Protein C occupies a uniquely important position in the biology of blood. Synthesised in the liver as an inactive zymogen, it circulates until the thrombin-thrombomodulin complex on endothelial cells cleaves it into activated protein C, or APC. Once activated, APC performs a remarkable triple duty: it proteolytically inactivates the clotting factors Va and VIIIa, damping down coagulation; it modulates inflammation through signalling pathways such as protease-activated receptor 1; and it exerts cytoprotective effects that help endothelial barriers survive injury. Because these functions depend on precisely choreographed interactions with other plasma and cell-surface proteins, the activity of protein C cannot be inferred from its concentration alone. Its structural conformation, its binding partners and the surrounding protein context all shape what it actually does in the bloodstream. This distinction matters enormously in paediatrics, where reference ranges for clotting proteins differ dramatically from adult norms and where children are neither small adults nor scaled-down versions of them.

The Melbourne team set out to interrogate this hidden dimension using co-immunoprecipitation, a technique that exploits the specificity of an antibody to fish a target protein, and everything physically bound to it, out of a complex biological mixture. Working with plasma samples pooled into age groups spanning neonates to adults, the researchers used an anti-protein C antibody to capture protein C along with its interacting partners. The captured proteins were then digested and identified by liquid chromatography coupled to tandem mass spectrometry, or LC-MS/MS, the workhorse technology of modern proteomics. By comparing which proteins were enriched in the protein C pull-downs relative to controls, and which were depleted, the team could build an age-resolved picture of the protein C interactome. Multivariate tools including principal component analysis and partial least squares discriminant analysis were then applied to test whether the co-precipitating proteomes segregated cleanly by age, while functional enrichment analysis asked what biological pathways the changing partners represented.

The answer was emphatic. The protein C co-precipitating proteome changed substantially across development, and the differences were most pronounced in the youngest cohorts. In neonates and infants, the enrichment of protein C itself in the pull-downs was the most striking feature, hinting that the molecular environment surrounding protein C in early life differs fundamentally from that of later childhood and adulthood. The number of identifiable interaction partners also fluctuated in unexpected ways: the largest complement of protein C associates appeared in the 11 to 16 year-old group, while the smallest was found in children aged 6 to 10. That the peak of interaction complexity occurs in adolescence, rather than adulthood, was among the more surprising patterns to emerge from the data, suggesting that the adolescent haemostatic system may be assembling a particularly rich set of protein C-centred complexes as it transitions toward the mature adult state.

What kinds of proteins were keeping company with protein C? The functional analyses pointed to age-specific networks spanning several biological domains. Beyond the expected haemostatic partners, the co-precipitating proteins included platelet-related proteins, components of the complement cascade, and immune-related proteins. This breadth is biologically provocative. It implies that protein C does not operate in a closed coagulation circuit but sits at the intersection of clotting, innate immunity and inflammatory signalling, and that the relative weight of these connections shifts with age. In early development, when the immune system is still calibrating its responses and the vascular system is maturing, protein C may be drawn into different protein communities than in adulthood, when the networks have settled into their mature configuration. The plasma proteome, in other words, is not a fixed backdrop but a shifting stage on which protein C plays a changing role.

Yet the study also uncovered a deeper layer of conservation beneath the surface variability. When the researchers performed pathway analysis on the age-specific interactomes, they found significant cross-over in the biological pathways represented across age ranges, even though the individual proteins identified as participating in those pathways differed. The same functional themes, coagulation regulation, platelet biology, complement activation and immune modulation, recurred from neonates to adults, but with different molecular actors filling the roles at each stage. This is a conceptually important finding: it suggests that the developing haemostatic system maintains its core functions by swapping components rather than by inventing new ones. The orchestra keeps playing the same symphony, but the instrumentation changes with age, and protein C sits at the centre of the score.

The technical achievement underlying these conclusions should not be understated. Co-immunoprecipitation from plasma is notoriously challenging, because abundant proteins such as albumin and immunoglobulins can swamp the signal, and transient or weak interactions may not survive the capture process. The team’s workflow, incorporating filter-assisted sample preparation and rigorous false discovery rate control in the mass spectrometric identifications, was designed to separate genuine co-precipitating partners from background. Pooling samples within age groups provided the statistical power needed to detect developmental trends, though it also means the study captures population-level patterns rather than individual variation. The raw data have been deposited in the ProteomeXchange repository under identifier PXD077907, allowing other researchers to reanalyse the interactomes and test the team’s interpretations independently, a transparency measure that strengthens confidence in the findings.

The clinical implications reach into one of the persistent puzzles of paediatric medicine: why children’s clotting systems behave so differently from adults’, and why therapies calibrated on adult data can misfire in young patients. Developmental haemostasis, the recognition that the coagulation system matures in a coordinated but non-uniform fashion, has reshaped how paediatric haematologists interpret clotting tests and dose anticoagulants. This study adds a new dimension to that framework. If protein C’s interaction partners change with age, then the anticoagulant, anti-inflammatory and cytoprotective functions of the protein C pathway may likewise vary across development, even when measured concentrations appear adequate. For conditions in which protein C plays a documented role, from sepsis-associated coagulopathy to thrombotic risk in children, an age-aware interactome could eventually inform more precise diagnostic thresholds and treatment strategies. The work was funded through an Investigator Initiated Research Agreement with Takeda Pharmaceutical, which the authors state had no role in the design, analysis or decision to publish.

There remain, of course, important caveats and open questions. Co-immunoprecipitation identifies association, not necessarily direct binding, and some detected partners may reflect indirect membership of larger complexes. Pooled samples smooth over individual differences, and the functional consequences of most of the observed interactions remain to be tested experimentally. Whether the adolescent peak in interaction partners reflects a genuine biological programme or a transient compositional state of plasma will require longitudinal follow-up. Nevertheless, the study opens a genuinely new window onto developmental haemostasis. It demonstrates that the protein C interactome is a moving target, remodelling itself across childhood and adolescence, and that the functional identity of a well-characterised anticoagulant protein cannot be fully understood without accounting for the company it keeps. As proteomic techniques grow more sensitive, the shifting social network of protein C may prove to be a key to understanding why the blood of a newborn, a ten-year-old and an adult are three different molecular worlds.

Subject of Research: Age-dependent changes in the protein-protein interaction partners of protein C in human plasma across development

Article Title: Age-dependent changes in protein interaction partners of protein C

Article References: Age-dependent changes in protein interaction partners of protein C. (n.d.). https://doi.org/10.1186/s12014-026-09638-9

Image Credits: AI Generated

DOI: 10.1186/s12014-026-09638-9

Keywords: protein C, co-immunoprecipitation, LC-MS/MS, developmental haemostasis, interactome, protein-protein interactions, plasma proteome, paediatric haematology, coagulation, complement, mass spectrometry, proteomics

The EVA1 family has been attracting growing attention in recent years, although most of the spotlight has fallen on its siblings. EVA1A has been implicated in autophagy and cell death pathways and has been described as a prognostic biomarker in glioma and hepatocellular carcinoma, while EVA1C has been linked to immune infiltration in low-grade gliomas. EVA1B, by contrast, has remained comparatively obscure, with only a handful of studies hinting at oncogenic roles in colorectal cancer and esophageal squamous carcinoma. The new work is among the first to examine the protein systematically in lung cancer, combining large-scale bioinformatic analysis with hands-on experiments in human tissue samples, cultured cell lines, and animal models.

The investigation began in silico. By mining publicly available genomic datasets, the researchers found that EVA1B and LRG1 display subtype-specific expression patterns across NSCLC, differing between lung adenocarcinoma and squamous cell carcinoma. Crucially, the computational analysis suggested that the two molecules are not merely co-expressed by coincidence but may be functionally connected, a hypothesis that shaped the entire experimental program that followed. This kind of integrative transcriptomic approach has become a standard starting point for biomarker discovery, but the team went well beyond correlation, seeking physical and functional evidence in biological systems.

The first line of experimental evidence came from paired clinical samples obtained from patients at Fujian Cancer Hospital, work that was approved by the institution’s ethics committee with written informed consent from all participants. Using immunohistochemistry and Western blot analysis, the researchers compared tumor tissue with matched paracancerous tissue taken from the same patients. The results were unambiguous: EVA1B protein levels were consistently higher in the tumor specimens. The pattern was mirrored in cell culture, where two NSCLC cell lines, A549 and H1299, both expressed more EVA1B than BEAS-2B, an immortalized but non-malignant bronchial epithelial cell line used as a normal reference. Together, these findings established that EVA1B overexpression is a reproducible feature of NSCLC cells rather than an artifact of any single model system.

With overexpression confirmed, the researchers asked what happens when EVA1B is removed. Using knockdown techniques to suppress the protein in the A549 and H1299 cells, they ran a battery of functional assays designed to probe the hallmarks of malignancy. Proliferation, measured with CCK-8 assays, slowed markedly. Migration, assessed with wound-healing experiments, and invasion, measured with Transwell chambers, were both significantly impaired. In other words, cells stripped of EVA1B became less able to multiply, less able to move, and less able to chew their way through surrounding tissue, the three capabilities that together make a cancer dangerous.

The molecular signature behind these behavioral changes pointed to a well-known process called the epithelial–mesenchymal transition, or EMT, and its malignant cousin, the EMT-like phenotype. During EMT, epithelial cancer cells shed their adhesive, stationary identity and acquire the motile, invasive character of mesenchymal cells, a transformation that is widely regarded as a critical step in metastasis. When EVA1B was knocked down, the cells showed increased expression of E-cadherin, the epithelial marker that keeps cells glued together, and decreased expression of N-cadherin and Vimentin, two classic mesenchymal markers. Levels of matrix metalloproteinase 11, an enzyme that helps degrade the extracellular matrix and clear a path for invading cells, also fell. The coordinated shift in this marker panel indicates that EVA1B helps maintain the invasive, EMT-like state of NSCLC cells.

Cell culture can only say so much, so the team turned to animal models. In xenograft experiments, in which human cancer cells are implanted into immunocompromised mice, tumors with EVA1B knocked down grew significantly more slowly than controls. More strikingly, in a lung metastasis model, silencing EVA1B reduced the metastatic burden in the animals’ lungs, providing direct in vivo evidence that the protein contributes to the spread of disease, not just to growth of the primary tumor. These experiments, approved under the hospital’s animal ethics protocols, bring the findings closer to physiological relevance and strengthen the case that EVA1B is a genuine driver of NSCLC aggressiveness rather than a passive passenger.

The mechanistic heart of the study lies in its connection to LRG1, a secreted glycoprotein whose name reflects a leucine-rich repeat structure and whose levels have long been monitored as an inflammatory marker in clinical blood tests. In recent years, LRG1 has been recast as an active participant in disease, particularly in pathological blood vessel formation and in tumor biology. Previous work by other groups has shown that LRG1 derived from NSCLC cells can be packaged into exosomes and promote angiogenesis through transforming growth factor beta signaling, and that exosomal LRG1 can drive NSCLC proliferation and metastasis by binding the extracellular matrix protein fibronectin. The Fujian team’s bioinformatic analysis had flagged a functional association between EVA1B and LRG1, and co-immunoprecipitation experiments provided supporting evidence: EVA1B and LRG1 were found together in the same immunoprecipitated protein complex, suggesting that the two molecules associate within cells or in closely associated molecular assemblies. The authors are careful in their language, describing an LRG1-associated mechanism rather than claiming a fully defined direct interaction, an appropriate degree of caution given that co-immunoprecipitation detects co-complex membership and does not by itself prove direct physical binding between two purified proteins.

Even with that caveat, the convergence of evidence is compelling. EVA1B is overexpressed in patient tumors and cancer cell lines; removing it suppresses proliferation, migration, invasion, EMT-like marker switching, xenograft growth, and metastatic colonization; and the protein appears to operate in concert with LRG1, a molecule already implicated in NSCLC progression through independent lines of research. If future work confirms and elaborates the EVA1B–LRG1 axis, it could open several therapeutic avenues. LRG1 is a secreted protein, and secreted targets are generally more druggable than intracellular ones, with antibody-based approaches already being explored against LRG1 in ocular disease and cancer settings. Alternatively, EVA1B itself, or the downstream EMT programs it sustains, could serve as a biomarker to identify patients at high risk of metastasis, or as a target for combination strategies alongside existing EGFR inhibitors and immunotherapies, both of which face well-documented resistance problems in NSCLC.

For now, the study stands as a textbook example of how modern cancer biology progresses from computational hint to clinical sample to mechanistic model. It also adds a new branch to the growing tree of EVA1 family research, positioning EVA1B alongside its better-known relatives as a molecule worth watching. Lung cancer kills roughly 1.8 million people each year, and every newly validated node in its signaling network represents another potential vulnerability. The Fujian team’s work, supported by the Natural Science Foundation of Fujian, suggests that a once-overlooked protein and an inflammation-associated glycoprotein may together form one such node, and that dismantling it could help deprive NSCLC of its migratory edge.

Subject of Research: The role of the EVA1B protein and its association with LRG1 in driving non-small cell lung cancer progression and metastasis.

Article Title: EVA1B promotes non-small cell lung cancer progression through an LRG1-associated mechanism

Article References: Huang, Z., Wang, H., Jiang, K., & Zhuang, W. (2026). EVA1B promotes non-small cell lung cancer progression through an LRG1-associated mechanism. 3 Biotech, 16(10), Article 444. https://doi.org/10.1007/s13205-026-05034-0

Image Credits: AI Generated

DOI: 10.1007/s13205-026-05034-0

Keywords: non-small cell lung cancer, EVA1B, LRG1, tumor progression, metastasis, EMT-like phenotype, epithelial-mesenchymal transition, biomarker, xenograft model, co-immunoprecipitation, cancer biology, A549 cells