The Sodium-Potassium Pump

The Na⁺/K⁺-ATPase (sodium-potassium pump) is an integral membrane protein found in the plasma membrane of essentially all animal cells. It belongs to the P-type ATPase family, so named because the pump phosphorylates itself on an aspartate residue during its catalytic cycle. Each cycle hydrolyzes one molecule of ATP to ADP and inorganic phosphate, and couples that chemical energy to moving ions uphill: three sodium ions are expelled from the cytosol to the extracellular space, and two potassium ions are imported from outside to inside, all against their respective concentration gradients. Because unequal numbers of positive charges cross the membrane in opposite directions, the pump is electrogenic — it directly contributes a small hyperpolarizing current, though most of the resting membrane potential arises indirectly, from the ion gradients the pump creates being exploited by passive leak channels. The pump is a heterodimer built from a catalytic α-subunit (about 1,020 amino acids, ten transmembrane helices) and a smaller glycosylated β-subunit that is required for proper folding and membrane insertion; many tissues also express an auxiliary FXYD regulatory subunit that fine-tunes the pump's ion affinities. The α-subunit contains the ATP-binding site, the phosphorylation site, and the alternating ion-binding pocket that is the heart of the mechanism. Multiple isoforms of both subunits exist (α1–α4, β1–β3 in mammals), allowing different tissues — kidney, muscle, brain — to tune pump density and kinetics to their specific needs. Conceptually, the crucial idea is that this is primary active transport: the energy source is the chemical bond energy of ATP itself, not a pre-existing ion gradient (that would be Secondary Active Transport). The pump does not simply "let ions through" — it must physically bind ions on one side, change shape, and release them on the other side, at an energetic cost, because it is moving them the thermodynamically wrong way. This distinguishes it categorically from the passive leak channels and carriers that dissipate gradients rather than build them.

The reaction cycle of the Na⁺/K⁺-ATPase is called the Post-Albers cycle, and it explains how a single protein achieves strict directionality and a fixed stoichiometry rather than just shuttling ions back and forth randomly. The pump exists in two principal conformational states, E1 and E2, and phosphorylation/dephosphorylation toggles it between them.

Step 1 — Na⁺ binding (E1 state). In the E1 conformation, three high-affinity binding sites for Na⁺ face the cytosol. Three intracellular Na⁺ ions bind.

Step 2 — Phosphorylation and occlusion. ATP binds a cytosolic domain and transfers its terminal (γ) phosphate to a conserved aspartate residue in the pump, forming a high-energy aspartyl-phosphate intermediate (E1~P) and releasing ADP. This phosphorylation locks the three Na⁺ ions in an "occluded" state — trapped inside the protein, accessible to neither side of the membrane.

Step 3 — Conformational switch (E1P → E2P). The phosphorylated pump relaxes into the E2P conformation. This large conformational change reorients the ion-binding pocket so it now opens toward the extracellular space, and critically, it lowers the binding sites' affinity for Na⁺ while raising their affinity for K⁺. The three Na⁺ ions are released to the outside.

Step 4 — K⁺ binding and dephosphorylation. Two extracellular K⁺ ions bind the now high-affinity sites. This binding triggers hydrolysis of the aspartyl-phosphate bond, releasing inorganic phosphate (Pi) into the cytosol and again occluding the bound ions — this time the two K⁺ ions.

Step 5 — Return to E1 and K⁺ release. Loss of the phosphate group drives the pump back to the E1 conformation, reopening the ion pathway toward the cytosol and lowering K⁺ affinity while restoring Na⁺ affinity. The two K⁺ ions are released into the cytosol, and the cycle is ready to begin again. Each full cycle transports 3 Na⁺ out : 2 K⁺ in : 1 ATP hydrolyzed. A single pump molecule can complete roughly 100–150 cycles per second, and a typical cell membrane contains hundreds to thousands of pump molecules per square micrometer, giving enough aggregate flux to counteract the constant passive leak of Na⁺ in and K⁺ out through resting channels.

The affinity-switching trick in Step 3 is the crux of the whole mechanism. Because phosphorylation of the aspartate residue mechanically distorts the transmembrane helices that form the ion-binding cavity, the same physical pocket can have high affinity for Na⁺ in one conformation and high affinity for K⁺ in another. This is why the pump can never simply run in reverse under normal conditions — the ion that would need to rebind at each step has the wrong affinity for that conformation, making the whole cycle strongly favor one direction (it is, in principle, reversible, and can be driven backward experimentally by imposing very large opposing gradients plus high ADP and Pi, which is part of how the mechanism was confirmed). The pump's directionality also depends on its orientation in the membrane: it is inserted with its cytosolic loops containing the ATP/phosphorylation site facing the cytoplasm, and the ion-binding transmembrane core spanning the full bilayer. The three Na⁺-binding sites and two K⁺-binding sites are not fully identical geometric pockets — structural biology (X-ray crystallography of the related SERCA Ca²⁺-ATPase and later of the Na⁺/K⁺-ATPase itself) shows overlapping but distinct coordination geometries that explain the asymmetric 3:2 stoichiometry.

Two further physiological points matter. First, the pump is directly regulated: insulin and thyroid hormone increase the number of pumps inserted into the membrane, aldosterone upregulates pump expression in the kidney's principal cells to drive Na⁺ reabsorption, and phosphorylation of the pump by protein kinases modulates its turnover rate. Second, downstream of the pump, the steep Na⁺ gradient it creates is the "battery" that secondary active transporters (like the Na⁺-glucose cotransporter SGLT1, or the Na⁺/Ca²⁺ exchanger) discharge to move other solutes uphill — the pump does not directly move glucose or calcium, but nothing else can move them without the gradient the pump builds.

The pump was discovered by the Danish physiologist Jens Christian Skou in 1957, in an experiment that is now a textbook case of careful biochemical reasoning from an unglamorous tissue: crab nerve.

Skou's original experiment. Skou was studying local anesthetics and needed a simple, abundant source of nerve cell membrane. He used crab (Carcinus maenas) leg nerve, homogenized it, and prepared a membrane fraction rich in an ATPase activity. He found that this ATPase's activity was:

• stimulated by Na⁺ and K⁺ together, far more than by either ion alone
• further stimulated by Mg²⁺
• inhibited by the plant steroid ouabain, a known blocker of active Na⁺ transport in intact cells

This combination — obligatory co-requirement for both Na⁺ and K⁺, Mg²⁺-dependence, and ouabain sensitivity — matched exactly the pharmacological and ionic signature of the "active sodium transport" that physiologists had inferred indirectly from whole-cell flux studies (using radioactive ²⁴Na and ⁴²K to track ion movement against gradients) for over a decade, but had never isolated as a molecule. Skou correctly concluded that he had identified the actual enzyme responsible for active Na⁺/K⁺ transport across the plasma membrane. He received the Nobel Prize in Chemistry in 1997 for this discovery.

Establishing the stoichiometry. The precise 3 Na⁺ : 2 K⁺ : 1 ATP ratio was pinned down through radiotracer flux experiments in resealed red blood cell "ghosts" (erythrocyte membranes emptied of contents and resealed around a controlled internal solution, pioneered by groups including Joseph Hoffman and Robert Post). By loading ghosts with known concentrations of ATP, Na⁺, and K⁺ and measuring efflux of radiolabeled ²²Na and influx of ⁴²K per mole of ATP hydrolyzed, researchers consistently measured a ratio of 3 Na⁺ effluxed and 2 K⁺ influxed for every ATP consumed — a fixed, non-negotiable coupling ratio, not a variable one, which is the hallmark of a tightly coupled molecular machine rather than independent leak pathways.

Confirming the phosphoenzyme intermediate — Robert Post's experiments. Robert L. Post and colleagues in the 1960s demonstrated that the pump forms a covalent phosphorylated intermediate during its cycle. Using ATP labeled with ³²P in the γ-phosphate position, they showed that the enzyme becomes transiently radiolabeled (phosphorylated) in a reaction that requires Na⁺ and Mg²⁺, and that this phosphate is subsequently released only when K⁺ is added — exactly the sequence predicted by the E1~P → E2P → dephosphorylation model. This was direct chemical proof of the aspartyl-phosphate intermediate central to the Post-Albers cycle (named for Post and the Danish biochemist Robert Albers, who independently proposed the conformational-switching model in the same period).

Vanadate trapping. A further, elegant confirmation came from the observation that vanadate ion (VO₄³⁻), a transition-state mimic of phosphate, binds the pump in its E2 conformation and locks it there, freezing the cycle and inhibiting transport — strong independent evidence that E2 is a genuine, distinct conformational state through which the enzyme must pass, not a theoretical construct.

Ouabain-binding stoichiometry. Because cardiac glycosides like ouabain bind the pump specifically and essentially irreversibly at the extracellular K⁺-binding face, researchers used radiolabeled ³H-ouabain binding assays to directly count the number of pump molecules per cell and to correlate pump density with the maximal rate of active Na⁺ efflux — closing the loop between molecule counts, biochemistry, and whole-cell physiology.

Example 1 — Quantifying the pump's electrical contribution. A cardiac myocyte has a resting membrane capacitance of 100 pF and rests at −80 mV. Its Na⁺/K⁺-ATPase density is such that the pump current (the net outward current due to 3-in/2-out charge imbalance per cycle) is measured at 0.5 pA per cell at steady state. Question: how many pump cycles per second does this represent, and does this pump current meaningfully affect the resting potential on its own? Each cycle moves 3 Na⁺ out and 2 K⁺ in — a net movement of +1 elementary charge out of the cell per cycle (3 positive charges leave, only 2 positive charges enter, net +1 charge leaves).

Charge per cycle = 1 elementary charge = 1.6 × 10⁻¹⁹ C.

If the cell has roughly 30 million pump molecules (a plausible density for excitable tissue), this corresponds to about 100 cycles per second per pump molecule — right in line with the experimentally measured single-pump turnover rate of 100–150 s⁻¹. This cross-check (whole-cell current ÷ single-channel/pump turnover = expected pump number) is exactly the logic used experimentally to estimate pump density from ouabain-sensitive current, and it shows the direct electrogenic contribution is small compared to the large resting conductances set by leak K⁺ channels — the pump's major effect on membrane potential is indirect, via maintaining the gradients that those channels then exploit.

Example 2 — Clinical reasoning: digoxin toxicity. A patient is given digoxin (a cardiac glycoside chemically related to ouabain) to treat heart failure, because at therapeutic doses partial inhibition of the cardiac Na⁺/K⁺-ATPase raises intracellular Na⁺ slightly, which reduces the gradient the Na⁺/Ca²⁺ exchanger uses to expel Ca²⁺, so more Ca²⁺ accumulates in the myocyte and boosts contractile force. The patient is also started on a thiazide diuretic, which causes mild hypokalemia (low plasma K⁺, dropping from 4.0 mM to 3.0 mM). Question: why does hypokalemia dangerously potentiate digoxin toxicity, given the pump's mechanism? Digoxin binds competitively at the pump's extracellular K⁺-binding site (the same site occupied in the E2P state). Extracellular K⁺ normally competes with digoxin for this site — high extracellular K⁺ tends to displace and reduce digoxin binding, while low extracellular K⁺ removes this competition and allows more digoxin to bind at any given drug concentration. So a drop in plasma K⁺ from 4.0 to 3.0 mM (a 25% decrease) does not just reduce pump turnover independently — it directly increases the fraction of pumps occupied by digoxin at the same drug dose, causing greater pump inhibition than the digoxin dose alone would predict. This over-inhibition raises intracellular Na⁺ further, drives excess Ca²⁺ accumulation via the Na⁺/Ca²⁺ exchanger, and can trigger dangerous cardiac arrhythmias. This is precisely why clinicians monitor and correct serum potassium carefully in any patient on digoxin — the mechanism of toxicity is a direct, predictable consequence of the pump's competitive ion-binding chemistry, not an unrelated side effect.

Jens Christian Skou (1918–2018) was a Danish physician and biochemist working at Aarhus University. Trained originally in medicine, Skou became interested in the mechanism of local anesthetics, which led him, almost as a side project, to investigate the ATPase activity of crab nerve membranes as a way of understanding membrane physiology more generally. His 1957 paper, published in Biochimica et Biophysica Acta, reported the Mg²⁺- and Na⁺+K⁺-dependent ATPase and correctly proposed that it was the physical basis of active Na⁺/K⁺ transport — a hypothesis that had been argued about on purely physiological (flux-measurement) grounds since the 1940s but never demonstrated at the level of an actual enzyme. Skou spent much of the rest of his career elucidating the pump's kinetics and structure, and he remained scientifically active into his nineties. He shared the 1997 Nobel Prize in Chemistry (with Paul Boyer and John Walker, who were honored for ATP synthase) for the discovery of the ion-transporting enzyme, Na⁺,K⁺-ATPase.

Cardiology. Cardiac glycosides (digoxin, and historically ouabain) are still used in heart failure and certain arrhythmias precisely because they partially inhibit the cardiac Na⁺/K⁺-ATPase, indirectly boosting contractile Ca²⁺ via the Na⁺/Ca²⁺ exchanger. Because the therapeutic-to-toxic window is narrow and depends on plasma K⁺ (see Example 2 above), digoxin remains a classic teaching case in clinical pharmacology for drug-electrolyte interactions.

Nephrology and diuretics. In the kidney, basolateral Na⁺/K⁺-ATPase in tubular epithelial cells drives the entire cascade of Na⁺ reabsorption that many diuretic drug classes are designed around; aldosterone's action on blood pressure and volume works substantially by upregulating this pump's expression in collecting-duct principal cells.

Neurology and anesthesia. During ischemic stroke, ATP depletion causes pump failure within minutes, collapsing ionic gradients, depolarizing neurons, and triggering excitotoxic glutamate release — a central mechanism of ischemic neuronal injury, and a target of ongoing neuroprotective drug research.

Diabetes and metabolism. Insulin acutely stimulates Na⁺/K⁺-ATPase activity (partly by promoting pump insertion into the membrane), which is why insulin, along with glucose, is used clinically to treat dangerous hyperkalemia — it drives K⁺ from plasma into cells.

Agriculture and toxicology. Cardiac glycosides occur naturally in plants like foxglove and milkweed as a chemical defense; monarch butterflies famously sequester milkweed glycosides to poison predators, exploiting the same pump-inhibition mechanism used therapeutically in humans.

Cell biology of volume and transport. Because the pump is the ultimate energy source for essentially all Na⁺-coupled secondary transport, it underlies intestinal glucose and amino acid absorption (relevant to oral rehydration therapy design in treating diarrheal disease) and regulatory volume control in virtually every animal cell — direct extensions of its role established in Tonicity and Cell Volume.

Ask the tutor: How does the Na⁺/K⁺-ATPase's electrogenic current actually change during a single action potential, and why doesn't it interfere with the much faster voltage-gated Na⁺ and K⁺ channel currents?