| Summary: | Prostate cancer is the second leading cause of cancer death in men in the UK. Around 30% of patients with apparently localized disease at diagnosis and who receive radical surgery to remove the primary tumour, will present with skeletal metastasis, in many cases >5 years after initial treatment. This suggests that prostate cancer cells can persist in a dormant or indolent state, undetected in patients for many years. To investigate states of dormancy in prostate cancer cells, the growth of human prostate cancer cell lines in vitro was examined to test the hypothesis that there are sub-populations of dormant cells present even within widely used, rapidly proliferating cell lines. My studies successfully identified these cells, which were then extensively characterised. Parallel studies in our laboratory indicated that these cells had greater potency than other cells in the initiation of metastatic lesions in xenograft models. To identify dormant cells in three prostate cancer cell lines: PC-3NW1, LNCaP and C42B4, were stained with the lipophilic membrane dye Vybrant DiD. The retention of this dye was used as a marker of dormant or slow-cycling cells. Flow cytometry analysis and fluorescent microscopy identified a dormant cell sub-population in all three prostate cancer cell lines at a frequency of <2% after 14 days in culture. Dormancy was not an intrinsic characteristic of this sub-population, as this phenotype could re-emerge in cultures of separated re-cultured non-dormant cells. The frequency of dormant cells in cultures could be altered by changes in culture conditions and when isolated these cells could be released from dormancy to form colonies. Gene expression profiles for the non-dormant and dormant sub-population were compared by RT-qPCR and a set of differentially expressed, dormancy specific genes identified. Immunofluorescence microscopy was used to assess whether the observed differences in gene expression between populations were reflected in altered specific protein levels. The presence of a specific marker of dormancy, CXCR4, identified by these experiments, was then evaluated in 75 primary patient samples by immunohistochemistry. Elevated CXCR4, as measure by immunohistochemistry, in patient samples correlated to other indicators of poor prognosis but did not independently predict the presence of metastases.
|
|---|
