Screening And Identification Of Gene Expression Profile And Signal Pathways Related To Colorectal Cancer Stem Cells

Background/ObjectiveColorectal cancer is one of the major malignant diseases with high incidence in the world. Every year about 1 million cancer cases and 500,000 cancer deaths are estimated to have occurred worldwide. The incidence rate of colorectal cancer in male patients is higher than it in females. The highest incidence rates are found in many western countries. In China, the incidence rate of colorectal cancer also showed an upward trend year by year. Drug resistance, early relapse and metastasis result in poor prognosis and high mortality rate of colorectal cancer. Therefore, to clarify the pathogenesis of colorectal cancer, find specific targets for its therapy, and enhance the efficacy of radiotherapy and chemotherapy are essential for treatment of this disease.It was discovered in tumor reaserch that not all tumor cells can produce a new tumor, only a small part of cells with self-renewal and strongly tumorigenic ability can do so. These cells are considered to be cancer stem cells. Cancer stem cell theory subverts the people's traditional understanding of tumor, which is a milestone in the history of cancer research. Cancer stem cells were found in variety of tumors, including breast cancer, glioma, colorectal cancer, prostate cancer, lung cancer, pancreatic cancer, gallbladder cancer and so on. Malignant behavior of cancer, such as infiltration, invasion, metastasis and radiotherapy and chemotherapy resistance, may be related to cancer stem cells. However, the molecular mechanism of colorectal cancer stem cells are still not clear. Different gene expression may be the main cause for the unique biological characteristics of colorectal cancer stem cells. Therefore, analysis of gene expression profile of colorectal cancer stem cells and exploring the potential cell signaling pathway may be useful for understanding the mechanism of colorectal cancer stem cells and also essential for finding potential targets for eradicating colorectal cancer stem cells.Materials and Methods1. Collection of tissue samples of colorectal cancer After collection of 74 cases of colorectal cancer and adjacent non-cancer tissue, samples were pretreated by both methods:(1) into the tissue preservation fluid,-80℃low temperature refrigerator; (2) regular 10% neutral formalin fixed.2. Immunohistochemical detection of CD133 protein and proteins related to MAPK signal pathway in colorectal cancer tissue Tissue dehydration, transparent, dipping wax, embedded, producer, dewaxing, hydration, antigen retrieval, respectively, with a combination of resistance and the second antibody reaction, DAB was color, hematoxylin, dehydrated slices, transparent and steal.3. Colorectal cancer cell culture 6 types of colorectal cancer cell lines including SW480, Colo205, SW620, Lovo, HCT116 and HT29, were cultured in 10% FBS RPMI-1640 medium.4. Analysis proportion of CD133+ cell subgroup by Flow cytometry (1) Detecting the propotion of CD133+ cells in colorectal cancer cell lines:106 cells were collected and labeled with PE-CD133/1 antibody, while cells labeled with PE-IgG as blank control. (2) Detecting the purity of CD133+ cells after isolated:The method was the same, but cells were labeled with PE-CD133/2 antibody.5. Isolation of CD133+ cells by MACS MicroBeads Cells were label with CD133-coupled magnetic beads, separated in the magnetic field, and eluted remain CD133+ cells in the column. This process was repeated for the second time isolation.6. CD133+ colorectal cancer stem cells culture CD133+ cells were cultured in DMEM/F12 medium, added with B27(1:50), EGF(20ng/ml) and bFGF(10ng/ml). Tumor mammospheres were collected by short-term low-speed centrifuge. When the tumor mammospheres grew to 75-100μm, it was mechanically and enzymatically dissociated into a single-cell suspension. CD133+ cells were fitting for tumorigenicity test in NOD/SCID mice only after a week time culture.7. Xenograft transplantation assay in NOD/SCID mice After having set up multiple concentrations, CD133+ cells and CD133- cells with equal amounts (104) were injected s. c. in left and right abdominal wall of NOD/SCID mice, respectively. We compared tumorigenic capacity of two subsets at the same concentration.8. Analysis of gene expression profiles of CD133+ cells and CD133- cells by gene chip RNA samples of CD133+ and CD133- cells were preparated to make gene chip after extracted. After hybridization, elution, detection, image acquisition and data analysis, we found differentially expressed genes. Different gene functions and related signaling pathways were analyzed by MAS platform.9. Detecting expression level of genes by RT-PCR Extraction of total RNA of all cell samples, synthesis of cDNA, PCR reactions, agarose gel electrophoresis, semi-quantitative analysis of p53, AKT1, TGFBR2, ABCB6, CDC2, CHEK1, CHKN1A and CCNB1 mRNA level of gene expression.10. Testing expression levels of proteins by Western Blot Extraction of the cells and tissue total protein, determination of protein concentration, denatured protein, protein electrophoresis on the sample, transfer film, closed, with a combination of resistance and the second antibody reaction respectively, the final chemical ECL Detection of light and semi-quantitative analysis the expression level of ERK1/2, JNK, p38, phosphorylated-ERKl/2, phosphorylated-p38, phosphorylated-JNK, phosphorylated-cdc2, phosphorylated-p53, phosphorylated-Chek1, phosphorylated-Chek2, phosphorylated-Rb1 and phosphorylated-Rb2.11. Statistical analysis Numeration data were presented as mean±standard deviation(SD), while measurement data were presented by percentage(%). Analysis was completed with a statistical software package for desktop computers (SPSS 13.0). Using Mann-Whitney and Kmskall-Wallis test to analyze the relationchip of CD 133 expression and clinical features of colorectal cancer. CD133+ cell proportion analyses were performed with One-Way ANOVA analysis.Gene expression levels were analyzed by independent sample T test. P< 0.05 (two-sided) was considered significant.Results1. CD133 expression in colorectal cancer tissue 25 cases were CD133 positive in 74 cases of colorectal cancer specimens, the positive rate was 33.8%,8 of them were strongly positive. The expression of CD 133 was positive in tumor tissue but negative in the corresponding adjacent non-tumor tissue. We have not found any significant differences of CD 133 expression in different gender, age, location and histological type of tumor(P>0.05). The different positive rates of CD133 expression in high, medium and poorly differentiated colorectal cancer was extremely significant (P< 0.05). The positive rates of CD133 were 15.4%(2/13),33.3%(12/36) and 58.8% (10/17), respectively. The positive rate of CD133 was highest in poorly differentiated tumor tissues. CD133 expressed significantly different in colorectal cancer with different TNM stage (P<0.05). The positive rates of CD 133 inⅠ,Ⅱ,ⅢandⅣstage were 15.4%(2/13),40.0%(12/30),19.0%(4/21)and 70.0%(7/10), respectively. The positive rate of CD133 was highest in stageⅣ, and it was also significantly higher in metastasis cancer than that in non-metastasis one (P<0.05)2. The proportion of CD133+ cells in colorectal cancer cell lines The proportion of CD133+ cell were higher in SW480, SW620, and Lovo cell lines, but lower in the other three ones. The proportion of CD133+ cells was the highest in SW480 cell line which occupied 0.916%.3. The proportion of CD133+ cells after separation After one time separation, the purity of CD133+ cells was about 30%, which was more than 80% after the second time separation.4. CD133+ cells grow as tumor mammospheres CD133+ cells suspended in serum-free medium and grew as single-cell at beginning. In the first 2-3 days, most cells began to amplify, while part of cells was dead. We observed cell debris at the bottom of petri dish. On the fifth day, cells continued to grow, and formed mammospheres with good refraction and round shape. We could not accurately distinguish cell shape and cell-cell boundaries. After the size of mammosphere arrived to 75-100μm, it could be passaged. As the passage number increased, we found the sizes of mammospheres were almost the same and trending towards spherical ones. When CD133+ cells were re-cultured in serum medium, they adhered to the wall and formed monolayer cell.5. The oncogenicity of CD133+ cells was stronger than CD133- cells. Each NOD/SCID mouse was injected s. c. in the left and right abdominal wall with the same number of CD133+ and CD133- cells (104). We found tumor in left side abdominal of NOD/SCID mice in 2 weeks after subcutaneous injection of CD133+ cells. Tumor size was larger than lcm in 3-4 weeks. But in the right side of mice abdomen no turmor was found.6. Gene expression profiles of CD133+ stem cells We have detected 21,522 genes, and found 414 differently expressed genes was found in CD133+ and CD 133 cells, in which 185 genes were up-regulated in CD133+ cells, while 229 genes were down-regulated. BMI1, which was up regulated in CD133+ cells, was over-expressed in many tumors. BMI1 might play an important role in maintaining self-renew ability of colorectal cancer stem cells. Several kinds of ABC transporters were also upregulated in CD133+ cells, including ABCB6, ABCC2, ABCA7, ABCF3 and so on, which might be related to drug resistance of colorectal cancer stem cells. Moreover, the expression level of anti-apoptosis genes was higher in CD133+ cells, such as IRF-3, IRF-7 and DAPK3. We presumed that the anti-apoptotic effect of colorectal cancer stem cells was exerted by activating or inhibiting special cell signal pathways. These genes were likely relevant to the malignant behavior of colorectal cancer stem cells.7. Gene function analysis The function of different genes between CD133+ and CD133- cells associated with biological processes ocuppied 55.6%, including transcription regulation, cell cycle, mitosis, transcription, cell division, protein amino acid phosphorylation and so on. Genes related to cells composition accounted for 16.1%, including nucleus, cytoplasm, cytosol, cell membrane, the membrane integrity structure, composition and other cytoplasmic membrane, mitochondria and so on; In addition, genes related to molecular function were 28.3%, including protein binding, nucleic acid binding, transferase activity, ATP binding, zinc ion binding, GTP binding and activity of transcription factors and so on.7. Relationship between differential expressed genes and cell signal pathways Different gene expression in CD133+ and CD133- cells were mainly distributed in the cell cycle signaling pathway, p53 signaling pathway, tight junction signaling pathway, glioma signaling pathway, pancreatic cancer signaling pathway, Calcium signaling pathway, insulin signaling pathway, prostate cancer signaling pathway, proteasome signaling pathway, GnRH and MAPK pathway and other important signaling pathways.6 different genes involved in p53 pathway (CDKN1A, CCNB1, CHEK1, PMAIP1, CCNG1 and CDC2), while the other 6 different genes related in MAPK signal transduction pathway (TGFBR2, AKT1, PLA2G4B, JMJD7-PLA2G4B, CACNA2D2, RRAS2 and RAP1A).9. QRT-PCR verified microarray results We used QRT-PCR which was quantitative and more accurate to determine the reliability of the gene chip results. The mRNA levels of p53, AKT1, TGFBR2 and ABCB6 were increased in CD133+ cells, compared with it in CD133-cells, the difference was extremely significant (P< 0.05). The mRNA levels of CDC2, CHEK1, CHKN1A and CCNB1 were significantly downregulated in the CD133+ cells (P<0.05). The results were consistent with the gene chip results, which showed that the microarray results were reliable.10. The expression of MAPK signaling pathway related protein in CD133+ and CD133- cells We have not see any differences of ERK1/2, JNK and p38 protein expression in two cells. However, our results showed that the expression level of phosphorylated ERK1/2 and phosphorylated p38 protein were higher in CD133+ cells than that in CD133- cells, but not phosphorylated JNK protein. Immunohistochemistry results showed that ERK protein was expressed in both CD133- and CD133+ colorectal cancer tissue. The expression of phosphorylated ERK protein was weakly in CD133+ tumor tissue, but was loss in CD133- tumor tissue. p38 protein was positively expressed in CD133+ tumors, and strongly positive in some of them. But it was negative or only weakly positive in CD133 tumor tissue. We have not tested any expression of JNK protein and its phosphorylation protein in all samples.11. The expression of p53 signaling pathway related protein in CD133+ and CD133" cells The expression of p53 signaling pathway-related proteins, including phosphorylated p53 protein, phosphorylated cdc-2 protein, phosphorylated Chk1 (Ser68), phosphorylated Chkl (Tyr68), phosphorylated Rb (Ser795) and phosphorylated of Rb (Ser807/811) protein, were all inhibited in CD 133+ cells, but not in CD 133- cells.Conclusion1. CD 133 expression was related to the clinicopathological characteristics of colorectal cancer The expression of CD 133 which correlated with TNM stage and metastasis of colorectal cancer can be used to predict tumor malignancy. This also indirectly illustrated colorectal cancer stem cells might play a key role in the progression of tumors.2. CD133 could be used as a marker for screening colorectal cancer stem cells The growth characteristics of CD133+ cells under the serum-free culture system were consistent with the characteristics of cancer stem cells. CD133+ cell still showed the capability of differentiation in serum culture. It was confirmed in animal models that CD133+ cells had more powerful proliferation. This indicated that undifferentiated colorectal cancer stem cells were eriched in CD133+ cells. The majority of CD133-cells were differentiated whose proliferation was limited. From this we speculated that colorectal cancer stem cells might be one of the key initiating factors of colorectal cancer.3. Colorectal cancer stem cells have different gene expression profile Differential expressed genes between CD133+ and CD133- cells involved in multiple functions of proteins and metabolic pathways. It suggested the molecular mechanism of colorectal cancer stem cells was really complex.4. The relationship of MAPK and p53 signaling pathways and colorectal cancer Colorectal cancer stem cells might conduct its biological effects which included proliferation, differentiation, anti-apoptosis, drug resistance and so on, through activating MAPK signaling pathway and inhibiting of p53 signaling pathway. MAPK and p53 signaling pathways associated with the occurrence and progression of colorectal cancer were expected to be potential therapeutic targets for eradicating colorectal cancer stem cells.