Venous thromboembolism, a condition that encompasses deep vein thrombosis and pulmonary embolism, remains one of the leading causes of cardiovascular illness and death worldwide. Yet for all its clinical importance, the genetic architecture that determines who develops these dangerous clots, and who suffers recurrent episodes despite anticoagulant therapy, has remained only partially mapped. A new study published in the Journal of Translational Medicine now offers one of the most comprehensive attempts to close that gap, combining human genetics, proteomics, transcriptomics, single-cell sequencing and laboratory experiments to identify a single gene, GNS, as a striking candidate link between immunity, inflammation and venous clot formation.
The research, led by Wenjun Liu, Jijun Wu, Yuanhao Feng and colleagues across several Chinese hospitals and universities, began with a proteome-wide two-sample Mendelian randomization analysis. This statistical technique uses naturally occurring genetic variants that influence plasma protein levels as instruments to ask whether the proteins themselves causally influence disease risk, rather than merely correlating with it. The team drew on cis-acting protein quantitative trait loci, genetic variants located near the genes encoding the proteins, which reduces the chance of confounding by other traits. Their dataset spanned large population resources, including the FinnGen study, the UK Biobank and the deCODE study, and their outcomes of interest were venous thromboembolism overall, deep vein thrombosis and pulmonary embolism.
From the initial screen, twenty plasma proteins emerged with genome-wide significant Mendelian randomization associations with at least one venous thromboembolism-related outcome. But genetic associations of this kind can be misleading: a single genetic variant can influence a protein level and a disease through two separate but nearby signals, a phenomenon known as linkage. To resolve this, the researchers applied Bayesian colocalization, which tests whether the disease association and the protein-level association are driven by the same causal variant rather than by two different ones clustered in the same genomic region. Six proteins survived this stringent filter with strong colocalization evidence, measured as a posterior probability of the shared-signal hypothesis exceeding 0.8: BCAM, VWF, GNS, ABO, TUFT1 and CHST15.
Of these, GNS stood out. The gene encodes glucosamine N-acetyltransferase, a lysosomal enzyme, but its prioritization here rested on genetic, not biochemical, grounds. Crucially, GNS showed directionally consistent protective associations across all three outcomes: higher genetically predicted plasma levels of the protein were linked to lower risk of venous thromboembolism, deep vein thrombosis and pulmonary embolism alike. The consistency across related but clinically distinct outcomes, backed by shared genetic signals, suggested that the association was not a statistical artifact confined to a single dataset. The other five candidates, which include well-known clotting players such as von Willebrand factor and the ABO blood group glycosyltransferase, have established roles in thrombosis; GNS was the unexpected addition.
To place the genetic finding in a clinical context, the team turned to independent whole-blood transcriptomic cohorts stratified by recurrence history and by estimated recurrence risk. Among patients who had already experienced a venous thromboembolic event, those in the higher-risk and recurrence-associated categories showed reduced GNS expression in their blood. This alignment between the genetic data, where higher GNS appears protective, and the transcriptomic data, where lower GNS expression tracks with recurrence risk, lent plausibility to the signal. Notably, a phenome-wide association analysis, which scanned across a broad spectrum of diseases and traits for potential side effects or confounding associations, did not identify genome-wide significant phenotypic links for GNS, suggesting the signal is relatively specific to the thrombotic domain.
The next question was where, biologically, GNS might act. The researchers integrated single-cell RNA sequencing data from human thrombus tissue to determine which cell types express the gene. The answer was unexpected but immunologically suggestive: GNS expression localized predominantly to macrophages and fibroblast populations within the clot tissue. Macrophages are the immune system’s resident sentinels and cleanup cells, and they are increasingly recognized as active participants in thromboinflammation, the intertwined processes by which inflammation drives clotting and clotting amplifies inflammation. Fibroblasts contribute to the structural remodeling of the thrombus. A lysosomal enzyme expressed in macrophages could plausibly modulate how these cells polarize and signal during clot evolution.
To test whether GNS does more than merely co-occur with thromboinflammatory states, the investigators turned to the laboratory. Using human monocyte-derived macrophages isolated from healthy donors, they modulated GNS expression or activity in both directions and examined the consequences. Bidirectional manipulation of GNS was associated with measurable changes in macrophage polarization, the process by which these cells shift between pro-inflammatory and pro-resolving functional states, as well as altered inflammatory responses. In animal models of venous thrombosis, conducted under approved ethical protocols and in accordance with ARRIVE guidelines, similar effects on thromboinflammatory phenotypes were observed in vivo. The consistency of the in vitro and in vivo results strengthened the case that GNS participates in the immune regulation surrounding clot formation, although the authors are careful to note that the precise molecular mechanisms remain to be established.
The study’s design illustrates a broader shift in how complex cardiovascular diseases are being dissected. Rather than starting from a candidate pathway and testing it in isolation, the work moved in the opposite direction: from unbiased population-scale genetics to a prioritized protein, then outward through multi-layered biological evidence and finally into experimental perturbation. Each layer served as a filter, and the fact that GNS emerged from so many independent lines of evidence, genetic, transcriptomic, single-cell and experimental, is what distinguishes this from an ordinary association study. It also helps address the perennial weakness of Mendelian randomization, which can identify robust genetic associations but cannot by itself explain how a protein acts.
The clinical implications, while still distant, are tantalizing. Existing anticoagulants target the coagulation cascade directly and carry an inherent risk of bleeding. If the GNS-centered thromboinflammatory axis turns out to be modifiable, it could offer a therapeutic angle that modulates immune function within the thrombus rather than the clotting factors in the blood. The finding that GNS expression is reduced in recurrence-risk populations also raises the possibility of a biomarker use: measuring GNS-related signatures might, after validation, help stratify patients by their risk of recurrent events, informing decisions about the duration of anticoagulant therapy. However, the authors explicitly caution that further mechanistic and prospective studies are required before any biological or clinical relevance can be assumed.
The research team, spanning the Third Affiliated Hospital of Guangzhou Medical University, Nanfang Hospital of Southern Medical University and collaborating institutions, emphasizes that their results are associative at the human level and modulatory in the experimental setting, not yet causal in the full mechanistic sense. The work was supported by the Clinical Characteristic Technology Project of Guangzhou and made use of publicly available genome-wide association summary statistics, transcriptomic datasets and single-cell RNA sequencing resources, with patient tissue collection conducted under institutional ethics approval and written informed consent. As an open-access publication, the study, published on 08 September 2026, allows other groups to interrogate the datasets and attempt replication.
For now, the message is one of a promising new thread in the biology of venous thromboembolism. Thrombosis has long been viewed primarily through the lens of coagulation proteins, platelets and blood flow, but the GNS findings add weight to the growing appreciation that immune cells, particularly macrophages, are woven into the fabric of the thrombus itself. If subsequent studies confirm and mechanistically explain how this lysosomal gene shapes macrophage behavior within clots, a gene first flagged by statistical genetics could become the seed of a genuinely new therapeutic concept in one of medicine’s most common cardiovascular emergencies.
Subject of Research: Genetic and multi-omic identification of GNS-associated thromboinflammatory signatures in venous thromboembolism
Subject of Research: Medicine
Article Title: Genetic prioritization and multi-omic characterization of GNS-associated thromboinflammatory signatures in venous thromboembolism
Article References: Liu, W., Wu, J., Feng, Y., Peng, J., Bai, Y., Huang, B., Liu, H., & Liang, J. (2026). Genetic prioritization and multi-omic characterization of GNS-associated thromboinflammatory signatures in venous thromboembolism. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08939-9
Image Credits: AI Generated
DOI: 10.1186/s12967-026-08939-9
Keywords: GNS, Venous thromboembolism, Proteome-wide Mendelian randomization, Colocalization, Single-cell transcriptomics, Macrophage polarization, Thromboinflammation, Deep vein thrombosis, Pulmonary embolism, Cardiovascular genetics
Cite Scienmag News
APA MLA Chicago
Juliet Wilcox. (September 9, 2026). Multi-omic study reveals thromboinflammatory signatures driving venous thromboembolism. Scienmag. https://scienmag.com/multi-omic-study-reveals-thromboinflammatory-signatures-driving-venous-thromboembolism/
Juliet Wilcox. “Multi-omic study reveals thromboinflammatory signatures driving venous thromboembolism.” Scienmag, 9 September 2026, https://scienmag.com/multi-omic-study-reveals-thromboinflammatory-signatures-driving-venous-thromboembolism/. Accessed 9 September 2026.
Juliet Wilcox. “Multi-omic study reveals thromboinflammatory signatures driving venous thromboembolism.” Scienmag. September 9, 2026. https://scienmag.com/multi-omic-study-reveals-thromboinflammatory-signatures-driving-venous-thromboembolism/
Copy citation Download RIS
Tags: cross-population genetic studies of thromboembolismdeep vein thrombosis and pulmonary embolism molecular mechanismsgenetic architecture of recurrent venous clotsgenetic basis of deep vein thrombosis and pulmonary embolismGNS gene role in immunity and inflammationidentifying biomarkers for venimmune-inflammatory pathways in venous clotsintegration of humanintegrative genomics in cardiovascular diseaselaboratory experiments on thromboinflammationlaboratory validation of thromboinflammatory pathwaysMendelian randomization in cardiovascular diseaseMendelian randomization in thrombosis researchmulti-omic analysis of thromboinflammatory signaturesproteomics and transcriptomics in clot formationproteomics and transcriptomics in venous thromboembolismrole of GNS gene in clot formationsingle-cell sequencing in thrombosis researchsingle-cell sequencing in venous thromboembolismVenous thromboembolism genetic risk factorsVenous thromboembolism genetics
