The body’s energy demands are driven by high energy phosphates in the form of adenosine triphosphate (ATP) and its equivalents. There are two ways in which ATP can be generated. The first is by the process of substrate-level phosphorylation where the phosphate group from a metabolic intermediate (substrate) is transferred to a molecule of adenosine diphosphate (ADP) to generate ATP. Such reactions occur in glycolysis, are catalysed by kinase or phosphorylase enzymes, and occur in both the cytoplasm and the mitochondria. The second, and by far the most important mechanism for the production of ATP, is the process of oxidative phosphorylation which, in eukaryotes, only occurs in mitochondria. Oxidative phosphorylation is the process in which ATP is formed as a result of the transfer of electrons from the reduced cofactors, NADH or FADH2, to O2 by a series of electron carriers.
There are two stages in the process. The first involves the electron transport chain, a series of four protein complexes that re-oxidize the NADH and FADH2 cofactors generating a proton motive force that pumps protons (hydrogen ions) out of the mitochondrial matrix into the intermembrane space, creating a proton (pH) gradient and an electrical potential across the inner mitochondrial membrane. The second stage involves ATP synthase, a large, multi-enzyme complex which mechanically couples the chemical synthesis of ATP to the electrochemical flow of these protons down the concentration gradient back across the inner mitochondrial membrane into the mitochondrial matrix. ATP synthesis is driven by the proton flow through the ATP synthase complex which is composed of two rotator motors mounted on a central shaft and held together by an eccentric bearing. The theoretical yield from oxidative phophorylation is 3 ATP per NADH and 2 ATP per FADH2. Inefficiencies reduce the actual yield to ~2.5 per NADH and 1.5 per FADH2.
The structure of ATP
ATP was proposed to be the intermediary molecule between energy-yielding and energy-requiring reactions in cells by the German biochemist Fritz Lipmann between 1939 and 1941, who coined the phrase "energy-rich phosphate bonds". ATP was first chemically synthesized by the British biochemist Alexander Todd in 1948 for which he was awarded the 1957 Nobel Prize in Chemistry.
Metabolic processes that use ATP as an energy source convert it back into its precursors ADP or AMP and ATP is continuously recycled in living organisms. The human body, which on average contains only 250 grams of ATP, turns over its own body weight equivalent in ATP every day. ATP is the source of energy for: muscle contraction (by the linear motor protein myosin as it moves along an actin filament); maintaining body heat (thermogenesis) in
Figure 1. The structure of ATP. ATP consists of a purine base (adenine) attached to the 1' carbon atom of ribose, a pentose (5 carbon) sugar, creating the nucleoside adenosine, to which three phosphates are attached at the 5' carbon atom of the ribose creating the triphosphate nucleotide ATP. Nucleotides are phosphorylated glycosyl amines. Reproduced from ref .warm-blooded animals; pumping ions across cell membranes; the synthesis of nucleic acids and proteins; nerve transmission(by the protein kinesin as it pulls synaptic vesicles along microtubules) and numerous intracellular signaling pathways (such as the insulin signaling pathways), by protein and lipid kinases which activate or inactivate signaling molecules by phosphorylation. ATP is also used by adenylate cyclase to produce the second messenger molecule cyclic AMP. The ratio between ATP and AMP is used by the cell to sense how much energy is available and to control the metabolic pathways that produce and consume ATP.
The structure of ATP with its three phosphate groups attached to the 5' carbon atom of the ribose moiety of adenosine (a glycosyl amine) is shown in Figure 1. ATP, ADP and AMP are interconverted by the addition and removal of these phosphate groups.
The overall process of oxidative phosphorylation
Oxidative phosphorylation is the process in which ATP is formed as a result of the transfer of electrons from NADH or FADH2 to O2 by a series of electron carriers. Like the tricarboxylic acid cycle (TCA cycle), oxidative phosphorylation takes place in mitochondria (Figure 2).
Figure 2. Cartoon of mitochondrion structure. The internal membrane of a mitochondrion is elaborately folded into structures known as cristae. Cristae increase the surface area of the inner membrane, which houses the components of the electron-transport chain. Reproduced from Wikipedia - Mitochondrial intermembrane space with permission. Click to enlargeThe main oxidizing agent used during aerobic metabolism of carbohydrates, fat and proteins is NAD+ with the cofactor FAD used in only one reaction (succinate dehydrogenase) in the TCA cycle. These get reduced to NADH and FADH2 during glycolysis and the TCA cycle. Unless the NAD+ and FAD can be regenerated, glycolysis and the TCA cycle would grind to a halt. NADH and FADH2 are oxidized back to NAD+ and FAD not directly by oxygen, but indirectly as electrons flow, via a series of electron carriers, from these reduced cofactors to oxygen forming water. The process is called electron transport and is the first stage of oxidative phosphorylation.
The NADH and FADH2 formed in glycolysis, fatty acid oxidation and the TCA cycle are energy-rich molecules because each contains a pair of electrons having a high transfer potential. When these electrons are used to reduce molecular oxygen to water, a large amount of free energy is liberated, and used to generate ATP.
Figure 3. Complexes involved in oxidative phosphorylation. The enzymes corresponding to the boxed Enzyme Commission numbers are as follows: Complex I - 184.108.40.206, NADH dehydrogenase (ubiquinone); 220.127.116.11, NADH dehydrogenase; 18.104.22.168, NADH dehydrogenase (quinone); Complex II - 22.214.171.124, succinate dehydrogenase (ubiquinone); 126.96.36.199, succinate dehydrogenase; Complex III - 188.8.131.52, ubiquinol-cytochrome-c reductase; Complex IV - 184.108.40.206, cytochrome-c oxidase; Complex V - 220.127.116.11, H+-transporting two-sector ATPase; 18.104.22.168, hydrogen/potassium-exchanging ATPase; 22.214.171.124, proton-exporting ATPase; 126.96.36.199, inorganic diphosphatase; 188.8.131.52, polyphosphate kinase. Reprinted with permission from Kanehisa Laboratories and the KEGG project: www.kegg.org. Click to enlargeAs summarized in Figure 3, the flow of electrons from NADH (or FADH2) to O2 is carried out by a set of four protein complexes located in the mitochondrial inner membrane called the electron transport chain. Within proteins, electrons are transferred between flavin cofactors, iron–sulfur clusters, and heme groups. Iron–sulfur clusters can be simple with two iron atoms joined by two atoms of inorganic sulphur, [2Fe–2S] clusters; or the more complex [4Fe–4S] clusters, which contain a cube of four iron atoms and four sulfur atoms. Each iron atom in these clusters is coordinated by an additional amino acid, usually the sulfur atom of cysteine. The Fe-S metal ion cofactors undergo redox reactions without binding or releasing protons, so in the electron transport chain they serve solely to transport electrons through proteins. Electrons move quite long distances through proteins by hopping along chains of these cofactors by a process of quantum tunneling.
The energy released by electrons flowing through this electron transport chain is used to transport protons out of the mitochondrial matrix, across the inner mitochondrial membrane, in a process called electron transport. The resulting uneven distribution of protons generates a pH gradient and a transmembrane electrical potential that creates a proton-motive force. This electrochemical gradient has two components: a difference in proton concentration (a H+ gradient, ΔpH) and a difference in electric potential, with the inner side of the membrane having a negative charge.
This store of energy is tapped and ATP synthesized when protons flow back across the membrane and down this gradient to the inner mitochondrial matrix through a fifth multiprotein complex, the enzyme ATP synthase. This process where energy is transferred from the electron transport chain to the ATP synthase by movements of protons across this membrane, is known as chemiosmosis. ATP synthase uses this electrochemical energy to synthesise ATP from ADP in a phosphorylation reaction that is driven by the proton flow. The kinetic energy of this proton flow forces the rotation of two components of the ATP synthase complex and couples this mechanical motion to the chemical synthesis of ATP
Comparative energy yields
Oxidative phosphorylation is the major source of ATP in aerobic organisms. It is a highly efficient way of releasing energy, compared to alternative fermentation processes such as anaerobic fermentation. Glycolysis produces a net gain of only 2 ATP molecules per molecule of glucose. In contrast the combination of glycolysis (2 NADH), the conversion of pyruvate to acetyl-CoA (2 NADH ) and the complete oxidation of acetyl-CoA via the TCA cycle (6 NADH + 2 FADH2 ) generates a total of 10 NADH + 2 FADH2 which when re-oxidised by oxidative phosphorylation can potentially produce 36 ATPs (net 2 ATPs from glycolysis, 30 ATPs from the 10 NADH, 4 ATPs from the 2 FADH2). These ATP yields are theoretical maximum values; in practice, some protons leak across the membrane, lowering the oxidative phosphorylation yield to ~30 ATP plus 2 ATP from glycolytic substrate-level phosphorylation.
Complexes involved in Oxidative phosphorylation
Complexes I - IV, the electron transport chain
Complex I : NADH-coenzyme Q oxidoreductase. This is the first protein in the electron transport chain and is a giant complex in mammals having 46 protein subunits and a molecular mass of about 1,000 kDa (Figure 3). The reaction catalyzed by this enzyme is the two electron oxidation of NADH by coenzyme Q10 or ubiquinone (represented as Q in the equation below), a lipid-soluble quinone that is found in the mitochondrion membrane.
NADH + Q +5 H+matrix -> NAD+ + QH2 + 4H+intermembrane
The start of the reaction, and indeed of the entire electron chain, is the binding of an NADH molecule to complex I and the donation of two electrons. The electrons enter complex I via the prosthetic group flavin mononucleotide (FMN) which is attached to the complex. The addition of electrons to FMN converts it to its reduced form, FMNH2. The electrons are then transferred through the second kind of prosthetic group in complex I - a series of [2Fe–2S] and [4Fe–4S] iron-sulphur clusters.
As the electrons pass through this complex, four protons are pumped from the mitochondrial matrix into the inter-membrane space. Finally, the electrons are transferred from the chain of iron–sulfur clusters to a ubiquinone molecule in the membrane. Reduction of ubiquinone also contributes to the generation of the proton gradient, as a further two protons are taken up from the matrix as ubiquinone is reduced to ubiquinol (QH2).
Complex II - Succinate-Q oxidoreductase.
Succinate–Q oxidoreductase also known as complex II or succinate dehydrogenase, is a second (alternative) entry point to the electron transport chain. It is unusual as it is part of both the tricarboxylic acid cycle and the electron transport chain. Complex II consists of four protein subunits (Figure 3) and contains a bound flavin adenine dinucleotide (FAD) cofactor, iron–sulfur clusters, and a heme group that does not participate in electron transfer to coenzyme Q10, but is believed to be important in decreasing the production of harmful reactive oxygen species. Complex II oxidizes succinate to fumarate and reduces ubiquinone to ubiquinol as shown in the equation below.
Succinate + Q -> Fumarate + QH2
This reaction releases less energy than the oxidation of NADH by complex I. Consequently complex II does not transport protons across the membrane and does not contribute to the proton gradient.
Complex III: Q-cytochrome c oxidoreductase
Q-cytochrome c oxidoreductase is also known as cytochrome c reductase, cytochrome bc1 complex, or simply complex III. In mammals, this enzyme is a dimer, (Figure 3) with each complex monomer containing 11 protein subunits, a [2Fe-2S] iron–sulfur cluster and three cytochromes (one cytochrome c1 and two b cytochromes). A cytochrome is a kind of electron-transferring protein that contains at least one heme group. The iron atoms inside the heme groups of complex III, alternate between a reduced ferrous (+2) and oxidized ferric (+3) state as the electrons are transferred through the protein.
The reaction catalyzed by complex III is the oxidation of one molecule of ubiquinol (QH2, reduced coenzyme Q10) and the reduction of two molecules of cytochrome c, a heme protein that is loosely associated with the mitochondrion. Unlike coenzyme Q10, which carries two electrons, cytochrome c carries only one electron.
QH2 + 2 Cyt cox + 2 H+matrix -> Q + 2 Cyt cred + 4 H+intermembrane
As only one of the electrons can be transferred from the QH2 donor to a cytochrome c acceptor at a time, the reaction mechanism of complex III is more elaborate than those of the other respiratory complexes, and occurs in two steps called the Q cycle. In the first step, the enzyme binds three substrates – ubiquinol (QH2), cytochrome c, ubiquinone (Q). The QH2 is oxidized, with one electron being passed to the second substrate, cytochrome c. The two protons released from QH2 pass into the intermembrane space. The third substrate, Q, accepts the second electron from QH2 and is reduced to Q−, which is the ubesemiquinone free radical. The first two substrates (QH2 and cytochrome c) are released, but the ubisemiquinone intermediate remains bound. In the second step, a second molecule of QH2 is bound and again passes its first electron to a cytochrome c acceptor while its second electron is passed to the bound ubisemiquinone, reducing it to QH2 as it gains two protons from the mitochondrial matrix. This QH2 is then released from the enzyme.
As coenzyme Q10 is reduced to ubiquinol on the inner side of the membrane and oxidized to ubiquinone on the outer side, a net transfer of protons across the membrane occurs, adding to the proton gradient. The rather complex two-step mechanism by which this occurs is important, as it increases the efficiency of proton transfer. If, instead of the Q cycle, one molecule of QH2 were used to directly reduce two molecules of cytochrome c, the efficiency would be halved, with only one proton transferred per cytochrome c reduced.
Complex IV: Cytochrome c oxidase
Cytochrome c oxidase, also known as complex IV, is the final protein complex in the electron transport chain. The mammalian enzyme has an extremely complicated structure (Figure 3) and contains 13 protein subunits, two heme groups, and multiple metal ion cofactors (three atoms of copper, one of magnesium and one of zinc). This enzyme mediates the final reaction in the electron transport chain and transfers electrons to oxygen, while pumping protons across the membrane. The final (terminal) electron acceptor, oxygen, is reduced to water in this step. Both the direct pumping of protons and the consumption of matrix protons in the reduction of oxygen contribute to the proton gradient. The reaction catalyzed by complex IV is the oxidation of cytochrome c and the reduction of oxygen.
4 Cyt cred + O2 + 8H+matrix -> 4 Cytcox + 2 H2O + 4H+intermembrane
Complex V - ATP synthase
ATP synthase, also called complex V, is the final enzyme in the oxidative phosphorylation pathway (Figure 3). The enzyme uses the potential energy stored in a proton gradient across a membrane to drive the synthesis of ATP from ADP and phosphate (Pi). Estimates of the number of protons required to synthesize one ATP have ranged from three to four.
ADP + Pi + 4H+intermembrane <--> ATP + H2O + 4H+matrix
As shown in Figures 3 and 4, ATP synthase is a massive protein complex with a mushroom-like head and stalk shape. The mammalian enzyme complex contains 16 protein subunits and has a mass of approximately 600 kilodaltons. It has a bipartite structure with the membrane –intrinsic portion F0 responsible for proton transport and the F1 portion (stalk and ball-shaped headpiece) performing the task of ATP synthesis/hydrolysis. When F1 is detached from F0 and thus decoupled from the proton motive force, it catalyses only ATP hydrolysis.
The portion embedded within the membrane, F0, contains a ring of 10 ‘c’ subunits and the proton channel. The F1 portion of ATP synthase is composed of a ball-shaped head made up of six proteins (three α subunits and three β subunits) arranged alternately like the segments of an orange, around a central, 90 Å long, α-helix formed by the C-terminal residues 209-272 of the γ subunit. This α-helix extends through the (αβ)3 ball with its C-terminal residue emerging in a dimple 15 Å deep in the surface at the top of the F1 assembly(Figure 4). The lower half of the γ209-272 helix forms a left-handed anti-parallel coiled-coil with a second α-helix composed of amino acids 1-45 from the N-terminus of the γ subunit. In the crystal structure this helical structure was seen to protrude as a stem about 30 Å from the bottom of the main body of F1 and represents most of the known 45 Å stalk of ATP synthase. The authors concluded that the δ and ε subunits are also probably in this exposed stalk region but were not seen owing to disorder in the crystal.
Figure 4. Structure of ATP synthase. (A) 3D model from the crystal structure of ATP synthase,(B) cartoon of the ATP synthase complex. The enzyme complex is composed of two subunits, F0 and F1. The F0 subunit is embedded in the inner mitochondrial membrane, and F1 subunit, containing the catalytic parts, protrudes into the mitochondrial matrix. The F1 subunit consists of five polypeptide chains, α, β, γ, δ and ε. The α and β chains, which make up the bulk of F1 are arranged alternatively in a hexameric ring. Both α and β bind nucleotides, but only β participates in catalysis. The γ subunit is a long helical coiled coil that extends into the middle of the αβ trimer. The γ subunit breaks the symmetry of the α/β hexamer, because each α/β monomer interacts with a different face of the γ subunit. The F0 subunit contains the proton channel formed by ten ‘c’ units. An ‘a’ subunit binds to F0. The F0 and F1 units are connected by the γ/ε stalk and by an exterior column made up of the ‘a’ subunit, two ‘b’ subunits, and the δ subunit. Reproduced from refswith permission. Click to enlargeThere are three catalytic nucleotide binding sites in the (α/β)3 head of F1. Both the α and β subunits bind nucleotides, but only the β subunits catalyze the ATP synthesis reaction. Reaching along the side of the F1 portion and back into the membrane is a long rod-like subunit that anchors the α and β subunits into the base of the enzyme. This unit acts as a stator and is composed of the ‘a’ subunit, two ‘b’ subunits from F0 and the δ subunit from F1 (Figure 4).
As protons cross the membrane through the channel in the base of ATP synthase, both the F0 proton-driven motor and the engaged γ/ε F1 stalk rotate relative to the (αβ)3 head as illustrated in the video ATP synthase from John Walkers laboratory at the University of Cambridge, UK. Other details and key papers of the structure of ATP synthase can be found at that site.
Rotation might be caused by changes in the ionization of amino acids in the ring of ‘c’ subunits causing electrostatic interactions that propel the ring of ‘c’ subunits past the proton channel. The ‘c’ ring is tightly linked to the γ subunit, so that the rotating ‘c’ ring drives the rotation of the central axle (the γ/ε subunit stalk) within the α and β subunits. The α and β subunits are prevented from rotating themselves by the side-arm (comprised of two ‘b’ chains and the δ subunit), which acts as a stator. This movement of the shaft of the γ subunit within the ball of α and β subunits provides the energy for the active sites in the β subunits to undergo a cycle of movements that produces and then releases ATP.
This ATP synthesis reaction is called the binding change mechanism and involves each of the three active sites of the (αβ)3 head cycling between three states. In the ‘open’ state, ADP and phosphate enter the active site. The protein then closes up around the molecules and binds them loosely – the ‘loose’ state. The enzyme then changes shape again and forces these molecules together, with the active site in the resulting ‘tight’ state binding the newly produced ATP molecule with very high affinity. Finally, the active site cycles back to the ‘open’ state, releasing ATP and initiating the next cycle by binding more ADP and phosphate. The portion of the γ subunit that passes through the (αβ)3 head is an asymmetric curved α-helix  whose rotation drives the inter-conversion of each β position. The three reactive sites are filled with ATP, ADP or empty depending on their respective positions relative to the concave, neutral or convex sides of the γ subunit shaft. In this way each β subunit repeatedly cycles from open --> loose-->tight, and no two β subunits will ever be in the same conformation at the same time. Each 360 degree rotation leads to the synthesis of 3 molecules of ATP. Given there are 10 ‘c’ subunits then each ATP generated requires the transport of 10/3 = 3.33 protons.
Historical aspects of oxidative phosphorylation
The field of oxidative phosphorylation began with the report in 1906 by British biochemist Arthur (later Sir Arthur) Hayden of a vital role for phosphate in cellular fermentation, but initially only sugar phosphates (substrate-level phosphorylation) were known to be involved. From 1937 to the early 1940s the link between the oxidation of sugars and the generation of ATP was firmly established by the Danish scientist Herman Kalckar confirming the central role of ATP in energy transfer that had been proposed by Fritz Lipmann in 1941. The term oxidative phosphorylation was coined by the Russian biochemist Vladimir Belitser in 1939. In 1949, Americans Morris Friedkin and Albert Lehninger proved that the coenzyme NADH linked metabolic pathways such as the tricarboxylic acid cycle and the synthesis of ATP.
For the next decade, the mechanism by which ATP is generated remained mysterious, with scientists searching for an elusive "high-energy intermediate" that would link oxidation and phosphorylation reactions. This puzzle was solved when in 1961, British scientist Peter Mitchell showed that cell respiration leads to differing concentrations of hydrogen ions (pH) inside and outside the mitochondrial membrane and developed the chemiosmotic hypothesis. At first, this proposal was highly controversial, but it was slowly accepted and Mitchell was awarded the Nobel Prize in Chemistry in 1978. At the same time international research concentrated on purifying and characterizing the enzymes involved, with major contributions being made by American scientist David Green on the complexes of the electron-transport chain, and by the Austrian-born, American biochemist Efraim Racker, who isolated the F1 part of the ATP synthase complex in 1961.
In 1964 Paul Boyer proposed that ATP is synthesized through structural changes in the ATP synthase enzyme and in 1973 discovered that the step in ATP synthesis which requires energy, is the release of ATP and the binding of ADP together with Pi (The Binding Change Mechanism). This was a critical step towards solving the mechanism by which ATP synthase generated ATP and was followed by his radical proposal of rotational catalysis in 1982.
In 1981 British chemist John E Walker determined the DNA sequence of the genes encoding the proteins in ATP synthase and, in 1994, solved the crystal structure of the F1 part of the ATP synthase. In 1997 Walker and Boyer were awarded the Nobel Prize in Chemistry for ‘their elucidation of the enzymatic mechanisms underlying the synthesis of ATP’.
In 1996-1997 the hypothesis that parts of ATP synthase rotate during the synthesis and hydrolysis of ATP was demonstrated spectroscopically (Wolfgang Junge, Germany), chemically (Richard Cross, USA) and microscopically (Masasuke Yoshida and Kazuhiko Kinosita, Japan).
^ Lord (Alexander R) Todd - http://www.nobelprize.org/nobel_prizes/chemistry/laureates/1957/index.html
^ Jakubowski H, ‘Biochemistry Online: An Approach Based on Chemical Logic’, Chapter 8, Oxidation/Phosphorylation, C: ATP and Oxidative Phosphorylation Reactions - http://employees.csbsju.edu/hjakubowski/classes/ch331/oxphos/olcouplingoxphos.html
^ ATP synthase - Mitochondrial Biology Unit [http://www.mrc-mbu.cam.ac.uk/research/atp-synthase]
^ Berg JM et al., A Family of Membrane Proteins Uses ATP Hydrolysis to pump ions across membranes, Biochemistry, 5th edition, Section 13.2, WH Freeman, NY.
^ Lodish H, et al., Molecular Cell Biology. 4th edition, Section 4.3, ‘Nucleic Acid Synthesis’, 2000, WH Freeman and Company.
^ Lodish H, et al., Molecular Cell Biology. 4th edition, Section 4.5, ‘Stepwise Formation of Proteins on Ribosomes’, 2000, WH Freeman and Company.
^ Protein Synthesis: The Nine Steps [dwb.unl.edu/Teacher/NSF/C08/C08Links/www.../Nine_steps.html]
^ Asbury CL, Fehr AN, Block SM, Kinesin Moves by an Asymmetric Hand-Over-Hand mechanism. Science 2003; 302:2130-4.
^ Berg JM, 2002, Oxidative Phosphorylation - Biochemistry - NCBI Bookshelf
^ Wikipedia - Mitochondrial intermembrane space - https://en.wikipedia.org/wiki/Mitochondrial_intermembrane_space
^ Mitchell P, Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of mechanism. Nature, 1961; 191(4784):144–8. Bibcode:1961Natur.191..144M. doi:10.1038/191144a0. PMID 13771349
^ Abrahams JP et al., Structure at 2.8 Å resolution of F1-ATPase from bovine heart mitochondria. Nature, 1994; 370:621-8.
^ Junge W, et al., ATP synthase: an electrochemical transducer with rotary mechanics. Trends Biochem Sci, 1997; 22:420-3
^ Junge W, ATP synthase and other motor proteins. Proc Natl Acad Sci U S A. 1999; 96(9):4735–7. PMCID: PMC33562
^ Cytochrome -http://en.wikipedia.org/wiki/Cytochrome
^ Proton Gradient and ATP Synthase - ACE Biochemistry - Yola [acebiochemistry.yolasite.com/proton-gradient-and-atp-synthase.php]
^ pH gradient leads to ATP synthesis - http://biowiki.ucdavis.edu/Biochemistry/Oxidation_and_Phosphorylation/ATP_and_Oxidative_Phosphorylation/pH_Gradient_Leads_to_ATP_Synthesis
^ Nobel Media AB 2015 - http://www.nobelprize.org/nobel_prizes/chemistry/laureates/1997/illpres/history.html
^ Sabbert D, et al., Intersubunit rotation in active F-ATPase. Nature, 1996; 381:623–6.
^ Zhou Y, et al., Subunit rotation in Escherichia coli F0F1–ATP synthase during oxidative phosphorylation, Proc Natl Acad Sci U S A. 1997; 94(20):10583–7.
^ Noji, H, et al., Direct observation of the rotation of F1-ATPase, Nature, 1997; 386:299 - 302; doi:10.1038/386299a0