Chapter 6 — Inside the Cell: Metabolic Pathways Without the Biochemistry Degree

The Hook: The sign on the treadmill

Every commercial gym I have ever been in has the same chart.

It's printed on the treadmill console or laminated on the wall beside it. Two coloured bands. The lower one, at 60–70% of your maximum heart rate, is labelled FAT BURNING ZONE. The upper one, at 70–85%, is labelled CARDIO ZONE.

The implication is unmistakable and it is what everybody takes away: if you want to lose fat, stay in the lower band. Work harder and you'll stop burning fat.

I want to be careful here, because that chart is not a lie. The underlying physiology it's pointing at is real, and I'll show you exactly what it is in §6.6. At lower exercise intensities, a higher proportion of the energy you use comes from fat. That's a genuine, measurable, reproducible fact.

The problem is the word zone — because a zone implies a place you enter, a mode you switch into, a state you're either in or out of.

There is no zone. There is no switch. There has never been a switch.

Right now, sitting still reading this, you are burning fat and carbohydrate, simultaneously, in a ratio of roughly 60:40. You were burning both when you woke up. You'll be burning both while asleep. When you stand up and walk to the kitchen the ratio will shift, and it will shift again when you eat, and again in four hours, and none of these transitions involve entering or leaving anything.

Your metabolism is not a set of gears. It's a mixing board — a row of sliders, all of them partially up, continuously adjusting.

Once you have that picture, an enormous amount of confident nonsense stops working on you. The fat-burning zone. "Metabolic switching." "Your brain runs on glucose." "Ketosis means you're in fat-burning mode." "Fasting flips you into autophagy." Every one of those claims depends on you believing in a switch, and by the end of this chapter you won't.

A word before we start, because this is the chapter students most often describe as the moment they stopped enjoying nutrition.

I'm not going to draw you a metabolic pathway chart. You've seen them — the wall poster with four hundred arrows, which is genuinely magnificent and genuinely useless to a person trying to decide what to have for lunch.

What you need is about seven ideas, each of which has a good analogy, and none of which requires you to remember a single enzyme name. That's what this chapter is. If you've bounced off biochemistry before, give me twenty minutes — I think this one will land.

🏃 Fast Track: §6.1 (ATP), §6.2 (the three fuel stores), and §6.6 (the mixing board — the threshold concept) are the irreducible core. Twenty minutes, and they're the three that defend you against the most claims.

🔬 Deep Dive: §6.7 (gluconeogenesis), §6.8 (ketosis) and §6.9 (mTOR and AMPK) are what you need to properly evaluate the keto argument (Chapter 10), the fasting argument (Chapter 21), the protein argument (Chapter 8), and the longevity claims you'll meet everywhere.

6.1 The currency: ATP

Your cells don't run on glucose. They don't run on fat either. They run on ATP, and everything else is a way of making it.

ATP — adenosine triphosphate — is the only energy currency your cells actually spend. Muscle contraction spends ATP. Nerve signalling spends ATP. Building a protein, pumping ions across a membrane, synthesizing a hormone: ATP, ATP, ATP.

Here's the analogy that makes the rest of the chapter work.

ATP is the cash in your pocket. Glycogen is your current account. Body fat is your pension.

• Cash (ATP) — you carry almost none. Your entire body's ATP stock would fuel a few seconds of hard effort. It's spent and regenerated continuously, thousands of times a day. You don't store energy as ATP for the same reason you don't keep your life savings in your pocket.
• Current account (glycogen) — accessible in seconds, limited balance, and it runs out.
• Pension (body fat) — enormous, slower to access, and the entire point of the system.

Everything in this chapter is about converting the pension and the current account into cash. That is the whole of metabolism, stated once.

6.2 The three stores, and how unequal they are

Now put numbers on it, because the asymmetry is the single most under-appreciated fact in nutrition.

Sit with the fourth row. A lean adult carries somewhere around 90,000 calories of fat. Someone carrying more body fat carries proportionally more. Total glycogen is perhaps 2,000.

Your fat stores exceed your carbohydrate stores by a factor of forty or more.

Two consequences follow immediately, and they're load-bearing for the rest of the book.

One: you cannot run out of fat. Not in any circumstance a person survives. Fatigue during prolonged exercise is essentially never a shortage of fat — it's glycogen depletion, dehydration, or central fatigue. Chapter 23 builds on this.

Two: the muscle glycogen is not shared. This surprises people. Muscle glycogen can only be used by the muscle that stores it — muscle cells lack the enzyme needed to release glucose back into the blood. Your quadriceps cannot lend glucose to your brain. Only liver glycogen can raise blood glucose, and there's only about 100 g of it — roughly a day's worth, which is why the liver's glycogen is the strategic reserve and the muscle's is a local fuel tank.

And the last row matters most. There is no protein store. Every gram of protein in your body is doing a job — it's a muscle fibre, an enzyme, an antibody, a transporter. There's a small circulating "amino acid pool," and that's it. So when the body breaks down protein for energy, it is dismantling functional tissue. Not spending savings — selling furniture.

That fact is why protein intake matters so much during weight loss (Chapter 8), why energy availability matters so much for athletes (Chapter 23), and why the elderly losing weight is a clinical concern rather than a success (Chapter 25).

🔄 Check your understanding. A marathon runner "hits the wall" at mile 20. They have 18 kg of body fat, which is over 150,000 kcal. Why can't they simply use it?

Answer

They are using it — continuously, throughout the race. The problem isn't availability, it's rate. Fat oxidation supplies ATP more slowly than glycolysis can, so at marathon pace the required power output exceeds what fat alone can deliver. When muscle glycogen runs low, the athlete must slow to a pace that fat can sustain, which feels like hitting a wall.

Add the liver-glycogen problem: with liver glycogen depleted, blood glucose falls, and the brain — which cannot use fatty acids (§6.5) — is affected too, producing the disorientation runners describe.

The wall is a rate limitation and a glycogen limitation, never a fat shortage. Chapter 23 covers what to do about it.

🧾 Cost check. The asymmetry in §6.2 quietly prices an entire product category. Exogenous ketone supplements — ketone salts and esters — run roughly $60–$150 a month, and ester products considerably more. They do what they say: drinking them raises blood ketones, measurably, within about half an hour.

What they don't do is make you burn body fat. Raising blood ketones from the outside is the opposite of producing them from the inside — the whole reason your body makes ketones is that it's running low on carbohydrate and drawing on fat stores. Supplying ketones exogenously provides an additional fuel, which if anything reduces the demand on your own fat. There's genuine research interest in exogenous ketones for specific applications — certain neurological conditions, some athletic contexts — and the weight-loss marketing rests on a confusion between a marker and a mechanism. The number on the meter goes up; the reason you wanted it to has gone away.

6.3 Glycolysis: splitting sugar

The first pathway, and the oldest — glycolysis is so ancient that essentially every living thing does it, from bacteria to you.

What it does: takes one molecule of glucose and splits it into two molecules of pyruvate, producing a small amount of ATP directly.

Where: in the cytoplasm — the general interior of the cell, not inside any organelle.

Why it matters: it's fast, and it doesn't require oxygen.

That combination is the whole point. When you sprint, or lift something heavy, or run up stairs, your demand for ATP spikes faster than oxygen can be delivered. Glycolysis covers the gap.

Then pyruvate goes one of two ways:

• Oxygen available → pyruvate enters the mitochondrion and is fully burned (§6.4). Efficient.
• Oxygen limiting → pyruvate is converted to lactate, which regenerates a molecule glycolysis needs so it can keep running. Fast but low-yield.

🔍 Why this works — and a myth that needs killing. You have been told that lactic acid causes muscle burn and next-day soreness. Both are wrong, and the correction is genuinely useful.

Lactate is not a waste product. It's a fuel. Produced in a hard-working muscle, it's exported into the blood and taken up by the heart, by other muscles, and by the liver — which can convert it back into glucose (the Cori cycle) and send it out again. Lactate is one of the body's more elegant recycling systems, and elite athletes are better at clearing and using it, not at avoiding it.

The burn during hard effort is associated with the accumulation of hydrogen ions and other metabolites, not lactate itself. And next-day soreness — delayed onset muscle soreness — is mechanical damage and the inflammatory repair response; lactate is long gone by then. This matters beyond trivia: it's why "flush out the lactic acid" is a meaningless instruction, and why the supplement industry sells products to remove a substance your body is actively trying to use.

6.4 The mitochondrion: where things actually burn

Pyruvate — or a fat fragment, or a broken-down amino acid — enters the mitochondrion, and this is where the real energy is extracted.

Two stages, and you need the shape rather than the detail.

Stage one: the citric acid cycle (also called the Krebs cycle or TCA cycle). Fuel fragments arrive as a molecule called acetyl-CoA, and the cycle strips them apart, capturing high-energy electrons onto carrier molecules and releasing carbon dioxide — the CO₂ you exhale, which as Chapter 4 noted is literally where lost fat goes.

Stage two: the electron transport chain. Those loaded carriers hand their electrons down a series of protein complexes in the mitochondrial membrane. The energy released pumps protons across the membrane, creating a gradient — and the protons flowing back through a molecular turbine drive the synthesis of ATP. Oxygen sits at the end of this chain, accepting the spent electrons and combining with hydrogen to form water.

That is what breathing is for. Not to bring in oxygen for its own sake — to provide the final electron acceptor without which the entire chain backs up and stops.

📊 Diagram (described). Picture a hydroelectric dam.

Fuel molecules — glucose fragments, fat fragments — arrive at the top and are dismantled by the citric acid cycle. That dismantling doesn't directly make ATP; instead it pumps water uphill into a reservoir, using electron carriers as the pumps. This is the electron transport chain: protons pushed across the mitochondrial membrane into a reservoir on the far side.

Then the reservoir drains. Protons flow back down through a turbine — an enzyme called ATP synthase, which physically rotates as they pass — and that rotation is what generates ATP.

Two things follow from the picture. First, oxygen is the drainage channel at the bottom. Without it, the water has nowhere to go, the reservoir backs up, the pumps stall, and the whole dam stops. That's why you can't survive minutes without breathing, no matter how much fuel you have. Second, the dam is why fat delivers more energy per gram than carbohydrate — 9 kcal versus 4. Fat molecules are more chemically reduced, meaning they carry more electrons to pump with, so each gram fills more reservoir.

And note where the dam is: mitochondria. More and better mitochondria means more capacity to burn fat aerobically — which is precisely what endurance training builds, and precisely why trained people oxidize more fat at the same workload than untrained people. Not a supplement. An adaptation.

6.5 Beta-oxidation: taking fat apart

Fat is stored as triglyceride — three fatty acid chains attached to a glycerol backbone.

To use it:

1. Lipolysis. The triglyceride is split, releasing free fatty acids into the blood.
2. Transport. Fatty acids travel to a cell and are moved into the mitochondrion. (This transport step requires carnitine — remember that, because it's about to matter.)
3. Beta-oxidation. Inside the mitochondrion, the long fatty acid chain is chopped, two carbons at a time, into acetyl-CoA units.
4. Those acetyl-CoA units enter the citric acid cycle, exactly as pyruvate's do.

Notice the convergence: carbohydrate and fat arrive at the same place. Both become acetyl-CoA; both feed the same cycle and the same dam. They differ in how fast they get there and how much they yield.

Two consequences worth knowing.

The brain can't do this. Fatty acids don't cross the blood–brain barrier well. So the brain — the organ using about 20% of your resting energy — cannot run directly on fat. That constraint is the entire reason §6.7 and §6.8 exist.

And fat oxidation is slower than glycolysis. More steps, more transport, more oxygen required. Fat is the high-capacity, low-rate fuel. Carbohydrate is the limited-capacity, high-rate fuel. Neither is better. They're for different jobs, which is why you have both.

🔬 Claim → Evidence → Verdict

The claim: "L-carnitine transports fat into the mitochondria to be burned. Supplementing with it increases fat burning and helps you lose weight."

Where it comes from: The mechanism is completely correct, which is what makes this one instructive. Carnitine genuinely is required to shuttle long-chain fatty acids into the mitochondrion. Step two above is real and carnitine really does it.

What the evidence actually shows: The mechanism is real and carnitine is not the rate-limiting step in healthy people. Your body synthesizes it, you obtain it from food, and muscle carnitine concentrations are not readily raised by oral supplementation — the transport into muscle is itself tightly regulated. Supplementation trials in non-deficient people generally do not show meaningful increases in fat oxidation or weight loss. Genuine carnitine deficiency exists — in certain rare genetic conditions, in some dialysis patients, with certain medications — and in those people supplementation matters a great deal.

📉 Evidence quality: Mechanism established (rung 1–2). Human supplementation trials in healthy people: multiple, small, largely null.

Verdict: ❌ Not supported for fat loss in healthy people.

💡 And this is the pattern worth extracting, because you'll meet it a hundred times: "X is required for Y, therefore more X increases Y." It is almost always wrong, because biological pathways have a rate-limiting step and it's usually not the ingredient being sold. A road being required to reach a city doesn't mean building a wider road gets you there faster if the traffic jam is at the bridge.

6.6 🚪 The mixing board: there is no switch

Here it is. The threshold concept, and the reason this chapter exists.

🚪 Threshold concept.

You are never burning one fuel exclusively. At every moment you are oxidizing carbohydrate and fat simultaneously, in a continuously shifting ratio. There is no switch, no mode, and no zone — only a mixing board whose sliders move.

The evidence, from Chapter 5

You already met it. The respiratory quotient — CO₂ produced divided by O₂ consumed:

A resting, several-hours-fasted human sits around 0.80. Not 0.7. Not 1.0.

You have never in your life been in a "fat-burning mode" and you have never been out of one.

What moves the sliders

Notice these are continuous variables, all acting at once. That's the mixing board.

So what about the fat-burning zone?

Now we can answer the treadmill.

What's true: at lower exercise intensities, the proportion of energy from fat is higher. At higher intensities the proportion shifts toward carbohydrate, because glycolysis can supply ATP faster than beta-oxidation can. This is the crossover concept, and it's well established. There's even an intensity — often somewhere in the region of 60–65% of maximal oxygen uptake in trained people, lower in untrained — at which absolute fat oxidation peaks, sometimes called Fatmax.

What's false: that this matters much for fat loss.

Work the arithmetic:

LOW INTENSITY (walking, 30 min)
  Total energy burned:            ~150 kcal
  Proportion from fat:            ~60%
  Fat oxidized:                   ~90 kcal worth

HIGH INTENSITY (running, 30 min)
  Total energy burned:            ~400 kcal
  Proportion from fat:            ~35%
  Fat oxidized:                   ~140 kcal worth

The higher-intensity session burned a smaller proportion of fat and a larger amount of it. The zone chart has you optimizing the percentage while ignoring the total.

And the deeper point, which is the one that actually settles it: fat loss over weeks is determined by energy balance, not by which substrate you oxidized during a particular half hour. If you burn carbohydrate during exercise, you will oxidize correspondingly more fat afterward as your body restores glycogen and handles the deficit. The books balance over days, not during the session.

🔬 Claim → Evidence → Verdict

The claim: "Exercise in the fat-burning zone (60–70% max heart rate) to lose fat. Working harder burns carbohydrate instead."

Where it comes from: A genuine, replicated physiological observation — the crossover concept. Lower intensity really does mean a higher proportion of energy from fat. The chart is built on a real finding.

What the evidence actually shows: The proportion is the wrong variable. Higher intensity burns more total energy and frequently more total fat; and regardless, substrate use during exercise is largely irrelevant to fat loss over weeks, which tracks energy balance. Trials comparing exercise intensities at matched energy expenditure generally find similar fat-loss outcomes.

📉 Evidence quality: The underlying physiology is solid (rung 6). The practical inference is contradicted by outcome trials.

Verdict: 🟠 Probably false as a fat-loss strategy — a true mechanism, wrong conclusion.

What to actually do: exercise at whatever intensity you'll sustain and enjoy. Lower intensity is easier to recover from and easier to do for longer; higher intensity is time-efficient and better for cardiorespiratory fitness. Both are fine. Neither is a fat-burning secret, and the chart on the wall has been quietly discouraging people from working hard for thirty years.

6.7 Gluconeogenesis: making sugar out of not-sugar

The brain needs glucose (§6.5). Liver glycogen holds about a day's worth. So what happens on day two of not eating?

Gluconeogenesis — literally "making new glucose." Your liver (and to a lesser extent your kidneys) can manufacture glucose from:

• Lactate, recycled from muscles (the Cori cycle, §6.3)
• Glycerol, the backbone released when triglycerides are broken apart
• Amino acids, from protein — dietary protein, or your own tissue

Note the conspicuous absence: you cannot make meaningful amounts of glucose from fatty acids. The chemistry runs one way — glucose can become fat, but the bulk of fat cannot become glucose. Only the small glycerol fraction can.

This has a consequence that resolves a very common argument.

🔬 Claim → Evidence → Verdict

The claim: "Carbohydrate is an essential nutrient — your brain runs on glucose, so you must eat carbohydrates."

Where it comes from: True premises. The brain does require glucose. Fatty acids don't cross the blood–brain barrier well. Liver glycogen lasts about a day.

What the evidence actually shows: The conclusion doesn't follow, because of gluconeogenesis. There is no dietary requirement for carbohydrate in the sense that there is for essential amino acids and essential fatty acids — your liver can manufacture the glucose your brain requires from lactate, glycerol, and amino acids. Populations have lived long-term on very low carbohydrate intakes. And in prolonged carbohydrate restriction the brain adapts further, deriving a substantial share of its energy from ketone bodies (§6.8), reducing its glucose requirement considerably — though not to zero; some glucose is still needed and gluconeogenesis supplies it.

The honest complication, which both sides ignore: making glucose from amino acids means using amino acids, which means either eating enough protein or breaking down your own tissue. So "you don't need carbohydrate" is true, and the cost of not eating it is paid in protein. That's a real trade-off, not a free lunch — and it's part of why very-low-carbohydrate diets tend to require higher protein intakes to protect lean mass.

📉 Evidence quality: Well-established biochemistry; population and trial evidence for viability.

Verdict: 🟠 Probably false as stated. Carbohydrate is not essential in the biochemical sense. Whether a low-carbohydrate diet is a good idea is an entirely different question — that's Chapter 10 — and the answer there is much less flattering to both camps than either expects.

🧩 Productive struggle. Four minutes before reading on.

A person eats zero carbohydrate for three days. Their blood glucose, measured each morning, stays within the normal range the whole time.

Where is that glucose coming from — and what is it costing them?

What I'd say

Day one: mostly liver glycogen, which holds roughly 80–120 g. That covers the brain's demand (very roughly 100–120 g/day) for about a day.

Days two and three: gluconeogenesis, from three sources — glycerol released as triglycerides are broken down, lactate recycled via the Cori cycle, and amino acids.

The cost is the amino acids. If dietary protein is adequate, they come from food. If it isn't, they come from body protein — and since there is no protein store (§6.2), that means functional tissue: muscle, and eventually organ protein.

This is why protein intake rises in importance as carbohydrate falls, and it's the mechanistic reason very-low-carbohydrate diets emphasize protein. It's also why prolonged fasting costs lean mass even in someone with plenty of fat — the fat can supply energy but it cannot supply the brain's glucose.

Meanwhile, over the following days, ketone production ramps up (§6.8) and progressively spares the glucose requirement — which is exactly what that adaptation is for.

6.8 Ketosis: the third fuel

If fat can't fuel the brain and gluconeogenesis costs protein, evolution had an obvious problem to solve. The solution is ketone bodies.

What happens: when carbohydrate is scarce and fat oxidation is high, acetyl-CoA accumulates in the liver faster than the citric acid cycle can process it. The liver converts the excess into ketone bodies — principally beta-hydroxybutyrate and acetoacetate.

Why it matters: unlike fatty acids, ketones cross the blood–brain barrier. The brain can run substantially on them. In sustained ketosis, a large fraction of the brain's energy — commonly estimated at over half — can come from ketones, sharply reducing how much glucose gluconeogenesis must supply, which spares protein.

This is not a hack. It's a survival adaptation, and it's the reason humans can go weeks without food while retaining the cognitive function needed to find some.

The distinction that matters clinically

The crucial difference is insulin. In nutritional ketosis, insulin is low but present, and it acts as a brake on ketone production. In DKA that brake is gone, ketones rise without limit, and blood pH falls. These are not points on the same scale; they are different situations.

⚠️ When to see a professional. People with type 1 diabetes should not undertake ketogenic or fasting protocols without specialist supervision — the risk is real and specific. People taking SGLT2 inhibitors for type 2 diabetes are at risk of euglycemic DKA — ketoacidosis with normal blood glucose, which is easy to miss — and should not begin very-low-carbohydrate diets without medical guidance. Pregnancy, breastfeeding, kidney disease, liver disease, a history of eating disorders, and pancreatitis are all situations where ketogenic diets need medical oversight rather than a podcast.

🔬 Claim → Evidence → Verdict

The claim: "In ketosis you're in fat-burning mode, so a ketogenic diet burns more body fat than other diets at the same calories."

Where it comes from: A real, measurable observation — in ketosis, fat oxidation is elevated and RQ does fall. You genuinely are oxidizing more fat. The claim is describing something true.

What the evidence actually shows: Oxidizing more fat is not the same as losing more fat, and conflating the two is the error. If you eat mostly fat, you burn mostly fat — that's substrate availability, not fat loss. Body fat only decreases if you oxidize more than you consume. Controlled feeding studies holding calories and protein constant, including work from Kevin Hall's group, generally find little or no fat-loss advantage for ketogenic diets — and in some cases a small advantage the other way once the initial glycogen-and-water shift (§4.9) is accounted for.

Meanwhile, ketogenic diets do often work well in practice — through appetite suppression, high protein, elimination of most ultra-processed food, and reduced food variety. Those are real mechanisms operating through energy intake, and they're why some people do extremely well on it.

📉 Evidence quality: Rung 6 controlled feeding trials on the metabolic question; substantial rung 5 and 6 evidence on real-world outcomes.

Verdict: 🟠 Probably false as a metabolic-advantage claim; the diet itself gets a fuller and considerably more sympathetic hearing in Chapter 10.

🔄 Check your understanding. Someone tells you: "I've been in ketosis for two weeks and my ketone meter reads 1.8 mmol/L, so I'm definitely burning fat — but I haven't lost any weight. The meter must be broken." What's actually happening?

Answer

The meter is fine and both facts are true simultaneously. A reading of 1.8 mmol/L is squarely within nutritional ketosis, and yes, they are oxidizing a lot of fat.

But oxidizing fat and losing fat are different things. If you eat mostly fat, you burn mostly fat — that's substrate availability. Body fat only decreases when you oxidize more than you consume, which is an energy balance question, not a ketone question. A person eating 3,000 kcal of predominantly fat can be deeply ketotic and gaining weight.

This is the §6.6 threshold applied: the ketone meter measures which slider is up, not whether the account is being drawn down. It's a fuel-mix reading, not a progress reading — and marketing it as the latter is one of the more effective bits of misdirection in the whole field.

🍽️ On your plate. The practical residue of §6.7 and §6.8, in three lines. (1) You don't need to eat carbohydrate, biochemically — but not eating it raises your protein requirement, because gluconeogenesis runs on amino acids. (2) Ketosis is a normal state, not an achievement, and a ketone reading tells you nothing about whether you're losing fat. (3) If you have type 1 diabetes, take an SGLT2 inhibitor, are pregnant, or have a history of disordered eating, ketogenic approaches need a clinician rather than a podcast — see the ⚠️ above.

6.9 mTOR and AMPK: the build/breakdown dial

Two more names worth knowing, because they're behind most of the longevity content you'll encounter.

Your cells have to decide, continuously, between two modes:

Build — synthesize proteins, grow, divide, repair. Expensive. Appropriate when energy and building materials are plentiful.

Conserve — stop building, break down damaged components, recycle. Appropriate when they're not.

Two sensor systems govern this.

Autophagy — literally "self-eating" — is the cellular recycling process AMPK promotes: damaged proteins and organelles are broken down and their components reused. It's genuinely important biology, and the researchers who worked it out won a Nobel Prize for it.

Here's where nutrition content goes off the rails.

🔬 Claim → Evidence → Verdict

The claim: "Fasting activates autophagy, which clears damaged cells, slows ageing, and prevents disease. You need to fast for 16 (or 24, or 72) hours to trigger it."

Where it comes from: Real and important science. Autophagy exists, matters, and is induced by nutrient deprivation. In model organisms — yeast, worms, flies, mice — interventions that activate these pathways, including caloric restriction and mTOR inhibition, extend lifespan, sometimes dramatically. The mechanistic story is genuinely one of the more exciting areas in biology.

What the evidence actually shows: Almost every specific claim you'll encounter is an extrapolation across three separate gaps. Gap one: species. Lifespan extension in yeast and mice has repeatedly failed to translate cleanly to longer-lived mammals. Gap two: measurement. Autophagy in living humans is genuinely difficult to measure — most human data is indirect, from markers rather than from autophagy itself. Gap three: the numbers. The specific fasting durations quoted in popular content — "autophagy starts at 16 hours," "peaks at 72" — are largely derived from rodent work or from mechanistic inference. Mice have a metabolic rate roughly seven times higher than ours; a 24-hour fast in a mouse is not a 24-hour fast in a human, and the confident hour-marks circulating online do not have solid human outcome data behind them.

What we do have decent human evidence for is more modest: fasting and caloric restriction produce measurable metabolic changes, and time-restricted eating helps some people manage intake. That's Chapter 21's territory and the verdict there is mixed.

📉 Evidence quality: Rung 1–2 for the mechanism (excellent). Rung 2 for lifespan (animals). Weak and largely indirect for human autophagy. Absent for the specific hour thresholds.

Verdict: ⚗️ Untested in humans for the health and longevity claims specifically. Note this is not the same as false — the mechanism is real and the hypothesis is serious. It means the human outcome data that would justify the confident claims does not yet exist, and anyone quoting you a specific hour is quoting a mouse.

🍽️ On your plate. Two practical things fall out of §6.9. (1) If your goal is building or keeping muscle, mTOR is the pathway you want stimulated — which means adequate protein, distributed across the day, with resistance training. Chapter 8 has the doses. (2) If someone tells you a supplement "activates AMPK" or "inhibits mTOR" for longevity, notice that these pathways do opposite things and that you want both, at different times. A product that permanently suppressed mTOR would prevent you from building or repairing anything — which is why the longevity story is far more complicated than the content suggests, and why it is not currently a product.

6.10 Protein's third job: the pool that's always draining

Carbohydrate and fat have one job between them: fuel. Protein has three, and only one of them is fuel — which is why it behaves differently from everything else in this chapter.

Job one: structure and function. Muscle fibres, enzymes, antibodies, transport proteins, receptors, collagen, hormones. This is what protein is for.

Job two: the substrate for gluconeogenesis (§6.7), when glucose is needed and carbohydrate isn't available.

Job three: fuel, when energy is short. The least important and the most costly.

Turnover: you are being rebuilt constantly

Here's the fact that reframes protein intake.

Your body breaks down and rebuilds a substantial quantity of its own protein every single day — on the order of 200–300 grams for a typical adult, far more than anyone eats. Muscle is dismantled and reassembled. Gut lining cells are replaced every few days. Enzymes are made and destroyed continuously.

This constant demolition and reconstruction is called protein turnover, and it's not waste. It's quality control: damaged proteins are recycled, and the body can rapidly change what it's made of in response to what it's doing.

The amino acids released by breakdown enter a small circulating amino acid pool, which is the only thing resembling a protein store you have — and it's genuinely small, a few grams, turning over constantly. Amino acids in that pool have three fates: be reused to build something new, be converted to glucose or fat, or be broken down for energy with their nitrogen excreted as urea by the liver and cleared by the kidneys.

Crucially, that last route is one-way. Once an amino acid's nitrogen has gone into urea and left in your urine, it isn't coming back. You leak protein continuously, and that leak has to be covered by what you eat.

🔍 Why this works. This is the mechanistic answer to a question people ask constantly: why do I need protein every day when I have plenty of muscle? Because the pool is small and the leak is continuous. Think of a bathtub with the plug slightly out. The tub is your amino acid pool, the tap is dietary protein, and the drain is irreversible oxidation and nitrogen loss. Your muscles aren't a reservoir feeding the tub — they're the bathroom, built out of what the tub supplies. When the tap runs slow, the level drops, and the body starts dismantling the bathroom to keep the tub full.

That's what "you can't store protein" actually means, and it's why protein requirements are daily in a way that carbohydrate and fat requirements simply aren't.

Essential and non-essential

Of the twenty amino acids that build human proteins, your body can synthesize about eleven. The other nine — the essential amino acids — it cannot make, and they must come from food.

This is the reason "protein quality" is a real concept rather than marketing (Chapter 8): a protein source is judged largely by whether it supplies all nine essentials in useful proportions. And it's why leucine keeps appearing — it's both an essential amino acid and the primary dietary trigger for mTOR (§6.9).

Nitrogen balance is how this gets measured: nitrogen in from food versus nitrogen out in urine, faeces, sweat, and skin. Positive balance means net protein gain (growth, pregnancy, recovery, training adaptation). Negative balance means net loss (illness, injury, severe restriction, ageing). The RDA for protein was set largely from nitrogen balance studies — and Chapter 8 explains why that method almost certainly underestimates what's optimal, as opposed to what's minimally adequate.

🧾 Cost check. Because §6.9's mTOR discussion inevitably leads somewhere expensive: leucine and BCAA supplements typically run $30–$60 a month, marketed on the entirely real fact that leucine triggers muscle protein synthesis. The mechanism is correct. The problem is §6.5's pattern again — leucine is not the rate-limiting step if you're eating adequate protein, because you're already getting far more leucine from food than a scoop provides, alongside the other eight essential amino acids that isolated BCAAs don't supply. Building a protein from leucine alone is like ordering a pallet of bricks and no mortar.

A serving of BCAAs might cost $1.50 and deliver a few grams of three amino acids. Two eggs cost about $0.60 and deliver roughly 12 g of complete protein containing all nine essentials plus choline, selenium, and B12. There are real uses for isolated amino acids in specific clinical settings; a healthy person who eats enough protein is not one of them.

🪞 Learning Check-In

Second of these. Two minutes.

On Part I as a whole, which you've now finished:

• Which chapter felt hardest? For most readers it's Chapter 2 or this one. Now ask a different question: which will you still be using in five years? Almost certainly Chapter 2. Difficulty and value are not the same axis, and the fact that they feel like the same axis is worth noticing.
• Have you done the checkpoints — the belief inventory, the claim filter card, the digestion log, the three-day diary, your own RMR? If you've done fewer than three, that's the most common pattern and it's worth being honest about now rather than at Chapter 38, when the project assembles.
• Did you attempt the 🧩 problems before reading the answers? The gluconeogenesis one in §6.7 is genuinely answerable from §6.2 and §6.5 alone. If you read straight through, try going back and doing it properly — the difference in what you retain is larger than it feels.

One thing to carry into Part II: you now have enough machinery to evaluate a macronutrient claim on your own. When Chapter 7 tells you something about carbohydrate, try to predict the verdict before you read it. Being wrong is fine and informative; the prediction is what does the work.

Spaced Review

Answer before reading on.

1. (Chapter 5) What is the respiratory quotient of a resting human, and what did we say it implied before we'd explained why?

About 0.80 — a mixture, roughly two-thirds fat. It implied there is no fuel switch, and §6.6 has now given you the mechanism: carbohydrate and fat both converge on acetyl-CoA and feed the same citric acid cycle, so there is nothing to switch between. The sliders just move.

2. (Chapter 3) Long-chain fat leaves the intestine by a different route from carbohydrate and protein. What was it, and how does that connect to this chapter?

Fat is packaged into chylomicrons and enters the lymphatic system, joining the bloodstream near the heart — bypassing the liver's first pass. Carbohydrate and amino acids go via the portal vein straight to the liver. The connection: the liver is the organ that does gluconeogenesis and ketogenesis, and it gets first look at absorbed sugar and amino acids but not at absorbed fat. That asymmetry is part of why the liver is so central to metabolic regulation.

3. (Chapter 4) Why does "you burn more fat at low intensity" fail as fat-loss advice, in terms of the framework from Chapter 4?

Because fat loss is determined by energy balance over weeks, not by substrate use during a session. Chapter 4's threshold concept applies directly: the accounting happens over days, and burning carbohydrate now simply means oxidizing more fat later. The zone chart optimizes a proportion while ignoring the total.

Project Checkpoint: Your Fuel-Mix Reflection

Component six, and the last of Phase 1 (Baseline). This one is thinking rather than measuring — fifteen minutes with a pen.

The task: map one ordinary day of yours onto the fuel systems in this chapter.

Take yesterday. Write down four or five moments and, for each, answer: what were my energy demands, and which sliders were where?

Three questions to answer in writing:

1. How many hours of your day were genuinely high-intensity? For most people the honest answer is minutes. What does that tell you about which fuel dominated your day?
2. When was your liver glycogen lowest? (Hint: it's whenever your longest gap without eating ended.) What was your brain running on at that point?
3. Where in your day did the "no switch" idea contradict something you previously believed? Be specific. This is the entry that matters.

Then, the reflection that makes this a Framework component:

Write one paragraph on what you previously believed about fat burning, and what you believe now.

Almost everyone has something here — the fat-burning zone, "your body burns muscle after X hours," "you have to eat every three hours or you go catabolic," "ketosis means you're burning fat," "cardio on an empty stomach burns more fat." Name yours. Write down where you got it. Then add it to your Chapter 1 Belief Inventory as a late entry — because you now know something about it that you didn't when you sealed the list, and Chapter 17 will be more interesting with it there.

Next checkpoint (Chapter 7): your carbohydrate audit — total, type, refined versus intact, and where they sit in your day. Phase 2 begins.

Chapter Summary

The seven ideas:

1. ATP is the only currency. Cash in your pocket. Glycogen is the current account; body fat is the pension.
2. The stores are wildly unequal. ~2,000 kcal of glycogen vs. 90,000+ of fat — a factor of forty or more. You cannot run out of fat. And there is no protein store — using protein for energy means dismantling functional tissue.
3. Glycolysis is fast and doesn't need oxygen. Lactate is a fuel, not a waste product, and it doesn't cause next-day soreness.
4. Mitochondria are the dam. Citric acid cycle pumps the reservoir; the electron transport chain drains it through a turbine; oxygen is the drainage channel. Fat yields 9 kcal/g because it carries more electrons to pump with.
5. Beta-oxidation converges with glycolysis at acetyl-CoA — same cycle, same dam. Fat is high-capacity/low-rate; carbohydrate is limited-capacity/high-rate. The brain can't use fatty acids, which is why gluconeogenesis and ketosis exist.
6. 🚪 There is no switch. RQ ~0.80 at rest means you're always burning both. The mixing board's sliders move with intensity, feeding status, training, and glycogen — continuously.
7. Protein turnover never stops. You break down and rebuild 200–300 g of your own protein daily. The amino acid pool is small, the nitrogen leak is one-way, and that is precisely why protein requirements are daily in a way carbohydrate and fat requirements are not.

The pattern to extract from §6.5: "X is required for Y, therefore more X increases Y" is almost always wrong. The rate-limiting step is rarely the ingredient being sold.

This chapter's verdicts:

Nutritional ketosis (0.5–3 mmol/L, insulin present) and diabetic ketoacidosis (>10–15 mmol/L, insulin absent, acidotic) are different situations, not points on a scale.

The one thing to remember: your metabolism is a mixing board, not a gearbox. Every claim that depends on you entering or leaving a mode is describing a machine you don't have.

And a useful corollary, since three of this chapter's five verdicts turn on it: a measurement of which fuel you are using is not a measurement of whether your energy stores are shrinking. The ketone meter, the RQ reading, and the treadmill's zone chart all report the position of a slider. None of them reports the balance of the account.

What's Next

Part I is done. You can evaluate a study, place it on a ladder, spot healthy-user bias, hold both halves of energy balance, calculate your own requirements with honest error bars, and describe what your cells actually do with food.

That's the machinery. Part II is where you use it, one macronutrient at a time.

Chapter 7 starts with the one that got demonized worst. Sugars, starches, fiber; why "simple versus complex" is nearly useless and "intact versus refined" is genuinely useful; glycemic index and why it under-delivers; and the question this chapter just half-answered — how much carbohydrate does a human actually require?

You now know enough to predict that answer. Try, before you read it.