Heart attacks leave scars, and those scars are built by cardiac fibroblasts, the connective-tissue cells that swarm into damaged heart muscle and lay down stiff extracellular matrix. For decades these cells were treated as a single, uniform population of scar-builders. A new study published in the Journal of Translational Medicine dismantles that assumption. By integrating single-cell transcriptomic data from 73 samples and 236,748 individual cardiac fibroblasts, a team led by researchers at Renji Hospital, Shanghai Jiao Tong University School of Medicine, has produced one of the most detailed maps yet of fibroblast diversity in the healthy and diseased heart, and in doing so has uncovered a metabolic switch that appears to drive the scarring process itself.
The scale of the analysis is what sets the work apart. Single-cell RNA sequencing allows researchers to read the gene-expression fingerprint of thousands of individual cells, but combining datasets across laboratories, species of injury models and disease states is a formidable computational challenge. The team integrated, clustered and annotated the full complement of 236,748 fibroblasts, then applied a battery of analytical tools: differential gene-expression testing to compare healthy and fibrotic states, RNA velocity to infer the direction of cellular transitions, and non-negative matrix factorization, or NMF, to decompose the transcriptional landscape into coherent gene-expression programs that recur across cells and samples.
The clustering revealed eight major populations of cardiac fibroblasts, which the researchers organized into two broad lineages. One lineage, which they termed the steady-state lineage, corresponds to the fibroblasts that maintain the structural scaffolding of the healthy heart. The second, the fibrotic-state lineage, encompasses the activated cells that expand after injury and drive pathological remodeling. This two-lineage framework gives the field a vocabulary for what has been a confusing picture: fibroblasts in a scarred heart are not simply more numerous versions of their healthy counterparts, but a collection of distinct cell states with different gene-expression signatures and, presumably, different functions.
The NMF analysis added a second layer of insight by identifying twelve meta-programs, recurring modules of co-expressed genes that cut across individual cell clusters. Among these, one connection stood out: a strong association between PDK4-mediated glycolysis and extracellular matrix deposition. PDK4, pyruvate dehydrogenase kinase 4, is an enzyme that sits at a critical metabolic junction. By phosphorylating and inhibiting the pyruvate dehydrogenase complex, it blocks the flow of pyruvate into the mitochondria and the tricarboxylic acid cycle, pushing cells toward glycolysis and the conversion of pyruvate into lactate. The finding that this metabolic program tracks with matrix production in fibroblasts suggested, provocatively, that scarring might be fueled by a specific metabolic rewiring rather than being an incidental byproduct of inflammation.
To test that idea, the researchers turned to neonatal mouse primary cardiac fibroblasts grown in the laboratory, and deployed a series of mechanistic experiments: immunoblotting to measure protein levels, metabolomics to profile metabolic intermediates, co-immunoprecipitation to detect protein interactions, and ChIP-qPCR to examine binding to chromatin. The results traced a complete signaling chain. PDK4 activity promoted the production of lactate, the metabolic end-product that accumulates when cells rely heavily on glycolysis. That lactate, in turn, drove histone H3 lactylation, a chemical modification in which lactate-derived lactyl groups are attached to histone proteins, the spools around which DNA is wound.
Histone lactylation is a relatively recent addition to the catalog of epigenetic marks, first described as a link between cellular metabolism and gene regulation. By modifying histones, lactate does not merely accumulate as waste; it can actively reshape which genes are accessible for transcription. In the fibroblasts studied here, histone H3 lactylation increased the expression of TGF-β1, transforming growth factor beta 1, the canonical profibrotic cytokine that activates fibroblasts and stimulates them to synthesize collagen and other extracellular matrix components. The chain from PDK4 to glycolysis to lactate to histone lactylation to TGF-β1 to matrix synthesis therefore constitutes a coherent mechanistic model in which a metabolic enzyme acts upstream of the central molecular driver of fibrosis.
The critical question, of course, is whether interrupting this chain in a living animal can protect the heart. The researchers addressed this in a mouse model of myocardial infarction, in which the left anterior descending coronary artery is ligated to reproduce the ischemic injury of a human heart attack. Using immunofluorescence, flow cytometry and echocardiography, they evaluated what happened when PDK4 was inhibited. The outcome was striking: blocking PDK4 alleviated cardiac fibrosis and improved cardiac function in the infarcted mice. The team also demonstrated antifibrotic effects from fibroblast-targeted PDK4 knockdown delivered by AAV9, an adeno-associated viral vector commonly used to shuttle genetic payloads into cardiac tissue, which strengthens the case that the benefit arises from acting on fibroblasts specifically rather than through indirect effects on other cell types.
These findings arrive at a moment when the metabolic control of immune and stromal cells is one of the fastest-moving areas of biology. The idea that lactate is not a dead-end metabolite but a signaling molecule capable of reprogramming gene expression has transformed how researchers think about tumors, inflammation and now fibrosis. The new study extends that framework to the heart and, importantly, connects it to a druggable target. Kinases such as PDK4 are among the most tractable enzyme classes for pharmacological inhibition, and the demonstration that PDK4 inhibition improves cardiac function after infarction in mice provides a concrete proof of concept for a therapeutic strategy that would not merely slow scarring but intervene at its metabolic source.
The single-cell atlas itself is likely to prove as valuable as the mechanistic discovery. With eight fibroblast populations, two lineages and twelve meta-programs catalogued across healthy and diseased hearts, the resource gives investigators a reference against which future studies can map their own data, identify which fibroblast states expand in particular disease contexts and design experiments that target specific subpopulations rather than fibroblasts in general. The heterogeneity that once frustrated mechanistic understanding becomes, in this framing, a map of opportunities: if only a subset of fibroblast states drives pathological matrix deposition, therapies can be aimed at those states while sparing the cells that maintain normal cardiac architecture.
Cautions remain, as they always do in translational research. The mechanistic experiments were performed largely in neonatal mouse fibroblasts and mouse models of infarction, and human cardiac fibrosis involves additional layers of comorbidity, age and medication exposure that animal models only approximate. The published version of the paper was still subject to final editorial processing at the time of its early release, and the field will want to see the findings replicated independently. Nevertheless, the study delivers something rare: a large-scale descriptive atlas, a specific molecular mechanism and an in vivo therapeutic validation, all pointing at the same target. If PDK4-driven lactylation of histones proves as central to human cardiac fibrosis as it appears to be in the mouse, the enzyme that once seemed an obscure regulator of mitochondrial fuel choice may become one of the most closely watched targets in cardiovascular medicine.
Subject of Research: Fibroblast heterogeneity and PDK4-driven metabolic activation in cardiac fibrosis after myocardial infarction
Article Title: Single-cell transcriptomics landscape decodes cardiac fibroblast heterogeneity and reveals PDK4-driven profibrotic activation during cardiac fibrosis
Article References: Qi, X., Guo, S., Sun, Y., Zhang, L., Lu, K., Huang, X., Tang, R., Zhang, N., Yu, D., & Lian, F. (2026). Single-cell transcriptomics landscape decodes cardiac fibroblast heterogeneity and reveals PDK4-driven profibrotic activation during cardiac fibrosis. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08985-3
Image Credits: AI Generated
DOI: 10.1186/s12967-026-08985-3
Keywords: cardiac fibroblasts, single-cell transcriptomics, myocardial fibrosis, PDK4, histone lactylation, lactate, TGF-β1, extracellular matrix, myocardial infarction, glycolysis, AAV9, non-negative matrix factorization
Cite Scienmag News
APA
MLA
Chicago
Juliet Wilcox. (October 2, 2026). Single-Cell Atlas Reveals How the Metabolic Enzyme PDK4 Drives Scarring After Heart Attack. Scienmag. https://scienmag.com/single-cell-atlas-reveals-how-the-metabolic-enzyme-pdk4-drives-scarring-after-heart-attack/
Juliet Wilcox. “Single-Cell Atlas Reveals How the Metabolic Enzyme PDK4 Drives Scarring After Heart Attack.” Scienmag, 2 October 2026, https://scienmag.com/single-cell-atlas-reveals-how-the-metabolic-enzyme-pdk4-drives-scarring-after-heart-attack/. Accessed 2 October 2026.
Juliet Wilcox. “Single-Cell Atlas Reveals How the Metabolic Enzyme PDK4 Drives Scarring After Heart Attack.” Scienmag. October 2, 2026. https://scienmag.com/single-cell-atlas-reveals-how-the-metabolic-enzyme-pdk4-drives-scarring-after-heart-attack/
Copy citation
Download RIS
Tags: AAV9cardiac fibroblastscellular metabolic switches in cardiac fibroblastscomputational analysis of heart tissueextracellular matrixextracellular matrix remodeling after heart injuryfibroblast diversity in heart diseasefibroblast subpopulations in cardiac fibrosisgene expression profiling of heart scarsglycolysisheart attack-induced cardiac fibrosishistone lactylationlactatemyocardial fibrosismyocardial infarctionmyocardial scarring mechanismsnon-negative matrix factorizationPDK4PDK4 metabolic enzyme in heart repairrole of PDK4 in cardiac tissue remodelingsingle-cell RNA sequencing in cardiologysingle-cell transcriptomicssingle-cell transcriptomics of cardiac fibroblastsTGF-β1
