Expansion of mouse castration-resistant intermediate prostate stem cells in vitro

Abstract

Background: Most castration-resistant prostate cancers (CRPCs) have a luminal phenotype with high androgen receptor (AR) and prostate-specific antigen (PSA) expression. Currently, it is difficult to culture castration-resistant luminal cells with AR and PSA expression.

Methods: We formulated a custom-made medium and isolated primary cells from the prostate of adult wild-type (WT) and TRAMP mice. The cells were characterized by immunofluorescence staining, transcriptomic analysis, and qRT-PCR verification. Their self-renewal and differentiation potential in vitro and in vivo were examined. We treated the cells with androgen deprivation and enzalutamide and performed immunofluorescence staining and western blotting to analyze their expression of AR and PSA.

Results: We isolated a novel type of castration-resistant intermediate prostate stem cells (CRIPSCs) from adult WT and TRAMP mice. The mouse CRIPSCs proliferated rapidly in two-dimensional (2D) culture dishes and can be cultured for more than six months. The mouse CRIPSCs expressed luminal markers (AR, PSA, and Dsg4), basal markers (CK5 and p63), Psca, and the intermediate cell marker (Ivl). Transcriptomic analysis showed that the mouse CRIPSCs had upregulated signaling pathways related to cancer development and drug resistance. In the long-term culture, TRAMP CRIPSCs had higher expression of the genes related to stem cells and cancers than WT mice. Both WT and TRAMP CRIPSCs formed organoids in Matrigel. WT CRIPSCs did not form prostate tissues when transplanted in vivo without urogenital sinus mesenchyme (UGM) cells. In contrast, TRAMP CRIPSCs formed prostate ducts in NOG mice without UGM cells and differentiated into luminal, basal, and neuroendocrine cells. Androgens regulated AR translocation between the nucleus and cytoplasm in the mouse CRIPSCs. Treatment of androgen deprivation (ADT) and enzalutamide reduced AR expression in WT and TRAMP CRIPSCs; however, this treatment promoted PSA expression in TRAMP, while not WT CRIPSCs, similar to the clinical observations of CRPC.

Conclusions: Our study established a method for isolating and expanding mouse CRIPSCs in 2D culture dishes. Mouse CRIPSCs had markers of basal and luminal cells, including AR and PSA, and can differentiate into prostate organoids and tissues. TRAMP CRIPSCs had elevated PSA expression upon ADT and enzalutamide treatment. Our method can be translated into clinical settings for CRPC precision medicine.

Keywords: Androgen deprivation; Androgen receptor; Enzalutamide; Intermediate cell; Prostate-specific antigen; Stem cell; castration-resistant prostate cancer.

Fig. 1

Expansion of CRIPSCs from adult mice. Phase-contrast images of primary mouse prostate epithelial cells isolated from wild-type (WT) and TRAMP mice at primary passage (P0, 7 days, A and B), P1 (14 days) (C and D), P5 (2 months) (E and F), and P25 (6 months) (G and H). Arrows point to small epithelial cells. Arrowheads point to mesenchymal cells. Scale bars, 100 μm

Fig. 2

Marker expression of mouse CRIPSCs. The prostate epithelial cells isolated from WT mice at P0 (7 days) (A and B) and P5 (2 months) (C–J) were immunostained by the antibodies against CK8 (A and C), CK18 (B and D), AR (E), PSA (F), E-cadherin (G), CK5 (H), p63 (I), and Sox2 (J). DAPI-stained nuclei. Scale bars, 100 μm

Fig. 3

Gene expression profiles of mouse CRIPSCs. A KEGG pathway analysis. B Heatmap of the differentially expressed genes in prostate tissue and WT CRIPSCs. C qRT-PCR analysis of the expression of some genes in prostate tissue, the CRIPSCs cultured for 2 (P5) and 6 (P25) months in vitro, from WT and TRAMP mice. Data were presented as mean ± SD. Two-way ANOVA was performed on the data, followed by Bonferroni post hoc tests. “ns” = not significant. **p < 0.01. ***p < 0.001

Fig. 4

Organoid formation in vitro. A Phase-contrast images of the organoids formed by the CRIPSCs (P25, 6 months) derived from WT and TRAMP mice. B Quantification of organoid diameter. Data were presented as mean ± SD. Two-way ANOVA was performed on the data, followed by Bonferroni post hoc tests. ***p < 0.001. C–E, Immunofluorescence images of the organoids stained by the antibodies against CK8, CK18, PSA, p63, CK5, and Sox2. DAPI-stained nuclei. Scale bars, 100 μm

Fig. 5

Differentiation of mouse CRIPSCs in vivo. The CRIPSCs (P25, 6 months) isolated from TRAMP mice were transplanted into NOG mice for eight weeks, followed by cryosection and immunostaining (A–H). The antibodies included CK8 (A), CK18 (B), AR (C), PSA (D), CK5 (E), p63 (F), Chromogranin A (CHGA) (G), Synaptophysin (SYP) (H). DAPI-stained nuclei. Scale bars, 100 μm

Fig. 6

AR translocation between the nucleus and cytoplasm. WT CRIPSCs (P25, 6 months) were cultured in the media supplemented with different concentrations of DHT for 12 days (A, 0 nM; C, 0.1 nM; E, 1 nM) or 10 days of low DHT followed by two days of 10 nM DHT (B, D, and F). DAPI-stained nuclei. Scale bars, 100 μm

Fig. 7

AR and PSA regulation by androgen deprivation and enzalutamide treatment. Western blots and quantifications of AR and PSA of WT (A–C) and TRAMP (D–F) CRIPSCs (P25, 6 months) treated with different concentrations of DHT and enzalutamide (Enza) for one month. Data were presented as mean ± SD. One-way ANOVA was performed on the data, followed by Bonferroni post hoc tests. *p < 0.05. **p < 0.01. ***p < 0.001