| BACKGROUND:Lung cancer, especially non-small cell lung carcinoma, has become the global serious health problem in21st century, for the incidence and mortality rate have been increasing fast during the past decades. Although comprehensive treatment programs have led to improved efficacy at present, the5-year survival rate is still less than20%. Tumor invasion and metastasis is the leading cause of death in patients with lung cancer.Most patients with lung cancer prognosis is poor, so early detection, early diagnosis, early treatment is essential for improving the cure rate and survival rate.HAb18G is a recently identified hepatomaassociated antigen and its association with tumor growth, invasion, and angiogenesis has been studied in variety of tumors.Studies indicates that HAb18G play an important role in invasion and metastasis mainly via regulating fibroblasts as well as tumor cells themselves to disrupt hepatocellular carcinoma(HCC) microenvironment and can be a drug target for preventing metastasis.Its antibodies, especially LICARTIN, can effectively inhibit HCC tumor growth and metastasis in vivo and may be used as a promising drug for antimetastasis and recurrence therapy of HCC.HAb18G belongs to a member of immunoglobulin family. Its cDNA and amino acid sequences are identical to a new member of CD147. CD147, also known as extracellular matrix metalloproteinase inducer(EMMPRIN), basigin, M6, and tumor cell-derived collagenase stimulating factor, has been shown to be involved in the progression of malignancy by regulating expression of vascular endothelial growth factor(VEGF) and matrix metalloproteinases(MMPs).By analysis of HAb18G molecular structure, some scholars have pointed that it is a unique IgC2-I domain arrangement of immunoglobulin superfamily(IgSF) members, and that it could represent the general structure of the other CD147family new members. Immunohistochemical staining showed that the protein was mainly localized with the cell membrane or cytoplasm, studies have shown that overexpression HAb18G prompt cancer patients with poor prognosis, but the scholars did not notice whether HAb18G localization in cancer cells is associated with patients’ clinicopathological characteristics and prognosis or not. However, studies of CD147expression on the prognosis of lung cancer reported is inconsistent. Therefore, This study is to explore the role of HAb18G in non-small cell lung cancer by quantitative and semi-quantitative analysis of its expression and cellular localization, and make more reliable judgments of the prognosis of lung cancer patients.This study use RNA interference to explore the effect of SiHAb18G on its downstream gene MMP-2、MMP-9and VEGF, and biological behaviors of lung cancer cell line. We also cocluture SiHAb18G lung cell A549and human lung fibroblasts to detect the effect of HAb18G on fibroblasts by stimulating MMP-2and MMP-9.Our previous studies showed that low expression of PTEN(Phosphatase and tensin homolog deleted on chromosome ten) in has a relationship in the development of lung cancer. So whether PTEN and HAb18G has relationship in non-small cell lung cancer or not, which is one of our studies.OBJECTIVE:1. To explore the expression of HAb18G in non-small cell lung carcinoma, including its expression level and cellular localization. Analysis of their significance in the progression and prognosis of non-small cell lung carcinoma. 2. To explore the relationship between PTEN and HAb18G in non-samll cell lung cancer.3. To explore the influence of HAb18G on its downstram gene MMP-2, MMP-9and VEGF, and and its coculture with fibroblasts on stimulation MMP-2and MMP-9.4. To explore the effects of HAbl8G on biological behaviors of lung cancer cell line A549.METHODS:1. Immunohistochemistry SP methods were performd to detect the expression of HAb18G protein in208non-small cell lung cancer,19benign lung tissues and41normal lung tissues, the intensity (positive unit, PU) of HAb18G was assessed quantitatively to analysis the relationship between HAb18G PU value and Patients’ clinicopathological characteristicsusing by using ImagePro Plus image analysis softwore.2. For evaluation of the prognosis, we used semi-quantitative analyisis, negative(no positive staining at all) or weak (1+positivity in any of the epithelial cells or2+positivity in<30%of the epithelial cells) expression was defined as low expression; moderate (2+positivity in≥30%of the epithelial cells or3+positivity in≤50%of the epithelial cells) or strong (3+positivity in>50%of the epithelial cells) expression as high expression. Staining results were evaluated independently by two investigators without any prior knowledge of each patient’s clinical information.3. The expressions of HAb18G protein in cancer cell lines were detected by Immunohistochemistry SP methods, and its mRNA level was assessed by quantitive real time qRT-PCR. So, we can decide the cell lines that express HAb18G higher to perform RNA interference.4. SiRNA targeting HAb18G was transiently transfected into A549cell. Then we detected the inhibition efficiency by quantitive real time PCR. mRNA level of MMP-2, MMP-9and VEGF were detected by quantitive real time PCR, and the protein level was detected by Western blot. After coculture A549cell with fibroblasts, qRT-PCR was used to detect the effect of HAb18G on fibroblasts to secret MMP-2, MMP9in mRNA level.5. The mobile ability of A549cell was detected by wound healing assay. MTT was used to detecte the cell proliferation. Ki-67were detected by Immunohistochemisty. Tunel assay was used to detected the degree of apoptosis afer SiRNA targeting HAbl8G.6. All statistical analysis were carried out with the use of SPSS13.0statistical software package. One-way ANOVA test was used in the comparison of multiple sample averages. When the variances are heterogeneous, Dunnett’s T3test were used for multiple comparisons between groups. Otherwise, LSD test were used; Paired samples used paird samples’t test; Two independent-sample t-tests were used between the two groups; Correlation of non-normal distribution between two variable used Spearman method; Five-year survival rate was checked with Pearson χ2text. Kaplan-Meier method was used for univariate survival analysis, log-rank test was for the significance of differences in survival, and Cox proportional hazards regression model was for the identification of relevant prognostic factors. A2-sided P<0.05was considered statistically significant in all the analysis.RESULTS:1.HAb18G protein in non-small cell lung cancer, benign lung disease and normal lung tissuesThe expression of HAb18G in non-small cell lung cancer tissue was mainly localized in the lung cancer cell membrane or cytoplasm, which were clearly brown-yellow stained. It showed low or no expression in keratin pearls of lung squamous cells. It was almost no expression in the non-neoplastic internal controls and benign control tissues. It showed a small amount expression of HAb18G in mucosal epithelium cells and wall glands epthelial cells in bronchietasis.In208NSCLC tissues,35cases(16.8%) showed low expression,173cases(83.2%) showed high expression.94(45.2%) expressed in the membrane, including51squamous cell carcinoma,33adenocarcinoma and10large cell lung carcinoma;106(51.0%)expressed in the cytoplasm, including39squamous cell carcinoma,62adenocarcinoma and5large cell lung carcinoma;8(3.8%) showed no staining, including7adenocarcinoma and1large cell lung carcinoma.PU value in non-small cell lung carcinoma was obviously higher than that in non-neoplastic internal controls and benign control tissues(t=-25.34, P=0.000). There were no differences between non-neoplastic internal controls and benign control tissues(t=-1.43. P=0.158). In208non-small cell lung carcinoma tissues, there were difference among different types of lung cancer(F=8.51, P=0.000), PU value in squamous cell carcinoma was higher than that in adenocarcinoma (P=0.000). PU value in squamous cell carcinoma and large cell carcinoma were similar(P=0.407), it also was similar in adenocarcinoma and large cell carcinoma(P=0.171).In200non-small lung carcinoma with staining HAb18G, membranous localization of HAb18G PU value was higher than that of cytoplasmic localization(t=8.76, P=0.000). PU value in membranous localization of squamous cell carcinoma and adenocarcinmoma was higher than that in cytoplasmic localization (t=6.51,P=0.000; t=2.26, P=0.000; respectively), but membranous localization and cytoplasmic localization did not show significant differences statistically in large cell carcinomoma (t=8.76, P=0.864). In membranous group, PU showed differentce among different types of non-small cell lung cancer(F=4.08. P=0.02), PU value of HAb18G in squamous cell carcinoma was higher than that in adenocarcinoma and large cell carcinomoma(P=0.025and P=0.027, respectively), PU was similar in adenocarcinmoma and large cell carcinmoma(P=0.464). In cytoplasmic group, PU value was basically the same among squamous cell carcinoma, adenocarcinoma and large cell carcinmoma(F=2.53, P=0.085).2.HAb18G expression in different adenocarcinoma subtypesHAb18G PU was associated with was adenocarcinama subtypes(F=34.45, P=0.000), it was increased in lepidic predominant, papillary predominant, acinar predominant, micropapillary predominant, solid predominant with mucin production.3.Association of expression and localization of HAb18G protein with its clinical pathological characteristics of non-small cell lung cancerStatistical results of the IHC showed that the quantitative expression of HAb18G was correlated with differentiation(t=-5.26, P=0.000), lymph node status(t=-2.45, P=0.015,), distant metastasis status(t=-5.58.P=0.000) and clinical stage(F=22.47, P=0.000). PU was higher in moderately to poorly differention, primary tumor with lymph node metastasis or distant metastasis and high clinical stage. The quantitative expression of HAb18G was also associated with gender, history smoking, timor position and tumor diameter(t=2.21,P=0.028; t=-2.71, P=0.007; t=3.38, P=0.000and t=-4.44. P=0.000, respectively). PU of HAb18G was higher in male, patients with history smoking, and tumor located in the central.IHC staining was performed on13tissue sections of lymph node metastasis deposits in NSCLC. The pathologists evaluated and compared the expression levels of HAb18G in lymph node metastasis tissues with the corresponding primary cancer tissues. The expression of HAb18G in lymph node metastatic tissues were almost equal to that in the primary tumor tissues (t=-0.34, P=0.74).In200non-small cell lung cancer with stained HAb18G, HAb18G PU in NSCLC with membranous localization was associated with tumor differention (t=-3.26, P=0.003). Both of cytoplasmic localization and membranous localization were associated with tumor stage(t=5.79. P=0.004; t=7.72,P=0.000, respectively) and distant metastasis(t=-2.71, P=0.014; t=-3.46, P=0.001, respectively).4.Association with patient’s prognosisThe prognostic value of HAb18G expression and localization in136NSCLC patients with5-year follow-up data was analyzed.5-year overall survival rate in NSCLC was30.9%(42/136) in this study.The survival rate of HAb18G for low expression at the5th year was75.0%(18/24), and for high expression was21.4%(24/112), difference in survival rates between the two groups was obvious (χ2=26.57, P=0.000). Results of survival analysis using the Kaplan-Meier method indicated that the overall survival time in136patients was significantly shorter in patients with high expression of HAb18G than that in low expression (P=0.000, log rank test). Kaplan-Meier also indicated that the overall survival time in squamous cell carcinoma and adenocarcinoma patients with low HAb18G expression was significantly higher than that in patients with high HAb18G expression (P=0.001and P=0.003, log rank test, respectively). Patients with high expression of HAb18G also predicts poor prognosis in tumor stage I and stage Ⅰ-Ⅲ(P=0.001and P=0.01, log rank test, respectively), whereas patients of stage Ⅳ showed a poor prognosis both in low and high expression(P=0.306, log rank test).Five year survival rate was85.7%(6/7),45.2%(33/73) and5.4%(3/56) in patients with no staining, cytoplasmic and membranous localization respectively. difference in survival rates among this three groups was obvious (χ2=33.97, P=0.000). The intensity in membranous expression of HAb18G was associated with poor prognosis, while the intensity in cytoplasmic expression predicted a favorable outcome for136NSCLC patients (P=0.000. log rank test). Membranous expression of HAb18G predicted poor prognosis in both squamous cell carcinoma and adenocarcinoma(P=0.000,P=0.004, respectively, log rank test).5.Cox proportional hazards multivariate analysisBased on the results of multivariable Cox proportional hazards analysis, adjusting for age, sex. smoking history, tumor size, position, differentiation, lymph node status, distant metastasis, TNM stage, expression and localization, it was showed that lymph node status, distant metastasis status, expression and localization of HAb18G were independent prognostic indicators for patients of lung cancer(P=0.019, P=0.000, P=0.000, P=0.003, respectively, Cox regression analysis).The final choice of the four-variable model results suggest that lymph node metastasis, distant metastasis, HAb18G expression and localization are independent prognostic factors affecting overall survival (P=0.012, P=0.000, P=0.000, P=0.001, respectively).6.Receiver operation characteristic curve(ROC) curve for HAb18G of non-small cell lung cancer detected by IHCROC curve showed that the sensitivity and specificity of HAbl8G for diagnosis of non-small cell lung cancer by IHC is better(P=0.000), the area under the ROC curve is0.989.7.Correlation of PTEN and HAb18G expression in non-samll cell lung cancerPTEN positive intensity in lung carcinoma were significantly lower than that in non-small lung carcinoma (t=4.67,.P=0.000), it was not associated with histological differentiation, lymph node metastasis and TNM stage. PTEN in peripheral types were higher than that in central types(t=2.46,P=0.016).HAb18G positive intensity in non-small lung carcinoma were significantly higher than that in control tissues(t=-17.37, P=0.000), and it was related to histological differentiation, lymph node metastasis and TNM stage(P<0.05). PTEN was correlated negatively with HAb18G expression(r=-0.312, P=0.002).8. Effects of HAb18G SiRNA on MMP-2、MMP-9and VEGF expression in lung cancer cell A549The mRNA expression of HAb18G was decreased by94%afer SiRNA interfering48hours(t=-92.96, P=0.000). The mRNA expression of MMP-2、MMP-9and VEGF was decreased by62%(t=-10.92, P=0.000),70%(t=-7.27, P=0.000) and94%(t=-45.32, P=0.000) afer siHAb18G respectively. Western blot showed that HAb18G protein level was decreased by40%afer siHAb18G48h later(t=58.55, P=0.000), MMP-2protein level was decreased by34%afer siHAb18G48h later(t=27.23, P=0.000), MMP-9protein level was also decreased by31%afer siHAb18G48h later(t=28.64, P=0.000) and VEGF was decreased by35%(t=27.25, P=0.000).9.Effects of HAb18G on MMP-2、MMP-9expression of coculture lung cancer cell with fibroblastExpression of MMP-2mRNA level in coculture negative control A549lung cancer cell with fibroblast was increased56%(t=-174.20,P=0.000) than culture negative controlA549cell, and the increased was77.36%(t=-426.03, P=0.000)when coculture SiHAb18G lung cancer cell A549with fibroblast. MMP-2was increased57%(t=-178.16, P=0.000) in Negative control A549cell with fibroblast than SiHAb18G lung cancer cell A549with fibroblast. Expression of MMP-9mRNA level in coculture lung cancer cell A549with fibroblast was increased79%(t=-199.14, P=0.000) than culture A549,and the increased was91.67%(t=-311.81, P=0.000) when coculture SiHAb18G lung cancer cell A549with fibroblast.MMP-9was increased40%(t=-72.39, P=0.000) in Negative control A549cell with fibroblast than SiHAb18G lung cancer cell A549with fibroblast.10.Effects of HAb18G SiRNA on migration, proliferation and apoptosis of human lung cancer cellMigration distance of SiHAb18G lung cell A549was slower than control group(t=11.39, P=0.000), and the number of migration cells was smaller than negative control group(t=36.22, P=0.000). MTT proliferation of lung cell A549was decreased after SiRNA targeting HAb18G at24h,48h, and72h, compared with blank group(P<0.05), it also was decreased by41%,43.72%,41.98%, compared with negative control group(P<0.05). ki67was also decreased after siRNA targeting HAb18G(t=17.14, P=0.000) and degree of apoptosis was increased compared with control group(t=-8.85, P=0.000).CONCLUSIONS:1. Increased HAb18G was associated with differentiation, TNM stage, lymph node metastasis, tumor size history smoking and gender. HAb18G expression was also associated with different adenocarcinoma substypes. HAb18G with higher PU value was predominantly localized at the tumor cell membranes than cytoplasms. Both of membranous and cytoplasmic localization was linked to TNM stage and distant metastasis. Membranous localization was associated with tumor differentiation. In136non-small lung cancer with follow-up data, univariate analysis indicated that patients with HAb18G high expression, membrous localization predicted poor prognosis. Multivariate analysis showed that lymph node metastasis, distant metastasis, expression, and cellular localization were independent predictors of patient survival.2. There is a negative correlation between HAb18G and PTEN. The decrease or deletion of PTEN expression may be related to the abnormal activation of HAb18G protein, and paly an important role in the initiation and the development of non-small lung cancer.3. Down-regulation of HAb18G may participate in the development of lung cancer by stimulating fibroblasts as well as tumor cells themselves to express MMPs and VEGF. 4. Expression of HAb18G can promote migration and proliferation of lung cancer cell, and inhibit apoptosis to prompt tumor development. |
PDF Full Text Request