ProMPT - Prostate Cancer: Mechanisms of Progression and Treatment

NCRI logo

Research Progress in the Last 12 Months

a) Research

Programme I: Molecular Pathology of Progressing Prostate Cancer (Cambridge & Newcastle)

I.A. Provide information on temporal recruitment of androgen receptor (AR) co-regulators to androgen responsive gene promoters (including PSA promoter/ enhancer regions). Determine the effects of siRNA knock-down of specific AR co-regulators on the expression of AR regulated genes.

I.B. To undertake promoter array analysis for AR recruitment following androgen treatment of the LNCaP cell-line. To combine with existing expression array data for androgen treatment of the same line at the same time-point to identify pioneer factors



In our studies we have identified the DEAD box RNA helicase, p68, as an important co-regulator of the AR. p68, implicated in a wide range of cellular processes including pre-mRNA, rRNA and miRNA processing, ribosome biogenesis and cell proliferation, and transcription has become the focus of this project. Confocal microscopy and immunoprecipitation have been used to demonstrate a p68-AR interaction. Reporter assays, RNA interference and chromatin immuno-precipitation (ChIP) experiments have shown p68 enhances AR transcriptional activity and is recruited with AR, in response to hormone, to androgen responsive genes. Tyrosine phosphorylation of p68, mediated by c-Abl, was shown to stimulate ligand-dependent transcription of AR regulated genes through tyrosine 593 - this effect could be completely blocked by the tyrosine kinase inhibitor Imatanib. Because p68 plays key roles in alternative splicing, we used splicing assays to further demonstrate that p68 enhanced the AR-dependent repression of a hormone responsive CD44 variable exon minigene. Using our prostate cancer tissue microarray we have shown that there is both an increased frequency and expression of p68 in prostate cancer compared with benign tissue. A manuscript is in the final stages of writing prior to submission (Over-expression of p68 (Ddx5) RNA Helicase in Prostate Cancer: a role in tumour progression. Clark EL, Coulson A, Dalgliesh C, Rajan P, Elliott DJ, Fleming S, Heer R, Gaughan L, Leung HY, Fuller-Pace FV and Robson CN).


The androgen receptor (AR) initiates important developmental and oncogenic transcriptional pathways. The AR is known to bind as a homodimer to 15-base pair bipartite palindromic androgen-response elements; however, few direct AR gene targets are known. To identify AR promoter targets, we used chromatin immunoprecipitation with on-chip detection of genomic fragments. We identified 1,532 potential AR-binding sites, including previously known AR gene targets. Many of the new AR target genes show altered expression in prostate cancer. Analysis of sequences underlying AR-binding sites showed that more than 50% of AR-binding sites did not contain the established 15 bp AR-binding element. Unbiased sequence analysis showed 6-bp motifs, which were significantly enriched and were bound directly by the AR in vitro. Binding sequences for the avian erythroblastosis virus E26 homologue (ETS) transcription factor family were also highly enriched, and we uncovered an interaction between the AR and ETS1 at a subset of AR promoter targets. These discoveries offer the potential to uncover interactions between the AR and novel genes which may be useful as biomarkers or therapeutic targets (see Massie et al, Embo Reports, 2007).

Programme II: Mechanisms of skeletal metastases (Sheffield & Manchester)

In Sheffield, prostate cancer cells have been engineered to stably express GFP and then serially passaged in the tibias of athymic mice to generate strains with enhanced tumourogenicity. These cells have been further engineered to either express very high levels of OPG or to have stably suppressed OPG expression using shRNA and tested for sensitivity to TRAIL. Although the OPG overexpressing clones were shown to be producing OPG at up to 100ng/ml in conditioned medium, this OPG was not biologically active. Studies with recombinant OPG added to cultures of these cells and the parent cell line suggest that the cells produce a factor that can sequester OPG activity. This is currently under investigation. OPG knock down studies do show that suppression of OPG can result in increased TRAIL sensitivity in some clones. However considerable heterogeneity in TRAIL sensitivity that was unrelated to OPG production has been demonstrated in both engineered cell strains and the parent PC3 cell line as well as in other tumour cell lines and strains in our laboratory. This suggests that production of OPG is just one of a number of mechanisms by which induction of apoptosis by TRAIL is avoided by tumour cells. Induction of TRAIL resistance in vivo is being evaluated and resistant populations characterized to determine the dominant pathways exploited for resistance in vivo. Much of this work has been published or has been submitted for publication:

OPG transcripts and proteins have been evaluated from both stromal cells and from tumour populations. All cells produced mRNA for mature OPG plus varying levels of 2 additional shorter splice variants. Specific knock-down of alternatively spliced mRNAs indicated that only siRNAs directed toward full length OPG mRNA were effective in suppressing OPG protein production, as measured by ELISA or Western blot, in tumour and stromal populations. This suggests that the shorter splice variants did not encode secreted OPG. We also examined the available anti-OPG antibodies for differences in binding proteins on 1 and 2D Western blots. Several different apparently specific proteins were found in these experiments that were bound by antibodies. All proteins except for the ~55kDa mature OPG protein, present on all blots, were shown not to be OPG by purification and MALDI-TOF analysis. In doing this work we realized that one of the antibodies commonly used crossreacted very strongly with carbonic anhydrase II. This and the data above related to function of splice variant have been reported in a recent manuscript.

Manchester has shown that Arachidonic acid is a potent stimulator of prostate epithelial invasion and that it induces a switch from "mesenchymal" to "amoeboid" movement. We have now determined that prostate epithelial cells take up lipids from their environment as a first step in consolidating cellular survival and propagation at the metastatic site. The CD36 / caveolin axis is critical to this process and studies of this mechanism are ongoing. Our experimental programme is going on to study the changes in lipid composition during this process using mass labelled lipids and FTIR/Bio ToF-SIMS analysis in combination with the advanced gas chromatography facility mass spectroscopic analysis of bone marrow aspirates form patients (Prof Goodacre/ P Gardner et al Analytical Chemistry, Manchester). The effect of AA on cellular morphology and trans-migrational movement across endothelial barriers in 2D or in 3D co-culture has been followed by timelapse 4D microscopy in collaboration with the CRUK advanced imaging facility (Paterson Institute). We have demonstrated that amoeboid movement is essential to endothelial transmigration (Rho mediation) and the mesenchymal switching is the means of stromal migration (Rac Mediation). Both of these pathways are amenable to blockade by therapeutic intervention. (eg TAT-C3 or Y297632 inhibition: Rac-1 pathway or NSC 23766 Rho pathway).

Programme III - York & Manchester

Initial studies have established that tumour frequency is higher from a2b1hi/CD133+ cells relative to the progenitor populations a2b1lo/CD44+ and a2b1lo/CD24+. However, only 10% of patient samples tested have resulted in tumours. These results were obtained with two patient's samples; a Gleason grade 9 tumour and a lymph node metastasis of the prostate. Since these results were obtained the amount of tissue per sample has decreased to small core biopsies of 2mm x1cm on average. We initially resorted to expanding the cells in culture and subsequently selecting the cell population for grafting. However, this approach was not successful, and tumours were not obtained from any of the phenotypes grafted. It is unlikely that this was due to the presence of normal cells in culture as we detected expression of the fusion product TMPRSS:ERG (see Figure X). Where material is limited, we initially expand the tissue sample sub-cutaneously prior to implanting of specific phenotypes intraprostatically. This approach has proved successful as tumour frequency has increased to 50% n=4. Of the 7 patients tissue, 4 have been grafted sub-cutaneously and 3 orthotopically. One sample in particular, from a Gleason 7 tumour, has grown sub-cutaneously and intraprostatically resulting in metastatsis of the kidneys and liver, with a latency of 8 weeks (sub-cut) and 3 months (intraprostatically). This sample is now maintained as a xenograft and can be successfully grown in vitro. We are awaiting results from the selected populations for this and other samples.

We have invested a lot of time and event in our mouse model, and have now a successful breeding colony of Rag2-/- GC-/- mice. We initiated this colony in Jan07 from one breeding pair, and have produced 325 mice to date. This mouse is particularly immuno-compromised and as such tumour take has increased significantly as well as the incidence of metastasis. It is very rare to get metastases when grafting sub-cutaneously, even with cell lines, and is rarer with patient material. The phenotype is normal with no increased incidence of lymphomas previously found in NOD/SCID hosts after long latent periods.


The Manchester team has isolated a "Side Population" from both primary human benign and malignant Prostate epithelial cells (Bhatt et al., 2003). We have shown that this population is enriched for putative prostate stem cells (Brown et al The Prostate 2007). The technical problems relating to small cell number analysis have been overcome in collaboration with EpiStem Ltd and the CRUK Affymetrix/Bioinformatics Facilities. We have now been able to identify comparative genetic profiles from SP and Non Sp populations and have shown that quiescent cells within the SP are small with little cytoplasm, enabling sorting criteria to be expanded to select for cells of low forward and side scatter characteristics on FACs analysis (Brown et al 2007). We have identified >700 significant gene differences between SP and Non-SP. The isolated stem population is also to be characterised chemometrically as part of a BBSRC post doctoral collaborative fellowship programme from 2008-2011 in collaboration with Manchester University Analytical Chemistry and stem cell research groups in the USA and France.

Programme 4:Newcastle Medicinal Chemistry programme 12 months funding for post doctoral RA1A

To discover small-molecule inhibitors of the MEK5-ERK5 signalling pathway for the treatment of prostate cancer.

  1. HTS and Hit-validation. In collaboration with CRT Development Laboratories, high-throughput screening for ERK5 inhibitors will be carried out in order to identify hit-series. Initial structure activity relationships will be established by the purchase of structurally related compounds. Key compounds from selected hit-series will be resynthesised in excellent purity and the hit-activity confirmed from the genuine sample. This will identify a number of validated-hits as deliverable 1.
  2. Hit-to-lead development. Validated hit series will be explored and expanded by the synthesis of small, focussed libraries to develop structure activity relationships (SARs). Library design will be guided by molecular modelling (CRT Development Labs) using an ERK5 homology model developed from the published ERK2 X-ray structure. Lead compounds will display inhibitory activity in the 10-100 nM range for ERK5 and at least 10-fold selectivity vs closely related kinases.


  1. HTS and Hit-validation: HTS screening has been completed for ERK5 at 30M. A diverse library of 48,479 compounds returned 195 confirmed hits (0.6% hit rate). A kinase focussed library of 9,138 compounds returned 50 confirmed hits (0.41% hit rate). Additional compounds from commercial suppliers have been assayed to expand preliminary SARs. Three promising hit series have been identified, with other interesting compounds rated as lower priority. Key examples of these series have bee resynthesised and purified. Compounds will be supplied to CRT DL for evaluation in January 08.
  2. Hit-to-Lead development: Small molecule libraries are in preparation around one selected hit-series in order to complete the SAR from the purchased libraries. Hit compounds have been docked into the ERK5 model and are providing useful suggestions for future compound libraries, subject to the hit-validation results.

Programme VI

The Bristol ProMPT group in the Department of Social Medicine has achieved its objective of generating a large case-control dataset from ProtecT study data (2,036 cases, 12,782 controls). This case-control dataset is being used to study molecular, genetic, clinical and lifecourse aspects of prostate cancer epidemiology, including: the roles of insulin growth factor (IGF), vitamin D and folate pathway molecules in prostate cancer aetiology and progression; environmental (diet, alcohol, tobacco), clinical (diabetes, obesity, symptoms, medications) and anthropometric (height, leg length) factors associated with the risk of prostate cancer; and genotypes related to diet, alcohol and prostate cancer. The Bristol ProMPT group has also established links with researchers in the Department of Biochemistry, who will focus on the role of Rho GTPases and interactions between IGF-IR and EGFR systems in prostate cancer cells, and has obtained access to data from long-term cohort studies (UK Boyd-Orr, Caerphilly, PROCESS, and CLasP studies; Norwegian HUNT study) to explore factors associated with prostate cancer incidence.

b) Infrastructure


We have continued our collaboration with Sheffield, developing a new project (funded by a new White Rose graduate student network) to study DNA damage and repair in cancer stem cells. This also has an International dimension with the involvement of Dr Rob Bristow, from Toronto. We are also seeking to expand our tissue collection in Sheffield to acquire particular tumour subtypes. A joint collaborative funded student (with Cambridge), Claire Brinham, has recently graduated with a PhD from York. Between collaboratives we have continued to work with Dr David Hudson, analysing the expression phenotypes of prostate cancer stem cells, and are actively pursuing the identification of TMPRSS2-erg fusion genes in the cancer stem cell fraction using FISH technology provided by Dr J Clark and Prof C Cooper, ICR.

Biological repository

Bristol has performed the epidemiological collections, and we now have a linked clinical database. Bristol has established a repository of data derived from the ProtecT study, including the case-control series, and accompanied by statistical files, molecular assay and genetic data, and standard operating procedures. A full-time ProMPT Research Associate has been appointed in the Department of Social Medicine to undertake and coordinate epidemiological research. The case-control series generated in Bristol from ProtecT data is available to researchers across the collaborative, together with a repository of derived data and statistical files. We provided Dr Chris Parker with our CaP TMA to determine expression of novel markers in CaP. We also provided this TMA resource to Prof C Cooper for his work on the TMPRSS2 fusions.

About Us
Current Research

Funded Jointly by: Cancer Research UK logo Department of Health logo MRC logo