Alveolar rhabdomyosarcoma (RMS), an aggressive pediatric soft tissue cancer, is driven by the oncogenic fusion transcription factor PAX3::FOXO1 (P3F) or PAX7::FOXO1. In a subset of fusion-positive (FP)-RMS cases, amplification of the MIR17HG locus leads to overexpression of the miR-17-92 cluster of microRNAs (miRNAs). However, miR-17-92 is also highly expressed in FP-RMS tumors lacking this amplification, suggesting alternative regulatory mechanisms. Here, we show that P3F and MYCN cooperatively drive miR-17-92 expression in FP-RMS. CRISPR/Cas9-mediated knockout of P3F or MYCN in FP-RMS cell lines substantially reduced miR-17-92 expression. Using a human myoblast line or low P3F FP-RMS variant with inducible P3F or MYCN expression, P3F or MYCN alone induces minimal to low miR-17-92 expression whereas introduction of both MYCN and P3F leads to robust activation of the miR-17-92 cluster and acquisition of oncogenic phenotypes. Chromatin immunoprecipitation sequencing (ChIP-seq) revealed a P3F binding motif located 1.84 Mb upstream of the MIR17HG promoter. CRISPR-mediated deletion of this region in the myoblast system resulted in marked reduction of miR-17-92 expression and impaired oncogenic transformation. Functional inhibition of mature miRNAs of this cluster in FP-RMS cells using miRNA-sponge constructs suppressed proliferation and transformation. In the myoblast model system, transduction studies with exogenous miR-17-92 or miRNA-sponge expression constructs indicated that miR-17-92 is necessary but not sufficient for oncogenic transformation. Implications: These findings establish a cooperative transcriptional axis in FP-RMS involving P3F and MYCN that activates MIR17HG through a distal regulatory element, thereby contributing to oncogenic behavior and uncovering a novel mechanistic vulnerability.

Breast cancer metastasis remains the leading cause of disease-related mortality, yet the regulatory mechanisms enabling tumor cells to adapt to distant tissue environments remain poorly defined. Here, we generated a multi-omic resource integrating transcriptomic (RNA-seq) and chromatin accessibility (ATAC-seq) profiles from estrogen receptor-positive primary breast tumors and matched liver and lung metastases obtained from the same patients, including diagnostic and postmortem specimens. This design enabled direct intra-patient comparisons of primary and metastatic regulatory landscapes. We identified extensive reprogramming of chromatin accessibility and gene expression in metastatic lesions relative to primary tumors, indicating widespread regulatory remodeling. Peak-to-gene correlation analysis uncovered thousands of metastasis-associated enhancer-gene linkages, with metastatic samples exhibiting significantly increased enhancer connectivity. These rewired enhancer networks preferentially targeted clinically relevant genes, including RIPK4 , ISYNA1, and RAB3D, which were associated with three or more enhancers in metastases but not in primary tumors. Furthermore, ATAC-seq footprinting combined with gene expression analysis revealed distinct, tissue-specific transcription factor activity profiles in liver and lung metastases, including ESR1, that were not explained by transcription factor expression levels alone. Together, these data provide a resource of metastasis-specific regulatory elements and their cognate transcription factors, revealing new regulatory vulnerabilities that may be therapeutically exploitable. Implications: These findings support a model in which breast cancer metastasis is driven by enhancer regulatory reprogramming, allowing tumor cells to adapt to a variety of microenvironments.

Neuroendocrine prostate cancer (NEPC) is an aggressive and therapy-resistant subtype of advanced prostate cancer characterized by poor prognosis and limited treatment options. Enzalutamide, an androgen receptor (AR) pathway inhibitor, is a standard second-line therapy for prostate adenocarcinoma (PrAd). However, enzalutamide resistance (EnzR) promotes not only neuroendocrine differentiation (NED) but also the induction of stemness-associated programs in PrAd cells. We identified the de novo DNA methyltransferase DNMT3B as a stemness-related gene upregulated in EnzR-PrAd cells and found that its expression is further significantly elevated in NEPC compared with PrAd. Our studies reveal that DNMT3B is a critical driver of NEPC development by coordinating the regulation of neuroendocrine and stemness transcriptional networks. Genetic inhibition of DNMT3B suppresses NEPC cell proliferation by impairing cell-cycle progression and triggering apoptosis both in vitro and in vivo. DNMT3B loss altered NED-associated, stemness, and epithelial-associated genes. We further uncover a reciprocal regulatory relationship between DNMT3B and the neuroendocrine transcriptional repressor REST. Using a human prostate cell transformation model that recapitulates NEPC evolution, we demonstrate that DNMT3B is required for both the initiation and maintenance of the NEPC phenotype. Finally, pharmacologic inhibition of DNMT3B with the selective inhibitor Nanaomycin A reduces NEPC tumor growth and suppresses NED and stemness marker expression without apparent acute toxicity in vivo. Implications: DNMT3B drives therapy-induced lineage plasticity in prostate cancer, and its inhibition suppresses NEPC growth and differentiation, highlighting it as a promising therapeutic target for treatment-resistant disease.

Metabolic plasticity driven by mitochondrial oxidative phosphorylation (OXPHOS) is increasingly recognized as a key determinant of therapeutic tolerance in hepatocellular carcinoma (HCC), but the upstream regulators that preserve electron transport chain stability during treatment remain poorly defined. In this study, we identified paraoxonase-1 (PON1) as a clinically relevant regulator of mitochondrial metabolism and lenvatinib response in HCC. PON1 was markedly upregulated in HCC and independently associated with poor overall and recurrence-free survival. Functionally, PON1 promoted tumor growth and conferred robust tolerance to lenvatinib. Mechanistically, PON1 directly interacted with and stabilized NDUFA4, a key component required for complex IV assembly, thereby maintaining mitochondrial membrane potential, complex IV integrity, and OXPHOS-dependent adenosine triphosphate production while limiting reactive oxygen species accumulation. Genetic silencing of PON1 or NDUFA4 impaired mitochondrial respiration, increased oxidative stress, and restored lenvatinib sensitivity in HCC cells and xenograft models. Structure-guided virtual screening identified the Food and Drug Administration-approved CFTR corrector lumacaftor as a potent modulator of PON1 that disrupted the PON1-NDUFA4 interaction and enhanced the antitumor efficacy of lenvatinib in vivo. These findings identify the PON1-NDUFA4 axis as a previously unrecognized metabolic vulnerability that sustains mitochondrial respiratory fitness and lenvatinib resistance in HCC. Targeting mitochondrial protein-stabilizing mechanisms such as PON1-NDUFA4 may offer a broadly applicable strategy for overcoming therapy resistance in liver cancer and other aggressive malignancies. Implications: These findings establish mitochondrial protein stabilization as an actionable therapeutic vulnerability and provide a rationale for combination strategies to overcome targeted therapy resistance in HCC.

Dihydrotestosterone (DHT), the principal driver of prostate cancer, is inactivated via reduction of its 3-keto group to a 3β-hydroxyl (3β-OH). 3β-Hydroxysteroid dehydrogenase type 1 (3β-HSD1), encoded by HSD3B1, catalyzes 3β-OH oxidation and is stabilized by an adrenal-permissive germline variant, thereby enhancing androgen biosynthesis. Here, we tested whether 3β-HSD1 maintains androgen receptor (AR) signaling by back-converting the inactive metabolite 3β-androstanediol (3β-diol) to DHT, and whether it similarly oxidizes the abiraterone metabolite 3β-OH-5α-abiraterone (3β-OH-5α-Abi) to 3-keto-5α-Abi, a weak AR agonist. Using HSD3B1-overexpression and knockout in vitro models, we quantified steroid interconversion by mass spectrometry and AR-responsive gene expression by qPCR and RNA sequencing following 3β-diol exposure. HSD3B1 overexpression efficiently enabled the conversion of 3β-diol to DHT and 3β-OH-5α-Abi to 3-keto-5α-Abi. In C4-2 prostate cancer cells, 3β-diol-derived DHT regeneration and subsequent AR gene induction were blocked by the 3β-HSD1 inhibitor trilostane or abolished by HSD3B1 knockout. Clinically, in chemotherapy-naïve metastatic castration-resistant prostate cancer (mCRPC) patients treated with abiraterone, apalutamide, and prednisone in the PANTHER trial, men harboring the adrenal-permissive HSD3B1 genotype showed significantly greater serum depletion of 3β-OH-5α-Abi. These findings demonstrate that 3β-HSD1-mediated steroid back-conversion can maintain intratumoral androgens, providing further mechanistic insight into the poor outcomes associated with the adrenal-permissive HSD3B1 genotype and highlighting 3β-HSD1 as a critical therapeutic target in prostate cancer. Implications: 3β-HSD1-mediated back-conversion of inactivated androgens (3β-diol→DHT) and abiraterone metabolites (3β-OH-5α-Abi→3-keto-5α-Abi) regenerates AR-active ligands, supporting sustained AR signaling and motivating 3β-HSD1 targeting and HSD3B1 genotype-informed treatment strategies in CRPC.

Oral squamous cell carcinoma (OSCC) is a lethal malignancy characterized by frequent metastasis. Utilizing a 4-NQO-induced murine model and single-cell RNA sequencing, we identified a distinct Collagen Type I Alpha 1 Chain (Col1a1+) epithelial subpopulation that is significantly expanded in OSCC, exhibits the highest epithelial-mesenchymal transition (EMT) signature, and correlates with poor patient survival. Functional studies demonstrated that Col1a1 knockdown in OSCC cell lines (HSC3, SCC25) attenuated TGF-β-induced EMT, impaired migration and invasion in vitro, and reduced tumor burden in a conditional knockout mouse model. Transcriptomic profiling following Col1a1 silencing revealed the suppression of TGF-β signaling and ECM-receptor interaction pathways. We further identified BMP1 as a key downstream effector, demonstrating a direct physical interaction between COL1A1 and BMP1, and showed that COL1A1 potentiates BMP1's proteolytic activity. Rescue experiments confirmed that BMP1 is essential for COL1A1-driven tumor growth, metastasis, and TGF-β pathway activation in vivo. Furthermore, we uncovered an immunomodulatory role for the COL1A1-BMP1 axis, whereby it promotes M2 macrophage polarization via TGF-β secretion to foster an immunosuppressive tumor microenvironment. This effect was reversible using either BMP1 knockdown or TGF-β neutralization. In conclusion, our study establishes a novel COL1A1-BMP1-TGF-β signaling cascade that drives OSCC progression by autonomously enhancing tumor cell malignancy and non-autonomously reprogramming the immune landscape, nominating this axis as a promising therapeutic target. Implications: These findings identify a novel COL1A1-BMP1 axis driving both OSCC malignancy and immunosuppression, offering potential diagnostic markers and therapeutic targets for the clinic.

ERG fusions occur in about 50% of prostate cancers (PCa) and represent early somatic events that co-initiate tumorigenesis. Yet, the molecular programs unleashed by ERG at disease onset remain elusive. Here, the inducible, temporally controlled expression of ERG enabled the accurate mapping of early ERG-driven processes associated with an apparent decrease in cell fitness across a panel of prostate cells. We observed that ERG triggers senescence and epithelial-mesenchymal transition (EMT), generating distinct cellular states in which EMT and senescence coexist in the same cell or occur independently. ERG-mediated activation of canonical and non-canonical TGF beta signaling differentially regulates EMT and senescence markers. Moreover, ERG drives context-dependent changes in cell identity and promotes an intermediate basal-luminal cell population with stem cell gene expression. Notably, p53 loss enhances ERG-driven expression of cell plasticity markers without affecting senescence. Transcriptomic investigation of ERG-expressing epithelial cells revealed activation of inflammatory signals. Conditioned medium experiments demonstrated paracrine-mediated inflammation in epithelial ERG-negative, stromal, and myeloid cells. Consistently, patient-derived expression data from 5 independent cohorts and 2,048 men showed activation of TNFA-NFKB and TGF beta signaling in ERG-rearranged tumors. Further, deconvolution analysis indicated a higher infiltration of inflammatory macrophages in ERG-positive tumors. Altogether, we uncovered a previously unrecognized triad of ERG-driven processes -dedifferentiation, senescence, and inflammation- that may underpin its oncogenic potential and shape PCa initiation and therapeutic response. Implications: Temporal control of ERG expression in prostate cells enabled accurate mapping of ERG-driven processes, identifying dedifferentiation, senescence, and inflammation as candidate contributors to ERG-dependent tumorigenesis and therapeutic response.

Desmoplastic small round cell tumors (DSRCTs) are rare, aggressive, fusion-driven sarcomas with poor outcomes despite intensive chemotherapy. Utilization of targeted therapies in DSRCT remains limited, underscoring the need for deeper characterization of patient tumors. To address this, we performed multi-omic profiling on nine patient-derived tumor biopsies from five patients enrolled in a precision oncology program. We consistently observed elevated mRNA and protein expression of human epidermal growth factor receptor 2 (HER2), androgen receptor (AR), and DNA damage response (DDR) markers, and characterized these molecular features using an integrated assay suite including bulk and single-cell RNA sequencing, protein profiling, immunohistochemistry, immune analyses, and functional homologous recombination deficiency (HRD) testing. We define a replication stress-associated DNA damage landscape and identify functional HRD in a subset of tumors lacking genomic HRD scar signatures. Single-cell analyses reveal intra- and intertumoral heterogeneity, while longitudinal sampling uncovers treatment-dependent shifts in expression and activity that may contribute to adaptive resistance. The immune microenvironment is characterized by dysfunctional T-cell states and sparse antigen-presenting cells. Collectively, these analyses delineate recurrent, biologically targetable features and patient-specific vulnerabilities, establishing a foundation for biomarker-guided therapeutic strategies in DSRCT. These findings support the investigation of rational combination approaches informed by sensitive detection methods and functional testing to address resistance in ultra-rare cancers. Implications: Integrative multi-omic profiling combined with functional testing in DSRCT reveals patient-specific vulnerabilities and biologically targetable receptor and DNA damage response dependencies, while defining immune states that may inform therapeutic response and rational combination strategies in this rare, fusion-driven cancer.

Stomach adenocarcinoma (STAD) is among the most prevalent malignant gastrointestinal tumors worldwide. Ferroptosis, a relatively new form of programmed cell death, has garnered significant attention for its role in cancer molecular mechanisms and therapeutic strategies. Therefore, it is particularly important to explore the mechanism of key molecules in ferroptosis on STAD. In our study, CUGBP Elav-like family member 1 (CELF2) was identified to be lowly expressed in STAD tumor tissues at both mRNA and protein levels. Next, CELF2 effectively inhibited the growth and migration of STAD cells both in vitro and in vivo. RNA-seq analysis of AGSCELF2 and AGSNC cells showed that the functions of differentially expressed genes were enriched in the glutathione metabolism and ferroptosis pathway. CELF2 increased the levels of key ferroptosis markers, including ROS, MDA, Fe[2+] and 4-HNE, while reducing GSH levels. Furthermore, CELF2 enhanced the susceptibility of STAD tumor cells to ferroptosis. Interestingly, the expression levels of CELF2 and Solute carrier family 7 member 11 (SLC7A11) were negatively correlated in RNA-seq data and TCGA database. RIP and RNA pull-down assays confirmed the interaction between CELF2 and SLC7A11. Overexpression of SLC7A11 reversed the effects of CELF2 on STAD growth and ferroptosis levels. Moreover, Ferroptosis inhibitor Ferrostatin-1 effectively rescued the inhibitory effect of CELF2 overexpression on proliferation of tumor cells. CELF2 is a tumor suppressor that promotes the level and susceptibility of ferroptosis in STAD cells by targeting SLC7A11. Implications: This study offers valuable molecular insights for the development of ferroptosis-targeted therapies for STAD.

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal malignancy primarily driven by oncogenic KRAS signaling. The splicing factor SRSF1 plays a key oncogenic role in PDAC through reciprocal cross-interactions with KRAS signaling. However, the mechanisms regulating SRSF1 protein stability remain poorly understood. Here, we identify the deubiquitinase USP39 as a critical regulator of SRSF1 stability. It interacts with SRSF1 in an RNA-independent manner and suppresses its ubiquitination. USP39 is upregulated in PDAC and correlates with poor patient prognosis. Functional analyses demonstrate that USP39 promotes PDAC cell progression, in part through stabilization of SRSF1. Mechanistically, MYC activates USP39 transcription through direct promoter binding. These findings define a MYC-USP39-SRSF1 regulatory axis that integrates transcriptional and post-translational mechanisms in PDAC and suggest USP39 as a potential therapeutic target. Implications: USP39 functions as a central regulator that integrates transcriptional and post-translational regulation in pancreatic cancer through the MYC-USP39-SRSF1 axis and represents a potential therapeutic target.

Despite the success of endocrine therapy (ET) in treating hormone receptor-positive breast cancer, a significant proportion of patients relapse during or after treatment, making ET resistance a major clinical challenge. Previously, we have shown that ET-resistant breast cancer cells exhibit reduced ceramide levels and an increased sensitivity to ceramide-induced cell death. In this study, we demonstrate that ceramides induce a distinct transcriptional reprogramming in ET-resistant cells, characterized by upregulation of endoplasmic reticulum stress (EnRS) pathways. Ceramide-induced EnRS is PERK-dependent and functionally linked to cell death in multiple models of ET resistance. Using a photoactivatable ceramide probe, we identify TRAM1 as a functionally important ceramide-interacting protein (CIP) in ET-resistant cells that correlates with worse relapse-free survival and a more aggressive breast cancer phenotype in patients with luminal breast cancer. Additionally, knockdown of TRAM1 phenocopies ceramide action in ET resistance, thereby suggesting its role in mediating ceramide-induced lethal actions in ET resistance. Together, our findings reveal that ET-resistant breast cancer cells are highly sensitive to PERK-mediated EnRS relative to ET-sensitive cells. Ceramides, likely via interactions with CIPs such as TRAM1, lead to PERK activation and consequential cell death in the ET-resistant breast cancer models. This sensitivity to ceramide-induced EnRS and cell death is a vulnerability that could be taken advantage of to treat ET-resistant breast cancer.

Breast cancers are molecularly heterogeneous, with subtype-specific differences in transcriptional programs, chromatin architecture, and therapeutic responses. Although poly(ADP-ribose) polymerase (PARP) 1 has been extensively studied in the context of DNA repair, emerging evidence implicates its catalytic activity in a broader set of cellular processes, including the regulation of gene expression. In this study, we employed a nicotinamide adenine dinucleotide (NAD+) analogue-sensitive PARP1 chemical genetics approach combined with mass spectrometry to map the ADP-ribosylated proteome across 6 human breast cancer cell lines representing luminal and basal/triple-negative subtypes. We identified thousands of PARP1 substrates and hundreds of Glu/Asp ADP-ribosylation (ADPRylation) sites, revealing both shared and subtype-specific modifications in cell lines maintained under basal growth conditions. Luminal-specific substrates were enriched in chromatin and transcriptional regulators, whereas basal-specific substrates were preferentially linked to translation and RNA processing, highlighting lineage-dependent PARP1 activity. Transcription factors emerged as major substrates, with TFAP2A serving as a proof of concept; it is selectively ADPRylated in luminal cells, and inhibition of PARP1-mediated ADPRylation modulates its promoter occupancy in a subtype-specific manner. Our data provide a new resource for studying subtype-specific PARP1-mediated ADPRylation in breast cancer cells. Collectively, our findings expand the conceptual framework for PARP1 function beyond DNA repair, offering mechanistic insights into subtype-specific gene regulation and potential determinants of PARP inhibitor sensitivity in breast cancer.