraction was not contaminated with mitochondria. Immunoblotting with the Bid antibody revealed that p21 FL-Bid translocated from the cytosol to the mitochondria-containing pellet fraction 4 h and more pronounced 8 h after termination of the NMDA exposure (Fig. 2A). This process was paralleled by the translocation of cyt-C from the mitochondria-containing pellet fraction to the cytosol (Fig. 2A) 4 h and 8 h after the NMDA exposure. Interestingly, despite a significant cyt-C decrease in the mitochondrial fraction by 8 h, the cyt-C content in the cytosolic fraction did not increase correspondingly, suggesting that cyt-C may be degraded upon release into the cytosol (see also 31). Indeed, treatment with the membrane permeable cathepsin inhibitor CA-074 methyl ester recovered the cyt-C content in the cytosolic fraction 8 h after the NMDA exposure (Fig. 2A).

These results were confirmed by immunofluorescence analysis of Bid redistribution during excitotoxic neuron death using the above described Bid antibody. Neurons of sham-exposed control cultures exhibited a diffuse Bid immunofluorescence signal (Fig. 2B). Cyt-C co-staining revealed a filamentous, punctate staining pattern in sham-exposed controls characteristic of mitochondria. In contrast, cells that had released cyt-C in response to NMDA exhibited a clustered Bid immunofluorescence around the nucleus. Cells with released cyt C and clustered Bid immunofluorescence also exhibited nuclear chromatin condensation as evidenced by staining with the chromatin dye Hoechst 33258 (Fig. 2B).

We subsequently addressed the question, whether FL-Bid was sufficient to induce cell death in the hippocampal neuron cultures, and whether this cell death can occur in the absence of caspase-8-mediated Bid cleavage. Cultured rat hippocampal neurons were infected with adenoviral vectors expressing either FL-Bid or a Bid mutant that can not be cleaved by caspase-8 (D59A) under the control of a tetracycline responsive promoter 32. Western blot analysis of cultures infected with the adenoviral vectors (50 MOI) and induced for 24 h with 1 μg/ml doxycycline demonstrated a mild overexpression of FL-Bid in the hippocampal neuron cultures (Fig. 3A). This overexpression was however sufficient to induce a massive cell death in the hippocampal neuron cultures that was characterized by cell shrinkage and nuclear condensation. Hoechst staining of nuclear chromatin revealed significant neuronal damage 24 h after induction with doxycycline, reaching a level of 80.1 ± 2.7% (Fig. 3B). FL-Bid-induced cell death was associated with the mitochondrial release of the pro-apoptotic factors cyt-C and AIF (Fig. 3C and data not shown). In contrast, hippocampal neurons that were infected with the adenoviral vectors but were not induced with doxycycline remained viable for up to 24 h, as were cells that were treated with doxycycline but not infected with the adenoviral vectors (Fig. 3B). Interestingly, overexpression of Bid(D59A) also potently induced cell death in the hippocampal neuron cultures, reaching a level of 73.5 ± 3.1% after 24 h (p > 0.1, no significant difference compared to FL-Bid induced cultures).

N-Methyl-D-aspartate (NMDA), recombinant tumor necrosis factor-α (TNF-α), glycin and cycloheximide (CHX) came from Sigma (Poole, Dorset, U.K.). Tetrodotoxin was purchased from Biotrend (Cologne, Germany), CA-074 methyl ester (CA-074-ME) from Calbiochem (Bad Soden, Germany), and staurosporine (STS) from Alexis (Grünberg, Germany).

Cultured hippocampal neurons were prepared from neonatal (P1) Fischer 344 rats (Luetjens et al., 2000). Dissociated hippocampal neurons were plated at a density of 2 × 10⁵ cells/cm² into poly-L-lysine-coated 6- or 24-well plates or glass chamber slides (Nunc). Cells were maintained in MEM medium supplemented with 10% NU-serum, 2% B-27 supplement (50 × concentrate), 2 mM L-glutamine, 20 mM D-glucose, 26.2 mM sodium bicarbonate, 100 U/ml penicillin and 100 μg/ml streptomycin (Life Technologies, Karlsruhe, Germany). Experiments were performed on 14 – 16 day-old cultures. Animal care followed official governmental guide lines. HeLa D98 cells were cultured in RPMI 1640 medium (Life Technologies, Germany) supplemented with 10% fetal calf serum (PAA, Cölbe, Germany).

Cultures were exposed for 5 min to Mg²⁺-free Hepes-buffered saline (HBS) supplemented with 300 μM NMDA, 0.5 μM tetrodotoxin and 1 μM glycine 78. Control cultures were exposed to Mg²⁺-free HBS devoid of NMDA (sham exposure). Cell death was determined after 24 h by trypan-blue uptake 7. Experiments were performed on 14 or 15 day-old cultures.

Tetracycline (tet)-on inducible adenovirus vectors expressing wild-type FL-Bid or the D59A mutant of FL-Bid that can not be cleaved by caspase-8 were generated as described previously 32. Hippocampal neurons were infected at a MOI (multiplicity of infection) of 50 with both the FL-Bid or Bid(D59A) containing viruses and the reverse tet transactivator rtTA containing virus. One μg/ml doxycycline (Sigma) was added to the medium 24 h post infection to activate gene expression from the tet-inducible promoter 32.

Hippocampal neurons were washed, fixed with formaldehyde, and permeabilized with Tween-20 (0.3%). The following primary antibodies were used: mouse monoclonal anti-cyt-C (6H2.B4, San Diego, CA; 10 μg/ml), rabbit polyclonal anti-Bid raised against amino acids 1–58 of mouse Bid (AR-53; 1:1,000), and rabbit polyclonal anti-Apoptosis-Inducing-Factor (AIF) (1:500; 46). Nuclei were counterstained with Hoechst33258. Primary antibodies were detected and fluorescent images acquired as described previously 7.

The release of cyt-C from mitochondria was analyzed by selective permeabilization of the plasma membrane 47. Briefly, cells were permeabilized with 100 μg/ml digitonin at 4°C for 10 min. The supernatant representing the cytosol and the mitochondria-containing pellet fraction were separated by centrifugation and denatured. SDS-PAGE and Western blot analysis was performed as described previously 8. Cyt-C was detected with a monoclonal Cyt-C antibody (7H8.2C12; Pharmingen), Bid with the rabbit polyclonal antibody diluted 1:2,000, VDAC (Porin) and HSP-90 with mouse monoclonal anti-Porin (31HL, Calbiochem) and anti-HSP90α/β (Santa Cruz, Heidelberg, Germany) antibodies, respectively, at a dilution of 1:5,000. α-spectrin and its calpain-generated cleavage products were detected with a mouse monoclonal antibody (1622; Chemicon) diluted 1:5,000, and α-tubulin with a mouse monoclonal antibody (DM-1A; Sigma) diluted 1:1,000.

Data are given as means ± S.E.M. For statistical comparison, one-way analysis of variance followed by Tukey's test were employed. P values smaller than 0.05 were considered to be statistically significant.