Serine for cognitive enhancement
Why would serine be important for preventing neurodegenerative diseases?
First some background:
A new animal study on combined metabolic activators (CMA) for the prevention of neurodegenerative diseases caught my eye.[ref] The study looked at using a combination of readily available supplements to improve cellular metabolism — and thus improve cognitive function in aging.
This animal study is a follow-up to last year’s clinical trial in older adults who tested out a “combined metabolic activators” stack consisting of:[ref]
12.35 g L-serine
1 g nicotinamide riboside
2.55 g N-acetyl-L-cysteine
3.73 g L-carnitine tartrate
(I wrote about the CMA phase III clinical trial here.)
The new study looked at both Alzheimer’s and Parkinson’s in animal models and compared two different combined metabolic activator (CMA) stacks - CMA1 (nicotinamide riboside, L-serine, N-acetyl cysteine, L-carnitine tartrate), and CMA2 (nicotinamide, L-serine, N-acetyl cysteine, L-carnitine tartrate). Both CMA1 and CMA2 (the difference being nicotinamide riboside vs nicotinamide) “markedly enhanced metabolic and behavioral outcomes, aligning with neuro-histological observations.” They also tested the individual components.
This got me wondering - why serine?
I understand that nicotinamide (or NR) improves cellular function by adding to the NAD+ pool. It seems like every reaction under the sun uses NAD. And I get that N-acetylcysteine (NAC) increases glutathione, staving off oxidative stress in cells. Similarly, carnitine will directly improve cellular energy production.
What is l-serine?
Serine is an amino acid - a building block of proteins - and it comes in either the L- or D-serine form. L-serine is the form found in food, and d-serine is made in the body.
The body can synthesize serine from glycine or from glycolysis, so it is considered a non-essential amino acid. Plus, the l-serine form can also be converted to d-serine for use as well. In fact, there’s an enzyme in astrocytes that specifically converts l-serine to d-serine in the brain.[ref]
Serine is then used in the biosynthesis of DNA molecules as well as a bunch of proteins in the body.
Ok - so serine is important for cellular growth and cellular health, but why is it critical in the brain? What goes wrong in aging?
What serine does in the brain:
Serine is a neuromodulator. NMDA receptors on neurons are activated by glutamate plus either glycine or serine. In other words, the NMDA receptors need both glutamate plus serine/glycine for activation, and serine actually binds better there.
Glutamate is an excitatory neurotransmitter, firing the glutamatergic neurons in the brain which are needed for memory, learning, and focus. But it needs to be in balance. Too much glutamate is excitotoxic, killing neurons. Too little activation of glutamatergic neurons gives you memory and learning problems.
I mentioned above that serine can be derived from glycolysis, the process of converting glucose to ATP and other metabolites.
Alzheimer's disease is often called type 3 diabetes (by carbohydrate-hating health gurus). It's called this because there is a dysregulation of the brain's ability to use glucose in Alzheimer's patients, but it's not due to a lack of insulin.
Putting the pieces together, reduced glycolysis in the brain also could mean reduced serine levels.
A study in Cell Metabolism explains: “A decrease in regional brain glucose consumption is observed in early Alzheimer’s disease, decades before neuronal death and clinical symptoms occur.”
The study goes on to explore how defective glycolysis in astrocytes causes decreased l-serine production. The decrease in brain glycolysis precedes the accumulation of amyloid-beta plaque and tau tangles in Alzheimer’s animal models. The authors of the study explain that not only does the decreased glycolysis reduce ATP availability, but it also decreases the serine needed for synaptic transmission.
The researchers measured brain glucose metabolism, glycolysis, and serine levels in the brains of Alzheimer’s-prone mice at young and middle age. They found that by middle age, there were regions of the brain that had lower glycolysis and lower serine levels. The researchers then looked at the NMDA receptors on glutamatergic neurons. They found that serine was lacking in the NMDA receptor binding sites, reducing synaptic stimulation. In addition, they were able to rescue the synaptic plasticity by adding supplemental serine. The extra serine added to their drinking water helped restore the mice’s spatial memory.
In Parkinson’s, serine is likely also important in its interaction with the NMDA receptor.[ref - another interesting study]
To sum up:
Serine is preferentially needed for NMDA receptors, which impacts learning, memory, and focus. Impaired glycolysis in older brains not only reduces ATP but also reduces serine levels.
To me, this explains why a ketogenic diet, which provides ATP without glycolysis, helps a little in Alzheimer’s but doesn’t really fix the whole problem. This also explains why glycine supplementation can be helpful for brain function. (Longevity Lifehacks article on glycine)
More studies on serine:
In animals, d-serine supplementation clearly reverses age-related declines in brain function.[ref]
Serine for Alzheimer’s, Parkinson’s, and ALS has been studied for decades by the Brain Chemistry Labs out of Jackson, WY. Check out their website for the background of how they found the serine link to neurodegeneration and their current research.
A clinical trial in healthy older adults (average age 73) in Brazil found that serine supplementation helped improve executive function and spatial memory. The study participants took a 30mg/kg dose of serine 1.5 hours before the cognitive tests.[ref]
The NMDA receptor is likely involved in schizophrenia. A number of clinical trials over the last twenty years have looked at adding serine to antipsychotic treatments for schizophrenia. A meta-analysis showed that d-serine levels are lower and that supplemental serine helped with symptoms.[ref]
Serine levels likely play a role in response to ketamine for depression. A clinical trial showed that participants with low serine had a much better response to ketamine for treatment-resistant depression. Ketamine also interacts with glutamatergic neurons.[ref]
Adding in more serine:
Safety first: what are the side effects?
Most people get 3.5-8 g/day of serine from their diet. Doses of up to 25g/day are used in clinical trials. Possible side effects are bloating or upset stomach at high doses.[ref] A trial in healthy adults found no observable adverse effects at 12g/day.[ref]
Keep in mind that supplemental serine may affect mood (lots of clinical trials for schizophrenia and mood disorders). Talk with your doctor if you’re on medications, especially psychiatric medications. Excess activation of NMDA receptors can be bad.
The Alzheimer’s trial with serine as part of the CMA stack used 12g/day.
Serine supplements:
Serine is readily available online in powdered form or capsules. It doesn’t have much taste, so you may want to avoid the capsules and simply add the powder to a drink. (Here’s why I avoid cellulose capsules.)
Note that the conversion of l-serine to d-serine uses the enzyme serine racemase along with vitamin B6 as a cofactor.[ref] Consider whether you could be low in B6 and if you need to add more in through supplements or diet.
Foods containing serine:
Foods high in serine include:[ref]
beef, pork
soy protein, tofu
seal meat, whale meat
cod, whitefish, caviar, mollusks
milk, cheese
pumpkin seeds, mixed nuts, peanut butter
egg whites
spirulina
Final thoughts:
Studies show that serine levels in the brain decline with age.[ref] My takeaway from the new study of CAM in mice is that low serine alone isn’t the whole problem in neurodegenerative diseases, but it plays a more important role than I had thought.
The studies linking decreased glycolysis in the brain to low serine — and thus changes to NMDA receptor function make a lot of sense. To me, this makes more sense than the idea that decreased glycolysis was causing a lack of ‘brain energy’. And it explains why a keto diet doesn’t cure dementia. (Yes, I know that ketones in the brain can be important in brain function in aging… I had just been puzzled as to why keto wasn’t more of a cure-all, if lack of energy was the whole problem.)
The questions that remain for me are - at what age does supplemental or increasing serine become important? is there a way to test it?
There is a variant of MTHFD1L (rs11754661) that you mention in a genetic lifehacks article on depression as a source of rumination, and I am homozygous (AA), which is pretty rare for that polymorphism. Because of that, I did a deep dive into it, and I believe I was able to trace its biggest impact back to serine, which might be of interest to you in this context. That SNP has also turned up in GWAS for alzheimer’s, by the way. Here is my understanding of how that polymorphism relates to serine. The rs11754661 A allele appears to reduce activity of the MTHFD1L enzyme, reducing the rate at which 1-carbon-units can be extracted from serine or glycine, since it catalyzes the final step in the mitochondrial pathway to produce formate. The "workaround" is to shift more of the burden to the cytosol, which does not have the ability to extract 1-carbon units from glycine, only serine, since there is no glycine cleavage system in the cytosol. The end result, I believe, is that it forces your cells to synthesize serine through glycolysis even when there is glycine available that could otherwise be used in the mitochondrial pathway. Another way to look at it is that one molecule of pyruvate can produce one molecule of serine, which has two carbons that can be extracted in the folate cycle. But if demand for 1-carbon-units is high, the mitochondrial pathway gets backed up. The cytosolic pathway can only get one carbon from each serine, and therefore demands twice as much pyruvate be converted to serine. I think the end result is an excess of glycine, and not enough lactate, or ATP, or D-serine, or whatever else the pyruvate could have been used for. So when demand for 1-carbon-units is high, the cytosolic pathway is like the afterburners of a jet engine, an expensive way to boost output. And for people with the A allele of rs11754661, their astrocytes or other cells may need to use their “afterburners” more than most people. I don't know all the consequences, but this article on serine made me think about that rabbit hole I went down after I read your article "Depression, genetics, and mitochondrial function" on genetic lifehacks. 😄 (thank you for these articles, I really enjoy them!)