Right to Adequate Choline

“Choline is a required nutrient for humans; it must be consumed in the diet to maintain health. Yet current dietary intakes for most people are well below the levels considered adequate.” — Steven Zeisel, Annual Review of Nutrition, 2006

A molecule the twentieth century did not name

Choline was isolated from bile in 1862 by Adolph Strecker and given a name that recorded its origin (χολή, bile). For the next hundred and thirty-six years, it circulated in biochemistry textbooks as a component of phosphatidylcholine and acetylcholine, without any official status as a nutrient. It became an essential nutrient in the United States in 1998, when the Institute of Medicine issued its first Adequate Intake, on the basis of the depletion studies conducted through the 1990s by Steven Zeisel at the University of North Carolina. Zeisel and his colleagues had shown that healthy male adults fed a choline-deficient diet for a few weeks developed measurable hepatic dysfunction, reversible on repletion. The reversibility was the criterion. Without adequate choline, the human liver fails within weeks in a controlled experimental setting.

The IOM set the Adequate Intake at 550 mg per day for men and 425 mg per day for women, and raised it to 450 mg during pregnancy. In 2016, the European Food Safety Authority followed with an Adequate Intake of 400 mg per day for adults and 480 mg during pregnancy. In the United States, national dietary survey data from the NHANES cycles show that fewer than one adult in ten reaches the AI (Wallace and Fulgoni, Nutrients 2016), and the fraction reaching the level suggested by more recent obstetric research is smaller still. A nutrient whose experimentally demonstrated deficit produces measurable organ dysfunction, whose recommendation was set in the closing years of the twentieth century, and whose population intake falls short of that recommendation across an entire continent, is not a marginal question. It is a right in the sense this site uses the word.

The pleiotropy that should be startling

Most nutrient descriptions open with a list of what the nutrient does. What is missed by this format is the vertigo. A molecule of three carbons and one quaternary nitrogen is at once architecture, signal, methyl donor, osmolyte, transporter, and regulator of inflammation. Each of these roles is structurally different from the others. The same small molecule is doing them all, in parallel, in every human body, at every moment. The interest of choline is not that it has many uses. It is that the physiological world happens to have organized several of its most different tasks around the same three-carbon frame.

Membrane architecture

Phosphatidylcholine is the dominant phospholipid of animal cell membranes, representing between forty and fifty per cent of total membrane phospholipid in most tissues. Sphingomyelin, another choline-bearing phospholipid, is the principal sphingolipid of the plasma membrane and, in the central and peripheral nervous system, of the myelin sheath. Every membrane through which a nutrient enters a cell, every membrane across which a receptor signals, every membrane that isolates a mitochondrial compartment from its cytosol, contains choline as a structural component. When dietary choline is inadequate, the cell breaks down its own phosphatidylcholine reserves to release free choline for the synthesis of acetylcholine, at the cost of membrane integrity. Richard Wurtman named this the autocannibalism hypothesis. The name is not gentle, and it is not intended to be.

Hepatic lipid export

The liver assembles very-low-density lipoproteins for the export of triglyceride from hepatocytes into circulation. VLDL assembly requires phosphatidylcholine for the surface monolayer of the particle. When choline is deficient, VLDL export fails, triglyceride accumulates in the hepatocyte, and non-alcoholic fatty liver disease develops. This is not a hypothesis. It is what Zeisel’s depletion studies showed in the 1990s and what subsequent work has confirmed in a range of populations, including postmenopausal women, whose oestrogen-dependent endogenous choline synthesis is diminished. NAFLD is currently the most prevalent liver disease in industrialized populations. The extent to which its epidemic is a choline deficit is not settled, and the extent to which the question is being asked in gastroenterology clinics is small.

Methyl donation

Choline is oxidized to betaine, which donates a methyl group to homocysteine to regenerate methionine. This pathway is one of two remethylation routes in the one-carbon cycle, the other being folate- and vitamin-B12-dependent. When folate and B12 status are marginal, choline becomes the load-bearing methyl donor, and when choline is inadequate, folate and B12 must bear the load alone. The three nutrients are functionally coupled, and a deficit in one increases the required intake of the others. This coupling is entirely absent from routine clinical reasoning, in which homocysteine is rarely measured, folate is treated in isolation, and choline is not on the map.

Central cholinergic system

Acetylcholine is the neurotransmitter of the basal forebrain projection to the cortex, the septum-hippocampal projection, and the striatal interneurons. Its synthesis depends on the availability of choline at the terminal, and the high-affinity choline transporter is the rate-limiting step. The cholinergic hypothesis of Alzheimer’s disease, formulated by Bartus and colleagues in 1982, rests on the observation that the earliest and most severe neuronal loss in Alzheimer’s is in the cholinergic projection from the nucleus basalis of Meynert. The clinical acknowledgment of this hypothesis is the class of acetylcholinesterase inhibitors (donepezil, rivastigmine, galantamine) that constitute one of the few licensed pharmacological options in the disease. These drugs raise synaptic acetylcholine by blocking its degradation. What they cannot do is provide substrate. If substrate is inadequate, blocking degradation raises a smaller pool.

Bile

Choline received its name from bile because bile is exceptionally rich in it. Phosphatidylcholine is the essential phospholipid of bile, forming with bile salts and cholesterol the mixed micelles that keep cholesterol in solution during intestinal transit. When biliary phosphatidylcholine falls, cholesterol crystallizes and the biliary epithelium is exposed to the detergent action of bile salts unbuffered by phospholipid. Cholesterol gallstone formation and primary biliary epithelial injury both follow. The organ that first yielded choline to nineteenth-century chemistry continues to depend on it for its own function.

Renal osmoregulation

The medullary cells of the kidney live in an osmotic environment that would destroy most tissues. They defend themselves by accumulating betaine as a compatible osmolyte. Betaine is the oxidation product of choline. The molecule that governs vagal tone in the peripheral nervous system also protects the cells that manage the osmolarity of the entire body. The physiological logic that arranged this economy is not obvious. It is nonetheless the economy.

Erythrocyte membrane and lifespan

The lipid composition of the red blood cell membrane, in particular the ratio of phosphatidylcholine to sphingomyelin, governs the deformability of the erythrocyte and its ability to traverse the capillary bed without lysis. Membrane composition drifts with age and with the availability of dietary choline. The erythrocyte lifespan, ordinarily around one hundred and twenty days, is not fixed by intrinsic biology alone. It is co-determined by the material available to build and maintain the membrane.

REM sleep architecture

Rapid eye movement sleep depends on cholinergic activation of the pontine tegmentum. The alternation of REM and non-REM cycles across the night is regulated by reciprocal interactions between cholinergic and aminergic populations. Pharmacological blockade of muscarinic transmission suppresses REM. Age-related reduction of cholinergic tone tracks with the age-related shortening and fragmentation of REM. The nightly reorganization of memory, and much of what a life feels like from the inside, depends on the availability of choline at cholinergic terminals hours after the last meal.

Gametogenesis and reproduction

Sperm membranes are phosphatidylcholine-rich, and choline is required for spermatogenesis. In oogenesis and early embryogenesis, choline supply governs the availability of phosphatidylcholine for the extraordinary membrane synthesis of the developing organism. The reproductive relevance of choline is not a footnote to pregnancy. It begins upstream of conception.

Epigenetic regulation

The phosphatidylethanolamine N-methyltransferase (PEMT) pathway in the liver synthesizes phosphatidylcholine from phosphatidylethanolamine using three methyl groups donated by S-adenosylmethionine. PEMT is the largest single consumer of SAM in the mammalian body. When choline intake is inadequate, PEMT flux increases, SAM is depleted, and methylation reactions throughout the genome are affected, including methylation of promoters governing BDNF and other neurotrophic and metabolic genes. Choline is therefore not only a substrate for methyl donation via betaine, it is also a determinant of methyl availability in the entire epigenetic system.

The parasympathetic nervous system as the compensatory substrate

The list above is already long. The section that follows would justify a fiche of its own. It is placed here because the parasympathetic system is where the material of choline becomes the material of the very capacity the site keeps naming in editorials like The Tragedy of Diagnosis and The Life We Call Normal: the compensatory reserve that keeps a body whole in conditions the body’s environment has not made easy.

Acetylcholine and the vagus

Acetylcholine is the principal neurotransmitter of the parasympathetic nervous system, both preganglionic and postganglionic. The vagus nerve (tenth cranial nerve) carries roughly eighty per cent of the parasympathetic outflow of the body and roughly eighty per cent of the afferent traffic from the viscera to the brainstem. Cardiac deceleration, bronchial tone, gastrointestinal motility and secretion, biliary release, pancreatic exocrine secretion, sexual arousal, lacrimation, salivation, and pupillary constriction are all downstream of cholinergic transmission. When choline supply to cholinergic terminals is inadequate, every one of these functions is provisioned from a smaller pool.

Vagal tone and heart rate variability

Heart rate variability, and in particular its high-frequency component, is the standard non-invasive index of vagal tone. Kleiger and colleagues showed in 1987 that reduced HRV after myocardial infarction predicts all-cause mortality independently of ejection fraction. The Framingham cohort, the ARIC study (Tsuji and colleagues, 1996), and dozens of subsequent cohorts have confirmed that low HRV predicts sudden cardiac death, cardiovascular events, and all-cause mortality in populations without prior cardiac disease. HRV is a physiological parameter with a documented prognostic weight larger than that of most laboratory measures ordered in routine adult medicine. It is nonetheless essentially absent from that medicine. It is used in cardiology of the athlete, in autonomic testing suites, and in research protocols. It is not on the panel a general practitioner orders for a fatigued patient in her forties.

The cholinergic anti-inflammatory pathway

In 2000, Kevin Tracey and colleagues at the Feinstein Institute published in Nature the demonstration that stimulation of the efferent vagus nerve suppresses systemic inflammatory response through nicotinic acetylcholine receptors of the α7 subtype on macrophages, which in turn suppress tumor necrosis factor release. The pathway has since been mapped in detail. It is now understood that a substantial component of the anti-inflammatory reserve of a healthy organism is a cholinergic reflex, dependent on the availability of acetylcholine at nerve terminals contacting immune cells in the spleen and the liver. Chronic subclinical inflammation, which is the pathophysiological common ground of a large fraction of age-related disease, is in part a failure of this reflex. Choline supply is the substrate of the reflex.

The diffuse signature of dysautonomia

The clinical presentation of impaired parasympathetic function is not a syndrome. It is a distribution of complaints that reach different specialties independently. Palpitations reach cardiology, where the imaging is often normal. Postural intolerance reaches internal medicine or is dismissed. Irritable bowel reaches gastroenterology. Insomnia reaches psychiatry or sleep medicine. Brain fog reaches nobody in particular. Depression and anxiety reach psychiatry, where the diagnosis will be made in the vocabulary of mood disorders rather than of autonomic imbalance. The argument of The Tragedy of Diagnosis is nowhere more literal than here. A patient with reduced vagal tone will typically be given four or five diagnoses across four or five services, none of them naming the underlying autonomic architecture, and none of them investigating whether the substrate of that architecture is adequate.

The industrial mirror of Behind Every Test

A modest industry produces implanted vagus nerve stimulators for epilepsy, treatment-resistant depression, cluster headache, and inflammatory conditions. LivaNova, SetPoint Medical, ElectroCore, and GammaCore are its principal actors. The devices work by delivering electrical current to the vagus in order to increase the frequency of vagal firing. What none of them asks, and what no protocol accompanying them requires, is whether the vagus has substrate to release at each stimulated firing. The industrial architecture around the parasympathetic system has captured the nerve as intervention target while leaving the substrate that the nerve releases entirely unaddressed. This is the argument of Behind Every Test, an Industry, played back in inversion: where there is an industry aligned with a nutrient, the test is common and the nutrient is measured. Where the industry is aligned with a device that bypasses the nutrient, the nutrient is orphaned.

Pregnancy as the paradigmatic case

The physiological right that the pregnancy case establishes is not an argument by analogy. It is the strongest and most literal instance of the general claim. In 2018, Marie Caudill and colleagues at Cornell published in FASEB Journal a randomized controlled trial in which pregnant women received either the IOM recommended intake of 480 mg per day of choline or 930 mg per day. The children of the mothers who received the higher dose demonstrated enhanced sustained attention on tests administered up to seven years later. Prior work by Caudill’s group and by Zeisel’s group had documented the effect of maternal choline on hippocampal development in the offspring, with permanent enhancement of memory and plasticity in animal models given prenatal choline supplementation.

The AI of 450 to 480 mg during pregnancy was set in the absence of a trial designed to identify the dose that optimizes neurodevelopmental outcome. The Caudill 2018 trial provided such data. The AI has not been revised. Prenatal supplementation of choline is not part of standard obstetric care in any industrialized country. Folate has been fortified into the wheat supply of most industrialized countries since the mid-1990s, following the demonstration of its role in neural tube closure. Choline has not been fortified. The comparison is instructive. Folate had a demonstrable acute defect visible at birth, a small industry aligned with cereal fortification, and a public health administration prepared to act. Choline had a subtler outcome visible in developmental trajectory, no industry aligned with a fortifiable food matrix, and no institutional actor prepared to move. The mechanism by which one nutrient enters the population food supply and another does not is not a question of biology.

The genotype question

Zeisel’s group has shown that variants in the PEMT gene, in MTHFD1, and in choline dehydrogenase (CHDH) substantially modify the individual requirement for dietary choline. Some women with common PEMT variants develop signs of organ dysfunction on intakes that are adequate for others. The concept of a single population-wide AI, in a nutrient with this degree of demonstrated genetic modification of requirement, is a compromise imposed by the format of a recommendation, not a physiological fact. The physiological reality is a distribution. The right to choline is therefore, in this exact sense, the right to a supply calibrated to the individual, with the diagnostic capacity to identify who needs more.

Sources and forms

Dietary choline is concentrated in a small number of foods. The egg yolk provides approximately 150 mg of choline per large egg, and remains, by density, the single most efficient dietary source. Beef liver provides approximately 350 mg per hundred-gram serving. Soybeans, salmon, chicken, and cruciferous vegetables provide smaller but non-negligible quantities. A population that consumes few eggs, avoids organ meats, and derives its protein primarily from muscle meats and grains will be structurally short of choline. Vegan populations without deliberate supplementation are the extreme case.

Supplemental forms differ in their target tissue. Choline bitartrate is the least expensive and provides substrate for general phosphatidylcholine synthesis. Phosphatidylcholine (lecithin) is delivered as an intact phospholipid and preferentially incorporated into membrane and hepatic pools. Alpha-glycerylphosphorylcholine (alpha-GPC) crosses the blood-brain barrier more efficiently and is used clinically for central cholinergic support. Citicoline (CDP-choline) is a phosphorylated intermediate with independent cytoprotective effects, used in trials of ischaemic stroke and cognitive decline. The correct form depends on the deficit being addressed. The correct dose depends on the individual. Neither is discoverable within the current clinical infrastructure, because neither is being asked.

Measurement, or its absence

Plasma free choline is a poor reflector of tissue status. Whole blood choline is somewhat more informative. Erythrocyte phosphatidylcholine content, hepatic proton spectroscopy for intrahepatic triglyceride, and urinary trimethylamine metabolites provide indirect windows. None of these is available to a general practitioner in a routine panel. A functional index that is available, and that reports on the peripheral cholinergic system without requiring biochemistry, is heart rate variability. HRV is a physiological read-out of vagal tone, and vagal tone is a functional read-out of acetylcholine availability at terminals. It is not a substitute for a choline assay. It is a signal from downstream, and it is currently unused outside sports cardiology and autonomic research.

What the right to choline requires

• Recognition of choline in food guides at intake levels that reflect the demonstrable requirement, revised to incorporate the neurodevelopmental data available since 1998.
• A prenatal protocol that includes assessment of choline status and supplementation to a dose supported by the neurodevelopmental evidence, on the model that folate fortification created for its own indication.
• Inclusion of heart rate variability in the routine assessment of the fatigued adult, of the adult with unexplained palpitations, of the adult with insomnia and cognitive complaints, of the ageing patient at risk of cardiovascular events, as a functional index of parasympathetic tone.
• Explicit consideration of choline status in the management of non-alcoholic fatty liver disease, of dysautonomic states, of cognitive decline, and of the many clinical presentations whose common substrate is a failure of cholinergic reserve.
• The recognition, in medical training, that the parasympathetic nervous system is a physiological parameter of which the person is titulary, and that its material substrate is a nutrient the medical system does not measure, does not teach, and does not supply.

References

• Zeisel SH. Choline: An Essential Nutrient for Public Health. Annu Rev Nutr. 2006;26:229-250.
• Institute of Medicine. Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. National Academies Press, 1998.
• Wallace TC, Fulgoni VL. Assessment of Total Choline Intakes in the United States. Nutrients. 2016;8(1):50.
• Caudill MA, Strupp BJ, Muscalu L, Nevins JEH, Canfield RL. Maternal choline supplementation during the third trimester of pregnancy improves infant information processing speed: a randomized, double-blind, controlled feeding study. FASEB J. 2018;32(4):2172-2180.
• Bartus RT, Dean RL, Beer B, Lippa AS. The cholinergic hypothesis of geriatric memory dysfunction. Science. 1982;217(4558):408-14.
• Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature. 2000;405(6785):458-62. (Tracey group.)
• Tracey KJ. The inflammatory reflex. Nature. 2002;420(6917):853-9.
• Kleiger RE, Miller JP, Bigger JT, Moss AJ. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol. 1987;59(4):256-62.
• Tsuji H, Larson MG, Venditti FJ, et al. Impact of reduced heart rate variability on risk for cardiac events. The Framingham Heart Study. Circulation. 1996;94(11):2850-5.
• da Costa KA, Kozyreva OG, Song J, Galanko JA, Fischer LM, Zeisel SH. Common genetic polymorphisms affect the human requirement for the nutrient choline. FASEB J. 2006;20(9):1336-44.