尿液lncRNA在前列腺癌早期诊断以及新型miRNA在进展机制中的研究

发布时间：2018-03-17 11:27

本文选题：前列腺癌　切入点：长链非编码RNA　出处：《第二军医大学》2016年博士论文　论文类型：学位论文

【摘要】：研究背景:中国前列腺癌(Prostate cancer,PCa)的发病率和死亡率急剧上升,其增长率位居肿瘤之首。同时,PCa的防治面临着两大临床难题,即早期诊断指标特异性差和晚期治疗方法效果不佳。因此,临床上亟需找到新型的敏感性和特异较高的PCa诊断标志物或诊断策略来代替血清PSA或弥补血清PSA在早期诊断上的不足;基础研究上亟需阐明PCa进展的关键分子机制,从而研发出抑制去势抵抗性前列腺癌(Castration-resistant prostate cancer,CRPC)的新型药物,实现CRPC治疗上的突破。研究目的:研究尿液中长链非编码RNA MALAT1和PCA3评分在前列腺初次穿刺患者中的诊断作用,建立基于尿液MALAT1或PCA3评分的综合诊断模型;探索新型mi R-n5在PCa进展中的作用机制,寻找CRPC干预的新靶点。研究方法:全转录组扩增试剂盒扩增尿沉渣RNA,q RT-PCR检测MALAT1和PCA3的表达,单因素logistic回归预测PCa的危险因素,多因素logistic回归建立PCa诊断模型,Spearman rank检测分析尿液MALAT1或PCA3评分与其它临床变量之间的关系,运用受试者工作曲线(Receiver operating characteristic curve,ROC)确定诊断指标的曲线下面积(Area under the ROC,AUC)、最佳截断值、特异度和敏感度,决策曲线分析(Decision curve analysis,DCA)来估算尿液MALAT1和PCA3评分的临床价值。Northern blot检测新型mi RNA的存在与表达,细胞增殖、迁移、侵袭和克隆形成实验来评价mi R-n5在进展性PCa细胞系中的生物学功能,生物信息学软件预测mi R-n5的靶基因,双荧光素酶报告基因法验证靶基因,Graph Pad Prism5进行数据展示并计算药物的IC50,皮下种植和胫骨内注射建立CRPC小鼠移植瘤模型。用SPSS v.17.0(SPSS Inc.,Chicago,IL,USA),Med Calc v.10.4.7.0(Med Calc Software bvba,Mariakerke,Belgium)和R软件包v.3.1.1(The R Foundation for Statistical Computing)对数据进行统计学分析,运用Hem I软件进行数据的Heatmap分析;以P0.05为差异有统计学意义。结果:尿液MALAT1和PCA3评分可以明确区分肿瘤组和穿刺阴性组患者并与PCa的检出率高度相关,但进一步的相关性分析发现,尿液MALAT1和PCA3评分与临床危险因素以及Gleason评分没有相关性。年龄、血清总PSA(Total PSA,t PSA)、前列腺体积(Prostate volume,PV)、游离PSA比值(Percentage of free PSA,%f PSA)和直肠指诊结果(Digital rectal examination,DRE)是总体人群的独立预测因子,而在PSA=4-10ng/ml(PSA“诊断灰区”)人群中,年龄、PV、%f PSA和DRE是其独立预测因子。尿液MALAT1评分在总体人群中的PCa诊断效能与t PSA相比虽然稍高但没有统计学意义(p=0.510),但在PSA“诊断灰区”人群中则明显优于t PSA(p=0.034)。PCA3评分在PSA“诊断灰区”人群中的PCa诊断效能则明显优于t PSA(p=0.0407)和%f PSA(p=0.046)。进一步的logistic回归分析得出,在发现组的PSA“诊断灰区”人群中,基于尿液MALAT1评分综合模型的预测准确性为79.79%和AUC为0.853,预测准确性比基础模型提高了5.32%,AUC提高了0.0318。将这一模型应用到验证组也得到了相同的结果。同样,在PSA“诊断灰区”人群中,基于尿液PCA3评分综合模型的AUC也高达0.819,高于基础模型的0.788。应用DCA,在PSA“诊断灰区”人群中,不管是在发现组还是在验证组,基于尿液MALAT1评分综合模型的临床预测能力都优于基础模型。将临床穿刺风险定为25%,发现组人群中基于尿液MALAT1评分综合模型的总获益率高于基础模型(13.97%vs.11.47%),且可以避免更多不必要的前列腺穿刺(47.32%vs.39.78%);验证组人群基于尿液MALAT1评分综合模型的总获益率同样高于基础模型(15.73%vs.13.11%),且可以避免更多不必要的前列腺穿刺(30.33%vs.22.47%),更为重要的是,综合模型没有漏诊1例高级别PCa而基础模型则漏诊1例高级别PCa。同样,在PSA“诊断灰区”人群中,基于PCA3评分综合模型几乎在所有临床穿刺阈值上优于基础模型,特别是在25%-40%这一区间显示出了更好的预测作用。应用30%这一临床穿刺阈值,基于PCA3综合模型可以避免60.3%不必要的穿刺活检,但以漏诊4例PCa为代价,其中2例为高级别PCa。此外,应用65对PCa及其癌旁正常组织进行RNA-seq发现300多个新型mi RNA,并使用Northern blot验证了其中表达量最高的8个新型mi RNA。通过在进展性PCa与惰性PCa细胞系中的表达情况比较,筛选出4个与PCa进展相关的新型mi RNA。再通过临床样本的验证,最终选取mi R-n5作为后续的研究对象。体外的功能实验表明,mi R-n5能够明显抑制进展性PCa细胞的增殖、迁移、侵袭和克隆形成能力,说明mi R-n5的缺失或低表达可能促进了PCa的进展。进一步的体内实验证明了肿瘤内注射mi R-n5能够抑制小鼠移植瘤的生长。此后,生物信息学分析并找到了mi R-n5的作用靶基因KDM6B,同时,通过双荧光素酶报告基因及q RT-PCR进行验证。进一步的研究表明,KDM6B的特异性抑制剂——小分子化合物GSK-J4能够抑制KDM6B的活性并明显抑制PCa细胞的生长,体内实验也证明了这一小分子化合物能够明显抑制小鼠移植瘤的生长。结论:尿液MALAT1和PCA3是PCa的独立预测因子,能明确区分前列腺穿刺阳性和阴性患者,基于尿液MALAT1或PCA3建立的临床综合诊断模型可大大提高PCa的临床诊断效能。尤其在PSA“诊断灰区”,尿液MALAT1和PCA3的诊断效能明显高于血清总PSA,基于尿液MALAT1或PCA3建立的临床综合诊断模型的预测能力高于应用临床危险因素建立的基础模型,并能够避免更多不必要的前列腺穿刺且不会漏诊更多的高级别PCa。此外,本研究证明了mi R-n5在体内和体外都能明显抑制进展性PCa的生长,它通过抑制其靶基因KDM6B的表达发挥抗肿瘤作用。同时,GSK-J4能够抑制mi R-n5的靶基因KDM6B的活性并明显抑制PCa细胞的生长。新型mi R-n5有望成为一个新的CRPC干预的靶点,而GSK-J4也有望成为治疗进展性PCa的一种新型靶向药物。
[Abstract]:Background: prostate cancer Chinese (Prostate cancer, PCa) incidence and mortality rates rose sharply, the growth rate ranked first in cancer prevention and treatment of PCa. At the same time, facing two major clinical problems, namely early diagnosis effect index of poor specificity and late treatment is poor. Therefore, the clinical need to find high sensitivity and specificity PCa diagnosis of new markers or diagnostic strategies instead of serum PSA or serum PSA in early diagnosis for lack of the basic research to clarify the PCa; the key advances in molecular mechanism, thus developed a suppression of castration resistant prostate cancer (Castration-resistant prostate, cancer, CRPC) of the new drug, to achieve a breakthrough in the treatment of CRPC. Objective: To study the effect of long chain non diagnostic MALAT1 encoding RNA and PCA3 score in patients with primary prostate puncture in the urine, urine MALAT1 or PCA3 score is established based on Ensemble Synthetic diagnosis model; to explore the mechanism of the new mi R-n5 in the progression of PCa, a new target for CRPC treatment. Methods: whole transcriptome amplification kit amplified urine RNA, expression of Q MALAT1 and PCA3 RT-PCR test, single factor Logistic regression prediction of PCa risk factors, logistic regression to establish PCa diagnosis Spearman model, rank test to analyze the relationship between urine MALAT1 or PCA3 score and other clinical variables, using the receiver operating curve (Receiver operating characteristic curve, ROC) to determine the diagnostic indexes of the area under the curve (Area under the ROC, AUC), the optimal cut-off value, the sensitivity and specificity of decision curve analysis (Decision curve analysis, DCA) to estimate the clinical value of.Northern blot mi RNA to detect new urine MALAT1 and PCA3 scores of the existence and expression, cell proliferation, migration, formation of MI R experiment to evaluate the invasion and clone The biological function of -n5 in the progression of PCa cell lines, target genes were predicted by bioinformatics software mi R-n5, dual luciferase assay validation of target gene, Graph Pad Prism5 for data display and calculation of the drug IC50, the establishment of CRPC transplantation tumor mouse model of subcutaneous implantation and intratibia injection. Using SPSS (SPSS Inc., v.17.0 Chicago, IL, USA), Med Calc v.10.4.7.0 (Med Calc Software BVBA, Mariakerke, Belgium) and R (The R Foundation software v.3.1.1 for Statistical Computing) for statistical analysis of data using Hem I software for data analysis by P0.05 Heatmap; the difference was statistically significant. Results: the urine MALAT1 and PCA3 the score can clearly distinguish the tumor group and negative group patients with PCa puncture and the detection rate was highly correlated, but further correlation analysis found that the urine MALAT1 and PCA3 score and the clinical risk factors and Gl The score of eason. No correlation between age, serum total PSA (Total PSA, t PSA), prostate volume (Prostate volume, PV), the ratio of PSA (Percentage of free free PSA,%f PSA) and rectal examination results (Digital rectal examination, DRE) were independent predictors of the overall population, and in the PSA=4-10ng/ml (PSA "diagnostic gray zone") population, age, PV,%f PSA and DRE are the independent predictors of MALAT1 score PSA. Urine PCa diagnostic efficacy and T in the general population than although slightly higher but not statistically significant (p=0.510), but in PSA "diagnostic gray zone" in the crowd was significantly better than t PSA (p=0.034) PSA ".PCA3 score in the diagnosis of PCa diagnostic gray zone" in the crowd was significantly better than t PSA (p=0.0407) and%f PSA (p=0.046). Further logistic regression analysis, found in the group of PSA diagnosis of "gray zone" in the population, based on urine MALAT1 score model The prediction accuracy is 79.79% and AUC is 0.853, the prediction accuracy is 5.32% more than the base model, AUC improved 0.0318. applied this model to the validation group also got the same result. Similarly, in the PSA diagnosis of "gray zone" in the crowd, urine PCA3 scores model of AUC is as high as 0.819 based on the above basis the application of 0.788. model in DCA, PSA diagnosis of "gray zone" in the crowd, no matter is found in groups or in the validation group, the clinical prediction ability of urine MALAT1 scores were better than the model based on the basic model. The clinical risk of puncture was 25% and the total benefit based on MALAT1 score found urine synthesis model is higher than the rate of group based model the crowd (13.97%vs.11.47%), and can avoid more unnecessary prostate biopsy (47.32%vs.39.78%); the total benefit groups based on the verification of the integrated model of the rate of urine MALAT1 score higher than that of basic models (15.73% Vs.13.11%), and can avoid more unnecessary prostate biopsy (30.33%vs.22.47%), more importantly, no comprehensive model of misdiagnosis of 1 cases of high grade PCa based model of missed diagnosis in 1 cases of high grade PCa., PSA in diagnosis of "grey zone" in the crowd, PCA3 score of the comprehensive model in almost all clinical on the basis of model based on the threshold of puncture is better than 25%-40%, especially in this interval showed better predictors. The clinical application of 30% puncture threshold, PCA3 model can avoid unnecessary biopsy of 60.3% based on 4 cases of misdiagnosis but with the price of PCa, including 2 cases of high grade PCa. in application 65 pairs of PCa and normal tissues were found in RNA-seq more than 300 new mi RNA, and use Northern blot to validate the expression level of the highest of the 8 new mi RNA. through the situation of comparative expression in the progress of PCa and inert PCa cell lines, screen Select the new mi RNA. 4 and PCa related through the validation of clinical samples, the final selection of MI R-n5 as a follow-up study. Functional experiments in vitro showed that MI and R-n5 could significantly inhibit the progression of PCa cell proliferation, migration, invasion and clone formation ability, lack of MI or low expression of R-n5 may promote the progress of PCa. In vivo experiments proved that intratumoral injection of MI R-n5 can inhibit tumor growth. Since then, the bioinformatics analysis and find the target genes of KDM6B, MI and R-n5, were verified by dual luciferase reporter gene Q and RT-PCR. Further studies show that the activity of a specific inhibitor of KDM6B the small molecule compound GSK-J4 can inhibit KDM6B and inhibit the growth of PCa cells in vivo experiments have proved that the small molecular compounds can inhibit the growth of transplanted tumor in mice The growth of MALAT1 and PCA3. Conclusion: the urine was an independent predictor of PCa, can make a clear distinction between positive and negative prostate biopsy in patients with clinical diagnosis of urinary MALAT1 or PCA3 model established can greatly improve the diagnostic efficiency of PCa based on PSA. Especially in the diagnosis of "grey zone", the diagnostic performance of urine MALAT1 and PCA3 were significantly higher than those of the serum total PSA based application of clinical risk factors for the predictive ability of clinical comprehensive diagnosis model established by the PCA3 or MALAT1 in urine was higher than that based on high level PCa. and can avoid more unnecessary prostate biopsy and not missed more in addition, this study demonstrated that MI R-n5 could inhibit the progression of PCa in vivo and in vitro growth, its antitumor effect by inhibiting the expression of the target gene of KDM6B. At the same time, the target gene KDM6B GSK-J4 can inhibit the activity of MI R-n5 and inhibit PCa cell Growth. The new mi R-n5 is expected to be a new target for CRPC intervention, and GSK-J4 is also expected to be a new target drug for the treatment of progressive PCa.

【学位授予单位】：第二军医大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：R737.25

【相似文献】