Fates of Pyruvate and Fermentation

Industrially, ethanol fermentation by Saccharomyces cerevisiae is the chemical foundation of brewing, winemaking, and biofuel production, where pyruvate decarboxylase and alcohol dehydrogenase are optimized (through strain selection and increasingly through metabolic engineering) to maximize ethanol titer and tolerance, since ethanol itself becomes toxic to the yeast at higher concentrations by disrupting membrane fluidity. Lactate fermentation by Lactobacillus and related bacteria underlies the entire fermented-dairy and fermented-vegetable industries — yogurt, cheese curd formation, sauerkraut, kimchi — where the acidification from lactate production (dropping pH) both flavors the food and inhibits spoilage organisms, an ancient form of biological food preservation that predates any understanding of the underlying biochemistry by millennia. Clinically, blood lactate concentration is a routine and important diagnostic marker: elevated lactate (lactic acidosis) signals tissue hypoxia in conditions ranging from severe sepsis and cardiogenic shock to simple strenuous exercise, and serial lactate measurements are used in emergency medicine to track the adequacy of tissue perfusion and guide resuscitation. Cancer biology has rediscovered fermentation-adjacent metabolism in the Warburg effect, where many tumor cells preferentially ferment glucose to lactate even in the presence of ample oxygen ("aerobic glycolysis"), a metabolic reprogramming that supports rapid biosynthesis of nucleotides and lipids for proliferation rather than maximal ATP yield, and that is now exploited diagnostically (FDG-PET imaging tracks the elevated glucose uptake this reprogramming causes) and is an active target for anticancer drug development against enzymes like LDH-A. In red blood cells, which lack mitochondria entirely, lactate fermentation is not optional or emergency-only — it is the sole and permanent mode of glucose metabolism, making LDH indispensable to erythrocyte energy supply for the entire ~120-day lifespan of the cell. And pharmaceutically, disulfiram (Antabuse), used to treat alcohol use disorder, works by inhibiting aldehyde dehydrogenase (the enzyme that normally clears acetaldehyde produced from ethanol in the liver), causing toxic acetaldehyde accumulation and unpleasant symptoms upon alcohol consumption — a direct clinical exploitation of the same aldehyde chemistry that governs yeast's own alcohol dehydrogenase step, run in reverse in the human liver.

Ask the tutor: How does the malate-aspartate shuttle let aerobic cells reoxidize cytosolic NADH without ever needing lactate fermentation at all?

Picture the cytoplasm as a small, closed workshop with exactly one pair of tools that must never both be checked out at once: an oxidized hydride acceptor, NAD⁺, and its reduced partner, NADH. The total pool of NAD⁺ + NADH in a cell is minuscule — typically tens to low hundreds of micromolar — compared with the millimolar fluxes of carbon that glycolysis processes every second. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), the payoff-phase enzyme that oxidizes glyceraldehyde-3-phosphate to 1,3-bisphosphoglycerate, consumes one NAD⁺ and produces one NADH for every triose it processes. Two trioses come from each glucose, so glycolysis nets two NADH molecules per glucose that must go somewhere. Under aerobic conditions, that NADH is shuttled into mitochondria (via the malate-aspartate or glycerol-3-phosphate shuttle) and reoxidized by the electron transport chain, ultimately handing its electrons to O₂. But when oxygen is scarce or absent — inside a contracting muscle fiber outrunning its blood supply, inside a yeast cell fermenting grape must, inside an anaerobic bacterium in your gut — that route is closed. The NADH has nowhere to unload its electrons except back onto an organic acceptor already sitting in the cytoplasm: pyruvate itself, the very molecule glycolysis just finished making. This is the logical elegance of fermentation. Pyruvate is glycolysis's terminal product and, simultaneously, its own electron acceptor of last resort. Reducing pyruvate (or a decarboxylated derivative of it) regenerates exactly the NAD⁺ that GAPDH needs, closing the redox loop with no external oxidant required. The reaction is intramolecular in the sense that matters — the electrons removed earlier in glycolysis are returned to a carbon skeleton derived from the same glucose molecule, so the overall process from glucose to lactate, or glucose to ethanol + CO₂, involves zero net change in the oxidation state of carbon summed over the whole pathway. Fermentation product formation is best understood as an oxidation-state-conserving disposal reaction, not a further stage of fuel combustion.

The two dominant fermentation routes differ only in what happens to the carbon skeleton after the electrons are removed, or before they're returned. Lactate fermentation is the minimalist solution: a single enzyme, lactate dehydrogenase, transfers a hydride directly from NADH to the carbonyl carbon of pyruvate, producing lactate in one step, no carbon lost. Ethanol fermentation is a two-step solution used by yeast and many plants: pyruvate is first decarboxylated (losing a CO₂ and shortening the chain by one carbon) to acetaldehyde, and only then is acetaldehyde reduced by NADH to ethanol. Both routes cost the cell one NADH per pyruvate and return one NAD⁺; both are, from an ATP standpoint, complete non-events — no ATP is made or consumed in either fermentation step itself.

The overall stoichiometry of each pathway, starting from glucose, is:

where: lactate fermentation carries no NAD⁺/NADH term because glycolysis's NADH production and lactate dehydrogenase's NADH consumption exactly cancel over the whole pathway; the two ATP are those from substrate-level phosphorylation in the payoff phase only.

where: again NAD⁺/NADH cancels overall, and the two CO₂ come from decarboxylation of the two pyruvate molecules, one per triose branch. The redox-neutrality of fermentation is captured by tracking the NAD⁺/NADH couple explicitly across the two half-reactions that must balance:

where: adding these two reactions cancels NAD⁺ and NADH completely, leaving only the carbon transformation — this cancellation is the entire reason fermentation exists. The driving force for hydride transfer in both LDH and alcohol dehydrogenase (ADH) is set by the standard reduction potentials of the relevant carbonyl/alcohol couples versus the NAD⁺/NADH couple:

where: $n = 2$ electrons transferred as a hydride ion (H⁻, formally one proton plus two electrons delivered as a hydride equivalent), $F = 96{,}485\ \text{C}\,\cdot\,\text{mol}^{-1}$ is the Faraday constant, and $\Delta E^{\circ\prime}$ is the potential difference between the two redox couples at pH 7.

Reading: the first two equations show why fermentation is invisible to the cell's overall carbon-oxidation-state ledger — pyruvate reduction is chemically the mirror image of G3P oxidation, so summing them over the whole glycolytic-fermentative pathway leaves NAD⁺/NADH untouched net. The $\Delta G^{\circ\prime} = -nF\Delta E^{\circ\prime}$ relation is the standard bridge between electrochemical potentials (tabulated for biological half-reactions) and Gibbs free energies, letting us predict that hydride transfer to pyruvate or acetaldehyde is thermodynamically favorable because both carbonyl/alcohol couples have less negative (more positive) standard reduction potentials than the NAD⁺/NADH couple itself, so electrons flow spontaneously from NADH to the organic acceptor.

Example 1 — Foundational: NAD⁺/NADH mass balance across one full turn of glycolysis plus lactate fermentation. Start with one glucose molecule. The payoff phase produces 2 pyruvate, 2 ATP (net, after the investment phase's −2 ATP, so 4 gross − 2 invested = 2 net), and 2 NADH (one per triose at the GAPDH step). Track NAD⁺/NADH explicitly as a running balance rather than assuming it, to confirm the claim that fermentation contributes zero net ATP. Step A: GAPDH reaction, run twice (once per triose):

This consumes 2 NAD⁺ from the free cytosolic pool and deposits 2 NADH into it. Step B: the remainder of the payoff phase (phosphoglycerate kinase, phosphoglycerate mutase, enolase, pyruvate kinase) converts the two 1,3-BPG into two pyruvate, generating the 4 ATP of substrate-level phosphorylation (net 2 ATP once the 2 ATP invested in hexokinase and PFK-1 steps are subtracted). No NAD⁺/NADH appears in these four enzymes. Step C: LDH reaction, run twice (once per pyruvate):

This consumes exactly the 2 NADH generated in Step A and returns exactly 2 NAD⁺ to the free pool. Net NAD⁺ balance: $-2$ (Step A) $+2$ (Step C) $= 0$. The free NAD⁺ pool is completely unchanged after one full turn of glucose → 2 lactate. Net ATP: 2, entirely from the payoff phase's substrate-level phosphorylation, with fermentation itself contributing exactly zero ATP of its own. This confirms numerically what the overview claimed: fermentation's entire functional output, in terms of the cell's energy currency, is the restored NAD⁺, not any additional ATP.

Example 2 — Applied: comparing lactate accumulation in a sprinting muscle fiber over 10 seconds. A fast-twitch muscle fiber during a maximal 10-second sprint effort consumes ATP at roughly $150\ \mu\text{mol}\cdot\text{g}^{-1}\cdot\text{min}^{-1}$ (a typical measured value for human vastus lateralis under maximal voluntary contraction, sustained briefly). Assume nearly all of this ATP demand over the 10-second bolus is met anaerobically via glycolysis (a simplifying but realistic assumption for the first several seconds of maximal effort, before oxidative phosphorylation ramps up). Convert the rate to a 10-second window:

Since glycolysis nets 2 ATP per glucose (via lactate fermentation, anaerobic), the glucose (or glucosyl-unit, since muscle draws on glycogen) consumption is:

Each glucosyl unit fermented to lactate yields 2 lactate molecules, so:

produced in the tissue over that 10-second sprint. This is consistent with measured intramuscular lactate accumulations of roughly 20–30 µmol/g wet weight after brief maximal sprints, confirming the estimate is physiologically sensible — and it explains why blood lactate spikes sharply (often to 10–20 mM, versus a resting ~1 mM) within a minute of an all-out sprint: lactate diffuses out of the fiber into the bloodstream as it accumulates faster than local buffering and clearance (largely via the liver's Cori cycle, converting lactate back to glucose during recovery) can remove it.

Example 3 — Exam/competition level: computing $\Delta G^{\circ\prime}$ for the LDH reaction from tabulated reduction potentials, then checking consistency with the reaction quotient at physiological (non-standard) concentrations. The two relevant half-reactions and their standard reduction potentials (pH 7, 25 °C) are:

For the overall LDH reaction, pyruvate is the electron acceptor (its half-reaction runs as written, as a reduction) and NADH is the electron donor (its half-reaction runs in reverse, as an oxidation), so:

With $n = 2$ and $F = 96{,}485\ \text{C/mol}$:

This matches the commonly cited experimental value of roughly $-25\ \text{kJ/mol}$ for the LDH reaction, confirming the electrochemical route and the direct thermodynamic tabulation agree within normal rounding of literature $E^{\circ\prime}$ values. Now check this against a non-standard, asymmetric physiological scenario, to see whether the reaction remains favorable under realistic intracellular concentrations rather than the artificial 1 M standard state. Take a hypoxic muscle cell with $[\text{pyruvate}] = 0.15\ \text{mM}$, $[\text{lactate}] = 8.5\ \text{mM}$, $[\text{NADH}] = 0.06\ \text{mM}$, $[\text{NAD}^+] = 0.55\ \text{mM}$, and pH 7 (so $[\text{H}^+]$ is folded into the $\Delta G^{\circ\prime}$ term already, standard biochemical convention). The reaction quotient is:

Even under this deliberately unfavorable-looking, asymmetric concentration regime (product-heavy on both lactate and NAD⁺ sides), the reaction remains spontaneous, $\Delta G \approx -10\ \text{kJ/mol}$, confirming that LDH reliably pulls the reaction forward across the wide range of NADH/NAD⁺ ratios a real cell experiences during intense anaerobic work — it does not stall even as lactate and NAD⁺ accumulate substantially above their resting levels.

Otto Fritz Meyerhof and Gustav Embden, working largely independently in Germany in the 1920s and 1930s, established the core enzymatic sequence that we now call glycolysis — sometimes still called the Embden-Meyerhof-Parnas pathway — and, critically, Meyerhof's physiological chemistry laboratory demonstrated the tight, quantitative relationship between lactate production in contracting frog muscle and the anaerobic breakdown of glycogen, work for which he shared the 1922 Nobel Prize in Physiology or Medicine with Archibald Hill (who had independently characterized the heat production and oxygen debt of muscle contraction). Meyerhof showed that lactate appears stoichiometrically from glycogen in oxygen-deprived muscle and disappears again (being resynthesized to glycogen) during recovery in the presence of oxygen — an early, precise experimental anticipation of what would later be formalized as the Cori cycle by Carl and Gerty Cori, who won their own Nobel Prize in 1947 for working out the full glucose–lactate–glucose shuttle between muscle and liver.