Scientists Uncover Hidden Feedback Loop That Drives Colon Cancer Spread

Colon cancer remains one of the deadliest malignancies worldwide, with more than 1.9 million new cases and roughly 935,000 deaths recorded each year. While surgery, chemotherapy and targeted drugs can control many primary tumours, it is metastasis—the escape of cancer cells from the original tumour to distant organs such as the liver—that claims the majority of lives. New research published in the Journal of Cellular and Molecular Medicine has now revealed a previously underappreciated molecular circuit that helps colon cancer cells migrate, invade and seed new tumours, offering fresh clues about how doctors might one day intercept the spread before it starts.

The study, led by researchers at the First Medical Centre of the Chinese PLA General Hospital, focused on a secreted protein called SFRP2, a member of the secreted frizzled-related protein family best known for modulating Wnt signalling. Although SFRP2 has traditionally been described as a Wnt antagonist, accumulating evidence has painted a more complicated picture, linking it to tumour progression in glioblastoma, osteosarcoma and lung cancer. What remained unclear was exactly how SFRP2 might contribute to colon cancer metastasis, and whether the protein could be exploited as a biomarker or therapeutic target.

To answer those questions, the team began by mining publicly available genomic datasets. Comparing primary tumours with metastatic colon cancer tissues in the GEO dataset GSE40367, they found SFRP2 among the genes most significantly elevated in metastatic samples. Data from The Cancer Genome Atlas confirmed that SFRP2 expression was higher in colon tumours than in normal tissue, while Kaplan–Meier survival analysis revealed a sobering pattern: patients with high SFRP2 levels had worse overall survival, worse recurrence-free survival and worse post-progression survival. Immunohistochemistry and western blotting of patient specimens showed the strongest SFRP2 signal in metastatic tumour tissue, above even primary tumours and healthy tissue.

The researchers then turned to laboratory models. When they exposed colon cancer cells to TGF-β, a molecule that drives epithelial-mesenchymal transition, SFRP2 levels rose in step with the loss of the epithelial marker E-cadherin and the gain of mesenchymal markers N-cadherin and Vimentin. Using a clever in vitro invasion system, they separated LoVo colon cancer cells into highly metastatic populations that penetrated a coated membrane and poorly metastatic populations that did not; the aggressive cells carried far more SFRP2. Functional tests drove the point home: overexpressing SFRP2 in HCT116 cells boosted migration and invasion, while silencing the gene in LoVo cells suppressed both. Strikingly, SFRP2 manipulation had no effect on cell proliferation, suggesting the protein’s role is specifically in the physical business of spreading rather than growing.

The mechanistic core of the study lies in SFRP2’s interaction with Snai1, a transcription factor that orchestrates epithelial-mesenchymal transition by repressing epithelial genes and activating mesenchymal ones. Gene set enrichment analysis pointed toward focal adhesion and cell adhesion pathways, and Pearson correlation analysis showed SFRP2 and Snai1 levels rising together in patient data. Co-immunoprecipitation experiments confirmed that the two proteins physically bind to each other, both at endogenous levels and when tagged versions were introduced into cells. Most importantly, when the team blocked new protein synthesis with cycloheximide, Snai1 lingered far longer in SFRP2-overexpressing cells, demonstrating that SFRP2 protects Snai1 from degradation. Treating SFRP2-deficient cells with the proteasome inhibitor MG132 restored Snai1 levels, pinpointing the ubiquitin-proteasome system as the route of destruction.

That discovery led the investigators to USP11, a deubiquitinating enzyme predicted by the BioGRID database to interact with Snai1. Ubiquitin tags normally mark proteins for destruction by the 26S proteasome, and deubiquitinases like USP11 remove those tags, extending a protein’s lifespan. The team showed that USP11 binds Snai1 and that SFRP2 strengthens this partnership: in the presence of SFRP2, the USP11–Snai1 interaction intensified, whereas silencing SFRP2 weakened it. When the researchers inhibited USP11’s enzymatic activity with mitoxantrone, Snai1 stability collapsed, even in SFRP2-overexpressing cells. Conversely, forcing USP11 expression in SFRP2-knockdown cells rescued Snai1 protein levels. The picture that emerges is of SFRP2 as a facilitator that recruits USP11 to Snai1, allowing the deubiquitinase to strip away the ubiquitin marks that would otherwise doom Snai1 to the cellular recycling bin.

Perhaps the most striking finding is that the relationship runs in both directions. Scanning the SFRP2 promoter with the JASPAR tool revealed a binding motif for Snai1, and follow-up experiments showed that Snai1 directly binds the SFRP2 promoter to drive its transcription. Overexpressing Snai1 raised SFRP2 levels; knocking it down lowered them. Mutating the binding site abolished the effect in luciferase reporter assays, and chromatin immunoprecipitation confirmed the interaction on native DNA. In other words, SFRP2 stabilizes Snai1, and Snai1 in turn manufactures more SFRP2—a self-reinforcing positive feedback loop that could lock cancer cells into an invasive, metastatic state once it is triggered.

The team then traced what this loop does downstream. Gene set enrichment analysis flagged the PI3K/AKT/mTOR pathway, a central growth and survival cascade frequently hijacked by tumours. In cells, SFRP2 overexpression increased phosphorylation of PI3K, AKT and mTOR, and manipulating Snai1 reversed those changes. Treatment with the PI3K inhibitor LY294002 dampened the pathway without altering Snai1 or SFRP2 levels, indicating the cascade lies downstream of the SFRP2–Snai1 axis rather than upstream. Functional assays confirmed the consequences: silencing Snai1 partially blocked the migratory and invasive boost conferred by SFRP2, while restoring Snai1 in SFRP2-deficient cells rescued their aggressive behaviour, along with mesenchymal markers such as N-cadherin, Slug and Vimentin.

Crucially, the story held up in living animals. In mouse models of metastasis, SFRP2-overexpressing colon cancer cells produced dramatically more liver metastases than controls, an effect abolished when Snai1 was simultaneously knocked down. Tumours from these animals showed elevated Snai1 and phosphorylated PI3K, AKT and mTOR, while SFRP2-silenced cells yielded fewer metastases unless Snai1 was re-expressed. The authors acknowledge limitations, including the reliance on cell lines and xenografts that cannot fully reproduce the complexity of the human tumour microenvironment, and they caution that other pathways may contribute. Even so, the identification of a SFRP2–Snai1–USP11 feedback loop wired into PI3K/AKT/mTOR signalling gives researchers a concrete set of intervention points. Given that elevated SFRP2 already correlates with poor survival, the loop could serve both as a prognostic marker for aggressive disease and as a target for drugs designed to break the circuit before colon cancer cells ever leave home.

Subject of Research: The role of SFRP2, Snai1 and USP11 in a feedback loop promoting colon cancer metastasis via PI3K/AKT/mTOR signalling

Article Title: SFRP2 Potentiates Metastasis of Colon Cancer by Enhancing Snai1 Protein Stability via USP11

Article References: Kuang, Y., Liu, X., Ke, M., Chang, Z., Jia, B., Li, B., & Qian, H. (2026). SFRP2 Potentiates Metastasis of Colon Cancer by Enhancing Snai1 Protein Stability via USP11. Journal of Cellular and Molecular Medicine, 30(17), Article e71318. https://doi.org/10.1111/jcmm.71318

Image Credits: AI Generated

DOI: 10.1111/jcmm.71318

Keywords: colon cancer, metastasis, SFRP2, Snai1, USP11, deubiquitinase, EMT, PI3K/AKT/mTOR, protein stability, biomarker, TGF-beta, liver metastasis