Fig 1: Functional characterization of FOSL1 as a key regulator in double-negative prostate cancer subtype. A Transcription factor activity in malignant epithelial subtypes. Heatmap displaying the top transcription factors with the highest inferred activity (based on SCENIC analysis) in each of the four malignant epithelial subtypes. Color scale (blue to red) represents low to high regulon activity. B Spatial activity of FOSL1 across subtypes. UMAP of malignant epithelial cells colored by the SCENIC-derived regulon activity score of FOSL1, showing its specific enrichment in Subtype 4. Gradient from white to blue indicates low to high activity. C Effect of FOSL1 knockdown on cell proliferation (CCK-8 assay). Growth curves of PC-3 cells transfected with control siRNA or three independent siRNAs targeting FOSL1. Data are presented as mean ± SD; ***P < 0.001, **P < 0.01 (two-way ANOVA). D Effect of FOSL1 knockdown on clonogenic survival. Representative images of colony formation assays for PC-3 cells treated as in (C). E Effect of FOSL1 knockdown on apoptosis. (Left) Representative flow cytometry plots of Annexin V/PI staining. Quadrants: viable cells (Annexin V⁻/PI⁻), early apoptotic (Annexin V⁺/PI⁻), late apoptotic (Annexin V⁺/PI⁺), and necrotic (Annexin V⁻/PI⁺). (Right) Quantification of total apoptotic cells (early + late apoptosis). Data are mean ± SD; ***P < 0.001, **P < 0.01 (one-way ANOVA). F Identification of FOSL1-regulated DNPC genes. Venn diagram showing the overlap between predicted FOSL1 target genes (from SCENIC) and the DNPC gene signature, yielding 8 candidate genes. G FOSL1 binding to candidate gene promoters (ChIP-qPCR). ChIP-qPCR analysis showing enrichment of FOSL1 at the promoters of the 8 candidate genes in control vs. FOSL1-knockdown PC-3 cells. Results are presented as % input. Data are mean ± SD; ***P < 0.001, **P < 0.01, ns = not significant (two-tailed Student’s t-test). H Expression of candidate genes upon FOSL1 knockdown (qRT-PCR). mRNA expression levels of four selected candidate genes in PC-3 cells after FOSL1 knockdown. Data are normalized to control and presented as mean ± SD; ***P < 0.001, **P < 0.01, ns = not significant (two-tailed Student’s t-test). I Correlation between FOSL1 and HMGA1 expression. Scatter plot showing a significant positive correlation between FOSL1 and HMGA1 expression across all malignant epithelial cells (n = 152,872). The red line indicates the linear regression fit (R² = 0.212, P < 0.0001, Pearson correlation). J Spatial co-expression of FOSL1 and HMGA1. (Left three panels) Spatial mapping of FOSL1 and HMGA1 expression and their co-localization in a representative section (Patient 01). Red spots indicate high co-expression, yellow indicates high FOSL1 alone, green indicates high HMGA1 alone, and gray indicates low expression of both. (Right panel) Correlation analysis of FOSL1 and HMGA1 expression specifically within the co-expressing spots (red spots, left panel), showing a strong positive correlation (R = 0.923, P = 1.28e-10, Pearson correlation). K-M Western blot analysis of downstream pathways. Protein levels of (K) stemness markers (CD44, OCT4, SOX2, NANOG), (L) EMT markers (ZO-1, N-cadherin, E-cadherin, Vimentin, SNAIL), and (M) DNPC markers (DSG3, KRT6A, KRT5), along with FOSL1 and HMGA1, in PC-3 cells under four conditions: Control, FOSL1 knockdown (siFOSL1), HMGA1 overexpression (oeHMGA1), and FOSL1 knockdown combined with HMGA1 rescue (siFOSL1 + oeHMGA1). GAPDH served as a loading control. N Rescue of proliferation by HMGA1 overexpression. CCK-8 proliferation assay of PC-3 cells under the four conditions described in (K-M). Data are mean ± SD; ***P < 0.001, **P < 0.01, ns = not significant (two-way ANOVA). O Rescue of apoptosis by HMGA1 overexpression. Quantification of total apoptosis (Annexin V⁺ cells) by flow cytometry under the four conditions described in (K-M). Data are mean ± SD; ***P < 0.001, **P < 0.01, ns = not significant (one-way ANOVA). P Pharmacological inhibition of FOSL1 reduces HMGA1. Western blot showing dose-dependent decrease of FOSL1 and its downstream target HMGA1 in PC-3 cells treated with increasing concentrations (0, 0.2, 2, 5, 10 µM) of the FOSL1 degrader T-5224. Q In vivo combination therapy schematic. Workflow of the xenograft study. Nude mice bearing PC-3 subcutaneous tumors were treated starting at day 13 post-inoculation with vehicle, Docetaxel (1 mg/kg), FOSL1 degrader T-5224 (10 mg/kg), or the combination via intraperitoneal injection every 48 h. Treatment continued until day 28. R Representative images of resected tumors. Photographs of excised tumors from each treatment group at the study endpoint. S Tumor growth curves. Tumor volume (mm³) was measured over time for each treatment group. Data are presented as mean ± SEM; ***P < 0.001 (two-way ANOVA). T Tumor weight at endpoint. Final tumor weights (g) for each group. Data are mean ± SEM; ***P < 0.001, **P < 0.01 (one-way ANOVA). U Assessment of tumor cell proliferation (IHC). Representative immunohistochemistry images showing Ki67 expression in tumor sections from each treatment group. Scale bar, 100 μm