This interesting study discovered an enzyme that temporarily lowers BG without requiring insulin:
Medical News Today: Could this amino acid improve glucose control in diabetes?.
It appears that the amino acid [alanine] activates an enzyme called AMP kinase (AMPK) that increases energy production in cells. This results in a short-term reduction in blood glucose that does not involve insulin.
This could turn into a pretty cool pre-meal pill someday 
Yep, see post here and references to 5’ AMP-activated protein kinase => How exercise increases the effect of insulin and carb usage
Basically the same thing you get while exercising, but would be easier to activate with a pill for a meal.
Alanine supplements (beta and alpha isomers, but this article is about alpha) have been on the market for decades.
Unfortunately, taking Alanine straight actually increases blood sugar, and is indicated to treat T1D hypos https://www.webmd.com/vitamins/ai/ingredientmono-1247/alpha-alanine
But it is dirt cheap, so anyone can test this out themselves https://www.bulksupplements.com/l-alanine.html
How intriguing! Although it is not the first time than mice and human metabolisms lead to opposite effects…
It is two differnt things, beta-alanine versus l-alanine (the most common form of alpha-alanine).
In l-alanine (alpha-alanine) the amino and the carboxy groups are attached to the alpha carbon, which is why it is called an alpha amino acid.
In the beta-alanine, the amino group is not attached to the alpha carbon, it is attached to the carbon that is adjacent.
@Eric Both the research and myself are talking about L-Alpha-Alanine.
I only stated that the other, similar molecule exists. I take it (beta) and it has no noticeable effect on my blood sugar.
I’m familiar with Alanine from research into Glucagon long ago. Researchers use it to stimulate glucagon and hepatic release to test alpha cell and liver function.
I’m equally surprised they think it will improve glycemic control, unless they mean restoring glucagon response to hypoglycemia. But it didn’t sound like it…
Here’s another
Yes, I know. I just wanted to illustrate that there are two different things, so that when people hear “alanine” they know there are two distinct amino acids, and are not confused by that.
I was disappointed that the original article that was referenced did not specify of which one they were speaking!
Regardless of all of that, I do not have a personal opinion on whether it works. I do know that I see a lot of change from the contraction mediated pathway, which has an effect on AMPK that is similar to what the supposed effect of l-alanine is, which is why I found this topic interesting.
As for this study, the full article is here (and is properly titled, science journalists are lazy):
Mice were fed orally 1.5g/kg. The human equivalent dose would be 1.5g/12.3*60 for a 60kg human. So if going to design an experiment, the starting dose should be between 5-10g (about 0.50 cents worth, cheap experiment!)
Chemicals on the way. Will attempt to replicate their study for humans over next month. Will update when have results.
Self-Study Design
Purpose:
To see the effects of the oral administration of l-alpha-alanine & beta-alanine on hepatic release and glucose tolerance.
Methods:
TEST 1. After 4hrs of stable fasting BGs, but no longer than a 10hr fast to not deplete liver glucagon, take human-equivalent dose (HED) of 8.5g L-alanine. Measure response with CGM over next 2-3 hours.
TEST 2. Same for 4.8g Beta-Alanine
TEST 3. Do oral glucose tolerance test (OGTT) with 75g dextrose/glucose.
TEST 4. Take 8.5g L-Alanine 15min before OGTT.
TEST 5. Consider the same with Beta-alanine, depending on test 2 result.
Visually compare (overlay) graphs to each other. Estimate AUCs if possible or necessary.
OGTT is the standard for testing diabetic response, but uses over 12x the amount of glucose than the predicted HED (6g). The 75g dextrose may overwhelm Alanine’s assumed positive effects on glucose tolerance, but if there is an effect, it should be easier to compare CGM results than very small doses of glucose.
The Intraperitoneal glucose tolerance test (IPGTT) of original study cannot be replicated. I don’t want to inject myself in the guts with these substances. Sorry.
HEDs are a guideline for safe starting doses in human studies, but current human dosing recommendations of L-Alanine for other treatments are much higher:
Schizophrenia: 7g/day for 6 weeks
Glucagon Storage Disease Type 2: 10g 3x/day
Insulin-induced hypoglycemia: 20-40g
Prevent nighttime hypoglycemia: 40g + 10g glucose
My diabetes is very atypical so results may not have relevance for diabetics with different systemic dysfunction.
For example, AFAI and doctors know, my alpha-cell and glucagon response is normal, but never tested. Also, my HOMA-IR indicates very high insulin sensitivity, which may not allow further stimulation by AMPk. More typical diabetics would need to replicate these tests.
However, original study found significant effects for both diabetic and normal mice.
Study will be n=1, mouse study had multiple groups of n=6, thus underpowered. Multiple tests would be necessary to improve results and not willing!
Multiple OGTTs may exhaust β-cells for subsequent tests.
May need to retest at higher doses if no effect found at HED of 8.5g. And consider the 1.5:1 ratio of Alanine:Glucose used in the study.
However, if such a high daily dose of Alanine is necessary (up to 0.5 kilograms/day for 300g carbs), it would be extremely impractical as a treatment. Perhaps even dangerous or toxic to the liver and glucagon system.
This is quite interesting to me. I would love to see what you report when you are done.
If we can obtain some here, and if you report some interesting results, we could do another test as well 
