P-type ATPase

The P-type ATPases, also known as E1-E2 ATPases, are a large group of evolutionarily related ion and lipid pumps that are found in bacteria, archaea, and eukaryotes. They are α-helical bundle primary transporters referred to as P-type ATPases because they catalyze auto- (or self-) phosphorylation of a key conserved aspartate residue within the pump. In addition, they all appear to interconvert between at least two different conformations, denoted by E1 and E2.

Most members of this transporter family are specific for the pumping of a large array of cations, however one subfamily is involved in flipping phospholipids to maintain the asymmetric nature of the biomembrane.

Prominent examples of P-type ATPases are the sodium-potassium pump (Na+,K+-ATPase), the plasma membrane proton pump (H+-ATPase), the proton-potassium pump (H+,K+-ATPase), and the calcium pump (Ca2+-ATPase).

Discovery
The first P-type ATPase discovered was the Na+,K+-ATPase, which Nobel laureate Jens Christian Skou isolated in 1957. The Na+,K+-ATPase was only the first member of a large and still-growing protein family, which in May 2013 had around 500 confirmed and unique members in Swiss-Prot (Prosite motif PS00154).

Phylogenetic classification
A phylogenetic analysis of 159 sequences made in 1998 by Axelsen and Palmgren showed that P-type ATPases can be divided into five subfamilies, based strictly on a conserved sequence kernel excluding the highly variable N and C terminal regions. The phylogenetic analysis grouped the proteins independent of the organism from which they are isolated and showed that the diversification of the P-type ATPase family occurred prior to the separation of eubacteria, archaea, and eucaryota. This underlines the significance of this protein family for cell survival.


 * Type I consists of the transition/heavy metal ATPases.
 * Type IA ATPases are involved in K+ import. They are atypical P-type ATPases because, unlike other P-type ATPases, they function as part of a heterotetrameric complex (called KdpFABC), where the actual K+ transport is mediated by another subcomponent of the complex.
 * Type IB ATPases are involved in transport of the soft Lewis acids: Cu+, Ag+, Cu2+, Zn2+, Cd2+, Pb2+ and Co2+. They are key elements for metal resistance and metal homeostasis in a wide range of organisms.


 * Type II ATPases are split into four groups.
 * Type IIA transports Ca2+. SERCA1a is a type IIA pump.
 * Type IIB transports Ca2+.
 * Type IIC consists of the closely related Na+/K+ and H+/K+ ATPases from animal cells.
 * Type IID contains a small number of fungal ATPases of unknown function.


 * Type III ATPases contains the plasma membrane H+-ATPases from plants and fungi (IIIA) and a small subdivision with Mg2+-ATPases from three bacterial species (IIIB).


 * Type IV ATPases have been shown to be involved in the transport of phospholipids, such as phosphatidylserine, phosphatidylcholine and phosphatidylethanolamine.


 * Type V ATPases have unknown specificity. This large group are found only in eukaryotes and are believed to be involved in cation transport in the endoplasmic reticulum.

Human genes
Human genes encoding P-type ATPases or P-type ATPase-like proteins include:


 * Na+/K+ transporting: ATP1A1, ATP1A2, ATP1A3, ATP1A4, ATP1B1, ATP1B2, ATP1B3, ATP1B4
 * Ca++ transporting: ATP2A1, ATP2A2, ATP2A3, ATP2B1, ATP2B2, ATP2B3, ATP2B4, ATP2C1
 * Cu++ transporting: ATP7A, ATP7B
 * Class I, type 8: ATP8A1, ATP8B1, ATP8B2, ATP8B3, ATP8B4
 * Class II, type 9: ATP9A, ATP9B
 * Class V, type 10: ATP10A, ATP10B, ATP10D
 * Class VI, type 11: ATP11A, ATP11B, ATP11C
 * H+/K+ transporting, gastric: ATP4A;
 * H+/K+ transporting, nongastric: ATP12A
 * type 13: ATP13A1, ATP13A2, ATP13A3, ATP13A4, ATP13A5
 * ATP2C2;
 * KIAA0195

Structure
Most of our knowledge about the structure and function of P-type ATPases originates from SERCA1a, a sarco(endo)plasmic reticulum Ca2+-ATPase of fast twitch muscle from adult rabbit. It is generally acknowledged that the structure of SERCA1a is representative for the family of P-type ATPases.

Structures of the Na+/K+-ATPase and H+-ATPase are also available.

Structure of SERCA1a
SERCA1a is composed of a cytoplasmic section and a transmembrane section with two Ca2+-binding sites. The cytoplasmic section consists of three cytoplasmic domains, designated the P, N, and A domains, containing over half the mass of the protein. The transmembrane section has ten transmembrane helices (M1-M10), with the two Ca2+-binding sites located near the midpoint of the bilayer. The binding sites are formed by side-chains and backbone carbonyls from M4, M5, M6, and M8. M4 is unwound in this region due to a conserved proline (P308). This unwinding of M4 is recognised as a key structural feature of P-type ATPases.

The P domain contains the canonical aspartic acid phosphorylated during the reaction cycle. It is composed of two parts widely separated in sequence. These two parts assemble into a seven-strand parallel β-sheet with eight short associated a-helices, forming a Rossmann fold.

The N domain is inserted between the two segments of the P domain, and is formed of a seven-strand antiparallel β-sheet between two helix bundles. This domain contains the ATP-binding pocket, pointing out toward the solvent near the P-domain.

The A domain is the smallest of the three domains. It consists of a distorted jellyroll structure and two short helices. It is the actuator domain modulating the occlusion of Ca2+ in the transmembrane binding sites, and it is pivot in transposing the energy from the hydrolysis of ATP in the cytoplasmic domains to the vectorial transport of cations in the transmembrane domain. The A domain dephosphorylates the P domain as part of the reaction cycle using a highly conserved TGES motif located at one end of the jellyroll.

ATP hydrolysis occurs in the cytoplasmic headpiece at the interface between domain N and P. Two Mg-ion sites form part of the active site. ATP hydrolysis is tightly coupled to Ca2+ translocation through the membrane, more than 40 Å away, by the A domain.

It is interesting to note that the folding pattern and the locations of the critical amino acids for phosphorylation in P-type ATPases has the haloacid dehalogenase fold characteristic of the haloacid dehalogenase (HAD) superfamily, as predicted by sequence homology. The HAD superfamily functions on the common theme of an aspartate ester formation by an SN2 reaction mechanism. This SN2 reaction is clearly observed in the solved structure of SERCA with ADP plus AlF4-.

Differences from SERCA1a
Various subfamilies of P-type ATPases also need additional subunits for proper function. Both P-IA and P-IV pumps needs extra subunits to function. The functional unit of Na+/K+-ATPase consists of two additional subunits, beta and gamma, involved in trafficking, folding, and regulation of these pumps. SERCA1a and other P-IIA ATPases are also regulated by phospholamban and sarcolipin in vivo. It is presumed that other subfamilies need additional subunits for the proper function in vivo, also.

Some members of the family have additional domains fused to the pump. Heavy metal pumps can have several N- and C-terminal heavy metal-binding domains that have been found to be involved in regulation.

The proton pumps (IIIA) have a C-terminal regulatory domain (called the R domain), which, when unphosphorylated, inhibit pumping.

While most subfamilies have 10 transmembrane helices, there are some notable exceptions. The P-IA ATPases are predicted to have 7, and the large subfamily of heavy metal pumps (IB) is predicted to have 8 transmembrane helices. Type V appears to have a total of 12 transmembrane helices.

Mechanism
All P-type ATPases use the energy derived from ATP to drive vectorial transport. They form a high-energy aspartyl-phosphoanhydride intermediate in the reaction cycle, and they interconvert between at least two different conformations, denoted by E1 and E2. The E1-E2 notation stems from the initial studies on this family of enzymes made on the Na+,K+-ATPase, where the sodium form and the potassium form are referred to as E1 and E2, respectively, in the "Post-Albers scheme".

The E1-E2 schema has been proven to work, but there exist more than two major conformational states. However, the E1-E2 notation highlights the selectivity of the enzyme. In E1, the pump has high affinity for the exported substrate and low affinity for the imported substrate. In E2, it has low affinity of the exported substrate and high affinity for the imported substrate.

Four major enzyme states form the cornerstones in the reaction cycle. Several additional reaction intermediates occur interposed. These are termed E1~P, E2P, E2-P*, and E1/E2, described below.

In the case of SERCA1a, energy from ATP is used to transport 2 Ca2+-ions from the cytoplasmic side to the lumen of the sarcoplasmatic reticulum, and to countertransport 1-3 protons into the cytoplasm.

Starting in the E1/E2 state, the reaction cycle begins as the enzyme releases 1-3 protons from the cation-ligating residues, in exchange for cytoplasmic Ca2+-ions. This leads to assembly of the phosphorylation site between the ATP-bound N domain and the P domain, while the A domain directs the occlusion of the bound Ca2+. In this occluded state, the Ca2+ ions are buried in a proteinaceous environment with no access to either side of the membrane.

The Ca2E1~P state becomes formed through a kinase reaction, where the P domain becomes phosphorylated, producing ADP. The cleavage of the β,-phosphodiester bond releases the gamma-phosphate from ADP and unleashes the N domain from the P domain.

This then allows the A domain to rotate toward the phosphorylation site, making a firm association with both the P and the N domains. This movement of the A domain exerts a downward push on M3-M4 and a drag on M1-M2, forcing the pump to open at the luminal side and forming the E2P state. During this transition, the transmembrane Ca2+-binding residues are forced apart, destroying the high-affinity binding site. This is in agreement with the general model form substrate translocation (cf. 1.2), showing that energy in primary transport is not used to bind the substrate but to release it again from the buried counter ions. At the same time the N domain becomes exposed to the cytosol, ready for ATP exchange at the nucleotide-binding site.

As the Ca2+ dissociate to the luminal side, the cation binding sites are neutralised by proton binding, which makes a closure of the transmembrane segments favourable. This closure is coupled to a downward rotation of the A domain and a movement of the P domain, which then leads to the E2-P* occluded state. Meanwhile, the N domain exchanges ADP for ATP.

The P domain is dephosphorylated by the A domain, and the cycle completes when the phosphate is released from the enzyme, stimulated by the newly bound ATP, while a cytoplasmic pathway opens to exchange the protons for two new Ca2+-ions.