Push or Block: Both Directions of PKM2 Activity Kill Leukemia Cells

Acute myeloid leukemia cells carry a metabolic fingerprint that sets them apart from their healthy neighbors: they abandon oxidative phosphorylation in favor of anaerobic glycolysis, a shift that fuels their relentless growth. At the heart of this rewiring sits pyruvate kinase, the enzyme that catalyzes the final step of glycolysis. In healthy, energy-hungry tissues, the PKM gene produces the M1 isoform, a constitutively active tetramer that churns out pyruvate for mitochondrial ATP production. Cancer cells, including leukemia cells, replace it with the M2 isoform, a predominantly dimeric, low-activity enzyme that leaves upstream glycolytic intermediates to pile up and divert into biosynthetic pathways such as the pentose phosphate pathway and the serine synthesis pathway.

PKM2 protein levels were elevated in AML cells, yet overall pyruvate kinase enzymatic activity was significantly reduced compared with peripheral blood mononuclear cells from healthy donors. The researchers quantified PKM2 mRNA by quantitative reverse transcription polymerase chain reaction in patient-derived AML samples and healthy donor PBMCs, finding significantly higher PKM2 transcript levels in the leukemic cells, and measured baseline enzymatic activity with a lactate dehydrogenase-coupled fluorometric assay.

The paradox at the center of the paper is that both suppressing and boosting PKM2 activity produce the same lethal outcome for leukemia cells. The team describes this as a “Goldilocks effect”: like blood glucose, which must stay within a narrow range for the body to survive, PKM2 activity must sit within an optimal window. Push it in either direction and the cell dies. To test this, the researchers used lentiviral-mediated knockdown with two distinct short hairpin RNA sequences in TEX and OCI-AML2 leukemia cell lines, confirming reduced protein levels and reduced enzymatic activity by immunoblotting and activity assays.

The consequences of genetic inhibition were stark. Cells with reduced PKM2 proliferated significantly more slowly and formed fewer colonies than controls transduced with a scrambled sequence. When the researchers injected equal numbers of control or knockdown cells into the tail veins of immunodeficient NSG mice, the knockdown cells showed roughly fourfold less bone marrow engraftment after five weeks, measured as the human CD33-positive and CD45-positive fraction by flow cytometry. Mice carrying the knockdown cells also lived significantly longer, with a median survival of 42 days compared with 33 days for mice injected with control cells.

Pharmacological activation told the other half of the story. Because few PKM2-targeting compounds exist, the team ran a cell-based drug screen comparing control and PKM2-knockdown leukemia cells, reasoning that compounds losing potency in knockdown cells must depend on PKM2 for their activity. That screen identified plumbagin, a 5-hydroxy-2-methyl-1,4-naphthoquinone, as a PKM2 modulator. Plumbagin’s cytotoxicity was significantly diminished in knockdown cells, indicating its killing effect depends on the presence of the enzyme. Using recombinant enzymes, the researchers showed that plumbagin increased PKM2 activity in a dose-dependent manner while leaving PKM1 unaffected, and gas chromatography-mass spectrometry detected the compound within PKM2-enriched fractions isolated from treated cells. Cellular thermal shift assays, which detect ligand binding as an increase in a protein’s melting temperature, showed plumbagin raised the thermal stability of PKM2 but not PKM1, supporting it as a specific PKM2-targeting activator.

In patient-derived AML cells, plumbagin elevated both total pyruvate kinase and PKM2-specific activity, with no change in healthy donor PBMCs. Leukemic cell viability fell with increasing plumbagin concentrations, with half-maximal inhibitory concentrations between 1.40 and 2.73 micromolar across a panel of hematological malignant cell lines, while PBMCs showed no significant effects. At one micromolar, plumbagin reduced clonogenic growth of patient-derived AML cells without impairing colony formation by healthy PBMCs. The selectivity carried into vivo models: mice engrafted intrafemorally with patient-derived AML cells and treated with 2.5 milligrams per kilogram of plumbagin three times per week for five weeks showed approximately 85 percent reduction in leukemic engraftment in the opposite femur, whereas mice engrafted with normal CD34-positive bone marrow cells showed no significant change. Body weight and toxicity markers including bilirubin, creatine kinase and alanine aminotransferase showed no significant changes, suggesting the treatment was well tolerated in these models.

The mechanism that reconciles why opposite directions of modulation converge on the same fate lies in PKM2’s oligomeric state and its second, non-metabolic life in the nucleus. Only dimeric PKM2 can translocate into the nucleus, where it acts as a protein kinase, phosphorylating targets such as STAT3 and histone H3 and serving as a transcriptional co-activator that upregulates oncogenes including c-Myc. In AML, c-Myc is a major oncogenic driver that promotes leukemia cell growth and blocks myeloid differentiation. Using disuccinimidyl suberate cross-linking to separate dimeric from tetrameric PKM2 by molecular weight, the researchers found that genetic knockdown reduced both oligomeric forms, while plumbagin treatment increased tetrameric PKM2 and raised the tetramer-to-dimer ratio.

Both scenarios starved the nucleus of dimeric PKM2. Nuclear-enriched fractions showed reduced nuclear PKM2 accumulation after genetic inhibition and after pharmacological activation alike, since tetramerization locks the enzyme out of the nucleus just as effectively as deleting the protein. The downstream molecular consequence was a drop in c-Myc expression in both models, with genetic inhibition producing an approximate 50 percent decrease. The authors propose that this interruption of PKM2’s nuclear protein kinase function is the converging point through which both inhibition and activation trigger leukemia cell death, linking the enzyme’s metabolic and transcriptional roles into a single regulatory circuit that determines cell fate.

The findings position PKM2 as a promising therapeutic target in AML, a disease with poor prognosis in which current research into pharmacological modulation remains sparse. Prior work has shown that PKM2 inhibitors such as 3k and shikonin modestly impair solid tumor growth, while activators such as ML-265 reduce lung tumor volume in mice but require high dosing of 50 milligrams per kilogram twice daily, underscoring the need for more potent, PKM2-specific compounds. By demonstrating that a single molecule, plumbagin, can selectively activate PKM2 in leukemic cells while sparing normal blood cells, and by explaining mechanistically why both directions of modulation are lethal, the study offers a framework for targeting an enzyme that leukemia cells cannot afford to have too little of, or too much of. The work remains at the preclinical stage, based on cell lines, patient-derived samples and mouse engraftment models, and clinical application will require further validation, but it clarifies why an enzyme long viewed simply as a glycolytic workhorse may instead be a finely tuned switch governing leukemia cell survival.

Subject of Research: The role of the pyruvate kinase isoform PKM2 in acute myeloid leukemia cell survival and metabolism

Article Title: PKM2 modulation regulates leukemia cell survival

Article References: Vrdoljak, N., Parfenova, E. N., Roma, A., Mosca, D. A., Minden, M. D., & Spagnuolo, P. A. (2026). PKM2 modulation regulates leukemia cell survival. Cancer & Metabolism, 14(1), Article 26. https://doi.org/10.1186/s40170-026-00440-7

Image Credits: AI Generated

DOI: 10.1186/s40170-026-00440-7

Keywords: acute myeloid leukemia, PKM2, pyruvate kinase, cancer metabolism, glycolysis, plumbagin, c-Myc, nuclear translocation, drug screen, preclinical study, Cancer & Metabolism, leukemia research