| Background and objectives:Cancer/testis antigen (CTA) has became a research focus in tumor immunotherapy due to its highly frequent and strict expression in tumor, and the ability that it could induce specific and antineoplastic humoral and/or cellular immunity.The testes-specific protease 50 (TSP50) is a CTA. Its gene was discovered from a human testes cDNA library on a hypomethylated DNA fragment isolated from human breast cancer cells via the methylation sensitive-representational difference analysis technique. It encodes a threonine protease which is homologous to serine proteases, but its crucial catalytic triad has a substitution of threonine at the serine residue site. TSP50 is normally and specifically expressed in the spermatocytes of testes, abnormally activated and expressed in most (more than 90%) breast cancer biopsies. Knockdown of TSP50 gene expression could inhibit cell proliferation, colony formation and migration, induce cell apoptosis, and enhance cell sensitivity to doxorubicin, and the underlying molecular mechanisms might be related to activation of the NF-κB signaling pathway. Previous investigations found that TSP50 gene was negatively regulated by the p53 gene, and basic fibroblast growth factor (bFGF) could downregulate TSP50 expression via the ERK/Sp1 pathway due to TSP50 gene promoter containing Sp1 binding site. Therefore, these results imply that the TSP50 gene should be an oncogene. However, to our knowledge, there is no report that TSP50 has been detected in other human malignancies except breast cancer.Previous studies have demonstrated that the TSP50 gene promoter's DNA methylation status most likely control the gene expression in different types of tissues. DNA methylation is associated with TSP50 gene silencing in many normal tissues such as breast, lung and kidney. Conversely, DNA demethylation is associated with elevated levels of TSP50 gene expression in the testes and breast cancer. Global hypomethylation is common and prominent in colorectal carcinoma (CRC) as compared to normal colorectal tissue. Therefore, we speculated that TSP50 could be expressed in CRC. However, to date, the expression state of TSP50 gene in CRC and its clinical significance is unknown. We aimed to analyze the expression status of TSP50 in CRCs compared with colorectal adenomas and normal tissues, determine its relationship with clinicopathological parameters, and investigate its prognostic value for CRC patients based on tumor stage (early and advanced stage). In addition, p53 protein expression was examined to investigate its correlation with TSP50 expression in CRCs, and the prognostic significance of carcinoembryonic antigen (CEA), a well established prognostic factor for CRC, was analyzed to verify the reliability of this cohort of CRC patients.Methods:1. Total RNAs and proteins were extracted from CRC cell line SW480,SW620,LS 174T,HT-29,LoVo,CaCo-2 and HCT 116. TSP50 mRNAs and proteins were detected in the 7 CRC cell lines via RT-PCR and Western blot analysis.2. TSP50 proteins were detected in 8 pairs of CRC specimens and adjacent normal colorectal tissues using Western blot analysis.3. Immunohistochemical analysis of TSP50 with tissue microarrays composed of 95 CRCs, 20 colorectal adenomas and 20 normal colorectal tissues were carried out to observe TSP50 subcellular location (nucleus, cytoplasm or membrane) and analyze the relationship between TSP50 expression and type of colorectal tissues. Immunohistochemical stainings of TSP50, CEA or p53 were evaluated by staining intensity and/or the percentage of positive epithelial cells and graded as follows: -, +, ++ and +++. For statistical analysis, the - and + cases were pooled into the low-expression group, and the ++ and +++ cases were pooled into the highexpression group.4. Chi-square test was used to analyze the univariate associations of clinicopathological features (patient age and sex, and tumor location, size, depth of invasion, Lymph node metastasis, stage and grade) with the expression status of TSP50, p53 or CEA. Spearman's rho was carried out to analyze the correlation between the expression levels of TSP50 and p53 or CEA.5. The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value of TSP50 on distinguishing CRCs from colorectal adenomas and normal tissues. The sensitivity, specificity, positive predictive value, negative predictive value and Youden index (sensitivity + specificity - 1) were calculated. 6. The overall duration of survival was measured from the date of surgery to the end of our study. 5-year survival rate and median live time were calculated. Deaths from CRC were the outcomes (events) of interest. Kaplan-Meier method, log-rank test, and multivariate Cox regression analysis were performed based on tumor stage to evaluate the prognostic value of TSP50, p53, CEA and clinicopathological features in CRC patients.Results:1. Aberrant expression of TSP50 was detected in CRC cell lines SW480,SW620,LS 174T,HT-29,LoVo,CaCo-2 and HCT 116 by RT-PCR and Western blot analysis.2. TSP50 was expressed in all the 8 CRC samples, and not or weakly expressed in the adjacent normal colorectal tissues. TSP50 expression levels were obviously higher in 7 CRC samples than those in the adjacent normal colorectal tissues.3. Immunohistochemical analysis of TSP50 expression in colorectal tissue microarrays displayed: 65 in 95 (68.4%) CRC cases highly expressed TSP50, 3 in 20 (15.0%) colorectal adenoma cases highly expressed TSP50, and none of 20 colorectal normal tissues highly expressed TSP50. TSP50 expression levels in CRCs were significantly higher than those in colorectal normal tissues or adenomas (P<0.0001). TSP50 proteins were observed predominantly in the cytoplasm, but exhibited in the membrane and cytoplasm of some CRC samples.4. There was no significant correlation between TSP50 and p53 or CEA expression (P>0.05). There was also no significant association between TSP50 expression status in CRCs and all the clinicopathologic features including patients age and sex, and tumor location, size, depth of invasion, lymph node metastasis, stage and grade (P>0.05). p53 overexpression was significantly associated with tumor location (P=0.033), and CEA expression was negatively correlated with tumor grade (P=0.020).5. The sensitivity, specificity, PPV, NPV and Youden index of TSP50 overexpression to distinguish CRCs from colorectal adenomas and normal tissues were 68.4%, 92.5%, 95.6%, 55.2% and 60.9%, respectively.6. The 5-year survival rate and median live time of CRC patients with high TSP50 expression were 42.8% and 48 months (95% CI=30.804-65.196), respectively, but those parameters of CRC patients with low TSP50 expression were 73.3% and 108 months (95% CI=96.921-119.079), respectively. Kaplan-Meier survival analysis demonstrated that the survival of CRC patients with high TSP50 expression was significantly shorter than that of the patients with low TSP50 expression (P=0.010), specifically in patients who had early-stage tumors (stage I and II; P=0.004). Multivariate Cox regression analysis indicated that high TSP50 expression was a statistically significant independent risk factor (HR=2.205, 95% CI=1.214-4.004, P=0.009).Conclusions:1. TSP50 is abnormally highly expressed in CRC, and weakly or not expressed in colorectal normal tissues. The TSP50 expression levels in CRCs are obviously higher than those in the adjacent normal colorectal tissues.2. TSP50 overexpression distinguishes CRCs from colorectal adenomas and normal tissues with high specificity and PPV, it is an attractive and potential molecular target for CRC.3. High TSP50 expression in CRC could be a novel independent factor for unfavorable prognosis, especially for those with early-stage tumors. |
PDF Full Text Request