| Part I.Individualized pathway analysis and screening of differential expression genesBackgroundColon cancer(CC)occurs from the colon and usually origins as benign tumors,often existing as polyps and becoming cancerous over time,with high incidence and poor prognosis.This process is affected by genetic and environmental factors.Globally,it is the third most common cancer,accounting for 10%of all cancer deaths.It is estimated that there are 1.36 million new cases each year,and about 694,000 people die from the disease.Colon cancer is more common in developed countries,where more than 65%of cases are found.Currently,the extensively applied treatment management of colon cancer is the combination of surgery,radiotherapy,and chemotherapy.However,many patients acquire resistibility to the current existing therapeutic strategies.Thus,looking for more reliable and effective biomarkers for early screening and treatment options to improve cancer patients' prognosis and the postoperative survival rate is extremely important.It is shown that compared with normal tissue sample,there are hundreds of aberrantly expressed pathways in colon cancer tissues.There is a simple method that determines aberrantly expressed pathways through comparing tumor samples with the accumulated normal sample data.The basic principle is to quantify individual samples of abnormal pathways.Traditional individualized pathway analysis is a strict method,requiring exactly matched tumor and normal tissues from an individual patient.Khatri and coworkersdivided these methods into three categories:overrepresentation analysis(ORA),functional class scoring(FCS),pathway topology(PT).These path analysis methods are used to determine whose expression in tumor samples vary compared to normal tissue samples,and cannot determine a specific gene in different amount of different expression in the tumor samples,so the study of gene expression data classification has been limited.There are two studies on individual pathway analysis.PARADIGM is a tool that uses known functional structures to infer the state of a pathway.This method takes a set of interrelated variables as model.Variable interactions can describe the functional structure of the pathway,when using multiple levels of data analysis was outstanding performance.Because it works with known functional connections between gene DNA and protein,this method is not applicable to use a single level of data for analysis.Another method is pathway deregulation score(PDS)that the Drier and coworkers put forward,with a certain cancer samples and the main curve distance,the master curves drawn by normal sample median.To accurately calculate the PDS score,they used a whole group of samples from different stages of normal and/or cancer to reduce the dimensions and draw the optimal master curve.But calculating the PDS requires a series of data preprocessing steps,including selecting the number of principal components to use and filtering background genetic data to obtain the optimized master curve.The PDS needs to use the whole set of data to explain individual pathways,which is a drawback because it requires a large amount of data to extract the master curve to interpret data for a single patient.It also has a disadvantage when it comes to explaining a single sample,such as a patient's recurrent tumor and there is no corresponding set of data to extract the main curve.ObjectiveIn our study,individualized pathway aberrance score(iPAS)method was applied to the analysis of colon cancer samples,by virtue of comparing with the accumulated normal sample data.We took advantage of differential co-expression network(DCEN),accompanied wifh pathway analysis,to screen out the expression genes of the strongest associations.Materials and methodsWe first obtained the transcriptional data from ArrayExpress database,which contained 111 samples of colon tissue,including 56 colorectal cancer tissues and 55 peripheral non-tumor tissue samples.Linear models for microarray data(LIMMA)were used to filter differentially expressed genes.Furthermore,the enrichment pathway of differentially expressed genes was acquired through gene enrichment.Individual pathways of significance analysis,statistical samples of different pathways in the disease and the number and frequency of the differentially expressed in general.Finally,we based on the expression,the use of classical bayesian model to identify the common expression of genes on and get abnormal patterns of gene expression in total,constructing interconnected network structure,namely the difference between total express network.Calculation between groups within the data clustering coefficient,betweenness,build differences in all express network topology,for a total of express network structure that difference of abnormal genes related to express network(DCEN).Combined with individual pathways and express network,the difference of gene express network was analyzed,to further obtain potential regulation pathways within the relationship and screen most relevant genes.ResultsAccording to our screening criteria that p ? 0.01 and the log value of fold change should not be less than 2,a total of 485 DEGs were identified in patients with colon cancer.Among them,194 genes are up-regulated(with increased expression in colon cancer patients)and 291 genes are down-regulated(with reduced expression in colon cancer patients).Among the 485 DEGs,the most obviously up-regulated gene was interleukin 8(IL8),which is usually deemed as an expression variable of tumor necrosis factor;while the most obviously down-regulated gene was GCG,associated with a variety of intestinal diseases.The occurrence of colon cancer is caused by multi-gene and multi-step mutation.Therefore,the analysis of gene pathways and co-expression networks is more meaningful than the study of individual differential genes.Among them,245 genes are enriched in some special pathways,which are the intersection of 485 differently expressed genes and 1004 keypathways.The individualized pathway analysis found that three pathways,namely the chemical carcinogenic pathway,p53 signaling pathway and cytochrome P450 metabolic pathway,may be closely related to colon cancer.Through co-expression analysison 485 DEGs identified from 111 microarray datasets,we got a total of 4612 co-expressedgene pairs,with 480 DEGs were involved in.These results showed that our DCEN generally covers a multiple of signaling pathways,and there are interdependent functional relationships between the associated signaling pathways.There are many genes involved in different coexpression relationships,that is,the same gene may have a co-expression relationship with many different genes,and in multiple co-expression relationships.The top 5 genes are listed as follows:QPCT,SFN,SPINT1,GOT1,SLC22A18.All these genes are associated with cancer.The five differentially expressed genes in the difference pathway expression genes are listed as follows:SFN,GOT 1,CFI,STAM,VDR,and all are related to cancer.Among them,the connectivity of SFN is highest,and SFN was the center of co-expression network.Although SFN does not exist in chemical carcinogenic pathway and cytochrome P450,there was a complex relationship between SFN and these two pathways according to the circuit diagram.These three pathways are rich in differentially expressed genes,which may play an important role in the development of colon cancer,and most of the genes in the network are associated with the occurrence of many cancers.Conclusion1.Differential gene pathway analysis found that three pathways,namely the chemical carcinogenesis pathway,p53 signaling pathway and cytochrome P450 metabolic pathway,may be closely related to the pathogenesis of colon cancer.The individualized pathway analysis showed that the difference genes in p53 pathway may be related to the occurrence of colon cancer.2.SFN is the co-expression network center and has complex associations with multiple genes in the three pathways,which may play an important role in the pathogenesis of colon cancer.Part II.Expression and significance of SFN genes in colon cancer BackgroundTo sum up,the Stratifin(SFN)gene has a strong correlation with cancer occurrence,and it has obvious relationship with other genes,and it may be involved in the process of colorectal cancer because of its enrichment in many cancers related pathways,especially in p53 pathways,such as cell division.The expression of SFN gene in many types of cancer tissues is very different,which has potential effect on the incidence of colon cancer,breast cancer,cervical cancer,head and neck cancer and lung cancer.ObjectivesAccording to the conclusions we have derived from bioinformatics analysis,the SFN gene tends to be reduced in colorectal cancer.Therefore,we further study its expression in colorectal cancer and its related biological significance.MethodsThe tumor tissues and adjacent normal colonic tissues from 35 patients with colon cancer were collected in our hospital.Quantitative RT-PCR and western blot were used to detect the expression of SFN mRNA and protein levels respectively,to determine whether SFN was reduced in colon cancer tissue.Then we summarized the clinicopathological features of 35 patients with colon cancer,including sex,age,tumor size,tissue type,degree of differentiation and TNM staging,in order to determine whether the expression of SFN in colon cancer tissues is related to any of these clinicopathological features.We evaluated the expression of SFN in colon cancer cell lines and observed its effects on cell proliferation and apoptosis:three colon cancer cell lines and normal colonic epithelial cells were selected,the expression of SFN protein and mRNA in each cell was measured with western blot and quantitative PCR assays,to verify whether SFN is reduced in colon cancer cells in vitro.According to the experimental results,the representative colon cancer cell line was selected for follow-up experiment.We overexpressed and silenced SFN gene in SW480 cells.The CCK-8 assay was used to determine the growth of each cell at different perious after transfection.The apoptosis of each cell was determined by flow cytometry.The activity of apoptosis-related pathway was detected after SFN expression:it was known that SFN was enriched in p53 pathway and western blot was used to detecte p53 pathway activation.In addition,the expression of MDM2 and apoptosis factor Bax was detected to explore the molecular mechanism whereby SFN affecting cell survival.SW480 cells transfected with different plasmids were inoculated into nude mice.The weight of the nude mice was measured,and the length and width of the tumor were measured and the tumor volume was calculated.The cervical vertebrae were dislocated and the nude mice were executed,the tumor tissue was stripped out and used for subsequent trials:to compare the weight of subcutaneous transplanted tumor in each group,to calculate the tumor inhibition rate,to detect the expression of Bax and Ki67 in tumor tissue,and to detect the apoptosis of tumor cells by the method of in situ end labeling(TUNEL).ResultsSFN expression in colon cancer tissues and cells:in 35 patients with colorectal cancer,the expression of SFN at protein and mRNA level was significantly lower than that in the surrounding normal tissue,and the difference was statistically significant(P<0.05).The expression of SFN in colorectal cancer cells was validated in vitro:the mRNA and protein expression of SFN in SW480,SW620 and HCT116 colorectal cancer cell lines were significantly lower than that of NCM460 normal colorectal epithelial cells(P<0.05).Among them,SW480 cells were the most obvious,so we used SW480 cell line as a follow-up study of colorectal cancer cell model.SFN overexpression can inhibit cell proliferation,strengthen cell apoptosis and influence cell cycle in vitro:CCK-8 assay showed that PIRES-SFN cells were inhibited by transfection,and the proliferation of GIPZ and shRNA cells increased significantly in transfected cells.It is indicated that SFN inhibits cell proliferation in colorectal cancer cells.The analysis of V-FITC/PI flow cytometry showed that the proportion of apoptotic cells increased significantly after SFN expression.In addition,the apoptosis of GIPZ slow viral shRNA decreased,and the SFN gene in the cell was silenced,so it was shown that the SFN gene played an important role in apoptosis.The effect of SFN transfection on cell cycle was analyzed by flow cytometry:after 48 hours of SFN transfection,the proportion of cells in the G2/M phase increased.The number of G1 cells was significantly reduced by SFN overexpression.SFN by modulating p53 pathway activity inhibits cell vitality:Western Blot test results showed that the expression of SFN mRNA in SW480 cells increased and the proportion of p53 was increased,and the expression level of p53 decreased significantly,At the same time,the expression level of Bax protein in apoptosis was obviously increased.Since MDM2 can be combined with p53 to degrade it,and p53 phosphorylation form and acetylation are the basic forms of its activation,we judge that SFN overexpression can increase and activate MDM2 pathway by lowering p53 to regulate cell activity.The effects of SFN on tumor growth,apoptosis and related pathways were studied in vivo allogeneic transplanted tumor cells:There was no significant difference in the body weight of nude mice in each group.By recording the volume of subcutaneous transplanted tumor,the growth rate of SFN transfection group was significantly slower than that of the other two groups,which was statistically different.On the other hand,after one months of growth,the average tumor weight of the unchanged group reached 1.27g,the tumor cells transfected with SFN showed inhibitory effect on tumor,the inhibitory rate reached 65.35%,and compared with those in the blank control group,there was no significant difference between the transfection empty carrier group and the Transfection group.The results suggest that SFN has strong inhibitory effect on the growth of tumor cells in vitro.In the Tunel experiment,the cell nucleus was dyed with brown-yellow granules as apoptotic cells.The apoptosis rate was calculated,and the results showed that the apoptotic degree of SFN transfected cells was higher than that in other groups,which was statistically different from the other groups(P<0.05).The results showed that SFN can induce apoptosis of tumor cells.ConclusionThe expression of SFN gene in colon cancer tissues and cells was significantly down-regulated.SFN gene can inhibit the proliferation of colon cancer cells by regulating p53 pathway to induce cell cycle arrest in G2/M stage,inhibit cell proliferation,and induce apoptosis.Although SFN gene in the process of colon cancer mechanism is not clear,but with the development of the related gene technology,its role in the tumor could be further understanding,may become a new colon cancer gene therapy targets. |
PDF Full Text Request