DiscussionOur findings suggest that 3OHB induces expression of the
Bdnf gene and increases BDNF protein levels in cerebral cortical neurons via activation of the
Bdnf gene promoter IV by a mechanism involving the transcription factor NF-κB and the histone acetyltransferase p300. Inhibition of monocarboxylic acid transporter 2 prevented 3OHB-induced
Bdnf expression indicating a requirement for cellular uptake of 3OHB for stimulation of
Bdnf expression. Cultured primary neurons are typically maintained in media containing concentrations of glucose (10–25 mM) that would be considered pathologically hyperglycemic (i.e., diabetic)
in vivo (e.g., Brewer
et al.
1993).
We found that whereas 3OHB stimulated
Bdnf expression in neurons maintained in a relatively low concentration of glucose (1 mM), it did not increase
Bdnf expression in neurons maintained in a high concentration of glucose (10 mM). During prolonged fasting, plasma glucose concentrations are maintained low (3–5 mM) while 3OHB concentrations are elevated greatly (5–10 mM). However, under the latter conditions the extracellular glucose concentration in the brain is believed to be considerably lower that plasma glucose concentration (de Vries
et al.
2003). Previous studies have shown that BDNF levels are reduced in the hippocampus in rodent models of type 2 diabetes including leptin receptor mutant mice (Stranahan
et al.
2009) and rats maintained on a high-fat plus glucose diet (Stranahan
et al.
2008).
Conversely, intermittent fasting induces BDNF expression in the hippocampus, which may mediate beneficial effects of intermittent fasting on hippocampal neurogenesis, synaptic plasticity, and neuroprotection (Lee
et al.
2002; Arumugam
et al.
2010). Our findings therefore suggest the possibility that 3OHB contributes to increased BDNF expression in response to fasting.
It was previously reported that NF-κB induces N-methyl-D-apartate receptor-mediated
Bdnf gene transcription in cultured cerebellar granule cells (Lipsky
et al.
2001; Marini
et al.
2004).
We found that 3OHB induces the nuclear translocation of the NF-κB p65 subunit and increases the interaction of the transcriptional co-activator p300 with p65 in cortical neurons. Consistent with our findings in neurons, it was recently reported that 3OHB increases NF-κB activation in calf hepatocytes (Shi
et al.
2014). We found that cortical neurons treated with 3OHB exhibited enhanced recruitment of p300 to a
Bdnf promoter IV sequence, and inhibitors of NF-κB and p300 blocked the ability of 3OHB to induce
Bdnf gene expression.
Increased mitochondrial ROS generation appears to mediate 3OHB-induced
Bdnf expression because we found that agents that reduced mitochondrial superoxide levels prevented 3OHB-induced
Bdnf promoter activity. Interestingly, we found that levels of two antioxidant enzymes (SOD2 and heme oxygenase 1) that are encoded by genes previously shown to be induced by NF-κB (Kiningham
et al.
2001; Naidu
et al.
2008) were increased in 3OHB-treated neurons.
We found that 3OHB increased mitochondrial respiration in cortical neurons suggesting that up-regulation of antioxidant defenses in response to 3OHB may represent an adaptive response to cope with elevated mitochondrial ROS generation.
We found that neurons treated with BDNF exhibited an increased mitochondrial respiration rate, and that a BDNF blocking antibody prevented 3OHB-induced increases in mitochondrial respiration. It was previously reported that BDNF can increase mitochondrial respiratory coupling in rat brain mitochondria (Markham
et al.
2004) and increases respiratory coupling efficiency in mouse brain synaptosomes (Markham
et al.
2012). It has also been shown that when cultured rat cortical neurons are treated with 3OHB, their respiratory capacity increases and they are better able to sustain mitochondrial function when exposed to high levels of glutamate (Laird
et al.
2013).
Our findings suggest the neuroprotective actions of 3OHB are mediated, at least in part, by BDNF signaling. Numerous studies have shown that fasting is neuroprotective in animal models relevant to AD, PD, and HD (Duan and Mattson
1999; Duan
et al.
2003; Halagappa
et al.,
2007; Griffioen
et al.
2013), as well as stroke (Arumugam
et al.
2010) and traumatic brain injury (Davis
et al.
2008). If and to what extent 3OHB contributes to such neuroprotection remains to be established. However, the ability of ketogenic diets (Gasior
et al.
2006; Maalouf
et al.
2009) and 3OHB supplementation (Kashiwaya
et al.
2013) to protect neurons and improve functional outcome in animal models of neurodegenerative conditions is consistent with an important role for 3OHB in the neuroprotective effects of fasting.
Evolutionary considerations and experimental findings suggest that cognitive function is bolstered by vigorous physical activity and food scarcity/fasting (Mattson
2015). The ability to outsmart one's competitors in the battle for limited food resources has been fundamental to the evolution of the brains of most species including humans. The integration of signaling pathways by which the brain and other organ systems respond adaptively to evolutionarily fundamental bioenergetics challenges are undoubtedly complex involving activation of CNS neural circuits, neuroendocrine pathways, and signals from peripheral organs to the brain. Emerging evidence suggests that BDNF is key mediator of adaptive responses of the brain and peripheral organ systems to bioenergetic challenges (Marosi and Mattson
2014).
Our findings suggest a role for 3OHB in up-regulation of BDNF signaling in brain cells in response to exercise and fasting. We found that voluntary exercise increases plasma 3OHB levels and that there is a significant positive correlation between the concentration of circulating 3OHB and levels of BDNF in the hippocampus. Consistent with a role for 3OHB-induced BDNF expression in the neuroprotective effects of fasting and exercise in vivo, it was reported that fasting engages TrkB signaling, promotes neuroplasticity, and improves behavioral recovery after spinal cord injury (Plunet
et al.,
2008). For example, BDNF signaling in the CNS regulates appetite (Kernie
et al.
2000), enhances peripheral insulin sensitivity (Nakagawa
et al.
2000), and enhances parasympathetic regulation of heart rate (Wan
et al.
2014).
Because BDNF expression in multiple brain regions is increased in response to energetic challenges that elevate 3OHB levels, our finding that 3OHB acts directly on neurons to stimulate BDNF expression suggests potential roles for 3OHB in up-regulating BDNF expression under such conditions and could, by this mechanism, contribute to the beneficial effects of fasting and vigorous exercise on cognitive performance, and to improved peripheral energy metabolism and cardiovascular fitness (Wan
et al.
2003; van Praag
et al.
2014).
Социальные закладки