Milk Protein Gene CSN3 Emerges as a Driver of Colorectal Cancer Growth Through AKT Signaling

A protein better known for its role in milk has taken an unexpected turn onto the cancer research stage. CSN3, the gene encoding kappa-casein, a structural component of the casein micelles that give milk its characteristic protein organization, has now been implicated in the progression of colorectal cancer, one of the most common and lethal malignancies worldwide. In a study published in Molecular Biology Reports, a team of researchers from The Second People’s Hospital of Lianyungang in China reports that CSN3 is significantly elevated in colorectal tumor tissues, that its abundance correlates with lymph node metastasis, advanced TNM stage, and poor patient prognosis, and that the protein actively pushes colorectal cancer cells toward more aggressive behavior by switching on the PI3K/AKT/mTOR signaling cascade, one of the most frequently dysregulated growth pathways in human cancer.

The finding is striking precisely because CSN3 has historically been studied in an entirely different context. Kappa-casein is essential for the formation and stability of casein micelles in mammary epithelial cells, and the gene has long been a subject of interest in animal breeding and dairy science, where polymorphisms in CSN3 influence milk composition in cattle and water buffalo. Yet accumulating evidence has suggested that dysregulated casein gene expression may extend beyond lactation. Previous work has linked abnormal CSN3 activity to tumorigenesis in other tissues, including a recent report that CSN3 promotes gastric cancer progression and is associated with immune infiltration and inflammatory signaling. The new study extends that emerging picture to the lower gastrointestinal tract, where colorectal cancer remains a leading cause of cancer-related death globally.

To establish the clinical relevance of CSN3 in colorectal cancer, the research team, led by Cheng Chen and Tingting Tu of the hospital’s Department of Radiotherapy, with correspondence to Yingzhi Lu and Guanhong Huang, combined large-scale public data with direct tissue analysis. They interrogated gene expression datasets from the Gene Expression Omnibus and The Cancer Genome Atlas, and complemented these computational surveys with immunohistochemistry and quantitative reverse transcription polymerase chain reaction on patient samples. Across these independent approaches, CSN3 messenger RNA and protein levels were consistently higher in colorectal cancer tissues than in normal counterparts. Crucially, the elevated expression was not merely a molecular curiosity: patients whose tumors expressed high levels of CSN3 were more likely to have cancer that had spread to lymph nodes and to present at an advanced TNM stage, and their overall outcomes were poorer.

Correlation alone, however, does not establish causation, and this is where the study’s functional experiments become central. The researchers used gain- and loss-of-function approaches in a panel of colorectal cancer cell lines, overexpressing CSN3 in HT29 and SW480 cells while silencing it with small interfering RNA in HCT116 and SW620 cells. They then measured the consequences using a battery of standard assays: Cell Counting Kit-8 assays for proliferation, Annexin V/propidium iodide flow cytometry for apoptosis, wound-healing assays for migratory capacity, and Transwell assays for both migration and invasion. The results were unambiguous. When CSN3 was overexpressed, colorectal cancer cells proliferated faster, moved farther, invaded more aggressively, and resisted programmed cell death. When CSN3 was knocked down, all of these malignant behaviors were reversed. The team also validated their knockdown results with an independent siRNA sequence, a methodological safeguard that strengthens confidence that the observed effects were genuinely attributable to CSN3 rather than off-target artifacts.

A particularly important dimension of the work concerns epithelial-mesenchymal transition, or EMT, the developmental program that cancer cells hijack to become motile and invasive. During EMT, epithelial cells lose their adhesive, polarized character and acquire the deformability and migratory behavior of mesenchymal cells, a transformation widely regarded as a key step in metastatic dissemination. Western blot analysis in the study tracked the canonical molecular signature of this transition, showing changes in E-cadherin, the epithelial adhesion molecule, and in N-cadherin and vimentin, hallmark mesenchymal markers, in response to CSN3 manipulation. CSN3 overexpression promoted the EMT program, while its knockdown suppressed it, linking the milk protein gene directly to the machinery that colorectal tumors use to invade surrounding tissue and seed distant sites.

The mechanistic heart of the paper lies in the identification of the signaling pathway through which CSN3 exerts these effects. Using transcriptomic enrichment analysis of the GEO dataset GSE21510, the researchers identified CSN3-associated signaling pathways, and western blotting then confirmed that CSN3 activates the PI3K/AKT/mTOR axis. This pathway functions as a central growth-control circuit in cells: phosphoinositide 3-kinase generates lipid second messengers that recruit and activate AKT, also known as protein kinase B, which in turn activates mechanistic target of rapamycin and its downstream effectors p70S6K and 4EBP1, driving protein synthesis, cell growth, survival, and proliferation. In the study, CSN3 manipulation was accompanied by increased phosphorylation of PI3K, AKT, mTOR, p70S6K, and 4EBP1, indicating that the entire cascade was switched on in cells with high CSN3 and dialed down when CSN3 was removed.

To test whether this pathway activation was functionally responsible for CSN3’s tumor-promoting effects, the researchers turned to pharmacological inhibition. MK-2206, a selective AKT inhibitor, was applied to CSN3-overexpressing cells, and the results were decisive: the drug attenuated CSN3-induced proliferation, migration, and invasion, partially restored apoptosis, and reduced AKT phosphorylation. In other words, when the AKT node of the pathway was blocked, CSN3 lost much of its power to drive malignant behavior, satisfying a key criterion for causal mechanism. The PI3K/AKT/mTOR pathway is already recognized as a major therapeutic target in colorectal cancer, and this work positions CSN3 as an upstream contributor to its activation in a subset of tumors.

Perhaps the most intriguing mechanistic discovery, however, involves a second casein family member. Protein-protein interaction analysis using the STRING database, together with co-immunoprecipitation experiments, supported a physical association between CSN3 and CSN1S1, the gene encoding alpha-S1 casein. When the researchers knocked down CSN1S1, the tumor-promoting effects of CSN3 were attenuated: proliferation, migration, invasion, EMT, and AKT activation all diminished. This rescue-style experiment suggests that CSN1S1 contributes to CSN3-mediated AKT activation, hinting at a cooperative role for casein proteins within cancer cells that echoes, in a distorted form, their cooperative function in milk production. The idea that two milk structural proteins may physically interact to activate a growth signaling pathway inside a tumor cell is among the more unexpected findings to emerge from colorectal cancer research in recent years.

The study also touched on the tumor immune microenvironment, using the CIBERSORT algorithm to explore correlations between CSN3 and immune-infiltrating cell populations, an analysis that follows the authors’ earlier work linking CSN3 to immune infiltration in gastric cancer. While the abstract’s conclusions center on the AKT-dependent malignant phenotypes, the immune dimension adds a further layer of potential relevance, given the importance of immunological context in colorectal cancer prognosis and treatment response. The retrospective study was approved by the Ethics Committee of the Second People’s Hospital of Lianyungang, and the authors declare no competing interests.

As with any cell-line and tissue-based investigation, the findings will need to be extended through larger clinical cohorts and in vivo models before they translate into clinical practice, and the authors themselves frame CSN3 as having potential prognostic relevance rather than presenting it as an established therapeutic target. Even so, the study adds a genuinely novel name to the growing catalog of molecules that drive colorectal cancer through PI3K/AKT/mTOR signaling, a pathway already targeted by existing drug classes. If subsequent work confirms CSN3’s role and clarifies how the casein protein complex engages AKT signaling in tumor cells, a gene long confined to dairy science textbooks could find itself on the shortlist of biomarkers used to stratify colorectal cancer patients, and perhaps even as a point of therapeutic vulnerability in tumors that depend on it.

Subject of Research: The role of the kappa-casein gene CSN3 and the PI3K/AKT/mTOR pathway in colorectal cancer progression

Article Title: CSN3 promotes colorectal cancer cell progression through activation of the PI3K/AKT/mTOR pathway

Article References: Chen, C., Tu, T., Xu, H., Wang, G., Lu, Y., & Huang, G. (2026). CSN3 promotes colorectal cancer cell progression through activation of the PI3K/AKT/mTOR pathway. Molecular Biology Reports, 53(1), Article 1666. https://doi.org/10.1007/s11033-026-12825-4

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12825-4

Keywords: colorectal cancer, CSN3, kappa-casein, PI3K/AKT/mTOR pathway, AKT signaling, CSN1S1, epithelial-mesenchymal transition, metastasis, MK-2206, tumor biomarker, Molecular Biology Reports, cell signaling