Genetically engineered mouse models (GEMMs) are extremely valuable in revealing novel biological insights into the initiation and progression mechanisms of human diseases such as cancer. Transgenic and conditional knockout mice have been frequently used for gene overexpression or ablation in specific tissues or cell types in vivo. However, generating germline mouse models can be time-consuming and costly. Recent advancements in gene editing technologies and the feasibility of delivering DNA plasmids by viral infection have enabled rapid generation of non-germline autochthonous mouse cancer models for several organs. The bladder is an organ that has been difficult for viral vectors to access, due to the presence of a glycosaminoglycan layer covering the urothelium. Here, we describe a novel method developed in lab for efficient delivery of DNA plasmids into the mouse bladder urothelium in vivo. Through intravesical instillation of pCAG-GFP DNA plasmid and electroporation of surgically exposed bladder, we show that the DNA plasmid can be delivered specifically into the bladder urothelial cells for transient expression. Our method provides a fast and convenient way for overexpression and knockdown of genes in the mouse bladder, and can be applied to building GEMMs of bladder cancer and other urological diseases.

Prostate epithelial basal cells are highly plastic in their luminal differentiation capability. Basal stem cells actively produce luminal cells during organogenesis, but become restricted in the adult prostate unless receiving oncogenic or inflammatory stimuli. Given that the number of luminal cells increases relative to basal cells through development and that equilibrium is reached in the adulthood, we hypothesize that a negative-feedback mechanism exists to inhibit basal-to-luminal differentiation. We provide evidence supporting this hypothesis by comparing murine prostatic growth in a tissue reconstitution assay with cell recombinants of different basal-to-luminal ratios. Additionally, in organoid culture, hybrid organoids derived from adjacent basal and luminal cells showed reduced basal stem cell activities, suggesting contact inhibition. Importantly, removal of adult luminal cells in vivo via either an inducible Cre/loxP-Dre/rox dual-lineage-tracing system or orthotopic trypsin injection led to robust reactivation of basal stem cell activities, which acts independent of androgen. These data illustrate the prostate organ as a distinctive paradigm where cell contact from differentiated daughter cells restricts adult stem cell multipotency to maintain the steady-state epithelial architecture.

Both the canonical Wnt and androgen receptor (AR) signaling pathways are important for prostate organogenesis and homeostasis. How they crosstalk to regulate prostate stem cell behaviors remains unclear. Here, we show in lineage-tracing mouse models that although Wnt is essential for basal stem cell multipotency, ectopic Wnt activity promotes basal cell over-proliferation and squamous phenotypes, which are counteracted by elevated levels of androgen. In prostate basal cell organoids, dihydrotestosterone (DHT) antagonizes R-spondin-stimulated growth in a concentration-dependent manner. DHT down-regulates the expressions of a Wnt reporter and target genes, and RNA sequencing (RNA-seq) analyses identify Wnt signaling as a key altered pathway. Mechanistically, DHT enhances AR and β-catenin protein binding, and CUT&RUN analyses reveal that ectopic AR sequesters β-catenin away from its Wnt-related cistrome. Our results suggest that an intermediate level of Wnt activity in prostate basal stem cells, achieved via AR-β-catenin interaction, is essential for normal prostate homeostasis.

Androgen signals through androgen receptor (AR) to influence prostate development and cancer. How stromal and epithelial AR regulate prostate homeostasis remains unclear. Using genetic lineage tracing, we systematically investigated the role of cell-autonomous AR in different prostate epithelial cell types. Here we show that AR is dispensable for basal cell maintenance, but is cell-autonomously required for the luminal differentiation of rare basal stem cells. In contrast, AR deletion in luminal cells alters cell morphology and induces transient over-proliferation, without affecting androgen-mediated luminal cell survival or regeneration. However, AR is selectively required for the maintenance of daughter cells produced by castration-resistant Nkx3.1-expressing luminal stem cells (CARNs). Notably, Pten loss can override AR-loss effects in both basal and luminal compartments to initiate tumours. Our data reveal distinct cell-type-specific roles of epithelial AR in orchestrating prostate homeostasis, and question the notion that epithelial AR serves as a tumour suppressor in early cancer initiation.

Androgen receptor (AR) is expressed in both the prostate epithelium and the prostate stroma and plays diverse roles in prostate physiology. Although low expression of stromal AR is clinically associated with advanced cancer stage and worse outcome, whether stromal AR inhibits or promotes prostate cancer progression remains controversial. Here, we specifically delete AR in smooth muscle cells of the adult mouse prostate under two tumorigenic conditions, namely, the Hi-Myc genetic model and the T + E2 hormonal carcinogenesis model. Histology analyses show that stromal AR deletion exacerbates tumor progression phenotypes in both models. Furthermore, single-cell analyses of the tumor samples reveal that secretory luminal cells are the cell population particularly affected by stromal AR deletion, as they transition to a cellular state of potentiated PI3K-mTORC1 activities. Our results suggest that stromal AR normally inhibits prostate cancer progression by restraining secretory luminal cells and imply possible unintended negative effects of androgen deprivation therapy.