Hidden Protein From Circular RNA Shields Colon Cancer Cells From Fiery Death

Scientists in China have uncovered a molecular trick that colon cancer cells use to survive what should be an explosive death. A team led by researchers at China Pharmaceutical University and Nanjing Hospital of Chinese Medicine reports in Molecular Cancer that a circular RNA made from the PIAS1 gene carries hidden instructions for building a small protein, and that this protein helps tumor cells dodge pyroptosis, a violent, inflammatory form of cell suicide. The finding, published as an open-access research article, adds a new layer to the biology of circular RNAs and points toward fresh biomarkers and drug targets for one of the world’s most common malignancies.

Circular RNAs, or circRNAs, are unusual molecules. Unlike ordinary messenger RNAs, which are linear transcripts read straight from DNA, circRNAs are produced when the cellular splicing machinery joins a gene’s exons in a backward loop, covalently sealing the ends into a closed circle. For decades these loops were dismissed as transcriptional noise. More recently, however, biologists have realized that some circRNAs can be translated into proteins, often producing peptides that differ from anything encoded by the parent gene. The new study shows that this is exactly what happens with a circRNA derived from PIAS1, a gene previously linked to control of the NF-κB signaling pathway, a master regulator of inflammation and immunity.

The researchers began by mapping the molecule itself. Using polymerase chain reaction assays and RNase R tolerance tests, which exploit the fact that circular molecules resist exonuclease digestion far better than linear ones, they confirmed that circPIAS1 is generated by back-splicing of exons 3 through 6 of the linear PIAS1 transcript. Fluorescence in situ hybridization revealed where the circle resides inside cells, and the team then asked the critical question: does it actually code for a protein? Liquid chromatography tandem mass spectrometry, a technique that identifies proteins by shredding them into fragments and weighing the pieces, demonstrated that circPIAS1 is translated into a 108-amino-acid product the authors named circPIAS1-108aa. Immunofluorescence staining corroborated the protein’s existence and helped localize it.

What the team found next was striking. Immunohistochemistry on patient tissue samples showed that circPIAS1-108aa is significantly enriched in colon cancer tissue compared with healthy tissue, marking the protein as a potential player in the disease. Functional experiments followed. When the researchers manipulated levels of the protein in colon cancer cell lines, using small interfering RNAs to knock it down and plasmids to boost it, they observed that circPIAS1-108aa promoted tumor cell proliferation. The effect held up in living animals as well: after lentiviral transfection, nude mice carrying colon cancer cells with elevated levels of the protein developed tumors that grew more aggressively, confirming the pro-tumor activity in vivo.

The mechanistic heart of the paper concerns how circPIAS1-108aa protects cancer cells from pyroptosis. Pyroptosis is a form of regulated cell death driven by the NOD-like receptor family of innate immune sensors. When these sensors are activated, they trigger caspase enzymes that punch gasdermin pores into the cell membrane, causing the cell to swell, burst, and release a flood of inflammatory signals. In principle, this fiery demise is a defense against tumors, so cancer cells have strong incentives to suppress it. RNA sequencing and western blotting of cells with altered circPIAS1-108aa levels showed that the protein dampens both the NF-κB pathway and the NOD-like receptor signaling pathway, thereby holding pyroptosis in check.

To find out how the protein exerts this effect, the team used immunoprecipitation followed by mass spectrometry to fish for binding partners. The search pulled out TAB2, an adaptor protein best known for relaying signals from cell surface receptors to the NF-κB cascade. The interaction was then interrogated with AlphaFold3, the artificial intelligence system that predicts protein structures and interaction interfaces, which helped the researchers pinpoint the sites where the two molecules contact each other. The result was unexpected: circPIAS1-108aa does not chemically modify TAB2. Instead, it physically binds the adaptor and escorts it into the nucleus.

That nuclear translocation is the pivot of the whole mechanism. TAB2’s usual role is to help activate NF-κB, which then enters the nucleus to switch on inflammatory and survival genes, including components of the pyroptosis machinery. By dragging TAB2 into the nucleus in a way that short-circuits normal signaling, circPIAS1-108aa prevents the proper activation of NF-κB and the NOD-like receptor pathway downstream. The authors emphasize that this regulation is mediated by direct physical binding rather than by post-translational modification of TAB2, a distinction that matters because it suggests the protein acts as a trafficking partner rather than an enzyme, and it defines a specific molecular contact that could in principle be disrupted by drugs.

The study’s conclusions tie the threads together. CircPIAS1, transcribed from the PIAS1 gene, encodes circPIAS1-108aa, which substitutes for functions associated with the parental gene in suppressing pyroptosis of colon cancer cells. Because the protein is abundant in patient tumors and its levels track with tumor growth in animal models, the authors propose it as a candidate prognostic biomarker and a therapeutic target. If clinicians could measure circPIAS1-108aa in biopsy tissue, they might gauge how effectively a patient’s tumor has silenced its own inflammatory death program. Conversely, molecules designed to break the circPIAS1-108aa–TAB2 interaction could restore pyroptosis and turn the tumor’s shield into a vulnerability.

The work also carries broader implications for the circRNA field. It reinforces the emerging view that circular transcripts are not merely regulatory RNAs but can serve as a parallel protein-coding genome, with circRNA-derived proteins sometimes stepping in for their host genes. Here, a circle born from PIAS1, a gene already implicated in NF-κB control, yields a protein that modulates the very same pathway, hinting at an evolutionary logic in which back-splicing generates specialized regulators of a gene’s own circuitry. The methodological pipeline the team used, from RNase R validation and FISH through mass spectrometry, RNA sequencing, immunoprecipitation, AlphaFold3 prediction, and mouse xenografts, offers a template other labs can apply to the hundreds of circRNAs whose coding potential remains untested.

Caution is warranted, as with any preclinical study. The experiments were conducted in cell lines, patient cohorts of defined size, and immunodeficient mice, and the paper was released as an accepted manuscript subject to further editorial revision. Translating the findings into the clinic will require validating circPIAS1-108aa as a biomarker in large, independent patient populations and developing safe ways to interfere with the TAB2 interaction in human tumors. Still, the study opens an intriguing front in colon cancer research, a disease where poor prognosis remains common and new molecular handles are badly needed. By revealing how a small protein stitched together from a circular RNA molecule helps tumors evade one of the immune system’s most dramatic killing mechanisms, the researchers have turned a once-overlooked RNA loop into a lead worth following.

Subject of Research: A circPIAS1-encoded protein that regulates NF-κB signaling and pyroptosis in colon cancer

Article Title: PIAS1 circRNA-translated novel protein inhibits NF-κB signaling pathway by promoting the nuclear translocation of TAB2 and impedes pyroptosis in colon cancer

Article References: Wang, M.-Y., Wang, S.-M., Zang, X., Liu, C.-Y., Xiao, C.-M., Hou, H., Wu, X., Wang, Y., Kong, L.-Y., & Xia, Y.-Z. (2026). PIAS1 circRNA-translated novel protein inhibits NF-κB signaling pathway by promoting the nuclear translocation of TAB2 and impedes pyroptosis in colon cancer. Molecular Cancer. https://doi.org/10.1186/s12943-026-02803-6

Image Credits: AI Generated

DOI: 10.1186/s12943-026-02803-6

Keywords: colon cancer, circular RNA, circPIAS1-108aa, pyroptosis, NF-κB signaling, TAB2, NOD-like receptor pathway, non-coding RNA translation, biomarker, therapeutic target, AlphaFold3, tumor immunology