Glycolysis

Glycolysis is the highly conserved metabolic pathway that converts one molecule of glucose (six carbons) into two molecules of pyruvate (three carbons) in the cytosol, yielding a net gain of two ATP and two NADH per glucose while operating in virtually every human cell ^[17]D5. Also known as the Embden-Meyerhof-Parnas (EMP) pathway, it is the central catabolic route for glucose and is the only energy-yielding pathway available to erythrocyte s, which lack mitochondria. The pathway proceeds in ten enzymatic steps organized into two phases: a preparative (energy-investment) phase that consumes 2 ATP, and a payoff (energy-generation) phase that produces 4 ATP and 2 NADH.

Beyond ATP production, glycolysis functions as a critical signaling hub ^[17]D5. Metabolic intermediates feed into biosynthetic pathways: glucose-6-phosphate enters the pentose phosphate pathway for NADPH and ribose synthesis ^[3]D5; 3-phosphoglycerate contributes to serine and glycine synthesis; and pyruvate serves as a precursor for oxaloacetate (anaplerosis) and lactate . Lactate, long dismissed as a waste product, is now recognized as a major circulating fuel, a gluconeogenic precursor, and a signaling molecule that regulates gene expression via lysine lactylation ^{[2]D5[10]D5[22]D5}. Glycolytic flux also controls cell fate decisions in immune cells ^[20]D5, stem cell s ^[12]D5, and cancer cells ^[21]D5. In the brain, aerobic glycolysis accounts for 10-12% of glucose use and supports synaptic growth during development ^[9]A1a. During hypoxia , cells shift to anaerobic glycolysis to sustain ATP production, a program orchestrated by hypoxia-inducible factor (HIF) ^[11]D5[18]D5.

Clinical significance

Disruptions in glycolysis underlie a spectrum of human diseases. Inborn errors of glycolytic enzymes cause hemolytic anaemia ( pyruvate kinase deficiency ), exercise-induced rhabdomyolysis (phosphofructokinase deficiency, aldolase A deficiency ^[5]D5), and congenital disorders of glycosylation (PGM1 deficiency, treatable with galactose|d-galactose ^[1]A1c). In cancer, the Warburg effect , aerobic glycolysis even in the presence of oxygen, supports biosynthesis and tumour growth ^[21]D5. In ischemia and hypoxia , excessive lactate accumulation can cause lactic acidosis ^[18]D5. The pathway is also a target for therapy; for example, inhibition of glycolysis is being explored in cancer and inflammatory diseases ^[30]D5[31]D5, while the metabolite-repair systems that safeguard glycolysis are an emerging area of clinical investigation ^[7]D5.

The molecular structures of the key intermediates and enzymes are described in the next section.

Pearl: Glycolysis is not a single linear pathway but a hub supplying ATP, biosynthetic precursors, and redox equivalents; its regulation by PFK-1 and pyruvate kinase dictates metabolic flux in health and disease.

Building on this framework, the molecular machinery of glycolysis comprises small-molecule intermediates and protein catalysts whose structures dictate pathway flux, regulation, and integration with cellular energy status.

Glucose and its phosphorylation

D-Glucose circulates predominantly as the six-membered pyranose ring (α- and β-anomers), which interconvert spontaneously via mutarotation. Entry into the glycolytic path requires phosphorylation at C6 by hexokinase or glucokinase, producing glucose-6-phosphate. This step traps glucose inside the cell and forces the ring to open transiently, a structural prerequisite for the subsequent isomerization to fructose-6-phosphate. The conformational switch from cyclic to linear geometry is a recurring theme: the enzyme active sites are built to recognize the open-chain form of their sugar substrates.

Key intermediates and their chemical logic

Glycolysis proceeds through a series of phosphorylated intermediates, each with a distinct functional group that determines its reactivity:

• Glucose-6-phosphate - a stable phosphate ester; commits glucose to the pathway.
• Fructose-1,6-bisphosphate (F1,6BP) - two phosphate groups (C1 and C6) create a molecule primed for scission into two trioses. F1,6BP is also a potent allosteric activator of pyruvate kinase.
• 1,3-Bisphosphoglycerate (1,3-BPG) - a mixed anhydride between carboxylate and phosphate; its high ΔG° of hydrolysis drives the first substrate-level ATP synthesis.
• Phosphoenolpyruvate (PEP) - an enol phosphate with extremely high phosphoryl-transfer potential (ΔG°' = -61.9 kJ/mol), powering the second ATP-generating step.
• Pyruvate - the final α-ketoacid product; its fate (oxidation in mitochondria or reduction to lactate) depends on oxygen availability.

Cofactors and metal ions

• ATP/ADP - the universal energy currency; ATP not only donates phosphate but also chelates Mg²⁺, forming the true substrate for all glycolytic kinases.
• NAD⁺/NADH - NAD⁺ accepts a hydride ion in the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) reaction. Maintaining the cytosolic NAD⁺/NADH ratio (≈700:1 in well-oxygenated cells) is essential; lactate dehydrogenase regenerates NAD⁺ under anaerobic conditions.
• Mg²⁺ - required for ATP binding and catalytic activity of kinases, enolase, and other Mg²⁺-dependent enzymes.

Enzyme catalytic domains

Despite catalysing very different reactions, glycolytic enzymes share modular folds that recur across metabolism:

• Kinases (hexokinase, PFK-1, pyruvate kinase) use an ATP-grasp or related fold. Hexokinase undergoes large domain closure upon glucose binding, excluding water to prevent ATP hydrolysis.
• Isomerase (glucose-6-phosphate isomerase) employs an enediol intermediate to interconvert aldose and ketose.
• Dehydrogenase (GAPDH) contains a catalytic cysteine that forms a thiohemiacetal with the substrate; subsequent oxidation reduces NAD⁺.
• Mutase - phosphoglycerate mutase (dPGM) belongs to the histidine phosphatase superfamily ^[38]D5. The catalytic histidine becomes transiently phosphorylated during the transfer of the phosphate group from C3 to C2. This superfamily also includes fructose-2,6-bisphosphatase and TIGAR (TP53-induced glycolysis and apoptosis regulator), linking glycolysis to tumour suppression and apoptosis ^[38]D5.
• Enolase - dehydrates 2-phosphoglycerate to PEP using two Mg²⁺ ions that stabilise the enolate intermediate.

PFKFB2 as a bifunctional regulator

A pivotal structural component is the bifunctional enzyme phosphofructokinase-2/fructose-2,6-bisphosphatase (PFKFB or PFK-2), which both produces and degrades fructose-2,6-bisphosphate (Fru-2,6-P₂) ^[37]D5. Fru-2,6-P₂ is the most potent allosteric activator of PFK-1 (phosphofructokinase-1), the committed step of glycolysis. PFKFB2, the cardiac isoform, integrates hormonal and metabolic signals through multi-site phosphorylation by AMPK, PKA, and other kinases. Its loss drives pathophysiology in metabolic heart disease, but PFKFB2 is also expressed in multiple extra-cardiac tissues and is implicated in cancer ^[37]D5. The dual catalytic domain organisation (kinase domain + phosphatase domain) allows fine bidirectional control of Fru-2,6-P₂ levels.

Pyruvate entry into mitochondria

The journey of glycolytic pyruvate ends at the mitochondrial pyruvate carrier (MPC), a heterodimer of MPC1 and MPC2 (or the alternative MPC1L/MPC2) from the SLC54 family ^[39]D5. Recent cryo-electron microscopy structures have captured the carrier in outward-open, occluded, and inward-open conformations, revealing an alternating access mechanism. The substrate binding site lies at the dimer interface. Three chemically distinct inhibitor classes (represented by UK-5099, zaprinast, and 7ACC2) all occupy the same binding pocket in the outward-open state, blocking transport ^[39]D5. These structural insights are now being leveraged to develop drugs for diabetes, neurodegeneration, metabolic dysfunction-associated steatotic liver disease (MASLD), and certain cancers ^[39]D5.

Pearl: The structural versatility of glycolytic components, from the induced-fit closure of hexokinase and the phosphorylated histidine of dPGM to the alternating access of the MPC, is the molecular basis for the pathway's tight regulation and its ability to integrate with cellular energy, redox, and growth signals.

From the phosphorylated glucose molecule, the glycolytic pathway proceeds through ten ordered reactions, each catalyzed by a specific enzyme that drives the sequential conversion of a six-carbon sugar into two three-carbon pyruvate molecules ^[17]D5. The pathway is organized into two phases. The energy investment phase (steps 1-5) consumes 2 ATP to phosphorylate the intermediates, while the energy payoff phase (steps 6-10) generates 4 ATP and 2 NADH, yielding a net gain of 2 ATP and 2 NADH per glucose. A mermaid flowchart visualizes the stepwise transformation:

The Energy Investment Phase (Steps 1-5)

Step 1: Hexokinase (HK) or Glucokinase (GCK) phosphorylates glucose to glucose-6-phosphate (G6P), committing glucose to intracellular metabolism. In most tissues, hexokinase (Km ~0.1 mM) binds glucose with high affinity and is inhibited by G6P. In pancreatic β‑cells and hepatocytes, glucokinase (hexokinase IV, Km ~8 mM) functions as a glucose sensor; its unique kinetic properties allow flux to vary proportionally with blood glucose, coupling glycolytic rate to insulin secretion ^[46]D5. The reaction consumes 1 ATP and is irreversible.

Step 2: Phosphoglucose isomerase (GPI) reversibly isomerizes G6P to fructose-6-phosphate (F6P), moving the carbonyl from carbon 1 to carbon 2 to prepare for the next phosphorylation.

Step 3: Phosphofructokinase-1 (PFK-1) phosphorylates F6P to fructose-1,6-bisphosphate (F1,6BP), consuming a second ATP. This is the first committed, rate-limiting step of glycolysis and the primary regulatory node. PFK-1 is allosterically activated by fructose-2,6-bisphosphate (F2,6BP) and inhibited by ATP and citrate ^[17]D5[54]D5. The human liver PFK structure, recently solved by cryo‑EM, revealed a tetrameric arrangement that mediates cooperative ATP binding ^[54]D5. A second PFK enzyme, PPi-dependent PFK, is found in some organisms but not in mammalian glycolysis ^[54]D5.

Step 4: Aldolase (ALDO) cleaves F1,6BP into two triose phosphates: dihydroxyacetone phosphate (DHAP) and glyceraldehyde-3-phosphate (G3P). Three aldolase isoforms exist; aldolase A is the predominant muscle isoform, and its deficiency causes exercise‑induced rhabdomyolysis ^[5]D5.

Step 5: Triosephosphate isomerase (TPI) reversibly converts DHAP to G3P, ensuring both trioses enter the payoff phase. TPI is catalytically perfect (diffusion‑limited), yet it also produces a damaging side‑product: methylglyoxal from the spontaneous elimination of phosphate from DHAP. The glyoxalase system (GLO1, GLO2) detoxifies methylglyoxal; TPI deficiency, a rare inborn error, causes severe hemolytic anemia and neurodegeneration due to methylglyoxal accumulation ^[57]D5.

The Energy Payoff Phase (Steps 6-10)

Step 6: Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) oxidizes G3P to 1,3-bisphosphoglycerate (1,3-BPG) while reducing NAD⁺ to NADH. The thioester intermediate generates a high‑energy acyl‑phosphate bond. The NADH must be re‑oxidized to NAD⁺ to sustain flux; under aerobic conditions the malate‑aspartate shuttle or glycerol‑3‑phosphate shuttle transfers electrons into mitochondria ^[40]D5, while under hypoxia NADH is re‑oxidized by lactate dehydrogenase.

Step 7: Phosphoglycerate kinase (PGK) transfers the phosphate from 1,3-BPG to ADP, producing 3‑phosphoglycerate (3‑PG) and the first ATP. This substrate-level phosphorylation is reversible and generates 2 ATP per glucose (one from each triose).

Step 8: Phosphoglycerate mutase (PGM) isomerizes 3‑PG to 2‑phosphoglycerate (2‑PG) via a 2,3‑bisphosphoglycerate intermediate.

Step 9: Enolase (ENO) dehydrates 2‑PG to phosphoenolpyruvate (PEP), creating another high‑energy enol‑phosphate bond. Enolase requires Mg²⁺; fluoride inhibition of enolase underlies the clinical use of fluoride as a glycolysis blocker in blood collection tubes.

Step 10: Pyruvate kinase (PK) transfers the phosphate from PEP to ADP, producing pyruvate and the second ATP (2 ATP per glucose). PK is the third irreversible step and is allosterically activated by F1,6BP (feed‑forward) and inhibited by ATP and alanine. Four PK isoforms exist: PKL (liver), PKR (erythrocytes), PKM1 (muscle, brain), and PKM2 (embryonic and cancer cells). PKM2 can switch between a highly active tetramer and a less active dimer, diverting glycolytic intermediates into biosynthetic branches ^[17]D5.

Key Enzymes and Their Unique Roles

The PFKFB family (PFKFB1-4) synthesizes and degrades F2,6BP, the potent allosteric activator of PFK‑1. PFKFB2 (cardiac isoform) is regulated by multi‑site phosphorylation and is critical for cardiac stress responses ^[37]D5. PFKFB3 is highly expressed in proliferating cells and is short‑lived due to ubiquitin‑mediated degradation by APC/C‑Cdh1; its release in mid‑to‑late G1 phase drives a glycolytic burst that supports cell cycle progression ^[52]D5. Neurons constitutively degrade PFKFB3 via APC/C‑Cdh1, directing glucose toward the pentose phosphate pathway for antioxidant defense instead of glycolysis ^[55]D5.

Branch Points and Side Reactions

Glucose‑6‑phosphate stands at a major metabolic fork. It can enter the pentose phosphate pathway to generate NADPH and ribose‑5‑phosphate for nucleotide synthesis ^[3]D5. This branch is especially active in erythrocytes, where NADPH maintains reduced glutathione, and in cancer cells supporting anabolic growth ^[28]D5.

Dihydroxyacetone phosphate (DHAP) can be reduced to glycerol‑3‑phosphate for triglyceride synthesis or, as noted, spontaneously eliminate phosphate to form the toxic metabolite methylglyoxal. The glyoxalase system converts methylglyoxal to D‑lactate, protecting the proteome from glycation damage ^[57]D5[7]D5.

Phosphoenolpyruvate (PEP) can be diverted to gluconeogenesis (via PEP carboxykinase) in the liver, but in glycolysis it funnels exclusively toward pyruvate.

Metabolic Fate of Pyruvate

Pyruvate itself is a junction. Under aerobic conditions it enters mitochondria, where pyruvate dehydrogenase converts it to acetyl‑CoA for the TCA cycle. Under anaerobic conditions (or in erythrocytes, which lack mitochondria) lactate dehydrogenase reduces pyruvate to lactate, regenerating NAD⁺ to sustain glycolysis ^[17]D5. In cancer cells, even in the presence of oxygen, pyruvate is preferentially fermented to lactate (the Warburg effect), a phenomenon that supports rapid proliferation through biosynthetic intermediates ^[42]D5.

Pearl: The PFK-1 step is not only rate-limiting but the metabolic commitment point; once fructose-1,6-bisphosphate forms, the cell must complete glycolysis, the carbon cannot exit the pathway except through distal branches, and the net ATP yield depends on efficient regeneration of NAD⁺ via lactate or mitochondrial shuttles.

From the stepwise enzymatic sequence, attention now turns to the mechanisms that tune glycolytic flux in response to cellular energy status, oxygen availability, hormonal signals, and cell-cycle demands. The pathway is not constitutively active; its rate is set at three irreversible steps, hexokinase/glucokinase, phosphofructokinase-1 (PFK-1), and pyruvate kinase, where allosteric, hormonal, transcriptional, and post-translational controls converge.

Allosteric and Hormonal Regulation

The committed step, PFK-1, is the primary flux-determining enzyme. ATP and citrate inhibit PFK-1; AMP and fructose 2,6‑bisphosphate activate it ^[54]D5. Fructose 2,6‑bisphosphate is itself the product of a bifunctional enzyme, PFK-2 (PFKFB), which is subject to hormonal control. In liver, glucagon via cAMP inactivates PFK-2, lowering fructose 2,6‑bisphosphate and slowing glycolysis, while insulin has the opposite effect ^[37]D5. The cardiac isoenzyme PFKFB2 undergoes multi-site phosphorylation that integrates adrenergic, metabolic, and ischemic stress signals to match glucose use to workload ^[37]D5. Hexokinase is inhibited by its product glucose 6‑phosphate; the liver/pancreatic isoenzyme glucokinase, which lacks this inhibition, has a higher Km and serves as a glucose sensor. In β‑cells, glucokinase couples blood glucose to insulin secretion as part of a glycolytic flux regulatory unit that includes the two fructose ester steps ^[46]D5. Pyruvate kinase is activated by fructose 1,6‑bisphosphate (feed‑forward) and inhibited by ATP and alanine.

Transcriptional Regulation

Hypoxia-inducible factor 1α (HIF‑1α) is the master transcriptional driver of glycolysis under low oxygen. Stabilized by hypoxia, HIF‑1α upregulates GLUT1, hexokinase 2, PFKFB3, lactate dehydrogenase A, and MCT4, shifting ATP production away from oxidative phosphorylation ^[18]D5. Even under normoxia, this program can be engaged by oncogenic signaling. c‑Myc directly induces many of the same glycolytic genes, while p53 opposes glycolysis through targets such as TIGAR and by repressing PFKFB3; loss of p53 thus amplifies the Warburg effect ^[58]D5[65]D5. NF‑κB also promotes a glycolytic switch, acting in crosstalk with p53 to regulate mitochondrial and nuclear gene expression ^[68]D5. The coactivators YAP/TAZ drive a TEAD-mediated transcriptional program that includes glycolytic enzymes, coupling nutrient availability to cell growth ^[64]D5.

Post‑translational and Cell‑Cycle Control

Glycolysis is also gated by the cell cycle. The E3 ubiquitin ligase APC/C‑Cdh1 targets PFKFB3 for proteasomal degradation, keeping glycolytic flux low in quiescent cells. In mid‑to‑late G₁, APC/C‑Cdh1 activity falls, PFKFB3 accumulates, and a burst of glycolysis precedes S‑phase entry ^[52]D5. This surge is transient: at the onset of S‑phase, PFKFB3 is degraded by a second ubiquitin ligase, SCF‑β‑TrCP ^[52]D5. In neurons, APC/C‑Cdh1 constitutively degrades PFKFB3, explaining why these cells cannot upregulate glycolysis and instead shunt glucose into the pentose phosphate pathway for antioxidant defense ^[55]D5. In the heart and other tissues, PFK‑1 itself is allosterically regulated; the structural basis of human liver PFK allostery has been resolved by cryo‑electron microscopy, revealing how ATP and fructose 2,6‑bisphosphate induce distinct conformational states ^[54]D5. Broader post‑translational modifications, including acetylation, succinylation, and malonylation, modulate mitochondrial enzymes and indirectly affect glycolytic flux ^[27]D5.

Metabolite Repair and Damage Control

Enzyme promiscuity and spontaneous reactions generate abnormal metabolites that can poison glycolysis. Dedicated metabolite repair enzymes reverse or prevent such damage, defining a growing class of inborn errors of metabolism ^[7]D5. Preservation of metabolic integrity is therefore an underappreciated layer of regulation.

Tissue‑Specific and Disease Contexts

Brain metabolism relies on aerobic glycolysis during development, when synaptic growth rates peak ^[9]A1a; in Alzheimer disease, region‑specific declines in glucose metabolism precede clinical symptoms ^[4]D5. The healthy heart derives only 5 % of ATP from glycolysis but switches to greater glucose use under ischemia or pressure overload, a shift critically dependent on PFKFB2 ^[27]D5[37]D5. Immune cells exhibit striking metabolic plasticity: activated T lymphocytes engage aerobic glycolysis (the Warburg effect) but concurrently run oxidative phosphorylation, allocating diverse nutrient sources to support proliferation and effector function ^[66]D5. Tumor‑associated macrophages similarly polarize with distinct glycolytic or oxidative profiles that shape the tumor microenvironment ^[23]D5. Lactate, long dismissed as a waste product, is now recognized as a signaling molecule that is exported via MCT4, taken up by neighboring cells, and can modify histones and proteins via lysine lactylation, thereby coupling glycolytic flux to epigenetic and immune regulation ^{[10]D5[30]D5[15]D5[59]D5}. The aryl hydrocarbon receptor integrates metabolic cues from diet, microbiome, and glycolysis to modulate immune responses ^[62]D5. Cancer cells exploit these regulatory layers: oncogenic mutations (PI3K/Akt/mTOR, Myc, HIF‑1) lock glycolysis in the “on” state, while p53 loss removes a brake, yielding the stereotyped metabolic hallmarks of malignancy ^[21]D5[65]D5. Therapeutic strategies targeting glycolysis, including MCT and LDH inhibitors, as well as agents that restore mitochondrial function, are under clinical investigation ^{[61]D5[60]D5[13]D5}.

From this regulatory architecture, the net ATP and NADH yield of glycolysis emerges, the subject of the next section.

Pearl: Metabolite repair systems safeguard glycolytic integrity, and their defects cause inborn errors of metabolism.

Having examined the regulatory switches that control flux through glycolysis, the question of net energy return becomes central to understanding its role. The pathway’s balance sheet, accounting for every high-energy phosphate bond and reducing equivalent, reveals why it serves as both a rapid ATP source and a hub for redox and biosynthetic intermediates.

Net ATP and NADH Yield per Glucose Molecule

Glycolysis converts one molecule of glucose (6 carbons) into two molecules of pyruvate (3 carbons each) through a sequence of ten enzyme-catalyzed reactions. The net energy gain is modest but energetically efficient: 2 ATP and 2 NADH per glucose. The investment phase (steps 1-5) consumes 2 ATP (hexokinase/glucokinase and phosphofructokinase-1), while the payoff phase (steps 6-10) generates 4 ATP via substrate-level phosphorylation at phosphoglycerate kinase and pyruvate kinase, yielding the net of 2 ATP. Simultaneously, 2 NAD+ molecules are reduced to NADH at the glyceraldehyde-3-phosphate dehydrogenase step.

This 2-ATP yield is the anaerobic ceiling; further ATP extraction from glucose requires mitochondrial oxidation of pyruvate and NADH.

The Energetic Cost of Redox Balance

Glycolysis cannot proceed without a continuous supply of oxidized NAD+. In the absence of oxygen or when mitochondrial capacity is exceeded, lactate dehydrogenase reduces pyruvate to lactate, regenerating NAD+ and allowing glycolysis to sustain ATP production - albeit with the same 2 ATP yield. Lactate, long viewed as a waste product, is now recognized as a major circulating carbohydrate fuel and a redox buffer that equilibrates the NADH/NAD+ ratio across tissues ^[2]D5. Circulating lactate can be taken up by heart, liver, and kidney and reconverted to pyruvate for oxidation or gluconeogenesis, effectively shifting the energetic burden from the glycolytic cell to the whole body.

Under aerobic conditions, the 2 NADH produced per glucose are oxidized back to NAD+ by shuttling reducing equivalents into mitochondria. Two principal shuttles operate: the glycerol-3-phosphate shuttle (yielding 1.5 ATP per NADH via FADH₂) and the malate-aspartate shuttle (yielding 2.5 ATP per NADH). The malate-aspartate shuttle is particularly important in liver and kidney because it also supports gluconeogenesis and ureagenesis. Defects in this shuttle, as in citrin deficiency (SLC25A13 mutations), impair hepatic glycolysis, cause an energy deficit in hepatocytes, and lead to failure to thrive, dyslipidemia, and hyperammonemia ^[69]D5[70]D5. Medium-chain triglyceride (MCT) supplementation bypasses the glycolytic block by providing an alternative mitochondrial fuel, correcting the cytosolic NAD+/NADH ratio via the malate-citrate shuttle, and restoring lipogenesis ^[69]D5[70]D5.

Clinical Implications of the Energy Budget

The stoichiometric yield of glycolysis dictates its biological roles. Erythrocytes, which lack mitochondria, rely entirely on glycolysis for ATP and produce lactate as the end product. A single red blood cell consumes about 0.01 pmol ATP per second, all derived from the net 2 ATP per glucose ^[17]D5. During hypoxia, cells upregulate glycolytic enzymes via HIF-1α stabilization, shifting ATP production from oxidative phosphorylation to glycolysis to maintain viability ^[11]D5[19]D5. This metabolic flexibility is essential in ischemic tissue but, in cancer, drives the Warburg effect: tumor cells avidly consume glucose and excrete lactate even in the presence of oxygen, yielding only 2 ATP per glucose but providing abundant glycolytic intermediates for biosynthesis ^[72]D5.

The net 2 ATP per glucose, modest compared to the ~30-32 ATP from complete oxidation of glucose, underscores why glycolysis is integrated with other pathways to meet whole-body energy demands, as explored next.

Pearl: The net ATP yield of glycolysis is only 2 per glucose, but the 2 NADH produced represent an additional ~3-5 ATP if shuttled into mitochondria. In cells lacking mitochondria (erythrocytes) or with defective shuttles (citrin deficiency), the 2 ATP ceiling becomes a critical bioenergetic bottleneck.

The energetic yield of 2 ATP per glucose masks a more critical dimension: how glycolytic flux is partitioned and coordinated across organs to meet whole-body demands. Under normal physiology, insulin, glucagon, catecholamines, and substrate availability direct tissue-specific glycolytic activity, while pathological states such as cancer and fibrosis hijack this coordination.

Fed vs Fasted Continuum

In the fed state, insulin drives glucose uptake and glycolysis in muscle and adipose tissue. As fasting progresses, glucagon and cortisol suppress hepatic glycolysis and activate gluconeogenesis, sparing glucose for the brain. The brain itself maintains a high glycolytic rate but shifts to ketone body oxidation during prolonged fasting. This metabolic flexibility depends on the ability of tissues to toggle between glycolysis and oxidative phosphorylation (OXPHOS), a balance governed by pyruvate flux. At the pyruvate branch point, the mitochondrial pyruvate carrier (MPC) directs carbohydrate-derived carbon toward oxidation, while lactate dehydrogenase (LDH) diverts it toward lactate, a key mechanism in the Warburg effect ^[79]D5.

Tissue-specific Glycolytic Profiles

RBCs rely exclusively on glycolysis because they lack mitochondria, producing lactate as the obligate end product. Skeletal muscle possesses a high glycolytic capacity, especially during intense contraction, and shifts to OXPHOS at rest. Liver is uniquely bidirectional: glycolytic in the fed state, gluconeogenic in fasting. Heart muscle preferentially oxidizes fatty acids but switches to glycolysis under ischemia.

Mechanotransduction Links Mechanics to Glycolysis

Mechanical forces regulate glycolysis through the YAP/TAZ signaling axis. Downstream of integrin-FAK-Src and RhoA/ROCK, YAP/TAZ activate transcription of key glycolytic enzymes including GLUT1, HK2, and PFKFB3. This mechanometabolic coupling is implicated in cardiovascular remodeling, fibrosis, and tumor progression, where stiff extracellular matrix drives a pro-glycolytic phenotype ^[44]D5.

Lactate as a Systemic Signal and the Cori Cycle

Lactate produced by glycolysis in muscle or RBCs is shuttled via the Cori cycle to the liver, where it is reconverted to glucose at a cost of 6 ATP per glucose. In cancer, this cycle is pathologically amplified: tumors export large quantities of lactate through monocarboxylate transporters, acidifying the tumor microenvironment, suppressing anti-tumor immunity, and promoting regulatory T cell expansion and MDSC recruitment ^[76]D5[77]D5. This lactate-rich milieu also impairs effector T cell metabolism, driving mitochondrial dysfunction and reducing OXPHOS capacity ^[77]D5.

Therapeutic Strategies Targeting Whole-body Glycolytic Coordination

The Warburg effect, aerobic glycolysis even when oxygen is sufficient ^{[75]B2a[76]D5}, creates a therapeutic window. Ketogenic diet (KD) as adjunctive therapy for glioblastoma exploits this glucose dependency: adherent patients achieve a median OS of 29.4 months (vs 14.6 months historical control) with a 66.7% 3-year survival rate, and no Grade 3/4 diet-related toxicities ^[75]B2a. Pharmacologic inhibitors targeting HK2, G6PD, PFKFB3, and LDHA are in preclinical and early clinical investigation ^[76]D5, and metabolic inhibitors combined with immune checkpoint blockade (e.g., LDHA or IDO1 inhibitors plus anti-PD-1) aim to overcome acquired immunotherapy resistance by rebalancing nutrient competition in the TME ^[77]D5. In the cardiovascular realm, platelet mitochondria undergo an OXPHOS-to-glycolysis shift that drives thrombus formation, suggesting metabolism-targeting antiplatelet strategies distinct from traditional agents ^[78]D5.

Pearl: The shift from OXPHOS to glycolysis in cancer is not a mere epiphenomenon, it is an exploitable liability; therapies that restrict glucose (ketogenic diet) or block key glycolytic enzymes (HK2, LDHA, PFKFB3) aim to collapse this metabolic adaptation, and emerging data show meaningful survival gains without major toxicity ^{[75]B2a[76]D5}.

Having traced how glycolysis integrates with whole-body glucose homeostasis, we now examine the specific cofactors, vitamins, and minerals required for each catalytic step. These dependencies explain why nutritional deficiencies and inborn errors of cofactor synthesis produce characteristic metabolic derangements.

NAD⁺ (Vitamin B₃/Niacin) and the GAPDH Step

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is the only glycolytic enzyme that directly uses a vitamin-derived cofactor: NAD⁺ (from niacin , vitamin B₃). Oxidation of GAP to 1,3-BPG reduces NAD⁺ to NADH, which must be reoxidized to sustain glycolytic flux ^[27]D5. The malate-aspartate shuttle (MAS) and glycerol-3-phosphate shuttle perform this regeneration in most tissues; MAS deficiencies cause redox imbalance with elevated lactate and glycerol-3-phosphate ^[40]D5. Niacin deficiency (pellagra) reduces NAD⁺ availability and can impair GAPDH flux, shifting the redox state toward NADH accumulation and contributing to lactic acidosis under high glycolytic demand ^[83]D5.

Magnesium (Mg²⁺): A Universal Cofactor

Magnesium is a required cofactor for hexokinase, phosphofructokinase-1, phosphoglycerate kinase, enolase, and pyruvate kinase, every ATP-consuming or ATP-generating step ^[85]D5. Mg²⁺ forms the MgATP²⁻ complex that is the true substrate for all kinases. Under anoxic conditions, elevated cytosolic Mg²⁺ activates adenylate kinase and PPi-dependent reactions, saving ATP and enabling glycolysis to persist ^[85]D5. Hypomagnesemia (common with diuretics, alcoholism, or diarrhea) can theoretically impair glycolysis, though clinical effects are usually overshadowed by concurrent electrolyte disturbances.

2,3-Bisphosphoglycerate (2,3-BPG) and Phosphoglycerate Mutase

The cofactor-dependent form of phosphoglycerate mutase (dPGM), a member of the histidine phosphatase superfamily, requires 2,3-BPG to phosphorylate its catalytic histidine and enable conversion of 3-PG to 2-PG ^[38]D5. In erythrocytes, 2,3-BPG is synthesized via the Rapoport-Luebering shunt from 1,3-BPG. Shifts in 2,3-BPG levels (e.g., in anemia, hypoxia, or certain enzyme defects) may alter PGAM activity, but adaptive mechanisms generally preserve flux.

Thiamine (Vitamin B₁) and Pyruvate Entry into the TCA Cycle

Although thiamine pyrophosphate (TPP) is not a cofactor for glycolysis itself, it is essential for pyruvate dehydrogenase (PDH), which converts glycolytic pyruvate to acetyl-CoA. Thiamine deficiency (beriberi) inhibits PDH, causing accumulation of pyruvate and lactate - often manifesting as lactic acidosis >2 mmol/L. This creates an indirect but clinically critical dependency of glycolytic output on thiamine status.

Pantothenic Acid (Vitamin B₅) and Coenzyme A

Coenzyme A , derived from pantothenic acid , is required for acetyl-CoA synthesis from pyruvate (via PDH) and for entry into the TCA cycle. CoA esters, particularly malonyl-CoA, also function as allosteric regulators of glycolysis by controlling fatty acid oxidation ^[27]D5.

Table: Glycolytic Enzyme-Cofactor Mapping

* Potassium is an established monovalent cation cofactor for pyruvate kinase.

These dependencies lay the foundation for the inborn errors and nutritional deficiencies discussed in the next section.

Pearl: The only vitamin-derived cofactor required directly by a glycolytic enzyme is NAD⁺ (vitamin B₃); all other cofactor dependencies involve minerals (Mg²⁺, K⁺) and the metabolite 2,3-BPG, while thiamine and pantothenate support downstream pyruvate metabolism.

From the cofactor dependencies of glycolysis, the clinical consequences of pathway disruption become immediately apparent. Inborn errors, circulating biomarkers, and pharmacologic targets each trace back to the same enzymatic steps and regulatory nodes.

Inborn Errors of Glycolysis

Glycolytic enzyme deficiencies are rare but instructive. They typically present with exercise-induced rhabdomyolysis , hemolytic anemia, or multisystem disease depending on tissue-specific isoform expression ^[5]D5[6]D5. The following table summarizes the major disorders:

Citrin deficiency illustrates how a single transporter defect disrupts multiple pathways. Citrin is essential for the malate-aspartate NADH shuttle, which is required for hepatic glycolysis. Without it, hepatocytes cannot use glucose or fatty acids as energy sources, leading to an energy deficit that drives steatosis, cholestasis, and hyperammonemia ^[70]D5[86]D5. MCT supplementation bypasses the defect by providing medium-chain fatty acids that enter mitochondria independently of carnitine shuttle, rapidly supplying energy and restoring redox balance ^[69]D5. The recommended dose is equivalent to the liver's energy requirements, given in divided doses with meals ^[70]D5. Intravenous fructose is contraindicated, and persistent hyperglycemia must be avoided ^[70]D5.

PGM1-CDG is notable because it is one of the few congenital disorders of glycosylation with an effective treatment: d-galactose improves glycosylation, liver function, and hypoglycemia ^[1]A1c. Early diagnosis is critical, as untreated cardiomyopathy can be life-threatening.

Metabolite repair enzymes also protect glycolysis from toxic side-products. Deficiency of these enzymes can cause disease, highlighting the fragility of the pathway ^[57]D5.

Biomarkers

Lactate is the most clinically relevant biomarker of glycolytic flux. Once considered a waste product, lactate is now recognized as a major energy source, gluconeogenic precursor, and signaling molecule ^[22]D5. In critical illness, elevated blood lactate correlates with severity and is used as a "strain" biomarker rather than simply a marker of hypoxia ^[22]D5.

Lactate-to-pyruvate ratio helps distinguish causes of lactic acidosis : a ratio >25 suggests impaired mitochondrial oxidation (e.g., respiratory chain defects), while a ratio <10 favors increased glycolysis (e.g., hypoxia, exercise) ^[22]D5.

Glucose-6-phosphate dehydrogenase (G6PD) activity is measured in patients with unexplained hemolytic anemia. Deficiency is common worldwide and can be detected by fluorescent spot test or quantitative assay ^[3]D5.

Citrulline and ammonia are elevated in citrin deficiency, especially during metabolic decompensation. Newborn screening for citrullinemia can identify NICCD ^[86]D5.

Drug Targets

Glycolysis is an attractive therapeutic target in cancer, where the Warburg effect drives aerobic glycolysis even in the presence of oxygen. Key nodes under investigation include:

• GLUT1 (glucose transporter): Inhibitors reduce glucose uptake in tumors ^[89]D5[91]D5.
• Hexokinase 2 (HK2): Overexpressed in many cancers; inhibition impairs glycolysis and sensitizes to chemotherapy ^[89]D5[91]D5.
• PFKFB3: Regulates glycolysis through fructose-2,6-bisphosphate; inhibitors are in preclinical development ^[91]D5.
• Pyruvate dehydrogenase kinase (PDK1): Inhibits pyruvate dehydrogenase, promoting lactate production; dichloroacetate (DCA) reactivates PDH and shifts metabolism toward oxidative phosphorylation ^[91]D5.
• Lactate dehydrogenase A (LDHA): Converts pyruvate to lactate; knockdown reduces tumor growth and metastasis ^[91]D5.
• Monocarboxylate transporters (MCT1/4): Export lactate from cells; inhibitors disrupt tumor microenvironment acidification and immune evasion ^[30]D5[89]D5.

Lactate itself is now targeted as an immunosuppressive oncometabolite. Accumulation in the tumor microenvironment impairs T cell proliferation and cytotoxic function while expanding regulatory T cells and myeloid-derived suppressor cells ^[61]D5[91]D5. Inhibiting lactate production or transport can restore antitumor immunity.

Combination strategies are essential. Glycolytic inhibitors are being tested with immune checkpoint blockade, chemotherapy, and radiotherapy. For example, in anaplastic thyroid carcinoma, targeting HK2 or MCT enhances radiosensitivity and improves response to targeted therapies ^[89]D5. In ovarian cancer, glycolytic reprogramming supports chemoresistance, and combining glycolysis inhibitors with platinum-based chemotherapy shows promise ^[91]D5.

Traditional Chinese medicine compounds also modulate glycolysis. Ginsenosides (Rg3, Rh3, compound K) inhibit Warburg-type glycolysis in cancer cells, though their clinical translation is limited by pharmacokinetics and microbiota-dependent activation ^[60]D5. Astragalus-based formulas and berberine also attenuate glycolytic flux ^[87]D5.

Citrin deficiency offers a rare example of a metabolic disorder treatable by dietary intervention. MCT supplementation is the mainstay, but sodium pyruvate, ursodeoxycholic acid, and nitrogen scavengers address specific aspects ^[69]D5. Liver transplantation remains the only curative option for severe cases ^[69]D5.

Pearl: In any patient with unexplained rhabdomyolysis, hemolytic anemia, or neonatal cholestasis, consider an inborn error of glycolysis, the diagnosis is treatable (e.g., d-galactose for PGM1-CDG, MCT for citrin deficiency) and early intervention prevents irreversible organ damage ^[1]A1c[69]D5.

Building on these clinical correlations, distinguishing glycolysis from its reciprocal and parallel pathways is essential to avoid diagnostic and conceptual errors. Several isozyme differences, regulatory loops, and transport steps are routinely conflated.

Glycolysis vs. Gluconeogenesis: Not Simply Reversible

Although seven of the ten glycolytic enzymes are shared with gluconeogenesis, the three irreversible steps catalyzed by hexokinase/glucokinase, phosphofructokinase-1 (PFK-1), and pyruvate kinase are bypassed by distinct gluconeogenic enzymes, glucose-6-phosphatase, fructose-1,6-bisphosphatase, and pyruvate carboxylase plus phosphoenolpyruvate carboxykinase. Flux through these reciprocal pathways is reciprocally regulated by insulin and glucagon; their simultaneous activation (futile cycling) is prevented by tight allosteric control. The clinical relevance of this distinction is evident in colorectal adenoma versus cancer, where the glycolysis/gluconeogenesis pathway is among the most altered metabolic modules ^[94]B3a.

Hexokinase vs. Glucokinase: Affinity, Inhibition, and Tissue Role

Hexokinase I-III have a low Km (~0.1 mM) for glucose and are inhibited by glucose-6-phosphate, trapping glucose in most tissues. Glucokinase (hexokinase IV) has a high Km (~10 mM), lacks product inhibition, and functions as a glucose sensor in hepatocytes and pancreatic β-cells. Confusing these isozymes leads to misunderstanding of hepatic glucose uptake and the pathogenesis of MODY (maturity-onset diabetes of the young) type 2, caused by glucokinase mutations.

The Citrate Transporter: A Glycolysis-Lipogenesis Link

NaCT/SLC13A5, the sodium-coupled citrate transporter, delivers citrate into hepatocytes, where it fuels both lipogenesis and gluconeogenesis. Citrate also allosterically inhibits PFK-1, coupling glycolytic flux to citrate availability. Notably, SLC13A5 deficiency in humans causes severe epileptic encephalopathy (EIEE25/DEE25), while in mice the same deficiency protects against diet-induced obesity and diabetes, a species difference explained by the high-capacity human transporter versus low-capacity mouse ortholog ^[92]D5. This distinction is critical when translating rodent metabolic findings to humans.

Pearl

Pearl: The three irreversible steps of glycolysis are always bypassed in gluconeogenesis, never assume an enzyme’s presence in both pathways implies reversibility; the isozyme-specific regulation of hexokinase versus glucokinase determines tissue-specific glucose sensing and is a common source of board-question error.

From the distinctions between aerobic glycolysis and the Warburg effect, several unifying principles emerge that consolidate the pathway's role from molecular structure to clinical application. Glycolysis is not merely an anaerobic ATP source but a signaling hub whose intermediates, lactate, NADH, and glycolytic metabolites, regulate gene expression, immune function, and redox balance ^[17]D5[98]D5. The lactate shuttle paradigm reframes lactate as a major energy substrate and gluconeogenic precursor, not a waste product; blood lactate elevation signals metabolic strain rather than mere stress ^[22]D5.

Key Metabolic Principles

• Aerobic glycolysis is a kinetically regulated ATP supply pathway operating alongside oxidative phosphorylation to meet cellular demand, not a pathological contingency ^[98]D5.
• The pentose phosphate pathway (PPP) branches from glycolysis to supply NADPH for redox homeostasis and ribose 5-phosphate for nucleotide synthesis; G6PD deficiency (affecting ~400 million people worldwide) causes hemolytic anemia and impaired leukocyte oxidative burst ^[3]D5.
• Lactate functions as a signaling molecule via lysine lactylation (a post-translational modification) and N-lactoyl amino acids, translating transient glycolytic flux into durable downstream effects ^[10]D5.

Clinical Pearls

• PGM1 deficiency (PGM1-CDG): Presents with cleft palate, hypoglycemia, and cardiomyopathy; treatable with d-galactose supplementation ^[1]A1c.
• Citrin deficiency: Causes hyperammonemia and cholestasis due to malate-aspartate shuttle dysfunction; management includes medium-chain triglycerides (MCT) to supply hepatic energy and restore redox balance ^[69]D5[86]D5.
• Lactic acidosis : In critical illness (e.g., ARDS), immunometabolic reprogramming with increased glycolysis and impaired oxidative phosphorylation drives inflammation; targeting NAD⁺ homeostasis and mitochondrial function is a therapeutic frontier ^[13]D5.
• Cancer metabolism: The Warburg effect (aerobic glycolysis) supports anabolic demands and redox defense; total lesion glycolysis (TLG) on FDG-PET predicts survival in cervical cancer ^{[28]D5[30]D5[97]B2a}.

Therapeutic Implications

• Brain energy rescue: In Alzheimer disease, presymptomatic glucose hypometabolism can be countered with ketone-based interventions and lifestyle modifications ^[4]D5.
• Cardiac metabolism: The heart relies on flexible substrate use (fatty acids 40-60%, glucose 20-40%); malonyl-CoA allosteric control is a target for ischemic heart disease ^[27]D5.
• Immunometabolism: Glycolytic reprogramming in T cells and macrophages determines effector function; targeting lactate transport (e.g., MCT inhibitors) shows promise in autoimmunity and infection ^[20]D5[99]D5.

Pearl: When interpreting a high blood lactate, remember the lactate shuttle: it is a strain biomarker reflecting increased glycolytic flux and inter-organ shuttling, not necessarily tissue hypoxia, a distinction that guides resuscitation in sepsis and trauma ^[22]D5.