Evidence for caspase effects on release of cytochrome c and AIF in a model of ischemia in cortical neurons.

Neuronal apoptosis following ischemia can be mediated by a caspase-dependent pathway, which involves the mitochondrial release of cytochrome c that initiates a cascade of caspase activation. In addition, there is a caspase-independent pathway, which is mediated by the release of apoptosis-inducing factor (AIF). Using caspase inhibitor gene therapy, we investigated the roles of caspases on the mitochondrial release of cyt c and the release of AIF. Specifically, we used herpes simplex virus-1 amplicon vectors to ectopically express a viral caspase inhibitor (crmA or p35) in mixed cortical cultures exposed to oxygen/glucose deprivation. Overexpression of either crmA or p35 (but not the caspase-3 inhibitor DEVD) inhibited the release of AIF; this suggests that there can be cross-talk between the caspase-dependent and the ostensibly caspase-independent pathway. In addition, both crmA overexpression and DEVD inhibited cyt c release, suggesting a positive feedback loop involving activated caspases stimulating cyt c release.

Effects of overexpression of antioxidants on the release of cytochrome c and apoptosis-inducing factor in the model of ischemia.

Apoptosis arises from neuronal damage following an ischemic insult. Apoptosis-inducing factor (AIF) is a protein released from mitochondria in response to pro-apoptotic signals which then translocates to the nucleus and triggers DNA fragmentation. In parallel with this, pro-apoptotic signals cause the release of cytochrome c from mitochondria, activating caspase-dependent apoptosis. During post-ischemic reperfusion, reactive oxygen species (ROS) are formed in excess in mitochondria and can play a role in initiating apoptosis. In cultures, ROS are formed during post oxygen glucose deprivation (OGD) normoxia/normoglycemia that is used as a model for ischemia. In this study, we delivered viral vectors to overexpress antioxidants (GPX, catalase, CuZnSOD, or MnSOD) in mixed cortical cultures, in order to investigate the effects of ROS-reduction on the release of cytochrome c and AIF. Overexpression of MnSOD, CuZnSOD, catalase or GPX all prevented AIF translocation from mitochondria to the nucleus. Potentially, this could reflect broadly non-specific protection due to reducing ROS load. Arguing against this, overexpression of the same antioxidants did not inhibit cytochrome c release. These findings suggest a specific interaction between ROS formation and the caspase-independent route of apoptosis.

gp120 neurotoxicity fails to induce heat shock defenses, while the over expression of hsp70 protects against gp120.

gp120, the coat glycoprotein of HIV, can damage CNS neurons. This appears to mostly involve an indirect pathway in which gp120 infects microglia, triggering the release of cytokines and glutamatergic excitotoxins which then damage neurons. A well-characterized response of cells to insults is to mobilize the heat stress response, a defense that has a number of protective consequences. We tested the capacity of gp120, at a dose well-documented to be neurotoxic, to activate the heat shock response in cultures from cortex and hippocampus, two brain regions sensitive to the neurotoxic effects of gp120. We found that gp120 failed to induce expression of hsp70, hsp25 or hsp90 in cortical or hippocampal cultures, under conditions where induction can be demonstrated in response to other insults. The failure of gp120 to induce a heat shock response is significant because we subsequently demonstrated that such an induction would have been beneficial. Specifically, over expression of hsp70 with a herpes viral amplicon vector protected cultured hippocampal neurons from gp120 neurotoxicity.

Effects of glucocorticoids in the gp120-induced inhibition of glutamate uptake in hippocampal cultures.

Studies examining the development of AIDS Related Dementia have concentrated on neurotoxic properties of the HIV viral coat protein, gp120. We have previously shown that this neurotoxicity can be exacerbated by glucocorticoids (GCs), the stress hormones secreted by the adrenal. Moreover, GCs also worsen several of the mechanisms mediating gp120 neurotoxicity, such as increased calcium flux, ROS generation, and energy depletion. Gp120 interferes with the reuptake of glutamate in glia cultures, another possible mechanism by which it can be neurotoxic. This paper examines the role of GCs in exacerbating this phenomenon. It was found that while GCs do not exacerbate the decrease in reuptake of glutamate in glia cultures, they do enhance the decrease in mixed neuronal cultures and this latter effect appears to be energy-dependent.

Protection against gp120-induced neurotoxicity by an array of estrogenic steroids.

gp120, the coat protein of HIV, can be neurotoxic and is thought to contribute to AIDS-related dementia complex. Such toxicity involves activation of glutamate receptors, mobilization of free cytosolic calcium, and generation of oxygen radicals. We have previously shown that the estrogen 17beta-estradiol, in concentrations of 100 nM or higher, lessens the neurotoxicity of gp120 in hippocampal and cortical cultures, blunts gp120-induced calcium mobilization, and lessens the oxidative consequences. In this study, we examined the protective potential of other estrogens. We found gp120 neurotoxicity in hippocampal cultures to be significantly lessened by estrone, equilin and estriol, although with an order of magnitude less potent than 17beta-estradiol. We also found all four estrogens to blunt gp120-induced calcium mobilization, with estriol being more efficacious than the other three estrogens. These findings give insight both into the mechanisms of estrogenic protection (e.g. receptor-dependent versus independent actions) as well as into the potential therapeutic use of estrogens against AIDS-related dementia complex.

Glucocorticoid exacerbation of gp120 neurotoxicity: role of microglia.

Gp120 protein, part of the HIV coat, may be a causative agent in AIDS-Related Dementia (ARD) because of its demonstrated neurotoxicity in vitro and in vivo. There are two possible mechanisms for this toxicity, namely through release of toxins from the microglia or through direct action on neuronal chemokine receptors. In tissue culture, glucocorticoids (GCs), the adrenal steroids released during stress, exacerbate gp120 neurotoxicity. In this paper, we examine the means by which GCs may increase toxicity, focusing on interactions with microglia and glia. Media from microglia treated with gp120 was toxic to neurons but not to glia. The effects of GCs upon the extent of gp120-induced release of toxins by microglia seemed to be dependent on the time of exposure to the hormone. Twenty-four-hour exposure of microglia to GCs decreased the toxicity of gp120-treated microglial conditioned media. In contrast, longer-term GC exposure enhanced neurotoxicity. There also appeared to be a component of gp120 neurotoxicity in hippocampal cultures that was exacerbated by GCs, independent of the amount of microglia present. Thus, GCs appear to act at a number of different sites in the multi-cellular pathway to exacerbate the neurotoxic effects of gp120.

Effect of GP120 on glutathione peroxidase activity in cortical cultures and the interaction with steroid hormones.

GP120 (the protein component of the HIV viral coat) is neurotoxic and may contribute to the cell loss associated with AIDS-related dementia. Previously, it has been shown in rat cortical mixed cultures that gp120 increased the accumulation of hydrogen peroxide and superoxide, two reactive oxygen species (ROS). We now demonstrate that gp120 increased activity of the key antioxidant glutathione peroxidase (GSPx), presumably as a defensive mechanism against the increased ROS load. Both estrogen and glucocorticoids (GCs), the adrenal steroid released during stress, blunted this gp120 effect on GSPx activity. The similar effects of estrogen and of GCs are superficially surprising, given prior demonstrations that GCs exacerbated and estrogens protected against gp120 neurotoxicity. We find that these similar effects of estrogen and GCs on GSPx regulation arose, in fact, from very different routes, which are commensurate with these prior reports. Specifically, estrogen has demonstrated antioxidant properties that may prevent the ROS increase (therefore acting as a neuroprotective agent) and rendered unnecessary the compensatory GSPx increased activity. To verify this we have added H2O2 to estrogen + gp120-treated cells, and GSPx activity was increased. However, with addition of H2O2 to GCs + gp120-treated cells there was no increase in activity. GCs appeared to decrease enzyme production and or activity and therefore under insult conditions ROS levels rose in the cell resulting in increased neurotoxicity. Overexpression of GSPx enzyme via herpes vector system reversed the GCs-induced loss of enzyme and eliminated the GCs exacerbation of gp120 neurotoxicity.