If you ask the average person, ketosis is primarily about carb restriction and fat intake. Go on a low-carb diet, eat more fat, allow your body to burn its own reserves. Pretty straightforward. Ketones are supposed to replace glucose.
But what do we make of protein? Some keto dieters avoid it like the plague, worried anything more than a quarter pound of animal flesh will knock them back into sugar-burning purgatory. Some have even likened it to “chocolate cake.” Others eat it freely. Who’s right?
The most restrictive therapeutic ketogenic diets, the ones used to treat childhood intractable epilepsy, are very low protein—around 5-10% of calories. These diets are designed to maximize ketone production. Any more protein than that and those kids might not make enough ketones to treat their condition.
The most ketogenic state of all—fasting—is also very low in protein. Zero, to be exact.
Okay, so protein can inhibit ketosis. Why? What’s going on?
One common assumption is that too much protein converts to glucose via gluconeogenesis. This is the “steak is just chocolate cake” hypothesis. It makes sense and sounds reasonable. It’s also completely wrong.
It turns out that gluconeogenesis follows an “as needed” schedule. Our livers don’t just mindlessly produce glucose anytime protein reaches a certain threshold. Our livers convert protein into glucose when we—for whatever reason—need more glucose. Demand-driven, not supply-driven. Keto-adapted individuals running over 70% of their brain on ketones and most of their muscle on fatty acids (which spares ketones for the brain) don’t demand a whole lot of glucose. Even under “optimal conditions “—giving a bunch of adults who just fasted overnight a big dose of radio-labeled protein and then tracking its fate through the body—humans convert very little dietary protein into glucose.
This isn’t a real issue.
What Causes Protein To Inhibit Ketosis?
It all starts with the Krebs cycle, that metabolic pathway that converts fatty acids into useable energy. In a “normal” cycle, fatty acids are broken down into acetyl-CoA. The liver pairs acetyl-CoA with oxaloacetate to complete the cycle and produce ATP energy. That’s basic energy generation.
Without oxaloacetate, the Krebs cycle cannot continue. Without oxaloacetate, acetyl-CoA has a different energetic fate: conversion into ketones. Where does oxaloacetate come from?
Carbs, usually. But protein can also be a source. Like carbohydrates, protein has the potential to donate oxaloacetate during the Krebs cycle. The more protein you eat, the more oxaloacetate you’ll have ready and willing to inhibit ketogenesis. This is how protein inhibits ketosis. Not by increasing gluconeogenesis. Not by spiking insulin.
By donating oxaloacetate.
How Much Protein Can You Eat and Still Remain Keto?
It depends on your goals and requirements.
If you’re dealing with serious epilepsy, creeping dementia, general inflammation, or anything else that requires or may improve with deep ketosis, aim for a lower protein content (10-15% of calories). Get those high ketone levels, see how it feels, and see if that’s the protein intake for you. Start low, really revel in those high ketone readings.
If you’re losing weight (or trying to), eat closer to 15-20%. For you, the ketone readings aren’t the biggest focus. How you look, feel, and perform are your main concern. Eating slightly more protein will increase satiety, making “eating less” a spontaneous, inadvertent thing that just happens. It will also stave off at least some portion of the lean mass accretion that occurs during weight loss; you want to lose body fat, not muscle.
If you’re trying to gain large amounts of muscle, eat closer to 20-25%.
Why You Shouldn’t Over-Restrict Protein
Just don’t go below 15% of your calories unless you absolutely need to. There’s a bottom. Protein is an incredible essential macronutrient. Fat is plentiful, even when you’re lean. Carbs we can produce from protein, if we really must, or we can just switch over to ketones and fats for the bulk of the energy that would otherwise come from carbs. Protein cannot be made. We have to eat it.
If we stop eating dietary fat, we’ll burn what we have on our bodies and—to a point—get healthier.
If we stop eating carbs, we’ll burn through our glycogen stores and then get better at burning fat. And we’ll be healthier.
If we stop eating protein, our organs, muscles, and bones will atrophy. Our health will suffer.
Another reason it’s so important (and so satiating) is that protein contains the most micronutrients. Fat-soluble vitamins are great, but the real good stuff we like—the B vitamins , the minerals—come packaged with protein.
How I’ve Changed My Approach To Protein
I think I need less than I used to think I needed. I eat maybe 80-100 grams a day max now. Some days a fair amount less, some days a fair amount more.
I also don’t think about protein meal-to-meal or even day-to-day. I tend to think of protein averages over three- or four-day chunks. If I get 200-250 grams in three days, I’m good and it doesn’t matter when or how I got it.
I know I’m in protein-sparing mode. We usually think of ketones as glucose-sparing, and they are. Generating (and being able to utilize) enough ketones to replace a large portion of the rare and flighty glucose is an invaluable asset in diseases of dysfunctional brain glucose metabolism like Alzheimer’s. Ketones are also protein-sparing. For one, if we aren’t burning through glucose, we don’t need any extra.
I make sure to eat a significant amount of collagen . Collagen  reduces amino acid requirements. It’s not enough by itself to stimulate muscle protein synthesis or provide the essential amino acids. It does help balance out muscle meat intake, reduce inflammation, improve sleep, speed up joint and connective tissue healing, and reduce the amount of protein I need to reach my nutritional goals.
Important to note, though…
I’ve been doing this fat-adapted thing  for a long time. My body is finely tuned to this kind of diet. It’s what it expects. People who are on week 2.5 of their keto journey might not have the same dynamic and may need more protein.
Keep in mind, too, that I’m not actively trying to gain muscle mass. The name of the game (for me) is to maintain: my body comp, my physical performance, my organ reserve, my health, my basic functionality. If I got the urge to put on lean muscle, I’d increase my protein intake.
Final Takeaways For Considering Protein Intake
- If you crave protein, you should eat it. Cravings for a natural, relatively unadorned food can usually be trusted.
- Know there aren’t any hard-and-fast rules about protein and ketosis. Everyone’s different. “Modified ketogenic” diets—higher in carbs and protein—are still effective against epilepsy . In one study , obese men ate an ad libitum (they ate what and how much they wanted) ketogenic diet consisting of 4% carb, 30% protein, 66% fat. They got into and remained in ketosis and ended up losing more weight with less hunger than another group on a high-carb diet with the same amount of protein.
- If you’re going to severely restrict a vital macronutrient like protein, you’d better have a good reason. You’d better be seeing measurable, obvious benefits that disappear when you eat more protein. Don’t wed yourself to the numbers or to the idea of a thing. Always ground your dietary excursions in tangible, verifiable feedback—both subjective and objective. Do what works. Don’t do what doesn’t work, even if it’s “supposed to” be working.
- As always (especially if you’re using keto to address a medical condition), make sure to consult your doctor.
Now I’d love to hear from you. How does protein affect your ketogenic diet? Do you even notice—or consider the question—in your process?
Take care, everyone.
- Fromentin C, Tomé D, Nau F, et al. Dietary Proteins Contribute Little to Glucose Production, Even Under Optimal Gluconeogenic Conditions in Healthy Humans . Diabetes. 2013;62(5):1435-1442.
- Johnstone AM, Horgan GW, Murison SD, Bremner DM, Lobley GE. Effects of a high-protein ketogenic diet on hunger, appetite, and weight loss in obese men feeding ad libitum . Am J Clin Nutr. 2008;87(1):44-55.
- El-rashidy OF, Nassar MF, Abdel-hamid IA, et al. Modified Atkins diet vs classic ketogenic formula in intractable epilepsy . Acta Neurol Scand. 2013;128(6):402-8.
Mark Sisson is the New York Times bestselling author of The Keto Reset Diet  and a dozen other healthy living books. He is one of the leading voices in the evolutionary health movement.