A36
Proteomic Dissection of an Early Colon Cancer Model: HMGB1 a Potential Biomarker
Abeer Hammoudi1, Rosalind Jenkins1, Jane Hamlett1, Karen Reed2, Alan Clarke2, Alastair Watson1, John Jenkins1
1University of Liverpool, UK, 2University of Cardiff, UK
Background
Colon cancer is the result of accumulation of multiple genetic changes. Mutation of APC gene or other components of the Wnt pathway occurs in 95% of sporadic colon cancers. Inactivation of APC stabilises β-catenin which translocate to the nucleus with subsequent transactivation of many target genes.
A novel murine colon cancer model with inducible intestinal epithelial APC deletion was developed by Clarke and his group. They have shown that loss of APC causes immediate activation of the Wnt pathway with failure in cell migration and cell differentiation. A DNA microarray gene analysis of this unique model has been undertaken and a cohort of APC-regulated gene changes identified for which there has been little previous association with neoplasia.
We aim to complement the previous studies and identify the proteins that are associated with the phenotypic changes that immediately follow the loss of APC. The ultimate purpose being to identify potential Bio-markers for early stage colon cancer detection.
Method
We have established techniques for generating extracts from murine intestinal epithelial cells (APC-WT: APC-KO): proteomic profiling analysis using iTRAQ-QSTAR. Immunohistochemistry, western blotting, ELISA and Taqman RT-PCR were performed to confirm iTRAQ data.
Results
Controlled iTRAQ analysis enabled the identification of 900 proteins, 600 of those were identifiable in all samples. Our analysis revealed 125 proteins whose levels were altered upon APC KO; 50 being up-regulated and 75 were down-regulated. Validation, IHC and WB on the murine samples for 23 selected upregulated proteins. In addition, HMGB1 elevation was detected in serum from day5 APCfl/fl mice.
HMGB1 expression was tested in 15 matched human colon cancer tissue samples using Taqman RT-PCR and showed up-regulation in all the samples (P= 0.0001).
Conclusion
Integration with the DNA microarray data only showed 40% agreement, highlighting the importance of a combined approach. Using proteomic and transcriptome data, HMGB1 has been identified as a potential early stage biomarker.
