miR-9对鼻咽癌增殖、肿瘤干细胞“干性”、EMT和转移的调控作用及机制

  本文关键词：miR-9对鼻咽癌增殖、肿瘤干细胞“干性”、EMT和转移的调控作用及机制，，由笔耕文化传播整理发布。

        背景和目的鼻咽癌是一种发生于鼻咽粘膜的恶性肿瘤,其恶性程度较高,且具有极强的转移能力。我国是鼻咽癌发病率最高的国家,以广东、广西、湖南、福建等南方地区为著。早在80年代初,姚开泰就提出了鼻咽癌发病的三击／多步假说,认为EB病毒（EBV）、化学致癌物和胚性击中（即遗传因素或自发突变）是构成NPC的主要病因,不仅为一系列后续研究提供了理论基础,而且对鼻咽癌的防治具有重要的实践指导意义。生活方式已被证实和鼻咽癌的发病有关,尤其吸烟不仅是个体鼻咽癌发病的危险因素,也与EB病毒血清阳性的健康男性的发病相关,吸烟能诱导EB病毒的活动。因此,鼻咽癌的发生是多因素、多基因、多阶段、多途径的。鼻咽癌治疗方法包括放射治疗、外科手术治疗、化学药物治疗、免疫和生物治疗,基因靶向治疗也为鼻咽癌的治疗提供新的前景。2011年Weinberg指出肿瘤具有自给自足的生长信号、抗生长信号的不敏感、抵抗细胞死亡、潜力无限的复制能力、持续的血管生成、组织浸润和转移、避免免疫摧毁、促进肿瘤的炎症、细胞能量异常、基因组不稳定和突变十种特征,并综述了针对以上特征的靶向治疗方法。大多数实体瘤的肿瘤微环境中存在各种类型的细胞,如肿瘤细胞、炎症细胞、肿瘤干细胞等。肿瘤干细胞（CSCs）是指可以自我更新和分化成为不同表型的肿瘤细胞。肿瘤干细胞假说认为,肿瘤细胞中有一小部分细胞,导致肿瘤发生和生长,具有无限的自我更新和复制能力,并对放疗和化疗产生抵抗。不同肿瘤的肿瘤干细胞表面标记物不同。基于CSCs的特性,有效杀灭清除CSCs被认为是根治肿瘤的有效途径之一。目前针对CSCs的靶向治疗主要有靶向抑制其表面标志物,调节CSCs通路,诱导CSCs向肿瘤细胞分化,细胞治疗（如诱导生成y8T cells杀伤CSC）等。转移是恶性肿瘤最基本的生物学特征,肿瘤的转移是影响患者生存期的首要因素,已成为肿瘤研究领域的热点和难点,亦是迫切要解决的重大问题之一。肿瘤转移最突出的特征是不同的肿瘤能在人体内相同或不同的部位形成转移,即肿瘤转移的组织特异性。上皮-间充质转化（Epithelial-mesenchymal transition, EMT）是指上皮细胞通过特定程序转化为具有间充质表型细胞的生物学过程,是上皮细胞来源的恶性肿瘤细胞获得迁移和侵袭能力的重要生物学过程,是肿瘤转移的重要阶段。miRNAs是一种小的内源性非编码RNA分子,大约由21—25个核苷酸组成,在动物和植物,它们是重要的调节分子,每个miRNA可以有多个靶基因,而几个miRNAs也可以调节同一个基因,它们通过与其靶基因的相互作用来调节基因表达,进而调控生物体的生长发育。最近的研究表明,一些miRNA调节细胞增殖和凋亡的过程,在肿瘤的形成中是很重要的,一些miRNAs可作为癌基因或抑癌功能,miRNA表达谱可能成为肿瘤诊断的有用的生物标志物,而miRNAs可能是一个功能强大的工具,用于肿瘤的预防和治疗。研究表明,miR-9在多种人体肿瘤组织中异常表达,如miR-9在乳腺癌、胃癌、卵巢癌等中表达下调,但在神经系统肿瘤中表达上调。miR-9在肿瘤细胞增殖、凋亡、EMT、侵袭、转移等方面发挥重要作用。有研究人员的基因芯片结果显示,miR-9在人NPC组织标本上表达下调,但功能不详。我们的预实验结果显示,miR-9在人NPC细胞株中表达下调,但在NPC组织标本上表达却上调：同时,miR-9过表达抑制NPC细胞体内外增殖,而miR-9过表达却促进NPC细胞发生EMT和迁移。以上这些似乎自相矛盾的实验结果促使我们去深入解析miR-9在人NPC发病中的作用及其机制。方法第一章miR-9靶向调控CCNG1抑制鼻咽癌细胞的增殖1.质粒提取试剂盒提取质粒后,NaAC、乙醇沉淀法纯化质粒。2.稳定细胞株建立包装载体PMD2G和穿梭载体PPAX2与目的质粒lipotemin2000共转染至293FT细胞,得到的病毒上清以浓度梯度方式感染细胞,嘌呤霉素2μg/ml （GIBCO）作用约2周杀死未感染细胞,或者流式细胞仪分选出成功感染的细胞。3. TRIzol--氯仿抽提法提取RNA,按照Bio-Rad逆转录试剂盒操作说明书,采用两步法逆转录,实时荧光定量PCR（Realtime qPCR）检测miRNA和基因mRNA水平的表达。Western blot检测基因蛋白水平表达。免疫组化检测组织中基因表达位置和含量。4.CCK-8法绘制细胞生长曲线,平板克隆实验观测细胞增殖情况。5.流式细胞仪细胞周期检测。6.荧光素酶活性检测：萤火虫荧光素酶测定得到的RLU值除以Renilla荧光素酶测定得到的RLU值验证3’-UTR靶向结合位点活性。7.皮下成瘤观测细胞在体内的成瘤和增殖情况,肝包膜异位成瘤观测细胞的成瘤和转移情况。8.统计学分析采用SPSS13.0统计学软件进行数据分析。荧光定量PCR各细胞株2-△△Ct比较采用单因素方差分析（One-way ANONA）;转染后miR-9表达、细胞周期、平板克隆的两组间比较及免疫组化结果采用两独立样本t检验；CCK-8生长曲线及动物实验中皮下成瘤体积比较采用析因设计的方差分析。P<0.05为有统计学差异,数值大小以均值±标准差（X±S）来表示。第二章miR-9靶向调控Hesl在鼻咽癌肿瘤干细胞干性维持中作用及机制1. TRIzol-氯仿抽提法提取RNA,按照Bio-Rad逆转录试剂盒操作说明书,采用两步法逆转录,实时荧光定量PCR（Realtime qPCR）检测miRNA和基因mRNA水平的表达。2. Western blot检测基因蛋白水平表达。3.细胞免疫荧光观察基因在细胞中表达的定位和表达水平。4.免疫组化检测组织中基因表达位置和含量。5.肿瘤球培养、SP细胞比例检测观测肿瘤细胞的“干性”。6.荧光素酶活性检测：萤火虫荧光素酶测定得到的RLU值除以Renilla荧光素酶测定得到的RLU值验证3’-UTR靶向结合位点活性。7.miR-9的治疗作用：皮下成瘤后,miR-9mimics通过转染试剂注射至瘤体内,观察瘤体生长情况。8.统计学分析采用SPSS13.0统计学软件进行数据分析。肿瘤球计数及免疫组化结果采用两独立样本t检验。P<0.05为有统计学差异。第三章c-Myc通过miR-9靶向调控Klf4促进鼻咽癌细胞EMT、侵袭和转移的作用1.质粒扩增提取和纯化、稳定细胞株建立、荧光素酶活性检测、Western blotting、免疫组化、细胞免疫荧光见第一、第二章。2. Transwell迁移试验、Boyden侵袭试验和划痕试验检测细胞的转移能力。3.荧光标记鬼笔环肽染色试验和激光共聚焦观察细胞骨架变化。4.扫描电镜观察细胞形态。5.细胞粘附试验和倒置显微镜观测细胞粘附能力。6.肝包膜下移植后观测移植位点的成瘤情况及远处转移情况。7.统计学分析：采用SPSS13.0统计学软件进行数据分析。荧光定量PCR2-ΔΔCt两两比较比较采用两独立样本t检验；Transwell和Boyden小室结果多组间比较采用单因素方差分析（One-way ANONA）,两组间比较采用两独立样本t检验；细胞粘附能力比较采用析因设计的方差分析。P<0.05为有统计学差异,数值大小以均值±标准差（X±S）来表示。第四章miR-9调控鼻咽癌细胞中免疫和炎症相关基因的表达1.人类基因U133Plus2.0阵列分析基因表达谱。2.RNA提取、逆转、实时荧光定量验证芯片结果。3.统计学分析采用SPSS13.0统计学软件进行数据分析。荧光定量PCR各细胞株2-△△Ct比较采用两独立样本t检验。P<0.05为有统计学差异,数值大小以均值±标准差（X±S）来表示。结果第一章miR-9靶向调控CCNG1抑制鼻咽癌细胞的增殖1.miR-9在鼻咽癌细胞株中的表达我们采用实时荧光定量PCR （Real-time qPCR）检测了miR-9在NP-69、6-10B、5-8F、CNE2、HNE1、HONE1、SUNE1中的表达水平。结果显示：miR-9在鼻咽癌细胞株上表达较NP-69明显下降（F=370.010,P=0.000）。2.构建稳定过表达miR-9细胞株将三质粒包装系统（LV-con或LV-miR-9, PMD2G和PPAX2）共转染至293T细胞,产生两种病毒,即空白对照（LV-con）病毒和过表达miR-9（LV-miR-9）的病毒。转染72h后,收病毒上清感染细胞CNE2、HONE1、SUNE1,48h后观察绿色荧光,嘌呤霉素杀死未感染的细胞后,用qRT-PCR检测miR-9表达情况,结果示各种细胞较对照细胞miR-9表达均显著升高（t=7.355、5.451、5.794,P=0.018、0.032、0.029）。3. qRT-PCR检测鼻咽癌细胞株瞬时转染miRNAs后miR-9的表达采用脂质体法将miR-9mimics和inhibitor及其对照mimics con和inhibitor con转染鼻咽癌细胞株。转染48h后,提取RNA,逆转录后qRT-PCR检测miR-9的表达。结果示,miR-9mimics成功使CNE2、HONE1、SUNE1中miR-9过表达（t=7.619、11.224、5.228, P=0.017、0.000、0.035）而inhibitor降低CNE2和5-8F中miR-9的表达水平（t=10.924、14.95,P=0.000、0.004）。4.miR-9抑制鼻咽癌细胞增殖4.1CCK-8和平板克隆形成实验表明miR-9抑制鼻咽癌细胞增殖我们采用CCK-8法检测细胞生长曲线,发现CNE2细胞过表达miR-9后细胞生长变慢,而抑制miR-9后细胞生长加快（P=0.000）。平板克隆形成实验示,miR-9可显著减少CNE2、HONE1、SUNE1细胞克隆形成（t=7.721、2.744、16.545,P=0.000、0.034、0.000）。以上结果表明,miR-9过表达可抑制鼻咽癌细胞增殖。4.2miR-9影响鼻咽癌细胞的周期分布为研究miR-9抑制增殖的机制,我们进一步检测了miR-9对鼻咽癌细胞周期的影响。结果显示,miR-9过表达后,CNE2和HONE1细胞呈现G0-G1期阻滞,而miR-9抑制后,G0-G1期的细胞较对照组显著减少。5.miR-9过表达抑制CNE2细胞的体内成瘤能力为研究miR-9对鼻咽癌细胞在体内增殖的影响,我们首先将过表达miR-9的CNE2细胞及其对照细胞接种至4只裸鼠背部皮下,细胞数为1×106个,接种后第34天取材。结果对照组均成瘤,而miR-9过表达组均未成瘤。接着,我们将过表达miR-9的CNE2细胞及其对照细胞的皮下接种细胞数提高至1.5x106个,接种后第21天取材。结果示,对照组均成瘤,而miR-9过表达组仅一只成瘤。此外,我们进行了肝包膜下CNE2细胞异位移植成瘤,实验组与对照组移植细胞数均为1.2x106个,接种第23天取材,发现miR-9过表达组在肝包膜下均未成瘤,而对照组全部成瘤。6.miR-9过表达抑制SUNE1细胞裸鼠皮下成瘤能力如上所述,miR-9过表达的CNE2细胞皮下成瘤能力及肝包膜下异位移植能力均显著下降。接下来我们研究了miR-9过表达的SUNE1细胞的皮下成瘤能力。将过表达miR-9的SUNE1细胞及其对照细胞接种至5只裸鼠背部皮下,细胞数为1×106个,接种后第19天取材。结果发现miR-9过表达的SUNE1细胞成瘤后瘤体较对照组显著减小（t=4.279,P=0.003）,且生长速度显著减慢（F=34.112,P=0.000）。此外,miR-9过表达组组织中BrdU+和Ki67+的细胞均较对照组的少。7. CCNGl为miR-9的靶基因我们发现在miR-9过表达的鼻咽癌细胞株中CCNG1表达下降,而抑制miR-9后,CCNG1表达升高,在皮下成瘤的瘤体组织中,miR-9过表达组的组织中CCNG1表达下降。生物信息学软件预测发现,miR-9序列5’端2-9位核苷酸与CCNG1mRNA3’端-UTR完全互补。双荧光素酶报告基因系统的检测结果表明,重组质粒wt3’-UTR和miR-9mimics共转染可显著降低荧光素酶活性,而将wt3’-UTR和miR-9inhibitor共转染可显著增强荧光素酶活性（t=32.542、6.238,P=0.000、0.003）。突变质粒mut3’-UTR和miR-9mimics或miR-9inhibitor共转染均未明显改变荧光素酶活性（t=0.455、0.241,P=0.673、0.821）。以上数据表明,CCNG1为miR-9的靶基因。8.miR-9靶向抑制CCNG1表达而使细胞增殖降低为探讨CCNG1是否介导了miR-9抑制鼻咽癌细胞增殖的功能,我们首先把HONE1细胞株中CCNG1沉默,以明确CCNG1表达下调是否亦抑制了细胞增殖;Western blot显示,miR-9mimics和siCCNG1均能下调CCNGl表达,而siCCNG1和miR-9mimics均能抑制细胞增殖,二者均能使细胞周期G0-G1期细胞增多（F=14.134,P=0.002）,S期减少（F=24.175,P=0.000）。为研究CCNG1是否有拮抗miR-9的周期阻滞作用,我们在过表达miR-9的HONE1细胞的基础上进一步过表达CCNG1（LV-miR-9+pCl-CCNG1）,结果示,CCNG1表达升高,细胞增殖增快,G0-G1期的细胞减少（F=93.810,P=0.000）,S期和(G2-M期细胞显著增多（F=39.597、25.712,P=0.000、0.001）。综上,CCNG1介导了miR-9抑制鼻咽癌细胞增殖的功能。第二章miR-9靶向调控Hesl在鼻咽癌肿瘤干细胞干性维持中作用及机制1.miR-9过表达显著抑制NPC CSCs的自我更新miR-9过表达可显著下调CNE2和SUNE1细胞中干性相关基因（如Nanog、Oct4和ABCG2）表达,而下调NPC细胞中miR-9表达则导致干性基因表达上调；miR-9过表达可显著降低NPC细胞中SP细胞含量,并极显著抑制NPC细胞形成肿瘤球,且过表达组所形成肿瘤球的直径小于对照细胞的（t-39.436、31.577,P=0.000、0.000）。以上结果表明,miR-9至少抑制了NPC CSCs的自我更新。2.肿瘤细胞和肿瘤细胞所培养出的肿瘤球间miR-9和Hesl表达差异为了明确Hesl与鼻咽癌肿瘤干细胞的潜在关系,我们对miR-9和Hesl在鼻咽癌细胞（如CNE2和SUNE1细胞）和由其培养出的肿瘤球间基因表达差异进行了分析。结果显示,miR-9的表达在肿瘤球和肿瘤细胞间无明显差异（t-0.765、0.653,P=0.487、0.533）,而肿瘤球中Hes1、Sox2、Oct4、Nanog和ABCG2的表达均较肿瘤细胞的高。这预示Hesl在鼻咽癌肿瘤干细胞上发挥一定的功能。3.Hesl过表达显著促进了NPC CSCs的自我更新Hesl过表达可显著上调CNE2和SUNE1细胞中干性相关基因（即Nanog和ABCG2）表达,而下调NPC细胞中Hesl表达则导致干性基因表达下调；流式细胞仪分析表明,Hesl过表达可显著增加NPC细胞（CNE2和SUNE1细胞）中SP细胞含量,而下调Hes1则显著下调NPC细胞（CNE2和SUNE1细胞）中SP细胞含量；肿瘤球形成实验表明,过表达Hes1的CNE2和SUNE1细胞较对照组的成球能力显著增强（t=14.902、20.412,P=0.003、0.000）,而抑制Hesl组的成球能力显著下降（t=14.248、18.251,P=0.000、0.000）。以上数据提示,Hes1参与了NPC CSCs自我更新的调控。4.Hesl是miR-9的靶基因既然miR-9参与了NPC CSCs自我更新的调控,其通过下游哪一个靶基因实现?我们的研究发现,miR-9过表达降低NPC细胞（即CNE2和SUNE1细胞）中Hesl表达,而下调miR-9表达则导致Hesl表达上调；CNE2和SUNE1细胞的皮下成瘤组织中,miR-9过表达组的Hes1表达下调。生物信息学软件预测发现,miR-9序列5’端2-7位核苷酸与Hes1mRNA3’端-UTR完全互补。双荧光素酶报告基因系统的检测结果表明,重组质粒wt3’-UTR和miR-9mimics共转染可显著降低荧光素酶活性,而将wt3’-UTR和miR-9inhibitor共转染可显著增强荧光素酶活性（t-=23.089、11.573,P=0.000、0.000）；突变质粒mut3’-UTR和miR-9mimics或miR-9inhibitor共转染均未明显改变荧光素酶活性（t==1.075、0.218,P=0.343、0.838）。以上数据表明,Hes1为miR-9的靶基因。5.Hesl介导了miR-9抑制鼻咽癌肿瘤干细胞自我更新的功能在明确Hesl是miR-9的靶基因基础上,拟进一步阐明miR-9是否通过下调Hesl表达抑制了鼻咽癌肿瘤干细胞的自我更新。Western blot结果显示,miR-9和Hesl共同过表达组中Nanog和ABCG2的表达水平较miR-9组的高,SP细胞含量较miR.-9组的高,且Hesl过表达后有效阻断了miR-9抑制肿瘤球形成的能力（F=162.277,P=0.000）。综上,miR-9抑制鼻咽癌肿瘤干细胞自我更新的功能可由其靶基因Hes1介导。6.探讨miR-9潜在的治疗价值为研究miR-9对鼻咽癌的治疗作用,我们进行了miR-9的裸鼠皮下瘤内注射实验,首次预实验中,注射miR-9Agomir组的三只裸鼠均发现沿进针方向形成空洞且迁延不愈,而对照组未见空洞（数据未展示）。第二次重复实验中使用miR-9mimics与MaxSuppressorTM In vivo RNA-LANCErⅡ共注射,结果发现miR-9可抑制瘤体生长,且部分瘤体中心出现空洞。两组分别提取组织RNA和蛋白进行分析,结果显示,实验组中miR-9成功过表达；且其靶基因CCNG1和Hes1均被抑制。此部分结果正在进行第三次实验验证。第三章c-Myc通过miR-9靶向调控Klf4促进鼻咽癌细胞EMT、侵袭和转移的作用1. c-Myc在鼻咽癌细胞中上调miR-9表达c-Myc过表达上调鼻咽癌细胞株中miR-9表达,而下调c-Myc则抑制miR-9表达。可见,c-Myc正向调控了鼻咽癌细胞中miR-9的表达。2. c-Myc靶向上调miR-9表达引起EMT相关基因表达变化为明确c-Myc对EMT相关基因是否有影响,我们首先行qRT-PCR检测EMT相关基因表达情况,结果显示,c-Myc过表达显著下调HONE1细胞中E-cadherin表达（t=-44.880,P=0.000）,而通过RNAi抑制c-Myc表达则升高E-cadherin表达（t=3.149,P=0.035）;此外,c-Myc无论过表达还是通过RNAi抑制c-Myc表达均在mRNA水平对N-cadherin和Vimentin表达无显著影响（P>0.05）。Western blot检测结果表明,c-Myc过表达显著下调HONE1细胞中E-cadherin表达,上调Vimentin表达,而抑制c-Myc表达则上调E-cadherin表达,并下降Vimentin表达。此外,miR-9inhibitor能阻断c-Myc对E-cadherin和Vimentin的影响,并在CNE2细胞中得到进一步验证。3. c-Myc通过上调miR-9表达促进了NPC细胞迁移和侵袭如上所述,鼻咽癌细胞中miR-9介导了c-Myc对E-cadherin和Vimentin表达的影响,接下来我们研究了c-Myc对鼻咽癌细胞迁移和侵袭的影响及miR-9在其中所起的作用。Transwell结果显示,c-Myc促进细胞迁移（t=12.231,P=0.000）, Boyden小室结果显示c-Myc可促进细胞侵袭（t=10.000,P=0.000）,而miR-9inhibitor可阻断c-Myc促迁移和侵袭的作用（F=147.016、59.900,P=0.001、0.001）。4.miR-9对鼻咽癌细胞中EMT相关基因表达的调控作用在培养稳定过表达miR-9的细胞株过程中,发现CNE2和HONE1细胞形态发生了变化,部分细胞变得细长且触角增多,细胞失去原有的铺路石样或规整的形态,这提示miR-9过表达可能诱导了EMT。随后,在基因层面进一步明确miR-9是否确实诱发了EMT。结果表明,miR-9过表达导致SUNE1、CNE2、 HONE1、HNE1细胞中上皮相关基因E-cadherin和a-catenin表达下降,而间充质相关基因N-cadherin和Vimentin表达升高；下调miR-9表达则引起CNE2和HONE1细胞中E-cadherin表达升高,而N-cadherin和Vimentin表达下降。以上数据提示,miR-9过表达在鼻咽癌细胞上诱导了EMT。5.miR-9促进鼻咽癌细胞的迁移和侵袭我们进一步采用Transwell和Boyden小室明确miR-9过表达对CNE2和HONE1细胞的迁移和侵袭能力的影响。结果显示,过表达miR-9的CNE2和HONE1细胞的迁移和侵袭能力较对照组的显著增强（t=6.533、10.983：13.131、12.351,P=0.000、0.000；0.000、0.000）。划痕实验亦显示,miR-9过表达可促进CNE2和HONE1细胞迁移。综上,miR-9可促进鼻咽癌细胞迁移和侵袭。6.miR-9过表达降低鼻咽癌细胞的粘附能力为研究miR-9对肿瘤细胞的粘附能力的影响,我们用0.05nM EDTA处理过表达miR-9的CNE2和HONE1细胞及其对照细胞后,miR-9过表达的鼻咽癌细胞较对照细胞粘附能力下降（F=846.528、751.057,P=0.000、0.000）,提示过表达miR-9的鼻咽癌细胞粘附降低。7.miR-9对细胞骨架的影响我们用鬼笔环肽标记法检测了miR-9过表达对CNE2和HONE1细胞细胞骨架的影响,并用扫描电镜观测了细胞的微观形态。结果显示,miR-9过表达的细胞表面纤维增粗增多,伪足增多。Western blot检测Rho酶和Rac的表达较对照的增多。提示miR-9与细胞形态改变和伪足形成等转移起始步骤相关。8.Klf4为miR-9的靶基因miR-9过表达降低NPC细胞（即CNE2和SUNE1细胞）中Klf4表达,而下调miR-9表达则导致K1f4表达上调；CNE2和SUNE1细胞的皮下成瘤组织中,miR-9过表达组的Klf4表达下调。生物信息学软件预测发现,miR-9序列5’端2-7位核苷酸与Klf4mRNA3’端-UTR完全互补。双荧光素酶报告基因系统的检测结果表明,重组质粒wt3’-UTR和miR-9mimics共转染可显著降低荧光素酶活性,而将wt3’-UTR和miR-9inhibitor共转染可显著增强荧光素酶活性（t=4.380、13.777,P=0.012、0.000）；突变质粒mut3’-UTR和miR-9mimics或miR-9inhibitor共转染均未明显改变荧光素酶活性（t=0.921、0.302,P=0.409、0.777）。以上数据表明,Klf4为miR-9的靶基因。9.miR-9靶向抑制Klf4进而增强鼻咽癌细胞的迁移和侵袭能力我们的前期研究表明,Klf4在鼻咽癌细胞中具有抑制EMT、迁移和侵袭的功能,为研究Klf4是否具有拮抗miR-9促进迁移和侵袭的作用,我们在HoNE1细胞过表达miR-9的基础上过表达Klf4,结果过表达Klf4组（LV-miR-9+LV-Klf4）的Klf4表达升高,同时迁移和侵袭能力较LV-miR-9组显著增强（F=137.015、12.438,P=0.000、0.001）。以上提示,Klf4拮抗了miR-9促迁移和侵袭的作用,即miR-9通过靶向抑制Klf4促进细胞迁移和侵袭。10. c-Myc过表达促进鼻咽癌细胞体内转移为明确c-Myc对鼻咽癌细胞体内转移的影响,我们把过表达c-Myc的CNE2细胞及对照细胞（细胞剂量：1×106/只）分别接种于裸鼠肝包膜下,每组7只裸鼠,结果示接种处全部成瘤,而发生淋巴结转移的过表达组有5/7只,而对照2/7只。第四章miR-9调控鼻咽癌细胞中免疫和炎症相关基因的表达1.miR-9引起鼻咽癌细胞中干扰素相关基因表达变化微阵列分析显示,miR-9过表达的CNE2细胞中多种IFN调节的基因如IFI44L、PSMB8、IRF5、PSMB10、IFI27、PSB9HUMAN、IFIT2、TRAIL、IFIT1、 PSB8HUMAN、IRF1和B2M的表达都发生了变化。qRT-PCR进一步验证了基因芯片结果（P<0.05或0.01）。为了进一步证实miR-9对IFN调节因子的作用,我们又进行了miR-9mimics和inhibitor的瞬时转染,qRT-PCR发现miR-9mimics组中IFI44L、PSMB8、IRF5、 PSMB10、IFI27、IFIT2、TRAIL、IFIT1、IRF1、B2M和GBP1表达升高,而ISG20和AIM2表达下降。相反,miR-9inhibitor组中IFI44L、PSMB8、IRF5、PSMB10、 IFI27、IFIT2、TRAIL、IFIT1、IRF1、B2M和GBP1表达下降,而ISG20和AIM2表达则升高（P<0.05或0.01）。2.miR-9引起鼻咽癌细胞中主要组织相容性复合体Ⅰ型（MHCI）分子表达变化微阵列分析显示,miR-9过表达的CNE2细胞中多种主要组织相容性复合体Ⅰ型（MHC Ⅰ）分子如HLA-B、HLA-H、HLA-C、HLA-F、Q8WW48HUMAN、 NP001004349.1、Q6ZUW0HUMAN、O19682HUMAN和TAP1表达升高。qRT-PCR证实了HLA-B、HLA-F和TAP1表达的变化（P<0.05或0.01）。为了进一步证实miR-9对MHC1分子的作用,我们又进行了miR-9mimics和inhibitor的瞬时转染,qRT-PCR发现HLA-B、HLA-F和TAP1在miR-9mimics组中的表达升高,而在miR-9inhibitor组中表达下降（P<0.05或0.01）。3.miR-9引起鼻咽癌细胞白介素（IL）相关基因表达变化微阵列分析显示,miR-9过表达的CNE2细胞中多种IL相关基因IL20RB、 GALT、IL7、IL1B、IL11、IL1F8、IL1A、IL6和IL7R等表达发生变化,结果用qRT-PCR证实（P<0.05或0.01）。为了进一步证实miR-9对IL相关基因的作用,我们又进行了miR-9mimics和inhibitor的瞬时转染,qRT-PCR发现miR-9mimics可以显著上调一些IL相关基因（IL20RB、GALT、IL7）（P<0.05或0.01）,而下调另一些IL相关基因（IL1B、 IL11、IL1F8、IL1A、IL6、IL7R）（P<0.05或0.01）。相反,miR-9inhibitor下调IL20RB、GALT和IL7（P<0.05或0.01）,而上调IL1B、IL11、IL1F8、ILIA、IL6和IL7R（P<0.05或0.01）。结论1.miR-9靶向下调CCNG1抑制NPC细胞增殖；2.miR-9的靶基因Hesl介导了miR-9抑制鼻咽癌肿瘤干细胞自我更新的功能；3. c-Myc通过miR-9靶向调控Klf4促进了鼻咽癌细胞EMT、侵袭和转移；4.miR-9可调节NPC细胞中免疫和炎症相关基因表达,提示miR-9可能在炎症和肿瘤间起桥梁作用；5.miR-9靶向调控CCNG1表达抑制了NPC细胞增殖,而miR-9靶向下调Klf4表达则促进了鼻咽癌细胞EMT、侵袭和迁移,这些预示miR-9可通过靶向调控不同的靶基因在肿瘤细胞的增殖以及EMT、侵袭和转移中发挥不同的功能,这也预示了miRNAs功能的多样性和复杂性。

    Background and objectiveNasopharyngeal carcinoma （NPC） is the most common cancer arising from the mucosal epithelium of the nasopharynx, and is extremely common in southern regions of China, particularly in Guangdong, Guangxi, Hunan, Fujian. NPC frequently metastasizes to regional lymph nodes. The etiology of NPC seems to follow a multi-step process, in which Epstein-Barr virus infection, dietary, smoking, genetic factors and environmental carcinogens seem to play important roles. Typical conventional treatments of NPC include radiotherapy, surgery and chemotherapy. Recently, molecular targeted therapy has become the hotspot and focus of comprehensive treatment of NPC.As was shown by Weinberg in2011, The hallmarks of cancer comprise ten biological capabilities acquired during the multistep development of human tumors including sustaining proliferative signaling, evading growth suppressors, resisting cell death, enabling replicative immortality, inducing angiogenesis, activating invasion and metastasis, genome instability, inflammation, reprogramming of energy metabolism and evading immune destruction. There are many kinds of cells in "tumor microenvironment", such as tumor cells, inflammatory cells, cancer stem cells （CSCs）. CSCs may generate tumors through the stem cell processes of self-renewal and differentiation into multiple cell types. Such cells are proposed to persist in tumors as a distinct population and cause relapse and metastasis, moreover, they can resistant to chemotherapy and radiotherapy. Therefore, development of specific therapies targeted at CSCs holds hope for improvement of survival cancer patients.Metastasis is a basic biological characteristics of malignant tumor, and the primary factors affecting patient survival, which has become a hot and difficult field of cancer research. The most prominent feature of tumor metastasis is the tissue specificity, which is different tumor can form metastases in the same or different parts of the human body. Epithelial-mesenchymal transition or transformation （EMT） is a process characterized by loss of cell adhesion, repression of E-cadherin expression, and increased cell motility. EMT is an important contributor to the invasion and metastasis of epithelial-derived cancers.miRNAs are small non-coding RNA molecules, and important regulatory molecules in animals and plants. Each miRNA may have a plurality of target genes, while several miRNAs can also target the same gene. Recent studies show that some miRNAs regulate cell proliferation and apoptosis processes in cancer. In addition, some miRNAs may function as oncogenes or tumor suppressors. miRNA expression profiles may become useful biomarkers for cancer diagnostics. Moreover, miRNA therapy could be a powerful tool for cancer prevention and therapeutics. Therefore, aberrant expression of miRNAs has been implicated in numerous disease states, and miRNA-based therapies are under investigation.miR-9, initially demonstrated to function in neurogenesis, has been confirmed to be implicated in cancer. miR-9is under-expressed in many types of cancers, including breast cancer, hepatocellular carcinomas, gastric cancer, ovarian cancer, colon cancer, melanoma, clear cell renal cell carcinoma and medulloblastoma, all of which is indicative of a tumor suppressor potential, whereas miR-9is over-expressed in brain cancer and Hodgkin’s lymphoma, suggesting oncomir activity for miR-9in these two cancers. Some evidence indicates that the functions of miR-9in cancer cells are recently implicated in regulating proliferation, EMT, metastasis, apoptosis and tumor angiogenesis, etc. However, the contribution of miR-9which is commonly down-regulated in NPC to the pathogenesis of NPC remains to be characterized.MethodsPart I miR-9inhibits cell growth and tumorigenesis through repressing CCNG1in NPC1Establishment of miR-9stably over-expressed NPC cells:Particles expressing vector along with packaging plasmids psPAX2and pMD2.G were transfected into HEK293T cells （maintained in10%FBS） using Lipofectamine2000reagent （Invitrogen） according to the manufacturers’instruction.48h after transfection, virus supernatant was harvested from these cells, and then used to infect NPC cells. To attain100%percentage of infected NPC cells based on EGFP assay and FACS analysis, lentiviral infected cells were selected by2μg/ml puromycin （GIBCO） for2weeks to kill non-infected cells.2RNA isolation, reverse transcription and qRT-PCR:For miRNA and mRNA analyses, total RNA from NPC cells was extracted with Trizol Reagent （TaKaRa） according to the protocol provided by the manufacturer. Total RNA was reversely transcribed with the PrimeScript RT reagent Kit （TaKaRa）. The expression levels of mature miRNA were determined by SYBR Green quantitative PCR amplifications performed on the Stratagene Mx3000P Real-Time PCR system （Agilent Technologies, Inc., USA）. U6was used for normalization. Expression of mRNA analysis was performed using SYBR Green Master Mix （TaKaRa） as described using GAPDH for normalization on a Stratagene Mx3000P qRT-PCR System. The primers used for the amplification of the indicated genes were listed in Table S1-4. All samples were normalized to internal controls and fold changes were calculated through relative quantification （2-ΔΔCt）.3CCK-8assay:A total of800-1000cells were seeded in a96-well plate and then allowed to grow in normal medium for96hours. Cells were incubated in100ul normal+10ul CCK-8. The absorbance in each well was measured at450nm by a microplate reader. 4Colony formation assay:200cells were seeded in6-well plates and grown for2weeks. Then washed with PBS, fixed with methanol for10min, and stained with hematoxylin for10min.5Cell-cycle analysis:Pipet cell suspension （EtOH70%） over night at-20℃. Treated with RNase A to remove RNAs from the cells, cell-cycle distribution was analyzed by propidium iodide staining and flow cytometry.6Luciferase activities detection:Insert the3’-UTR into3’downstream to the luciferase reporter in the pGL3vector which carries activated promoter sequences. The association of miRNA with3’-UTR will cause the attenuation of translation of luciferase mRNA, which shows the decline of luciferase activities.7Western blot and high sensitivity chemiluminescence imaging system were used to survey the proteic level of genes.8Animal experiment:Cells were injected subcutaneously or into hepatic subcapsular of nude mice, tumors in subcutaneous were measured every2-3days, tumor volunes were calculated as follows:D×d×d/2（D meant the longest diameter and d meant the shortest diameter）. Tumors were used to extract RNA and proptein, and fixed with4%paraformaldehyde for HE staining and Immunohistochemistry.9Statistical analysis:Data were presented as mean±SEM unless otherwise indicated of at least3independent experiments. Statistical analysis was performed using a SPSS13.0software package. Two-tailed Student’s t test was used for comparisons of2independent groups. Statistical significance was assessed by the Student’s t-test （*p<0.5;#p<0.01）.Part II miR-9impairs cancer stem cells through negative regulation of Hesl in NPC1Establishment of Hes1and shHes1stably over-expressed NPC cells, Western blot and Cell Immunofluorescence were used to analyze protein expression of "stemness "genes.2Luciferase activities were detected to verify Hesl was a direct target of miR-9in NPC cells after Hes1wt3’UTR （mut3’UTR） and miRNAs cotransfection.3Tumor sphere formation and ratio of stem cell populations （SP） detected by flow cytometry and hoechst33342were performed to evaluate the "sternness" of NPC cells.4Intratumoral delivery of miR-9in a nasopharyngeal cancer xenograft model: Nude mice were subcutaneously inoculated with CNE2cells and maintained until the tumor cells had formed solid, palpable tumors with an average volume of80to100mm3,7days following inoculation, miR-9or a negative control miRNA was repeatedly administered by intratumoral injections every3days. All mice were killed on day16.5Statistical analysis:Data were presented as mean±SEM unless otherwise indicated of at least3independent experiments. Statistical analysis was performed using a SPSS13.0software package, Two-tailed Student’s t test was used for comparisons of2independent groups.Part Ⅲ miR-9triggers EMT and migration by repressing Klf4in NPC1Establishment of c-Myc and miR-9stably over-expressed NPC cells, Western blot and Cell Immunofluorescence were used to analyze protein expression of genes, Immunohistochemistry was used to investigate the protein expression and the definite localization of proteins.2Transwell, Scratching test and Boyden were used for migration and invasion of cells.3Fluorescently labeled phalloidin staining test and Scanning electron microscopy were used to survey the Skeleton and morphological variation of NPC cells.4PBS with0.05nM EDTA and inverted fluorescence microscope were used for adhesion ability of cells.5Statistical analysis:Data were presented as mean±SEM unless otherwise indicated of at least3independent experiments. Statistical analysis was performed using a SPSS13.0software package.Two-tailed Student’s t test was used for comparisons of2independent groups; One-way ANONA was used for comparisons of several groups; Factorial ANOVA performed an analysis of variance for multiple classification variables. Part IV miR-9modulates the expression of interferon-regulated genes and MHC class I molecules in NPC1RNA extract and qRT-PCR were carried out to detect the mRNA expression of genes。2mRNA microarray analysis:Expression microarray analysis was carried out with commercially available Affymetrix Human Gene U133Plus2.0array according to the Affymetrix standard protocol, which carried47,000transcripts representing38,500well-characterized human genes. All the hybridization procedures and data analysis were performed by Capital Bio Corp.（Bejing, China）. Total RNA samples were isolated from NPC cells （CNE2cells） using Trizol reagent （Invitrogen）. Briefly, total RNA was used to synthesize cDNA in an in vitro transcription reaction, and then cDNA was fluorescently labeled by Cy5or Cy3-CPT with Klenow enzyme. Labeled cDNA was then hybridized to Affymetrix Human Gene U133Plus2.0arrays. Hybridization was processed at45℃, with rotation for16h （Affymetrix GeneChip Hybridization Oven640）. Chips were then washed and stained in the Affymetrix Fluidics Station450. Hybridization signals were scanned with a Lux-Scan3.0scanner （Capital Bio. Corporation, Beijing, China）. The resultant images were digitized with Genepix Pro6.0software （Axon Instruments, Foster City, CA, USA）.3Statistical analysis:Data were presented as mean±SEM unless otherwise indicated of at least3independent experiments. Statistical analysis was performed using a SPSS13.0software package. Two-tailed Student’s t test was used for comparisons of2independent groups.ResultsPart I miR-9inhibits cell growth and tumorigenesis through repressing CCNG1in NPC1. qRT-PCR data demonstrated the miR-9expression was significantly lower in NPC cells （i.e.,6-10B,5-8F, CNE2, HNE1, HONE1, SUNE1cells） than that in NP-69cells （F=370.010, P=0.000）.2. We established miR-9-expressing NPC cell lines （i.e., CNE2, HONE1and SUNE1cells）.3. The levels of miR-9in CNE2, HONE1and SUNE1cells transfected with miR-9mimics was much higher than that in CNE2, HONE1and SUNE1cells transfected with control RNA （t=7.619,11.224,5.228, P=0.017,0.000,0.035） Moreover, the levels of miR-9in CNE2and5-8F cells transfected with miR-9inhibitor was much lower than that in CNE2and5-8F cells transfected with inhibitor control （t=10.924,14.95, P=0.000,0.004）, indicating that miR-9inhibitor efficiently down-regulated the endogenous miR-9expression in CNE2and5-8F cells.4. miR-9inhibits cell growth and affect cell-cycle distribuionCCK-8assay showed that miR-9-expressing NPC cells inhibited cell growth, whereas the proliferation of NPC cells transfected with miR-9inhibitor was enhanced （P=0.000）. Flat cloning formation experiments also indicated that the ability of proliferation was significantly inhibited in miR-9-expressing groups （t=7.721,2.744,16.545, P=0.000,0.034,0.000）. In sum, the above two experiments illustrates that miR-9overexpression suppresses the proliferation of NPC cells in vitro.We studied the effects of miR-9on cell cycle using FACS. The results showed that compare to the control group, the ratio of G0-G1phase in miR-9-expressing cells were significantly increased （P<0.05）, while the ratio of G0-G1phase in NPC cells transfected with miR-9inhibitor were significantly decreased （P<0.05）, indicating that miR-9could cease cell cycle at phase G0/G1.5. The suppressive effects of miR-9overexpression on tumorigenicity of NPC cellsTo explore the effects of miR-9on NPC tumorigenicity, miR-9-expressing CNE2cells （1×106） we firstly subcutaneously injected into the dorsal flank of nude mice. At34days after implantation, mice injected with vector-expressing CNE2cells carried large tumor burdens, while mice injected with miR-9-expressing CNE2cells did not developed tumor. Then, we increased the number of injected cells from1×106to1.5×106. At21days after implantation, all the mice injected with vector-expressing CNE2cells developed tumors, but only one of four mice injected with miR-9-expressing CNE2cells developed tumor which is much smaller than the controls. Additionally, miR-9-expressing CNE2cells （1.2×106） were transplanted subcapsularly into the livers of nude mice. At23days after implantation, all the mice transplanted with vector-expressing CNE2cells formed tumors in livers, while none of those mice injected with miR-9-expressing CNE2cells developed tumor.Nextly, miR-9-expressing SUNE1cells （1×106） cells were subcutaneously transplanted into the dorsal flank nude mice. The tumor became palpable between7and9days after inoculation, and all the mice injected with vector-expressing cells and4/5mice injected with miR-9-expressing CNE2cells formed tumors at the end of the experiment （t=4.279, P=0.003）. As early as9days postimplantation, the growth of transplanted tumors between2groups became statistically significant （P=0.021）. At19days after implantation, those mice injected with vector-expressing cells carried larger burdens. We also showed that both the staining intensity and the number of hyperproliferative BrdU and Ki67tumor cells were significantly decreased compared with control.6. CCNG1is a direct target gene of miR-9To explore the mechanism of growth inhibition induced by miR-9, we want to idenfity the target gene of miR-9. Firstly, we found that miR-9overexpression down-regulated CCNG1expression in NPC cells, while miR-9inhibitor up-regulated CCNG1expression in NPC cells. Secondly, we performed luciferase reporter assay to determine whether miR-9could directly target the3’-UTR of CCNG1in NPC cells. CNE2cells were transfected with wt or mt3’-UTR vector and miR-9mimics. The results showed a significant decrease of luciferase activity when compared with miR control （t=32.542, P=0.000）. The activity of mt3’-UTR vector was unaffected by a simultaneous transfection with miR-9mimics or miR-9inhibitor （t=0.455,0.241, P=0.673,0.821）. Moreover, cotransfection with miR-9inhibitor and wt3’-UTR vector in CNE2cells led to an increase of luciferase activity （t=6.238, P=0.003） Taken together, all these results strongly suggest that CCNG1is a direct target gene of miR-9in NPC cells. 7. miR-9affects cell proliferation and cell cycle via CCNG1in NPC cells.To elucidate whether the growth-suppressive effect of miR-9could be mediated by CCNG1in NPC cells, we firstly silenced CCNG1to investigate whether the reduced expression of CCNG1could mimic the suppressive effect of miR-9. HONE1cells were transfected with si-CCNG1, and then we examined cell proliferation rate and cell cycle distribution （F=14.134, P=0.002）. CCNG1knockdown led to significant cell growth inhibition and cell-cycle arrest, similar to those induced by miR-9（P<0.01）. Subsequently, we evaluated whether ectopic expression of CCNG1could rescue the suppressive effect of miR-9. HONE1cells overexpressing miR-9were transfected with pC3-CCNG1. We observed that the ectopic expression of CCNG1in miR-9-expressing NPC cells significantly rescued miR-9-induced cell growth inhibition and cell-cycle arrest （F=93.810, P=0.000）. In summary, the growth-suppressive effect of miR-9could be mediated by CCNG1in NPC cells.Part II miR-9impairs cancer stem cells through negative regulation of Hesl in NPC1. miR-9overexpresssion down-regulates the expression of stemness-related genes in NPC cellsWestern blot showed that Nanog, Oct4and ABCG2was down-expressed in miR-9-expressing CNE2and SUNE1cells, and up-expressed in5-8F cells transfected with miR-9inhibitor. Immunohistochemistry also showed a low-expression of Nanog and Sox2in subcutaneous xenotransplanted tumors overexpressing miR-9. We found that the ability of tumor spheres of miR-9-expressing cells reduced （t=39.436,31.577, P=0.000,0.000）, and the ratio of SP cells was decreased in miR-9-expressing cells.2. The expression of stemness-related genes between adherent cells and tumor spheresFor further study, we tested the expression of stemness-related genes between adherent cells and tumor spheres （the second generation）. Western blot data resulte indicated that the expression of Hesl, Sox2, Oct4, Nanog and ABCG2was higher in tumor spheres than in adherent cells, while miR-9expression （detected by qRT-PCR） showed no disparity between adherent cells and tumor spheres （t=0.765,0.653, P=0.487,0.533）3. Effects of Hesl on "sternness" of NPC cellsWe found a high expression of Nanog and ABCG2in Hesl-expressing （LV-Hes1） NPC cells, and low expression in Hesl-silenced （shHesl） cells. The SP cells in SUNE1overexpressing Hes1could up to8.6%compared to the control5.3%, while1.0%in Hes1-silenced cells; the SP cells in CNE2overexpressing Hes1could be up to5.3%compared to the control2.0%, while0.4%in Hesl-silenced CNE2cells. Moreover, the ability of tumor sphere formation was enhanced in Hes1-expressing CNE2and SUNE1cells （t=14.902,20.412, P=0.003,0.000）4. Hes1is a direct target of miR-9in NPC cellsAccording to the targeted gene prediction performed by TargetScan and studies reported by Siok-Lay Tan, to explore the mechanism of "sternness" inhibition induced by miR-9, we investigated whether miR-9could regulate Hesl expression in NPC cells, and we found that miR-9overexpression down-regulated Hesl expression in NPC cells, while miR-9inhibitor up-regulated Hesl expression in NPC cells.We performed luciferase reporter assay to determine whether miR-9could directly target the3’-UTR of Hesl in NPC cells. CNE2cells were then transfected with wt or mt3’-UTR vector and miR-9mimics. The results showed a significant decrease of luciferase activity when compared with miR control （t=23.089, P=0.000）. The activity of mt3’-UTR vector was unaffected by a simultaneous transfection with miR-9mimics or miR-9inhibitor （t=1.075,0.218, P=0.343,0.838）. Moreover, cotransfection with miR-9inhibitor and wt3’-UTR vector in CNE2cells led to an increase of luciferase activity （t=11.573, P=0.000）. Taken together, these results strongly suggest that Hesl is a direct target gene of miR-9in NPC cells.5. Hesl rescues the suppressive effect of miR-9on "sternness"Subsequently, we evaluated whether over-expression of Hes1could rescue the suppressive effect of miR-9. miR-9-expressing CNE2cells were infected with LV-Hes1. We showed that ectopic expression of Hesl significantly increased expression of Hesl and other genes related to "sternness", and rescued miR-9-induced reduction of SP cells and tumor spheres （F=162.277, P=0.000）6. Intraturnoral delivery of miR-9leads to regression of tumors in a NPC xenograft modelAs the in vitro and in vivo data showed an antitumorigenic role for miR-9in NPC, we examined the therapeutic potential of synthetic miR-9Agomir or mimics in vivo. In the first pre-experiment, we found that the tumor injected with miR-9Agomir formed cavities along the direction of the needles after injection with a delayed healing, but the controls healed rapidly （data not shown）. In the repeated test, we injected miR-9mimics into two tumors with MaxSuppressorTM In vivo RNA-LANCErⅡ, and found that intratumoral delivery of synthetic miR-9induced a specific inhibitory response and it formed a hollow in the center of one tumor which was interesting. Now, we are carrying out a third time repeat.Part III miR-9triggers EMT and migration by repressing Klf4in NPC1. c-Myc activates miR-9expressionWe established c-Myc-expressing NPC cells, levels of c-Myc and miR-9expression were evaluated by qRT-PCR. The result showed that c-Myc and miR-9levels in HONE1and SUNE1cells harboring c-Myc transgene was significantly higher than that in cells harboring vector control （t=5.199,4.769, P=0.035,0.041） and （t=2.536,4.690, P=0.039,0.005）. Moreover, the levels of c-Myc and miR-9in HONE1and SUNE1cells transfected with sic-Myc was much lower than that transfected with inhibitor control （t=11.545,11.744, P=0.000,0.000） and （t=6.585,10.473, P=0.003,0.000）. In summary, c-Myc efficiently up-regulated the endogenous miR-9expression in NPC cells.2. miR-9-mediated c-Myc induced EMTTo study whether c-Myc can affect EMT process in NPC cells, we detected EMT-related genes using qRT-PCR. The level of E-cadherin seemed to be variable and became higher when c-Myc was highly expressed（t=-44.880, P=0.000）, whereas there was no such relationship with N-cadherin and Vimentin （P>0.05）. Interestingly, we found expreesion of vimentin at protein level was higher when c-Myc was highly expressed, suggesting that c-Myc may have some correlation with E-cadherin and vimentin.3. miR-9-mediated c-Myc enhanced motilityAs shown above, miR-9-mediated c-Myc induced E-cadherin and vimentin alteration. Transwell assay and Boyden chamber assay displayed the enhanced ability of migration and invasion in c-Myc-expressing cells （t=12.231,10.000; P=0.000,0.000）, while miR-9inhibitor could rescue c-Myc-induced enhancement of motility （F=147.016,59.900, P=0.001,0.001）4. miR-9induces EMT-like changes in NPC cells.During culture of miR-9-expressing cells, we were surprised to find that the cells gradually lose the original cobblestone-like or neat form and appeard morphological diversification, some cells became elongated and the antennas of cells were increased. Previous studies suggested that miR-9was associated with EMT in breast cancer, further study on the role of miR-9in EMT was carried out. Western blot showed that E-cadherin and a-catenin which represent epithelial factors were down-regulated in NPC cells overexpressing miR-9, however, up-regulated by miR-9inhibitor. While the expression of N-cadherin and vimentin were up-regulated in miR-9-expressing cells, but down-regulated by miR-9inhibitor, indicating that the ectopic expression of miR-9lead to an EMT-like conversion.5. miR-9promotes cell migration and invasion in NPC.Transwell assay and Boyden chamber assay demonstrated that miR-9overexpression increased the migration and invasion of CNE2and HONE1cells （t=6.533,10.983,13.131,12.351,P=0.000,0.000,0.000,0.000）. Scratch-Migration assay also showed the increased migration of cells overexpressing miR-9as compared with controls.6. miR-9reduces the ability of cell adhesionCell adhesion was performed and analysed by counting the adherent cells, experiments showed that miR-9-expressing cells had lower rates of adhesion than the control cells （F=846.528,751.057, P=0.000,0.000） 7. Effect of miR-9on cytoskeletonWe investigated whether or not miR-9induces changes in cytoskeleton by detect ing F-actin at the light and electron microscopic levels through the use of the actin-binding protein phalloidin, and we also observed the microscopic morphology of the cells using scanning electron microscopy. We found an increased expression of F-actin in miR-9overexpressed cells, with increased pseudopodia.As is known, Rac and Rho regulate the formation of distinct actin filament-based structures, and Cdc42and Rac are also required for the assembly of adhesion sites to the extracellular matrix. We detected an increased expression of Rho and Rac in miR-9-expressing cells. All above results suggested that miR-9was correlated with changes in shape of cells and pseudopodia formation, which is the initial step of tumor metastasis.8. Klf4is a direct target gene of miR-9in NPC cellsWe found that Klf4promoted mesenchymal-epithelial transition （MET） and inhibited migration and metastasis of NPC cells in other studies of our group. According to the targeted gene prediction performed by microrna.org, we investigated whether miR-9could regulate Klf4expression in NPC cells. Western blot suggested a decreased expression of Klf4in miR-9-expressing NPC cells and subcutaneous xenotransplanted tumors, whereas the expression of Klf4was up-regulated by miR-9inhibitor.We performed luciferase reporter assay to determine whether miR-9could directly target the3’-UTR of Klf4in NPC cells. The target sequence of Klf43’-UTR （wt3’-UTR） or the mutant sequence （mt3’-UTR） was cloned into a luciferase reporter vector. CNE2cells were then transfected with wt or mt3’-UTR vector and miR-9mimics. The results showed a significant decrease of luciferase activity when compared with miR control （t=4.380, P=0.012）. The activity of mt3’-UTR vector was unaffected by a simultaneous transfection with miR-9mimics or miR-9inhibitor （t=0.921,0.302, P=0.409,0.777）. Moreover, cotransfection with miR-9inhibitor and wt3’-UTR vector in CNE2cells led to a increase of luciferase activity （t=13.777, P=0.000）. Taken together, all these results strongly suggest that Klf4 was a direct target gene of miR-9in NPC cells.9. miR-9-induced EMT and enhanced motility are mediated by Klf4.Subsequently, we evaluated whether over-expression of Klf4could rescue the improvement of miR-9in motility. HONE1cells overexpressing miR-9were infected with LV-Klf4. Western blot showed that ectopic expression of Klf4significantly increased expression of Klf4and E-cadherin, but decreased expression of N-cadherin. Furthermore, Transwell assay and Boyden chamber assay showed that Klf4rescued the increased ability of cell migration and invasion induced by miR-9（F=137.015, P=0.000）. In sum, we suggest that miR-9promotes EMT and cell motility by targeting Klf4.10. c-Myc promotes tumor metastasis in vivoTo fully explore the effects of c-Myc on metastasis in vivo, we injected1×106CNE2cells infected with LV-c-Myc or the control plasmid （LV-con） subcapsularly into the livers of nude mice,7mice for each group,23days after implantation, all the mice developed tumors in liver at the end of the experiment. The5per7mice in LV-c-Myc group developed lymph node metastases, while2per7mice in control group developed lymph node metastases, suggesting c-Myc can promote tumor metastasis in vivo.In summary, we have found that miR-9triggers EMT and migration in NPC by repressing Klf4.Part IV miR-9modulates the expression of interferon-regulated genes and MHC class I molecules in NPC1. miR-9altered IFN-regulated gene expression in NPC cellsMicroarray analysis of CNE2cells infected with LV-miR-9showed the induction of many IFN-regulated target genes （e.g., IFI44L, PSMB8, IRF5, PSMB10, IFI27, PSB9HUMAN, IFIT2, TRAIL, IFIT1, PSB8HUMAN, IRF1and B2M）, which was further confirmed by qRT-PCR analyses.To fully explore the effects of miR-9on IFN-regulated target genes, CNE2cells were transiently transfected with miR-9mimics or anti-miR-9, respectively. The results of qRT-PCR demonstrated that miR-9mimic upregulated the expression of some IFN-regulated genes （such as IFI44L, PSMB8, IRF5, PSMB10, IFI27, IFIT2, TRAIL, IFIT1, IRF1, B2M and GBP1） in CNE2cells （P<0.05or0.01）, and downregulated the expression of some IFN-regulated genes （such as ISG20and AIM2） in CNE2cells （P<0.05or0.01）, respectively, while anti-miR-9correspondingly decreased the expression of miR-9mimics-upregulated IFN-related genes （i.e., IFI44L, PSMB8, IRF5, PSMB10, IFI27, IFIT2, TRAIL, IFIT1, IRF1, B2M and GBP1） in CNE2cells （P<0.05or0.01）, and increased the expression of miR-9mimics-downregulated gene （i.e., ISG20and AIM2） in CNE2cells （P<0.01）, separately. Collectively, the most significant alteration after miR-9overexpression in NPC cells was the increased expression of genes involved in IFN induction, including IFI44L, PSMB8, IRF5, PSMB10, IFI27, IFIT2, TRAIL, IFIT1, IRF1, B2M and GBP1.2. Induction of the expression of MHC class I molecules by miR-9in NPC cellsMicroarray analysis of CNE2cells infected with LV-miR-9showed the up-regulated expression of MHC class I molecules （HLA-B, HLA-H, HLA-C, HLA-F, Q8WW48HUMAN, NP001004349.1, Q6ZUW0HUMAN and O19682HUMAN） and TAP1gene （encoding antigen peptide transporter1, as ATP-binding cassette （ABC） transporter）. qRT-PCR of HLA-B, HLA-F and TAP1confirmed the microarray changes.Furthermore, when CNE2cells were transiently transfected with miR-9mimics, the expression of MHC Class I molecules （such as HLA-B and HLA-F） and TAP1was significantly upregulated, while CNE2cells transfected with anti-miR-9indicated the decreased expression of HLA-B, HLA-F and TAP1（P<0.05or0.01）. In summary, miR-9plays a significant role in regulating the expression of MHC Class I molecules.3. miR-9overexpression enhanced or reduced IL-related gene expression in NPC cellsThe microarray data derived from CNE2cells infected with LV-miR-9demonstrated the significantly altered expression in human IL-related genes （for example, IL20RB, GALT, IL7, IL1B, IL11, IL1F8, ILIA, IL6and IL7R）, which was further confirmed by qRT-PCR analyses. To fully explore the effects of both miR-9upregulation and endogenous miR-9downregulation on IL-related genes, CNE2cells were transiently transfected with miR-9mimics or anti-miR-9, respectively. The results of qRT-PCR illustrated that miR-9mimics upregulated the expression of some IL-related genes （i.e., IL20RB, GALT, IL7） in CNE2cells （P<0.05or0.01）, while other interleukin-related genes （i.e., IL1B, IL11, IL1F8, ILIA, IL6and IL7R） were remarkably downregulated by miR-9mimics in CNE2cells （P<0.05or0.01）. Conversely, anti-miR-9correspondingly decreased the expression of IL20RB, GALT and IL7in CNE2cells （P<0.05or0.01）, and increased the expression of IL1B, IL11, IL1F8, IL1A, IL6and IL7R in CNE2cells （P<0.05or0.01）, separately. Collectively, the most significant alteration after miR-9overexpression was the decreased expression of IL-related genes, including IL1B, IL11, IL1F8, IL1A, IL6and IL7R.Conclusions1. miR-9inhibits cell proliferation and tumorigenesis by targeting CCNG1, while miR-9promotes EMT and migration by repressing Klf4, suggesting that miR-9plays different roles in modulating cell proliferation, and EMT, migration&metastasis by directly targets CCNG1and Klf4in NPC, respectively;2. miR-9significantly suppresses cell proliferation and tumorigenesis of NPC by directly downregulating Hesl to deplete cancer stem cells;3. miR-9modulates the expression of interferon-regulated genes and MHC class I molecules in NPC.

miR-9对鼻咽癌增殖、肿瘤干细胞“干性”、EMT和转移的调控作用及机制

摘要3-17ABSTRACT17-32前言35-44    参考文献40-44第一章 MIR-9靶向调控CCNG1抑制鼻咽癌细胞的增殖44-74    一 材料与方法44-55    二 结果55-62    三 讨论62-63    参考文献63-65    附图65-74第二章 M[R-9对鼻咽癌肿瘤干细胞的调控作用及机制研究74-91    一 材料与方法74-76    二 结果76-80    三 讨论80-82    参考文献82-84    附图84-91第三章 C-MYC通过MIR-9靶向调控KLF4促进鼻咽癌细胞EMT、侵袭和转移的作用及机制研究91-114    一 材料与方法91-94    二 结果94-100    三 讨论100-102    参考文献102-104    附图104-114第四章 MIR-9调控鼻咽癌细胞中免疫和炎症相关基因的表达114-133    一 材料与方法114-115    二 结果115-124    三 讨论124-126    参考文献126-129    附图129-133全文小结133-134附录134-135攻读学位期间成果135-136致谢136-138统计学合格证明138

  本文关键词：miR-9对鼻咽癌增殖、肿瘤干细胞“干性”、EMT和转移的调控作用及机制，由笔耕文化传播整理发布。