Glycolysis: Payoff Phase and ATP Generation

The payoff-phase enzymes are not museum pieces; they are active drug targets and diagnostic tools. Pyruvate kinase M2 (PKM2), a splice isoform expressed preferentially in rapidly dividing and cancer cells, has an unusual allosteric regulation that keeps its activity low, favoring the accumulation of upstream glycolytic intermediates that feed biosynthetic side pathways (the pentose phosphate pathway, serine synthesis) — this is central to the Warburg effect and is an active area of oncology drug development, with small-molecule PKM2 activators and inhibitors both explored as anticancer strategies. GAPDH itself, beyond its glycolytic role, has documented "moonlighting" functions in nuclear signaling and apoptosis, and its reactive active-site cysteine makes it a sensitive target for reactive oxygen and nitrogen species, linking glycolytic flux to redox stress signaling. Inherited pyruvate kinase deficiency is one of the most common causes of hereditary non-spherocytic hemolytic anemia, because mature red blood cells — lacking mitochondria — depend entirely on glycolysis (and thus on the ATP made in steps 7 and 10) for their energy supply; without functional pyruvate kinase, red cells cannot maintain the ion pumps that preserve their membrane and shape, leading to premature hemolysis. Enolase inhibitors are being explored against certain parasites and cancers because tumor cells and some pathogens depend heavily on glycolytic ATP even in the presence of oxygen. Clinically, elevated serum enolase (particularly neuron-specific enolase, NSE) serves as a tumor marker for small-cell lung cancer and neuroendocrine tumors. And industrially, the entire payoff-phase enzyme set is harnessed (via whole-cell or cell-free systems) in bioethanol and biochemical fermentation processes, where maximizing net ATP and NADH throughput per glucose is directly tied to production economics.

Ask the tutor: How does the actual cellular [NADH]/[NAD⁺] ratio limit glycolytic flux through GAPDH when oxidative phosphorylation is impaired?

The five reactions of the payoff phase act on glyceraldehyde-3-phosphate (GAP), and because Glycolysis: Preparatory Phase produces two molecules of GAP from every glucose (one directly, one via isomerization of dihydroxyacetone phosphate), every stoichiometric number from here on must be doubled to describe one glucose's fate. The chemistry falls into two blocks. Steps 6 and 7 generate the pathway's first ATP by oxidizing an aldehyde into a carboxylic acid derivative and trapping the released free energy in an acyl-phosphate bond. Steps 8, 9, and 10 then take a "merely good" phosphate ester and, through positional isomerization and dehydration, concentrate its transfer potential into a phosphoenol group so extraordinarily reactive that its hydrolysis releases roughly twice as much free energy — energy spent making the pathway's second and final ATP.

Step 6 — Glyceraldehyde-3-phosphate dehydrogenase (GAPDH). This is the mechanistic heart of the payoff phase. GAPDH oxidizes the aldehyde of GAP while simultaneously phosphorylating it with inorganic phosphate (not ATP) to form 1,3-bisphosphoglycerate (1,3-BPG). The enzyme uses an active-site cysteine thiolate as a nucleophile: it attacks the aldehyde carbon to form a covalent thiohemiacetal. A hydride is then transferred from this intermediate to NAD⁺ bound in an adjacent site, generating a high-energy thioester (an acyl-enzyme intermediate) and NADH. Inorganic phosphate then attacks the thioester carbonyl, displacing the cysteine thiolate and generating the mixed anhydride 1,3-bisphosphoglycerate. The overwhelming thermodynamic point is that the energy of the aldehyde-to-carboxylic-acid oxidation — normally released as heat if it happened in free solution — is instead conserved as a covalent thioester bond and then handed off intact into a phosphoanhydride bond. Nothing is wasted.

Step 7 — Phosphoglycerate kinase. This enzyme transfers the high-energy phosphoryl group from the mixed-anhydride carbon (C1) of 1,3-BPG directly to ADP, producing 3-phosphoglycerate and ATP. This is a textbook example of substrate-level phosphorylation: a phosphoryl group moves from one substrate to another in a single active site, with no intervening free Pi and no membrane potential involved. Because the reaction is so favorable, the coupled sum of steps 6 and 7 (aldehyde oxidation plus two phosphoryl transfers) is strongly exergonic overall, even though step 6 alone, in isolation, is close to thermoneutral.

Step 8 — Phosphoglycerate mutase. 3-phosphoglycerate's phosphate sits on C3, but the pathway needs it on C2 to set up dehydration in step 9. The mutase shuttles the phosphoryl group via a phosphohistidine enzyme intermediate: a histidine residue is first phosphorylated by the substrate's C3 phosphate, then the same intermediate transfers phosphate to C2 of the same molecule (a 2,3-bisphosphoglycerate intermediate is formed transiently and then acts as the leaving substrate for the second transfer). Net result: 2-phosphoglycerate, with no change in free energy of any consequence and no ATP involved.

Step 9 — Enolase. This enzyme catalyzes a dehydration: loss of water from 2-phosphoglycerate creates a carbon-carbon double bond conjugated with the phosphate ester, producing phosphoenolpyruvate (PEP). This is the crucial energy-concentrating step. 2-phosphoglycerate is only a modest phosphate donor (ΔG°′ of hydrolysis ≈ −17.6 kJ/mol), but removing water creates an enol phosphate whose keto tautomer (pyruvate) is so much more stable that the phosphate ester bond in PEP becomes ferociously reactive (ΔG°′ of hydrolysis ≈ −61.9 kJ/mol). Enolase requires Mg²⁺ to stabilize the carbanion-like transition state during the elimination.

Step 10 — Pyruvate kinase. The second and final substrate-level phosphorylation: PEP transfers its phosphoryl group to ADP, generating ATP and enol-pyruvate, which tautomerizes rapidly (and essentially irreversibly) to the keto form, pyruvate. This tautomerization is what makes the whole step so thermodynamically one-way — the reaction is pulled hard downhill because the immediate product of phosphoryl transfer relaxes into a much more stable form, and that stability can never be recovered by reversing the phosphorylation.

The central quantitative relationship governing this whole stretch of the pathway is the additivity of coupled reactions' standard free energies, since GAPDH and phosphoglycerate kinase operate as a functionally coupled pair.

where:

• $\Delta G^{\circ\prime}_{\text{GAP}\to\text{1,3-BPG}}$ is the standard free energy for oxidative phosphorylation of GAP by NAD⁺ and Pi (≈ +6.3 kJ/mol, slightly unfavorable alone),
• $\Delta G^{\circ\prime}_{\text{1,3-BPG}\to\text{3-PG}+\text{ATP}}$ is the free energy of phosphoryl transfer from 1,3-BPG to ADP (≈ −18.8 kJ/mol, strongly favorable),
• the sum, ≈ −12.5 kJ/mol, describes the net two-step conversion of GAP + NAD⁺ + Pi + ADP → 3-PG + NADH + H⁺ + ATP.

Reading: neither reaction needs to be favorable in isolation. What matters is that the shared intermediate, 1,3-BPG, links an endergonic oxidative-phosphorylation step to an exergonic phosphoryl-transfer step. Because 3-PG and ATP are rapidly consumed downstream, Le Chatelier pull keeps the whole sequence running forward — a recurring strategy in metabolism where a common intermediate makes an otherwise unfavorable step "invisible" thermodynamically. The second governing relationship is the standard free energy of phosphoryl-group hydrolysis itself, which sets the ranking of donors that determines whether a given transfer to ADP is favorable:

where:

• $\Delta G^{\circ\prime}_{\text{hydrolysis, donor}}$ is the standard free energy of hydrolysis of the phosphorylated substrate (1,3-BPG ≈ −49.3 kJ/mol; PEP ≈ −61.9 kJ/mol),
• $\Delta G^{\circ\prime}_{\text{hydrolysis, ATP}}$ is the standard free energy of ATP hydrolysis to ADP + Pi (≈ −30.5 kJ/mol),
• a negative result means transfer of phosphate from donor to ADP is thermodynamically favorable under standard conditions.

Reading: both 1,3-BPG (−49.3 − (−30.5) = −18.8 kJ/mol) and PEP (−61.9 − (−30.5) = −31.4 kJ/mol) sit comfortably "above" ATP on the phosphoryl-transfer-potential ladder, which is precisely why phosphoglycerate kinase and pyruvate kinase can drive ATP synthesis directly, without any membrane-based machinery.

Example 1 — Foundational: complete carbon and energy bookkeeping for the payoff phase. Starting material: 2 mol glyceraldehyde-3-phosphate (from 1 mol glucose). Track ATP, NADH, and pyruvate explicitly, being careful about the factor of two. Per single GAP molecule, steps 6–10 produce: 1 NADH (step 6), 1 ATP (step 7), 1 pyruvate (after step 10, 1 ATP made in step 10 as well). So per GAP: 1 NADH + 2 ATP + 1 pyruvate. For 2 GAP (one glucose):

Now fold in the preparatory phase's investment of 2 ATP (one consumed by hexokinase, one by PFK-1, both per glucose, not per triose):

Limiting-case check: if, hypothetically, only one triose (say DHAP were not isomerized and were simply discarded) proceeded through the payoff phase, net ATP would be $2 - 2 = 0$ and only 1 NADH + 1 pyruvate would form — this correctly shows why triose phosphate isomerase's quantitative conversion of DHAP to GAP is essential to the pathway's real yield; losing half the carbon flux would eliminate the net energy profit entirely. Final tally for one glucose: 2 pyruvate, 2 NADH, and a net gain of 2 ATP (4 gross from substrate-level phosphorylation, minus 2 invested earlier) — the canonical numbers every textbook reports, now traced to their mechanistic origin.

Example 2 — Applied: actual (non-standard) free energy of the GAPDH/PGK coupled step in a resting muscle cell. Standard free energies (ΔG°′) describe 1 M concentrations, but real cells never sit at standard state. Suppose a muscle cell has (approximate physiological concentrations): [GAP] = 2 × 10⁻⁵ M, [NAD⁺] = 2 × 10�10⁻³ Wait — use realistic numbers: [NAD⁺]/[NADH] ≈ 700 (cytosolic, oxidized-favoring), [Pi] = 1 × 10⁻³ M, [ADP] = 1 × 10⁻⁴ M, [ATP] = 3 × 10⁻³ M, [3-PG] = 5 × 10⁻⁵ M. We evaluate the coupled reaction:

with $\Delta G^{\circ\prime} = -12.5$ kJ/mol (from the coupling result above), using the actual-concentration correction:

Take $[\text{NAD}^+]/[\text{NADH}] = 700$ so $[\text{NADH}]/[\text{NAD}^+] = 1/700 \approx 1.43\times10^{-3}$. Compute the reaction quotient:

Numerator: $5\times10^{-5} \times 1.43\times10^{-3} = 7.14\times10^{-8}$; times $3\times10^{-3}$ gives $2.14\times10^{-10}$. Denominator: $2\times10^{-5} \times 1\times10^{-3} = 2\times10^{-8}$; times $1\times10^{-4}$ gives $2\times10^{-12}$.

At T = 310 K, $RT = 8.314 \times 310 = 2577$ J/mol = 2.577 kJ/mol. So:

Reading: even after folding in realistic cellular concentrations that push the reaction quotient well past its standard value, this coupled reaction remains slightly exergonic — consistent with glycolysis running forward under physiological conditions, though this pair of steps operates much closer to equilibrium than the strongly irreversible hexokinase, PFK-1, and pyruvate kinase steps. This is precisely why GAPDH/PGK are not major control points for Regulation of Glycolysis — a near-equilibrium reaction cannot be efficiently regulated by allosteric throttling, since small perturbations relax rapidly in either direction.

Example 3 — Exam/competition level: given only ΔG°′ values, predict the effect of a GAPDH mutation that abolishes the phosphate-binding pocket (forcing water, not Pi, to attack the thioester). Suppose a mutant GAPDH loses its Pi-binding arginine residues, so the thioester intermediate is instead hydrolyzed by water rather than phosphorolyzed by Pi. Predict: (a) the direct product, (b) whether NADH is still made, (c) the ATP yield of glycolysis under this mutation, and (d) the fate of the free energy that would normally have gone into ATP. (a) Water attacking the thioester carbonyl produces 3-phosphoglycerate directly (the phosphate already on C3 of the original GAP skeleton remains) — but no C1 phosphate is installed, so the product is simply the carboxylic acid, i.e., a 3-phospho-glycerate-equivalent lacking the mixed anhydride; effectively, this step now produces 3-phosphoglycerate without ever forming 1,3-BPG. (b) Yes — the hydride transfer to NAD⁺ occurs during formation of the thioester itself, upstream of the phosphate-versus-water branch point; NADH production is mechanistically decoupled from what happens next to the acyl-enzyme intermediate, so 1 NADH per triose is still generated. (c) Because no 1,3-BPG forms, phosphoglycerate kinase has no substrate, and the first substrate-level phosphorylation step is entirely lost. The pathway would still reach 3-phosphoglycerate (just via a different chemical route, hydrolysis rather than phosphorolysis) and could in principle continue through mutase, enolase, and pyruvate kinase to still yield 1 ATP per triose (2 ATP per glucose) — but this offsets only half the priming investment. Net ATP yield per glucose becomes:

(d) The free energy that would have been captured as the acyl-phosphate bond of 1,3-BPG is instead released as heat during ordinary thioester hydrolysis by water — chemically identical in kind to Step 1 of the first mechanism derivation above, where uncoupled aldehyde oxidation simply dissipates its energy. This example demonstrates concretely why the specificity of the phosphate-binding pocket, not just the redox chemistry, is what makes GAPDH an energy-conserving enzyme rather than a simple dehydrogenase — losing that one binding pocket collapses net glycolytic energy yield to zero, even though carbon flow to pyruvate and NADH production continue unimpeded.

Otto Warburg, though best known for oxygen metabolism and the "Warburg effect" in cancer cells, made foundational contributions to identifying glycolytic intermediates and enzymes, including work central to understanding GAPDH's role and NAD-linked oxidations in cellular respiration broadly. Gustav Embden and Otto Meyerhof, whose names are permanently attached to the pathway (the Embden–Meyerhof–Parnas pathway), worked through the 1920s and 1930s to establish the sequence of phosphorylated intermediates from glucose to lactate in muscle extracts, at a time when the very existence of enzyme-bound covalent intermediates was still speculative. Meyerhof's laboratory, and later Carl and Gerty Cori's biochemical dissection of muscle glycogen metabolism, established much of the quantitative stoichiometry — ATP yields, phosphate balances — that this article's bookkeeping relies on. The specific mechanistic resolution of GAPDH's thiohemiacetal/thioester chemistry came later, through mid-20th-century enzymology using group-specific reagents (iodoacetate, which alkylates the catalytic cysteine and abolishes activity) and pre-steady-state kinetic studies that could detect the transient acyl-enzyme intermediate spectroscopically before phosphate addition.