An important P-class pump that is present in the plasma mem branes of all animal cells is the Na+/K+ ATPase. This ion pump is a tetramer of subunit composition α2β2 and shares structural homology with the Ca2+ pump (see Figure 1). The small, glycosylated β transmembrane polypeptide apparently is not involved directly in ion pumping. During its catalytic cycle, the Na+/K+ ATPase moves three Na+ ions out of and two K+ ions into the cell per ATP molecule hydrolyzed. The mechanism of action of the Na+/K+ ATPase, outlined in Figure 2, is simi lar to that of the muscle SR calcium pump, except that ions are pumped in both directions across the membrane, with each ion moving against its concentration gradient. In its E1 conformation, the Na+/K+ ATPase has three high-affinity Na+-binding sites and two low-affinity K+-binding sites accessible from the cytosolic surface of the protein. The Km for binding of Na+ to these cytosolic sites is 0.6 mM, a value considerably lower than the intracellular Na+ concentration of ~12 mM; as a result, Na+ ions normally fully occupy these sites. Conversely, the affinity of the cytosolic K+-binding sites is low enough that K+ ions, transported inward through the protein, dissociate from E1 and enter the cytosol despite the high intracellular K+ concentration. During the E1 → E2 transition, the three bound Na+ ions gain access to the exoplasmic face, and simultaneously, the affinity of the three Na+-binding sites drops. The three Na+ ions, now bound to low-affinity Na+ sites, dissociate one at a time and enter the extracellular medium despite the high extracellular Na+ concentration. Transition to the E2 conformation also generates two high-affinity K+ sites accessible from the exoplasmic face. Because the Km for K+ binding to these sites (0.2 mM) is lower than the extracellular K+ concentration (4 mM), these sites will fill with K+ ions as the Na+ ions dissociate. Similarly, during the subsequent E2 → E1 transition, the two bound K+ ions are transported inward and then released into the cytosol.
Fig1. Structural comparison of Na+/K+ ATPase and muscle Ca2+ ATPase. Three-dimensional structure of the Na+/K+ ATPase (gold) compared with that of the muscle Ca2+ ATPase (purple), as seen from the cytoplasmic surface. αM1–αM10 denote the 10 membrane-spanning α helices of the Na+/K+ ATPase. [Data from J. P. Morth et al., 2007, Nature 450:1043, PDB ID 3b8e; and C. Toyoshima, H. Nomura, and T. Tsuda, 2004, Nature 432:361–368, PDB ID 1wpg.]
Fig2. Operational model of the plasma-membrane Na+/K+ ATPase. Only one of the two catalytic α subunits of this P-class pump is depicted. It is not known whether just one or both subunits in a single ATPase molecule transport ions. Ion pumping by the Na+/K+ ATPase involves phosphorylation, dephosphorylation, and conformational changes similar to those in the muscle Ca2+ ATPase (see Figure 3). In this case, hydrolysis of the E2–P intermedi ate powers the E2 → E1 conformational change and concomitant transport of two K+ ions inward. Na+ ions are indicated by red circles; K+ ions, by purple squares; high-energy acyl phosphate bond, by ~P; low-energy phosphoester bond, by –P.
Fig1. Structure of the catalytic α subunit of the muscle Ca21 ATPase. (a) Ca2+-binding sites in the E1 state (left), with two bound calcium ions, and the low-affinity E2 state (right), without bound ions. Side chains of key amino acids are white, and the oxygen atoms on the glutamate and aspartate side chains are red. In the high-affinity E1 conformation, Ca2+ ions bind at two sites between helices 4, 5, 6, and 8 inside the membrane. One site is formed out of negatively charged oxygen atoms from glutamate and aspartate side chains and from water molecules (not shown), and the other is formed out of side- and main-chain oxygen atoms. Seven oxygen atoms surround the Ca2+ ion in both sites. (b) Three-dimensional model of the protein in the E1 state based on the structure determined by x-ray crystallography. There are 10 transmembrane α helices, four of which (purple) contain residues that participate in Ca2+ binding. The cytosolic segment forms three domains: the nucleotide-binding domain (N, blue), the phosphorylation domain (P, green), and the actuator domain (A, beige), which connects two of the membrane-spanning helices. (c) Models of the pump in the E1 state (left) and in the E2 state (right). Note the differences between the E1 and E2 states in the conformations of the N and A domains. Movements of these domains power the conformational changes of the membrane-spanning α helices (purple) that constitute the Ca2+-binding sites, converting them from a conformation in which the Ca2+-binding sites are accessible from the cytosolic face (E1 state) to one in which the now loosely bound Ca2+ ions gain access to the exoplasmic face (E2 state). [Data from C. Toyoshima and G. Inesi, 2004, Annu. Rev. Biochem. 73:269–292, PDB ID 1su4; and K. Obara et al., 2005, P. Natl. Acad. Sci. USA 102:14489–14496, PDB ID 1agv.]
Certain drugs (e.g., ouabain and digoxin) bind to the exoplasmic domain of the plasma-membrane Na+/K+ ATPase and specifically inhibit its ATPase activity. The resulting disruption in the Na+/K+ balance of cells is strong evidence for the critical role of this ion pump in maintaining the normal K+ and Na+ ion concentration gradients.
