The phenomenon at the heart of the study is called intraplaque angiogenesis, the abnormal sprouting of microvessels within the wall of an atherosclerotic plaque. These tiny vessels are not benign plumbing. They leak, they recruit inflammatory cells, and they are widely regarded as a hallmark of plaques that are prone to rupture, the catastrophic event that triggers most heart attacks. What has remained murky is precisely how these vessels are driven to proliferate so aggressively. The Fudan team suspected the answer lay in a poorly explored dialogue between cellular metabolism and the epigenetic machinery that controls gene expression, a dialogue that has become one of the hottest frontiers in cardiovascular biology.

To find the culprit, the researchers combined bioinformatic screening of public gene expression datasets with profiling of actual atherosclerotic plaque specimens from patients and experiments in ApoE-knockout mice, a standard model of human atherosclerosis. Their search converged on TXNIP, a protein already known for its roles in oxidative stress and glucose metabolism, which they found to be significantly downregulated in the endothelial cells of unstable plaques. That pattern was consistent across human datasets and was confirmed by immunohistochemistry in both human plaque tissue and the aortas of mice fed a high-fat diet. In other words, wherever plaques turned dangerous, this protective protein seemed to vanish.

The mechanistic story that followed is where the study becomes genuinely striking. Using co-immunoprecipitation and targeted mutagenesis, the team showed that TXNIP binds directly to LC3, the central engine of autophagy, through a conserved LC3-interacting region known as a LIR motif. This binding allows TXNIP to escort hexokinase 2, or HK2, a glycolytic enzyme that sits on the outer mitochondrial membrane, into the mitophagy pathway, the cellular quality-control system that degrades damaged or surplus mitochondria and their associated proteins. Crucially, this degradation route is ubiquitin-independent, meaning TXNIP provides the targeting signal that conventional degradation pathways would otherwise lack.

When TXNIP is present, HK2 is kept in check. When TXNIP is absent, the brake comes off. HK2 accumulates on mitochondria, aerobic glycolysis accelerates, and the endothelial cells begin churning out lactate in large quantities. The team documented this shift using Seahorse metabolic flux analysis, which measures real-time proton efflux and oxygen consumption in living cells, and confirmed that a TXNIP mutant lacking its LIR motif lost the ability to restrain glycolysis. The result is a metabolic reprogramming of endothelial cells toward the Warburg-like behavior that characterizes rapidly proliferating cells, from tumor tissue to healing wounds.

But lactate, it turns out, is not merely metabolic exhaust. Over the past several years, biologists have recognized that lactate can chemically modify histones, the protein spools around which DNA is wound, through a process called lactylation. The Fudan group used CUT&Tag, a technique for mapping histone modifications across the genome, to show that the lactate flood in TXNIP-deficient cells enriched histone H3 lysine 9 lactylation, or H3K9la, specifically at the promoters of pro-angiogenic genes, including EGR3, PFN1, and MAPK3. This epigenetic mark loosens chromatin at those sites and drives the transcriptional programs that push endothelial cells into excessive proliferation, migration, and tube formation, the three behaviors that define pathological angiogenesis.

The in vivo experiments tied the whole axis together. The researchers knocked down TXNIP in ApoE-knockout mice and watched pathological intraplaque angiogenesis flourish and plaque stability deteriorate. Then came the therapeutic test: the team treated the animals with 2-deoxyglucose, or 2DG, a pharmacological inhibitor of hexokinase activity. Blocking HK2 effectively reversed the pathological vessel growth and improved plaque stability, even in mice lacking functional TXNIP. That result is significant because it demonstrates that HK2 sits downstream of TXNIP in the pathway and can be drugged independently of the upstream protein, offering a realistic pharmacological entry point.

What makes this work resonate beyond the vascular biology community is its framing of a complete chain of causation, from mitochondrial quality control through glycolytic reprogramming to epigenetic regulation of gene expression. The TXNIP-HK2-H3K9 lactylation axis, as the authors describe it, is a rare example of a single pathway connecting three of the most actively studied layers of cell biology. It also provides a mechanistic account of why cholesterol lowering alone cannot fully eliminate cardiovascular risk: plaque vulnerability is not only a lipid problem but a metabolic and epigenetic one, driven by processes that continue operating regardless of circulating LDL levels.

The therapeutic implications are tantalizing but still early. 2DG is a blunt instrument with known toxicity at higher doses, and translating mitophagy modulation or histone lactylation biology into safe human therapies will require years of additional work. Yet the study, funded by the Qingpu Hospital Affiliated to Fudan University and the Natural Science Foundation of Shanghai, points toward a non-cholesterol-dependent strategy for stabilizing vulnerable plaques, a goal that has eluded cardiology for a generation. If subsequent studies confirm the axis in larger cohorts and identify clinically suitable ways to intervene, the humble protein TXNIP, long a supporting character in stories of oxidative stress, may find itself recast as a central player in the battle against the world’s leading cause of death.

Subject of Research: The TXNIP-HK2-H3K9 lactylation axis linking mitophagy, glycolytic reprogramming, and epigenetic regulation in pathological angiogenesis and atherosclerosis.

Article Title: TXNIP restricts angiogenesis and atherosclerosis by targeting HK2 for mitophagic degradation to repress H3K9 lactylation

Article References: Li, L., Li, Y., Gao, H., Li, X., Gao, Y., Wu, W., & Fan, L. (2026). TXNIP restricts angiogenesis and atherosclerosis by targeting HK2 for mitophagic degradation to repress H3K9 lactylation. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08935-z

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08935-z

Keywords: TXNIP, HK2, mitophagy, H3K9 lactylation, angiogenesis, atherosclerosis, glycolysis, endothelial cells, plaque vulnerability, epigenetics, 2-deoxyglucose, cardiovascular disease