Glycolysis

Glycolysis, from the Greek for splitting sugar, is the pathway that breaks one molecule of glucose into two molecules of pyruvate. It takes ten steps, each catalysed by its own enzyme, and it runs in the cytosol of almost every cell of almost every organism on Earth. It needs no oxygen. In its course the cell gains two molecules of ATP and two of NADH for each glucose. That is a small fraction of the energy in glucose, but glycolysis can run fast, and for some cells, such as red blood cells, which have no mitochondria, it is the only source of ATP. It was also the first metabolic pathway to be worked out, between about 1897, when Eduard Buchner showed that a cell-free yeast extract could ferment sugar, and the 1940s; Gustav Embden, Otto Meyerhof and Jakub Parnas made major contributions, and the pathway is often called the Embden–Meyerhof pathway.

Glycolysis falls into two stages. In the first, the preparatory or investment phase of five steps, the cell spends two ATP to phosphorylate glucose and split it into two three-carbon sugar phosphates. In the second, the payoff phase of five steps, each three-carbon unit is oxidized and converted to pyruvate, yielding two ATP and one NADH per unit, so four ATP and two NADH in all. Net, the cell gains two ATP per glucose.

The preparatory phase

Step 1. Hexokinase transfers a phosphate from ATP to the hydroxyl group on carbon 6 of glucose, forming glucose 6-phosphate. The reaction is strongly favourable (ΔG°′ = −16.7 kJ/mol), as Lesson 9 showed, and in the cell it is effectively irreversible. Adding a charged phosphate traps glucose inside the cell, because the glucose transporters do not carry glucose 6-phosphate. In the liver a related enzyme, glucokinase, does the same job; it has a much higher KM for glucose, about 10 mM compared with less than 0.1 mM for hexokinase, so it works mainly when blood glucose is high after a meal, letting the liver take up and store the excess.

Step 2. Phosphoglucose isomerase converts glucose 6-phosphate, an aldose, into fructose 6-phosphate, a ketose. The reaction is readily reversible. Its purpose is to move the carbonyl group to carbon 2, which prepares the molecule to be split symmetrically two steps later.

Step 3. Phosphofructokinase-1 (PFK-1) transfers a second phosphate from ATP to carbon 1, forming fructose 1,6-bisphosphate. This is the committed step of glycolysis: once fructose 1,6-bisphosphate has been made, it has no other use than to continue down the pathway. The reaction is strongly favourable and irreversible in the cell, and PFK-1 is the main point at which the speed of glycolysis is controlled.

Step 4. Aldolase splits fructose 1,6-bisphosphate between carbons 3 and 4 into two three-carbon sugar phosphates: dihydroxyacetone phosphate (DHAP), from carbons 1 to 3, and glyceraldehyde 3-phosphate (GAP), from carbons 4 to 6. The reaction has a large positive ΔG°′ of +23.8 kJ/mol, so under standard conditions it would run backward. In the cell, the concentrations of the products are kept very low, and because one molecule becomes two, the low concentrations favour the forward direction strongly: the actual ΔG is close to zero.

Step 5. Triose phosphate isomerase converts DHAP into GAP. Only GAP can continue down the pathway, so this step ensures that both halves of glucose are used. Triose phosphate isomerase is one of the catalytically perfect enzymes mentioned in Lesson 5, working at the limit set by diffusion. At the end of the preparatory phase, one glucose has become two molecules of GAP, at a cost of two ATP.

The payoff phase

Every step of the payoff phase happens twice for each glucose, once for each GAP.

Step 6. Glyceraldehyde 3-phosphate dehydrogenase oxidizes the aldehyde group of GAP and, at the same time, attaches inorganic phosphate to it, forming 1,3-bisphosphoglycerate. The electrons removed reduce NAD⁺ to NADH. This is the only oxidation in glycolysis, and it is a beautiful example of energy coupling: the energy released by oxidizing the aldehyde is not lost as heat but captured in a new acyl-phosphate bond, which has a very high free energy of hydrolysis (−49.4 kJ/mol). The enzyme forms a covalent intermediate with the substrate through a cysteine in its active site, so that the oxidation and the phosphorylation are linked.

Step 7. Phosphoglycerate kinase transfers the high-energy phosphate of 1,3-bisphosphoglycerate to ADP, forming ATP and 3-phosphoglycerate. Making ATP by transferring phosphate directly from a substrate molecule like this is called substrate-level phosphorylation, to distinguish it from the oxidative phosphorylation of mitochondria. Since this step happens twice per glucose, the two ATP invested in the preparatory phase are now repaid.

Step 8. Phosphoglycerate mutase moves the phosphate from carbon 3 to carbon 2, forming 2-phosphoglycerate.

Step 9. Enolase removes a molecule of water, forming phosphoenolpyruvate (PEP). This small change has a large effect: the phosphate is now attached to an enol, a form that is unstable on its own. When the phosphate is removed, the enol rearranges into a ketone, and that rearrangement releases a great deal of energy. As a result, PEP has the highest free energy of hydrolysis of any common phosphate compound (−61.9 kJ/mol). Fluoride, once added to blood samples to stop glycolysis in the tube before glucose was measured, works by inhibiting enolase.

Step 10. Pyruvate kinase transfers the phosphate of PEP to ADP, forming ATP and pyruvate. The reaction is strongly favourable and irreversible, and it is the third point of control. It too happens twice per glucose, giving the net gain of two ATP.

The overall reaction of glycolysis is:

Control of glycolysis

In a working cell, seven of the ten steps run close to equilibrium, with an actual ΔG near zero, so they can run in either direction depending on concentrations. The other three, catalysed by hexokinase, PFK-1 and pyruvate kinase, are far from equilibrium and essentially one-way. These are the steps the cell regulates, because controlling an irreversible step controls the flow through the whole pathway, much as a dam controls the flow of a river.

PFK-1 is the most important control point. It is an allosteric enzyme with four subunits and sigmoid kinetics, and it reads the energy state of the cell. High ATP inhibits it: ATP binds to a regulatory site as well as the active site, so when the cell has plenty of energy, glycolysis slows. AMP and ADP, which accumulate when ATP is being used up, relieve that inhibition. Citrate, an intermediate of the citric acid cycle, also inhibits PFK-1, signalling that the cell has enough fuel entering its mitochondria. The most powerful activator is fructose 2,6-bisphosphate, made by a separate enzyme under the control of hormones. In the liver, insulin, released after a meal, raises fructose 2,6-bisphosphate and speeds glycolysis, while glucagon, released when blood glucose is low, lowers it and slows glycolysis so that the liver can instead make glucose for the rest of the body.

Hexokinase is inhibited by its own product, glucose 6-phosphate, so if the later steps slow down and glucose 6-phosphate accumulates, glucose uptake slows too. Pyruvate kinase is activated by fructose 1,6-bisphosphate, an example of feed-forward activation, in which an early intermediate prepares a later step for the material on its way. It is inhibited by ATP and alanine, and in the liver it is switched off by phosphorylation when glucagon is present.

The fates of pyruvate

Glycolysis can continue only as long as there is NAD⁺ for step 6. A cell contains only a small amount of NAD⁺, so the NADH made in glycolysis must be reoxidized for the pathway to keep running. What happens next depends on whether oxygen is available.

With oxygen, in cells with mitochondria, pyruvate enters the mitochondria and is converted to acetyl-CoA, which is oxidized completely to carbon dioxide in the citric acid cycle. The NADH passes its electrons, indirectly, to the respiratory chain, which regenerates NAD⁺ and makes much more ATP. This aerobic route, described in Lesson 11, extracts about fifteen times as much ATP from glucose as glycolysis alone.

Without enough oxygen, cells regenerate NAD⁺ by fermentation. In lactic acid fermentation, lactate dehydrogenase uses NADH to reduce pyruvate to lactate. Muscles do this during intense exercise when oxygen supply cannot keep up with demand, and red blood cells do it all the time. The lactate passes into the blood, and the liver converts it back into glucose, a cycle described by Carl and Gerty Cori, who shared the 1947 Nobel Prize in Physiology or Medicine. Lactate also helps make yogurt and sauerkraut, through bacteria that produce it. In alcoholic fermentation, yeast and some bacteria first remove carbon dioxide from pyruvate to form acetaldehyde, then reduce acetaldehyde to ethanol with NADH. The carbon dioxide raises bread, and the ethanol is the alcohol of beer and wine.

For many years the burning sensation and fatigue of hard exercise were blamed on lactic acid. The current view is that lactate itself does no harm and is a valuable fuel; heart muscle and slower muscle fibres take it up and oxidize it. Fatigue has several causes, including the accumulation of phosphate and changes in the handling of calcium and protons within muscle cells.

Other sugars and a famous quirk

Other sugars enter glycolysis at various points. Fructose, from fruit and from sucrose, is phosphorylated in the liver and split into DHAP and glyceraldehyde, bypassing PFK-1, which is one reason large amounts of fructose are metabolized differently from glucose and can promote fat synthesis in the liver. Galactose, from lactose, is converted to glucose 6-phosphate through a short series of reactions; people who lack one of those enzymes have galactosemia and must avoid milk from infancy.

Cancer cells show a quirk first noticed by Otto Warburg in the 1920s: many of them take up glucose avidly and ferment much of it to lactate even when oxygen is plentiful. The Warburg effect is thought to help rapidly dividing cells by supplying building blocks for new molecules as well as energy. It also has a practical use. In PET scanning, patients receive a radioactive glucose analogue, fluorodeoxyglucose, which cells take up and phosphorylate with hexokinase but cannot process further. Tumours, which take up more of it, appear as bright spots on the scan.

A worked example: the energy captured by glycolysis

A program: glycolysis by the numbers

The program tracks carbon atoms, ATP and NADH through the ten steps for one glucose, confirming the balance sheet, and then calculates the actual free energy of the aldolase step at typical concentrations in red blood cells, showing how a reaction with a large positive standard free energy can run forward in the cell.

import math

# step name, ATP change, NADH change (per glucose), copies per glucose
steps = [
    ("hexokinase", -1, 0, 1), ("phosphoglucose isomerase", 0, 0, 1),
    ("phosphofructokinase-1", -1, 0, 1), ("aldolase", 0, 0, 1),
    ("triose phosphate isomerase", 0, 0, 1),
    ("GAP dehydrogenase", 0, 1, 2), ("phosphoglycerate kinase", 1, 0, 2),
    ("phosphoglycerate mutase", 0, 0, 2), ("enolase", 0, 0, 2),
    ("pyruvate kinase", 1, 0, 2),
]
atp = nadh = 0
for name, d_atp, d_nadh, copies in steps:
    atp += d_atp * copies
    nadh += d_nadh * copies
    print(f"{name:28s} running totals: ATP {atp:+d}, NADH {nadh:+d}")
print(f"net per glucose: {atp} ATP, {nadh} NADH, 2 pyruvate (6 carbons in, 2 x 3 out)")

# Aldolase: FBP -> DHAP + GAP, dG0' = +23.8 kJ/mol; concentrations in red blood cells (M)
R, T = 8.314, 310.0
dg0 = 23.8
fbp, dhap, gap = 31e-6, 138e-6, 18.5e-6
q = dhap * gap / fbp
dg = dg0 + R * T * math.log(q) / 1000
print()
print(f"aldolase: Q = {q:.2e} M, dG = {dg0} + RT ln Q = {dg:.1f} kJ/mol")
# The same reaction if all three were at 1 mM
q1 = 1e-3 * 1e-3 / 1e-3
print(f"at 1 mM each: dG = {dg0 + R * T * math.log(q1) / 1000:.1f} kJ/mol")

hexokinase                   running totals: ATP -1, NADH +0
phosphoglucose isomerase     running totals: ATP -1, NADH +0
phosphofructokinase-1        running totals: ATP -2, NADH +0
aldolase                     running totals: ATP -2, NADH +0
triose phosphate isomerase   running totals: ATP -2, NADH +0
GAP dehydrogenase            running totals: ATP -2, NADH +2
phosphoglycerate kinase      running totals: ATP +0, NADH +2
phosphoglycerate mutase      running totals: ATP +0, NADH +2
enolase                      running totals: ATP +0, NADH +2
pyruvate kinase              running totals: ATP +2, NADH +2
net per glucose: 2 ATP, 2 NADH, 2 pyruvate (6 carbons in, 2 x 3 out)

aldolase: Q = 8.24e-05 M, dG = 23.8 + RT ln Q = -0.4 kJ/mol
at 1 mM each: dG = 6.0 kJ/mol

The running totals show the cell in debt by two ATP until step 7, breaking even after step 7 and ending with a net gain of two ATP and two NADH. For aldolase, the low concentrations in red blood cells bring the actual ΔG down from +23.8 kJ/mol to a small negative value, so the reaction runs forward. If all three compounds were at 1 mM, the reaction would still be uphill, at about +6 kJ/mol; it is the very low concentrations of the products, kept low by the steps that follow, that pull it forward.