Neutrophils have long been viewed as short-lived foot soldiers of the innate immune system, rushing to sites of infection to engulf bacteria before dying within hours. But over the past decade, tumor immunologists have recognized that neutrophils infiltrating tumors, known as tumor-infiltrating neutrophils, are far more than passive bystanders. In colorectal cancer, they promote tumor progression and resistance to therapy, yet the mechanisms that reprogram them inside tumors have remained poorly understood. The new study set out to answer a deceptively simple question: what tells a neutrophil arriving in a colon tumor to become an immunosuppressive cell rather than a tumor-killing one?

The answer, according to the researchers, begins inside the nucleus of the tumor cell itself. Using single-cell RNA sequencing of colon tumors from mice, the team identified distinct states of tumor-infiltrating neutrophils, including a population they designated T3 neutrophils. These T3 cells express high levels of two telling proteins: ferritin heavy chain 1, or FTH1, which protects cells from ferroptosis, an iron-dependent form of cell death driven by lipid peroxidation, and PD-L1, the molecular brake that tumors and their allies use to disable cytotoxic T cells. In other words, T3 neutrophils are both hard to kill and actively immunosuppressive, a dangerous combination for the host.

The mechanistic trail led back to GSK3β, a multifunctional kinase best known for its role in glycogen metabolism and the Wnt signaling pathway, both of which are frequently dysregulated in cancer. The team found that when GSK3β is active in the intestinal epithelial cells that give rise to tumors, it phosphorylates a linker histone called H1FX at a specific serine residue, serine 31 in the human protein. This phosphorylation event triggers the dissociation of a deubiquitinating enzyme called USP43 from H1FX, leaving the histone vulnerable to ubiquitination and destruction by the proteasome. Because linker histones help compact chromatin into inaccessible, transcriptionally silent configurations, the loss of H1FX opens up the genome, increasing chromatin accessibility specifically at the promoters of glycolytic genes.

The consequence is a metabolic shift. With glycolysis genes more accessible to transcriptional machinery, including the oncogenic transcription factor c-Myc and RNA polymerase II, tumor cells ramp up glucose consumption and lactate production. But the story does not end with lactate. The researchers discovered that the accumulating lactate chemically modifies fumarate hydratase, the enzyme that normally converts fumarate to malate in the tricarboxylic acid cycle, through a process called lactylation, which they traced to the enzyme AARS1 acting as a lactyltransferase. This lactylation inactivates fumarate hydratase, causing fumarate to build up in the tumor interstitial fluid. Fumarate, in this context, becomes an oncometabolite, a metabolic intermediate that, when accumulated aberrantly, actively drives malignancy.

The fumarate then acts on the neutrophils. Building on earlier work showing that fumarate can inhibit alpha-ketoglutarate-dependent enzymes, the team demonstrated that fumarate suppresses the activity of KDM2B, a histone demethylase, inside tumor-infiltrating neutrophils. With KDM2B inhibited, the repressive mark dimethylated histone H3 lysine 36, or H3K36me2, accumulates across the neutrophil genome, triggering an epigenetic reprogramming that pushes these cells toward the immunosuppressive T3 state. The fumarate also reinforces the survival of these cells by inhibiting the FTH1-dependent ferroptosis pathway, ensuring that the reprogrammed neutrophils do not simply die off but persist in the tumor, expressing PD-L1 and suppressing CD8-positive T cell activity.

The evidence supporting this metabolic-epigenetic axis is extensive. In mice engineered to lack GSK3β specifically in intestinal epithelial cells, colon tumors showed fewer T3 neutrophils, more functional CD8-positive T cells and reduced tumor burden. Conversely, tumors engineered to overexpress GSK3β recruited more T3 neutrophils and grew faster. The team also confirmed the axis in human colorectal cancer samples and in patient-derived organoids, where GSK3β levels correlated with fumarate production and the capacity of tumor-conditioned media to drive human neutrophils toward the T3 phenotype. Metabolite profiling, lactylproteomics and ATAC-seq experiments each independently corroborated the chain of causation, from histone phosphorylation through glycolysis and lactate accumulation to fumarate release and neutrophil polarization.

Perhaps the most clinically resonant finding involves tideglusib, a GSK3β inhibitor that has already been tested in human trials for neurological conditions. When the researchers treated mice bearing colorectal tumors with tideglusib, the drug reduced H1FX phosphorylation, curtailed glycolysis and fumarate accumulation, attenuated T3 neutrophil polarization and thereby relieved the immunosuppression that normally blunts checkpoint blockade therapy. In combination with immune checkpoint inhibitors, tideglusib produced markedly better tumor control than either approach alone in preclinical models. Because the drug is already clinically characterized, the path from these findings to a combination trial in colorectal cancer, a disease in which immunotherapy has historically benefited only a small subset of patients, appears unusually short.

The study also reframes how scientists think about communication between tumor cells and immune cells. Rather than relying solely on cytokines and surface receptors, colorectal cancer cells appear to broadcast their influence through metabolites, converting an intracellular kinase activity into a diffusible signal, fumarate, that rewrites the epigenetic identity of incoming neutrophils. This metabolic-epigenetic axis, as the authors describe it, suggests that oncometabolites may be far more than biomarkers of mitochondrial dysfunction; they can be deliberate instruments of immune evasion. It also raises the possibility that similar metabolite-mediated reprogramming operates in other tumor types, and that targeting the upstream metabolic events, rather than the exhausted immune cells themselves, could offer a more durable way to restore antitumor immunity.

Significant questions remain before these findings translate into the clinic. GSK3β performs essential functions in many tissues, and systemic inhibition carries potential toxicities that preclinical mouse models cannot fully anticipate. The precise contributions of individual neutrophil states to human disease, and whether fumarate levels in patient tumors can reliably predict immunotherapy response, will need prospective validation. Nonetheless, by connecting a chromatin-level event in tumor cells to an epigenetic switch in innate immune cells through a single metabolite, the work provides one of the most complete mechanistic narratives yet assembled for neutrophil reprogramming in cancer, and it hands researchers a concrete, druggable point of intervention in a disease that urgently needs new therapeutic angles.

Subject of Research: GSK3β-driven fumarate signaling and neutrophil reprogramming in colorectal cancer

Article Title: Intestinal epithelial GSK3β governs fumarate-dependent neutrophil reprogramming to promote colorectal cancer

Article References: Intestinal epithelial GSK3β governs fumarate-dependent neutrophil reprogramming to promote colorectal cancer. (n.d.). https://doi.org/10.1038/s43018-026-01253-9

Image Credits: AI Generated

DOI: 10.1038/s43018-026-01253-9

Keywords: colorectal cancer, GSK3β, neutrophils, fumarate, oncometabolites, ferroptosis, PD-L1, tumor microenvironment, epigenetics, lactylation, immunotherapy, tideglusib