| BackgroundNon-small cell ung cancer (NSCLC) is one of the most common cancers across theworld and consistutes approximately80-85%of all cases of primary lung cancers.Although surgical resection, chemotherapy, radiotherapy and molecular targeted therayrepresent the current standard of treatment for patients with NSCLC, the overall benefit ofthese patients is significantly limited due to cancer heterogeneity and multi-drug resistance.In conclusion, it is very urgent to elucidate the biological hehavior and pattern of NSCLC.Cumulative data show that the development and progression of NSCLC is closelyrelated to the desregulation of gene expression at transcriptional and post-transcriptionallevel. RNA-binding proteins (RBPs) are a family of newly discovered molecules that couldparticipate in the eukaryotic gene regulation via influencing mRNA stability atpost-transcriptional level. Human antigen R (HuR) and tristetraprolin (TTP) are twoimportant RNA-binding proteins of embryonic lethal abnormal vision (ELAV) family. Itwas reported that HuR and TTP play critical roles in carcinogenesis and tumor metastasisby interacting with the adenylate-uridylate rich elements (ARE) within the3’untranslatedregion (UTR) of mRNAs.Recently many studies have paid attention on miRNA functions. The miRNAs are akind of highly conserved sequences, and exert important roles in controlling eukaryoticgene expression that contributes to the cancer progression including cell proliferation,apoptosis and metastasis. Our previous results indicate that the drcrease or absence ofmiR-133b expression occurs in NSCLC tissues, which affects the outcome of patients withNSCLC. Considering the3’UTR of mRNA is the common recognition region for RBPsand miRNA, the two patterns could regulate each other, or cooperate in controlling somecritical molecules. ObjectiveIn the prestnt study, we sought to detect the expression of RNA-binding protein HuR,TTP and miR-133b in NSCLC tissues and control tissues, and analyze the prognostic valueof these molecules. Our study may provide evidence and explanations for the interacton ofRBPs and miRNA, also develop newly identified anti-cancer molecular targets.Meterials and methods1. A total of110formalin-fixed paraffin-embedded (FFPE) NSCLC tissues,80FFPEnon-cancer lung tissues,33fresh frozen (FF) NSCLC tissues and33FF paracancerouscontrol tissues were collected from General Hospital, Jinan Command of the People’sLiberation Army from January2010to August2013.2. HuR and TTP expression in110NSCLC tissues and paracancerous tissues weredetected by immunohistochemistry (IHC) method.3. Reverse-transcription polymerase chain reaction (RT-PCR) was used to detect theexpression of miR-133b in all tissues.4. Follow-up work were conducted through the outpatient physician visit or phone calltill March2014.The follow-up contents mainly contained therapeutic efficacy, relapse date,the therapeutic pattern after recurrence，survival time.5. Statistic analysis was performed by SPSS13.0. The data was represented by mean±SD. The correlation between two groups was analyzed by independent samples t test. Innonparametric statistics, Wilcoxon signed-rank test and Mann-Whitney U test were used.Enumeration data was analyzed by chi-square test. Kaplan-Meier methods, Log rank testand Cox’s proportional hazard regresssion were used to generate estimates of DFS and OS.A P value less than0.05was considered statistically significant.Results1. Cytoplasmic HuR positive expression in NSCLC tissues is higher than that inparacancerous tissuesThe positive expression rate of cytoplasmic HuR and nuclear HuR were61.82%(68/110) and100%(110/110) in NSCLC tissues, and3.64%(4/110) and98.18%(108/110)in paracancerous tissues, respectively.Cytoplasmic HuR expression was significant higherin NSCLC tissues than that in paracancerous tissues (P <0.0001). We did not findsignificant relationships between the cytoplasmic HuR expression and anyclinicopathological features.2. Cytoplasmic TTP positive expression in NSCLC tissues is lower than that inparacancerous tissues The positive expression rate of cytoplasmic TTP and nuclear TTP were5.45%(39/110) and35.45%(39/110) in NSCLC tissues, and49.09%(54/110) and29.09%(32/110) in paracancerous tissues, respectively. Cytoplasmic TTP expression wassignificant lower in NSCLC tissues than that in paracancerous tissues (P=0.028). Asignificant correlation between cytoplasmic TTP expression and pTNM stage was found (P=0.017), but there were no correlations between the TTP expression and sex, age, tumorsize, pathological type, differentiation stage, lymph metastasis, distant metastasis.3. The expression of miR-133b in FF NSCLC tisues is significantly decreasedcompared to that in FF paracancerous tissues, and was correlated with the advancedclinical stageIn33FF NSCLC tissues and33paracancerous tissues, the2-△CTvalues of miR-133bare0.0155±0.06616and0.0397±0.13634, and the△CT values of miR-133b are11.57±4.38and9.00±3.51, respectively, indicating that the miR-133b expression was lower inFF NSCLC tissues than that in FF paracancerous tissues (P=0.033). The miR-133bexpression level in FF NSCLC tissues was0.168-fold higher than that in FF paracanceroustissues.We used the relative expression value0.168of miR-133b in NSCLC as a standard;there were16cases with low miR-133b expression and17cases with high miR-133bexpression, respectively. No significant correlation was found between miR-133bexpression level and sex, age, tumor size, pathological types, differential degree, lymphmetastasis and diatant metastasis, but miR-133b expression in FF NSCLC tissuessignificantly correlated with advanced pTNM stage (P=0.032).4. The expression of miR-133b in FFPE NSCLC was significantly decreasedcompared to FFPE paracancerous tissues, and is corelated with tumor size and stageIn110FFPE NSCLC tissues and80non-cancer lung tissues, the2-△CTvalues ofmiR-133b were0.0041±0.0184and0.0100±0.0266, and the△CT values of miR-133bwere12.93±3.76and10.17±3.49, respectively, indicating that the miR-133b expressionlevel in FFPE NSCLC tissues was0.127-fold higher than that in FFPE non-cancer tissues.We used the relative expression value0.127of miR-133b in NSCLC tissue as astandard; there were56cases with low miR-133b expression. The decreased miR-133bexpression was closely related to lage tumor size (P=0.015) and advanced pTNM stage(P=0.032). There were no significante correlation between miR-133b expression leveland sex, age, pathological type, lymph metastasis and diatant metastasis. 5. Comparing the expression of miR-133b in FF NSCLC tissues and FFPE NSCLCtissuesIn29FF NSCLC tissues and29FFPE NSCLC tissues from the same cases, the2-△CTvalues of miR-133b are0.0160±0.07041and0.0103±0.03385, respectively. There is notsignificant difference regarding the miR-133b level between the two groups (P=0.443).6. The correlation between the cytoplasmic HuR expression, cytoplasmic TTPexpression and miR-133b expressionWe found a significant relationship between increased cytoplasmic expression of HuRand decreased cytoplamic expression of TTP (P=0.035). The increased cytoplasmicexpression of HuR also significantly related to the reduced expression of miR-133b (P=0.034). Patients with a low level of cytoplasmic TTP also had a low level of miR-133b (P<0.0001).7．The expression of HuR, TTP and miR-133b and NSCLC prognosisIn univariate analysis, patients with low miR-133b expression had significantly lowerdisease-free survival (DFS; P=0.025) and overall survival (OS; P=0.048). Unfortunately,cytoplamic HuR or TTP expression did not affect the outcome for NSCLC patients,although there was a tread towards significant difference.Conclusions1. There was a negative correlation in NSCLC tissues between these two RBPsexpression: cytoplasmic HuR and TTP expression.2. We detected lower miR-133b expression in FF NSCLC tissues or FFPE NSCLCtissues compared to normal tissues. The FFPE tissue can become a suitable sample sourcein the future study.3. The decreased expression of miR-133b was significantly with increasedcytoplasmic expression of HuR and decreased cytoplasmic expression of TTP, and is asignificant prognostic factor, suggesting that the interaction between RBPs and miRNAcould contribute to malignant biological behavior of NSCLC. However, large,well-designed prospective or retrospective studies with lage sample size are required toaddress some of these important issues. |
PDF Full Text Request