In one pass Your body makes and uses the glutamate in MSG every day, and it is the brain's main excitatory messenger. Not this — MSG is toxic ('Chinese Restaurant Syndrome') — Originates from a 1968 uncontrolled letter; Geha 2000 double-blind challenge found only 2.3% responding, non-reproducibly — glutamate is the same molecule as in tomato and breast milk, safe at dietary doses.
Educational content, not medical advice — consult a clinician.
In the gut it is the same ion as in a tomato MSG, the sodium salt of glutamic acid, dissociates into free glutamate and sodium once in the body; in the gut it is the same molecule as tomato glutamate.
Put the three doses side by side Natural foods bring far more glutamate each day than the MSG in a meal, which usually holds far less than the 5 g swallowed fasting in trials.
The first stop is also where most of it ends The small-intestinal epithelium burns passing glutamate on the spot as fuel, so the first stop for glutamate is also where most of it ends.
Your body makes and uses the glutamate in every day, and it is the brain's main excitatory messenger. But most of the glutamate you eat is used up on its first pass through the gut wall and the liver; the brain runs on glutamate it makes on the spot. Where that road breaks is taken apart in the later chapter on how MSG helps you use less salt.
The scare began in 1968. The New England Journal of Medicine printed a reader's letter in which Robert Ho Man Kwok described numbness at the back of his neck, palpitations and weakness after eating at Chinese restaurants, and guessed that MSG (monosodium glutamate) might be the cause. It was only a short letter, with no control group, no blinding and no record of dose, but it gave the world a catchy name: "Chinese restaurant syndrome."
Today the claim that "MSG is harmful" comes in many forms: headache, asthma, facial flushing, hyperactivity in children… The list of symptoms is so wide that any meal that left you feeling off can be made to fit. This story separates where the claim came from and what the evidence actually shows, then sets out who really needs to pay attention.
Evidence · A claim that explains everything
The name "Chinese restaurant syndrome" has lasted this long partly for reasons that have nothing to do with itself and everything to do with its shape. Learn the shape and you can use it on other claims later.
Its list of symptoms is open-ended: headache, palpitations, numbness, flushing, fatigue, thirst, trouble concentrating… After a meal, almost everyone has something going on. The longer the list, the higher the hit rate, and a claim that almost always hits tells you nothing new: it cannot predict anything, it can only be matched up after the fact.
It also comes with an attribution shortcut. A lot happens at once in a meal: more salt than usual, more oil, a bigger portion, some alcohol, eating faster than usual, a poor night's sleep beforehand. All of those variables moved, and MSG is the only one with a name, a label and the sound of a chemical. Attention naturally lands on the thing that has a name.
So there is only one way to test a claim like this: isolate that one variable and add it or leave it out without you knowing. That is why blinding exists, and it is the thread running through the chapters that follow.
Take this with you: when a claim can explain every discomfort you feel, what should make you wary is that it explains too much, not too little.
Chapter 2
Glutamate is already in your food
Glutamic acid is an amino acid, and one of the most abundant amino acids in the human body. The body makes it continuously and also uses it as the raw material for (the messenger that calms nerves down).
It is also widespread in natural foods, especially those rich in umami. Free glutamate in a few foods runs roughly:
Tomato: about 140–250 mg per 100 gParmesan cheese: about 1200 mg per 100 gSoy sauce: about 1000–1700 mg per 100 mlDried shiitake: about 1060 mg per 100 gHuman milk: about 19–22 mg per 100 ml, which is where a baby first gets glutamate The very first food a person drinks already contains it. So "is harmful?" is not really a question about a foreign substance. It is a question of whether the same molecule in a different package suddenly becomes something else.
Mechanism · Free versus locked in protein
(monosodium glutamate) is simply the sodium salt of glutamic acid. It is made by microbial fermentation, the same kind of process that makes soy sauce and cheese savory, and the end product is glutamate plus sodium. Once in the body it immediately splits into free glutamate ions and sodium ions, and in the gut that glutamate is the same molecule as the glutamate from a tomato. For the specific figures on natural glutamate in shiitake, see Mushrooms.
So is there any difference between the "natural" kind and the kind "in MSG"? Yes, but the difference is not the molecule. It is the state it is in when it reaches your tongue. Glutamate sits in food in two ways:
Locked inside protein: most of the glutamate in meat, eggs, milk and beans is one link in a long protein chain. While it is locked in, it has no umami taste and is not absorbed as it is; digestive enzymes in the stomach and small intestine have to cut it free first.Free: glutamate that has already come off the chain and floats on its own. All the umami your tongue can taste comes from this kind. What fermentation and long aging do is exactly this cutting in advance. Parmesan aged for a long time, soy sauce fermented for months, ham air-dried, tomatoes ripened on the vine in full sun: all of these take protein apart and release glutamate from the chain. These foods taste especially savory not because they contain something else, but because their glutamate has already been cut free.
Industry makes MSG the same way: microbes ferment sugars and produce free glutamate directly, which is then combined with sodium and crystallized. In other words, MSG is not a "chemical substitute" for natural food. It is the step that makes cheese and soy sauce savory, lifted out on its own and turned into a powder.
Mechanism · The body makes it and keeps it in check
Treating glutamate as just a seasoning misses how much weight it carries in the body. It is one of the busiest molecules in metabolism and in the nervous system:
The shared dock for amino groups: when the body turns one amino acid into another, it often unloads the amino group onto glutamate first, then passes it from glutamate to the next molecule. Nitrogen moving around the body mostly goes through its hands.The raw material for , the messenger that calms nerves: GABA is glutamate with one carboxyl group removed. In other words, the accelerator and the brake come from the same source.The brain's main excitatory messenger: the signal neurons pass to activate one another uses glutamate more than anything else. The pathways behind learning and memory run on it. That last point is exactly where the "excitotoxin damages the brain" scare lands, so it needs to be said in full: the glutamate in your brain does not come from your meal. Neurons and the astrocytes next to them make it on the spot and recycle it on the spot: used glutamate is taken up by the glial cells, turned into glutamine and handed back to the neurons, in a loop.
Why keep it under such tight control? Because the concentration itself is the signal. In the gap at a synapse, glutamate has to spike the instant a neuron fires and be cleared the instant it stops, or the message does not come through cleanly. If a meal could push up the background level, the system could not work. Put the other way, the fact that it works means it does not let diet steer it.
That is why the blood-brain barrier keeps a tight hold on glutamate. It is not a simple wall but a gate with a direction: the transporters there mainly move glutamate from the brain into the blood, not the other way round.
So where does the chain from eating it, to a rise in the blood, to a rush into the brain actually break? The later chapter on how helps you use less salt takes it apart link by link.
What the evidence says · double-blind trials failed repeatedly
In the decades after Kwok's letter, researchers tested the claim that " makes you feel unwell" again and again in double-blind, placebo-controlled trials. The conclusion has been consistent: at the amounts eaten with food, it does not show up.
The most rigorous was a multicenter trial that recruited people who believed they were sensitive to MSG and gave them a large 5 g dose on an empty stomach. With a large dose on an empty stomach, more people reported symptoms than with placebo; but the same person's reaction often did not match up on retesting, and no such reaction was seen when MSG was eaten with food. An independent review by FASEB, commissioned by the US FDA, reached the same conclusion: at normal dietary amounts MSG is safe for the general population; only when 3 g or more is taken at once on an empty stomach might a small number of people have brief, mild discomfort.
So the condition under which it "might cause trouble" is an empty stomach plus a very large dose, not a meal.
Numbers · The doses side by side
Put the doses side by side and it gets clearer. The glutamate that natural foods bring in every day (most of it locked inside protein) is far more than the added to a meal as seasoning, and the MSG in a meal is usually far less than the 5 g swallowed at once on an empty stomach in the trials. The dose that "might cause trouble" has to meet three conditions at once: an empty stomach, a large amount, and one of the few people who are sensitive. That is a completely different situation from an ordinary meal.
The "empty stomach" condition is not filler; it has a concrete physiological meaning. On an empty stomach, a swallow of water reaches the small intestine quickly, so the whole dose lands on a short stretch of gut wall at almost the same moment. Mixed into a meal, things are entirely different: food slows the emptying of the stomach, other amino acids released by protein digestion compete for the same transporters, and absorption is spread out over a long time. The same number of grams can give very different peaks.
That is why the trials used an empty stomach and a large dose: they were trying as hard as they could to force a reaction out. The researchers were not imitating a meal; they were deliberately creating conditions far more extreme than eating, to see whether symptoms could be provoked.
That changes what the result means. If a deliberately amplified condition produces only reactions that do not match up and do not repeat, then at a much gentler real dinner table it is even less likely to be the reason you feel unwell.
Which way the experimenters were pushing is a useful question to ask of any study. These trials were pushing to provoke a reaction; they could not provoke it reliably even so, and that is what gives the conclusion its weight.
Evidence · Why self-reported sensitivity fails
"Most people who believe they are sensitive cannot tell the real thing from the fake under blinding" can sound like calling them liars. It is not: their discomfort is usually real; they have just pinned it on the wrong cause. The mechanism in between is worth spelling out, because it applies to every judgment of the form "I feel unwell when I eat X."
Expectation itself produces symptoms. When you know you ate the thing that has been named, subjective feelings like headache and a racing heart really do come on more easily, and they are real feelings, not faked. That is why only a test in which you do not know counts.Too many things change at once in a meal. Salt, oil, portion size, eating speed, alcohol, that day's sleep: is just the one variable that has a name.People remember only the hits. Dozens of meals with MSG and no problem leave no impression; the one uncomfortable meal does. That is how memory works by default, not anyone's carelessness. So the most rigorous trial did one crucial thing: it picked out the people who reacted to real MSG but not to placebo and tested them again and again. That step subtracts the expectation effect first, so that the remaining signal is as clean as possible. Even so, only a small number of positive reactions were left, and they did not repeat on a second test.
True positive reactors do exist, and that should not be erased. They are just far fewer than the people who believe they are one, and the only way to tell the two apart is not memory but a blinded test.
Umami is a taste system of its own. The tongue has a pair of receptors (T1R1/T1R3) dedicated to recognizing glutamate; when glutamate docks on them they send a "savory, satisfying" signal, along a separate path that runs alongside salty and sweet.
That leads to a very practical conclusion: can help you eat less salt. The reason is the sodium content. By weight, table salt is about 39% sodium and MSG about 12%, roughly a third as much. Use a little MSG to add savoriness to a soup or dish while cutting back the salt, and total sodium falls without the dish necessarily tasting less satisfying. In taste tests of soups and dishes, replacing part of the salt with MSG allowed a good deal less salt without lowering the ratings for how good the food tasted.
For people who need to limit sodium (high blood pressure, chronic kidney disease, older adults), or who struggle with how bland a low-salt diet tastes, this is a sodium-cutting tool backed by taste tests; whether it lowers blood pressure over the long run has not been tested directly. It only works as a replacement, not if you salt as usual and then add MSG on top. For the full picture of sodium and blood pressure, see Potassium & Sodium.
A CLOSER LOOK
Umami lowers sodium through replacement
Glutamate activates a separate umami receptor; reducing salt is what can reduce sodium intake.
1Umami has its own route
Tongue T1R1/T1R3 receptors recognize glutamate and signal savory satisfaction alongside the separate salty-taste system.
2Replacement, not an extra addition
MSG contains less sodium by weight than salt. Taste tests support sodium reduction through partial replacement, but a long-term blood-pressure benefit has not been directly tested.
Illustration for understanding; not to scale. Saved figures include explanations and sources.
Mechanism · Where eaten glutamate stops
The chain behind "an excitotoxin rushes into the brain" has four links: you eat it, it is absorbed into the blood, the level in the blood rises, and it crosses the blood-brain barrier into the brain. At the amounts in a normal meal, most of it breaks at the first link.
Glutamate's first stop is the cells lining the small intestine, and that is also where most of it ends. The lining of the small intestine renews itself extremely fast and is among the most energy-hungry tissues in the body, and glutamate passing through the gut is one of its main fuels: it burns that glutamate on the spot for its own use. Tracer studies show that most of the glutamate in a meal is used up by the gut wall before it ever reaches the general circulation.
The little that gets past the gut wall enters the portal vein, and its next stop is the liver, where it is used further for converting amino acids and for energy. What actually reaches the general circulation is the small remainder left after these two checkpoints.
So the step "eat and blood glutamate shoots up" barely holds with a normal meal. And even if it rose a little, the blood-brain barrier's one-way gate is still there: the transporters there mainly move glutamate from the brain into the blood, not the other way.
Put the two layers together, and the "excitotoxin damages the brain" story behind "Chinese restaurant syndrome" cannot be followed to the end at the amounts people actually eat: it struggles to get past the gut wall.
That also fits with one fact. Babies take in glutamate from breast milk continuously, and breast milk is the only food during the most critical stretch of brain development. If glutamate in the diet could freely push up levels in the brain, that would be hard to make sense of.
Mechanism · Why umami can stand in for salt
"Swapping umami for salt" sounds like a sales line, but it has a quite specific pathway.
The tongue uses two different devices for salty and for umami. Saltiness comes from sodium ions passing straight through channels in the taste-cell membrane; umami comes from a signal that starts when glutamate docks on the T1R1/T1R3 receptor pair. The two lines travel separately all the way up to the brainstem and then the taste area of the cortex before they meet. Your final sense of whether this dish has enough flavor is the combined judgment after they meet, not the reading from either line alone.
That leaves room to work with. The hollow "something is missing" feeling after you cut salt can be filled in by the umami line, and the overall judgment is still "enough flavor."
The umami line also has an amplifier. Glutamate alone gives a limited umami taste, but together with nucleotides (the inosinate and guanylate family) the umami is clearly amplified. That is why so many traditional dishes bring two kinds of ingredients together: kombu with dried bonito, shiitake stewed with chicken, tomato stir-fried with egg, ham used to start a stock. Kitchens were using these pairings for centuries before anyone explained them.
An honest boundary: contains sodium too, so it is not "zero-sodium seasoning." What it does is deliver the same satisfaction for less sodium. If you salt as usual and then add MSG on top, total sodium only goes up; this tool works only as a replacement, not as an addition.
Who genuinely needs to be careful · an honest list
" is safe for most people" is not the same as "nobody ever needs to think about it." Here it is by group:
Most adults need no special limit: the amounts of MSG in an ordinary diet have not been shown to harm the general population in double-blind trials, and there is no need to avoid glutamate-containing foods or MSG seasoning.
People who believe they are sensitive to MSG can test it once: if you keep getting a headache or feeling unwell after meals, you can run a simple blinded test at home (someone else adds it or leaves it out without telling you) to see whether MSG is really involved. In the most rigorous double-blind trial, most people who believed they were sensitive could not tell MSG from placebo; true positive reactors exist, but they are far fewer than the people who believe they are one.
People with severe asthma: there are case reports of MSG possibly triggering asthma attacks (thin evidence). If you have severe asthma and are concerned, it is reasonable to talk it over with your doctor.
For any specific health problem, consult a doctor or a registered dietitian; the information on this site does not replace medical advice.
In practice · A blinded test at home
If you really suspect you are sensitive to , test it once rather than replaying the memory. A home version is not hard, but skip any of these and the test is worthless:
You must not know. Have someone else portion the food, add MSG or leave it out, write the answer down and put it away, and compare only after the test. If you handle it yourself, it means nothing: the expectation effect (knowing what you ate makes symptoms more likely to follow) walks straight into the result.Nothing else is allowed to change. The same dish, the same amount of salt, the same portion, eaten at the same time of day. Only one variable may move, or you will still not know what caused what.Repeat it several times, and shuffle real and fake. A single test is no better than a coin toss. Both the real and the fake need to come up over several rounds, in shuffled order.Decide what to record before you eat. Write down first what you are looking for: a headache? a racing heart? numbness? when does it start? how long does it last? Do not go hunting for a feeling after eating. "Looking after the meal" will almost always find something.Be ready to accept that nothing shows up. That does not mean your discomfort was fake. It means the cause lies elsewhere: the salt, the oil, the portion, eating too fast, or simply being tired that day. Pick the wrong target and you will keep avoiding something irrelevant while the real cause stays put. This method is not only for MSG. You can use it on any suspicion of the form "I feel unwell when I eat it." Being able to design one clean test yourself is more useful than memorizing any list of what to avoid.
Chapter 6
How to treat it · practical conclusions
Here is everything so far boiled down to judgments you can take with you:
" is toxic": no scientific support. The claim started with one uncontrolled personal letter, and decades of double-blind trials since have found no harm that reliably repeats.
"Natural glutamate is fine, MSG is not": does not hold. Once in the gut, the two are the same molecule. Worrying about MSG while happily eating ketchup, Parmesan and soy sauce makes no chemical sense.
Replacing some salt with MSG: a meaningful tool for cutting sodium, especially for people who need to limit salt but still want savory food.
"I feel unwell after eating at a certain Chinese restaurant": the more common explanations are that the dish itself was very salty or oily, the portion was too big, you drank alcohol on an empty stomach, or you do not tolerate some food component well (a particular protein, lactose, or , short-chain carbohydrates that are hard to digest), rather than MSG itself. Sorting it out takes a blinded test, not a gut feeling about the cause.
For how much natural glutamate shiitake contains, see Mushrooms; for sodium and blood pressure, see Potassium & Sodium.
Evidence · What is still unanswered
The most dangerous step in a debunking piece is turning "the charge does not stand" into "there is nothing left to think about." A few things are still open on this topic:
The mechanism of those rare true positive reactions is unclear. In the most rigorous trials, a very small number of people did have symptoms on real , but the reaction did not repeat on a second test. Whether that is leftover expectation, random fluctuation, or a real but unstable individual response has no answer yet. "No answer" and "does not exist" are two different sentences.The asthma line only reaches case reports. A case report is the weakest link in the chain of evidence: it can point to a direction worth checking, but it cannot prove cause. So it is written down here and marked as thin, neither deleted on your behalf nor blown up for you.Long-term, high-intake studies are scarce. The existing evidence centers on whether acute symptoms appear after a single dose. For the question "every day for decades," the data are much thinner, although natural foods already supply far more glutamate than seasoning amounts do, which makes a long-term risk hard to argue biologically. So this story's position is not "MSG is a good thing." It is that the charge against it does not stand up in the human evidence, while the things actually worth noticing (it contains sodium, and it often turns up in dishes that are already heavily seasoned) are the ones few people mention.
Keeping "the charge does not stand" apart from "nothing to think about" is more useful than remembering which seasoning is good or bad.
References · 8
Kwok, R. H. M. (1968). Chinese-restaurant syndrome [Letter to the editor]. New England Journal of Medicine, 278(14), 796. The anecdotal letter that originated the 'MSG symptom complex' narrative. 10.1056/NEJM196804042781419
Yamaguchi, S., & Ninomiya, K. (2000). Umami and food palatability. Journal of Nutrition, 130(4S Suppl), 921S-926S. Reviews the independent umami taste pathway and glutamate's natural occurrence in foods. 10.1093/jn/130.4.921S
Raiten, D. J., Talbot, J. M., & Fisher, K. D. (Eds.). (1995). Analysis of adverse reactions to monosodium glutamate (MSG), report prepared by FASEB for the U.S. FDA. Journal of Nutrition, 125(11), 2891S-2906S. Concluded MSG is generally safe at normal dietary intake. PubMed indexes the executive summary (PMID 7472671) without an abstract and the report text could not be re-read (publisher page blocked, 2026-09-25); FDA's MSG Q&A (fda-msg-qa) summarises this report's finding of transient, mild symptoms in some sensitive people after 3 g or more without food. 10.1093/jn/125.11.2891S
Geha, R. S., Beiser, A., Ren, C., et al. (2000). Review of alleged reaction to monosodium glutamate and outcome of a multicenter double-blind placebo-controlled study. Journal of Nutrition, 130(4S Suppl), 1058S-1062S. Double-blind placebo-controlled challenges failed to reliably reproduce MSG symptoms at dietary doses. Note: the id says JACI but this record is the J Nutr summary (PMID 10736382); the full trial report is geha-2000-msg-multicenter-challenge. Summary abstract: a multicenter double-blind placebo-controlled challenge in 130 people who reported MSG reactions; large doses given without food may elicit more symptoms than placebo, but responses were infrequent, inconsistent and not reproducible, and were not seen when MSG was given with food (abstract, PMID 10736382). 10.1093/jn/130.4.1058S
Geha, R. S., Beiser, A., Ren, C., Patterson, R., Greenberger, P. A., Grammer, L. C., et al. (2000). Multicenter, double-blind, placebo-controlled, multiple-challenge evaluation of reported reactions to monosodium glutamate. Journal of Allergy and Clinical Immunology, 106(5), 973-980. 130 self-selected people who reported MSG reactions were challenged with 5 g of MSG and with placebo without food (protocol A): 50 (38.5%) responded to MSG only, 17 (13.1%) to placebo only and 19 (14.6%) to both. Rising doses raised response rates, but only 19 of 37 MSG-only responders reacted the same way on retest; the 2 people who then responded to MSG but not placebo twice were challenged 3 times each with 5 g of MSG and with placebo in the presence of food, and each responded to only one MSG challenge. Conclusion: large doses without food may elicit more symptoms than placebo in people who believe they react, but no persistent or serious effects were seen and responses were not consistent on retesting (abstract, PMID 11080723). 10.1067/mai.2000.110794
U.S. Food and Drug Administration. (2026). Questions and answers on monosodium glutamate (MSG) (content current as of 17 July 2026). FDA considers the addition of MSG to foods 'generally recognized as safe' (GRAS). Adverse event reports led FDA to ask FASEB to examine MSG safety in the 1990s; FASEB's report concluded that MSG is safe, and identified short-term, transient and generally mild symptoms (headache, numbness, flushing, tingling, palpitations, drowsiness) that may occur in some sensitive individuals who consume 3 grams or more of MSG without food; consuming more than 3 grams of MSG without food at one time is unlikely. An average adult eats about 13 g of glutamate a day from food protein, against about 0.55 g a day of added MSG (Q&A page, read 2026-09-25). www.fda.gov/food/food-additives-petitions/questions-and-answers-monosodium-glutamate-msg
Jinap, S., & Hajeb, P. (2010). Glutamate: its applications in food and contribution to health. Appetite, 55(1), 1-10. Reviews glutamate/umami in food and the use of MSG to reduce added sodium while maintaining palatability. 10.1016/j.appet.2010.05.002
National Academies of Sciences, Engineering, and Medicine. (2019). Dietary Reference Intakes for Sodium and Potassium. National Academies Press. The PubMed book abstract (PMID 30844154) gives no reference values; it says the report updates the sodium and potassium DRIs with an expanded model that adds chronic-disease endpoints (the Chronic Disease Risk Reduction Intake). The numeric AIs and CDRRs were not re-read for this note (abstract). www.ncbi.nlm.nih.gov/books/NBK538102