【作者】 王磊；

【导师】 李玉华；

【作者基本信息】 南方医科大学 ， 内科学， 2013， 博士

【摘要】 髓系白血病是造血系统肿瘤细胞侵犯血液、骨髓以及其他组织的一种异质性疾病。未经治疗的髓系白血病按照疾病发展速度分为急性髓系白血病(acute myelocytic leukemia, AML)和慢性髓系白血病(chronic myelocytic leukemia, CML)。AML是任何一种具有自我更新能力但是不能够发育成熟的非淋巴系统造血祖细胞的无限增殖,包括粒细胞系、单核/巨噬细胞系、红细胞系以及巨核细胞系。髓系干细胞的生长失调导致骨髓抑制进而促进患者死亡,除非及时行化学治疗缓解这一过程。AML每年发病率约为3.7/100,000,并且随着年龄的增加而增加,<65时的发病率为1.9/100000,然而>65时的发病率为18.6/100000。经过年龄调整的发病率在男性(4.6/100000)中高于女性(3.0/100000)。在过去的10年中,AML的发病率有明显的增加。一旦患者确诊AML,应立即进行全身评估以及开始化学治疗。除了有明确的白血病分型以外,心、肺、肝、肾等全身系统功能的评估均应该进行。是否能够获得完全缓解(complete remission, CR)以及获得CR的持续时间等预后相关因子应该在开始治疗前进行评估。CR维持的时间越长患者获得治愈的机会越大。尽管在过去的30年里AML的治疗有了明显的进步,如新的细胞毒性药物、不同的治疗方案、信号转导通路抑制剂、免疫治疗、抗血管增生以及干细胞移植的应用,仍有20%～40%的AML患者经标准治疗后不能获得CR,预后差,更甚者仍有50%～70%的获得CR的患者会出现复发。一旦患者出现复发,在现有的治疗条件下,患者长期无病生存率小于10%。白血病复发或者复燃的主要原因包括对化疗药物的耐药以及微小残留病灶(minimal resitdual disease,MRD)。对于AML复发或者复燃的治疗仍然是白血病治疗的一大难题。单一的药物或者方案治疗AML是不可能的,因此新的可能的治疗方法还在继续寻找。表皮生长因子受体通路底物8(Epidermal growth factor receptor pathway substrate number8, Eps8)是一种表皮生长因子受体(epidermal growth factor receptor, EGFR)的酪氨酸激酶活化底物,首次由Fazioli于1993年在小鼠成纤维细胞NIH3T3中发现。Eps8能够被EGFR酪氨酸磷酸化,也能被其他几种受体酪氨酸激酶(receptor tyrosine kinases. RTKs)磷酸化。人Eps8基因位于染色体12q23-q24,有p97eps8和p68eps8两种亚型和Eps8L1-3(Eps8-related genes1-3)三种家族成员。Eps8广泛分布于上皮细胞、成纤维细胞以及部分血液细胞中。Scita等人,在血清饥饿的成纤维细胞中,应用细胞膜上皮质肌动蛋白染色的方法,证明Eps8主要成圆状分布在细胞核周围的细胞质中。Eps8具有典型的信号转导分子结构,包括(从N端到C端)公认的N端磷酸酪氨酸结合蛋白区域(PTB)、SH3区域和C端“功能区”(图1)。PTB结构域是蛋白-蛋白结合区,能够和多种磷酸酪氨酸依赖或者不依赖的多肽结合SH3区域可以选择性和特异性的与proline-X-X-asparate-tyrosine (PXXDY)形式的序列结合。Eps8的C端“功能区”通过激活GTPase和Rac可与F-actin直接结合,从而诱导Eps8定位于细胞内肌动蛋白聚集处,例如细胞膜发生皱折处。Eps8可参与多种信号通路。Eps8以Eps8-E3bl-Sosl、Eps8-Abil-p85-Sos-1和IRSp53-Eps8复合体的形式在体外被赋予Rac特异活性,并且是Rac活化诱导肌动蛋白细胞骨架重排过程中所必须的。同基因结构相对应,Eps8参与EGFR信号通路、转导通路以及有丝分裂信号通路。同时,Disanza等人证明Eps8可以通过帽化带倒钩结构末端的肌动蛋白丝,调控肌动蛋白导致的有丝分裂。也有研究证明Eps8参与非传统的Wnt信号通路,进而在脊椎动物表达ErbB4(一种酪氨酸酶受体,EGFR的一个亚家族)的神经元祖细胞的生长和有丝分裂中调控细胞的运动。在受到EGF刺激后Eps8的过度表达能够加速细胞的有丝分裂和转化能力,说明其参与RTK活化的信号通路(包括Ras/MAPK信号通路和P13K信号通路),从而促进肿瘤的发生和生长。近些年来,许多研究均证明Eps8在多种人肿瘤标本及细胞系中过表达,并参与肿瘤的形成、增殖和转移：(a)Eps8在宫颈癌、胰腺癌、甲状腺乳头状癌、乳腺癌肠癌、头颈部鳞状细胞癌(HNSCC)、口腔鳞状细胞癌(OSCC)中均过表达；(b)Eps8参与肠癌、宫颈癌、胰腺癌、HNSCC以及OSCC的发生和转移；(c)Eps8与宫颈癌、乳腺癌和甲状腺癌患者的预后密切相关,说明Eps8可能成为一种新型的公认的原癌基因。因此Eps8可能在人类肿瘤的发病机理中起着重要的作用,有望成为抑制肿瘤新的治疗靶点。Eps8在众多的实体肿瘤中具有重要的作用,且促进肿瘤的发生、增殖以及转移。但是目前为止,国内外均未有Eps8在AML中的作用的相关报道。在本次研究中,我们分四部分来研究Eps8在AML中的表达及作用：1.Eps8在AML患者骨髓单个核细胞(bone marrow mononuclear cells, BMMNCs)中以及AML细胞系中的表达情况；2.应用PCR芯片来研究Eps8相关基因在AML细胞系中的表达谱；3.PI3K/Akt抑制剂哌立福新对Eps8相关基因在AML细胞KGla中表达的影响；4.传统化疗药物柔红霉素对Eps8相关基因在AML细胞KGla中表达的影响。本研究的目的是研究Eps8在AML中的表达及作用,为Eps8成为AML治疗的新靶点提供理论依据。本研究分为四个部分：第一部分Eps8在AML患者骨髓和AML细胞系中的表达目的：本部分的研究目的是探索Eps8在AML患者和细胞系中的表达。方法：21例初发AML患者和13例健康志愿者(作为对照)入选本研究。所有参与实验的患者和志愿者均已签署知情同意书。医院资深的病理学家同时通过Wright-Giemsa染色、免疫组化以及免疫表型的分析再次确定患者的分型。每个参与实验的患者和健康志愿者,麻醉后髂脊穿刺,用含有EDTA抗凝剂的管子收集3-5ml骨髓液用于实验。Eps8在AML患者BMMNC中的表达用qRT-PCR检测,在AML细胞系中的表达采用western blot的方法检测。结果：1.1Eps8在AML患者中的表达qRT-PCR检测的结果表明Eps8在所有的AML患者及健康志愿者中均能检测到,但是AML患者中Eps8的表达高于健康志愿者(t=3.055,P=0.006)。进一步探讨Eps8与AML患者临床特征之间的关系,发现Eps8的表达水平跟患者的年龄、性别、WBC水平、血小板数、血红蛋白数以及肝脾肿大无关(P>0.05),但是跟患者在第一次化疗后是否获得CR相关,具有显著统计学意义(P=0.021)。将AML患者分为Eps8高表达组(相对对照组Eps8表达增高≥5倍)和Eps8低表达组(相对对照组Eps8表达增高<5倍)。结果显示Eps8高表达组的CR率低于Eps8低表达组(P=0.024)。这些结果说明Eps8可能能够成为AML判断预后的一个预测因子。1.2Eps8在AML细胞系中的表达Western blot结果显示Eps8在这6种细胞系中均表达,乳腺癌MCF-7细胞作为阳性对照。结果显示在AML细胞株中Eps8均有表达,但是在KGla中高表达。结论：与健康志愿者相比,Eps8在AML患者中高表达(t=3.055,P=0.006),且与患者在第一次化疗后是否获得CR相关,具有统计学意义(P=0.021)。Eps8高表达组的CR率低于Eps8低表达组(P=0.024)。Western blot检测显示AML细胞KG1a高表达Eps8。以上结果说明Eps8在AML的发病机制中有重要的作用,KG1a可以成为研究Eps8在AML作用的模型。第二部分AML细胞系KG1a中eps8相关的信号通路PCR芯片目的：这部分的目的是研究Eps8相关基因在AML细胞系KG1a中的表达。方法：应用PCR芯片研究84个代表性的Eps8相关基因在KG1a细胞中的表达概况。结果：2.1PCR芯片检测结果中高表达和低表达的基因结果显示有64个(包括Eps8)基因在KG1a细胞中高表达,有12个基因低表达。2.2KG1a细胞中相关的信号通路进一步,我们分析了高表达的基因的相关信号通路,发现几乎所有高表达基因都是细胞内信号调控分子,参与PI3K/Akt信号通路、Erk1/Erk2MAPK信号通路、IκB激酶/NFκB信号通路、JAK/STAT通路、JNK通路以及小GTPase调控的信号转导通路。部分高表达基因参与细胞的生长调控,包括调控细胞凋亡、细胞周期、细胞分化、细胞生长、细胞运动以及细胞增殖。结论：参与PI3K/Akt信号通路和小GTPase调控的信号转导通路的基因(例如AKT1, BCL2, CASP3, PIK3CA, TP53)在KG1a细胞中高表达。第三部分PI3K/Akt抑制剂哌立福新(perifosine)对KG1a的增殖、Eps8的表达、细胞凋亡以及细胞周期的影响目的：本部分的目的是探索PI3K/Akt抑制剂哌立福新对KG1a的增殖、Eps8的表达、细胞凋亡以及细胞周期的影响。方法：用台盼蓝染色法检测KG1a细胞的生长,用SPSS软件分析哌立福新作用的IC50。甲基纤维素克隆形成实验用来检测哌立福新对KGla克隆形成的影响。流式细胞术用来检测哌立福新对KGla细胞周期分布和凋亡的影响。Western blot检测哌立福新作用后KGla细胞中Eps8及细胞凋亡和周期调控因子(bcl-2,caspase-3,p21,cyclin E)的表达情况,流式细胞术检测KGla细胞中CD34+CD38-细胞的比例。结果：3.1哌立福新对KG1a细胞活性的影响结果显示不同浓度哌立福新对KG1a细胞存活率组间有显著差异(F=741.027,P=0.000),哌立福新对KG1a细胞存活率不同时间(24h、48h和72h)之间有显著差异(F=175.399,P=0.000),哌立福新作用浓度和作用时间对KG1a细胞的存活率存在交互效应(F=12.742,P=0.000)。各浓度哌立福新对KG1a24h、48h和72h的存活率均有显著意义(F=129.905,P=0.000:F=226.088,P:0.000； F=1033.382,P=0.000；),且随着浓度的增加存活率较对照组降低具有统计学意义(P<0.05)。除了1.25μmol/L哌立福新作用组外(F=4.143,P=0.074),其余浓度哌立福新作用组对KGla不同时间之间均具有显著差异(F=46.282,P=0.000; F=122.721,P=0.000:F=11.352,P=0.000；F=81.475,P=0.000；F=240.642,P=0.000)。哌立福新作用24h、48h和72h的IC50分别为13.28±3.63μmol/l、4.25±1.35μmol/l和3.65±0.85μmol/L。3.2哌立福新对KG1a细胞克隆形成的影响结果显示不同浓度哌立福新对KG1a细胞克隆形成率组间有显著差异(F=82.180,P=0.000),哌立福新对KG1a细胞克隆形成率不同时间(24h和72h)之间有显著差异(F=55.779,P=0.000),哌立福新作用浓度和作用时间对KGla细胞的克隆形成率存在交互效应(F=8.008,P=0.002)。各浓度哌立福新对KGla作用24h和72h后14天的克隆形成率均有显著意义(F=19.960,P=0.000;F=70.748,P=0.000),且随着浓度的增加克隆形成率降均较对照组降低有统计学意义(P<0.05)。各个浓度哌立福新作用组对KGla细胞14天克隆形成率不同时间之间均具有统计学意义(F=8.595,P=0.043:F=78.191,P=0.000；F=63.146,P=0.000)。3.3哌立福新对KG1a中Eps8表达的影响Western blott结果显示在Eps8在KGa中的表达随着哌立福新作用浓度及时间的增加,Eps8的条带明显减弱。3.4哌立福新对KGla细胞凋亡的影响3.4.1Wright-Giemsa染色分析哌立福新对KGla细胞形态的影响结果显示KGla细胞中出现典型的细胞凋亡形态,且凋亡率随10μmol/L哌立福新作用72h后明显增加。3.4.2哌立福新在KGla细胞诱导凋亡结果显示不同浓度哌立福新作用KGla细胞后细胞的总凋亡、早期凋亡和晚期凋亡组间均有显著差异(F=548.501,P=0.000；F=103.168,P=0.000;F=23.608, P=0.000):哌立福新作用不同时间(24h和72h)对KG1a细胞的细胞总凋亡以及早期凋亡之间均有统计学意义(F=5.198,P=0.037；F=6.713,P=0.020；),但是对细胞的晚期凋亡之间没有统计学意义(F=1.670,P=0.215)；哌立福新作用浓度和作用时间对KGla细胞的总凋亡率、早期凋亡率以及晚期凋亡率均存在交互效应(F=45.352,P=0.000；F=29.159,P=0.000；F=4.594,P=0.017)。各浓度哌立福新对KG1a24h和72h的总凋亡率均有显著意义(F=334.868, P=0.000；F=233.698,P=0.000),且随着浓度的增加总凋亡率增高(P<0.05)。只有10μmol/L和40μmol/L哌立福新作用组对KG1a的总凋亡的不同时间之间具有显著差异(F=25.122,P=0.007；F=103.933,P=0.001)。各浓度哌立福新对KG1a24h和72h的早期凋亡率均有显著意义(F=67.738,P=0.000; F=56.817,P=0.000),2.5μmol/L、10μmol/L、40μmol/L作用24h的早期凋亡率分别为(4.18±0.44)%、(6.09±0.35)%,(24.60±4.11)%,只有40μmol/L组跟对照组相比增加有统计学意义(P<0.05),作用72h后10μmol/L和40μmol/L组跟对照组相比增加有统计学意义(P<0.05)；各浓度哌立福新对KG1a24h和72h的晚期凋亡率均有显著意义(F=7.942, P=0.000; F=19.213,P=0.000),均较对照组增加有统计学意义(P<0.05)。2.5μmol/L、10μmolL和40μmol/L哌立福新作用组对KGla的早期凋亡的不同时间之间均具有统计学意义(F=10.236,P=0.033; F=67.800, P=0.001; F=23.270,P=0.008),但是只有40μmol/L哌立福新作用组对KGla的晚期凋亡的不同时间之间均具有统计学意义(F=10.164,P=0.033)。3.4.3哌立福新对KGla细胞中caspase-3和bcl-2的影响为了进一步明确哌立福新诱导凋亡的机制,采用western blot方法检测在1.25μmol/L、2.5μmol/L、5μmol/L、10μmol/L、20μmol/L、40μmol/L哌立福新作用72h后,KGla细胞中凋亡调控因子caspase-3和bcl-2的表达情况。结果显示caspase-3和bcl-2在哌立福新处理和不处理的KGla细胞中均有表达。随着哌立福新作用浓度的增加,caspase-3增加,bcl-2的表达降低。这些结果说明哌立福新能够通过增高caspase-3和降低bcl-2来诱导KGla细胞发生凋亡。3.5哌立福新对KGla细胞周期分布的影响3.5.1哌立福新对KGla周期分布的影响结果显示2.5μmol/L、10μmol/L、40μmol/L哌立福新作用KGla细胞72h后,不同浓度哌立福新作用组对Go/G1、S和G2/M期之间均有显著差异(F=741.027,P=0.000; F=380.969,P=0.000; F=42.050,P=0.000)。与对照组相比,哌立福新组细胞的Go/G1、G2/M期细胞的比例先增高后降低,S期的细胞比例先降低后升高(P<0.05)。以上结果说明哌立福新可以影响KGla细胞的周期,最终将其阻滞于S期。3.5.2哌立福新对KGla细胞中p21和cyclinE表达的影响为了研究细胞周期分布改变的机制,采用western blot方法检测在2.5μmol/L、10μmol/L、40μmol/L哌立福新作用72h后,KGla细胞中周期调控因子p21和cyclin E的表达情况。p21和cyclinE在哌立福新处理和不处理的KGla细胞中均表达。随着哌立福新浓度的增加,p21表达降低,cyclinE的表达增高。这些结果说明哌立福新能够通过降低p21以及增高cyclinE的表达来影响KGla细胞的周期分布。3.6KGla细胞中CD34和CD38的表达采用流式细胞术检测细胞表面CD34和CD38的表达情况,结果发现KGla细胞中CD34+CD38"的细胞占(98.40±1.52)%。结论：哌立福新能够剂量时间依赖性的抑制KG1a细胞的增殖、克隆形成以及Eps8的表达；能够通过增高caspase-3和降低bcl-2来诱导KG1a细胞发生凋亡,通过调节p21以及cyclinE的表达来影响KG1a细胞的周期分布；KGla细胞中CD34+CD38-的细胞占(98.40±1.52)%。第四部分传统化疗药物柔红霉素(daunorubicin)对KGla的增殖、Eps8的表达、细胞凋亡以及细胞周期的影响目的：本部分的目的是探索传统化疗药物柔红霉素对KGla的增殖、Eps8的表达、细胞凋亡以及细胞周期的影响。方法：用台盼蓝染色法检测KGla细胞的生长,用SPSS软件分析柔红霉素作用的IC50。甲基纤维素克隆形成实验用来检测柔红霉素对KG1a克隆形成的影响。流式细胞术用来检测柔红霉素对KGla细胞周期分布和凋亡的影响。Western blot检测柔红霉素作用后KG1a细胞中Eps8及细胞凋亡和周期调控因子(bcl-2, caspase-3,p21, cyclin E)的表达情况。结果：4.1柔红霉素对KGla细胞活性的影响结果显示不同浓度柔红霉素对KGla细胞存活率组间有显著差异(F=517.643,P=0.000),柔红霉素作用不同时间(24h、48h和72h)之间有显著差异(F=68.968,P=0.000),柔红霉素作用浓度和作用时间对KG1a细胞的存活率存在交互效应(F=25.673,P=0.000)。各浓度柔红霉素对KG1a24h、48h和72h的存活率均有显著意义(F=88.555,P=0.000；F=145.190,P=0.000；F=1052.825, P=0.000),且随着浓度的增加存活率降低有统计学意义(P<0.05)。除了0.1μmol/L柔红霉素作用组外(F=3.171,P=0.115),其余浓度柔红霉素作用组对KGla不同时间之间均具有显著差异(F=14.020,P=0.005；F=19.834,P=0.002；F=96.066, P=0.000；F=295.952,P=0.000;F=79.127,P=0.000).柔红霉素对KG1a细胞的24h、48h和72h的IC50分别为：(0.53±0.10)μmol/L、(0.26±0.07)μmol/L和(0.19±0.04)μmol/L。4.2柔红霉素对KGla细胞克隆形成的影响结果显示不同浓度柔红霉素作用24h后的KG1a培养14天后,发现只有0.05μmol/L、0.1μmol/L、0.2μmol/L柔红霉素作用组有克隆形成,0.4μmol/L、0.8μmol/L、1.6μmol/L组均无。0.05μmol/L、0.1μmol/L和0.2μmol/L柔红霉素对KGla细胞14天克隆形成率组间有显著差异(F=1437.188,P=0.000),柔红霉素对KGla细胞14天克隆形成率不同时间(24h,48h和72h)之间有显著差异(F=24.175,P=0.000),柔红霉素作用浓度和作用时间对KGla细胞的克隆形成率存在交互效应(F=5.189,P=0.002)。各浓度柔红霉素对KGla作用24h、48h和72h的后14天的克隆形成率均有显著意义(F=168.525,P=0.000；F=1239.212, P=0.000；F=2698.777,P=0.000),且随着浓度的增加克隆形成率降低具有统计学意义(P<0.05)。除了0.05μmol/L柔红霉素作用组(F=1.998,P=0.216)外,0.1μmol/L和0.2μmol/L浓度柔红霉素作用组对KGla不同时间之间均具有统计学意义(F=58.899,P=0.003；F=16.840,P=0.003)。4.3柔红霉素对KGla中Eps8表达的影响Western blot结果显示在Eps8在KGa中的表达随着柔红霉素作用浓度及时间的增加,Eps8的条带减弱。4.4柔红霉素对KGla细胞凋亡的影响4.4.1Wright-Giemsa染色分析柔红霉素对KG1a细胞形态的影响结果显示KGla细胞中出现典型的细胞凋亡形态,且凋亡率随0.8μmol/L柔红霉素作用72h后明显增加。4.4.2柔红霉素在KGla细胞诱导凋亡结果显示不同浓度柔红霉素作用KG1a细胞后细胞的总凋亡、早期凋亡和晚期凋亡组间均有显著差异(F=1002.926,P=0.000；F=520.100,P=0.000；F=611.588, P=0.000);柔红霉素作用不同时间(24h、48h和72h)对KGla细胞的细胞总凋亡、早期凋亡以及晚期凋亡之间均有统计学意义(F=3903.148,P=0.000;F=1164.274, P=0.020；F=1259.918,P=0.000);柔红霉素作用浓度和作用时间对KGla细胞的总凋亡率、早期凋亡率以及晚期凋亡率存在交互效应(F=627.257,P=0.000； F=324.245,P=0.000；F=541.471,P=0.000)。各个浓度柔红霉素对KGla24h和72h的总凋亡率均有显著意义(F=34.403,P=0.000；F=1280.583,P=0.000),且随着浓度的增加总凋亡率升高具有统计学意义(P<0.05).0.2μmol/L.0.8μmol/L和1.6μmol/L柔红霉素作用组对KG1a的总凋亡的不同时间之间均具有显著差异(F=1638.280,P=0.000；F=1825.615, P=0.000；F=1002.926,P=0.000)。各浓度柔红霉素对KG1a24h和72h的早期凋亡率均有显著意义(F=12.439,P=0.000；F=918.409,P=0.000),且随着浓度的增加早期凋亡率上升具有统计学意义(P<0.05)；0.2μmol/L、0.8μmol/L、1.6μmol/L柔红霉素对KG1a2411和72h的晚期凋亡率均有显著意义(F=6.612,P=0.015；F=800.619,P=0.000),各浓度柔红霉素作用24h的晚期凋亡率分别为(10.60±1.25)%、(11.66±0.83)%、(12.48±0.53)%,只有0.8μmol/L和1.6μmol/L组与对照组相比增加有统计学意义(P<0.05),作用72h后三组跟对照组相比均增加有统计学意义(P<0.05)。0.2μmol/L、0.8μmol/L和1.6μmol/L柔红霉素作用组对KGla的早期凋亡的不同时间之间均具有统计学意义(F=25.773,P=0.007；F=451.418,P=0.000；F=2789.011, P=0.000),对KGla的晚期凋亡的不同时间之间也均具有统计学意义(F=25.878, P=0.007；F-3018.753,P=0.000；F=149.171,P=0.000).4.4.3柔红霉素对KGla细胞中caspase-3和bcl-2的影响为了进一步明确哌立福新诱导凋亡的机制,采用western blot方法检测在0.05μmol/L,0.1μmol/L,0.2μmol/L,0.4μmol/L,0.8μmol/L,1.6μmol/L柔红霉素作用72h后,KGla细胞中凋亡调控因子caspase-3和bcl-2的表达情况。结果表明caspase-3和bcl-2在柔红霉素处理和不处理的KGla细胞中均有表达。随着柔红霉素作用浓度的增加,caspase-3增加,bcl-2的表达降低。4.5柔红霉素对KGla细胞周期分布的影响4.5.1柔红霉素对KGla周期分布的影响结果显示不同浓度柔红霉素作用KGla细胞后细胞的Go/G1、S和G2/M期细胞的比例间均有显著差异(F=83.353,P=0.000；F=163.987,P=0.000；F=392.806, P=0.000):柔红霉素作用不同时间(24h和72h)对KGla细胞的细胞Go/G1、S和G2/M期细胞的比例之间均有统计学意义(F=315.691,P:0.000:F=35.385, P=0.020；F=536.652,P=0.000)；柔红霉素作用浓度和作用时间对KGla细胞的Go/G1、S和G2/M期细胞的比例存在交互效应(F=57.981,P=0.000；F=163.987, P=0.000；F=86.004.P=0.000)。各浓度柔红霉素对KGla24h和72h的Go/G1期的细胞比例有显著意义(F=6.081,P=0.018；F=84.500,P=0.000),且随浓度的增加24h时细胞的比例变化不大,但是72h时细胞比例同对照组相比减少具有统计学意义(P<0.05)。0.2μmol/L、0.8μmol/L和1.6μmol/L柔红霉素作用组对KGla的G0/G1期细胞比例的不同时间之间均具有显著差异(F=175.145,P=0.000；F=72.017,P=0.001； F=869.490,P=0.000)。最终柔红霉素对KGla细胞G0/G1期的细胞比例降低。各浓度柔红霉素对KG1a24h和72h的S期的细胞比例有显著意义(F=43.251,P=0.000；F=126.041,P=0.000),且随着浓度的增加细胞比例先减少后增加,除了0.2μmol/L柔红霉素作用24h外,均与对照组相比具有统计学意义(P<0.05)。0.2μmol/L、0.8μmol/L和1.6μmol/L柔红霉素作用组对KG1a的S期细胞比例的不同时间之间均具有显著差异(F=13.336,P=0.022; F=100.644, P=0.001; F=256.210,P=0.000).最终柔红霉素对KG1a细胞S期的细胞比例增加。各浓度柔红霉素对KG1a24h和72h的G2/M期的细胞比例有显著意义(F=137.7651,P=0.000; F=270.731,P=0.000),且随着浓度的增加细胞比例先增加后降低,除了1.6μmol/L柔红霉素作用24h外,均与对照组相比具有统计学意义(P<0.05).0.2μmol/L、0.8pmol/L和1.6μmol/L柔红霉素作用组对KG1a的G2/M期细胞比例的不同时间之间均具有显著差异(F=280.308, P=0.000;F=238.568, P=0.000; F=59.312, P=0.002)。最终柔红霉素对KG1a细胞G0/G1期的细胞比例增加,但增加不如S期明显。柔红霉素能够影响KG1a细胞的周期分布,将其阻滞在S期。4.5.2柔红霉素对KG1a细胞中p21和cyclinE表达的影响为了研究细胞周期分布改变的机制,采用western blot方法检测在0.05μmol/L,0.1μmol/L,0.2μmol/L,0.4μmol/L,0.8μmol/L,1.6μmol/L柔红霉素作用72h后,KG1a细胞中周期调控因子p21和cyclin E的表达情况。p21和cyclinE在柔红霉素处理和不处理的KG1a细胞中均表达。随着柔红霉素浓度的增加,p21表达降低,cyclin E的表达增高。这些结果说明柔红霉素能够通过增高p21以及cyclinE的表达来影响KG1a细胞的周期分布。结论：柔红霉素能够剂量时间依赖性的抑制KG1a细胞的增殖、克隆形成以及Eps8的表达；通过增加caspase-3和降低bcl-2来诱导KG1a细胞发生凋亡,通过调节p21以及cyclin E的表达来影响KG1a细胞的周期分布。结论：1.与健康志愿者相比,Eps8在AML患者中明显高表达(t=3.055,P=0.006),且与患者在第一次化疗后是否获得CR相关,具有统计学意义(P=0.021)。Eps8高表达组的CR率明显低于Eps8低表达组(P=0.024)。Eps8在KGla细胞中明显高表达。以上结果说明Eps8在AML的发病机制中有重要的作用,KGla可以成为研究Eps8在AML作用的模型。2.参与PI3K/Akt信号通路和小GTPase调控的信号转导通路的基因(例如AKT1, BCL2, CASP3, PIK3CA, TP53)在KGla细胞中高表达。3.哌立福新能够剂量时间依赖性的抑制KGla细胞的增殖、克隆形成以及降低KGla细胞中Eps8的表达；哌立福新能够通过增加caspase-3和降低bcl-2来诱导KGla细胞发生凋亡。以及通过降低p21以及增加cyclinE的表达来影响KGla细胞的周期分布。KGla细胞中CD34+CD38-的细胞占(98.40±1.52)%。4.柔红霉素能够剂量时间依赖性的抑制KGla细胞的增殖、克隆形成以及降低KGla细胞中Eps8的表达；柔红霉素能够通过增加caspase-3和降低bcl-2来诱导KGla细胞发生凋亡。以及通过降低p21以及增加cyclinE的表达来影响KGla细胞的周期分布。更多还原

【Abstract】 The myeloid leukemias are a heterogeneous group of diseases characterized by infiltration of the blood, bone marrow, and other tissues by neoplastic cells of the hematopoietic system. Based on their untreated course, the myeloid leukemias have traditionally been designated acute or chronic. Acute myelocytic leukemia (AML) is a clonal expansion of any one of several nonlymphoid hematopoietic progenitors that retain the capacity of self-renewal, but are severely limited in their ability to differentiate into functional mature cells. These various progenitors include cells of granulocytic, monocyte/macrophage, erythroid, and megakaryocytic lineage. The disordered growth in the myeloid stem cell compartment leads to the patient’s death from bone marrow failure, unless a successful therapeutic strategy is employed. The incidence of AML is～3.7per100,000people per year, and increases with age; it is1.9/100000in individuals<65years and18.6/100000in those>65. The age-adjusted incidence is higher in men than in women (4.6/100000versus3.0/100000). A significant increase in AML incidence has occurred over the past10years. Once the diagnosis of AML is suspected, a rapid evaluation and initiation of appropriate therapy should follow. In addition to clarifying the subtype of leukemia, initial studies should evaluate the overall functional integrity of the major organ systems, including the cardiovascular, pulmonary, hepatic, and renal systems. Factors that have prognostic significance, either for achieving complete remission (CR) or for predicting the duration of CR, should also be assessed before initiating treatment. The length of CR, and the curability of AML.Although considerable progress has been made over the past3decades,, such as new cytotoxic agents、differentiation therapy、singnal transduction inhibitors、 immunotherapy、anti-angiogenesis and stem cell transplantation (HSCT),20%-40%AML patients still do not achieve CR with standard therapy and continue to have a poor prognosis, furthermore50%～70%patients achieved CR are likely to relapse after CR. Once patients have relapsed, with current therapies, the chance of long-term disease free survival is less than10%. The cause of leukemia refractoriness or relapse is often not clear but likely relates to multiple factors including resistance mechanisms to chemotherapy drugs and minimal resitdual disease(MRD). The treatment of patients with refractory or relapsed AML remains a major challenge for the leukemia community and it is not possible to identify a single regimen or approach as the standard of care in relapsed and refractory AML. New and promising approaches are being explored, however.Epidermal growth factor receptor pathway substrate number8(Eps8), a substrate for the tyrosine kinase activity of the epidermal growth factor receptor (EGFR), was first discovered by Fazioli in NIH3T3murine fibroblasts in1993. It was tyrosine-phosphorylated by EGFR, and also by several other receptor tyrosine kinases (RTKs). The human eps8gene locus was mapped to chromosome12q23-q24. There are two eps8isoforms (p97eps8and p68eps8) and other three eps8family members (Eps8-related genes:Eps8L1-3). Eps8was expressed in all epithelial and fibroblastic lines and in some hematopoietic cells. Scita et al studied, in serum-starved fibroblasts, Eps8displays a punctuate, cytoplasmic perinuclear distribution with some staining at sites of cortical actin accumulation of plasma membrane.Eps8is a structural organization typical of a signaling molecule, contains (from N to C terminus) a putative N-terminal phosphotyrosine binding protein (PTB) domain, a SH3domain and a C-terminal "effector region"(Fiure.l). The PTB domain is a protein-protein interaction module, which binds to a variety of peptides both in a phosphotyrosine-dependent and-independent fashion. SH3binds to proline-X-X-asparate-tyrosine(PXXDY) motifs both specifically and selectively. The C-terminal effector region of Eps8directly binds to F-actin, by activating the GTPase, Rac, and directs Eps8to localization within the cell where actin polymerization occurs, such as membrane ruffles.Eps8participanted in several signaling pathways. The complexs of Eps8-E3b1-Sosl, Eps8-Abil-p85-Sos-l and IRSp53-Eps8are endowed with Rac-specific activity in vitro and are required for Rac activation leading to actin cytoskeletal remodeling. According with the gene structure of Eps8, it participates in EGFR signaling, trafficking and mitogenic signaling. Disanza, et al. demonstrate that Eps8controls actin-based motility by capping the barbed ends of actin filaments. It is also proved that Eps8might participate in non-canonical Wnt signaling to control the movements of cells during vertebrate development and motility of neuronal progenitor cells expressing ErbB4. Overexpression of Eps8facilitates increased mitogenesis and transforming ability in response to EGF, indicating its participating in RTK-activated signaling pathways(including Ras/MAPK signaling pathway and PI3K signaling pathway) leading to tumorigenesis and tumor proliferation.Recently, many studies observed that Eps8is overexpressed in several human cancer samples and in human cancer cell lines, and it contributed to the tumorigenicity, tumor proliferation and tumor metastasis of them:(a) Eps8was overexpressed in human cancers, including cervical carcinomas, pancreatic cancers, papillary thyroid carcinomas, breast cancers, colon cancers and head and neck squamous cell carcinomas (HNSCC), and oral squamous cell carcinoma (OSCC);(b) Eps8contributed to thetumorigenicity and aggressiveness of colon cancers, cervical cancers, HNSCC, pancreatic cancer and OSCC, to the motility and mitogenesis of colon cancer cells;(c) Eps8could be a predictor factor of the survival of cervical cancer patients, breast patients and thyroid cancerpatients, identified Eps8as a novel putative oncogene. So Eps8is probably to be important in the pathogenesis of human cancers, may have potential to be a therapeutic target for inhibition of the cancer progression.Eps8has essential role in many human solid cancers and it contributed to their tumorigenicity, tumor proliferation and tumor metastasis, but there is no study about the interaction between the eps8and AML. In the present study, we detect four parts to investigate the relationship between Eps8and AML:1. The expression of Eps8in AML bone marrow cells and AML cell lines;2. The PCR array was used to detect the expression of Eps8-related genes in AML cell line KG1a;3. Effects of PI3K/Akt inhibitor perifosine for the expression of Eps8-related genes in AML cell line KG1a;4. Effects of traditional chemotherapy medicine daunorubicin for the expression of Eps8-related genes in AML cell line KG1a.The aim of the study was to explore the function of Eps8in AML, and to provide theoretical evidence for therapeutic target of Eps8in AML.The present study includes four parts:Part1The expression of Eps8in AML bone marrow and AML cell linesObjective:The aim of the part is to explore the possible expression of Eps8and in AML patients and cell lines.Methods:Twenty-one patients with do novo AML and10healthy volunteers (as the control) were enrolled in this study. All patients and volunteers were informed about the study and signed a form consenting to the procedure. The institutional pathologist examined the bone marrow samples simultaneously by review of the Wright-Giemsa-stained slide, enzyme histochemistry, and standard immunophenotype classified into the different types. About3-5mL bone marrow aspirate was collected into a tube containing EDTA by iliac crest puncture of each anesthetized patient before treatment and healthy volunteer. The expression of Eps8was detected by quantitative reverse transcription polymerase chain reaction (qRT-PCR) in AML patients and by western blot in AML cell lines.Results:1.1Expression of Eps8in AML patientsqRT-PCR analysis showed that the expression of Eps8was detected in all the AML patients and healthy volunteers, but Eps8was expressed significantly higher in AML patients(t=3.055, P=0.006). Furthermore, we have explored the correlation between Eps8and the patients’clinicopathological characteristics. The result showed that there was no significant association between the expression level of Eps8and sex, age, the WBC count, platelet count, hemoglobin, presence of splenomegaly and hepatomegaly (P>0.05), but we found that the patients with AML achieved CR or not after one course of chemotherapy significantly correlated with the expression of the Eps8(P=0.021). All the patients with AML were subgrouped as either Eps8high expression group (compared with control group, Eps8fold change≥5) or Eps8low expression group (compared with control group, Eps8fold change<5). The result showed that the CR rate in Eps8high expression group was significantly lower than low expression group (P=0.024). This indicated that Eps8may have broader implication in clinical prognostic value.1.2Expression of Eps8in AML cell linesWestern blot analysis showed that the expression of Eps8was detected in all the6AML cell lines, the breast cancer cell line MCF-7was used as the internal control. The result showed that Eps8was highly expressed in KG1a AML cell line.Conclusion:The expression of Eps8was significantly increased in AML patients compared with the healthy volunteers (t=3.055, P=0.006), and correlated with the AML patients achieved CR or not after one course of chemotherapy (P=0.021). The CR rate in Eps8high expression group was significantly lower than low expression group (P=0.024). Eps8was obviously high expressed in KG1a cell line, indicating Eps8may play an important role in the pathogenesis of AML and KG1a maybe a good study model for the investigation of the function of Eps8in the AML. Part2The PCR array was used to detect the expression of Eps8-related genes in AML cell line KGlaObjective:The aim of the part is to explore the expression of Eps8-relsted genes in AML cell line KG1a.Methods:We measured84representative Eps8-related genes using PCR array.Results:2.1The high and low expression genes in KGla cells detected by PCR assayThe result showed that there are64genes (including Eps8) high expressed in KG1a cells and12genes low expressed.2.2The releted signaling pathway in the KGla cellsFurthormore, we found that almost all the high expressed genes were intracellular signaling molecules, participating in the PI3K/Akt signaling pathway, Erkl/Erk2MAPK signaling pathway, IκB Kinase/NFκB cascade, JAK/STAT cascade, JNK cascade, small GTPase mediated signal transduction, and some of them can affect cell survival and growth, through the cell apoptosis, cell cycle, cell differentiation, cell growth, cell motility and cell proliferation.Conclusion:Genes (such as AKT1, BCL2, CASP3, PIK3CA, TP53) which participant in the PI3K/Akt signaling pathway and small GTPase mediated signal transduction high expressed in KG1a cells.Part3Effects of AKT/PI3K inhibitor perifosine for proliferation, the expression of Eps8, cell apoptosis and cell cycle in AML cell line KGlaObjective:The aim of the part is to explore the effects of AKT/PI3K inhibitor perifosine for proliferation, the expression of Eps8, cell apoptosis and cell cycle in AML cell line KG1a.Methods:The growth of KG1a cells was determined using Trypan Blue assay, then the IC50of the perifosine for KG1a cells was calculated by the SPSS. The effect of perifosine for the clone formation of KG1a cells after treated with different concentration perifosine was detected by methylcellulose colony-forming assay. The effect of perifosien for the distribution of cell cycle and the apoptosis rate were measured with flow cytometry. The effect of perifosine for the expression of Eps8and the cell apoptosis and cell cycle regulators (bcl-2, caspase-3, p21, cyclin E) were detected by western blot. The expression of CD34and CD38in the surface of KG1a cells were measured with flow cytometry.Results:3.1Effects of perifosine on the survival rate in KGla cellsThe results of the Trypan Blue assay showed the difference of the survival rate of KG1a cells between different concentration perifosine groups was significant (F=741.027, P=0.000), the difference of the survival rate of KGla cells between different treatment times of perifosine groups(24h,48h and72h) was significant (F=175.399,P=0.000), the interaction between the perifosine treatment concentration and time for the survival rate of KG1a cells was existed(F=12.742, P=0.000). The difference of the survival rates of KGla cells on24h,48h and72h after treated with different concentration perifosine was significant (F=129.905, P=0.000; F=226.088, P=0.000; F=1033.382, P=0.000), the survival rates were decreased significantly comparing with the control group when the concentration of perifosine was increased (P<0.05). Along with the1.25μmol/L perifosien group(F=4.143, P=0.074), the perifosien treatment groups for the survival rate of KGla cells were significantly different between the different treatment times(24h,48h and72h)(F=46.282, P=0.000; F=122.721,P=0.000; F=11.352, P=0.000; F=81.475,P=0.000; F=240.642, P=0.000). The half inhibition rates (IC50) of perifosine on the KGla cells were13.28±3.63μmol/L,4.25±1.35μmol/L and3.65±0.85μmol/L for24h,48h and72h, separately.3.2Effects of perifosine on the clone formation of KGla cellsThe results showed the difference of the clone formation rate of KG1a cells between different concentration perifosine groups was significant (F=82.180, P=0.000), the difference of the clone formation rate of KG1a cells between different treatment times of perifosine groups (24h,48h and72h) was significant (F=55.779,P=0.000), the interaction between the perifosine treatment concentration and time for the clone formation rate of KGla cells was existed (F=8.008, P=0.002). The difference of the clone formation rates of KGla cells on day14after treated with different concentration perifosine for24h and72h was significant (F=19.960, P=0.000; F=70.748, P=0.000), the clone formation rates were decreased significantly comparing with the control group when the concentration of perifosine was increased (P<0.05).. the perifosien treatment groups for the clone formation rate of KG1a cells on day14were significantly different between the different treatment times (F=8.595, P=0.043; F=78.191, P=0.000; F=63.146,.P=0.000).3.3Effects of perifosine on the expression of Eps8in the KGla cellsThe western blot assay showed that the expression of Eps8was decreased allowing with the concentration of perifosien and the treated time increasing.3.4Effects of perifosine on the apoptosis of KGla cells3.4.1Morphological analysis of the effects of perifosien by Wright-Giemsa stainKGla cells with characteristics of apoptosis was confirmed by Wright-Giemsa stain, and the results showed the percentage of apoptosis were increased after treatment with10μmol/L perifosine for72h.3.4.2Perifosien induced apoptosis in KGla cells The results showed the difference of the total apoptosis rate, early apoptosis rate and late apoptosis rate of KG1a cells between different concentration perifosine groups was significant (F=548.501,P=0.000; F=103.168,P=0.000; F=23.608, P=0.000); the difference of the total apoptosis rate and early apoptosis rate of KG1a cells between different treatment times of perifosine groups (24h,48h and72h) was significant (F=5.198, P=0.037; F=6.713, P=0.020), but the diference of the late apoptosis rate was not (F=1.670, P=0.215); the interaction between the perifosine treatment concentration and time for the total apoptosis rate, early apoptosis rate and late apoptosis rate of KGla cells was existed(F=45.352, P=0.000; F=29.159, P=0.000; F=4.594,P=0.017).The difference of the total apoptosis rates of KGla cells after treated with different concentration perifosine for24h and72h were significant (F=334.868, P=0.000; F=233.698, P=0.000), the total apoptosis rates were increased significantly comparing with the control group when the concentration of perifosine was increased (P<0.05). Only the10μmolL and40μmol/L perifosien treatment groups for the total apoptosis rate of KGla cells were significantly different between the different treatment times (F=25.122, P=0.007; F=103.933, P=0.001).The difference of the early apoptosis rates of KGla cells after treated with different concentration perifosine for24h and72h were significant (F=67.738, P=0.000; F=56.817, P=0.000), the early apoptosis rates of KGla after treated with2.5μmol/L,10μmol/L and40μmol/L perifosine were (4.18±0.44)%、(6.09±0.35)%,(24.60±4.11)%, respectively, but only the40μmol/L perifosine group was increased significantly comparing with the control group (P<0.05); when the treatment time was72h,10μmol/L and40μmol/L perifosine group was increased significantly comparing with the control group (F<0.05). The difference of the late apoptosis rates of KGla cells after treated with different concentration perifosine for24h and72h were significant (F=7.942, P=0.000; F=19.213, P=0.000), the late apoptosis rates were increased significantly comparing with the control group when the concentration of perifosine was increased (P<0.05). All the2.5μmol/L,10μmol/L and40μmol/L perifosine treatment groups for the early apoptosis rate of KG1a cells were significantly different between the different treatment times (F=10.236, P=0.033: F=67.800, P=0.001; F=23.270,P=0.008), but only the40μmol/L perifosien treatment groups for the late apoptosis rate of KGla cells were significantly different between the different treatment times (F=10.164, P=0.033).3.4.3Effects of perifosine on the expression of caspase-3and bcl-2in KG1a cellsTo study the underlying apoptotic mechanisms, expression of apoptosis regulators, caspase-3and bcl-2, were assessed after treatment with1.25μmol/L,2.5μmol/L,5μmol/L,10μmol/L,20μmol/L and40μmol/L perifosine for72h by western blot. The caspase-3and bcl-2was detected in all the groups treated with or without perifosine. According with the concentration of perifosine increasing, the expression of caspase-3was increased, and the expression of bcl-2was decreased. The result indicated that perifosine can induce the apoptosis of KG1a cells by activating caspase-3and decreasing bcl-2.3.5Effects of perifosine on the cell cycle distribution of KGla cells3.5.1Effects of perifosine on the cell cycle distribution of KGla cellsThe difference of the percentage of Go/G1, S and G2/M phase cells in KG1a cells after treating wirh2.5μmol/L,10μmol/L and40μmol/L perifosine for72h were significant (F=741.027, P=0.000; F=380.969,P=0.000; F=42.050, P=0.000). The percentage of Go/G1and G2/M phase cells were increased first and then decreased and the percentage of S phase cells was decreased first and then increased comparing with the control group when the concentration of perifosine was increased (P<0.05). These results indicated that perifosine could affect the cell cycle distribution and arrest KG1a cells at S phase.3.5.2Effects of perifosine on the expression of p21and cyclinE in KGla cellsTo study the underlying mechanisms of the cell cycle redistribution, expression of cell cycle regulators, p21and cyclinE, were assessed after treatment with2.5μmol/L,10μmol/L and40μmol/L perifosine for72h by weastern blot. P21and cyclinE were detected in all the groups treated with or without perifosine for72h. According with the concentration of perifosine increasing, the expression of p21was decreased and cyclin E was increased. The result indicated that perifosine can induce the cell cycle redistribution in KG1a cells by decreasing p21and increasing cyclin E.3.6The expression of the CD34and CD38in KGla cellsThe expression of CD34and CD38in the surface of KGla cells were measured with flow cytometry, and the result showed that the percentage of CD34+CD38-in KG1a cells was (98.40±1.52)%.Conclusion:Perifosine can significantly inhibite the proliferation, the clone formation of KGla cells and the expression of Eps8in KGla cells on a dose-and time-depended matter; Perifosine can induce the apoptosis of KGla cells by activating caspase-3and decreasing bcl-2, and induce the cell cycle redistribution of KGla cells by decreasing p21and increasing cyclin E; the percentage of CD34+CD38-in KG1a cells was(98.40±1.52)%.Part4Effects of traditional chemotherapy medicine daunorubicin (DNR) for proliferation, the expression of Eps8, cell apoptosis and cell cycle in AML cell line KGla Objective:The aim of the part is to explore the effects of traditional chemotherapy medicine daunorubicin(DNR) for proliferation, the expression of Eps8, cell apoptosis and cell cycle in AML cell line KG1a.Methods:The growth of KG1a cells was determined using Trypan Blue assay, then the IC50of the perifosine for KG1a cells was calculated by the SPSS. The effect of DNR for the clone formation of KGla cells after treated with different concentration DNR was detected by methylcellulose colony-forming assay. The effect of DNR for the apoptosis rate and the distribution of cell cycle were measured with flow cytometry. The effect of DNR for the expression of Eps8and the cell apoptosis and cell cycle regulators (bcl-2, caspase-3, p21, cyclin E) were detected by western blot.Results:4.1Effects of DNR on the survival rates in KGla cellsThe results showed the difference of the survival rate of KGla cells between different concentration DNR groups was significant (F=517.643, P=0.000), the difference of the survival rate of KGla cells between different treatment times of DNR groups (24h,48h and72h) was significant (F=68.968, P=0.000), the interaction between the DNR treatment concentration and time for the survival rate of KG1a cells was existed (F=25.673, P=0.000). The difference of the survival rates of KG1a cells on24h,48h and72h after treated with different concentration DNR was significant (F=88.555, P=0.000; F=145.190, P=0.000; F=1052.825,P=0.000), the survival rates were decreased significantly comparing with the control group when the concentration of perifosine was increased (P<0.05). Along with the0.1μmol/L DNR group (F=3.171, P=0.115), the DNR treatment groups for the survival rate of KGla cells were significantly different between the different treatment times (24h,48h and72h)(F=14.020,P=0.005; F=19.8345,PP=0.002; F=96.066, P=0.000; F=295.952, P=0.000; F=79.127,P=0.000). The IC50of DNR on the KGla cells were (0.53±0.10)μmol/L,(0.26±0.07)μmol/L and (0.19±0.04)umol/L for24h,48h and72h, separately.4.2Effects of DNR on the clone formation of KG1a cellsThe results showed when treated with different concentration of DNR for24h, the clone formation of KGla cells was founded only in the0.05μmol/L, O.lμmol/L and0.2μmol/L NDR treatment groups, not in0.4μmol/L,0.8μmol/L and1.6μmol/L DNR goups. The difference of the clone formation rate of KGla cells on day24between0.05μmol/L,0.1μmol/L and0.2μmol/L DNR groups was significant (F=1437.188, P=0.000), the difference of the clone formation rate of KGla cells between different treatment times of DNR groups (24h,48h and72h) was significant (F=24.175, P=0.000), the interaction between the DNR treatment concentration and time for the clone formation rate of KG1a cells was existed(F=5.189, P=0.002). The difference of the clone formation rates of KG1a cells on day14after treated with different concentration DNR for24h,48h and72h was significant (F=168.525, P=0.000; F=1239.212,P=0.000; F=2698.777, P=0.000), the clone formation rates were decreased significantly comparing with the control group when the concentration of DNR was increased (P<0.05). Along with0.05μmol/L DNR group (F=1.998,P=0.216),0.1μmol/L and0.2μmol/L DNR treatment groups for the clone formation rate of KGla cells on day14were significantly different between the different treatment times (F=58.899,P=0.003; F=16.840,P=0.003).4.3Effects of DNR on the expression of Eps8in the KGla cellsThe western blot assay showed that the expression of Eps8was decreased accompanied with the concentration of DNR and the treated time increasing.4.4Effects of DNR on the apoptosis of KGla cells4.4.1Morphological analysis of the effects of DNR by Wright-Giemsa stainKG1a cells with characteristics of apoptosis were confirmed by Wright-Giemsa stain, and the results showed the percentage of apoptosis were increased after treatment with10μmol/L DNR for72h.4.4.2DNR induced apoptosis in KGla cellsThe results showed the difference of the total apoptosis rate, early apoptosis rate and late apoptosis rate of KG1a cells between different concentration DNR groups was significant (F=1002.926, P=0.000; F=520.100,P=0.000; F=611.588, P=0.000), the difference of the total apoptosis rate, early apoptosis rate and late apoptosis rate of KGla cells between different treatment times of DNR groups (24h and72h) was significant (F=3903.148, P=0.000; F=1164.274, P=0.020; F=1259.918, P=0.000), the interaction between the DNR treatment concentration and time for the total apoptosis rate, early apoptosis rate and late apoptosis rate of KG1a cells was existed(F=627.257, P=0.000; F=324.245,P=0.000; F=541.471,P=0.000).The difference of the total apoptosis rates of KGla cells after treatment with different concentration DNR for24h and72h were significant (F=34.403, P=0.000; F=1280.583,P=0.000), the total apoptosis rates were increased significantly comparing with the control group when the concentration of DNR was increased (P<0.05).0.2μmol/L,0.8μmol/L and1.6μmol/L DNR treatment groups for the total apoptosis rate of KGla cells were significantly different between the different treatment times (F=1638.280,P=0.000; F=1825.615,P=0.000; F=1002.926,P=0.000).The difference of the early apoptosis rates of KGla cells after treated with different concentration DNR for24h and72h were significant (F=12.439, P=0.000; F=918.409, P=0.000), the early apoptosis rates were increased significantly comparing with the control group when the concentration of perifosine was increased (P<0.05); The difference of the late apoptosis rates of KGla cells after treated with0.2μmol/L,0.8μmol/L and1.6μmol/L DNR for24h and72h were significant (F=6.612, P=0.015; F=800.619, P=0.000), the late apoptosis rates of KGla with DNR treatment for24h were (10.60±1.25)%,(11.66±0.83)%,(12.48±0.53)%, respectively, but only the0.8μmol/L and1.6μmol/L DNR group were increased significantly comparing with the control group (P<0.05); when the treatment time was72h, all the three DNR groups were increased significantly comparing with the control group (P<0.05). All the0.2μmol/L,0.8μmol/L and1.6μmol/L DNR treatment groups for the early apoptosis rate of KG1a cells were significantly different between the different treatment times (F=25.773, P=0.007; F=451.418, P=0.000; F=2789.011, P=0.000), and also for the late apoptosis rate of KG1a cells were significantly different between the different treatment times (F=25.878, P=0.007; F=3018.753, P=0.000; F=149.171, P=0.000).4.4.3Effects of DNR on the expression of caspase-3and bcl-2in KGla cellsTo study the underlying apoptotic mechanisms, expression of apoptosis regulators, caspase-3and bcl-2, were assessed after treatment with0.05μmol/l,0.1μmol/L,0.2μmol/L,0.4μmol/L,0.8μmol/L and1.6μmol/L DNR for72h. Caspase-3and bcl-2was detected in all the groups treated with or without perifosine for72h. According with the concentration of DNR increasing, the expression of caspase-3was increased, and the expression of bcl-2was decreased. The result indicated that DNR can induce the apoptosis of KGla cells by increasing caspase-3and decreasing bcl-2.4.5Effects of DNR on the cell cycle distribution of KGla cells4.5.1Effects of DNR on the cell cycle distribution of KGla cellsThe results showed the difference of the percentage of Go/G1, S and G2/M phase cells in KGla cells between different concentration DNR groups was significant (F=83.353, P=0.000; F=163.987, P=0.000; F=392.806, P=0.000), the difference of the percentage of G0/G1, S and G2/M phase cells in KGla cells between different treatment times of DNR groups (24h and72h) was significant (F=315.691, P=0.000; F=35.385, P=0.020; F=536.652, P=0.000), the interaction between the DNR treatment concentration and time for the percentage of Go/G1, S and G2/M phase cells in KG1a cells was existed (F=57.981,P=0.000; F=163.987, P=0.000; F=86.004, P=0.000).The difference of the percentage of Go/G1phase cells in KG1a after treatment with different concentration DNR for24h and72h were significant (F=6.081, P=0.018; F=84.500, P=0.000), the percentage of Go/G1phase cells were decreased with DNR treatment for24h, and significantly decreased with DNR treatment for72h, comparing with the control group when the concentration of DNR was increased (P<0.05).0.2μmol/L,0.8μmol/L and1.6μmol/L DNR treatment groups for the percentage of Go/G1phase cells were significantly different between the different treatment times (F=175.145, P=0.000; F=72.017, P=0.001; F=869.490, P=0.000). DNR induced the percentage of Go/G1phase cells decreased.The difference of the percentage of S phase cells in KG1a after treatment with different concentration DNR for24h and72h were significant (F=43.251, P=0.000; F=126.041,.P=0.000), the percentage of S phase cells were increased first and then decreased significantly comparing with the control group when the concentration of DNR was increased (P<0.05), except the0.2μmol/L DNR goup for24h.0.2μmol/L,0.8μmol/L and1.6μmol/L DNR treatment groups for the percentage of S phase cells were significantly different between the different treatment times (F=13.336, P=0.022; F=100.644, P=0.001; F=256.210, P=0.000). DNR induced the percentage of S phase cells increased finally.The difference of the percentage of G2/M phase cells in KGla after treatment with different concentration DNR for24h and72h were significant (F137.7651, P=0.000; F=270.731, P=0.000), the percentage of G2/M phase cells were decreased first and then increased significantly comparing with the control group when the concentration of DNR was increased (P<0.05), except the1.6μmol/L DNR goup for 24h.0.2μmol/L,0.8μmol/L and1.6μmol/L DNR treatment groups for the percentage of G2/M phase cells were significantly different between the different treatment times (F=280.308, P=0.000; F=238.568, P=0.000; F=59.312, P=0.002). DNR induced the percentage of G2/M phase cells increased.These results indicated that DNR could affect the cell cycle distribution and arrest KG1a cells at S phase.4.5.2Effects of DNR on the expression of p21and cyclin E in KGla cellsTo study the underlying the cell cycle redistribution mechanisms, the expression of cell cycle regulators, p21and cyclin E, were assessed after treatment with0.05μmol/L,0.1μmol/L,0.2μmol/L,0.4μmol/L,0.8μmol/L,1.6μmol/L DNR for72h. P21and cyclin E were detected in all the groups treated with or without DNR treatment for72h. According with the concentration of DNR increasing, the expression of both p21was decreased and cyclin E was increased. The result indicated that DNR can induce the cell cycle redistribution of KG1a cells by decreasing p21and increasing cyclin E.Conclusion:DNR can significantly inhibite the proliferation, the clone formation of KG1a cells and in the expression of Eps8in KGla cells on a dose-and time-depended matter; DNR can induce the apoptosis of KGla cells by increasing caspase-3and decreasing bcl-2, and induce the cell cycle redistribution of KG1a cells by decreasing p21and increasing cyclin E.Conclusions:1.The expression of Eps8was significantly increased in AML patients compared with the healthy volunteers(t=3.055, P=0.006), and correlated with the AML patients achieved CR or not after one course of chemotherapy (P=0.021). The CR rate of high expression group was significantly lower than low expression group (P=0.024). Eps8was obviously high expressed in KG la cell line, indicating Eps8may play an important role in the pathogenesis of AML and KG1a maybe s good study model.2. Genes (such as AKT1, BCL2, CASP3, PIK3CA, TP53) which participant in the PI3K/Akt signaling pathway and small GTPase mediated signal transduction highly expressed in KG1a cells.3. Perifosine can significantly inhibite the proliferation, the clone formation and the ex更多还原