Story
Blood-brain barrier · Getting in is not the same as working
Brain capillary gaps are sealed, so small molecules get in only by dissolving through membranes or using dedicated transporters, and efflux pumps push them back out. Getting through is not the same as working, nor failing to as useless.
Last updated
Overview and key questions
In short: Most things in the blood cannot simply get into the brain, because the gaps between the cells of the brain's capillaries are sealed shut.
Educational content, not medical advice. See a clinician about your own case.
Story path
Chapter 1
The brain's vessels are sealed shut
The endothelial cells do not hold this seal on their own. A layer of pericytes sits against the outside of the vessel, and further out, astrocytes reach over with their end-feet and wrap the vessel; together they help maintain the barrier.
Mouse experiments show that once pericytes are lacking, the barrier leaks. The leak is not the gaps between cells being forced open: the endothelial cells carry material in small vesicles from the blood side across to the brain side. So the barrier is a state that living cells keep up all the time, not a wall that is built once and left alone.
Among brain supplements, taken with antidepressants can cause serotonin syndrome: with a high fever, confusion or rigid muscles, go to the emergency department right away.
Mechanism · Where it leaks without pericytes
Pericytes look after two more things: they regulate the expression of a set of barrier-specific genes in the endothelial cells, and they make the astrocyte end-feet line up around the vessel facing the same way.
To note: all of this was measured in genetically modified mice, with no human measurement, and the study did not work out how much of the human barrier depends on pericytes.
Chapter 2
A few ways in, and pumps pushing back out
The first way is to dissolve straight through: if a molecule is small enough, dissolves in fat, and does not hold on tightly to water, it can dissolve into the membrane of the endothelial cell and pass from the blood side to the brain side.
The second way is a dedicated door. Things the brain needs but that cannot dissolve through the membrane have to be carried in by transporters: glucose goes through GLUT1, and large neutral amino acids go mainly through LAT1.
The pumps work in the opposite direction. Efflux pumps such as P-glycoprotein sit in the barrier and push many molecules that have already slipped into the membrane back into the blood.
So whether a molecule can get into the brain comes down to three things: is it small enough and fat-soluble enough; is there a door that recognizes it; and will a pump push it back out.
Numbers · What it takes to diffuse across
To get into the brain by dissolving through the cell membrane, a molecule roughly has to meet three conditions: a molecular mass below 400–500 daltons; high fat solubility; and fewer than 8–10 hydrogen bonds with water. The author treats these as rules of thumb with exceptions.
To note: this is a narrative review by a single author. The author also warns that focusing only on P-glycoprotein misses the other efflux transporters in the same family.
Mechanism · Two dedicated doors
The LAT1 door does not carry only the amino acids from food. In the human brain endothelial cell line hCMEC/, the drug gabapentin is a substrate of LAT1; at near-physiological concentrations, transport through LAT1 was about 3 times higher than through other routes, and about 10 times higher in transfected cells made to express LAT1. To note: this is a cell model, not a measurement in people.
Evidence · The efflux pump in mice and people
There is a matching observation in people. The anti-diarrhea drug loperamide is a potent opioid, yet at usual doses it does not act on the central nervous system, because it is a substrate of P-glycoprotein. In a , 8 healthy men given loperamide alone had no depression of breathing; given it together with the P-glycoprotein inhibitor quinidine, their breathing was depressed, and a rise in blood loperamide levels did not explain the change. The study did not measure drug levels in the brain; from these results the authors concluded that P-glycoprotein efflux at the barrier had been blocked.
So being small and fat-soluble does not guarantee a way in: a pump may push the molecule back into the blood.
Chapter 3
Oral GABA: whether it reaches the brain is still unmeasured
The long-standing view is that GABA cannot cross the blood-brain barrier, but the animal studies contradict each other: some say it cannot get in, others say a little can; and there is no pattern linking how it was given, which animal was used and what was concluded. In mice it has also been measured that GABA leaves the brain about 17 times faster than it enters.
Not getting through does not mean doing nothing, but acting by way of the gut is only a hypothesis for now.
Evidence · What the human trials measured
What was measured was also some distance from the brain. Many studies looked at peripheral markers such as and ; GABA in the blood rises after an oral dose, but whether it rises in the brain is not known.
To note: 11 of the 14 trials had company employees among their authors, and the first author of this review is also employed by Unilever.
So whether oral GABA helps with stress or sleep is still unsettled, and no one has yet measured in people whether it gets into the brain.
Mechanism · The gut route is still a hypothesis
There is some evidence for a calming effect, but the 2015 review points out that most of it comes from researchers with a potential conflict of interest. The next step they propose is direct: give people GABA, then measure the GABA in their brains with magnetic resonance spectroscopy.
Chapter 4
Tryptophan queues; 5-HTP turns into serotonin first
To get into the brain, tryptophan has to compete with the other large neutral amino acids for the same transporter; when branched-chain amino acids rise in the blood, less tryptophan gets in.
5-HTP is the direct raw material for serotonin. Before it reaches the brain, the aromatic amino acid decarboxylase outside the brain turns part of it into serotonin, and that part never reaches the brain. So how much 5-HTP gets into the brain depends on how much is converted outside it. This step outside the brain can be blocked by decarboxylase inhibitors such as carbidopa.
If you have depression or take antidepressants, ask your doctor before adding 5-HTP: taken together they can cause serotonin syndrome, which is an emergency.
Evidence · When an effect looks large
Looking only at the 2 studies that had a placebo group and a risk of bias that was not very high, the effect becomes uncertain: there may or may not be one. The results of the studies also varied widely.
In the authors' assessment, the existing studies are weak overall and few have a placebo group; they call for placebo-controlled trials in patients with a clear diagnosis. So an effect that looks large does not mean strong evidence.
Safety · Two problems the review lists
One is eosinophilia-myalgia syndrome. By 1998 there were about ten suspected cases worldwide, and none was firmly linked to 5-HTP.
The other is serotonin syndrome, an emergency caused by too much serotonin in the body. Mild cases can bring tremor, diarrhea, a fast heartbeat and sweating; severe ones bring confusion, rigid muscles and a high fever, and can be life-threatening. 5-HTP can interact seriously with antidepressants, and with migraine or pain medicines that affect serotonin. If these signs appear while you are taking them, go to the emergency department right away.
The authors think 5-HTP deserves another look, but judge it too early to recommend wide clinical use; they urge use under medical supervision and further clinical trials.
So if you have depression or take antidepressants, ask your doctor before adding 5-HTP or saffron, and do not swap or stack medicines on your own.
Chapter 5
Crossing is not the same as working: phosphatidylserine and saffron
Phosphatidylserine (PS) runs into trouble on the third question, and the earlier and later trials did not test the same thing. The early positive trial used PS from bovine brain; the soy PS tested later differs from it in fatty acid makeup and may not be the same substance. In a controlled trial of soy PS, older people with memory complaints took it for 12 weeks, and their memory and other cognitive measures did not differ from placebo.
Saffron runs into trouble on the first question. Its pigment crocin can barely cross a layer of gut cells; it must first be cut into crocetin inside the gut lining cells before it is absorbed, and in lab models, the molecule that slowly crosses the barrier is also crocetin. In people, 23 pooled together showed an effect on depressive symptoms, but with publication bias and trials from a narrow set of regions. If you have depression, talk to your doctor before using it.
So with any claim about feeding the brain, first ask which molecule actually gets in, then ask what has been measured in people.
Evidence · Bovine-brain PS and soy PS
A 2001 trial switched to soy: 120 people over 57 who met the criteria for age-associated memory impairment were randomly assigned to placebo or one of two soy PS groups, took it for 12 weeks, and were then followed for 3 weeks off it. On learning, memory, reaction time, planning and attention, there were no significant differences between the groups; the authors concluded that soy PS does not affect memory or other cognitive functions in these older people.
The difference between the two sources lies in the molecule itself: the bovine-brain form contains about 8% , while in the soy form about 47% is linoleic acid and 7% alpha-linolenic acid. The US FDA considered that they may not be the same substance. In 2003 the FDA reviewed 10 intervention studies, judged that all had serious flaws or limited reliability, and found the evidence that PS lowers the risk of dementia very limited and preliminary.
So the positive results for bovine-brain PS cannot simply be carried over to soy PS, and the trial of soy PS itself found no effect.
Mechanism · With saffron, another molecule gets in
Crocetin is a substrate of P-glycoprotein. In two lab models of the barrier made from pig cells, it crossed slowly but steadily over 29 hours of observation. To note: all of these are lab and tissue experiments, with no data from people.
The question of effect needs human trials. In an analysis pooling 23 , saffron did better than placebo for depressive and anxiety symptoms, but there was publication bias and the trials came from a narrow set of regions, so the conclusion may still change. If you have depression, talk to your doctor before using it, and do not add it to your medicines on your own.
So for saffron the key point comes first: you swallow crocin, but the molecule that reaches the brain is crocetin.
References (14)
- Daneman, R., & Prat, A. (2015). The blood-brain barrier. Cold Spring Harbor Perspectives in Biology, 7(1), a020412. 10.1101/cshperspect.a020412
- Armulik, A., Genové, G., Mäe, M., Nisancioglu, M. H., Wallgard, E., Niaudet, C., He, L., Norlin, J., Lindblom, P., Strittmatter, K., Johansson, B. R., & Betsholtz, C. (2010). Pericytes regulate the blood-brain barrier. Nature, 468(7323), 557-561. 10.1038/nature09522
- Boyer, E. W., & Shannon, M. (2005). The serotonin syndrome. New England Journal of Medicine, 352(11), 1112-1120. 10.1056/NEJMra041867
- Mekonnen, S. (n.d.). 5-HTP safety concerns. webPOISONCONTROL, National Capital Poison Center. www.poison.org/articles/5htp-safety-concerns-173
- Pardridge, W. M. (2005). The blood-brain barrier: Bottleneck in brain drug development. NeuroRx, 2(1), 3-14. 10.1602/neurorx.2.1.3
- Boonstra, E., de Kleijn, R., Colzato, L. S., Alkemade, A., Forstmann, B. U., & Nieuwenhuis, S. (2015). Neurotransmitters as food supplements: The effects of GABA on brain and behavior. Frontiers in Psychology, 6, 1520. 10.3389/fpsyg.2015.01520
- Hepsomali, P., Groeger, J. A., Nishihira, J., & Scholey, A. (2020). Effects of oral gamma-aminobutyric acid (GABA) administration on stress and sleep in humans: A systematic review. Frontiers in Neuroscience, 14, 923. 10.3389/fnins.2020.00923
- Fernstrom, J. D. (2013). Large neutral amino acids: dietary effects on brain neurochemistry and function. Amino Acids, 45(3), 419-430. 10.1007/s00726-012-1330-y
- Turner, E. H., Loftis, J. M., & Blackwell, A. D. (2006). Serotonin a la carte: Supplementation with the serotonin precursor 5-hydroxytryptophan. Pharmacology & Therapeutics, 109(3), 325-338. 10.1016/j.pharmthera.2005.06.004
- Cenacchi, T., Bertoldin, T., Farina, C., Fiori, M. G., & Crepaldi, G. (1993). Cognitive decline in the elderly: A double-blind, placebo-controlled multicenter study on efficacy of phosphatidylserine administration. Aging Clinical and Experimental Research, 5(2), 123-133. 10.1007/BF03324139
- U.S. Food and Drug Administration. (2003, May 13). Phosphatidylserine and cognitive dysfunction and dementia (Qualified health claim: Final decision letter). wayback.archive-it.org/7993/20171114183649/https://www.fda.gov/Food/IngredientsPackagingLabeling/LabelingNutrition/ucm072999.htm
- Jorissen, B. L., Brouns, F., Van Boxtel, M. P., Ponds, R. W., Verhey, F. R., Jolles, J., & Riedel, W. J. (2001). The influence of soy-derived phosphatidylserine on cognition in age-associated memory impairment. Nutritional Neuroscience, 4(2), 121-134. 10.1080/1028415X.2001.11747356
- Lautenschläger, M., Sendker, J., Hüwel, S., Galla, H. J., Brandt, S., Düfer, M., Riehemann, K., & Hensel, A. (2015). Intestinal formation of trans-crocetin from saffron extract (Crocus sativus L.) and in vitro permeation through intestinal and blood brain barrier. Phytomedicine, 22(1), 36-44. 10.1016/j.phymed.2014.10.009
- Marx, W., Lane, M., Rocks, T., Ruusunen, A., Loughman, A., Lopresti, A., Marshall, S., Berk, M., Jacka, F., & Dean, O. M. (2019). Effect of saffron supplementation on symptoms of depression and anxiety: A systematic review and meta-analysis. Nutrition Reviews, 77(8), 557-571. 10.1093/nutrit/nuz023
Written from the sources listed above. Checked sentence by sentence against them in . No physician reviews this site: it explains how the body works, and it is not medical advice. How we make content
Spotted an error, or something unclear? Email us. We read every message, check it against the sources, and fix what's wrong.