【作者】 李金平；

【导师】 肖涛；

【作者基本信息】 中南大学 ， 临床医学， 2013， 博士

【摘要】 目的：1ncRNA在基因调控中起重要作用,目前它对骨肉瘤的发病机制的影响仍不清楚。本研究利用微矩阵基因芯片技术研究1ncRNA在骨肉瘤和癌旁组织表达谱的差异并初步探讨其生物学作用。方法：分别取9例骨肉瘤癌组织和对照癌旁组织,提取总RNA后进行质量测定,合格后杂交合成cDNA后经过Microarray基因芯片扫描仪扫描后得到原始数据,原始数据经过统计软件转换后进行统计分析,得到两种组织之间1ncRNA和mRNA的差异表达谱。分别选取上调的1ncRNA6个：ASLNC21868,ASLNC22124, ASLNC23844, ASLNC24587, BE503655, BC050642,下调的1ncRNA4个：ASLNC00339, ASLNC11435, ASLNC13387, ASLNC18814采用SYBR GREEN I实时荧光定量聚合酶链反应(RT-PCR)验证,并用GAPDH做内参。综合NCBI RefSeq, UCSC, RNAdb, LncRNAs等数据库资源,对芯片结果进行Gene Ontology、Pathways分析等初步生物信息学分析,并通过计算Pearson相关系数(≥0.95)建立1ncRNA与靶基因的基因网络图。结果：将差异表达的标准设为：Folgchange值≥2,P值≤0.05,发现两种不同组织芯片表达有明显差异。骨肉瘤组织与对照癌旁组织相比,在检测的1ncRNA表达谱中,9例样本平均上调的lncRNAs为3528个(从2723到4537),平均下调的1ncRNAs为3958(从3469到4368),其中在9例样本都上调的1ncRNAs403个,都下调的1ncRNAs798个,上调倍数最大的是ASLNC02724,7.23倍,下调倍数最大的是ASLNC05129,27.39倍。上调的1ncRNAs略少于下调的1ncRNAs。在检测的mRNA表达谱中,9例样本平均上调的mRNAs为2604个(从1394到3710),平均下调的mRNAs为2344(从1513到2867),其中在9例样本都上调的mRNAs986个,都下调的mRNAs933个。采用SYBR绿色荧光染料(SYBR GREENI)实时检测的聚合酶链反应(RT-PCR)验证表达上调的6个1ncRNAs: ASLNC21868, ASLNC22124, ASLNC23844, ASLNC24587, BE503655, BC050642,表达下调的4个1ncRNAs:ASLNC00339, ASLNC11435, ASLNC13387, ASLNC18814, RT-PCR结果显示ASLNC21868, ASLNC22124, ASLNC23844, ASLNC24587, BE503655, BC0506426个基因表达上调,ASLNC00339, ASLNC11435, ASLNC13387, ASLNC188144个基因表达下调,和芯片结果基本吻合。GO (Gene Ontology)分析显示,在上调表达的mRNAs中,cell adhesion、 extracellular matrix和protein binding分别在生物学进程(ontology: biological process)、细胞组件(ontology:cellular component)和分子功能(ontology:molecular function)三个方面富集程度最高,提示这些上调的mRNAs与这三者相关程度最高；在下调表达的mRNAs中,muscle system process、contractile fiber和structural constituent of muscle分别在生物学进程(ontology:biological process)、细胞组件(ontology:cellular component)和分子功能(ontology:molecular function)三个方面富集程度最高,同样提示这些下调的mRNAs与这三者相关程度最高。Pathways分析显示,在骨肉瘤组织中有32条通路(pathway)受下调的mRNAs调控,富集程度最高的是通路"Hypertrophic cardiomyopathy-Homo sapiens (human)”由24个目的基因组成；有34条通路(pathway)受上调的mRNAs调控,富集程度最高的是通路"ECM-receptor interaction-Homo sapiens (human)"由22个目的基因组成。我们选取了6个表达下调的lncRNA:ASLNC00339, ASLNC04683, ASLNC08248, ASLNC11435, ASLNC13387, ASLNC23339和6个表达上调的1ncRNA:ASLNC02419, ASLNC21868, ASLNC23844, ASLNC24079, BE050642, BE503655,通过计算与目的基因的Pearson相关系数(≥0.95),得到227个目的基因并共同建立编码-非编码基因共表达网络(coding-noncoding gene co-expression network)。结论：(1)1ncRNA microarray基因芯片是筛选骨肉瘤差异表达基因一种理想有效的方法；(2)首次利用1ncRNA microarray基因芯片揭示在骨肉瘤中存在异常表达的1ncRNA基因,利用实时荧光定量PCR对异常表达的部分1ncRNA基因进行验证进一步提高了芯片结果的可信度,生物信息学初步分析显示这些1ncRNA在骨肉瘤的发生发展中发挥复杂的调控作用：(3)1ncRNA可能是全新的骨肉瘤肿瘤标记物和潜在的基因治疗靶点更多还原

【Abstract】 Objective:Long noncoding RNAs （lncRNAs） have a critical role in genome regulation. The contributions of lncRNAs to osteosarcoma remain unknown. Here, we describe the expression profile of lncRNAs in osteosarcomas compared with paired adjacent noncancerous tissue using microarray analysis and study their biology funtions prelininarily.Methods:Firstly, we extracted total RNA from nine clinical samples of primary osteosarcoma and their paired adjacent noncancerous tissues.Labeling and hybridization of RNA were precessed and we got the differentially expression profiles after a series of procedures. Secondly, Six over-regulated lncRNA and four under-regulated from the differentially expression profiles were validated by RT-PCR in nine clinical samples of primary osteosarcoma and their paired adjacent noncancerous tissues, using GAPDH as inter reference. Thirdly, Bioinformatic analysis (gene ontology analysis, pathway analysis was used for further research on the differentially expression profiles based on NCBI RefSeq, UCSC, RNAdb, LncRNAs databases and we constructed a coding-noncoding gene co-expression network that included the differentially expressed lncRNAs and targeted coding genes by calculating PCC （PCC≥0.95）.Results:Compared with adjacent noncancerous tissues, the lncRNA and mRNA expression profiles are significantly differentially expressed in the primary osteosarcoma tissues. An average of3528lncRNA （range from2723-4537） was significantly over-regulated and an average of3958lncRNA （range from3469-4368） was significantly under-regulated （foldchange≥2.0, P≥0.05）.403lncRNAs were consistently over-regulated and798lncRNAs were consistently under-regulated in all samples analyzed. ASLNC02724was the most over-regulated lncRNA （foldchange:7.23） and ASLNC05129was the most under-regulated lncRNA （foldchange:27.39）. Over-regulated lncRNA was more common than under-regulated lncRNA. An average of2604mRNA （range from1394-3710） was significantly over-regulated and an average of2344mRNA （range from1513-2867） was significantly under-regulated （foldchange≥2.0, P≦0.05）.986mRNAs were consistently over-regulated and933mRNAs were consistently under-regulated in all samples analyzed. NM₀13227was the most over-regulated lncRNA （foldchange:33.17） and NM₀05963was the most under-regulated lncRNA （foldchange:170.32）. Over-regulated mRNA was equal to under-regulated mRNA. Six over-regulated lncRNAs:ASLNC21868, ASLNC22124, ASLNC23844, ASLNC24587, BE503655, BC050642and four under-regulated lncRNAs:ASLNC00339, ASLNC11435, ASLNC13387, ASLNC18814were validated by RT-PCR in in nine clinical samples of primary osteosarcoma and their paired adjacent noncancerous tissues. The results demonstrated that ASLNC21868, ASLNC22124, ASLNC23844, ASLNC24587, BE503655and BC050642were over-regulated and ASLNC00339, ASLNC11435, ASLNC13387and ASLNC18814were under-regulated in the osteosarcoma samples compared with normal samples. The RT-PCR results and microarray data are consistent. We found that the highest enriched GOs targeted by the over-regulated lncRNAs were cell adhesion （ontology:biological process）, extracellular matrix （ontology:cellular component） and protein binding （ontology:molecular function） and that the highest enriched GOs targeted by under-regulated lncRNAs were muscle system process （ontology:biological process）, contractile fiber （ontology:cellular component） and structural constituent of muscle （ontology:molecular function）. Pathway analysis indicated that34pathways corresponded to over-regulated transcripts and that the most enriched network was "ECM-receptor interaction-Homo sapiens （human）" composed of22targeted genes. Furthermore, this analysis showed that32pathways corresponded to under-regulated transcripts and that the most enriched network was "Hypertrophic cardiomyopathy-Homo sapiens （human）" composed of24targeted genes. We constructed a coding-noncoding gene co-expression network that included six over-regulated lncRNAs: ASLNC02419, ASLNC21868, ASLNC23844, ASLNC24079, BE050642, BE503655and six under-regulated lncRNAs:ASLNC00339, ASLNC04683, ASLNC08248, ASLNC11435, ASLNC13387, ASLNC23339and227targeted coding genes. The lncRNAs and coding genes that had Pearson correlation coefficients equal to or greater than0.95were chosen to construct the network.Conclusions:（1） LncRNA microarray analysis was ideal and effective way to screen the differentially expression profiles in osteosarcoma;（2） Our results were the first to reveal differentially expressed lncRNAs in osteosarcoma and some of the differentially expressed lncRNAs were validated by RT-PCR to promote the credibility of results. Prelininary bioinformatic analysis revealed that lncRNA was involved in the occurrence and development of osteosarcoma;（3） lncRNAs may be novel candidate biomarkers for the diagnosis of osteosarcoma and potential targets for gene therapy更多还原