Dual FOXA1 Mutations Drive Prostate Cancer or Therapy Resistance via Opposite Mechanisms
Knock-in mouse models reveal two FOXA1 mutation classes divergently fuel prostate tumor formation or castration-resistant progression.
Summary
FOXA1 is mutated in 10–40% of prostate cancers globally, yet its in vivo oncogenic roles were poorly understood. Researchers at the University of Michigan generated knock-in mouse models carrying distinct human FOXA1 mutation classes. Class 1 mutations, combined with p53 loss, drove AR-positive invasive adenocarcinomas through co-activation of mTORC1/2 and oncogenic AR signaling via chimeric AR-half enhancers. Class 2 mutations, by contrast, reprogrammed differentiated luminal cells into a progenitor-like state by activating KLF5 and AP-1 neo-enhancer circuits, enabling tumor cell survival under castrate androgen levels. These findings establish FOXA1 as a multifaceted oncogene whose distinct mutational classes divergently evolve—one driving tumor initiation, the other enabling therapy-resistant progression.
Detailed Summary
Prostate cancer is predominantly a transcription factor-driven disease, with FOXA1—a pioneer factor essential for androgen receptor (AR) activity in prostate epithelium—mutated in 10–40% of cases in Western populations and over 40% in Asian cohorts. Despite this clinical prevalence, no in vivo model had causally linked specific FOXA1 mutations to cancer biology. This study addresses that gap with rigorous genetic, transcriptomic, and epigenomic approaches.
Researchers engineered knock-in mouse models at the Rosa26 locus carrying either Class 1 (R265–71del, Wing 2 forkhead domain missense/indel) or Class 2 (C-terminal frameshift) human FOXA1 mutant transgenes, expressed specifically in prostate epithelia via Pb-Cre4. Class 1 mice crossed with Trp53-floxed animals developed high-grade invasive adenocarcinomas with 30–40% Ki67-positive proliferating cells, stromal invasion, and full AR/CK8 luminal marker expression—closely recapitulating human primary prostate cancer. Single-cell RNA-sequencing confirmed hyperproliferative transcriptional programs, MYC upregulation, NKX3.1 loss, and parallel co-activation of AR and mTORC1/2 signaling. Mechanistically, Class 1 mutants were shown to create chimeric AR-half enhancers—novel regulatory elements formed by juxtaposing FOXA1 motifs with AR binding sites—that drive oncogenic AR transcriptional output, explaining their functional concordance with PI3K/PTEN-loss tumors in human cohorts.
Class 2 mutations told a strikingly different story. Rather than driving overt carcinogenesis, Class 2 mutants induced intra-luminal cellular plasticity, reprogramming mature luminal cells into a progenitor-like state. Multi-omic analyses of prostate organoids and tissues revealed that Class 2 mutants gain broad chromatin occupancy, activating neo-enhancer circuits centered on KLF5 and AP-1 transcription factor networks. This reprogramming enabled luminal cells to survive and proliferate under castrate androgen conditions—a hallmark of castration-resistant prostate cancer (CRPC). Importantly, Class 2 mutations are clonally enriched in metastatic/CRPC human specimens, consistent with their role in therapy-resistant progression rather than tumor initiation.
Cross-species validation using TCGA (n=503), SU2C (n=657), and Asian CPGEA (n=206) cohorts confirmed that gene signatures derived from each mouse model were significantly enriched in their respective human counterparts. The study also demonstrates that Class 1 and Class 2 mutations are largely mutually exclusive genomically, underscoring their divergent oncogenic trajectories.
These findings reframe FOXA1 not as a single-function oncogene but as a context-dependent master regulator whose gain-of-function consequences are entirely dictated by mutation location. The mechanistic distinction—Class 1 co-opting mTORC1/2 and AR via chimeric enhancers versus Class 2 activating KLF5/AP-1 progenitor programs—opens distinct therapeutic windows for each disease stage.
Key Findings
- Class 1 FOXA1 mutations plus p53 loss drive invasive AR+ adenocarcinoma via chimeric AR-half enhancers and mTORC1/2 co-activation.
- Class 2 FOXA1 mutations reprogram luminal prostate cells to a progenitor-like state via KLF5 and AP-1 neo-enhancer circuits.
- Class 2 mutants confer survival and proliferation under castrate androgen levels, mechanistically explaining therapy-resistant CRPC.
- Cross-species genomic analyses in TCGA, SU2C, and Asian CPGEA cohorts validate mouse model findings in human prostate cancer.
- FOXA1 mutations are the predominant driver alteration in Asian prostate cancer patients, affecting over 40% of primary tumors.
Methodology
Knock-in mouse models with human FOXA1 Class 1 (R265–71del) and Class 2 C-terminal frameshift mutations were generated at the Rosa26 locus and crossed with Pb-Cre4 and Trp53-floxed mice. Prostate tissues and organoids underwent histopathology, single-cell RNA/ATAC-seq, bulk transcriptomics, and ChIP-seq. Findings were validated in three independent human cohorts (TCGA, SU2C, CPGEA).
Study Limitations
The study relies on Rosa26 knock-in overexpression rather than endogenous locus mutation, which may not fully replicate heterozygous human mutation contexts. Class 2 mouse models did not develop overt carcinoma, limiting direct comparison of tumorigenicity. Human cohort analyses are correlative and cannot establish causality independent of the mouse data.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.