Atp Synthase Structure And Mechanism

The F0F1-ATP synthase, an enzyme found in all organisms, has intrigued scientists for almost 40 years. This highly asymmetric protein assembly drives one of Nature's most challenging and probably the most important chemical reaction, namely the synthesis of ATP. The challenge of this reaction is the formation of an anhydride bond between adenosine-5'-diphosphate (ADP) and inorganic phosphate (Pi) in an aqueous environment that would usually favor ATP hydrolysis and not the synthesis reaction. The energy

Copyright © 2004 by American Scientific Publishers All rights of reproduction in any form reserved.

Encyclopedia of Nanoscience and Nanotechnology Edited by H. S. Nalwa Volume 7: Pages (83-89)

for the synthesis reaction is provided by the flow of protons down an electrochemical gradient and across the corresponding energy coupling membrane. In bacteria this membrane is the plasma membrane, in eukaryotes it is the mitochondrial membrane, and in photosynthetic organisms this process mostly takes place in the thylakoid membrane of chloroplasts. In mitochondria and bacteria, the synthase efficiently couples the oxidation of NADH to the formation of ATP. Nicotinamide adenine dinucleotide in the reduced form (NADH) is a redox equivalent that is generated when an organisms digests nutritious compounds like sugars, fats, or proteins. The reducing energy (in the form of two electrons) that is stored in NADH is utilized in the respiratory chain (Fig. 1), where the electrons are transferred in step-wise redox processes to oxygen. This process leads to the reduction of oxygen to form water while NADH is oxidized to NAD+ (Fig. 1). During these redox processes protons (H+) are pumped across the energy-coupling membranes by the membrane enzymes involved. The energy stored in the generated proton gradient drives the ATP synthesis as we will see next.

In photosynthetically active organisms, a light-driven proton translocation is coupled to the endergonic formation of the terminal phosphoryl anhydride bond of ATP (ADP + Pi ^ ATP).

Under certain in vivo conditions, for example, in some bacteria when they live in an anaerobic environment, the reaction of the synthase can be reversed so that an ATP hydrolysis-driven proton pumping occurs.

Under physiological conditions, the free energy that is stored in the terminal (3—y)-anhydride bond of an ATP molecule is about —9 x 10—20 J, equaling about -50 kJ/mol of ATP, depending on the cellular conditions.

The ATP synthases from different organisms are highly conserved. Much of the research on F^-ATP synthases has focused on the bacterial enzyme from Escherichia coli, mainly because molecular biological tools are easily

Figure 1. The respiratory chain. Electrons are transferred from NADH and H+ to an enzyme called NADH dehydrogenase. From there the electrons are further transported along the membrane systems using the mobile carriers, coenzyme Q and cytochrome c. At the respective enzymes, NADH dehydrogenase, cytochrome bct complex, and cytochrome c oxidase, the energy from the electron transfer is transduced into pumping of protons across the membrane. The resulting proton gradient drives the synthesis of ATP by the F0Fj-ATP synthase. Where possible, the X-ray structural models of the corresponding proteins were created using Cn3D version 4.0 from the NCBI. The molecular coordinates of the protein structures can be found in the Protein Data Bank (cytochrome bcj complex, PDB-file: 1BGY; cytochrome c, PDB-file: 1GIW; cytochrome c oxidase, PDB-file: 1OCZ; and F0FrATP synthase, PDB: 1QO1).

Figure 1. The respiratory chain. Electrons are transferred from NADH and H+ to an enzyme called NADH dehydrogenase. From there the electrons are further transported along the membrane systems using the mobile carriers, coenzyme Q and cytochrome c. At the respective enzymes, NADH dehydrogenase, cytochrome bct complex, and cytochrome c oxidase, the energy from the electron transfer is transduced into pumping of protons across the membrane. The resulting proton gradient drives the synthesis of ATP by the F0Fj-ATP synthase. Where possible, the X-ray structural models of the corresponding proteins were created using Cn3D version 4.0 from the NCBI. The molecular coordinates of the protein structures can be found in the Protein Data Bank (cytochrome bcj complex, PDB-file: 1BGY; cytochrome c, PDB-file: 1GIW; cytochrome c oxidase, PDB-file: 1OCZ; and F0FrATP synthase, PDB: 1QO1).

available, but also because the E. coli enzyme is simpler in subunit composition and can be readily purified in rather large quantities. A second favored source for purifying the ATP synthase or substructures thereof is the thermophilic bacterium PS3. In addition to easy availability, the PS3 enzyme is also very stable at ambient and even elevated temperatures.

In all organisms the synthase consists of two components: the membrane embedded F0 sector that is responsible for the translocation of protons and the membrane associated F-l part that contains the sites for nucleoside triphosphate synthesis (Fig. 2). Both parts of the enzyme are highly asymmetric. The F0 part of the E. coli enzyme consists of three different polypeptides a, b, and c with a stoichiometry of 1a, 2b, and 9-12 copies of the c subunit per functional unit [2]. Subunits a, b, and c are membrane integral and a and c are directly involved in proton movement through the membrane (for review, see [3]). Subunit b has an N-terminal membrane anchor. The residual amino acids of the protein protrude into the cytoplasm and interact, at least partially, with the F1 part of the synthase, linking it to the membrane.

F1 is the catalytic part of the synthase and consists of the five different subunits a through e with three copies each of the major subunits a and fi and one copy each of the minor subunits, y, 8, and e [4]. If this F1 part is removed from the membrane and uncoupled from the proton gradient, it is only capable of net ATP hydrolysis, owing to the lack of driving force provided by the proton gradient. This part of the enzyme is therefore also named the F1-ATPase.

It has been known since the 1980s that six nucleotide binding sites are located on the a and fi subunits. The three binding sites that are located on the fi subunits are

Figure 2. Model of the F0Fj-ATP synthase. The diagram shows the two portions of the ATP synthase. The polypeptides a, b, and c are at least partially embedded in the membrane and are constituents of the F0 portion of the membrane. Subunit a and the ring of c subunits are involved in the translocation of protons across the membrane. The membrane-associated Fj part consists of subunits a, ¡3, y, 8, and s and is responsible for the synthesis of ATP. The subunits ab2a3338 are considered to be the stator to the rotary motor. Subunits c9-14ys were shown in different experiments to rotate during catalysis (for more information see text).

Figure 2. Model of the F0Fj-ATP synthase. The diagram shows the two portions of the ATP synthase. The polypeptides a, b, and c are at least partially embedded in the membrane and are constituents of the F0 portion of the membrane. Subunit a and the ring of c subunits are involved in the translocation of protons across the membrane. The membrane-associated Fj part consists of subunits a, ¡3, y, 8, and s and is responsible for the synthesis of ATP. The subunits ab2a3338 are considered to be the stator to the rotary motor. Subunits c9-14ys were shown in different experiments to rotate during catalysis (for more information see text).

catalytically active (catalytic sites) and show strong positive catalytic cooperativity. When all three sites are filled with substrate, the rate of catalysis is at least 105 times faster than if only one substrate is bound to the catalytic sites; for more details see [5-7]. The function of the three sites located on the a subunits remains unclear, and they are usually referred to as noncatalytic sites.

In the late 1970s, Boyer presented a model for the reaction mechanism of the ATPase that was termed the "binding change mechanism." One of the tenets of the theory stated that every one of the catalytically active nucleotide binding sites of the ATPase is in a different structural and functional state at any given time during catalysis and during sequential catalytic cycles; all of the sites go through the same states. While one of the sites is ready to bind the substrates, ADP and Pi, a second site has both these substrates bound, and a third site is in a conformation that allows the actual bond formation between ADP and Pi, resulting in elimination of water and the synthesis of ATP. The binding change mechanism model further states that the energy for the synthesis of ATP is solely required to release the product from the active site and not for the actual formation of the (fi-y)-anhydride bond. Experiments indicated that the equilibrium between ADP + Pi and ATP in the catalytic site is close to unity. The idea that these sites sequentially participate in this mechanism suggested that binding of substrate at one site promotes the release of product at a second site. This proposed mechanism beautifully accommodated the data that had been gathered about structure, subunit composition, and catalysis of the enzyme. One important question that remained at this point was what type of conformational transition enables the catalytic sites to sequentially change their conformation and function. The first suggestion that a rotation of internal subunits was part of the catalytic mechanism was published in the early 1980s, also by Boyer [8, 9] (for review see [10]).

Tremendous progress in our knowledge about the mechanism of ATP hydrolysis and synthesis was made in the past decade, gaining even more momentum when a high resolution structural model of the beef heart mitochondrial Fx-ATPase was presented in 1994 by Walker's group [11]. An X-ray crystallographic model of the beef heart mitochondrial ATPase is shown in Figure 3. The model gives detailed information about the structure of the nucleotide binding sites. In the model two of the catalytic sites are rather similar in conformation, whereas the third catalytic site differs significantly from the others in structure. This third side does not have a nucleotide bound (at least in the crystal structure), and a whole section of the binding site is rotated to form a very open conformation (able to release product). The model shows all the noncatalytic sites to be very similar in structure.

Several other structural models of various subassemblies of F1 or F0F1 from a variety of different sources have become available over the last few years [12-16], increasing our knowledge about the structural basis of the ATPase and ATP synthase.

The structural models made rational mutagenesis of key residues within the nucleotide binding sites possible and allowed researchers to monitor the occupancy of the nucleotide sites during turnover using biophysical techniques

Figure 3. X-ray structural model of the mitochondrial F1-ATPase [10] Left: side view, right: view from the membrane surface. The molecular coordinates of a variety of different F1-ATPases from different sources are available in the Protein Data Bank. Here the model is based on the coordinates of the beef heart mitochondrial enzyme as described in [10]. The model was created using the program Rasmol and the molecular coordinates from the Protein Data Bank.

[17]. These biophysical data suggested that at maximal turnover conditions and saturating ATP, all three catalytic sites are occupied in Fx as well as in solubilized F0FX [17-20]. Experiments from our group using electron spin resonance spectroscopy and spin-labeled nucleotides suggested that under equilibrium conditions almost all nucleotide binding sites are filled [21-23].

Was this article helpful?

0 0

Post a comment