Immune gene KDM6A drives heart failure after attack, mouse study shows

Losing a single epigenetic regulator from blood-forming cells can turn the immune system against a healing heart. In a study published in Nature Cardiovascular Research, researchers report that mice lacking the X-linked histone demethylase KDM6A in their hematopoietic compartment recover far more poorly from myocardial infarction than littermate controls, and that a neutralizing antibody against the inflammatory messenger IL-1β largely rescues the deficit. The work, led by Emmanouil G. Solomonidis and Sebastian Cremer together with Stefanie Dimmeler’s group in Frankfurt, links a common category of age-related blood mutations, known as clonal hematopoiesis, to a specific and druggable inflammatory mechanism in heart disease.

Clonal hematopoiesis occurs when blood cell clones carrying acquired mutations expand with age. It is an established independent risk factor for cardiovascular disease, and while mutations in DNMT3A and TET2 have been studied extensively, the cardiovascular consequences of KDM6A mutations remained unclear. KDM6A encodes a histone demethylase that removes repressive H3K27me2/3 marks and is commonly mutated in patients with heart failure. To test what this means for the injured heart, the team crossed mice carrying a floxed Kdm6a allele with Vav-iCre mice, producing animals in which the gene is deleted specifically in blood-forming cells, referred to as Kdm6aΔ-Haem mice.

At baseline, before any injury, the knockout animals looked deceptively normal. Ejection fraction, the standard measure of pumping efficiency, was indistinguishable between groups, and neither peripheral blood nor the uninjured heart showed an excess of mature myeloid cells. Beneath the surface, however, the bone marrow was already remodeled: granulocyte–monocyte progenitors were increased, LSK progenitor cells were reduced, and long-term hematopoietic stem cells trended upward. The authors interpret this as a primed or imbalanced progenitor state established by KDM6A loss before disease ever strikes.

After permanent ligation of the left anterior descending coronary artery, the picture changed dramatically. Twenty-eight days after infarction, Kdm6aΔ-Haem mice had significantly lower ejection fractions than controls, with a P value of 0.001, larger left ventricular scars, and elevated cardiac expression of the inflammatory genes Il1b and Nlrp3. Both sexes were affected, but the phenotype was more pronounced in females. Three days after infarction, the peak of inflammatory activity, female knockout mice developed monocytosis, and flow cytometry revealed increased infiltration of myeloid cells, particularly monocytes, into the ischemic myocardium. Circulating levels of matrix metalloproteinases MMP-8 and MMP-12, enzymes linked to adverse post-infarction remodeling, were also elevated, suggesting that KDM6A loss amplifies the systemic inflammatory and remodeling response to injury.

To understand what these recruited cells were actually doing, the researchers performed single-cell RNA sequencing and single-cell ATAC sequencing on CD45-positive immune cells isolated from infarcted hearts of female mice three days after MI. In CCR2-positive recruited macrophages, the most heavily expanded population, KDM6A deficiency produced a robust pro-inflammatory transcriptional response: 245 genes were upregulated and 34 downregulated compared with wild type. Upregulated genes included inflammatory mediators such as Il1b, Nlrp3, Cxcl2 and Saa3 alongside oxidative stress genes like Hif1a, Sod2 and Txnip. Chromatin accessibility analysis revealed more regions opening than closing in these macrophages, with increased accessibility linked to genes involved in inflammation and myeloid activation, including Il1a, Nfkb1, Cd74, Fos and Junb. Predicted transcription factor binding shifted toward the AP-1 and C/EBP families, while binding of the chromatin architectural protein CTCF diminished, hinting at disrupted chromatin architecture.

Neutrophils told a complementary story. Although most differentially accessible regions in these cells became less accessible upon KDM6A loss, a focused program of inflammatory and chemotactic genes was selectively upregulated, including Saa3, Junb, Ccl6, Cxcr1, Ccr1, S100a8, Sell and Cxcl2, with gene ontology enrichment for leukocyte chemotaxis and myeloid cell migration. Ly6c-high monocytes, the precursors of inflammatory macrophages, showed a metabolic shift toward glycolysis, with upregulation of Gapdh, Pkm, Aldoa and related genes, alongside increased Il1b, Ccr2 and oxidative stress markers. The authors conclude that KDM6A loss reprograms multiple myeloid lineages toward inflammatory, migratory and glycolytic states through lineage-specific chromatin rewiring that converges on shared AP-1 and C/EBP transcription factor programs.

The mouse findings translated to humans. The team profiled peripheral blood mononuclear cells from patients with ischemic heart failure with reduced ejection fraction, comparing those with KDM6A-driven clonal hematopoiesis to those without. The patient cohort was small, three patients without CH and two with KDM6A CH, a limitation the authors acknowledge explicitly. Still, patients carrying KDM6A mutations showed a marked expansion of classical CD14-positive monocytes, the subset most associated with inflammatory responses, and upregulation of genes involved in myeloid differentiation, immune regulation and leukocyte activation, including CCR2, NFKB1, FOS, JUN, CCL5, STAT5B and multiple Toll-like receptors, closely mirroring the signatures seen in the knockout mice. No co-occurring CH driver mutations were detected in these patients, ensuring the signal reflected KDM6A-driven clonal hematopoiesis alone.

Computational modeling of cell–cell communication predicted that KDM6A-mutant monocytes send increased outgoing signals toward cardiac fibroblasts and cardiomyocytes, with IL-6, CCL, HGF and MHC-I/II pathways among the top enriched circuits. To test this experimentally, the researchers silenced KDM6A in human monocyte-derived macrophages and exposed rat cardiomyocytes and human cardiac fibroblasts to the conditioned media. Cardiomyocytes treated with supernatant from KDM6A-deficient macrophages beat faster and grew larger, signs of hypertrophy, while fibroblasts produced more collagen type I alpha 1, indicating enhanced fibrotic activation. These in vitro results, each with three biological replicates per group, suggest that KDM6A-mutant monocytes can aberrantly reshape paracrine signaling in the cardiac microenvironment.

The therapeutic experiment delivered the study’s most striking result. Kdm6aΔ-Haem mice received a neutralizing anti-mouse IL-1β antibody starting two hours after infarction, followed by weekly injections for four weeks, while controls received isotype immunoglobulin. At day one after MI, both groups showed similar drops in ejection fraction, confirming comparable initial injury. By week two the curves diverged, and by week four the treated mice had significantly better ejection fractions, a benefit seen in both males and females. Compared with the original untreated dataset, IL-1β blockade produced a near-complete rescue of the cardiac functional phenotype, with ejection fraction approaching levels observed in wild-type animals. Notably, scar size and interstitial fibrosis did not differ between treated and control groups at the terminal timepoint, which the authors interpret to mean the improvement came from better function of the surviving myocardium and attenuation of pathological inflammatory signaling rather than reduced scar formation.

Bulk RNA sequencing of female knockout hearts at four weeks after MI supported this interpretation at the molecular level. Anti-IL-1β treatment downregulated inflammatory pathways including TNF, IL-6 and IFNγ signaling as well as apoptosis, while pathways related to oxidative phosphorylation and fatty acid metabolism increased and glycolytic signatures declined. Because adverse post-infarction remodeling is characterized by mitochondrial dysfunction and impaired substrate oxidation, this partial restoration of myocardial energy metabolism provides a molecular correlate of the improved pumping function. The authors note that their findings align with a follow-up analysis of the CANTOS randomized trial, which suggested that patients with infarction harboring TET2-driven clonal hematopoiesis may respond better to the IL-1β antibody canakinumab, and with the ongoing TECTONIC trial testing canakinumab in coronary artery disease patients with and without TET2 mutations. No KDM6A-specific anti-inflammatory clinical trial currently exists, but the authors argue that KDM6A-driven clonal hematopoiesis represents another genetically defined inflammatory subtype that could, in principle, be targeted with IL-1β-directed drugs such as canakinumab or anakinra. For now, the evidence remains confined to mice and a small human observational cohort, and the authors caution that clinical heterogeneity of KDM6A mutations, including missense variants whose functional consequences are not fully validated, complicates direct extrapolation. Whether IL-1β blockade benefits patients with KDM6A-driven clonal hematopoiesis after infarction will require dedicated clinical testing.

Subject of Research: Clonal hematopoiesis driven by KDM6A loss impairs cardiac recovery after myocardial infarction through inflammatory reprogramming of myeloid cells

Article Title: Hematopoietic loss of KDM6A impairs cardiac recovery in heart failure via epigenetic reprogramming of myeloid cells

Article References: Solomonidis, E. G., Hecker, D., Shumliakivska, M., Leberzammer, J., Fischer, A., Katschke, M.-T., Bendel, M., Korth, L., Schuhmacher, B., Rumpf, L., Glaser, S.-F., Schmitz, K., Luxán, G., Frenz, S., Pennoyer, P., Wilken-Schmitz, A., Rasper, T., Zimmermann, S., Ganß, F., … Cremer, S. (2026). Hematopoietic loss of KDM6A impairs cardiac recovery in heart failure via epigenetic reprogramming of myeloid cells. Nature Cardiovascular Research. https://doi.org/10.1038/s44161-026-00887-3

Image Credits: AI Generated

DOI: 10.1038/s44161-026-00887-3

Keywords: clonal hematopoiesis, KDM6A, myocardial infarction, heart failure, macrophages, neutrophils, IL-1β, epigenetics, monocytes, cardiac inflammation, myeloid cells, Hematopoietic