The F0F1-ATPase

This substantial enzyme (Mr approximately 500,000) is readily visible in electron micrographs as 8.5 nm spheres attached to the matrix side of the mitochondrial inner membrane. The spheres can be detached by a variety of methods, after which they act as an ATPase. The physiological function of this enzyme is the synthesis of ATP, using the energy stored in the transmembrane pH and potential gradients.

The complete assembly contains at least 12 different types of polypeptide chain, several of which are present in multiple copies. The catalytic head group is connected by an oligomycin sensitive stalk to a proton conducting baseplate in the mitochondrial inner membrane. Three protons are thought to pass through the membrane from the external P phase to the internal N phase for each molecule of ATP manufactured by the complex.

The F1 head group contains three nucleotide binding sites, and the enzyme probably performs a three-phase catalytic cycle. In the first phase, ADP and phosphate bind to one active centre, which catalyses the formation of bound ATP. This step is energetically possible because the free energy released by tightly binding the ATP to the active centre compensates for the instability of the new phospho anhydride bond. The energy from the proton motive force is required to prise the ATP from the active centre.

The F0 base piece embedded in the mitochondrial inner membrane is a molecular turbine driven by the trans-membrane proton gradient. Proton entry forces a central camshaft to rotate within the F0 baseplate and the F1 head group, altering the subunit conformation as this movement takes place. A second, off-centre protein tether connects the head group to the base piece and prevents the head piece spinning uselessly as the central shaft rotates. Energy is transmitted to the catalytic subunits in the ATP synthase F1 headpiece by the rotation of the camshaft. The "cam" distorts the protein subunits, destroying their ability to bind ATP. The energy input is used to drive ATP release, not for bond formation.

It is presumably necessary to disable the catalytic mechanism on the centre which has just formed ATP (to stop this centre hydrolysing its own product) before destroying its ability to bind ATP. This allows the product to be released. Meanwhile, the two other active centres are performing their own parts of the catalytic cycle. The three active centres operate simultaneously, but 120o out of phase. It takes at least 9 protons (possibly as many as 12) to drive one revolution of the camshaft and produce 3 ATP molecules.

Remember that the whole complex is reversible. Normally the energy from the proton gradient is used to manufacture ATP, but it is equally possible in vitro to do things the other way round, and use the hydrolysis of ATP to drive the camshaft, and ultimately pump protons back through the turbine and into the extramitochondrial compartment. If the F0 base piece is not attached to a membrane nothing useful will be accomplished, and the complex will simply act as an ATPase, as was originally observed.

It is possible to directly observe the rotation of the F1ATPase cam shaft using a fluorescence microscope, although considerable ingenuity is required. Noji et al (1997) Nature 386, 299-302 used a genetically modified F1ATPase from a thermophilic bacterium expressed in E. coli. They discarded the F0 basepiece and tethered the F1 motor head groups to a glass plate using polyhistidine tags attached to the N-termini of all three beta subunits. The glass plate had been pre-treated with horseradish peroxidase conjugated with the nickel complex of nitrilotriacetic acid, to which polyhistidine binds with high affinity. [Nitrilotriacetic acid looks like half an EDTA molecule, so it leaves un-coordinated nickel positions available for external ligands.]

The motors were glued down by their large catalytic subunits, leaving the motor shafts exposed, and facing away from the glass. The gamma subunits which form the shaft were modified by site directed mutagenesis to remove the original Cys193 (which is inconveniently far down the shaft) and replace it with serine. These workers also replaced Ser107 in the stalk region with cysteine.

This single cysteine residue (the only one in the molecule) could then be biotinylated, and linked using streptavidin to fluorescently labelled, biotinylated actin filaments. [Streptavidin has four biotin binding sites.]

The fluorescent actin filaments were many times larger than the tethered motors and could be visualised in a light microscope.

Addition of 2mM ATP caused a small number of  the motor shafts marked by the actin filaments to rotate in a counter- clockwise direction. The movie shows the results they obtained.

ATPase movie

click the arrow on the
left to start the movie

Circular motion also occurs in the proteins which rotate bacterial flagellae, another important enzyme system which is driven by the proton motive force. It is apparent that the wheel has been in continuous use for at least 2000 million years.

Royal Route
Home Page