Use of fluorescent imaging to monitor drug responses in mouse models of tumourigenesis

Abstract

As our understanding of the complexities of cancer biology has increased, the ability to exploit unique features of tumour cells with molecularly targeted therapies has become a reality. However, despite unprecedented volumes of new molecules in clinical trials, the number of highly effective drugs approved by the regulatory authorities remains disappointingly low. Moreover, oncology drug development is plagued by high levels of attrition in late phase clinical development. Failure due to poor efficacy and toxicity issues are not believed to be a result of the development of molecules with inadequate pharmaceutical properties, but rather due to a lack of understanding of their full mechanism of action. All of this points to imprecise analysis of the drugs during the preclinical phase, highlighting the need for better preclinical drug development tools. Animal models provide a key preclinical tool, and as a therapeutic area, oncology is characterised by models which are not predictive of the true human pathology. Overexpression of the human epidermal growth factor receptor two (HER2) oncogene, and inactivation of the phosphatase and tensin (PTEN) tumour suppressor, are two important events in human breast cancer. A novel conditional mouse model driven by overexpression of HER2 coupled with / without the loss of PTEN has been characterised to interrogate the importance of these two cellular perturbations. Multifocal tumours arose in mice from both lines, while luminal tumour characteristics were shown to be reduced and basal characteristics increased with a reduction in PTEN expression. Disruption of PTEN rapidly accelerated tumour onset (from 138 to 82 days) and tumour growth (with the time from tumour onset to maximum tumour size reduced from 38 to 21 days), significantly reducing overall survival (from 165 to 102 days). The ability of tumour cells to colonize the lungs was not significantly affected by the loss of PTEN. Tumours arising in both mice genotypes were utilized to generate cell lines. These failed to provide an in vitro representation of the tumours, and found little utility in drug efficacy studies with HER family targeted agents, a situation which could be improved by the use of different culture methods. Since suppression of apoptosis is a hallmark of human cancer, and a desired endpoint of many anticancer therapies is the induction of cell death, the generation of cell lines inherently capable of sensing caspase-mediated apoptotic cell death would be a valuable drug development tool. Given that fluorescence imaging is also emerging as a potentially powerful modality for preclinical drug development, a novel fluorescent in house apoptosis reporter construct was generated (pCasFSwitch). Initial validation of pCasFSwitch by transient transfection into murine mammary carcinoma cells proved difficult due to transfection associated toxicity, yet proof-of-principle was indicated. Transfer of pCasFSwitch into a retroviral backbone vector enabled the generation of stably transfected squamous carcinoma cells more suitable for further analysis. Incubation of lysates from these cells with recombinant enzymes revealed the construct could be cleaved by caspase-3, but not by other members of the cysteine protease family. Furthermore, assessment of apoptosis levels in the cells upon staurosporine treatment proved the utility of the construct to quantify cell death, and was validated against data generated with a commercial competitor, NucView. Further comparison of the specificity of the imaging agents using caspase inhibitors was limited by the functionality of currently available inhibitors, but did reveal that in common with NucView, construct quantified levels of apoptosis were affected by inhibition. This thesis details the development of two preclinical drug development tools. A novel mouse model enables biological interrogation of two key events in human breast carcinogenesis. Since PTEN loss is associated with resistance to HER2 targeted therapies, it is ideally suited for efficacy testing to overcome such resistance. The in house fluorescent apoptosis imaging agent allows a temporal read-out of drug effects in live single cells. While the use of intravital imaging of stable cell lines implanted under imaging windows would allow in vivo validation of in vitro data. Taken together, such facilitation of thorough evaluation of therapies at the preclinical stage, will reduce the adverse effects felt by the pharmaceutical industry of failure late in the drug development pipeline.