| Colorectal cancer is a common cause of cancer death in China, whose morbidity has increased over the decade. It is generally believed that the development and progression of colorectal cancer is a complex multi-step process involving numerous molecular events, such as activation of oncogenes and inactivation of suppressor genes. Though many genes have been reported to be closely related to the progression of colorectal carcinogenesis, they are far from enough to explain the transformation and still a lot of other genes, known or unknown, remain to be discovered.Three subtracted cDNA libraries were constructed by SSH in our lab. They were colonic adenocarcinoma - normal mucosa cDNA subtracted library (subtracted T-N library), colonic adenoma - normal mucosa cDNA subtracted library (subtracted A-N library), and colonic adenocarcinoma - adenoma cDNA subtracted library (subtracted T-A library). Some of the genes screened out from the three libraries, such as IGFBP7 and Reg IV, had been further studied and many meaningful results had been obtained.To further identify and analyze the candidate differentially expressed sequences in subtracted T-N library, we constructed a Linux-based platform to analyze nucleic sequences automatically. The serial softwares and non redundant nucleotide databases are obtained from the authors by email or downloaded from the free Internet resources. We programmed to combine the softwares and the databases in the Linux operation system to make a platform which could remove the vector and repeat sequences, assemble these sequences, then BLAST with the databases, and finally export the results to Excel form. One hundred ESTs in the subtracted T-N library were analyzed by the platform and the method based on Internet (that is, removing the vector and repeat sequences, the submitting to BLAST with the public database of NCBI ) respectively. 15 contigs were obtained from the subtracted T-N library and when BLAST them with the nr00, nr01 databases, we found most of them were matched to known genes which were closely related to carcinogenesis such asIGFBP7; or to known genes that had not reported to do with rumors such as DIO2. Some sequences were not matched to any known genes in the database such as contig2 and contigll, which implied the possibility of new genes and a new area valuing further investigation. Most of the resulted from the two methods were identical. The Linux-based nucleotide analytic platform surpassed the Internet-based one, in term of the high throughput and strong secrecy.Considering the underlying defects of a single technique and the differential situation of single patient not deductive to the general, differentially expressed clones obtained by SSH need to be validated by other methods such as reverse Northern Blot, RT-PCR and so on. We used Digoxin-labeled probes to screen the clones in the subtracted T-N library by reverse Northern Blot. In the 91 blot signals, there were 21 up-regulated (signals of the adenocarcinoma stronger 2 times than those of normal mucosa). The rate of differentially expression was 23.1%. We selected two of the differentially expressed genes (THY, eIF4A), to analyze their levels of expression by semi-quantitative RT-PCR in 18 samples of colonic adenocarcinoma. It was found that they were both up-regulated in colonic adenocarcinoma and had statistical significance (THY: P=0.018; eIF4A: P=0.005).In summary, we drew the conclusion in this thesis as follows: l.The construction of the Linux-based automatic analysis platform of nucleic acids is successful. 2. Two genes, THY and eIF4A, were found to be up-regulated in colonic adenocarcinoma. 3. Involvement and possible significance of known and unknown genes show the complexity of colonic carcinogenesis. It is necessary to have a comprehensive study of these genes, which might connect and interact to form a module. It will throw light on the mechanisms underlying the malignancy and will be proved to have a new role in the future oncology research. |
PDF Full Text Request