Adaptation to periods of prolonged hypoxia involves modulation of gene expression, which controls functional remodelling of a variety of different cell types. Classically, chronic hypoxia is known to cause increased production of erythropoeitin to stimulate red blood cell number and so the O₂ carrying capacity of blood. However, there are numerous detrimental effects of chronic hypoxia, amongst them an increased incidence of dementias, particularly Alzheimer’s disease (e.g. Moroney et al. 1996).

At the cellular level, a major factor in Alzheimer’s disease is the production of the plaque-forming amyloid β peptides (AβPs) which can cause cell death during the neurodegenerative progression of this dementia (e.g. Mattson, 1997). The mechanisms underlying cell death caused by AβPs remain to be determined, but are believed to involve production of reactive oxygen species (ROS) and disturbances of Ca²⁺ homeostasis (reviewed by LaFerla, 2002). We found that prolonged hypoxia (10 % O₂, 24 h) caused a marked potentiation of stimulus-evoked catecholamine release from model O₂-sensing PC12 cells, in part by inducing a Cd²⁺ resistant Ca²⁺ influx pathway. This effect was associated with (indeed, appeared to require) formation of AβPs, and direct exposure to AβPs exerted remarkably similar effects in these cells (Taylor et al. 1999). Subsequent electrophysiological studies revealed that hypoxia (via AβP formation) appeared to induce a small Cd²⁺-resistant Ca²⁺ influx pathway, and also selectively up-regulated L-type voltage-gated Ca²⁺ channels in these cells (Green & Peers, 2001). These two separate effects were further distinguished, since inhibition of the transcription factor NFk-B prevented up-regulation of L-type channels, but not the hypoxic induction of the Cd²⁺-resistant Ca²⁺ influx pathway coupled to exocytosis (Green & Peers, 2002). However, both of these effects could be inhibited by a variety of antioxidants (Green & Peers, 2002; Green et al. 2002), indicating that formation of ROS, most likely from the AβPs themselves, was an essential step in this pathophysiological response to hypoxia.

These observations in turn raise numerous further questions. For example, what switches on the production of AβPs in hypoxia? A likely explanation is that there is a switch in the normal, non-amyloidogenic processing of amyloid precursor protein (APP) (which normally generates the neuroprotective, soluble fragment, sAPPα) to the pro-amyloidogenic processing pathway. This may involve altered expression of secretases, the enzymes that cleave APP (Mattson, 1997). Another question of major importance is whether the above-described observations, all made using a continuous cell line, can be reproduced in cells of the central nervous system. We are currently addressing this latter question, and have recently shown that both hypoxia and AβPs augment Ca²⁺ currents in cerebellar granule neurones. Furthermore, we have reported that Ca²⁺ signalling in primary cultures of astrocytes is disturbed following a period of chronic hypoxia, primarily because hypoxia appears to cause excessive Ca²⁺ loading of mitochondria (Smith et al. 2002), and this effect is associated with pro-amyloidogenic APP processing. These observations are amongst the first to provide insights into the cellular basis accounting for the increased incidence of Alzheimer’s disease following hypoxic episodes. Such information is likely to be important in the future development of interventions designed to prevent the long-term, deleterious effects of hypoxia.

The author’s own work was supported by The Wellcome Trust and the Medical Research Council.