The study, published in Frontiers in Molecular Biosciences, began with a straightforward clinical question: is RDM1 expressed differently in head and neck tumors than in healthy tissue? Drawing on both The Cancer Genome Atlas and tumor samples collected from patients at Guangdong Provincial People’s Hospital, the researchers found consistently elevated RDM1 levels in tumor tissue across three independent cohorts. Reverse transcription-quantitative PCR confirmed the pattern in patient-matched tumor and adjacent normal tissue. Survival analysis from the TCGA dataset hinted that patients with higher RDM1 expression fared worse, though the authors are careful to note that this trend did not reach statistical significance and should be treated as a hypothesis rather than a proven prognostic marker.

To understand what RDM1 actually does inside cancer cells, the team manipulated its expression in two head and neck squamous cell carcinoma cell lines, FaDu and Hep2, using both short hairpin RNA knockdown and lentiviral overexpression. The results were unambiguous. When RDM1 was removed, cell viability plummeted in MTT and image-based proliferation assays, and apoptosis surged, as measured by Annexin V staining and caspase-3/7 activity. When RDM1 was overexpressed, the opposite happened: cells proliferated faster and resisted programmed cell death. In other words, RDM1 behaves as an oncogene in this cancer type, consistent with its known tumor-promoting roles in lung, breast, ovarian, and thyroid cancers, though it acts as a suppressor in liver cancer, a context-dependent duality that has puzzled researchers for years.

Because RDM1 contains a RAD52 motif and an RNA recognition motif, its influence on DNA repair seemed plausible from the start. But the protein is predominantly cytoplasmic, while the DNA repair proteins it affects operate in the nucleus, raising the question of how a cytoplasmic protein could regulate nuclear processes. Transcriptome profiling using Affymetrix microarrays on RDM1 knockdown cells, followed by Ingenuity Pathway Analysis, revealed that DNA damage response and cell-cycle regulation pathways were among the most enriched when RDM1 was removed. Western blot experiments then showed that RDM1 knockdown decreased RAD52 protein levels while increasing TP53, the master tumor suppressor. It also raised CDKN1A, a cell-cycle inhibitor, while lowering CDK6, a kinase that drives DNA synthesis during the S phase. The downstream effector EZH2 was suppressed as well. Together, these findings suggest that RDM1 modulates the RAD52/TP53 axis to regulate DNA damage signaling and the CDKN1A/CDK6 axis to control cell-cycle progression, effectively acting as a cytoplasm-to-nucleus signal relay.

The real surprise came when the researchers turned their attention to what RDM1 does in the cytoplasm. KEGG pathway analysis of the transcriptomic data pointed to MAPK signaling, protein processing in the endoplasmic reticulum, and cell-cycle regulation as the top affected pathways. Subsequent experiments confirmed that RDM1 knockdown induced ER stress, as evidenced by increased expression of molecular chaperones. The team also found that the heme biosynthesis pathway, essential for cytochrome C and oxidative phosphorylation in mitochondria, was significantly inhibited when RDM1 was removed. These observations suggested that RDM1 might be involved in the cellular antioxidant response, the system that keeps reactive oxygen species in check. When the researchers measured ROS levels directly using a DCFDA assay, they found that RDM1 knockdown increased ROS while overexpression suppressed it, confirming that RDM1 helps cancer cells maintain a low-oxidative-stress environment.

To identify the molecular mechanism behind this antioxidant effect, the researchers examined NRF2, the transcription factor that orchestrates the antioxidant response, and its effector BCL2, a critical regulator of apoptosis. Western blot analysis showed that RDM1 knockdown dramatically reduced NRF2 and BCL2 protein levels, but crucially, their mRNA levels remained unchanged. This pointed to post-translational regulation, meaning RDM1 stabilizes NRF2 protein rather than activating its transcription. Rescue experiments reinforced the link: overexpressing NRF2 and BCL2 in RDM1 knockdown cells restored cell viability and reduced apoptosis, demonstrating that RDM1’s tumor-promoting function depends on the NRF2/BCL2 axis. Transcriptomic re-analysis also revealed that canonical NRF2 target genes, including BRCA1, OGG1, MUTYH, and MSH6, were downregulated upon RDM1 knockdown, consistent with reduced NRF2 transcriptional activity.

The next question was how RDM1 stabilizes NRF2. Using two complementary approaches, immunoprecipitation followed by mass spectrometry and a human proteome microarray, the researchers screened for proteins that physically interact with RDM1. After integrating the datasets and filtering for candidates associated with mitochondria or tumor proliferation, they narrowed the list to 57 key proteins. Among them, KEAP1, the E3 ubiquitin ligase that normally binds NRF2 and targets it for proteasomal degradation, showed the highest signal-to-noise ratio on the proteome microarray. Co-immunoprecipitation confirmed the RDM1/KEAP1 complex in head and neck cancer cells. Interestingly, RDM1 did not interact with RAD52, despite containing a RAD52 motif, suggesting that its primary cytoplasmic partner is KEAP1 rather than the nuclear repair machinery. Western blot analysis showed that RDM1 manipulation did not affect KEAP1 protein levels, consistent with a model in which RDM1 competes with NRF2 for KEAP1 binding, thereby preventing NRF2 ubiquitination and degradation. The authors acknowledge that direct competitive binding assays, cycloheximide-chase experiments, and NRF2 ubiquitination assays will be needed to definitively confirm this mechanism.

To validate these findings in living organisms, the team established a cell-derived xenograft model in nude mice using FaDu cells with stable RDM1 knockdown. Tumors in the RDM1 knockdown group grew significantly slower than controls, and immunohistochemistry revealed a marked decrease in NRF2 protein within the knockdown tumors. Notably, RDM1 knockdown had no impact on mouse body weight, suggesting that the protein’s tumor-promoting function can be disrupted without obvious systemic toxicity. The authors emphasize that an in vivo NRF2 rescue experiment will be necessary to definitively establish that the tumor-suppressive effect of RDM1 knockdown is mediated through NRF2, but the correlation between RDM1 loss, reduced NRF2, and slowed tumor growth is compelling.

What makes this study particularly significant is the crosstalk it reveals between the antioxidant pathway and the DNA damage response. These are not independent systems. First, NRF2-mediated ROS suppression indirectly reduces oxidative DNA damage and checkpoint activation. Second, NRF2 may transcriptionally regulate DNA repair genes such as RAD51 and RAD52, consistent with the transcriptomic changes observed upon RDM1 knockdown. Third, the NRF2 target BCL2 modulates both apoptosis and cell-cycle progression. This integrated model suggests that RDM1, by stabilizing NRF2, simultaneously influences antioxidant defense, DNA repair capacity, and cell-cycle control, three processes that tumors exploit to grow and resist therapy.

From a translational standpoint, the RDM1-KEAP1-NRF2 axis offers a promising therapeutic target. NRF2 hyperactivation is known to drive cisplatin resistance and radioresistance in head and neck cancer, and RDM1, acting upstream of NRF2 through its interaction with KEAP1, represents a viable point of intervention to suppress NRF2 signaling and sensitize tumors to chemotherapy and radiation. Because RDM1 is highly expressed in tumors but relatively low in most normal tissues, the therapeutic window appears favorable. The authors call for future preclinical studies combining RDM1 inhibition with standard-of-care therapy and for the development of RDM1-KEAP1 disruptors. They also stress that RDM1’s prognostic value must be validated in larger, well-characterized independent cohorts before it can be considered a definitive biomarker. For now, the study establishes RDM1 as an oncogenic driver in head and neck squamous cell carcinoma and reveals a novel molecular axis that links DNA repair, cell-cycle control, and antioxidant defense, offering a new direction for tackling one of medicine’s most challenging cancers.

Subject of Research: RDM1 stabilizes NRF2 via KEAP1 interaction to drive head and neck squamous cell carcinoma progression

Article Title: Unraveling the role of RDM1 in head and neck squamous cell carcinoma: insights into DDR regulation and antioxidant response

Article References: Wang, F., Wu, J., Wang, W., Zhao, R., Mao, Y., Wu, F., Li, Y., Ge, P., Zhu, W., Yan, X., Zheng, Y., & Lu, Z. (2026). Unraveling the role of RDM1 in head and neck squamous cell carcinoma: insights into DDR regulation and antioxidant response. Frontiers in Molecular Biosciences, 13, Article 1944888. https://doi.org/10.3389/fmolb.2026.1944888

Image Credits: AI Generated

DOI: 10.3389/fmolb.2026.1944888

Keywords: RDM1, head and neck squamous cell carcinoma, NRF2, KEAP1, DNA damage response, antioxidant response, oncogene, cancer therapy, Unraveling, role, head, neck