Biochemistry
Chemiosmosis
Cells store energy as a proton gradient (~200 mV), then let H+ flow back through ATP synthase to make ATP
Chemiosmosis is how cells turn a proton gradient into ATP: an electron transport chain pumps H+ across a membrane to build a proton-motive force of about 200 mV, and the protons then flow back through ATP synthase, whose rotor spins at over 100 revolutions per second to make roughly 3 ATP per turn. Peter Mitchell proposed the idea in 1961 and won the 1978 Nobel Prize; it produces about 90% of the ATP in aerobic cells.
- Energy storeProton gradient (PMF ~150–220 mV)
- MachineATP synthase (rotary motor)
- Rotor speed>100 rev/s
- Cost~3–4 H+ per ATP
- Share of aerobic ATP~90%
- Proposed byPeter Mitchell 1961 (Nobel 1978)
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
What chemiosmosis is, in one breath
Cells do not make ATP directly from the energy in food. They take a detour. First they burn fuel to pump hydrogen ions (protons, H+) across a sealed membrane, piling them up on one side like water behind a dam. That stored gradient — part chemistry, part voltage — is the cell's rechargeable battery. Then they open a single, exquisitely engineered gate in the membrane and let the protons rush back. The gate is a spinning molecular turbine called ATP synthase, and the rush of protons through it turns a rotor that physically cranks ATP into existence. That two-step trick — pump protons out, harvest them coming back — is chemiosmosis.
The word breaks down as "chemi-" (chemical) plus "osmosis" (movement across a membrane down a gradient). It is the central energy-converting process of life. Roughly 90% of the ATP in your body — the molecule that powers muscle contraction, nerve signaling, and essentially every active process — is made by chemiosmosis in your mitochondria. The same machinery, in mirror image, powers photosynthesis in chloroplasts and the entire metabolism of most bacteria. A human at rest turns over close to their own body weight in ATP every day, almost all of it through this one mechanism.
How the proton gradient is built and spent
Chemiosmosis has two physically separate halves, joined only by the membrane and the protons that cross it. This separation is the whole point — and was the radical part of the idea.
Step 1 — Pump protons out (build the gradient). High-energy electron carriers made by glycolysis and the Krebs cycle — NADH and FADH2 — drop their electrons into the electron transport chain, a series of large protein complexes embedded in the inner mitochondrial membrane (Complexes I, III, and IV in mitochondria). As electrons hop downhill from complex to complex toward their final acceptor, oxygen, each downhill step releases energy. The complexes use that energy to pump protons from the mitochondrial matrix out into the intermembrane space. Complex IV finishes the chain by combining electrons, protons, and O2 to make water. The net result: protons accumulate outside, the matrix becomes more alkaline and negatively charged, and a proton-motive force of roughly 150–220 mV builds across a membrane only about 5–6 nm thick. That is an electric field on the order of 30 million volts per meter — comparable to a lightning bolt, held steady inside every one of your cells.
Step 2 — Let protons back in through ATP synthase (spend the gradient). The inner membrane is otherwise nearly impermeable to protons, so the only easy way back is through ATP synthase. Protons entering its membrane-embedded Fo ring force the ring to rotate, one notch per proton. The ring is welded to a central γ-shaft that spins inside the mushroom-shaped F1 head sitting in the matrix. As the asymmetric γ-shaft turns, it squeezes each of three catalytic sites in turn, cycling them through grabbing ADP and phosphate, clamping them together into ATP, and ejecting the finished ATP. Three ATP come off per full 360° turn. Because the rotor spins more than 100 times a second, a single ATP synthase molecule can stamp out several hundred ATP per second.
The elegance is that the two halves never touch chemically. The electron transport chain has no idea what ATP synthase is doing; they communicate only through the shared proton reservoir. Pump faster than you spend and the gradient steepens; spend faster than you pump and it shallows. The membrane is the wire connecting the generator to the motor.
The players and the conditions it needs
- A sealed, ion-tight membrane. Without a closed compartment there is no gradient to store. A torn or leaky membrane discharges instantly and makes no ATP — this requirement is why Mitchell's idea was initially rejected.
- An electron transport chain. In mitochondria: Complex I (NADH dehydrogenase), Complex II (succinate dehydrogenase, feeds electrons but pumps no protons), Complex III (cytochrome bc1), and Complex IV (cytochrome c oxidase), with the mobile carriers ubiquinone (coenzyme Q) and cytochrome c ferrying electrons between them.
- A terminal electron acceptor. Usually O2 in aerobic respiration; some bacteria use nitrate, sulfate, or even iron. Without an acceptor, electrons back up, the chain stalls, and pumping stops — which is why cyanide (a Complex IV poison) is lethal in seconds.
- ATP synthase (Complex V). The rotary enzyme that converts proton flow into ATP. Conserved from bacteria to humans — one of the most ancient proteins known.
- A supply of ADP and inorganic phosphate (Pi) in the matrix, plus carriers (the ADP/ATP translocase and the phosphate carrier) to swap spent ADP in and fresh ATP out across the membrane.
Mitochondria vs chloroplasts vs bacteria
Chemiosmosis is the same physics everywhere, but the geography differs. Mitchell's deepest insight was that respiration and photosynthesis are the same trick run in different compartments.
| Property | Mitochondrion (respiration) | Chloroplast (photosynthesis) | Bacterium |
|---|---|---|---|
| Membrane used | Inner mitochondrial membrane | Thylakoid membrane | Plasma membrane |
| Energy source for pumping | Oxidizing NADH / FADH2 | Light absorbed by chlorophyll | Oxidizing fuel or light |
| Protons pumped into | Intermembrane space | Thylakoid lumen | Periplasm / outside |
| Protons flow back into | Matrix | Stroma | Cytoplasm |
| Terminal electron acceptor | O2 → water | NADP+ → NADPH | O2, nitrate, sulfate… |
| Process name | Oxidative phosphorylation | Photophosphorylation | Either |
| Dominant PMF component | Membrane voltage (ΔΨ) | pH gradient (ΔpH, ~3 units) | Variable |
| ATP synthase used | F-type (FoF1) | F-type (CFoCF1) | F-type |
The chloroplast case is a clean teaching example: in chloroplasts the proton-motive force is almost entirely a pH gradient (the lumen can hit pH 5 while the stroma sits near pH 8 — a 1000-fold proton concentration difference), with little membrane voltage. In mitochondria it is the reverse — mostly voltage, with a modest pH term. Same energy, different bookkeeping.
The real numbers
| Quantity | Value | Note |
|---|---|---|
| Proton-motive force | ~150–220 mV | Mitochondria; sum of ΔΨ and ΔpH |
| Membrane thickness | ~5–6 nm | Field ≈ 30 MV/m across it |
| ATP synthase rotor speed | >100 rev/s | Up to ~350 rev/s measured in vitro |
| c-ring subunits | 8 (mammals), 10–15 (others) | Sets protons per ATP — non-integer |
| Protons per ATP | ~3–4 (plus ~1 for transport) | c-ring size ÷ 3 catalytic sites |
| ATP per glucose (realistic) | ~30–32 | Old textbook value 36–38 is too high |
| Share of aerobic ATP | ~90% | Glycolysis + Krebs substrate-level = rest |
| Whole-body ATP turnover | ~50–75 kg/day | Recycled thousands of times; pool is tiny |
| Resting metabolism lost to proton leak | ~20–25% | Imperfect coupling, released as heat |
The torque the Fo motor generates is about 40 piconewton-nanometers — enough that experimenters could attach a fluorescent actin filament to the γ-shaft and literally watch it spin under a microscope (Hiroyuki Noji and colleagues, 1997), the first direct observation of a single rotary enzyme turning. Run in reverse, F1 alone will hydrolyze ATP and spin the rotor backwards, which is why the enzyme is formally named ATP synthase but can act as an ATPase.
Where it shows up: organisms, disease, and biotech
- Every breath you take. The oxygen you inhale exists chemically to be the final electron acceptor at the end of the mitochondrial chain. You breathe to keep chemiosmosis running; stop the chain (cyanide, carbon monoxide blocking Complex IV) and ATP production collapses in seconds despite plenty of oxygen in the blood.
- Brown fat and newborn warmth. Brown adipose tissue is packed with mitochondria expressing UCP1 (thermogenin), a deliberate proton-leak channel. It short-circuits the gradient so the energy comes out as heat instead of ATP — keeping hibernating mammals and human newborns warm. This is chemiosmosis intentionally "wasted."
- Mitochondrial diseases. Mutations in electron transport chain subunits or ATP synthase cause Leigh syndrome, MELAS, and Leber's hereditary optic neuropathy. Tissues with the highest ATP demand — brain, heart, retina, muscle — fail first, because they live closest to the edge of their chemiosmotic budget.
- Antibiotics and poisons. Oligomycin plugs the Fo proton channel of ATP synthase; rotenone blocks Complex I; cyanide and CO block Complex IV; antimycin A blocks Complex III. The uncoupler 2,4-dinitrophenol (DNP) was a 1930s diet drug that worked by discharging the gradient — and killed users by cooking them from the inside.
- Origin of life and evolution. Because chemiosmosis is universal — found in bacteria, archaea, mitochondria, and chloroplasts — many researchers (notably Nick Lane and William Martin) argue it is ancient and may predate the last universal common ancestor, possibly originating from natural proton gradients across the mineral walls of alkaline hydrothermal vents.
- Photosynthesis feeds the planet. The ATP that drives the Calvin cycle's carbon fixation in plants is made chemiosmotically across the thylakoid membrane. Every gram of plant biomass, and the oxygen in the air, traces back to light-driven chemiosmosis.
Common misconceptions and pitfalls
- "The electron transport chain makes ATP." It does not. The chain only pumps protons — it builds the battery. ATP itself is made exclusively by ATP synthase. Confusing the two is the single most common error. The chain and the synthase are separate proteins doing separate jobs, linked only by the gradient.
- "Protons flow through the electron transport chain to make ATP." No — protons flow out through the chain (pumped, uphill) and flow back in through ATP synthase (down the gradient). Two different routes, opposite directions.
- "Energy is stored in a high-energy phosphate intermediate." This was the dominant pre-Mitchell theory, and it was wrong. The energy is stored in the gradient across the membrane, not in any molecule. There is no missing intermediate to find — that is exactly why the search for one failed for years.
- "You get 38 ATP per glucose." The classic integer is an idealization. Real yields are about 30–32 because the proton-to-ATP ratio is non-integer, transport carriers cost extra protons, and membranes leak. The cell does not produce a fixed whole number.
- "Osmosis of water is involved." Despite the name, chemiosmosis is about protons crossing a membrane, not water. The "osmosis" refers loosely to movement down a gradient across a membrane, not to osmotic water flow.
- "The proton gradient is purely a concentration difference." It is electrochemical. Protons carry charge, so moving them builds both a chemical gradient (ΔpH) and an electrical gradient (membrane voltage, ΔΨ). In mitochondria the voltage term actually dominates. Ignoring the electrical component badly under-counts the stored energy.
Frequently asked questions
What is the proton-motive force?
The proton-motive force (PMF) is the stored energy in a transmembrane proton gradient — the same quantity that chemiosmosis spends to make ATP. It has two components: a chemical part, the pH difference (ΔpH) caused by protons being more concentrated on one side, and an electrical part, the membrane voltage (ΔΨ) caused by net positive charge accumulating on one side. In a respiring mitochondrion the matrix is alkaline and negatively charged relative to the intermembrane space, giving a total PMF of about 150–220 mV. Because both terms add, the PMF is dominated by the electrical component (ΔΨ ≈ 150–180 mV) with the chemical component (ΔpH ≈ 0.5–1 unit, worth about 30–60 mV) on top. ATP synthase taps this combined force; collapsing it — for example with the uncoupler DNP — stops ATP synthesis even though the electron transport chain keeps running.
How does ATP synthase actually make ATP?
ATP synthase is a rotary motor with two coupled parts. The membrane-embedded Fo portion contains a ring of 8 to 15 c-subunits, each carrying a proton-binding site. As protons flow down the gradient, each one binds a c-subunit on the high-concentration side, the ring rotates one notch, and the proton is released on the low-concentration side — turning the ring like a water wheel at over 100 revolutions per second. The ring is rigidly attached to a central γ-stalk that pokes up into the F1 head, a hexamer of three αβ pairs. As γ spins, its asymmetric shape forces each of the three catalytic β-sites through three conformations — loose (binds ADP + Pi), tight (forms ATP), and open (releases ATP) — Paul Boyer's binding-change mechanism. One full 360° rotation releases three ATP. Boyer and John Walker shared the 1997 Nobel Prize for working out this mechanism, and Walker's group solved the F1 crystal structure that confirmed it.
Where does chemiosmosis happen in a cell?
Chemiosmosis happens across any membrane that holds an electron transport chain and ATP synthase facing the same proton reservoir. In animal and plant mitochondria it occurs across the inner mitochondrial membrane — protons are pumped from the matrix into the intermembrane space and flow back into the matrix through ATP synthase (oxidative phosphorylation). In chloroplasts during photosynthesis it occurs across the thylakoid membrane — light-driven electron transport pumps protons into the thylakoid lumen, and they flow back out into the stroma through chloroplast ATP synthase (photophosphorylation). In bacteria and archaea, which lack organelles, it occurs across the plasma membrane. The unity of the mechanism across all three was Peter Mitchell's central insight: the same proton-gradient logic powers respiration and photosynthesis alike.
Why was Mitchell's chemiosmotic hypothesis so controversial?
When Peter Mitchell proposed it in 1961, biochemists were searching hard for a 'high-energy intermediate' — a phosphorylated molecule that would carry energy directly from electron transport to ATP synthesis, by analogy with substrate-level phosphorylation in glycolysis. Mitchell instead claimed the energy was stored not in any molecule but in a gradient across a membrane, requiring an intact closed compartment. This was radical: it meant the elusive intermediate did not exist and that membrane topology was essential. The hypothesis was disputed for over a decade. Decisive evidence came from experiments like André Jagendorf's 1966 acid-bath demonstration (chloroplasts made ATP in the dark after an artificial pH jump) and Efraim Racker's 1974 reconstitution of bacteriorhodopsin plus ATP synthase in liposomes, which made ATP from light alone with no respiratory chain. Mitchell received the 1978 Nobel Prize in Chemistry, sole laureate, for the idea.
How many ATP does chemiosmosis make per glucose?
Oxidizing one glucose in aerobic respiration feeds roughly 10 NADH and 2 FADH2 into the electron transport chain. Older textbooks said each NADH yields 3 ATP and each FADH2 yields 2 ATP, for about 34 ATP from oxidative phosphorylation plus 4 from substrate-level steps (about 38 total). Modern measurements give non-integer ratios — about 2.5 ATP per NADH and 1.5 per FADH2 — because ATP synthase needs roughly 3–4 protons per ATP plus another proton to import phosphate and exchange ADP for ATP, and the c-ring stoichiometry (8 in animals, more in plants and bacteria) is not a whole-number multiple of the pumping ratio. That gives a realistic total of about 30–32 ATP per glucose. The honest takeaway: the cell does not make a fixed integer of ATP per glucose; the yield depends on c-ring size, proton leak, and how much PMF is diverted to other transport jobs.
What happens if the proton gradient leaks?
If protons re-enter without passing through ATP synthase, the energy of the gradient is released as heat instead of being captured as ATP — this is called uncoupling. Brown adipose tissue does this on purpose: the protein UCP1 (thermogenin) provides a controlled proton leak that warms hibernating mammals and human newborns, burning fuel to make heat rather than ATP. Chemical uncouplers like 2,4-dinitrophenol (DNP) shuttle protons across the membrane and were sold as weight-loss drugs in the 1930s — they work by forcing cells to burn fat to maintain the gradient, but the runaway heat killed people, and DNP is now banned. Even normal mitochondria leak protons constantly, accounting for roughly 20–25% of resting metabolic rate as 'futile' heat production. This shows the gradient itself, not any chemical bond, is the energy currency that chemiosmosis is built on.