Androgen deprivation induces selective outgrowth of aggressive hormone-refractory prostate cancer clones expressing distinct cellular and molecular properties not present in parental androgen-dependent cancer cells

PURPOSE: The mechanism of progression of human prostate cancer (CaP) cells under androgen ablation therapy remains unclear. To study the alternative pathways of CaP cell growth under conditions of androgen deprivation, androgen-independent CaP variants were selected and expanded from an androgen-dependent CaP line via an in vitro androgen deprivation treatment. Cellular and molecular properties of these androgen-independent variants were characterized both in vitro and in vivo and compared with those of their parental androgen-dependent cells. METHODS: Androgen deprivation treatment of an androgen-dependent CaP cell line, LNCap, was carried out by replacing culture medium with RPMI 1640 medium plus 10% charcoal-stripped serum. Cells that survived through the androgen deprivation treatment were harvested and expanded in the androgen-deficient culture medium and were designated CL-1. The CL-1 cells were also re-cultured in androgen-containing medium and designated CL-2. The growth (cell cycle analysis, 3H-thymidine incorporation assay, growth expansion, and colonization efficiency), expression of CaP-associated markers (semiquantitative reverse transcriptase polymerase chain reaction), interaction with endothelial and bone marrow stromal cells, sensitivity to anticancer agents and radiation (growth inhibition), and tumorigenicity of CL-1 and CL-2 cells were determined and compared with these characteristics in parental LNCaP cells. RESULTS: CL-1 and CL-2 cells are fast-growing cells when compared with parental LNCap cells. They were capable of potentiating the growth of endothelial and bone marrow stromal cells in co-culture experiments and acquired significant resistance to radiation and to anticancer cytotoxic agents (Taxol® paclitaxel, vinblastine, and etoposide). In contrast to the poorly tumorigenic parental LNCap cells, CL-1 and CL-2 lines proved highly tumorigenic, exhibiting invasive and metastatic characteristics in intact and castrated mice or in female mice within a short period of 3 to 4 weeks. No growth supplements (e.g., Matrigel) were needed. When transfected with the green fluorescence protein (GFP) gene and transplanted orthotopically in the accessory sex gland, extensive metastatic disease from the primary CL tumor could be identified in bone, lymph nodes, lung, liver, spleen, kidney, and brain. Semiquantitative reverse transcriptase polymerase chain reaction analysis revealed a markedly distinct molecular expression profile in the CL lines: overexpression of basic fibroblast growth factor, interleukin-6, interleukin-8, vascular endothelial growth factor, transforming growth factor-β, epidermal growth factor receptor, caveolin, and bcl-2 messenger RNAs and marked down-regulation of E-cadherin, p-53, and pentaerythritol tetranitrate. CONCLUSIONS: Early administration of hormonal therapy after failure of first-line treatment is associated with a profound clonal selection of aggressive Al variants, such as CL-1 and CL-2 lines. These tumor lines, with their parental counterparts, can serve as valuable tools for studying the cellular and molecular mechanisms of CaP progression and metastasis under hormonal therapy. CL-1 and CL-2 offer a unique and reproducible model for the evaluation of drug sensitivity and for other therapeutic modalities for advanced prostate cancer.