Bcl-2 family proteins are critical regulators of programmed cell death. The balance and interactions between pro- and anti-apoptotic members at the outer mitochondrial membrane play a critical role in determining the release of apoptogenic molecules, such as cytochrome c (Ow et al. 2008). The Bcl-2 homology domain 3-only (BH3) proteins constitute an upstream signal-dependent, pro-apoptotic subgroup that function by inhibiting anti-apoptotic Bcl-2 family proteins and/or directly activating pro-apoptotic multi-BH domain Bax/Bak (Youle and Strasser 2008). At least seven members of the BH3-only sub-group have been identified, which are activated via either transcriptional induction or post-translational modifications such as phosphorylation or proteolytic cleavage (Youle and Strasser 2008).
Seizure-induced neuronal death is associated with some features of apoptosis, including mitochondrial release of cytochrome c and AIF, proteolytic activation of effector caspases, DNA fragmentation, chromatin condensation and nuclear pyknosis (for review see Engel and Henshall 2009). BH3-only proteins may be critical upstream initiators since mice lacking bim or puma are protected against seizure-damage (Engel et al. 2010a; b; Murphy et al. 2010). Bid may also be important because it is rapidly cleaved following prolonged seizures (status epilepticus) in rats (Henshall et al. 2001, 2002; Shinoda et al. 2004b; Li et al. 2006). Cleaved Bid has also been detected in the mitochondrial fraction of hippocampus from patients with temporal lobe epilepsy (Henshall et al. 2004). However, the functional significance of Bid for seizure-induced neuronal death in vivo has not yet been examined. We studied the role of Bid in a mouse seizure model and assessed seizure-induced neuronal death in Bid-deficient mice.
- Top of page
- Materials and methods
- Supporting Information
The BH3-only proteins are an emerging focus of interest as potential initiators of seizure-induced neuronal death. We show here that intra-amygdala KA-induced status epilepticus results in early Bid cleavage in the hippocampus. However, Bid-deficient mice are not protected against seizure-induced neuronal death in vivo. We also show that nuclear translocation of AIF does not appear to be Bid-dependent in this model. These studies demonstrate that Bid plays no role or a redundant role in seizure-induced neuronal death in vivo.
Biochemical hallmarks of apoptosis-associated signaling pathways are present in hippocampal and extra-hippocampal brain regions after experimental status epilepticus and in temporal lobe material from patients with long-standing epilepsy (Engel and Henshall 2009). However, the apical initiators of neuronal killing in this setting remain incompletely understood. Evidence already points to involvement of anti-apoptotic Bcl-2 family proteins as inhibitors of cell death after seizures. Indeed, loss of bcl-w or over-expression of Bcl-xL modulates hippocampal damage in seizure models (Murphy et al. 2007; Ju et al. 2008). BH3-only proteins may be critical for the initiation of cell death after seizures. Several have been shown to be induced or post-translationally modified within a few hours of seizures including Bim and Puma, and the pro-apoptotic multi-BH domain protein Bax accumulates at mitochondria (Henshall et al. 2001, 2002; Shinoda et al. 2004b; Li et al. 2005; Noh et al. 2006; Engel et al. 2010b; Murphy et al. 2010). Here we studied Bid, which is particularly interesting because it can be activated downstream of the death receptor pathway, which is induced after seizures (Henshall et al. 2001; Shinoda et al. 2003). Bid activation has previously been reported in multiple in vivo models of neurologic insults, including ischemia, trauma and status epilepticus (Henshall et al. 2001; Plesnila et al. 2001; Franz et al. 2002; Yin et al. 2002; Bermpohl et al. 2006). Our studies show that full-length Bid is expressed in adult mouse hippocampus, in agreement with other mouse as well as rat and human data (Wang et al. 1996; Henshall et al. 2001; Franz et al. 2002; Krajewska et al. 2002; Yin et al. 2002; Shinoda et al. 2004b). Cleavage of Bid greatly enhances its cytochrome c releasing activity (Li et al. 1998; Luo et al. 1998) and our data show that Bid is cleaved into the p15 form after status epilepticus in mice. Thus, as in rats, Bid cleavage is a feature of seizure-induced neuronal death in vivo in mice. This contrasts with certain in vitro data reporting glutamate- and N-methyl-d-aspartate treatment of neurons does not result in Bid cleavage (Ward et al. 2006). Indeed, in vitro studies suggested that a translocation of full-length Bid to mitochondria may also occur during excitotoxic cell death, and that over-expression of a caspase 8 cleavage resistant mutant of Bid is sufficient to trigger release of cytochrome c and AIF in neurons (Ward et al. 2006; Konig et al. 2007). These differences likely reflect elements of the in vivo setting not being reproduced in vitro, such as caspase 8 involvement, and use of a variety of cell types, which may display differential BH3-only protein responses, as we have observed in vivo (Murphy et al. 2010). However, in the original studies on Bid, full-length Bid at high concentrations was also capable of releasing cytochrome c (Li et al. 1998; Luo et al. 1998). The amount of tBid we detected is certainly equivalent and in some cases greater than is induced after ischemia or trauma (Plesnila et al. 2001; Franz et al. 2002; Bermpohl et al. 2006), suggesting status epilepticus is particularly effective at generating tBid.
Our analysis of the temporal profile over which Bid cleavage occurred in mice parallels the time course observed in rats after status epilepticus (Henshall et al. 2001; Li et al. 2006). Bid cleavage is an early event in the present model, preceding up-regulation of both Bim and Puma, two other potent BH3-only proteins (Engel et al. 2010b; Murphy et al. 2010). The immediacy of tBid formation after status epilepticus creates a problem in causally linking Bid to mitochondrial release of cytochrome c or AIF because these only emerge at ∼4 h (Murphy et al. 2007 and present data). In contrast, Bim and Puma induction coincide with the release of apoptogenic factors from mitochondria in this model (Engel et al. 2010b; Murphy et al. 2010).
Caspase 8 is a potential mediator of Bid cleavage in the present model. Bid has previously been shown to be efficiently cleaved to tBid by caspase 8 in mouse brain (Plesnila et al. 2001), and caspase 8 is cleaved after seizures in our model (Shinoda et al. 2004a). Moreover, tBid formation after seizures in rats is reduced by treatment with a caspase 8 inhibitor (Henshall et al. 2001; Li et al. 2006). Caspase 3 is unlikely to be the cause, at least initially, because it is not activated until ∼4 h after seizures in the model (Engel et al. 2010b). Although the tBid fragment was 15 kD, which argues for caspase involvement, there are reports that calpain can produce a similar-sized fragment of Bid (Mandic et al. 2002). Nevertheless, the requirement for Bid to be cleaved to effect neuronal death in response to excitotoxic insults remains unproven (Konig et al. 2007).
Previous studies in Bid-deficient mice demonstrated significantly smaller infarcts compared with wild-type mice following ischemia and, at least transiently, reduced lesion size after trauma (Plesnila et al. 2001; Yin et al. 2002; Bermpohl et al. 2006). In the present study, we found that neuronal death after status epilepticus was not different in Bid-deficient compared with wild-type mice. Thus, Bid is not required for seizure-induced neuronal death in vivo in this model. The difference could rest with the functional significance of inflammation, which for ischemia and trauma is a critical contributor to damage via disruption of the blood–brain barrier and invasion of inflammatory cells, such as macrophages and cytotoxic lymphocytes (Waterhouse et al. 2005; Wang et al. 2007). Regardless, the lack of importance of Bid for seizure-induced neuronal death was surprising given its role in other neurologic injury models. Taken together, our data indicate that there is specificity for BH3-only proteins between neurologic insults with Puma, and possibly Bim being more important for seizure-induced neuronal death in vivo, and Bid being more important for ischemia and trauma.
Taken together, our data show cleavage of Bid is a conserved feature of seizure-induced neuronal death in vivo but it does not appear to be functionally important. This is not surprising because the pathogenic mechanisms driving cell death after seizures versus ischemia or trauma have overlap in some areas (e.g. excessive glutamate release) but also differences (e.g. cerebral blood flow, inflammation, the extent of tissue oxygen and glucose depletion, oedema and oxidative stress, among others) (Nedergaard 1988; Meldrum 1994; Liou et al. 2003). At least two predications can be made. First, pathophysiological features of the specific neurologic insult impose cell death stimuli with unique characteristics which drive to a greater or lesser extent the importance of one or more BH3-only protein, but that these may later converge on common post-mitochondrial downstream effectors (e.g. caspases or AIF). Second, one or more other members of the more potent BH3-only proteins, such as Puma or Bim, may be critical for seizure-induced neuronal death while being functionally redundant for ischemia- or trauma-induced neuronal death. As noted, seizure-damage in Puma- and Bim-deficient mice is reduced compared with wild-type animals (Engel et al. 2010a; b; Murphy et al. 2010). This suggests that tailoring of treatment to differences in the pathophysiology of the insult continues to be a likely requirement of any therapeutic strategy for acute neurologic insults.
In conclusion, the present study demonstrates that prolonged seizures in vivo activate the BH3-only protein Bid. However, cell death and AIF translocation in this model was not significantly influenced by the absence of Bid. These data further resolve the functional influence of cell death pathways during seizure-induced neuronal death which may be important for future approaches to neuroprotection.
- Top of page
- Materials and methods
- Supporting Information
Figure S1. Bid immunohistochemistry in bid−/− mouse hippocampus. Representative photomicrographs showing Bid-deficient (bid−/−) mouse hippocampus stained with two different Bid antibodies (left panel, R & D systems’ anti-Bid; right panel, anti-phosphoBid from www.antibodies-online.com). Note there is staining of many hippocampal cells with either antibody in Bid-deficient mice, in particular the CA3 pyramidal neurons and cells in the hilus (arrows). Inset shows a western blot confirming the absence of Bid in Biddeficient mouse hippocarnpus.
Figure S2. Bim and Puma expression following seizures in wild- type and Bid-deficient mice (a) Representative western blots (n = 1 per lane) showing expression of Bid, Bim and Puma 8 h following status epilepticus. Note, Bim and Puma staining are somewhat elevated in Bid- deficient (bid’) mice compared to controls. (b) Graphs semi-quantifying Bim and Puma levels in wild-type and Bid-deficient mice (n = 3 each). *p < 0.05 compared to wild-type (wt). ns, non-significant.
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.