Sunday, September 11, 2011

Carbs and Serotonin, A Connection After All?

A few weeks ago in Do Carbs Keep You Sane, I reported from a couple papers that disagreed with the textbook theory that a high carb, low protein and low fat diet would increase tryptophan in the brain.  The Wurtmans from MIT have designed a whole pharmacologic diet around this theory, so it was interesting to read the rebuttal, especially since the rebuttal included data from Dr. Judith Wurtman's own papers.

In short, the theory goes that carbohydrate ingestion stimulates insulin production, which in turn causes protein to be driven out of the bloodstream and into the cells.  Tryptophan, the rarest amino acid in the diet and the precursor for serotonin, is mostly bound in the blood to another protein called albumin, which makes it immune to insulin's effects.  Therefore a carb bolus will increase the ratio of tryptophan to other amino acids competing for the same receptor, tryptophan shoots into the brain, and you get a nice hit of satiating, serenity-making serotonin.

If we follow the lines of this theory, a high protein diet will increase the amount of other amino acids and increase the competition for the receptor, leaving tryptophan a loser and the brain relatively "low" in serotonin.  Fat in the diet will also delay gastric emptying and lower the overall glycemic index, lowering the insulin response and therefore reducing the insulin mechanism for driving tryptophan into the brain.  Pretty simple.

Except in nutrition, nothing is ever simple.  Turns out this mechanism works a bit differently in rodents than in humans or other primates, and any natural food and even flour and potatoes should have too much protein for it to work in humans.  You can get this effect after a night's fast by eating or drinking something that is pure carbohydrate - such as marshmallows or lemonade.  Not exactly an evolutionary model.  In fact, in the primate models, the amount of tryptophan that made it into the brain depended on a higher amount of protein, not a lower amount with higher carbs.

But Mr. Jamie Scott sent me the Pubmed link for this paper a few weeks ago, and I certainly don't like to ignore papers, even if they tell a different story than the majority of the papers I had seen thus far:

High glycaemic index and glycaemic load meals increase the availability of tryptophan in healthy volunteers

Interesting study.  10 healthy human male university student volunteers are given several diets of varying macronutrient composition on different days.  The first meal was a high GI meal consisting of 768 calories of jasmine rise and a tomato pureee - this meal was 1.6% fat, 8% protein, and 90.4% carbohydrate.  The glycemic index of the meal was 117 and the glycemic load 200.  That's 171 grams of carbohydrate, in case you were wondering.  The other two "mixed macronutrient" meals served were lower in calories (about 457 each) of either a lower glycemic rice + or a high glycemic rice with a Lean Cuisine chicken.  (I kid you not).  The latter two meals were about 16% fat, 18% protein, and 66% carbohydrate give or take a rounding up or down, and each meal had 75 grams of carbohydrate.

(Okay, another hysterical sentence in this paper - each volunteer was tested with a standardized glucose drink to calculate glucose and insulin response to the different meals - the standard was the 75 grams of glucose bolus, and the figures were extrapolated to estimate the response to the 171 grams of carbohydrate meal because "it was considered unethical to give a glucose reference drink of 171.4g CHO.")

The results?  The young men found the two mixed macronutrient meals palatable, whereas the (double calorie) high carb, high GI meal was more satiating, but less palatable.  Sleepiness did not differ when measured immediately after the meal.

Only seven of the participants participated in all the blood draws, so only seven data sets were used for comparison of the ratio of tryptophan (TRY) to other "large neutral" amino acids (LNAA) in the study.  At baseline (which was fasting), the ratio did not differ between the subjects.  After the high carbohydrate, high GI meal, TRY:LNAA ratios increased by 23% and remained high for the next 8 hours.  The Lean Cuisine folks with the low GI rice had an increase in TRY:LNAA of about 8%, and the high GI Lean Cuisine folks had an increase in TRY:LNAA of 17%.

The high carbohydrate bolus in several studies brings out robust high insulin responses - several hundredfold percentage increases over baseline (in this study the increase in insulin was 650% over baseline).  In another study, plasma platelet serotonin levels were increased 3.5 fold after a similar high carb meal.  Of course, high amounts of serotonin floating around in your periphery may not be exactly good for you - it is thought the high levels of serotonin caused by the diet drug combination fen-phen caused the aortic calcification risk from those drugs and possibly the risk of increased pulmonary hypertension.

In other studies, there are interesting implications.  Recall that serotonin is the precursor for melatonin - in this rather recent study, a high carbohydrate, high GI meal (high GI rice, very very low fat) four hours before bedtime decreased sleep onset by 50% compared to those who consumed a low GI rice meal (also very low fat).

So it looks that you can increase tryptophan ratios in the periphery with high glycemic index meals, even with a more realistic macronutrient mix than jasmine rice and tomato puree.  Presumably this may increase tryptophan intake into the brain.  Tryptophan has several fates in the brain, serotonin being one, kynurenic another depending upon your state of inflammation and what drugs you may be on.  Once again we have one fragment of a hugely complex picture.

And, since this mechanism depends on insulin, if one is contending that a high carb "serotonin cure" diet is helpful for depression, one must take into account that people with severe hyperinsulinemia are more likely to be depressed than people without, not less (though there are other confounding factors - since inflammation is one, the high carb diet in an (inflamed) type II diabetic might lead to increases in kynurenic rather than serotonin, explaining the diffference… but you see there is way too much unknown to make any general prescription for high carb diets in this context.)


  1. I wanna know what that ethics committee was huffing so I can get me some.

  2. …especially since the researchers turn right around and shovel 171g of starch down the subjects!

  3. Maybe I'm completely misunderstanding this, but can the high levels of tryptophan simply be tryptophan resistance?

  4. Dr. Deans, do our serotonin levels rise as our insulin levels rise or is there something in the actual carbohydrates that cause us to produce serotonin?

    I'm curious because I've been following a VLC meat based diet and I want to make sure I can still get enough serotonin on this diet. So far I've felt that my brain get's too revved up and I suspect it's related to too much dopamine/BDNF and too little serotonin/prolactin, but if I can just raise my glucose levels to get more serotonin then I won't worry

    Great blog!

  5. Love this!! Emily... I love to see the talk of high carb and insulin production. It would make sense that high carb does raise seratonin levels and that would help lift depression; however, if I, (with PCOS T2 Diabetes, Metabolic Syndrome) eat too much carb, my brain will physically crave more and more, and then the insulin will increase and that starts the whole vicious cycle of weight gain and metabolic disease. That's why I stick to not so much a high protien diet, but a balanced diet with "lower" carb and that carb is taken in a timed basis.

  6. Hey Emily,

    Very interesting post. It seems worth noting that people with severe hyperinsulinemia generally have it because they are deficient in insulin signaling. So perhaps that observation is very consistent with the idea that insulin has important anti-depressant action.


  7. js290 - interesting idea, haven't thought of it. My suggestion is that the ultra-high ratio of tryptophan (increased by 23%) is definitely not from the evolutionary history

    arbo - VLC/ketogenic diets have therapeutic uses but if you don't have to stay there, life can be easier if you eat some carbs. More variety, more options at restaurants and with friends and family, etc. That said, I don't know that research has the answer for you. Practically speaking, I've seen people who seem to have symptoms of "serotonin deficiency" on super low carb diets - which is NOT depression, by the way. More likely insomnia, agitation, and sometimes suicidal thoughts. This can also occur on higher carb diets, of course. Other people feel fantastic on ultra low carb.

    Lynn - interesting

    Chris - that makes too much sense. Sometimes I don't think holistically. Good thing my commenters will :)

  8. If you have depression and insomnia how about eating mostly low carb, protein and fat for breakfast and lunch and highish carb low fat dinner? Would that be a good treatment protocol?

  9. I am rather puzzled by one thing here. Judith Wurtman advocates eating 30g of starchy or sugary carbs, seeing how fructose has not got the same impact on blood sugar levels, fruit won't do. She of course forgets how not all fruit are created equal and whereas apples may be fructose heavy, other fruits are certainly not. Anyhow if what we need for tryptophan to be able to cross the blood-brain barrier is insulin, how could 30g of bran and 30g of rice krispies have the same effect, when their glycemic loads are different? And why can't dates, which are extremely high on the glycemic load scale work??

    Please someone clear this up for me...

  10. Late to this article, but I do have one thought to add, in response to this;

    "You can get this effect after a night's fast by eating or drinking something that is pure carbohydrate - such as marshmallows or lemonade. Not exactly an evolutionary model."

    Well, there is *one* evolutionary source of (nearly) pure carbohydrate - honey.
    And there is the grandma advice (including my own grandmother) to have a teaspoon of honey before bed for a good sleep.

    This would give a shot of (almost) pure carbs on an empty stomach, in the dark -seems consistent with the carb-serotonin-melatonin theory?

    Some people advocate taking honey with warm milk, which has tryptophan in it, but I don;t know how they whey and casein proteins would interfere with the honey.

  11. Has anyone read Potatoes not Prozac? Eating a high protien diet during the day and a small potato with its skin on before bed helps the brain utilize tryptophan. People swear by it!

  12. I get a an antidepressant effect from low carb diets and/or feel "revved up" as arbo put it. I know serotonin isn't the only factor that comes into play when it comes to depression, so I was just curious as to what you or anyone else thinks about why some people get such a huge antidepressant effect from low/no carb diets?


Tired of receiving spam comments! Sorry, no new comments on the blog

Note: Only a member of this blog may post a comment.