The Hippo pathway is a conserved signaling cascade involved in cell proliferation, apoptosis, organ size and tissue homeostasis. Its core kinases include mammalian Ste20-like kinases 1 and 2, also called MST1/2, and large tumor suppressor kinases 1 and 2, or LATS1/2. Scaffold proteins such as SAV1 and MOB1 help regulate this kinase system. When the pathway is active, LATS1/2 phosphorylates the transcriptional co-activators Yes-associated protein, or YAP, and transcriptional co-activator with PDZ-binding motif, or TAZ, leading to their cytoplasmic sequestration and degradation. When signaling is reduced, dephosphorylated YAP and TAZ enter the nucleus and work with TEA domain family transcription factors, including TEAD1–4, to drive genes associated with cell growth and survival. The pathway also responds to cell density, cell-cell contact, mechanical stress and extracellular-matrix stiffness.

CircRNAs are endogenous non-coding RNAs distinguished by their closed-loop structures and lack of conventional 5′ caps and 3′ poly(A) tails. Their unusual architecture makes them relatively resistant to exonuclease degradation, and their expression can vary by tissue, cell type and developmental stage. Most are generated through back-splicing, in which a downstream splice donor is joined to an upstream splice acceptor. Reverse-complementary sequences in flanking introns can bring the splice sites together, while RNA-binding proteins including Quaking and Muscleblind can promote circularization. CircRNAs can arise from exons, introns or exon-intron regions of precursor messenger RNA, and their biogenesis is controlled by both cis-acting sequences and trans-acting factors. Their regulatory functions include capturing microRNAs, binding RNA-binding proteins and influencing transcription or alternative splicing.

MicroRNA sequestration is the most extensively characterized mechanism connecting circRNAs with Hippo and YAP signaling. Because a circRNA can contain multiple binding sites for the same microRNA, it may act as a competing endogenous RNA and change the expression of messenger RNAs targeted by that microRNA. In gastric cancer, reduced circLARP4 acts as a miR-424 sponge, increases LATS1 expression and maintains Hippo activity that limits YAP activation; restoring circLARP4 inhibited malignant behaviors in experimental models. By contrast, YAP1 suppresses circRNA-000425, which allows miR-17 and miR-106b to exert stronger oncogenic effects in gastric cancer cells, creating a feedback loop between YAP activity and circRNA expression. In colorectal cancer, circ_0128846 captures miR-1184, increases AJUBA expression and promotes tumor growth. In lung adenocarcinoma, circ_16601 sequesters miR-5580-5p, preserving fibrinogen beta chain messenger RNA and enhancing YAP1-dependent signaling. In breast cancer, hsa_circ_0005273 captures miR-200a-3p and increases YAP1 expression, while hsa_circ_0091074 influences TAZ through miR-1297 in triple-negative breast cancer.

CircRNAs can also alter Hippo signaling through direct interactions with proteins. In hepatocellular carcinoma, circPAK1 is elevated in tissues and cell lines and is associated with larger tumors, lymph-node metastasis, advanced TNM stage and microvascular invasion. It binds 14-3-3ζ, reducing that protein’s ability to retain YAP in the cytoplasm and thereby increasing YAP nuclear translocation and downstream transcription. In nasopharyngeal carcinoma, circRILPL1 interacts with ROCK1, weakening LATS1-mediated inhibition of YAP, and cooperates with IPO7 to promote YAP nuclear transport. The reverse pattern appears in bladder cancer, where circXRN2 interacts with the SPOP degron region, protects LATS1 from degradation and supports YAP phosphorylation, thereby restricting nuclear YAP and YAP-dependent transcription. In hepatocellular carcinoma, hsa_circ_0098181 interacts with eukaryotic translation elongation factor 2 and restrains migration and invasion, while Quaking-5 promotes its biogenesis. These examples demonstrate that circRNAs can regulate the stability, localization or activity of core pathway components without necessarily changing microRNA availability.

Some circRNAs can also encode functional peptides that act directly on the Hippo pathway. In colorectal cancer, circPPP1R12A contains an open reading frame that produces a 73-amino-acid peptide called circPPP1R12A-73aa. Experimental studies found that this peptide increases YAP transcriptional activity, promotes colorectal cancer cell migration and invasion, and supports tumor growth and liver metastasis in patient-derived xenograft models. Another colorectal cancer circRNA, circ-YAP, encodes a 220-amino-acid peptide that directly interacts with LATS1, decreases YAP phosphorylation and increases nuclear YAP. This produces a positive feedback loop because YAP/TEAD activity promotes circ-YAP transcription, while the peptide produced from circ-YAP further strengthens YAP activity. Translation of circ-YAP depends on N6-methyladenosine, or m6A, recognized by YTHDF3, together with eIF4G2-mediated translation initiation. The relatively small number of established examples means that the prevalence of peptide-producing circRNAs in Hippo biology remains uncertain.

The chemical modification of circRNAs can regulate their localization and interactions even when translation is not involved. In chordoma, METTL3-mediated m6A modification promotes the cytoplasmic export of circTEAD1. Once exported, circTEAD1 forms a complex with IGF2BP3 and Yap1 messenger RNA, increasing the stability of the transcript and supporting YAP1-driven tumor development. CircTEAD1 is elevated in chordoma tissue, correlates with Yap1 expression and is associated with continuous disease-free survival in the reported analysis. These observations place circRNAs within regulatory networks involving methyltransferases, modification readers, RNA-binding proteins and translation machinery. The review emphasizes that the consequences of a circRNA alteration may therefore depend on where the molecule is located and which interacting proteins are present.

CircRNA-mediated Hippo regulation also extends to therapeutic resistance and communication among cells within the tumor microenvironment. Exosomes from lenvatinib-resistant hepatocellular carcinoma cells contain circPAK1; transferring these vesicles to sensitive recipient cells induced a resistant phenotype, associated with circPAK1’s interaction with 14-3-3ζ and enhanced YAP activation. In lung adenocarcinoma, extracellular-vesicle circ_16601 promoted recruitment and activation of myofibroblasts, contributing to a tumor-supporting stromal environment. A separate exosomal mechanism has been linked to immune checkpoint resistance in hepatocellular carcinoma: tumor-derived circUHRF1 enters natural killer cells and acts through the miR-449c-5p/TIM-3 axis to induce exhaustion. In gastric cancer, higher circLARP4 expression is associated with better responses to oxaliplatin and 5-fluorouracil-based chemotherapy, while its miR-424/LATS1 mechanism maintains Hippo activity. These findings show that circRNAs can participate in cell-autonomous signaling, intercellular transfer and treatment response, although the evidence comes from specific models rather than broad clinical comparisons.

Abnormal expression of Hippo-associated circRNAs is frequently associated with clinically observable features. Increased circADAMTS6 in esophageal squamous cell carcinoma is linked to advanced disease and lymph-node metastasis, while reduced circLARP4 in gastric cancer accompanies tumor progression and unfavorable outcomes. In hepatocellular carcinoma, elevated circPAK1 correlates with aggressive clinicopathological features, and elevated circUHRF1 is associated with tumor progression and immune-related treatment resistance. Reduced hsa_circ_0098181 correlates with unfavorable metastatic characteristics in hepatocellular carcinoma. In lung adenocarcinoma, circ_16601 and hsa_circ_0017109 are associated with features such as lymph-node metastasis, tumor classification, tumor size and overall survival. Because circRNAs can occur in tissues, plasma and extracellular vesicles, they are being investigated as diagnostic or prognostic markers. However, most studies use relatively small retrospective cohorts, and the review notes that standardized detection methods and large prospective studies are needed before routine clinical application.

The review’s central conclusion is that circRNA–Hippo interactions constitute a versatile regulatory layer whose effects depend on cancer type, molecular partners and disease stage. circPAK1, circPPP1R12A and circ-YAP can reinforce YAP activity, whereas circLARP4, hsa_circ_0098181 and circXRN2 can preserve Hippo-mediated inhibition. Much of the current evidence comes from cultured cancer cell lines, gain- or loss-of-function experiments and mouse xenografts, so it remains uncertain whether every reported interaction fully represents endogenous human tumor biology. Patient-derived organoids, genetically engineered models, single-cell sequencing, spatial transcriptomics, epitranscriptomic profiling and longitudinal clinical cohorts may help distinguish broadly conserved mechanisms from context-specific events. The review supports circRNAs as promising candidates for biomarker development and therapeutic investigation, but it does not establish clinical efficacy, diagnostic performance or a tested treatment strategy for people with cancer.

Subject of Research: Circular RNA regulation of Hippo-YAP/TAZ signaling in cancer

Article Title: Interactions between circular RNAs (circRNAs) and the Hippo pathway in cancer

Article References: Li, P., Huang, D., & Gu, X. (2026). Interactions between circular RNAs (circRNAs) and the Hippo pathway in cancer. Molecular Biology Reports, 53(1), Article 1532. https://doi.org/10.1007/s11033-026-12721-x

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12721-x

Keywords: circular RNAs, Hippo pathway, YAP, TAZ, cancer biology, microRNA sponging, peptide translation, m6A modification, drug resistance, cancer biomarkers, Interactions, between