Screening Of Genes Relalated To Carcinogenesis Of Mammary Gland Of TA2 Mice And Study On The Mechanisms Of Its Multiple-organ Metastasis

Objective:1. To investigate the expression profiles of spontaneous breast cancer tissues and their matched normal mammary gland tissues and to identify the candidate differentially expressed genes (candidate genes) between these two kind of tissues by different analyzing methods. To investigate the effects of some of the candidate genes on the carcinogenesis of mammary gland of TA2 mice and to provide theoretic bases for human breast cancer, especially for familial breast cancers.2. To investigate the effects of CXCR4, CXCL12, integrin a4, integrinαL, integrinβ2, and L-selectin on the multi-organ metastasis of spontaneous breast cancer of TA2 mice and to provide theoretic bases for the estimation of potency of tumor metastasis and the discovery of new therapeutic targets for metastasis in clinic.Methods:1.The expression profiles of spontaneous breast cancer tissues and their matched mammary gland tissues were detected by Affymetrix Mouse Genome 430 2.0 chips. The expression data were analyzed by MAS5.0, Array2BIO and BGX methods. The proper cut-off of each methods was acquired by compared the results of these three methods. The relationship between these differentially expressed genes identified by all three methods and cancer were judged by articles from PubMed database and notation of Gene Ontology.2. Imprinted gene Decorin (DCN), oncogene EGFR and Cyclin D1 were chosen to confirm their expression in mammary gland tissues of 5-6 month old TA2 mice (Group A), spontaneous breast tissues (Group C) and their matched normal mammary gland tissues (Group B) by immunohistochemical and Real-time PCR methods.3. The expression of CXCR4, CXCL12, L-selectin, Integrinα4,αL andβ2 in spontaneous breast cancer with multiple-organ metastasis (Group A) and without multiple-organ metastasis (Group B), and the metastatic tumors of Group A was detected by immunohistochemical and Real-time PCR methods.Results: 1. Results of Affymetrix expression array1) 3416 candidate genes were identified by MAS5.0 method(P<0.0025), and 1681 of them were up-regulated in tumor tissues.661 candidate genes were identified by Array2BIO method(P<0.01), and 217 of them were up-regulated in tumor tissues. 2456 candidate genes were identified by BGX method,1104 of them were up-regulated in tumor tissues. The result contained 420 TypeⅠplots,646 TypeⅡplots, and 1390 TypeⅢplots. The ranks were between 24.01 and 385.02. The upper bound of the three type plots were decreased from TypeⅠplot to TypeⅢplot.2) 260 genes were identified by all three methods and it took up 10.59%(BGX method),39.33% (Array2BIO method), and 7.61% (MAS5.0 method) respectively. The 260 genes contained 170 TypeⅠplots,74 TypeⅡplots, and 16 TypeⅢplots. The constituent ratio was significantly different between the 260 genes and the total genes identified by BGX method (χ2=368.29, P<0.0001). Most of the 260 genes were TypeⅠandⅡplots. Genes with TypeⅠandⅡplots had a higher possibility to be identified by all three methods. Choosing genes with TypeⅠandⅡplots could reduce the candidate genes by 56.60%.94.23% and 95% of 260 genes were identified by Array2BIO and MAS5.0 methods with P<0.00015 as the cut-off, while the candidate genes could be narrowed down by 10.59% and 21.60%, respectively.60.38% of the 260 genes were related to carcinogenesis, tumor progression, metastasis, prognosis and therapy, and 68 genes were related to human or mouse mammary gland tumors.2. Expression of DCN, EGFR, Cyclin D1 and PCNA in normal mammary gland tissues and tumor tissues1) The immunohistochemistry results showed that there was a significantly progressive decrease in the occurrence of tissues with high decorin expression when comparing Group A (high expression in 50%) to Group B (high expression in 17.86%) (χ2=4.35; P=0.037). Same tendencies were showed for tissues with high EGFR or nuclear EGFR expression when comparing Group B to Group A (χ2=7.56,20.49; P<0.01, respectively), and the high expression rate in Group B was significantly higher than that of Group C (χ2=4.38; P<0.05). The cyclin D1 labeling index in Group C was higher than that in Group B (Z=-2.25, P<0.05). In Groups B and C, the cyclin D1 labeling index was higher in samples with nuclear EGFR expression than in samples without nuclear EGFR expression (Z=-2.28,-2.07, P<0.05). A positive correlation was found between the Cyclin D1 labeling index and the expression level of nuclear EGFR in Groups B and C (rs=0.723,0.474, P<0.05), but no correlation was established between nuclear EGFR expression and the PCNA labeling index.2) The results of Real-time PCR showed that samples of Group A expressed the highest level of DCN mRNA (0.95±0.25) and samples of Group C expressed the lowest level of DCN mRNA (0.13±0.10).Its expression was progressive decreased from Group A to Group C. Samples of Group B expressed the highest level of EGFR mRNA (0.05±0.02) and samples of Group A expressed the lowest level of EGFR mRNA (0.02±0.01), that of Group B was significantly higher than that in Group A (Z=-3.56, P<0.05). Samples of Group C expressed the highest level of Cyclin D1 mRNA (0.42±0.22) and samples of Group A expressed the lowest level of Cyclin D1 mRNA(0.04±0.01). Its expression was progressive increased from Group A to Group C. Samples of Group B expressed the highest level of PCNA mRNA (0.38±0.24) and samples of Group A expressed the lowest level of PCNA mRNA (0.14±0.10), that of Group B was significantly higher than that in Group A (Z=-3.56, P<0.05).3. Pathologic character of breast cancer with multiple-organ metastasis of TA2 mice The metastatic abilities of these 7 cases were different. Some of them lost the metastatic abilities with the increased passage in vivo. Stable multiple-organ metastasis could be observed from only one of them after they were transplanted for more than 30-passage in vivo. Apparent metastatic tumor could be observed 3w after inoculation at the mammary gland fat pads. Vasculogenic mimicry could be observed in the primary and metastatic tumor of the 7 cases, and it was significantly higher than that in tumor without multiple-organ metastasis.4. Expression of CXCR4, CXCL12, L-selectin, and Integrin a4,αL and (32 in primary tumor with multiple-organ metastasis (Group A) and without multiple-organ metastasis (Group B), and the metastastic tumor.1) The results of immunohistochemistry showed that the average positive cell percentage of L-selectin, CXCL12, and integrinαL in primary tumor of Group A and Group B were 39.81% vs.16.75%,55.49% vs.33.13%,24.81% vs.10.47%, respectively. Those of Group A were significantly higher than those in Group B (Z= -2.41,-2.14,-2.31 respectively; P<0.05), while those of CXCR4, integrin a4 andβ2 didn't change significantly. Only the expression of L-selectin in liver metastatic tumor was significantly higher than that in primary tumor(Z=-2.24, P=0.025). Some metastatic tumors were negative for integrin aL and CXCR4 staining.2) The results of Real-time PCR showed that the expression level of L-selectin and integrin aL mRNA were significantly higher than those of Group B (Z=-3.32,-3.47, P=0.001). The expression level of Integrin a4 andβ2 mRNA didn't changed significantly between these two groups.Conclusions:1. Combining of different methods or single method with proper cut-off can narrow down the candidate differentially expressed genes when you analyze Affymetrix expression array and this can elevate the accuracy of verification, although it can lose some genes related to carcinogenesis.2. The expression of decorin, EGFR and cyclin D1 of mammary epithelial cells changes with increasing age. With increased age, decreased expression of decorin and increased expression of EGFR, especially nuclear EGFR, may have caused increased proliferation of mammary epithelial cells partly by increasing the expression of Cyclin D1. The abnormal expression of decorin, EGFR and cyclin D1 are one of the reasons for the mammary epithelial cells' high proliferative activivies fo middle and old aged TA2 mice, and might be partially responsible for their carcinogenesis.3. Metastatic abilities of some of the tumors are unstable. Some of them loss the high metastatic ablility. Only the expression of L-selectin might be related to liver metastasis among the expression of CXCR4, CXCL12, integrin a4, aL andβ2. The expression of these substances is not the key reason for their multiple-organ metastasis. The tumor cells can express integrin or selectin exclusively expressed by leucocyte or lymphocyte, cell fusion may emerged during the carcinogenesis, and this may promote multiple-organ metastasis. Vasculogenic mimicry can be observed both in the primary and metastatic tumor, especially in small metastatic tumor, and this may be one of the reasons for rapidly growth of metastatic tumor.