The gene CCM1 (also called KRIT1), best known for its role in the vascular disorder cerebral cavernous malformation, is upregulated specifically at the mRNA level in metastatic castration-resistant prostate cancer (mCRPC), where it functionally suppresses the RNA helicase DDX5 and thereby hyperactivizes YAP/TAZ signaling, which in turn potentiates androgen receptor (AR) signaling and metastatic behavior of prostate cancer cells. The mechanism that upregulates CCM1 and downregulates DDX5 in mCRPC is unknown (genomic rearrangement is excluded), and it is unresolved whether androgen-deprivation therapy itself induces CCM1, and whether CCM1-driven potentiation of AR signaling causally contributes to castration resistance and to the non-responsiveness of roughly 15–25% of CRPC patients to first-line abiraterone or enzalutamide. The problem is to determine how this axis is dysregulated in mCRPC and whether it causally drives resistance to androgen-directed therapy, in vitro and in vivo.

Background

Prostate cancer is the most common male malignancy, and its lethal phase is metastatic castration-resistant prostate cancer (mCRPC) — metastatic disease that progresses despite androgen deprivation therapy (ADT), the classical treatment that lowers male sex hormones (androgens). Most mCRPC remains driven by the androgen receptor (AR), the transcription factor that mediates androgen signals, so patients are treated with AR-axis-targeted agents such as abiraterone (which blocks androgen synthesis) and enzalutamide (an AR antagonist); however, about 15–25% of patients do not respond to these drugs at all, and the basis of this primary non-responsiveness is incompletely understood. YAP and TAZ are transcriptional co-activators downstream of the Hippo pathway: when Hippo signaling is off, YAP/TAZ enter the nucleus and bind the TEAD family of transcription factors to turn on genes promoting proliferation, migration, and invasion; their activity is conveniently read out with a TEAD-responsive luciferase reporter. Enhanced YAP/TAZ signaling had been correlated with extraprostatic extension of prostate cancer, but the upstream mechanism driving it was unclear. CCM1 (KRIT1) is a scaffold protein encoded by a gene mutated in cerebral cavernous malformation, a vascular disorder, and was not previously implicated in prostate cancer; DDX5 (p68) is a DEAD-box RNA helicase known to act as a transcriptional co-regulator, including for AR, and generally reported to be upregulated in several tumors. Park et al. (Cancers 2021; first posted as bioRxiv 2020.04.23.057778) analyzed expression across prostate cancer cohorts including the Prostate Cancer Transcriptome Atlas (PCTA), a large integrated transcriptomic dataset, and found that CCM1 expression is dramatically and uniquely increased only in mCRPC samples, correlating with worse metastasis-free and overall survival. Mechanistically, in multiple prostate cancer cell lines, silencing CCM1 reduced TEAD reporter activity, migration, invasion, and cadherin switching (a marker of epithelial–mesenchymal transition), while silencing DDX5 had the opposite effects; CCM1 suppressed DDX5 functionally (control of DDX5 tyrosine-593 phosphorylation), and DDX5 in turn restrained YAP/TAZ by interfering with the YAP–TEAD interaction, establishing a CCM1–DDX5–YAP/TAZ axis upstream of YAP/TAZ. CCM1 silencing reduced YAP/TAZ activity both with and without androgen (dihydrotestosterone) stimulation, suggesting the axis can potentiate both ligand-dependent and ligand-independent AR signaling. Whole-genome sequencing of 968 tumors found no copy-number or rearrangement basis for either the CCM1 upregulation or the observed DDX5 downregulation in mCRPC, and the PCTA cohort cannot separate treatment-naïve metastases from heavily treated mCRPC. The authors therefore posed the questions that define this problem: what non-genomic mechanisms (epigenetic changes or tumor-microenvironment cues) upregulate CCM1 in mCRPC; whether ADT itself induces CCM1 in metastatic tissue; and whether CCM1-mediated potentiation of AR signaling causally drives castration resistance and primary non-responsiveness to abiraterone/enzalutamide, in vivo and in patients. The accompanying DDX5 downregulation in mCRPC was reported not as a posed question but as an unexplained observation, with the note that the previously reported function of DDX5 in other cancer models may not be directly applicable to prostate cancer biology.

Problem Statement

Determine how the CCM1–DDX5–YAP/TAZ–AR axis is dysregulated in metastatic castration-resistant prostate cancer (mCRPC), and whether that dysregulation causally drives resistance to androgen-directed therapy. The single overarching objective is to establish the regulatory mechanism and causal disease role of this axis in mCRPC. All four parts below — the three questions posed by Park et al. and the DDX5 issue — are jointly required to resolve it. Park et al. (Cancers 2021, 13(5), 1125) posed the following open questions: (1) identify and demonstrate the molecular mechanism(s) that upregulate CCM1 in mCRPC — genomic rearrangement and copy-number change having been excluded — in particular epigenetic changes at the CCM1 locus or cues from the tumor microenvironment; (2) determine, with metastatic cohorts that distinguish treatment-naïve from heavily treated metastases, whether androgen-deprivation therapy itself induces CCM1 expression in metastatic prostate cancer tissue, as opposed to upregulation intrinsic to metastatic progression — a yes/no question completed by a definite answer either way; and (3) establish in vivo both (3a) whether elevated CCM1 potentiates ligand-dependent and ligand-independent androgen receptor (AR) signaling through YAP/TAZ, and (3b) whether that potentiation is causally responsible, not merely associated, for castration resistance and for the primary non-responsiveness of approximately 15–25% of CRPC patients to first-line abiraterone and enzalutamide. A related open issue arises from the same study's observations rather than from an explicit question of its authors: DDX5 expression was found downregulated in mCRPC samples without evidence of genomic rearrangement, and the mechanism responsible for this downregulation is unknown, as is whether and how it squares with subsequent reports that DDX5 is overexpressed in aggressive CRPC. This DDX5 issue is a required part of the objective on the same footing as the three questions above, not an optional aside.

References

  1. Park S, Lee HY, Kim J, Park H, Ju YS, Kim EG, Kim J. Cerebral Cavernous Malformation 1 Determines YAP/TAZ Signaling-Dependent Metastatic Hallmarks of Prostate Cancer Cells. Cancers 2021, 13(5), 1125. DOI: 10.3390/cancers13051125; https://doi.org/10.3390/cancers13051125 (published, peer-reviewed version; PMC7961486)
  2. Park S, Lee HY, Park H, Ju YS, Kim J, Kim EG, Kim J. Cerebral cavernous malformation 1 determines YAP/TAZ signaling dependent metastatic hallmarks of prostate cancer cells. bioRxiv preprint, posted 2020-04-24. DOI: 10.1101/2020.04.23.057778; https://doi.org/10.1101/2020.04.23.057778
  3. Seo E, Jee B, Chung JH, Song W, Sung HH, Jeon HG, Jeong BC, Seo SI, Jeon SS, Lee HM, Kang M. Repression of SLC22A3 by the AR-V7/YAP1/TAZ axis in enzalutamide-resistant castration-resistant prostate cancer. The FEBS Journal 2023, 290(6), 1645-1662. DOI: 10.1111/febs.16657; https://doi.org/10.1111/febs.16657
  4. Le TK, Cherif C, Omabe K, Paris C, Lannes F, Audebert S, et al. DDX5 mRNA-targeting antisense oligonucleotide as a new promising therapeutic in combating castration-resistant prostate cancer. Molecular Therapy 2023, 31(2), 471-486. DOI: 10.1016/j.ymthe.2022.08.005; https://doi.org/10.1016/j.ymthe.2022.08.005
  5. Olatunde D, De Benedetti A. TLK1>Nek1 Axis Promotes Nuclear Retention and Activation of YAP with Implications for Castration-Resistant Prostate Cancer. Cancers 2024, 16, 2918. DOI: 10.3390/cancers16162918; https://doi.org/10.3390/cancers16162918
  6. Bainbridge A, Walker S, Smith J, Patterson K, Dutt A, Ng YM, Thomas HD, Wilson L, McCullough B, Jones D, Maan A, Banks P, McCracken SR, Gaughan L, Robson CN, Coffey K. IKBKE activity enhances AR levels in advanced prostate cancer via modulation of the Hippo pathway. Nucleic Acids Research 2020, 48(10), 5366-5382. DOI: 10.1093/nar/gkaa271; https://doi.org/10.1093/nar/gkaa271
  7. Wasim S, Lee SY, Kim J. Complexities of Prostate Cancer. International Journal of Molecular Sciences 2022, 23(22), 14257. DOI: 10.3390/ijms232214257; https://doi.org/10.3390/ijms232214257
  8. Liu J, Zhao K, Wu S, Li C, You C, Wang J, Shu K, Lei T. The Dual Role of PDCD10 in Cancers: A Promising Therapeutic Target. Cancers 2022, 14(23), 5986. DOI: 10.3390/cancers14235986; https://doi.org/10.3390/cancers14235986
  9. Rödel CJ, Abdelilah-Seyfried S. CCM scaffolds at the frontier of biomechanics: their roles in vascular physiology and disease. Biological Chemistry 2026. DOI: 10.1515/hsz-2025-0246; https://doi.org/10.1515/hsz-2025-0246
  10. Paradisi L, Guazzi P, Macis M, Finetti F, Trezza A, De Paolo R, Roncetti M, Marshall JF, Poliseno L, Finetti F, Trabalzini L. Unveiling a Novel Molecular Interaction and Pro-Metastatic Signaling Cascades Driven by KRIT1. International Journal of Molecular Sciences 2026, 27(8), 3419. DOI: 10.3390/ijms27083419; https://doi.org/10.3390/ijms27083419