Numerous studies have shown that disordered Ca²⁺ homeostasis plays a role in AD pathology [5]. The balance between [Ca²⁺]_i and ER Ca²⁺ content is mediated through two distinct processes. On the one hand, Ca²⁺ is taken up by the ER against a gradient by Sarcoplasmic/Endoplasmic Reticulum ATP-ases (SERCA) and released from this organelle via RyR and InsP₃R [6]. Several authors have reported abnormal ER Ca²⁺ release from InsP₃R pathways [7-9]. The ER is critical in the regulation of [Ca²⁺]_i, in the folding and processing of many proteins and in the activation of cell death pathways [10]. It has been reported that when Ca²⁺ release from the ER is disturbed, extra-cellular accumulation of misfolded proteins can occur and ER-stress-induced apoptosis pathways can be triggered [11-13]. When cells are stressed (such as might occur under conditions of nutritive challenge), the ER and other organelles are also perturbed. Under these conditions, the ER chaperone GRP 78 is induced [14], Caspase-12 (residing on the cytoplasmic side of the ER), is activated [15] and this further activates Caspases-9 and -3 which commit the cell to apoptosis [16]. It has also been demonstrated that Caspase-12 mice are resistant to ER stress and to Aβ induced cell death [15]. In vitro studies have also shown that ER Ca²⁺ dyshomeostasis are synergistically related [11]. ER function itself is sensitive to the presence of reactive oxygen species (ROS) [17-19]. In addition, when Ca²⁺ is released from the ER is increased, the ion is increasingly taken up by mitochondria. The mitochondrial respiratory then is induced to generated further ROS [20]. Many studies have also demonstrated that oxidative stress is critical in mediating the neurotoxic properties of Aβ species [21, 22]. Furthermore, cells expressing AD-associated Presenilin-1 mutations have also been shown be susceptible to ER stress [23, 24]. Various laboratories have also reported that ER Ca²⁺-release channels (RyR and InsP3R) are involved in transduction of the apoptosis signal pathway [25-27]. Apoptosis induction has also been observed in a large volume of published data in response to diverse signals [28]. Ferreiro et al have also shown that Aβ25-35 and Aβ40 both induce an increase in [Ca²⁺]_i in neurons after a 24-hour incubation with these peptides [11, 21].

In 2005, McGowan and colleagues published the creation of a new transgenic model of AD. In fact, McGowan et al generated two related strains of mice, one overexpressing human Aβ40 (BRI-Aβ40) and the other Aβ42 (BRI-Aβ42), against the same WT background and without attendant upregulation of APP [29]. In 2008, our laboratory reported that cortical neurons cultured from new born BRI-Aβ40 and BRI-Aβ42 pups both showed greater vulnerability to trophic factor withdrawal (TFW)-induced apoptosis. Under these conditions both transgenic neurons had elevated oxidative stress, exacerbated mitochondrial dysfunction, upregulated apoptotic signaling, and increased DNA fragmentation and cell death compare to WT controls [30].

This report represents research that is a mechanistic extension to previously reported findings from our laboratory [30]. In the previous work we found that overexpression of Aβ40 and Aβ42 almost equally increased neuronal susceptibility to oxidative stress, mitochondrial dysfunction and apoptotic cell death when the cellular environment becomes hostile under nutritive stress induced by TFW, which simulates the conditions of an aged Alzheimer's brain. Under these conditions, vascular supply, energy utilization, cellular metabolism and waste clearance by the brain are all known to be compromised. Several lines of evidence have implicated neuronal changes in [Ca²⁺]_i being critical in AD neurodegenration. For example, Ca²⁺ homeostasis is altered in fibroblasts isolated from AD patients [36], in cells bearing the human presenilin 1 (PS1) AD mutation [7], and evidence for abnormal ER Ca²⁺ release through the ER InsP₃ receptor [9]. The findings from the present study showing that Aβ40 and Aβ42 statistically equally exacerbated calcium response to TFW are consistent with our cell death data from the previous study, and confirms that dysregulation of intracellular calcium is critically associated with neuronal cell death induced by both Aβ40 and Aβ42. These results are in sharp contrast to the published data that demonstrated significant difference in amyloidogenic properties between the two Aβ species. Aβ42 was shown to be much more amyloidogenic than Aβ40 despite a very similar property in neurotoxicity demonstrated in our studies. These observations are important because, contrary to traditional way of thinking, we show that a highly amyloidogenic Aβ species (such as Aβ42) might not necessarily be more neurotoxic than a non-amyloidogenic Aβ species (such as Aβ40).

Our data on InsP₃R are also somewhat unexpected, given the documented role of calcium release from ER via these receptors in several models of neurodegeneration. Our findings of exacerbated [Ca²⁺]_i response in both Aβ transgenic neurons are strongly suggestive of a mechanism involving disturbed Ca²⁺ homeostasis when these cells are exposed to nutritional stress. However, the hypothesis that this increase might be mediated through ER Ca²⁺ release via InsP₃R is not supported by our data where stimulation of InsP₃R with BK and blockade of these receptors with XeC did not yield an appreciable impact on calcium response following TFW in transgenic cells even though WT neurons responded to these treatments in a predictable manner. The mechanisms underlying the muted response to BK in Aβ transgenic neurons are not clear, but it may indicate a depletion of ER calcium stores or a desensitization of InsP₃R to BK in these neurons following TFW. A third possibility could be that in neurons transgenic for Aβ, Fluo-3 AM (the acetoxymethyl ester of Fluo-3) was already saturated with increased intracellular calcium following TFW to a point that stimulation with BK would not be able to increased the fluorescence further even when additional calcium was being released from ER. However, this scenario is highly unlikely for the following reasons: Based both on published literature as well as recommendations from the dye's manufacturer (ANASPEC; Fremont, CA 94555), we optimized the use of Fluo-3 AM by selecting recommended concentrations of the dye (5 μM), loading time (35 minutes) and incubation temperature (37°). Several lines of published evidence suggest that average basal level of intracellular free calcium in primary cortical neurons is about 80-90 nM, and that at this concentration of calcium, only about 15% of Fluo-3 is typically bound to calcium (Stephen Paddock et al., Confocal Microcopy Methods and Protocols, Humana Press, 1999). Saturation of Fluo-3 by calcium typically does not occur until free calcium is well over 3 μM, which would require an increase of over 30 times above the average resting calcium level in these neurons. The increase that we observed in Aβ transgenic neurons was ~3 folds over the baseline level, which was far below what would be expected for saturation of Fluo-3. The apparent lack of calcium response to XeC in Aβ transgenic neurons following TFW indicates that calcium dysreguiation in these cells following TFW are likely to be mediated by calcium channels other than ER InsP₃R receptors. The fact that we were able to significantly dampen the increase [Ca²⁺]_i with chelation of the free ion with BAPTA in all three types of neurons, coupled with the published observations that specific calcium channel blockers may confer neuroprotective actions, suggests that it is imperative to pursue cell survival strategies in AD models based on mechanisms that are critically involved in maintaining intracellular calcium homeostasis.