ATP Synthase and the Mitochondrial Turbine

Chemiosmotic Proton Gradients, Boyer's Binding-Change Rotor, and Magnetic Bead Rotation

A resting human body holds only about 50 grams of adenosine triphosphate (ATP) at any given moment. Yet our cells hydrolyze and resynthesize roughly 60 to 75 kilograms of ATP every twenty-four hours. Each ATP molecule turns over more than a thousand times a day as it cycles through mitochondrial rotary turbines.

For decades, biochemists searched in vain for a high-energy chemical intermediate linking respiratory electron transport to ATP synthesis. Peter Mitchell broke the impasse in 1961 with his chemiosmotic hypothesis. Electron transport chains pump protons across an ion-tight inner membrane, storing energy as a $180\text{-mV}$ electrochemical protonmotive force that drives ATP synthase.

André Jagendorf and Ernest Uribe proved Mitchell's chemiosmotic model in 1966 without using light or electron transport. They soaked isolated spinach chloroplast thylakoids in a pH 4.0 acid bath in total darkness, then injected them into a pH 8.0 buffer containing ADP and inorganic phosphate. The artificial four-unit pH jump alone synthesized bursts of ATP.

Paul Boyer discovered via oxygen-18 isotope exchange in 1973 that forging the bond between ADP and phosphate inside the ATP synthase active site requires almost zero free energy ($K_{\text{eq}} \approx 1$). Proton flow does not drive bond synthesis directly. Instead, mechanical rotation spends energy to pry the newly formed, tightly bound ATP loose from the catalytic pocket.

John Walker solved the $2.8\text{-\AA}$ X-ray crystal structure of the bovine mitochondrial $\text{F}_1\text{-ATPase}$ catalytic head in 1994, confirming Boyer's binding-change mechanism. A curved, asymmetric central $\gamma$ shaft pokes up through a stationary hexamer of three $\alpha$ and three $\beta$ subunits. Its asymmetry forces the three catalytic $\beta$ pockets into Open, Loose, and Tight states simultaneously.

Masasuke Yoshida, Kazuhiko Kinosita Jr., and Hiroyuki Noji filmed this molecular rotation directly in 1997. They anchored the $\alpha_3\beta_3$ stator ring of thermophilic $\text{F}_1\text{-ATPase}$ to a glass coverslip and bolted a one-micrometer fluorescent actin filament to the central $\gamma$ rotor. Adding ATP caused the filament to spin counterclockwise in discrete 120-degree steps.

Hiroyuki Noji and colleagues completed the mechanical proof in 2004 by attaching a magnetic microbead to the $\gamma$ shaft of $\text{F}_1\text{-ATPase}$ inside femtoliter glass chambers. Rotating electromagnets forced the bead to spin clockwise against its natural hydrolysis direction. Cranked backward mechanically, the enzyme synthesized ATP from ADP and phosphate at three molecules per revolution.

Inside the membrane-embedded $\text{F}_{\text{O}}$ turbine, protons enter a half-channel in the stationary $a$-subunit and bind a negatively charged glutamate or aspartate residue on the adjacent $c$-ring rotor. Neutralized so it can enter the oily lipid bilayer, the $c$-subunit rides a full 360-degree turn before releasing its proton out a second half-channel on the matrix side.

Because every 360-degree turn of the central $\gamma$ shaft produces three ATP molecules in the $\text{F}_1$ head, the enzyme's gear ratio is set by the number of subunits in its $\text{F}_{\text{O}}$ $c$-ring. Mammalian mitochondria use a compact 8-subunit $c$-ring (2.7 protons per ATP). Chloroplast turbines use 14 subunits (4.7 protons per ATP) to run under smaller voltages.

While the central $\gamma\varepsilon$ shaft spins at up to 100 revolutions per second inside the catalytic head, why doesn't the $\alpha_3\beta_3$ hexamer spin along with it? A slender peripheral stalk on the outside of the complex braces the $\text{F}_1$ crown to the membrane $a$-subunit. That outer arm flexes like an elastic torsion bar to smooth out stepping torque.

Single-molecule load measurements show that the $\text{F}_1$ motor stalls when an external magnetic trap applies $40\text{ pN}\cdot\text{nm}$ of backward torque. Multiplying that $40\text{-pN}\cdot\text{nm}$ stall torque by a 120-degree step ($2.09\text{ radians}$) yields $84\text{ pN}\cdot\text{nm}$ of mechanical work. The turbine captures nearly 100 percent of the physiological free energy released by one ATP molecule.

Electron cryo-tomography by Werner Kühlbrandt revealed that mitochondrial ATP synthases do not float solo in flat membranes. Instead, they pair into V-shaped dimers joined at an 86-degree angle in long rows. Those rigid rows physically bend the inner mitochondrial membrane into tight 20-nanometer cristae ridges that focus protons onto the turbine inlets.