Glycolysis, the ten-step pathway that breaks glucose down to pyruvate, is often presented as a fixed sequence of reactions with one enzyme per step. In reality, most of these steps are carried out by families of enzyme variants, each tuned to the needs of a particular tissue, developmental stage, or metabolic state. These variants differ in how they are regulated, where they sit inside the cell, and even what non-metabolic jobs they take on. Understanding glycolytic enzymes means looking past the textbook diagram and into the diversity that makes the pathway flexible enough to serve a beating heart, a dividing tumor cell, and a plant chloroplast.
Hexokinase Isoforms and the First Committed Step
Hexokinase kicks off glycolysis by slapping a phosphate group onto glucose, trapping it inside the cell. Mammals have four hexokinase isoforms (I through IV), and they are not interchangeable. Hexokinase II is the dominant version in insulin-sensitive tissues like the heart, skeletal muscle, and fat, and it is also the isoform that gets cranked up in many tumors to feed their hungry glucose metabolism.1PubMed Central. Hexokinase II integrates energy metabolism and cellular protection: Akting on mitochondria and TORCing to autophagy Hexokinase IV, better known as glucokinase, handles glucose sensing in the liver and pancreas and has a much higher threshold for glucose before it kicks into gear, which is why it can act as a glucose sensor. The picture gets more nuanced than textbooks usually let on: in human liver, hexokinase III turns out to have surprisingly high activity, roughly half of the total low-affinity hexokinase activity, and it shows strong inhibition at high glucose concentrations. That quirk actually complicates laboratory measurements of glucokinase, since standard assays can confuse the two.2PubMed. Hexokinase isoenzymes in normal and cirrhotic human liver: suppression of glucokinase in cirrhosis
The way hexokinase works at a molecular level is itself a classic story in enzyme science. When glucose binds to yeast hexokinase, one lobe of the enzyme rotates about 12 degrees relative to the other, moving parts of the protein backbone by as much as 8 angstroms and closing the cleft around the sugar. This glucose-triggered shape change is essential for catalysis and is one of the best-documented examples of “induced fit,” the idea that the substrate actively reshapes the enzyme rather than simply slotting into a pre-formed pocket.3PubMed Central. Glucose-induced conformational change in yeast hexokinase
PFK-1, the Gatekeeper of Glycolysis
If hexokinase opens the door to glycolysis, phosphofructokinase-1 (PFK-1) decides how far it swings open. PFK-1 catalyzes the first truly committed step, converting fructose-6-phosphate to fructose-1,6-bisphosphate, and the cell has piled more regulatory inputs onto this enzyme than almost any other. When cellular energy is low, molecules like AMP and ADP activate PFK-1, speeding glycolysis to produce more ATP. When energy is plentiful, ATP itself and citrate (a product of the next metabolic hub, the citric acid cycle) inhibit PFK-1, slowing glucose breakdown.4PubMed Central. Structural basis for allosteric regulation of human phosphofructokinase-1 Recent structural work on human PFK-1 has revealed that ATP binds at multiple inhibitory sites and that the enzyme’s own tail end helps lock it into the inactive state, details that differ from the bacterial version studied for decades.5PubMed Central. Structural basis for allosteric regulation of human phosphofructokinase-1 – Section: Structure of human PFKL in the T-state
The regulation goes beyond simple energy signals. Long-chain fatty acyl-CoAs, the activated forms of fatty acids, directly and potently inhibit PFK-1 at very low concentrations. This provides a direct link between fat metabolism and sugar metabolism: when the cell is burning plenty of fat, it dials down glycolysis. The protective molecules AMP and ADP can counteract this fatty-acid brake, but ATP cannot, which means the enzyme integrates both lipid availability and energy charge simultaneously.6PubMed Central. Reversible high affinity inhibition of phosphofructokinase-1 by acyl-CoA: a mechanism integrating glycolytic flux with lipid metabolism
Fructose-2,6-Bisphosphate and Hormonal Tuning
One of the most powerful activators of PFK-1 is not a direct product of glycolysis at all but a signaling molecule called fructose-2,6-bisphosphate (F2,6BP). This molecule is made by a separate family of bifunctional enzymes known as PFKFBs (6-phosphofructo-2-kinase/fructose-2,6-bisphosphatases), which can both produce and destroy F2,6BP depending on hormonal signals.7PubMed Central. Balancing glycolytic flux: the role of 6-phosphofructo-2-kinase/fructose 2,6-bisphosphatases in cancer metabolism The PFKFB3 isoform, for instance, converts fructose-6-phosphate to F2,6BP, which then allosterically activates PFK-1 and ramps up glycolysis. This makes PFKFB3 a tempting drug target in cancers that depend on high glycolytic flux.8PubMed. Treatment against glucose-dependent cancers through metabolic PFKFB3 targeting of glycolytic flux Insulin, growth factors, and other hormones ultimately control the balance between the kinase and phosphatase activities of PFKFBs, giving the body a way to tune glycolysis over minutes to hours in response to feeding, fasting, or stress.
Pyruvate Kinase and the M1/M2 Switch
Pyruvate kinase, the enzyme that catalyzes the final step of glycolysis, comes in tissue-specific isoforms. The one that has drawn the most attention is the M2 isoform (PKM2). Normal adult tissues mainly express the M1 isoform, but tumor cells overwhelmingly switch to M2. Both come from the same gene; the difference is which of two mutually exclusive exons gets included during RNA processing.9PubMed Central. Turning on a fuel switch of cancer: hnRNP proteins regulate alternative splicing of pyruvate kinase mRNA That single exon swap changes the enzyme’s behavior dramatically.
PKM2 can exist as an active tetramer (four subunits joined together) or as a less active dimer. In the dimeric form, PKM2 can actually leave the cytoplasm and enter the nucleus, where it moonlights as a regulator of gene expression involved in tumor growth and metastasis.10PubMed Central. PKM2, function and expression and regulation What tips the balance between tetramer and dimer? Post-translational modifications like phosphorylation and acetylation disrupt the binding site for fructose-1,6-bisphosphate, the allosteric activator that normally holds the tetramer together. Without that glue, the complex falls apart into dimers that head for the nucleus.11PubMed Central. Structural basis for allosteric regulation of pyruvate kinase M2 by phosphorylation and acetylation
High glucose concentrations accelerate this process by promoting acetylation of PKM2 at a specific lysine residue (K305). That acetylation both reduces the enzyme’s catalytic activity and tags it for degradation through a lysosome-dependent recycling pathway. The net result is an accumulation of glycolytic intermediates upstream of pyruvate kinase, which the tumor cell diverts into biosynthetic pathways for building new membranes, nucleotides, and amino acids.12PubMed Central. Acetylation targets the M2 isoform of pyruvate kinase for degradation through chaperone-mediated autophagy and promotes tumor growth Tumors transitioning to high-glucose, low-oxygen environments upregulate these glycolytic enzyme isoforms to sustain the Warburg effect, the paradoxical preference for glycolysis even when oxygen is available.13PubMed Central. Two transition states of the glycogen shunt and two steady states of gene expression support metabolic flexibility and the Warburg effect in cancer
Enzymes With Surprising Second Lives
Several glycolytic enzymes turn out to have functions entirely unrelated to sugar metabolism, a phenomenon sometimes called “moonlighting.” GAPDH (glyceraldehyde-3-phosphate dehydrogenase) is the most famous example. Beyond its day job of oxidizing glyceraldehyde-3-phosphate, GAPDH has been implicated in controlling gene expression and even programmed cell death in animal cells. The key feature enabling these side roles is a highly reactive cysteine in the active site that can be chemically modified by oxidative signals, including glutathionylation and nitrosylation. When stress conditions trigger these modifications, GAPDH’s behavior shifts from metabolic to regulatory. Plant cells show the same phenomenon: their cytoplasmic GAPDH carries the same reactive cysteine and accumulates the same kinds of stress-triggered modifications, suggesting these moonlighting functions evolved early.14PubMed Central. Plant cytoplasmic GAPDH: redox post-translational modifications and moonlighting properties
Triosephosphate isomerase (TIM) is at the other end of the spectrum, an enzyme so good at its one job that biochemists have called it “perfectly evolved.” It interconverts two three-carbon sugars so quickly that the reaction rate is limited only by how fast the substrates can physically diffuse to the enzyme’s active site.15PubMed Central. Triosephosphate isomerase: a highly evolved biocatalyst There is essentially no room left for natural selection to make TIM faster.
Aldolase Variants and Inherited Disease
Aldolase, which splits fructose-1,6-bisphosphate into two three-carbon fragments, comes in tissue-specific isoforms labeled A, B, and C. Aldolase B is the liver isoform and is responsible for metabolizing dietary fructose as well as participating in glycolysis. Mutations in the ALDOB gene cause hereditary fructose intolerance, an inherited condition in which the liver cannot properly process fructose-1-phosphate. Certain naturally occurring variants of aldolase B show particularly impaired activity toward fructose-1-phosphate, and the disease follows a recessive inheritance pattern, meaning both copies of the gene must be affected.16PubMed. Hereditary fructose intolerance: functional study of two novel ALDOB natural variants and characterization of a partial gene deletion People with hereditary fructose intolerance can manage the condition by strictly avoiding fructose and sucrose in their diet, but undiagnosed cases in infants can cause serious liver damage.
Pyruvate kinase deficiency, mentioned earlier for its M2 cancer connection, has its own clinical face in red blood cells. A deficiency in the red-cell pyruvate kinase isoform (PKR) leads to hemolytic anemia because red blood cells depend almost entirely on glycolysis for their energy. Without enough pyruvate kinase activity, these cells cannot maintain adequate ATP levels and break down prematurely. Interestingly, the same deficiency raises levels of 2,3-diphosphoglycerate, an intermediate that helps hemoglobin release oxygen more easily, which partially compensates for the anemia by improving oxygen delivery to tissues.17PubMed Central. The variable manifestations of disease in pyruvate kinase deficiency and their management
The Phosphoglycerate Mutase Mechanism
Phosphoglycerate mutase shuffles a phosphate group between two positions on the three-carbon backbone, converting 3-phosphoglycerate to 2-phosphoglycerate. It does this through a phospho-histidine intermediate: the enzyme first picks up a phosphate from the cofactor 2,3-diphosphoglycerate, temporarily attaching it to a histidine residue, then transfers it to the substrate. Classic experiments confirmed this by isolating the radioactively labeled phosphoenzyme and showing that the phosphate sits on a histidine, matching the chemical behavior of synthetic phosphohistidine.18Archives of Biochemistry and Biophysics. Evidence for a phosphohistidine protein intermediate in the phosphoglycerate mutase reaction Phosphohistidine chemistry is much less studied than the phosphoserine or phosphotyrosine modifications familiar from cell signaling, partly because the bond is unstable and hard to detect with standard techniques.
Glycolytic Metabolons and Spatial Organization
Textbooks draw glycolysis as a series of reactions floating freely in the cytoplasm, but cells sometimes organize the pathway into temporary multi-enzyme clusters called metabolons. In the roundworm C. elegans, researchers found that when neurons face energy stress, glycolytic enzymes shift from a diffuse distribution throughout the cell to concentrated puncta right next to synapses. These enzymes colocalize with one another, suggesting the formation of an on-demand glycolytic compartment that produces ATP exactly where it is needed to keep synapses firing.19PubMed Central. Glycolytic Enzymes Localize to Synapses under Energy Stress to Support Synaptic Function
Metabolons are not unique to animal neurons. They are a general biological strategy: transient assemblies of sequential enzymes held together by weak, non-covalent interactions. Unlike permanent multi-enzyme complexes, metabolons form and dissolve as conditions change, allowing the cell to channel substrates directly from one active site to the next without losing them to the bulk of the cytoplasm.20PubMed Central. Metabolons, enzyme-enzyme assemblies that mediate substrate channeling, and their roles in plant metabolism This substrate channeling can dramatically increase the effective speed of a pathway and prevent side reactions. The transient nature of these assemblies is part of what makes them hard to study: they fall apart during standard cell lysis, so many metabolons probably remain undiscovered.
An Ancient Pathway, Conserved Across Life
Glycolysis is often called one of the oldest metabolic pathways, and genomic data backs this up. An analysis of nearly a thousand prokaryotic genomes found that the “trunk” enzymes of glycolysis, those catalyzing the middle and lower steps, are present in over 98% of all bacterial and archaeal genomes sampled. Enolase appeared in about 98.5% of genomes, phosphoglycerate kinase in roughly 99%, triosephosphate isomerase in about 98.6%, and GAPDH in over 99%. Even aldolase, which exists in structurally unrelated class I and class II forms across different lineages, showed up in about 95% of genomes. The near-universal distribution across both bacteria and archaea suggests these enzymes were already present in the last universal common ancestor of all cellular life.21PubMed Central. The early evolution of the glycolytic pathway from autotrophic origins to glycogen and back – Section: Universality of trunk glycolytic enzymes
What is interesting is that this conservation applies to the core catalytic function, not necessarily to the regulatory bells and whistles layered on top. The allosteric sites on human PFK-1, for example, differ from those on the bacterial enzyme. The isoform switching of pyruvate kinase is an animal innovation. Evolution has preserved the catalytic chassis of glycolysis while continuously retooling the regulatory elements to suit new metabolic contexts.
Therapeutic Targeting in Cancer
Because tumors depend on ramped-up glycolysis, glycolytic enzymes have attracted attention as drug targets. Researchers have explored inhibitors against hexokinase II, PFK-1, GAPDH, PKM2, and lactate dehydrogenase A (LDHA), among others. The challenge is that glycolysis is not optional for normal cells either. Inhibitors that effectively starve a tumor tend to carry unacceptable toxicity for healthy tissues, and tumors frequently adapt by switching to alternative fuel sources. Current thinking favors combination strategies, pairing glycolysis inhibitors with chemotherapy, radiation, immunotherapy, or other targeted agents, to lower the dose of each drug while exploiting synergies between them.22PubMed Central. Targeting Glycolytic Metabolism in Cancer Therapy: Current Approaches and Future Perspectives The field remains in early development. No glycolysis-specific inhibitor has yet become a standard-of-care cancer drug, though several are in clinical trials.
Plant Glycolysis Runs in Two Compartments
Plants add yet another layer of complexity. Unlike animal cells, which run glycolysis exclusively in the cytoplasm, plant cells maintain a complete glycolytic pathway in both the cytosol and inside their plastids (the organelles that include chloroplasts). The two parallel pathways use distinct enzyme isoforms encoded by separate nuclear genes.23PubMed. THE ORGANIZATION AND REGULATION OF PLANT GLYCOLYSIS This dual localization allows plants to coordinate sugar metabolism with photosynthesis, fatty acid synthesis, and other plastid-specific processes without the two compartments competing for the same enzyme pools. It also means that plant biochemists face a doubled cast of characters when trying to map glycolytic regulation, since the cytosolic and plastidial isoforms can respond to different signals and have different kinetic properties. The existence of compartmentalized glycolysis in plants is one reason findings from animal models do not always translate directly to agricultural or biofuel research.