Abstract
Abstract: Background: Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive types of solid cancer with 5-year survival rates approaching a dismal 5%. Novel therapeutic targets need to be identified thus aiding and assisting the design of treatments which will improve survival rates that have not changed in the last 30 years. Of particular interest are homeobox (HOX) genes, a set of 39 evolutionarily conserved transcription factors involved in embryonic antero-posterior patterning. Although expressed in development, HOX genes have been found to be re-expressed and indeed dysregulated in several types of cancer including lung, breast, ovarian and renal neoplasia. Limited research has been undertaken on the dysregulation of HOX genes in PDAC. HOX genes can be antagonised using HXR9, a peptide which competitively inhibits the interaction between HOX genes and their co-factor PBX, subsequently preventing HOX genes to fulfill their role of transcription factors. Cancer-specific tumour-modelling is fundamental to drug testing. There are few animal models that recapitulate the unique tumour architecture and molecular signature of PDAC, particularly the desmoplastic reaction characteristic of this malignancy. The chorioallantoic membrane (CAM) assay, an in ovo model that utilises the immunologically naive properties of the developing chick embryo to grow a solid tumour derived from pancreatic cancer cell lines. The CAM model is not widely used in pancreatic cancer research and more work is needed to evaluate it’s efficacy for tumour remodelling and subsequent drug testing. We have found that the CAM model is suitable for drug testing as it recapitulates the architecture and molecular signature of PDAC. In order to establish the CAM model as appropriate for drug testing in this context, we assessed whether the mitogen-activated protein kinases (MAPK) pathway was conserved, due to the high frequency of mutational activation of the KRAS gene in this cancer. Global gene expression was also carried out to determine genetic changes between cells grown in vitro and cells in a tumour microenvironment in the CAM model. Experimental design We investigated whether there is a signature HOX gene profile unique to this disease. We measured HOX gene expression by RT-PCR in four well-described pancreatic cancer cell lines. HOX gene expression was also measured in commercially obtained RNA and snap-frozen pancreatic cancer tissue from surgical resections. The CAM model was set up by grafting 4 pancreatic cancer cell lines and tumour architecture and molecular signature was evaluated by H&E staining and IHC. Gene expression was assessed by microarray analysis to compare global gene expression in cell-lines and CAM tumours and conservation of mitogen-activated protein kinases (MAPK) pathway was addressed by western blotting. HOX gene expression was measured by RT-PCR in both cell lines and CAM tumours. Finally, the efficacy of HRX9 was measured by generating IC50s using MTS and LDH assays. Levels of apoptosis were measured in vitro using Annexin-V-PE assay and in vivo by cleaved-caspase activation, assessed by IHC. Results: We found that HOX gene expression was elevated in tumour samples compared with normal tissue and in particular a significantly higher expression of HOXA13 in PDAC samples compared with normal pancreas. This was confirmed at the protein level by Immunohistochemistry (IHC). We also showed that the CAM model is suitable for drug testing as it recapitulates the architecture and molecular signature of PDAC. Results showed that MAPK pathway was conserved, due to the high frequency of mutational activation of the KRAS gene in this cancer. Cleaved-caspase activation also supports the hypothesis that tumour cells are driven into apoptosis upon HXR9 treatment. Conclusions HOX gene expression is highly dys-regulated in pancreatic cancer and more work is need to individually evaluate HOX genes on interest highlighted in this study. HOX gene expression can be antagonised by using HXR9. Finally, we have demonstrated the potential for HOX gene targeting as a novel therapy for PDAC.