AMPK is a metabolic master switch activated by multiple stimuli(e.g. hypoglycaemia, muscle contraction and hypoxia) that increase cellular AMP:ATP ratio (Rutter et al. 2003). Once activated AMPK inhibits ATP consuming processes and enhances ATP production partly, at least, by changes in fuel selection. Little is known regarding the distribution and functional roles for AMPK in the brain. Thus, we are examining the distribution and nutrient dependence of AMPK in the arcuate nucleus (ARC), a hypothalamic area involved in nutrient sensing and energy homeostasis. The insulin-secreting cell-line, CRI-G1, was used as a positive control since AMPK is regulated by glucose in pancreatic β-cells (Rutter et al. 2003). Male Sprague-Dawley rats were killed humanely, hypothalamic slices (300-400 µm) obtained and equilibrated in aCSF at 22°C for 20 mins followed by 1hr at 35°C. Isolated ARC wedges or CRI-G1 cells were incubated in 10 mM or 0 mM glucose for up to 30 minutes, prior to lysis, SDS-PAGE and immunoblotting. Glucose deprivation for 15 minutes increased AMPK phosphorylation at threonine-172 (a measure of the activation state) in CRI-G1 lysates (n = 5). This was accompanied by increased acetyl CoA carboxylase (ACC) phosphorylation at serine 221, a major substrate for AMPK, over the same time course. In contrast, although AMPK was phosphorylated following glucose removal in ARC lysates, there was no detectable increase in ACC phosphorylation (n = 4). Dual fluorescence immunohistochemistry with antibodies to MAP2 and phosphorylated AMPK demonstrated that AMPK was present in a sub-population of ARC neurones and phosphorylation increased following glucose removal. Antibodies specific to the α1 and α2 catalytic subunits showed that α1 AMPK is prominent in processes from cells surrounding the third ventricle, whereas α2 AMPK is present intracellularly in various ARC neurones. Incubation of slices in 10 or 0 mM glucose did not change the intensity or cellular location of either isoform. Thus, both AMPK catalytic subunits are present, and are phosphorylated in response to glucopenia in certain ARC neurones. However, unlike insulin-secreting cells ARC AMPK phosphorylation does not drive ACC phosphorylation. The distribution of AMPK subunits may be important indicators of their actions as neurones staining positive for AMPK have a similar localisation to those known to contain proteins responsible for glucose-sensing (Penicaud et al. 2002).