Dr Ian Vela finds Potential New Target for Metastatic Prostate Cancer

Discovery by: Drs Ian Vela, Colleen Nelson and Judith A. Clements (APCRC - Q); Drs Colm Morrissey, Eva Corey and Robert L. Vessella; Dr Edith M. Gardiner

Metastatic prostate cancer is currently considered incurable and there is an urgent need to identify important factors in the spread of localised cancer to distant sites. Our main research focus is cancer-bone interactions, which are thought to be critical for the establishment of PCa bone metastases. In the search for novel molecular promotors of PCa metastasis we detected a developmental gene, PITX2, with limited expression in primary PCa and increased expression in PCa metastases. In addition, we identified a likely role for this gene in PCa progression and metastasis to bone.

Establishment of metastatic deposits in hard tissue involves complex interactions between cancer and bone cells, often exploiting developmental pathways important in normal skeletal formation and function. In particular the canonical Wnt pathway has been implicated in PCa progression and skeletal metastasis through its effects on cellular proliferation and interaction with the bone microenvironment. During our investigation of Wnt involvement in PCa progression and bone metastasis we performed a broad evaluation of Wnt pathway associated gene expression in experimental tumors and in clinical samples.

Initially we used real-time poly-merase chain reaction (PCR) arrays to profile RNA expression of 83 Wnt relevant genes in experimental subcutaneous tumors in mice produced from the human PC-3 and LNCaP PCa cell lines. The PC-3 line produced highly osteolytic characteristics. We then used PCR arrays to profile Wnt genes in clinical samples of normal, primary and metastatic PCa tissue. Based on our analysis of the array results we chose an outstanding candidate gene for further investigation by staining for the candidate protein on clinical tissue microarrays (TMAs) of primary PCa, bone and soft tissue metastases, and by functional analyses of cultured cells.

Although several significant differences between the two types of subcutaneous tumors were detected, we selected PITX for followup analysis because its RNA level in tumors formed by PC-3 cells was more than 40-fold higher than that in LNCaP tumors. Furthermore, in concurrence with the current literature PITX2 RNA levels in the normal prostate and primary PCa were low to absent in the majority of cases. However, in our expanded studies of clinical specimens the PITX2 RNA levels were 13 000-fold higher in bone metastases than in the normal prostate. Consistent with RNA transcript levels, we saw a similar pattern of expression at the protein level in primary and metastatic tumors stained by PITX2 immunohistochemistry.

We tested the role of PITX2 in cellular migration by experimentally varying the level of the gene in the PC-3 cell cultures. Cells with increased PITX2 expression had increased motility in a wound healing scratch assay, whereas decreasing PITX2 in these cells by RNA interfernece had the opposite effect and impaired cell motility in this assay. Knockdown of cellular PITX2 also decreased migration towards culture medium that had previously been conditioned in bone cell cultures, consistent with the possible role of PITX2 in the honing of the PCa cells in the bone microenvironment.

PITX2 is a homeodomain transcription factor involved in left-right determination during development through its interaction with the canonical Wnt pathway. Most functional investigations of PITX2 have focused on its role in eye, heart, gut and tooth development, particularly in the context of abnormalities in these organs in the Axenfeld Rieger syndrome type 1, which is characterised by a loss of function mutation in the PITX2 gene. In those invesitgations several PITX2 target genes important for the development of these organs were identified, including N-cadherin, Nodal, Sonic hedgehog, Lefty, and Wnt 11.

Interestingly hypermethylation of the PITX2 promoter in primary PCa was previously associated with an increased risk of biochemical recurrence, suggesting possible usefulness as a biomarker to stratify prostatectomy cases with regard to this risk.

Methylation of the PITX2 promoter would be expected to decrease PITX2 RNA and protein levels, yet our investigation of metastatic material revealed an increase in PITX2 production with PCa progression. This finding would be consistent with the functional involvement of PITX2 and/or its transcriptional regulators in the development of metastasis. It will be essential to understand what drives this increase in PITX2 expression and the significance of this change during cancer progression.

Our observations that PITX2 expression was increased in metastatic disease, and PCa cell motility and migration toward osteoblsat conditioned medium were proportional to PITX2 expression in cell culture models suggest that PITX2 may promote PCa metastasis by driving the cancer cells toward the bone metastatic niche in the local microenvironment. It will be important to define the mechanistic role of PITX2 in this process and idenitfy the chemoattractant factors in bone to which the PITX2 mechanism responds. Given our current lack of effective therapeutic interventions in metastatic disease, a new target related to PITX2 function might be an attractive option for the prevention and/or treatment of PCa bone metastases.