AMPK is a key regulator of metabolic rate and fat burning. The mainline thinking on AMPK often follows the puritanical view of classic obesity research: AMPK senses LOW cellular energy levels by monitoring the ratio of ATP to ADP and AMP.1
This suggests that the best way to stimulate AMPK is by reducing caloric intake. When you fast, after all, AMPK is activated. If you want to burn fat, eat less and AMPK will be stimulated.
The problem with this is that it fails to take into account the fact that starch eating cultures like the tsimane farmer-foragers – who subsist on low linoleic acid, high starch foods such as plantain and cassava – have large caloric consumption and very high metabolic rates.2 Clearly they didn’t get lean from lowering calories consumed. How does this work?
Quick review on ATP, ADP and AMP
AMP is adenosine mono-phosphate. ADP is adenosine di-phosphate and ATP is adenosine tri-phosphate. So to go from ADP to ADP to ATP you just add phosphate groups. Each phosphate bond is a high energy bond. So ATP has more stored energy than does ADP.
When glucose is converted to pyruvate in the cytosol (cell water), the first few steps actually use ATP as an energy source to break the glucose down into glyceraldehyde-3-phosphate (G3P). The second half of glycolysis – resulting in pyruvate which enters the mitochondria – produces more ATP than was used to make the G3P, so the net result of glycolysis is more ATP.
When your mitochondria burns pyruvate via the krebs cycle, much more ATP is produced. So after a starchy meal, glycolysis is proceeding all the way through the krebs cycle and ATP levels are maximal. AMPK is activated by LOW ATP levels, so of course high cellular energy (high ATP) means minimally activated AMPK. That’s the common thinking.
Peter Explains Cellular Satiety after a Hi-Carb meal
Peter at Hyperlipid recently wrote this amazing post explaining how a starchy meal can lead to hydrogen peroxide production via the activity of the glycerophosphate shuttle. At the time at which the hydrogen peroxide is produced, energy levels in the cell are maximal. Glucose is high, lactose is high, The NADH/NAD+ ratio is high. ATP levels are maximal. Some amount of lipogenesis (fat making) is undoubtedly happening due to the stimulation of lipogenic genes such as ACC, FAS and SCD1 from insulin signalling. All of these things are forms of cellular energy. Lactate, NADH, Glucose, ATP, fat from lipogenesis. Cellular energy is high.
As Peter explains, high cytosolic NADH levels mean that NADH will be passed into the mitochondria via the glycerophosphate shuttle. This acts as an FADH2 input to the mitochondrial electron transport chain and creates hydrogen peroxide for the same reasons that saturated fat does as I explain in The ROS Theory of Obesity.
Hydrogen peroxide is generated in response to high cellular energy after a starchy meal.
Hydrogen Peroxide Stimulates AMPK through indirect means
It has been known since 2001 that adding hydrogen peroxide to cell culture results in a rapid activation of AMPK.3 Recently it was shown that the activation of AMPK by hydrogen peroxide happens via an increase in the ADP/ATP ratio, NOT through direct activation.4 But how does that work in our cell after a starchy meal? We have seen that NADH, ATP, glucose and lactate levels are all high.
Glyceraldehyde-3-Phosphate Dehydrogenase is a redox sensitive enzyme
I’ve talked about redox sensitive enzymes in Hydrogen Peroxide Flips The Switch.
Remember how I said that the first part of glycolysis consumes ATP, converting it to ADP? It’s only the second half of glycolysis that creates ATP and provides pyruvate as a starting material for the TCA cycle, which produces even more ATP. The bridge between the ATP consuming part of glucose metabolism and the ATP producing parts is an enzyme called Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH). It is highly redox sensitive and it’s activity level goes to zero when hydrogen peroxide levels are elevated to normal post-prandial physiological levels.5
So when high cellular energy levels lead to ROS production, GAPDH activity stops, glucose is converted to Glyceraldehyde-3-Phosphate, which converts ATP to ADP and ATP production slows. This results in a rapid increase in the ADP to ATP ratio and AMPK is activated.
AMPK is activated due to HIGH cellular energy levels after a starchy meal.
NOTE: If you read Peter’s article, you’ll see that he talks about glycerol-3-phosphate dehydrogenase. This is DIFFERENT FROM Glyceraldehyde-3-Phosphate Dehydrogenase.
AMPK turns off glucose metabolism and turns on fat metabolism
AMPK activates a suite of metabolic changes that switch a cell over from using glucose in response to insulin, to using fat. It activates PPAR alpha (PPARa)6, which in turn turns on PDK4 and CPT17. PDK4 turns off pyruvate dehydrogenase, which is what converts pyruvate generated through glycolysis to acetyl-CoA, allowing it to enter the mitochondria. CPT1 is the rate limiting enzyme that controls how fast fat can enter the mitochondria. More PDK4 means less glucose oxidation and more CPT1 means more fat oxidation.
AMPK also phosphorylates ACC and turns it OFF.8 ACC converts Acetyl-CoA to malonyl-CoA – the rate limiting step of lipogenesis (fat making). AMPK also phosphorylates SREBP-1c and turns it OFF. SREBP-1c turns on the lipogenic (fat making) genes.
AMPK turns off glucose burning and lipogenesis, and turns on fat burning in response to the redox inactivation of GAPDH caused by high cellular energy levels after a starchy meal.
- 1.Lin S-C, Hardie DG. AMPK: Sensing Glucose as well as Cellular Energy Status. Cell Metabolism. Published online February 2018:299-313. doi:10.1016/j.cmet.2017.10.009
- 2.Gurven MD, Trumble BC, Stieglitz J, et al. High resting metabolic rate among Amazonian forager-horticulturalists experiencing high pathogen burden. Am J Phys Anthropol. Published online July 4, 2016:414-425. doi:10.1002/ajpa.23040
- 3.Choi S-L, Kim S-J, Lee K-T, et al. The Regulation of AMP-Activated Protein Kinase by H2O2. Biochemical and Biophysical Research Communications. Published online September 2001:92-97. doi:10.1006/bbrc.2001.5544
- 4.Hinchy EC, Gruszczyk AV, Willows R, et al. Mitochondria-derived ROS activate AMP-activated protein kinase (AMPK) indirectly. Journal of Biological Chemistry. Published online November 2018:17208-17217. doi:10.1074/jbc.ra118.002579
- 5.V. Danshina, E. V. Schmalhausen, A. P. Mildly Oxidized Glyceraldehyde-3-Phosphate Dehydrogenase as a Possible Regulator of Glycolysis. IUBMB Life (International Union of Biochemistry and Molecular Biology: Life). Published online May 1, 2001:309-314. doi:10.1080/152165401317190824
- 6.Lee WJ, Kim M, Park H-S, et al. AMPK activation increases fatty acid oxidation in skeletal muscle by activating PPARα and PGC-1. Biochemical and Biophysical Research Communications. Published online February 2006:291-295. doi:10.1016/j.bbrc.2005.12.011
- 7.Pettersen IKN, Tusubira D, Ashrafi H, et al. Upregulated PDK4 expression is a sensitive marker of increased fatty acid oxidation. Mitochondrion. Published online November 2019:97-110. doi:10.1016/j.mito.2019.07.009
- 8.Park SH, Gammon SR, Knippers JD, Paulsen SR, Rubink DS, Winder WW. Phosphorylation-activity relationships of AMPK and acetyl-CoA carboxylase in muscle. Journal of Applied Physiology. Published online June 1, 2002:2475-2482. doi:10.1152/japplphysiol.00071.2002
Continuing with the fire analogy, or internal combustion engines…
This is like revving up 1st gear on your starch, then pressing the clutch pedal (GADPH stops), shifting to 2nd gear (production of glyceraldehyde-3-phosphate), let’s assume the clutch pedal releases & our analogy breaks down briefly, then laying into the gas (ATP->ADP) with fat burning (^ AMPK). Wow!
Want to comment on one piece of his post-
“With plasma FFAs at around 0.1mM it doesn’t matter very much how many CD36 receptors are present on the cell surface, fatty acid oxidation will be a limited source of both NADH and FADH2 supply to the electron transport chain.”
That won’t exist for a lot of torpid individuals who are borderline type 2 diabetic, whose hypertrophic fat cells and ATGL(?) upregulation will be dumping FFAs liberally into the bloodstream even under high insulation conditions right? So for torpid people, this idea doesn’t really apply, just for the skinny/normo-adipotic individuals.
*High insulin conditions, not high insulation, did my brain turn into Office 365 spellcheck or something?
Yes, true. And there is an interesting question about what is entering the mitochondria in terms of fat:carb ratios in the event of one eating a starchy meal while having elevated FFA.
Some of this sounds classic Ray Peat, as I interpret it.
Yes. Some of it does sound like Dr. Peat. But if I remember correctly he would not advocate for starch but fructose. In regards to the farmers above, how much is starch and how much is fat in the diet. This is all very interesting.
Very interesting switches. I do notice when I go to a special occasion and eat WAY more starch than normal I usually start burning up and get sweaty several hours later. My wife often comments that I feel like a radiator. Always assumed this was uncoupling but nice to see the dots connected here.
I so badly wish I could go vegan! but if i have a starch meal a couple hours later my blood sugar drops like a stone and i have to RUSH to the kitchen to get food…
I eat grass fed grass finished ground beef with some rice and veggies… even with this combo my muscles are weaker than if i left out the rice
🙁
Interesting. That sounds like what Peter from Hyperlipid would call “pathological insulin sensitivity”. This is where your cells cannot generate ROS to prevent the glucose from being taken up by cells. This sounds like the pigs in the second column.
Some say tsimane are favoring high metabolism due to infectious burden. Metabolic rate has been slowing last few years. Tsimane Project out of UCSB data. Possible changes hygiene etc.
The researchers studied this and found that 1) infectious burden only accounted for a small of the increase in metabolic rate in general and 2) falling infectious rates failed to explain the decline in age-related drop in energy expenditure.
might need a chart for this post..
Wait a second. I posted on twitter asking you about Stearic acid and ROS and AMPK, now I just had an epiphany-
Does Stearic acid trigger AMPK via ROS – ONLY – in the context of high glucose/starchy meal?
As explained here, a starchy meal can trigger AMPK by ROS:
—
“So when high cellular energy levels lead to ROS production, GAPDH activity stops, glucose is converted to Glyceraldehyde-3-Phosphate, which converts ATP to ADP and ATP production slows. This results in a rapid increase in the ADP to ATP ratio and AMPK is activated.
AMPK is activated due to HIGH cellular energy levels after a starchy meal.”
—
Stearic acid can increase ROS further. So can alcohol(? … not sure if it interferes with glucose though) …
So now: Anything that increases ROS in the presence of elevated glucose/starch, will trigger AMPK, and cause a massive upregulation in energy production, with combined elevated fat burning at the same time?
..which will independently upregulate GLUT4 per your latest blog post, and cause even more sugar to come into the cell, keeping the party going?
(Going to cross-post this in your latest blog post for good measure btw)
Something like that. What is clear is that lean humans have a higher fasted RER (respiratory exchange ratio), indicating that they burn more glucose, probably due to higher glycogen availability.
Curious then – not sure if this has been covered before – if lean humans also have higher intracellular ROS, in general. (Instinctually I would assume yes, as per the China study, they have more saturated fat)
Lean people generate more ROS but are also are better at converting it back to water. So their cellular levels are not higher even though they’re generating more.
Brad