This article is based on Peters, A, Schweiger, U., Pellerin, L, Hubold, C., Oltmanns, K., Conrad, M., Schultes, B., Born, J and H. Fehm. 2004. The selfish brain: competition for energy resources. Neuroscience and Behavioral Reviews 28, 143-180.
In this paper the authors lay the foundation for their thesis that the brain not only regulates energy metabolism in the periphery but at times actually usurps it from the muscles and fat tissues. There are two very good reasons for the brain to be ‘selfish’ with regards to its energy needs; first, relative to its mass the brain uses far more energy than any other organ; second, despite its large energy needs the brain has trouble producing and storing energy.
While the peripheral organs can metabolize glucose, fat or proteins to produce energy, the restrictions the blood brain barrier (BBB) places leaves it almost exclusively dependent upon glucose. Since the brain has only a very limited ability to store energy it must ensure it receives a constant flow of glucose.
In order to do this the brain constantly monitors its own energy levels and when necessary takes energy away from other parts of the body.
The authors believe that reduced concentrations of ATP in the brain cause it to activate a stress program that increases its allocation of glucose from the periphery. At the same time it does this it activates the feeding centers of the brain and gives the signal to eat.
The authors argue that the plasticity of the part of the brain involved in the stress response can cause it to produce dysfunctional set points at which this stress program is activated. Too high a ‘set point’ will cause the brain to continually pull glucose from the periphery and result in anorexia. Too low a set point will result in reduced brain energy levels, increased glucose levels in the periphery and obesity.
The results of the CDC’s pharmacogenomics studies of allostatic stress in CFS suggest that the set point at which this stress response is induced is altered in CFS patients.
Ion Channels The regulation of ATP concentrations in the brain plays a central role in this paradigm. ATP is the central agent of energy production (aerobic respiration) in the mitochondria. Glucose and fatty acids are the main fuels for ATP synthesis. The authors propose that ATP sensitive potassium channels (ion channels) found on neurons and neuroendocrine cells are a kind of ‘energy sensor’.
These channels are closed when ATP is abundant (high energy state) and open when ADP is (low energy state). Adenosine di-phosphate (ADP) is what is left over after a phosphate group from adenosine tri-phosphate (ATP) is removed. It is the release of the phosphate that provides energy. The ATP/ADP ratio in a cell determines, therefore, whether these ion channels are open or closed.
The situation is actually more complicated than this. Both ATP and ADP bind to receptors on the ATP sensitive potassium channels called sulfonylurea receptors (SUR’s)
There are two types of SUR receptors on the ATP sensitive potassium channels; one has a high affinity for and one a low affinity for ATP.
High ATP affinity SUR receptors (SUR 1) – Cells with these receptors are active even when ATP levels are low. They are found throughout the brain on excitatory neurons – they serve to increase nervous system activity.
Low ATP affinity SUR receptors (SUR 2) – Because cells with these receptors become active only when ATP levels are high, they are inactive much of the time. These cells, not surprisingly, are mostly found on inhibitory neurons – they act to inhibitnervous system activity when it gets too high.
Given these traits the following scenario can be envisioned regarding the energy status of the brain.
During pathologically low energy states neurons seek to save themselves by becoming unresponsive to outside stimuli. During these periods both excitatory and inhibitory neurons are inactive.
During low but non-pathologic states of energy the excitatory neurons with high levels of high ATP SUR receptors are active but the inhibitory neurons are not. This results in increased glutamate production.
During very high energy states the inhibitory neurons with low affinity ATP SUR receptors become activated in order to moderate nervous system activity. These neurons produce the neurotransmitter GABA.
The key point is that the balance between nervous system excitation (glutamate production) and inhibition (GABA production) is determined by the ATP levels in the brain. If the brain needs more energy the excitatory neurons are activated and glutamate is produced. If the brain has too much energy the inhibitory neurons are activated and GABA is produced.
A side note: the receptors glutamate interacts with, called NMDA receptors, are found in the greatest abundance in three areas of the brain that are of great interest in CFS; the hippocampus, amygdala and basal ganglia. These regions are particularly vulnerable to a process called glutamic acid excitotoxicity which occurs when high glutamate levels damage or destroy neurons.
The selfish brain
Increased glutamate production is the key element in the brain’s attempt to increase its energy levels. It does this in three ways; by robbing the periphery of glucose, by restricting glucose uptake by the tissues in the periphery, and by activating the feeding centers of the brain.
Robbing Peter to pay Paul – altering the blood-brain barrier – The first thing the brain usually does to get more energy is to rob the periphery of glucose. This occurs when glutamate produced by the neurons located near the blood brain barrier (BBB) causes the astrocytes lining it to produce glucose transporters (GLUT1) that increase the uptake of glucose from the blood into the brain.
Increasing glucose levels and reducing glucose uptake in the periphery – the brain also increases glucose levels in the bloodstream by prompting muscle and fat cells in the periphery to inhibit their uptake of glucose (and switch to another energy source). It does this through activation of the limbic-hypothalamus-pituitary-adrenal system (LHPA). This system, which links the two major stress response systems, the HPA axis and the sympathetic nervous system (SNS), originates in two regions of the limbic portion of the brain.
The Limbic System
(adapted from Stedman’s Electronic Medical Dict.)
The limbic system describes an array of different brain structures including the hippocampus, amygdala, and fornicate gyrus, all of which connect to the hypothalamus. Through its connections with the hypothalamus the limbic system exerts an important influence upon the endocrine and autonomic motor systems; its functions also appear to affect motivation and mood.
First, a description of how this process works. Neurons from the hippocampus and amygdala activate both the HPA axis and the SNS. HPA axis activation occurs via the production of neuropeptides (corticotrophin releasing hormone (CRG), vasopressin) that stimulate the pituitary to produce adrenocorticotropin (ACTH), which in turn triggers the release of the major glucocorticoid, cortisol, produced by the adrenal gland. Cortisol, an antagonist of insulin, increases blood glucose concentrations. Insulin is a hormone that promotes, among other things, glucose utilization.
SNS neurons project from the limbic region to the adrenal gland and pancreas where they respectively stimulate the release of epinephrine (adrenaline) and suppress insulin release. They also project to the muscles where they suppress glucose uptake.
Activating the feeding centers of the brain – Another way the brain can also increase glucose availability in the bloodstream is by using glutamate to trigger the feeding centers of the brain.
In a nutshell the authors believe that during states of low brain energy high affinity ATP sensitive potassium channels increase glutamate production, thereby (a) increasing glucose transport across the BBB into the brain, and (b) activating the stress response (LHPA) which impedes glucose uptake to the muscles and fat tissues, and (c) triggering the feeding centers of the brain.
A final step
This system contains one more layer of complexity. It is not a straight shot from stressor to LPHA activation. One doesn’t want a stress response of this magnitude to become activated at any kind of stress; one wants it to become activated only when really needed.
The authors propose that the brain regulates the point or ‘set point’ at which the LPHA system is activated through two kinds of receptors, mineralocorticoid and glucocorticoid receptors (MR’s and GR’s) found on the neurons. The authors propose the brain regulates the ‘set point’ `at which the LPHA system becomes activated
These receptors operate in much the same way the ATP sensitive potassium channels do. Instead of reacting to ATP, however, they react to cortisol. Both are produced in the cell, and both regulate gene activity in the cell. The ability of the cell to adequately monitor its cortisol levels is key, therefore, to maintaining the proper energy levels in the brain.
MR’s and GR’s are known to regulate the transcription of many genes. One group of genes they interact with controls calcium ion channels, another group affects the activity of ligand gated ion channel (i.e. glutamate), and a third regulates intracellular signaling system involving G-protein coupled receptors.
It is through these genes that MR/GR’s regulate the excitability of the limbic neurons and the activity of LHPA response. In this model then MRs/GRs are ultimately responsible for the stimulation/inhibition of the master hormone, cortisol.
Mineralocorticoid receptors (MRs) – have a high affinity for cortisol; they bind to it even when low levels of intracellular cortisol are present. These receptors are excitatory, once bound they promote the production of more cortisol and other substances. If the MR’s are too abundant or too active they will call for cortisol production when it is not needed. This will lead to an hyper-responsive stress response in which the brain constantly activates the HPA axis and SNS.
Glucocorticoid receptors (GRs) – have a low affinity for cortisol; they bind to it only when intracellular cortisol levels are high. These receptors are inhibitory – they inhibit the production of cortisol and other substances. If these receptors dominate they will block the call for cortisol and lead to a hypocortisolic state. In this state the brain is not sufficiently responsive to its energy needs; it is locked in a hypo-responsive state.
But where does the pathology occur? How do things go wrong with this system? One way this system can go wrong is via an alteration of the set point at which the brain’s stress response is activated. The is where the limbic region of the brain and the hippocampus comes in.
The Hippocampus – A Locus of Dysregulation?
The stress response had long been thought to stop at the gateway between the brain and the endocrine system, the hypothalamus. Recent evidence indicating that the hippocampus contains high levels of receptors for adrenal hormones indicates that it too is activated during the stress response. Peters et. al. believe that the impact of stressful events on the hippocampus can cause it to permanently alter the LHPA activation set point in some people.
What makes this theory plausible is the high degree of plasticity the hippocampus displays. The hippocampus needs to display a high degree of plasticity because it appears to be the seat of an important process called long-term potentiation (LTP).
LTP is involved in our ability to apply lessons learned from an experience to similar experiences in the future. Experiments with laboratory animals suggest this type of learning may be centered in the hippocampus. The amygdala and hippocampus appear to decide what is stressful and how to deal with it.
Just as with the immune system short term elevations of epinephrine and cortisol appear to promote hippocampal functioning and learning but chronic elevations lead to impaired functioning. The hippocampus appears to respond to chronic stress with atrophy, memory impairment and increased fear.
A dysregulation centered in the hippocampus by an infection, trauma, psychological stress, low blood volume, etc could lead to an over or underactive response to all sorts of stimuli ranging from infection to low blood glucose levels to low blood volume etc.
Peter’s scenario posits that a chronically activated stress response can, through damage to the hippocampus (due to glutamate excesses?), alter the ‘set point’ as which the stress response of the brain kicks in.
Three models of an impaired brain response system
The authors describe three scenarios of altered LHPA set point activation. Not surprisingly given the connection between LHPA activation and glucose metabolism two of the three scenarios explain the genesis of abnormal feeding behaviors such as anorexia nervosa and obesity. These authors believe obesity and anorexia nervosa are brain diseases not behavioral problems.
The first factor on the road to obesity is, as noted above, a stressor that dysregulates the set point of the LHPA system. It doesn’t appear to matter what kind of stressor – Peters invokes a psychological stressor in his example – but also states that low cerebral glucose levels, exercise, infection, a hypothalamic lesion, (low blood volume, hypoxia) or endocrine disrupting chemicals could have the same effects.
In Peters’ scenario a person undergoing a period of high psychological stress develops atypical depression. Atypical depression – the kind of mood disorder most often found in CFS – is characterized by lethargy, fatigue, overeating, increased sleep, avoidance of social contacts and low CRH and cortisol levels. (See The Hypocortisolism in Chronic Fatigue Syndrome (ME/CFS) – Artifact or Central Factor?) for an alternative definition of ‘atypical depression’.)
Peters believes these findings are consistent with a decreased MR/GR ratio and a reduced set point of the LHPA system. In this scenario the brain is not selfish at all; it is in fact insufficiently selfish – it needs to pull more glucose from the periphery than it does. The low set point of the LHPA system results in extra allocations of glucose to the periphery and reduced allocations of glucose to the brain. This leads to fatigue, increased sleep and low cortisol levels.
The brain reacts to this by activating its feeding centers. This appears to be a process separate from the stress response – thus while the stress response is inhibited another part of the brain is still active – and telling the person to eat. Increased food intake increases glucose levels in the periphery and thus the brain but also results in increased mass in the periphery.
Since more muscle and fat tissue results in greater glucose demand overall this scenario simply results in greater competition for glucose with the brain. This further stimulates the feeding response, etc – the person is continually beset by cravings for food and they get more and more obese.
The impairment of the stress response stops the brain from doing what it should have been doing in the first place – increasing glucose transport across the BBB and reducing glucose uptake in the periphery.
Explaining anorexia nervosa
With regard to anorexia nervosa the opposite is true; the LHPA set point is too high. This increases glucose allocation to the brain but reduces it to the body. The low body mass indicates to the brain that it needn’t allocate much glucose to it. This results in the body wasting away without the brain activating the feeding response.
The unselfish brain and CFS
But how does all this relate to CFS? The authors of the two allostatic load papers in the Pharmacogenomic’s Journal proposed that a stress induced dysregulation of the LHPA system has resulted in an abnormally low LHPA ‘set point’ in CFS.
Their study indicated that the levels of two substances (aldosterone, cortisol) that bind to the MR and GR receptors were much lower in CFS patients than in the non-fatigued controls. This suggests a central metabolic dysregulation exists in CFS. The high rates of obesity in the CFS patients, and, in particular, the increased waist/hip ratios do, in fact, suggest a central metabolic dysregulation is present.
In this scenario the brains are selfish enough; the low set point in the brain precludes their brains from pulling enough glucose from the body. Low cerebral glucose levels have been found in CFS.
Maloney et. al. posit that a stressful event (psychological stress, infection, toxin, low blood volume, hypoxia) has lowered the set point of the brains stress response system. This has lead to an increased allostatic load not just with regard to CFS patients metabolism but in their cardiovascular system as well, and would, of course, have many other affects on the body.
McEwen, B. Plasticity of the hippocampus: adaptation to chronic stress and allostatic load. Annals New York Academy Science, 265-277.