Choline on the Brain? A Guide to Choline in Chronic Fatigue Syndrome

(This paper largely follows the outline of a paper by Chaudhuri and Behan that examined magnetic resonance studies on CFS (Chaudhuri and Behan 2004).

Brain metabolic activity in CFS

Three studies have examined metabolic functioning in the brain using proton magnetic spectroscopy (MRS) (Chaudhuri et. al. 2003, Tomoda et. al. 2000, Puri et. al. 2002). The levels of three metabolites (N-acetyl aspartate (NAA), choline, creatine) in the brain are examined using this technique.

Puri’s study found that CFS patients (a) have significantly higher levels of choline in the occipital region of the brain than do controls and (b) exhibit an abnormal choline gradient between the motor and occipital cortex (Puri et. al. 2002).

Chaudhuri’s study found increased choline levels in the basal ganglia (Chaudhuri et. al. 2003). A very small study (n=3) examining adolescents with CFS also found increased choline levels in the basal ganglia as well (Tomoda et. al. 2001).

Three studies then, all of them small, but most with highly significant findings (p<.05, p<.001, p<.008) have found increased brain choline levels mostly in the basal ganglia. Normal NAA levels in two studies indicated neuronal mass was not disturbed.

The basal ganglia

The basal ganglia are large masses of gray matter at the base of the cerebral hemisphere; i.e. they are near the base of the skull where it meets the spinal column. They provide a nexus for interactions combining limbic/motor activities with volition; i.e. they play a key role in internal motivational states. One of the aspects they effect is perception of effort.

The limbic system is a collective term that denotes an array of interconnected brain structures (hippocampus, amygdale, fornicate gyrus) at or near the edge (limbus) of the cerebral hemisphere that connect with the hypothalamus.

By way of these connections, the limbic system exerts an important influence upon the endocrine and autonomic motor systems and appears to effect motivation and mood. Several endocrine and autonomic nervous system abnormalities have been identified in ME/CFS.

Basal ganglia dysfunction often causes problems with  something called ‘tasking’. Sequential task processing, for instance, an important process used in initiating and following through complex tasks, is often impaired in people with basal ganglia dysfunction.

The ‘reward’ system which provides motivational impulses that in turn stimulate other parts of the brain is also often disrupted. These two abnormalities can increase the effort needed to carry out complex tasks, in particular.

A disease called akinesia which is defined as “poverty and slowness in willful movements” can also occur because of basal ganglia disease. It is believed to result from the inability of the brain to respond to environmental cues such as sight, sound and touch.


Choline is found in three forms in humans; phosphatidycholine (lecithin), acetylcholine and cytidine diphosphocholine. Most of the choline in the body is found in specialized fat cells called phospholipids that are abundant in the membranes of cells. Choline in used in the synthesis of three components in cell membranes; phospholipids, phosphatidycholine (lecithin) and sphingomyelin.

Causes of increased brain choline production

Elevated brain choline levels are usually associated with increased cell production (malignant tumors) and/or increased cell membrane turnover due to inflammation or ischemia (low blood flows) (Chaudhuri et. al. 2003).

Immune cell activation by macrophages and/or neuronal astrocytes has been shown to produce choline peaks in AIDS patients. Accumulations of neutrophils, lymphocytes and macrophages have also been associated with high choline peaks in patients with non-malignant tumors of the brain (Cummings et al. 2000).

Areas of inflammation and cell death are readily observed on MRI’s. Two MRI studies, however, have found no indication that either are present in the basal ganglia of CFS patients. Nor have systematic metabolic studies on lipid and peroxisomal function indicated systemic abnormalities in phospholid metabolism.

Given these findings the authors suggest the changes seen are due to local changes in the lipid composition of the membranes of the nerves.  These local changes could be due toreparative gliosis or altered ‘intramembranal signaling’ (Chaudhuri et. al. 2003).

Glial cells are non-neuronal components of the central nervous system that closely interact with neurons. Consisting of astrocytes, oligodendroctyes and microglia cells, they play an important role in neuron neuroprotection. Through their detoxification of glutamate, for instance they protects cells from being damaged due to glutamate toxicity.

Glial cells also secrete trophic factors that appear to protect against cerebral ischemia. Through their ability to siphon off excess potassium from neurons they also regulate potassium levels.

Most glial cells are astrocytes that surround the ends of the synapses and border endothelial cells in the capillaries. They help shuffle nutrients and metabolites from the blood into the neurons and play an important role in regulating the extracellular concentrations of ions, metabolites and neurotransmitters and in supporting neuronal functioning.

Olidgodendrocytes form myelin, an substance involved in the propagation of the action potential involved in nerve impulse transmission. Continually monitoring the nuural microenvironment, microglia rapidly react to pathological changes in their microenviroment with cytokine and/or trophic factor production (Hansson and Ronnback 2003).

Reparative gliosis or glial cell activation often occurs in response to tissue injury in the brain. Activated glial cells such (as astrocytes and the microgla) produce several cytokines (IL-1, S100B) that appear to ameliorate, at least at first, neuronal damage.

Chronic glial cell activation, however, appears to cause a cascade of events that eventuate in neurodegneration. Glial activation and cytokine production have been implicated in the progression of Alzheimers disease, for instance (Mrak and Griffin 1997).

While transient glial activation occurs in many viral encephalopathies, chronic glial cell activation occurs, for reasons that are not yet known, occurs in AIDS (Mrak and Griffin 1997). Mrak and Griffin’s suggestion that neurodegeneration could ‘persist or progress even following successful eradication of the virus’  is intriguing given the viral-like presentation of symptoms often initially seen in CFS.

Somewhat ominously they question whether chronic glial activation may place one at risk for Alzheimer’s (Mrak and Griffin 1997).  Chaudhuri and Behan are suggesting that damage to the brain has activated these important neuroprotective cells in ME/CFS patients brains.

Some other evidence suggests Chaudhuri and Behan may be correct in their assessment that reparative gliosis is occurring. Glial cells are the main source of transforming growth factor beta, (TBG-b), which is increased in the PBMC’s or serum of CFS patients (Peterson et. al. 1994, Bennett et. al. 2000, Kennedy et. al. 2004). TBG-b activity in the brain appears to contribute to neurodegeneration (Borlongan et. al. 2000).

An ATP connection?

Chaudhuri and Behan suggest there may be a connection between the high brain choline levels they’ve found and reduced ATP production. Evidence of impaired ATP production in two subsets of CFS patients and in another fatigue syndrome suggests reduced ATP production exists in at least some CFS patients.

A few small studies using magnetic resonance spectroscopy have found that at least a subset of CFS patient’s exhibit significantly reduced skeletal muscle exercise capacity that is accompanied by an early onset of intracellular acidification. In the one larger study (n=46) twenty percent of CFS patients had reduced phosphocreatine (PCr) /ATP ratios and higher levels of ADP upon exercise.

When energy is released ATP is transformed into ADP. Since higher rates of ADP imply higher ATP utilization – this study appears to indicate that CFS patients are using up ATP faster than normal. The breakdown of PCr furnishes the phosphate needed for the resynthesis of ATP from ADP. Thus low PCr ratios indicate that less phosphate is available to resynthesize ATP. Thus not only are some CFS patients using up ATP faster than normal they appear to be less able than normal to resynthesize ATP.

Two other studies have found reduced ATP concentrations during exercise and reduced PCr recovery during exercise in CFS. The common component to all these studies appears to be reduced availability of ATP probably because of increased breakdown of ATP; i. e. increased energy utilization in CFS patients. Intriguingly, Chaudhuri and Behan have also reported significantly increased resting energy expenditure (REE) levels in CFS patients.

CFS is not the only fatigue ‘syndrome’ that appears to be associated, at least in part, with reduced ATP levels. Cardiac syndrome X (CSX) is characterized by cardiac angina pain, a normal angiogram and in a significant portion of long term patients fatigue, muscle pain and exercise intolerance.

CSX is believed caused by reduced ATP levels in the cardiac cells. Twenty percent of CSX patients in a recent study exhibited PCr/ATP ratios in cardiac muscle similar to those found in the skeletal muscles of CFS patients.

Intriguingly a thallium cardiac scan of CFS patients indicated that CFS patients may display cardiac cell abnormalities similar to those found in CSX (Watson et. al. 1997). High outflows of cellular potassium may be responsible for the defects in thallium tracer distribution in the left ventricles of CFS patients (Watson et. al. 1997, Chaudhuri et. al. 2003).

An interesting finding given Dr. Cheney’s focus on energy production in the hearts of CFS patients and another positive study that was never followed up on.

Putting it all together – CFS is a disease of increased phospholipase (PLA) activity

Chaudhuri and Behan suggest increased choline levels contribute to cognitive dysfunction (effortful task processing) and reduced ATP levels impair aerobic metabolism and contribute to the exercise intolerance seen in CFS. What might increased brain choline and decreased ATP production have in common?

Chaudhuri and Behan believe both are due to increased phospholipase (PLA) activity. This appears to suggest they believe increased PLA activity occurs not just in the brain but is system wide. Since PLA is ubiquitous in the body increased PLA activity could affect a wide variety of tissues.


Phospholipases (PLAs) are a superfamily of esterases that release phospholipids ‘moieties’ (fractions) including choline when they hydrolyze (break) the ester bonds in lipid membranes. Aside from actual trauma or lipid peroxidation (free radical damage) phospholipid release from cell membranes is  usually caused by  phospholipase activity.

Phospholipids in the CNS cell membranes are high in polyunsaturated fatty acids (PUFA’s) and PUFA metabolism is ‘stringently controlled’ by PLA2 (and acetyltransferase). Normally when fatty acids are released by PLA2 they are rapidly taken up by membrane phospholipids by an energy dependent mechanism using CoA and ATP.

Phospholipase activity releases factors that exert widely varying effects in the cell. Phospholipids play a key role in regulating the release of arachidonic acid, the precursor of eicocanisoid synthesis. The eicocansoids (prostaglandins, thromboxanes, leukotrienes) mediate (trigger) the inflammatory process.

A marker of cellular injury, the inflammatory process begins with the release of AA. AA is broken up to produce pro-inflammatory mediators such as prostaglandins (COX 1, 2) and leukotrienes. Prostaglandins then combine with cellular receptors to initiate signaling cascades which utilize G-proteins and cyclic CMP (cAMP) to produce pro-inflammatory substances.

PLA2 activity has been implicated in the pathology of a number of neurodegenerative diseases including Alzheimer’s and is thought to play a role in neuronal plasticity (Sun et. al. 2004). The activation of the P2Y nucleotide receptor on astrocytes triggers ‘reactive gliosis’, a process implicated in these neurodegenerative diseases and which Chaudhuri and Behan believe may be occurring in CFS.

Triggering phospholipase activity

Why phospholipase activity would be increased in CFS patients is unclear. CFS patients appear to be subject, however, to several factors (infection, increased neurotransmitter/ cytokine levels, oxidative stress, neurotoxins) that could trigger phospholipase activity.

Chaudhuri et. al. note that infection and/or neurotoxins can produce prolonged changes in membrane functioning (Chaudhuri et. al. 2003). The authors suggest the adaptation of the host cell to either a pathogen or its exotoxin (neurotoxin) could result in a long term derangement of the membranes.


Viruses can induce phospholipase activation and the release of lipids including choline by effecting membrane permeability. Indeed, many bacteria, viruses and parasites utilize lipid rich membrane domains as routes of entry into the cell.

Does the lack of immune activation (inflammation) in the brains of CFS patients suggests either an ongoing pathogenic attack is not involved or that it occurs without evoking an immune response (?). The possibility that a neurogenic pathogen could trigger PLA activity immediately brings up the question of two viruses, HHV6 and EBV, that have been historically linked with CFS.


HHV6 infects the very cells Chaudhuri and Behan posit may be producing the choline peaks in CFS. The HHV6 foundation suggests HHV6 activity in the glial cells could cause fatigue by altering ion channel function. HHV6 is ubiquitous in the human population; almost everyone has been exposed to it by third year.

The lack of an antigenic response to HHV6 in healthy controls suggests it remains in its latent state most of the time. In vitro studies indicate HHV6 is able to infect a number of nervous system cells including neurons, astrocytes, oligodendrocytes, microglial cells).

Astrocytes, in particular, appear to function as latent reservoirs for the virus. Antigenic responses to HHV6 in MS and a type of encephalitis indicates HHV6 reactivation occurs in some neurological disorders. PCR analysis has indicated HHV6 is present in encephalitis, meningitis, febrile seizure and encephalopathy.

HHV6 reactivation occurs in two neurological diseases, multiple sclerosis (MS) and Guillain-Barre syndrome, in which fatigue is a prominent symptom (Dewhurst 2004).

Two variants of HHV6 exist, HHV6A – which has not been clearly associated with any disease, and HHV6-B – which is responsible for most of the symptomatic infections in childhood. HHV6A appears to occur more commonly in central nervous system (CNS) tissues than HHV6-B (Dewhurst 2004).

The history of HHV6 and CFS is decidedly mixed. Once thought to be a key factor in CFS, after many studies HHV6 activation is generally now thought not to play a major role in CFS.

Questions regarding the efficacy of most HHV6 testing procedures in establishing the presence of an active population have, however, left a window open for HHV6’s possible re-emergence as a vital research topic in CFS. The HHV6 foundation asserts that only early antigen testing of HHV6 is effective in CFS.

Epstein-Barr virus (EBV)

EBV has an even more tortured history with CFS than does HHV6. Once thought possibly to be the cause of CFS EBV is now considered only to be an opportunistic infection. Recent studies indicate that as with HHV6 antigen testing for antigens found on the coat of the EBV particle may miss substantial numbers of CFS patients who contain a virus that is active but fails to replicate. Glaser believes that though replication is never achieved the virus is nevertheless able to produce enzymes that negatively affect the body (Glaser et. al. 2005).

Oxidative stress – Studies indicate increased free radical production occurs in neurodegenerative disorders. Autopsies indicate the brains of people with neurodegenerative diseases (Alzheimer’s, Parkinson’s, ALS, Huntington’s, etc.) display signs of oxidative damage. Whether this damage is the causal in nature or simply a common endpoint of a chronic disease process has been a source of controversy (Klein and Ackerman 2003).

Melatonin, a free radical scavenger, has been an effective neuroprotector in rodents exposed to oxidative stresses (Borlongan et. al. 2000). One theory suggests increased levels of oxidants or (neuro)toxins can rearrange the position of lipids in membranes so that their ester bonds are more accessible to PLA attack.


Cytokines such as IL-1 and TNF-a and lipopolysaccharides (LPS) and neuronal components such as NMDA, glutamate and muscarinic cholinergic enhancers (agonists) can all trigger PLA2 release.


Neurotransmitters and growth factors can also trigger PLA activity.

Treatment recommendation

Because oxidative stress can trigger phospholipase activity the authors recommend the use of highly unsaturated fatty acids (HUFA’s) in CFS.

Ongoing research

During the 2004-5 period the CFIDS Association of America funded Dr. Shungu to examine brain metabolites using H MRS technology. This most intriguing project, which will be a great deal larger in scope than the previous studies, should provide valuable information on metabolite levels in the brains of CFS patients. Part of the abstract from the CAA’s website is below.

H MRS Neurometabolites as Diagnostic Markers for Chronic Fatigue Syndrome – During the past 3-4 years, our research group had the opportunity to use a brain imaging technique known as hydrogen magnetic resonance spectroscopic imaging (H MRSI) – an imaging technique that is similar to conventional MRI, except that it can measure levels of certain important brain chemicals or neurometabolites – to record the levels of such chemicals in the brain of 31 individuals suspected with CFS.

Comparison of the levels of these neurometabolites with those in normal people, showed about 50% of all CFS patients had abnormal levels of the chemicals.

This proposal’s overall objective will be to develop H MRSI as a tool for evaluating CFS. To accomplish this, we will test the hypotheses that (1) CFS will be associated with specific changes in the levels of certain brain chemicals, and that such changes will be measurable by H MRSI; by refining this technique, these measurable brain chemical changes could serve as diagnostic markers of CFS; and (2) that the profile of these brain chemicals in CFS will be significantly different from that in people with psychiatric diseases, such as generalized anxiety, that are very similar to and often confused with CFS.

Therefore, the results of this research will be able to establish not only that CFS has a distinct profile of certain brain chemicals than healthy human brain, but also that its profile is different from that of a very similar psychiatric disorder. This finding can be the basis for using the levels of brain chemicals measured by H MRSI as markers for chronic fatigue syndrome in clinical evaluations as well as in clinical trials of promising treatments.


The researchers suggest increased choline levels seen in the brains of CFS patients occur when pathogenic (or other) insults on the cell membranes in the brain trigger phospholipase activity and choline release. Increased choline levels in a part of the brain involved in task processing and motor activities (movement) could account for the fatigue associated with mental effort and movement found in CFS.

Reduced ATP levels due to high phospholipase activity in the muscles could also contribute to the exercise intolerance seen in CFS. Several processes that appear to be upregulated in CFS (cytokine production, oxidative stress) could trigger phospholipase activity.

The very small sample sizes in the studies noted automatically raises a red flag; too many small apparently significant studies of CFS patients have failed the test of replication. It is encouraging, however, when independent study groups, no matter how small their sample sizes, find similar results in two different areas of the brain.


A recent update on the CDC’s CFS webpage reported that a PET scan study found that interferon treatment resulted in increased activity in the basal ganglia, particularly in the putamen and globus pallidus nuclei. This ties together increased the increased immune activity (interferon) with the altered function in the basal ganglia circuitry hypothesized by Chaudhuri and Behan. Interferon activation is required for RNase L activation. Potentially it ties together the immune and CNS’s of CFS patients.


Bell, D. 2005. Noted in Lyndonville News Vol 2, No. 2.

Bennett AL, Chao CC, Hu S, Buchwald D, Fagioli LR, Schur PH, Peterson PK, Komaroff AL. 1997. Elevation of bioactive transforming growth factor-beta in serum from patients with chronic fatigue syndrome. J Clin Immunol.17(2):160-6.

Borolongan, C., Yamamoto, M., Takei, N., Kumazaki, M., Ungsuparkorn, C., Hida, H., Sanberg, P. and H. Nishino. 2000. Glial cell survival is enhanced during melatonin-induced neuroprotection agains cerebral ischemia. FASEB J. 14: 1307-1317.

Cummings, B., Mchowat, J. and r. Schnellmann. 2000. Phospholipase A2s in cell injury and death. Journal of Pharmacology and experimental therapeutics 294: 793-799.

Chaudhuri, A. and P. Behan. 2000. Fatigue and basal ganglia. Journal of Neurological Sciences 179: 34-42.

Chaudhuri, A,. Condon, B., Gow, J., Brennan, D. and D. Hadley. 2003. Proton magnetic resonance spectroscopy of basal ganglia in chronic fatigue syndrome. Brain Imaging 14: 225-228.

Chaudhuri, A., and P. Behan. 2004. In vivo magnetic resonance spectroscopy in chronic fatigue syndrome. Prostaglandins, Leukotrienes and Essential Fatty Acids 71: 181-183.

Chaudhuri, A. and P. Behan. 2004. Fatigue in neurological disorders. Lancet 363: 978-988.

Dewhurst, S. 2004. Human herpesvirus Type 6 and human herpesvirus type 7 infections of the central nervous system. Herpes 11, 105A-111A.

Filippi, M., Rocca, M., Falini, A., Codella, M., Scotti, G. and G. Comi. 2002. Functional magnetic resonance imaging correlates of fatigue in multiple sclerosis. Neuroimage 15: 559-567.

Glaser, R., Padgett, D., Litsky, M., Baiocchi R., Yang, E., Chen, M., Yeh, P., Klimas, N., Marshall, G., Whiteside, T., Herberman, R., Kiecolt-Glaser, J., and M. Williams. 2005. Stress-associated changes in the steady-state expression of latent Epstein-Barr virus; implications for Chronic Fatigue Syndrome and cancer. Brain, Behavior and Immunity 19: 91-103.

Hansson, E. and L. Ronnback. 2003. Glial neuronal signaling in the central nervous system. FASEB 17, 341-348.

Hokama, Y., Uto, G., Palafox, N. A., Enlander, D., Jordan, E. and A. Cocchetto 2003. Chronic Phase Lipids in Sera of Chronic Fatigue Syndrome (CFS), Chronic Ciguatera Fish Poisoning (CCFP), Hepatitis B, and Cancer With Antigenic Epitope Resembling Ciguatoxin, as Assessed With MAb-CTX. Journal of Clinical Laboratory Analysis 17:132–139.

Kennedy G, Spence V, Underwood C, Belch JJ. 2004. Increased neutrophil apoptosis in chronic fatigue syndrome. J Clin Pathol.57(8):891-3.

Klein, J. 2000. Membrane breakdown in acute and chronic neurodegeneration: focus on choline-containing phospholipids. J. Neural. Transm. 107: 1027-1063.

Klein, J and S. Ackerman. 2003. Oxidative stress, cell cycle and neurodegeneration. The Journal of Clinical Investigation 111: 785-793.

Mrak, R. E. and S. Griffin. 1997. The role of chronic self-propagating glial response in neurodegeneration: implications for long-lived survivors of human deficiency virus. Journal of NeuroVirology 3: 241-246.

Natelson B., Weaver, S., Tseng, S. and  J. Ottenweiler. 2005.  et. al. 2004. Spinal fluid abnormalities in patients with chronic fatigue syndrome. Clinical Laboratory Diagnostic Technology 12, 52-55.

Overstreet, D. and V. Djuric. A genetic rat model of cholinergic hypersensitivity: implications for chemical intolerance, chronic fatigue and asthma. Annals of New York Academy of Sciences 92-103.

Peterson PK, Sirr SA, Grammith FC, Schenck CH, Pheley AM, Hu S, Chao CC. 1994. Effects of mild exercise on cytokines and cerebral blood flow in chronic fatigue syndrome patients. Clin Diagn Lab Immunol. 1(2):222-6.

Puri, B., Counsell, S., Zaman, R., Main, J., Collins, A., Hajnal, J. and N. Davey. 2002. Relative increase in choline in the occipital cortex in chronic fatigue syndrome. Acta Psychiatrica Scandanavia 106: 224-226.

Scheean, G., Murray, N., Rothwell, J., Miller, D. and A. Thompson. 1997. An electrophysiological study of the mechanism of fatigue in multiple sclerosis. Brain 120: 299-315.

Sun, G., Jianfeng, W., Jensen, M. and A. Simonyi. 2004. Phospholipase A2 in the central nervous system: implications for neurodegenerative disease. J. Lipid Research 45: 205-213.

Tomoda A, Miike T, Yamada E, Honda H, Moroi T, Ogawa M, Ohtani Y, Morishita S. 2000. Chronic fatigue syndrome in childhood. Brain Dev 22 :60-4.

Tomoda A, Joudoi T, Rabab el-M, Matsumoto T, Park TH, Miike T. 2005. Cytokine production and modulation: comparison of patients with chronic fatigue syndrome and normal controls. Psychiatry Res. 2005 Mar 30;134(1):101-4

Watson WS, McCreath GT, Chaudhuri A, Behan P. Possible cell membrane transport defect in chronic fatigue syndrome? Journal of Chronic Fatigue Syndrome 1997; 3(3): 1-13.

Share this!