Glycolysis Pathway: Steps and Where It Occurs

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-09-19

Glycolysis is the ten-enzyme cytosolic pathway that converts one molecule of glucose into two molecules of pyruvate, harvesting a net two ATP and two NADH. Where does glycolysis occur? In the cytosol of every cell, from bacterial cytoplasm to human erythrocytes, never inside the mitochondrial matrix under standard eukaryotic conditions.

That single fact, the location, is the reason glycolysis matters so much. A cell with damaged mitochondria, a cell in a low-oxygen tissue, and a cell that simply needs fast ATP all fall back on the same ten reactions. Red blood cells have no mitochondria at all, so glycolysis is their only source of ATP. Cancer cells frequently shift their energy economy toward glycolysis even when oxygen is present, a pattern tied to the transcription factor HIF-1α and to the PI3K/Akt signaling axis [1][2]. Understanding the pathway means understanding both the chemistry of each step and the logic of why the cell runs it where it does.

Glycolysis Definition and Cellular Location

Glycolysis (from the Greek glykys, sweet, and lysis, splitting) is the anaerobic, oxygen-independent oxidation of glucose to pyruvate. The pathway is also called the Embden-Meyerhof-Parnas pathway, after the three biochemists who mapped it in the early twentieth century. It is nearly universal. The same core sequence operates in bacteria, archaea, and eukaryotes, which is why it is treated as one of the oldest energy-harvesting routes in biology [3].

The location is the cytosol. All ten enzymes are soluble cytosolic proteins, and no membrane separates the substrate from the catalyst. This is a deliberate contrast with oxidative phosphorylation, which requires the inner mitochondrial membrane, and with the TCA cycle, which runs in the mitochondrial matrix.

There are documented exceptions worth knowing because they sharpen the rule. In the stramenopile parasite Blastocystis, the payoff phase of glycolysis is branched, with some enzymes in the cytosol and some in the mitochondrial matrix, and a dedicated mitochondrial carrier shuttles intermediates such as dihydroxyacetone phosphate and glyceraldehyde-3-phosphate across the inner membrane [4]. In most eukaryotes, including humans, glycolysis is strictly cytosolic [4]. Plants run glycolysis in the cytosol and also possess the oxidative pentose phosphate pathway, having lost the alternative Entner-Doudoroff route [5].

Why the Pathway Matters

Glycolysis sits at the intersection of energy production and biosynthesis. It supplies ATP directly, feeds pyruvate into further oxidation, and provides carbon skeletons that branch into the pentose phosphate pathway, serine synthesis, and lipid production. The oxidative pentose phosphate pathway runs in parallel with glycolysis and generates NADPH, which T cells require for cystine reduction and antitumor immunity [6]. That competition for glucose-6-phosphate is one reason glycolytic flux is so tightly regulated.

Clinically, the pathway is a target. In liver disease, mitochondrial dysfunction forces hepatocytes toward glycolysis, and natural products that hit glycolytic enzymes are studied as therapeutic agents [7]. In cancer, the over-expression of glycolytic enzymes and transporters shifts which steps control flux, and metabolic control analysis has identified glucose transport, hexokinase, and hexose-6-phosphate isomerase as the dominant control points in cancer cells, whereas phosphofructokinase-1 dominates in normal cells [8]. That difference is the basis for trying to design drugs that hit tumor glycolysis without crippling healthy tissue [2].

The Two Phases of Glycolysis

The ten steps divide cleanly into two halves. The investment phase (steps 1 through 5) consumes two ATP and splits the six-carbon sugar into two three-carbon molecules. The payoff phase (steps 6 through 10) produces four ATP and two NADH. Subtract the investment and the net is two ATP and two NADH per glucose.

The logic is elegant. The cell spends ATP early to trap glucose inside the cell and to destabilize it enough that the later steps can extract energy. Every reaction in the payoff phase happens twice per glucose, because there are two three-carbon molecules after the split.

Investment Phase: Steps 1 to 5

Step 1: Hexokinase (or glucokinase). Glucose + ATP → glucose-6-phosphate + ADP. This is the first regulated step. Hexokinase phosphorylates glucose, and the added phosphate group traps the sugar inside the cell because phosphorylated sugars cannot cross the plasma membrane. In the liver and pancreatic beta cells, the isoenzyme glucokinase performs the same reaction with a lower affinity for glucose, which lets those cells sense blood glucose rather than simply consume it. Pathway transplantation experiments in Saccharomyces cerevisiae identified glucokinase, phosphofructokinase, and pyruvate kinase as the essential regulatory steps, and showed that expressing them together created imbalances that could only be corrected by lowering glucokinase activity [3]. Hexokinase is also a major control point in cancer glycolysis [8].

Step 2: Phosphoglucose isomerase. Glucose-6-phosphate → fructose-6-phosphate. An isomerization that moves the carbonyl from the aldehyde position to the ketone position, setting up the next phosphorylation.

Step 3: Phosphofructokinase-1 (PFK-1). Fructose-6-phosphate + ATP → fructose-1,6-bisphosphate + ADP. This is the committed step and the rate-limiting step of glycolysis. Once fructose-1,6-bisphosphate forms, the molecule is destined to continue through the pathway. PFK-1 is allosterically inhibited by ATP and by citrate, and activated by AMP and fructose-2,6-bisphosphate. The ATP inhibition is the key feedback logic: when ATP is abundant, the cell has no reason to burn more glucose, so it shuts the gate. Citrate inhibition signals that the TCA cycle is already saturated with carbon. The second ATP consumed in the investment phase is spent here.

Step 4: Aldolase. Fructose-1,6-bisphosphate → dihydroxyacetone phosphate + glyceraldehyde-3-phosphate. The six-carbon sugar is cleaved into two three-carbon triose phosphates. This is the reaction that gives glycolysis its name.

Step 5: Triose phosphate isomerase. Dihydroxyacetone phosphate → glyceraldehyde-3-phosphate. Only glyceraldehyde-3-phosphate can continue, so this isomerase converts the other product of aldolase into the same molecule. After this step, there are two molecules of glyceraldehyde-3-phosphate, and every subsequent reaction runs twice.

Payoff Phase: Steps 6 to 10

Step 6: Glyceraldehyde-3-phosphate dehydrogenase. Glyceraldehyde-3-phosphate + NAD+ + Pi → 1,3-bisphosphoglycerate + NADH + H+. This is the only redox reaction in glycolysis and the source of both NADH molecules. The enzyme uses inorganic phosphate, not ATP, to add the high-energy acyl phosphate. Because two trioses flow through, two NADH are produced per glucose.

Step 7: Phosphoglycerate kinase. 1,3-bisphosphoglycerate + ADP → 3-phosphoglycerate + ATP. Substrate-level phosphorylation. The high-energy phosphate from 1,3-bisphosphoglycerate is transferred directly to ADP, making ATP without any membrane or oxygen. Two ATP per glucose.

Step 8: Phosphoglycerate mutase. 3-phosphoglycerate → 2-phosphoglycerate. A relocation of the phosphate group from carbon 3 to carbon 2.

Step 9: Enolase. 2-phosphoglycerate → phosphoenolpyruvate + H2O. A dehydration that creates the highest-energy phosphate bond in the pathway.

Step 10: Pyruvate kinase. Phosphoenolpyruvate + ADP → pyruvate + ATP. The second substrate-level phosphorylation and the third regulated enzyme. Pyruvate kinase is activated by fructose-1,6-bisphosphate (feed-forward activation) and inhibited by ATP and alanine. The product, pyruvate, sits at a branch point: it can be reduced to lactate, converted to acetyl-CoA, or used for gluconeogenesis in tissues that perform it.

The three regulated enzymes, hexokinase, PFK-1, and pyruvate kinase, are the same three steps identified as essential in pathway transplantation experiments, which confirms that the classical textbook regulation map holds up under modern genetic tests [3]. The pathway is also subject to substrate cycling, in which kinases and phosphatases run in opposite directions at the same time. In sperm cells, where this was measured carefully, the rates of hexokinase and phosphofructokinase at low flux greatly exceeded the overall pathway flux, meaning the classical picture of a single rate-determining step is an oversimplification at low metabolic rates [9].

Step Table

Steps 6 through 10 run twice per glucose, so the per-step yields shown are the totals for the whole pathway.

flowchart TD
    A[Glucose enters cytosol] --> B[Hexokinase traps glucose]
    B --> C[Isomerase and PFK-1 commit the sugar]
    C --> D[Aldolase splits into two trioses]
    D --> E[G3P dehydrogenase makes NADH]
    E --> F[Phosphoglycerate kinase makes ATP]
    F --> G[Pyruvate kinase makes ATP]
    G --> H[Pyruvate exits to lactate or mitochondria]
    C --> I{ATP and citrate high}
    I --> J[PFK-1 inhibited]
    J --> C

How Glycolysis Is Measured in the Laboratory

Three methods dominate bench practice. The first is the coupled spectrophotometric assay, in which the production of NADH at step 6 is followed at 340 nm. Because NADH absorbs light at 340 nm and NAD+ does not, the rate of absorbance increase is a direct readout of glycolytic flux through glyceraldehyde-3-phosphate dehydrogenase. This is the workhorse assay for enzyme kinetics.

The second is extracellular flux analysis, which measures the extracellular acidification rate (ECAR) as a proxy for lactate production and therefore glycolytic rate, paired with oxygen consumption rate (OCR) for mitochondrial respiration. In SH-SY5Y neuroblastoma cells, ECAR and OCR show a reciprocal relationship under fixed energy demand. Raising glucose raises ECAR and lowers OCR (the Crabtree effect), and restricting glycolysis by glucose deprivation or by substituting galactose for glucose lowers ECAR and raises OCR [10]. That reciprocal behavior is why ECAR and OCR are always reported together.

The third is isotope tracing. Investigators feed cells specifically labeled glucose or fructose and follow the label into downstream metabolites. This approach was used to quantify substrate cycling in sperm, where the crossover plots of metabolites revealed that kinase rates exceeded net flux at low metabolic rates [9]. The same logic underlies nutrient-sensitized genome-wide screens, which showed that the ribose moiety of uridine or RNA can be salvaged through the non-oxidative pentose phosphate pathway into fructose-6-phosphate and glyceraldehyde-3-phosphate, entering glycolysis downstream of the regulated hexokinase step and supporting ATP production when glucose is absent [11].

Clinical and Comparative Relevance

Red blood cells rely entirely on glycolysis. Mature erythrocytes eject their nucleus and mitochondria during differentiation, so oxidative phosphorylation is unavailable. They generate ATP exclusively through substrate-level phosphorylation, and they reduce pyruvate to lactate to regenerate NAD+ so that step 6 can continue. This is why erythrocytes are the classic tissue for illustrating anaerobic glycolysis, and why inherited defects in glycolytic enzymes cause hemolytic anemia.

Cancer cells frequently increase glycolytic flux. The Warburg effect describes the conversion of pyruvate to lactate rather than its oxidation in mitochondria, even when oxygen is available [12]. This shift is driven in part by HIF-1α downstream of PI3K/Akt signaling, and it supports biosynthesis as well as ATP production because glycolytic intermediates feed nucleotide, amino acid, and lipid synthesis [1][2]. Metabolic control analysis of cancer glycolysis found that the steps with the highest control differ from normal cells, with glucose transport, hexokinase, and hexose-6-phosphate isomerase dominant in tumors and PFK-1 dominant in normal tissue [8]. That difference is the therapeutic opening.

Glycolysis also matters in non-neoplastic disease. In calcific aortic valve disease, valve endothelial cells are more glycolytic at baseline than valve interstitial cells, and blunting glucose metabolism reduces inflammatory signaling and monocyte adhesion [13]. In liver disease, mitochondrial dysfunction forces a shift toward glycolytic ATP production, and natural products that target glycolytic enzymes are under study [7]. In adenomyosis with heavy menstrual bleeding, expression of HIF-1α, GLUT1, HK2, PFKFB3, and PKM2 is reduced in patients with excessive bleeding and greater fibrosis [14].

Comparative biology adds caveats. Methanogens store glycogen despite being unable to grow on sugars, and energy charge appears to signal the switch between storage and glycolytic use of that polymer [15]. In Drosophila larvae, knocking down late glycolytic enzymes in the fat body causes atrophy, downregulates mTOR, and triggers distant muscle disorganization through TNF-α and ImpL2 signaling, showing that glycolytic status in one tissue can be communicated systemically [16]. Astrocytes illustrate the same principle in the mammalian brain: their fine processes are too narrow to hold mitochondria, so they depend on glycolysis and glycogenolysis for local ATP [17].

Common Mistakes and Limitations

The most frequent error is placing glycolysis in the mitochondrion. It is cytosolic. The TCA cycle and oxidative phosphorylation are mitochondrial, and students often merge the three into one compartment. Keep them separate in your mental model.

The second error is forgetting that steps 6 through 10 run twice. Students correctly list four ATP produced and two ATP consumed but then mislabel the per-step yield. The four ATP come as two ATP from phosphoglycerate kinase and two from pyruvate kinase.

The third error is confusing the two substrate-level phosphorylation steps with oxidative phosphorylation. Substrate-level phosphorylation transfers a phosphate directly from a high-energy metabolite to ADP. It needs no membrane, no proton gradient, and no oxygen.

The fourth error is treating PFK-1 as the only regulated step that matters. It is the rate-limiting step in most normal tissues, and it is inhibited by ATP and citrate, but hexokinase and pyruvate kinase are also regulated, and in cancer cells the control distribution shifts toward glucose transport and hexokinase [8]. In sperm at low flux, kinase rates far exceed net flux, which shows that regulation is not a single bottleneck [9].

The fifth error is assuming glycolysis always ends in lactate. It ends in pyruvate. Whether pyruvate becomes lactate, acetyl-CoA, oxaloacetate, or alanine depends on the tissue, the oxygen supply, and the cell's biosynthetic needs.

A limitation worth stating plainly: enzyme kinetics measured in a cuvette or a cultured cell line do not automatically predict flux in a whole tissue. Substrate cycling, compartmentalized isoenzymes, and transporter availability all modify the real rate. Individual clinical or veterinary cases require professional evaluation.

Quick Review

1. Glycolysis occurs in the cytosol of all cells, not in mitochondria.
2. Ten steps split into an investment phase (steps 1 to 5, uses 2 ATP) and a payoff phase (steps 6 to 10, makes 4 ATP and 2 NADH).
3. Net yield per glucose is 2 ATP and 2 NADH.
4. The three regulated enzymes are hexokinase, phosphofructokinase-1, and pyruvate kinase.
5. PFK-1 is the rate-limiting step and is inhibited by ATP and citrate.
6. Steps 6 through 10 run twice per glucose because aldolase and triose phosphate isomerase produce two trioses.
7. Red blood cells depend entirely on glycolysis because they lack mitochondria.

Frequently Asked Questions

Where does glycolysis occur in the cell?

Glycolysis occurs in the cytosol, the fluid compartment of the cytoplasm. All ten enzymes are soluble cytosolic proteins, and no mitochondrial membrane is involved.

Does glycolysis require oxygen?

No. Glycolysis is anaerobic and produces ATP by substrate-level phosphorylation, which needs no oxygen and no electron transport chain.

What is the net ATP yield of glycolysis?

The net yield is 2 ATP per glucose. Four ATP are produced in the payoff phase and two are consumed in the investment phase.

Which enzyme is the rate-limiting step of glycolysis?

Phosphofructokinase-1 is the rate-limiting step. It catalyzes step 3 and is allosterically inhibited by ATP and citrate.

Why do red blood cells rely entirely on glycolysis?

Mature red blood cells lack mitochondria, so oxidative phosphorylation is unavailable. They make ATP only through glycolysis and reduce pyruvate to lactate to regenerate NAD+.

How many NADH molecules does glycolysis produce?

Two NADH per glucose. Both come from step 6, glyceraldehyde-3-phosphate dehydrogenase, which runs once for each of the two triose molecules.

Related Articles

• Autophagy Pathway: Steps, Mechanism, and Regulation
• Apoptosis Pathway: Key Steps and Mechanisms
• Transcription Occurs in the Nucleus: Why and How
• MAP Kinase Pathway: Steps, Components, and Regulation
• DNA Replication Occurs in S Phase: Mechanisms and Evidence
• Hamster Adoption: Where and How to Adopt a Hamster
• Where Does Transcription Occur? Cell Locations
• Clotting Cascade: Coagulation Pathway Steps
• Pentose Phosphate Pathway: Steps and Functions

Sources

1. Significance of flavonoids targeting PI3K/Akt/HIF-1α signaling pathway in therapy-resistant cancer cells - A potential contribution to the predictive, preventive, and personalized medicine.
2. Tumor glycolysis as a target for cancer therapy: progress and prospects.
3. Synergistic regulatory mechanisms in glycolysis revealed by pathway transplantation.
4. A mitochondrial carrier transports glycolytic intermediates to link cytosolic and mitochondrial glycolysis in the human gut parasite Blastocystis.
5. Plastid ancestors lacked a complete Entner-Doudoroff pathway, limiting plants to glycolysis and the pentose phosphate pathway.
6. Oxidative pentose phosphate pathway is required for T cell activation and antitumor immunity.
7. Natural products target glycolysis in liver disease.
8. Metabolic control analysis as a strategy to identify therapeutic targets, the case of cancer glycolysis.
9. The effect of substrate cycling on the ATP yield of sperm glycolysis.
10. Glycolysis-respiration relationships in a neuroblastoma cell line.
11. Salvage of ribose from uridine or RNA supports glycolysis in nutrient-limited conditions.
12. Potential roles of plant metabolites and Traditional Chinese Medicine formulas in regulating glycolysis-OXPHOS plasticity in gastric precancerous lesions and gastric cancer: a critical appraisal of the evidence.
13. Glycolysis and hexosamine biosynthesis regulate inflammatory protein expression and maturation and key steps of monocyte recruitment by human aortic valve cells.
14. Reduced endometrial glycolysis concomitant with increased lesional fibrosis in patients with adenomyosis who complained of heavy menstrual bleeding.
15. Glycogen metabolism in methanogens: A key pathway for metabolic response to nutrient availability.
16. Fat body glycolysis defects inhibit mTOR and promote distant muscle disorganization through TNF-α/egr and ImpL2 signaling in Drosophila larvae.
17. Energy metabolism in astrocytes: high rate of oxidative metabolism and spatiotemporal dependence on glycolysis/glycogenolysis.