No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Colorectal cancer (CRC) is the third most common cancer worldwide. Colorectal polyps are pre-cursors of CRC; however hyperplastic polyps (HPs) lack malignant potential. The aim of this thesis was to describe differences in gene expression and pathway activation between colorectal tissues (normal mucosa and polyp) from early stages of CRC development and to validate novel candidate genes identified by qRT-PCR. Differential gene expression was investigated in 48 colorectal tissues from the early stages of the adenoma-carcinoma sequence using DASL Whole-Genome expression microarrays and appropriate bioinformatics. Particular emphasis was placed on the comparison between Adenomatous polyps (APs) and HPs as lesions with and without malignant potential. In the comparison between HP and AP tissues 1633 significantly differentially expressed genes (DEGs) (p<0.05) and 33 pathways were identified, which confirms the fundamental differences between these polyps. Moreover, DEGs associated with Wnt-Signalling, MAPK Signalling, p53 Signalling, cell cycle and apoptosis were noted between HPs and APs. In addition, a novel network was created using COXPRESSdb, which found connections between genes comparing HP and AP tissues. Six candidate genes were selected based on their differential expression across the range of colorectal tissues; ASCL2, ANXA2, AXIN2, ETS2, G3BP1 and TFF2. qRT-PCR was employed to investigate expression of these candidate genes in a larger Validation Set (n=143) of colorectal tissues. With the exception of TFF2, significant differential expression was identified for all genes, which supported the results of the DASL microarray. Again the most significant differences identified were between HPs and APs. In conclusion, HPs and APs have different malignant potential, with associated differential gene expression profiles, which could be exploited for screening and to provide potential therapeutic candidates.

Colorectal cancer (CRC) is the third most common cancer in the UK and Adenomatous Polyposis Coli (APC) mutations are the most common genetic abnormality encountered in the sporadic form of this disease. In this study, we have used an Apc fl/fl/ mouse model which has a conditionally regulated Apc gene in the intestinal epithelium. Our hypothesis was that analysis of the changes in protein expression which occur in the intestinal epithelial cells of adult mice following Apc deletion would provide insights into signalling pathways which are also involved in early human colorectal carcinogenesis. The aims of this study were to use proteomic analysis to identify potential biomarkers for the early stages of CRC and to identify key proteins or processes involved in early colorectal carcinogenesis. iTRAQ-QSTAR proteomic analysis of intestinal epithelial extracts from Apc+/+ and Apc f//f/ mice identified 125 proteins which were differentially expressed, 50 being upregulated and 75 downregulated. We focused our efforts on the upregulated proteins as the detection of a positive signal is better for a biomarker. Ingenuity Pathway Analysis identified 19 proteins that could be detected in serum/blood, of which 13 were selected for further validation based on review of the literature. Immunohistochemistry, Western blotting and qRT-PCR identified 7 of these proteins as potential serum biomarkers of colorectal carcinogenesis. Integration of our iTRAQ data with a previous cDNA microarray performed using this mouse model identified c-Myc-dependent proteins. These included the 7 proteins identified in earlier studies and 4 additional proteins that might also play roles in colon carcinogenesis. The resulting 11 potential biomarkers were HMGB1, NCL, KRT18, RPL6, DDX5, PHB, SFRS2, FABP6, NAP1 L 1, NPM-1 and CBX3. These were further validated using another transgenic mouse model with the conditional regulation of both Apc and c-Myc genes (ApcfllfiMycfllfi), for which another iTRAQ (8- plex) analysis was performed. Confirmation studies were then carried out using the ApcMin/+ mouse model which shows more resemblance to human CRC. qRT-PCR studies comparing colonic polyp tissue samples from 6 month old ApcMin/+ mice and colonic tissue samples from their Apc+/+ wild-type counterparts showed increased expression of our candidate biomarkers in polyp tissue. For one of our candidates, HMGB1, a commercial ELlSA kit was then used to assess its serum concentration in various mouse models. This showed a statistically significant increase in serum HMGB1 concentration in Apc fl/fl mice compared to Apc+/+ mice. At 6 months of age, serum HMGB1 concentration was shown to be 1.69 fold higher in ApcMin/+ than in Apc+/+ mice. Our candidate biomarkers have also subsequently been validated using human serum and colonic tissue samples. The proteomic analysis of samples from mouse models with aberrant Apc expression has therefore generated a series of candidate biomarkers which have been demonstrated to be transferable to human CRC, thus validating our overall approach.

Adhesion of circulating tumour cells to the blood vascular endothelium is a pivotal step in metastasis. This study shows that the levels of free circulating galectin- 2, -3, -4, and -8, but not galectin-9 and -1, were markedly increased up to 31-fold in the bloodstream of colon and breast cancer patients and in particular those with metastasis. The presence in vitro of each of these galectins at pathological concentrations induced dose-dependent increases of cancer cell adhesion to monolayers of human macro- and micro-vascular endothelial cells, an effect that was abolished by the presence of galectin inhibitor, by pre- fixation of the cells, or by pre- treatment of the cells with O-glycanase to remove cell surface TF (Galβ1,3GalNAcα-) antigen. Suppression of the TF-expressing mucin protein MUCl by siRNA reduced, while overexpression of MUC 1 increased, the galectin-mediated cancer cell adhesion. Higher levels of circulating galectin-2 were associated with a significantly increased mortality risk in colorectal cancer patients and this association was diminished by serum co-existence of auto-anti-MUCl antibody specifically against the TF epitope of MUCl. Thus, the increased circulations of galectin members are common features in cancer and promote metastatic spread. Circulating galectins therefore represent a novel class of therapeutic targets for the development of effective agents to reduce metastasis and increase patient's survival. The possible role of modified heparins as inhibitors of the galectin3-ligand interaction that leads to increased vascular adhesion was therefore investigated. ELISA assays showed that chemically modified heparin derivatives successfully blocked galectin-3 adhesion to asialo-bovine mucin, and also galectin-3-mediated cellular adhesion to endothelial cell monolayers and extracellular matrix components, which suggests a possible role for heparin derivatives in cancer therapeutics. Finally in this study, the functional importance of Core 1 Gal-transferase (C 1 Ga1T) was investigated. It has long been presumed that there is a competition between Core 1 Gal-transferase (C1GalT), Core 3 G1cNAc-transferase (C3GnT) and sialyl-transferase (ST6Ga1NAc- T) for elongation of O-linked mucin-type glycans that initiate with GalNAcα-Ser/Thr. However, evidence that supports such a competition among these glyco-transferases is surprisingly lacking. This study shows that selective suppression of the CIGalT caused over 80% reduction of Galβ1,3GaINAcα- (Core 1, Thomsen-Friedenreich, TF antigen) expression in human colon cancer HT29 ary-based lecti. Suppression of Cl GalT was also associated with 198±8%, 136±24% 136±24% and 231±6% increase of sialyl-GalNAca- (sialyl-Tn), G1cNAcβ1 ,3GaINAcα- (Core 3) and GalNAcα- (Tn) expression in HT29 and 174±11 %, 155±37% and 200±5% increase in SW620 cells. These results provide direct evidence of a competition between CIGalT, C3GnT and ST6GalNAcT transferases for modification of the GaINAcα-Ser/Thr in O-glycan biosynthesis. As Tn, TF and sialyl- Tn are all oncofetal carbohydrate antigens and over-expressed in up to 90% of all human cancers, this information may also be useful for future development of glyco-transferase-targeted therapeutic strategies for cancer treatment.

There is accumulating evidence that the micro-environment plays a pivotal role in the progression of malignancy and that key stromal changes can be identified which influence this progress. This investigation focussed on two specific areas; fibroblast phenotype and the composition of the extracellular matrix, especially regarding the protein-tenascin-C (TN-C).;The heterogeneity within colonic fibroblast populations is becoming increasingly apparent. Using immunohistochemical techniques this study defined specific populations and identified consistent phenotypic changes in these cells in malignancy. In particular CD34 expression was lost in sub-mucosal fibroblast and alpha smooth muscle actin (alphaSMA) expression increased within tumour stroma. There was also evidence of progressive loss of high molecular weight caldesmon (hCD) expression by the pericryptal myofibroblasts with the development of malignancy. The stromal changes were specific to malignancy and were not demonstrated in pre-invasive adenomas.;TN-C is an important component of the extracellular matrix (ECM) which can occur as several different isoforms. Immunohistochemical techniques were used to demonstrate alterations in the distribution of TN-C between normal and malignant colon and the presence of isoforms containing exon 14 in both. The distribution of TN-C in pre-invasive adenomas was the same as that in normal colon but some sub-mucosal TN-C expression including exon 14 containing isoforms was observed in acute inflammation. Polymerase chain reactions (PCR) identified a total of 7 TN-C isoforms in colonic tissue with no obvious association between isoforms profile and disease.;TN-C can also be expressed by tumour cells and PCR demonstrated multiple isoforms in the colorectal tumour cell lines SW480, SW620, HCT15 and HT29. However, TN-C expression did not appear to correlate with invasive capacity as demonstrated in preliminary in-vitro invasion assays.