Physiological oxygen level is critical for modeling neuronal metabolism in vitro

• Published in: Journal of neuroscience research Vol. 90; no. 2; pp. 422 - 434
• Main Authors: Zhu, Jing, Aja, Susan, Kim, Eun-Kyoung, Park, Min Jung, Ramamurthy, Santosh, Jia, Junling, Hu, Xueying, Geng, Ping, Ronnett, Gabriele V.
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01.02.2012
• Subjects: AMPK; Animals; Cells, Cultured; Cerebral Cortex - embryology; Cerebral Cortex - metabolism; Cerebral Cortex - physiology; culture; Energy Metabolism - physiology; Female; metabolism; Models, Neurological; neuron; Neurons - physiology; oxygen; Oxygen Consumption - physiology; Pregnancy; Rats; Rats, Sprague-Dawley; Reactive Oxygen Species - metabolism
• ISSN: 0360-4012; 1097-4547
• DOI: 10.1002/jnr.22765

In vitro models are important tools for studying the mechanisms that govern neuronal responses to injury. Most neuronal culture methods employ nonphysiological conditions with regard to metabolic parameters. Standard neuronal cell culture is performed at ambient (21%) oxygen levels, whereas actual tissue oxygen levels in the mammalian brain range from 1% to 5%. In this study, we examined the consequences of oxygen level on the viability and metabolism of primary cultures of cortical neurons. Our results indicate that physiological oxygen level (5% O2) has a beneficial effect on cortical neuronal survival and mitochondrial function in vitro. Moreover, oxygen level affects metabolic fluxes: glucose uptake and glycolysis was enhanced at physiological oxygen level, whereas glucose oxidation and fatty acid oxidation were reduced. Adenosine monophosphate‐activated protein kinase (AMPK) was more activated in 5% O2 and appears to play a role in these metabolic effects. Inhibiting AMPK activity with compound C decreased glucose uptake, intracellular ATP level, and viability in neurons cultured in 5% O2. These data indicate that oxygen level is an important parameter to consider when modeling neuronal responses to stress in vitro. © 2011 Wiley Periodicals, Inc.