This blog presents an idea that, based on current research, appears likely true and may save your life:
Taking intermittent “protein breaks,” when you eat very little or no protein for 2-4 consecutive days while eating plenty of carbohydrate, can slow aging and prevent, delay, or reverse — at least partially — many diseases including obesity, type 2 diabetes, autoimmune disorders, and brain diseases such as Alzheimer’s and Parkinson’s.
If you take protein breaks, you must follow each by a period of eating adequate amounts of protein.
This hypothesis awaits confirmation by randomized human clinical trials, but is consistent with prior studies in humans, other animals, and living cells.
Because no clinical trial has yet been done to establish the efficacy of protein breaks in humans, I cannot recommend them. But I recommend you consider them with your doctor.
protein.md 
Dr. Hill,
Can you quantify how much protein restriction constitutes a protein break?
Hi Eric,
During a protein break, less than 20 grams of protein per day should be most beneficial. It’s difficult to eat zero grams of protein and still obtain adequate calories, because so many foods contain some protein. Specially made foods could provide zero protein, though.
Thanks for your question.
If you are fully keto adapted, why on earth would one need to gorge on glycating carbohydrates? Surely part of the benefit of a protein break would be to allow the beta cells of the pancreas to restore and repair, which certainly won’t happen if they have to produce substantial insulin to deal with the “plenty of carbohydrates”.
Hi NM,
During amino acid starvation, even in the presence of glucose and insulin, mTOR is inhibited, activating autophagy. The reason for eating carbs (no need to gorge, though) is to inhibit glycogen synthase kinase 3 (GSK3) via insulin and thereby activate TFEB by promoting its nuclear translocation. This stimulates lysosomal biogenesis, likely critical for preventing and treating Alzheimer’s, in which an autophagic failure lies not in autophagy induction, but in lysosomal acidification and proteolysis.
Eating carbohydrates, particularly during protein restriction, should not result in significant glycation in persons with good insulin function. This is evidenced by low hemoglobin A1C (HbA1c) levels in nondiabetics — having good glucose tolerance — who eat high-percent carbohydrate diets.
You say to do high carbs for 2-4 days followed by adequate protein. How much protein would be considered adequate, and for how long before you switch back to carbs? Is there success in this theory so far?
Thanks. So what happens in traditionally low-carbohydrate tribes in respect to autophagy and so forth? You are postulating a fairly critical role for occasional dietary carbohydrate consumption, which even the USDA et al cannot find final evidence. Genuinely interested if strict keto-adaptation in the medium or long term is, in your opinion, dangerous. And, if so, what is the implication of this conclusion on our species’ long icy periods therein?
Hi Carlita,
The US RDA for protein of 0.8 grams per kg of bodyweight, or about 40 to 70 grams per day for average adults, is more than adequate for most.
Each period of adequate protein intake should probably be at least twice as long as any preceding period of severe protein restriction. This is roughly consistent with animal studies utilizing 30% protein restriction.
Intermittent protein restriction has proven beneficial in mice genetically engineered to have Alzheimer’s disease, in humans for lowering fasting blood glucose and insulin and improving insulin sensitivity, and in lowering insulin requirements in diabetics. Anecdotally, I have seen a number of overweight patients lose weight with intermittent periods of protein restriction and high-percent carbohydrate intake.
Hi NM,
“So what happens in traditionally low-carbohydrate tribes in respect to autophagy and so forth?”
Basal autophagy in everyone occurs independently of diet, especially in the brain, and declines with age. When low-carb-eating tribes have a break in their protein intake, starvation-induced autophagy is then also stimulated.
“You are postulating a fairly critical role for occasional dietary carbohydrate consumption, which even the USDA et al cannot find final evidence.”
I don’t postulate a need for carbohydrates to survive, as humans are well adapted through ketosis to eat zero carbohydrates indefinitely. Instead I propose a benefit to carbohydrate consumption during periods of very low protein intake, to stimulate lysosomal biogenesis via TFEB and consequent autophagic disposal of misfolded proteins and damaged organelles. This disposal is hampered in Alzheimer’s, a disease that evolution has had scant opportunity to prevent, as its usual onset after age 60 occurs when natural selection and sexual selection no longer operate on individuals and their genes.
“Genuinely interested if strict keto-adaptation in the medium or long term is, in your opinion, dangerous. And, if so, what is the implication of this conclusion on our species’ long icy periods therein?”
I have not heard of any medium- or long-term problems with keto-adaptation, though I have not recently surveyed the literature on this topic. Some of our ancestors evidently survived for long periods consuming virtually no carbohydrate — for example, eating just fish. But I suspect that many of their brains accumulated, or would have accumulated, aggregates of misfolded proteins, especially after age 70, as occurs today in many neurodegenerative diseases such as Alzheimer’s and Parkinson’s. Huntington’s and Lou Gehrig’s disease (ALS) are also diseases of protein misfolding in the brain, though their onset is usually earlier than age 70.
During periods of adequate protein intake between protein breaks, I see no reason why one can’t restrict carbohydrates.