Lactate modification drives mitochondrial damage and fibrosis in diabetic hearts

Lactate has long been dismissed as little more than metabolic exhaust, the acidic byproduct that burns in exercising muscles and lingers in the blood of critically ill patients. A new study argues that in the diabetic heart, this discarded molecule may be one of the driving forces behind a silent and dangerous transformation: the slow stiffening of cardiac tissue into scar. Reporting in Scientific Reports, a research team led by Bing Lian and Deliang Shen presents evidence that lactate chemically modifies proteins inside heart cells in ways that simultaneously cripple the cell’s mitochondrial cleanup machinery and switch on a powerful fibrosis-promoting signal. The work, which combines patient measurements, diabetic mice, and experiments in cultured cardiac fibroblasts, proposes a previously unrecognized molecular circuit linking sugar metabolism to structural heart damage.

The clinical stakes are considerable. Diabetic cardiomyopathy, a complication of diabetes that damages the heart muscle independently of clogged coronary arteries or high blood pressure, is thought to affect roughly 30 to 40 percent of diabetic patients even when blood sugar is well controlled. Its hallmark is myocardial fibrosis, the excessive deposition of collagen and other matrix proteins by overactivated fibroblasts, which gradually robs the ventricles of their ability to relax and contract. Targeted therapies remain scarce, largely because the molecular chain connecting metabolic derangement to structural remodeling has never been fully mapped. The new study set out to trace one possible link: lactylation, a recently discovered post-translational modification in which lactate-derived lactyl groups are chemically attached to lysine residues on histones and other functional proteins.

The clinical arm of the study enrolled 80 patients with type 2 diabetes who had confirmed myocardial fibrosis, defined as an extracellular volume fraction of 30 percent or higher on cardiac magnetic resonance imaging, along with 40 age- and sex-matched diabetic patients without fibrosis. All participants underwent T1 mapping on a 3.0 Tesla scanner before and after gadolinium contrast injection, together with echocardiography and fasting blood draws processed under strict cold-chain conditions. Serum lactate was markedly elevated in the fibrotic group, averaging 5.8 millimoles per liter compared with 3.2 in controls, a difference that persisted when the analysis was repeated excluding patients taking metformin, a drug known to raise lactate levels. Notably, all fibrotic patients had preserved ejection fractions, but they showed elevated NT-proBNP, reduced septal mitral annular velocities, and higher E/e’ ratios, a profile consistent with the heart failure with preserved ejection fraction phenotype that typifies early diabetic cardiomyopathy.

To probe causality rather than mere correlation, the researchers turned to a mouse model in which diabetes was induced by streptozotocin, a toxin that destroys insulin-producing beta cells. After eight weeks of hyperglycemia, the diabetic mice showed a 2.1-fold increase in myocardial lactate, and their hearts were visibly failing: ejection fraction fell from 68.5 percent to 48.2 percent, and collagen volume fraction, measured by Masson’s trichrome staining, surged from 3.2 percent to 15.7 percent, corroborated by a 2.9-fold rise in hydroxyproline, a biochemical proxy for total collagen. Beneath these structural changes lay a mitochondrial crisis. Protein levels of PINK1, the sentry kinase that tags damaged mitochondria for destruction, dropped by 62 percent; its partner Parkin fell by 57 percent; and the LC3-II/I ratio, a readout of autophagic activity, declined by 54 percent. ATP production fell by 45 percent while reactive oxygen species climbed 2.8-fold. Across individual animals, myocardial lactate correlated negatively with PINK1 and with ejection fraction, and positively with collagen deposition.

The first mechanistic thread the team pulled concerned HSP90, a molecular chaperone known to stabilize PINK1. In primary cardiac fibroblasts grown from neonatal mice and exposed to high glucose, adding lactate at 10 millimoles per liter deepened the mitophagy defect dramatically, cutting PINK1 by a further 75 percent and Parkin by 68 percent relative to high glucose alone. Co-immunoprecipitation assays showed that lactate treatment induced lactylation of HSP90 and reduced the physical interaction between HSP90 and PINK1 by 71 percent. The authors propose that when lactylated, HSP90 becomes a dysfunctional, dominant-negative chaperone: it can still clutch PINK1 but can no longer stabilize it properly, leaving damaged mitochondria untagged and unremoved. Intriguingly, silencing HSP90 with small interfering RNA partially restored mitophagy markers, an apparently paradoxical result that makes sense if removing the total pool eliminates the poisoned, lactylated fraction and allows remaining chaperones to take over.

The second thread ran through the genome. Using chromatin immunoprecipitation followed by quantitative PCR, the researchers found that histone H3K18 lactylation was enriched 4.2-fold at the promoter region of the TGF-β1 gene in fibroblasts exposed to high glucose plus lactate, compared with high glucose alone. This epigenetic mark, which generally flags active transcription, translated into a 3.8-fold surge in TGF-β1 messenger RNA and a 3.5-fold rise in TGF-β1 protein. Downstream, the fibrotic cascade ignited: phosphorylated Smad3, the principal effector of TGF-β signaling in fibroblasts, increased 3.1-fold, alongside coordinated upregulation of alpha-smooth muscle actin and both type I and type III collagen. Functionally, lactate-treated fibroblasts migrated 2.7-fold more readily through Transwell membranes, a behavioral hallmark of the fibroblast-to-myofibroblast transition that drives scarring.

Crucially, the two pathways appear not to run in parallel isolation but to feed each other. Impaired mitophagy allows dysfunctional mitochondria to accumulate, and their sustained reactive oxygen species output can itself enhance TGF-β1 expression and Smad3 phosphorylation. Conversely, TGF-β1 signaling is known to suppress mitochondrial biogenesis and function, potentially creating a self-reinforcing loop in which each arm of the pathology amplifies the other. Consistent with this integrated model, the authors observed that restoring mitophagy by knocking down HSP90 not only normalized mitochondrial markers but also suppressed TGF-β1 expression in lactate-treated fibroblasts, while pharmacological lowering of lactate improved both arms simultaneously.

That pharmacological test provided the study’s most translationally suggestive result. Diabetic mice treated with GSK2837808A, an inhibitor of lactate dehydrogenase A, the enzyme that catalyzes the final step of lactate production, showed a 49 percent reduction in myocardial lactate after six weeks of treatment. The molecular consequences followed: PINK1 rose 2.1-fold and Parkin 1.9-fold, collagen volume fraction fell by 48 percent, hydroxyproline dropped by 45 percent, and cardiac function measurably recovered, with ejection fraction improving by 35 percent and fractional shortening by 40 percent. In cultured fibroblasts, genetic silencing of either LDHA or HSP90 cut TGF-β1 transcription by 60 percent and fibroblast migration by 65 percent. The authors also point to existing compounds with lactate-modulating activity, including lactate transporter inhibitors and agents already known to influence lactylation, as candidate starting points for drug development.

The researchers are careful to flag the boundaries of their evidence. The streptozotocin model recapitulates features of type 1 diabetes with absolute insulin deficiency rather than the insulin resistance characteristic of the type 2 patients in their clinical cohort, and validation in diet-induced models is needed. Serum lactate reflects whole-body metabolism and cannot be attributed to the heart alone, and the team did not perform direct mitophagy flux assays, electron microscopy, or mass spectrometric confirmation of the specific lactylation site on HSP90, which was assigned based on literature precedent. The precise cell types in which each modification operates within the intact heart also remain unmapped. Still, the study reframes lactate from metabolic waste product to chemical messenger, and it sketches a coherent axis, from lactate through lactylation to failed mitochondrial quality control and activated fibrotic signaling, that may help explain why diabetic hearts scar. Whether blocking that axis can be done safely and specifically in patients is now the question that must be answered next.

Subject of Research: Lactylation-driven mitochondrial dysfunction and myocardial fibrosis in diabetic cardiomyopathy

Article Title: Lactylation-driven mitochondrial dysfunction promotes diabetic myocardial fibrosis via impairing mitophagy and activating TGF-β1

Article References: Lian, B., Li, J., Guo, Y., Li, X., Han, Y., Zhang, J., Lv, F., & Shen, D. (2026). Lactylation-driven mitochondrial dysfunction promotes diabetic myocardial fibrosis via impairing mitophagy and activating TGF-β1. Scientific Reports, 16(1), Article 34035. https://doi.org/10.1038/s41598-026-62184-0

Image Credits: AI Generated

DOI: 10.1038/s41598-026-62184-0

Keywords: diabetic cardiomyopathy, lactylation, lactate, mitophagy, PINK1, Parkin, HSP90, TGF-β1, H3K18, myocardial fibrosis, LDHA inhibition, mitochondrial dysfunction