通过对肿瘤细胞致瘤潜能及上皮—间充质转化的研究探讨肿瘤干细胞假说的本质
|
摘要
背景：
克隆演变学说是最早用于解释实体肿瘤起始和进展生物学且被大部分学者接受的学说,它主要以遗传变异的多样性为理论依据,认为所有的肿瘤细胞都具有生成新的肿瘤的潜能。随着对肿瘤研究的不断深入,科学家提出了肿瘤干细胞(cancer stem cells, CSCs)学说,认为肿瘤干细胞是肿瘤中极小部分具有自我更新、无限增殖及分化潜能的细胞,肿瘤之所以具有无限增殖能力,其根本在于肿瘤细胞中存在占极少数量的肿瘤干细胞。目前已从人白血病、乳腺癌、胶质瘤、肺癌、结直肠癌、前列腺癌,黑色素瘤及胰腺癌等14种常见肿瘤中分离、鉴定出CSCs。尽管目前大多数研究结果支持肿瘤起源于某类具有“干性”特征的肿瘤细胞,然而学术界对肿瘤干细胞学说还存在很大的争议。Shipitsin等通过分析基因表达谱发现尽管乳腺癌干细胞和普通乳腺癌细胞分别是更原始和更分化的细胞群体,且它们的基因表达差异可能存在预后相关性,但是这些细胞之间的遗传学差异支持克隆进化学说参与了肿瘤内的异质性。而Kelly等通过将小鼠来源的淋巴瘤和白血病细胞移植致组织兼容的小鼠中得出所有肿瘤细胞均具有高致瘤的结果,认为肿瘤的生长在于肿瘤细胞与其周围微环境的适应能力而不在于细胞是否具有干细胞特征。这些争议牵涉到传统的克隆演变学说,直接冲击了肿瘤干细胞假说的基础一致瘤性。这一系列的争议给肿瘤干细胞的进一步研究及对这一学说的应用前景提出新的挑战,然而,这些争议着实推动了肿瘤研究的进展。另一方面,Morel等发现通过EMT过程能将人永生化的人正常乳腺上皮细胞(HMLEs)诱导成乳腺癌干细胞。这些研究结果为EMT在肿瘤细胞的播散及持续生长中的作用提供了充分的证据,更重要的是,它们建立了EMT与肿瘤干细胞起源之间的联系。总之,肿瘤干细胞研究至今,仍存在不少争议；争议的关键在于其是否具有特有的致瘤性,而分化的肿瘤细胞却不再具有成瘤能力。其实,解决这一争议的方法在于证实单个肿瘤细胞能否最终发展成肿瘤。然而,因为技术上的原因,目前仍很难准确的将单个肿瘤细胞接种于体内模型进行研究。另一方面,如果这一学说是错误的,那么之前的研究为什么会得出支持肿瘤干细胞存在的证据?科学家是否忽视了产生这些证据背后的本质呢?此外,为什么会导致EMT具有产生肿瘤干细胞的能力?寻找这些问题的答案可能需要对肿瘤干细胞的致瘤性及EMT在介导肿瘤干细胞恶性生物学行为中的本质进行探讨。
目的
1)通过单细胞克隆扩增-成瘤法探讨不同乳腺癌细胞株中致瘤性细胞所占的比例。
2)探讨EMT与悬浮培养富集乳腺癌干细胞的关系。
3)探讨EMT在肿瘤干细胞特殊生物学行为中的作用。
4)分析导致肿瘤干细胞具有高致瘤性的潜在的机制。
5)乳腺癌临床样本中分析肿瘤干细胞表型与间充质表型的一致性。
材料与方法
1) CD44+/CD24-/low细胞的流式检测及分选：将需检测的细胞,制成单细胞悬液,以PBS调整细胞密度为每管1×106/100μl。设实验组和同型对照组。实验组加入CD44-FITC和CD24-PE标记的单克隆抗体,对照组加入相应的同型对照抗体,4℃避光温育,30min后加5mlPBS漂洗去除未结合的抗体,1小时内用流式细胞仪检测或者分选。
2)单细胞克隆扩增-成瘤法：将达到80%融合状态的贴壁培养细胞,经胰酶消化成单细胞状态,按有限稀释法,将细胞稀释成10cells/ml,混匀后将细胞悬液接种于96孔板,200μl/孔,之后在倒置光学显微镜下观察仅含1个细胞的孔并标记,接种完后将细胞置于培养箱中培养。待单个细胞长成适当大小的细胞集落后,按上述消化方法将孔里的细胞集落消化并将细胞悬液相应地接种于24孔板继续培养。待细胞生长至适当融合状态后,按上述方法将细胞转移至6孔板培养；相似地,细胞最后扩大培养于1Ocm×10cm培养皿或者T75培养瓶中。收集由单个细胞扩增出来的细胞,制成单细胞悬液并计数,用5ml PBS洗涤两次以去除残留血清,再将细胞重悬于含1:matrigel的无血清的培养基中,最后细胞以1×107cells/200μl的体积皮下接种于4周龄雌性免疫缺陷BALB/c裸鼠体内；对于雌激素受体阳性的细胞(包括T47D,MCF-7和BT474),受体小鼠同时在大腿肌注0.05mg17-β雌二醇,每周注射一次。每周两次触摸观察肿瘤生长情况。
3)3D培养技术检测单个肿瘤细胞的体外成瘤能力：水泥胶包埋96孔板的每小孔,利用有限稀释法将细胞制成10cells/ml,与含胶的培养液等比例混匀,100gl/孔接种于水凝胶包埋的孔里,显微镜下观察只含单个细胞的孔并标记；在37℃、5%CO2培养箱中静止培养。
4)乳腺癌微球体的培养：将需要进行悬浮球培养的细胞重悬于不含血清的DMEM/F12培养基,并添加20ng/ml EGF、20ng/ml bFGF、0.4%BSA和2%B27,以50,000cells/ml接种于六孔板(3ml/孔),在37℃、5%CO2培养箱中静止培养,每隔一天加1ml新鲜培养基。
5)EMT诱导因子共培养乳腺癌细胞：流式细胞术分选出非肿瘤干细胞亚群重悬于分别添加有50ng/ml IL-6,20ng/ml EGF和20ng/ml b-FGF、或者10ng/mlTGF-β1的相应培养基中,置于37℃,5%CO2培养箱培养。
6)MTT细胞增殖检测及阿霉素药物敏感性检测：细胞增殖检测：消化收集细胞,重悬于相应的培养基并计数,用有限稀释法将细胞悬液稀释成5×103cells/ml,吹打均匀后接种于96孔板,200μl/孔,置于37℃,5%CO2培养箱培养。在相应的96孔板上设调零孔,即加200μl/孔不含细胞的相应培养基。分别在培养12h、36h、60h、84h、108h及132h后每孔加入20ulMTT溶液(5mg/ml,即0.5%MTT),继续培养4h。终止培养,小心吸去孔内培养液。每孔加入150u1二甲基亚砜,置摇床上低速振荡10min,使结晶物充分溶解。在酶联免疫检测仪OD490nm处测量各孔的吸光值。
阿霉素药物敏感性检测：消化收集细胞,重悬于相应的培养基并计数,用有限稀释法将细胞悬液稀释成5×104cells/ml,吹打均匀后接种于96孔板,200μ1/孔,置于37℃,5%CO2培养箱培养。在相应的96孔板上设调零孔(培养基、MTT、二甲基亚砜)及对照孔(细胞、相同浓度的药物溶解介质、培养液、MTT、二甲基亚砜),培养12h待细胞贴壁后,检测孔每孔加10μ1阿霉素工作液至终浓度为10μg/ml,继续培养。分别在加药后4h、24h、48h、72h、96h及120h后每孔加入20ulMTT溶液(5mg/ml,即0.5%MTT),继续培养4h,同上述方法检测吸光值。
7)细胞照射：采用Varian2100C直线加速器6MV X线为放射源,源皮距照射,剂量率400cGy/min, SSD=100cm, PDD=80%,照射野100mm×100mm,按0,2,4,6,8Gy的剂量分别给予照射。
8)克隆形成实验及放射生物学参数及剂量存活曲线的拟合：将照射后的细胞置于37℃,5%C02的孵育箱进行培养,连续培养15-25天。终止培养后常规以PBS缓冲液洗涤细胞2次,去除培养液中的蛋白成分,每孔加入1.5m1100%甲醇,室温固定15分钟；PBS清洗两次,每孔加入1ml1%的结晶紫染色后,室温作用20分钟；流水缓慢洗去多余染液,晾干,在显微镜下计算细胞>50个的细胞克隆数,计算克隆形成率(克隆形成率=生成的克隆数/接种细胞数×100%)和存活分数(survival fraction, SF=受照射细胞的克隆形成率/对照组细胞克隆形成率×100%)。最后,对所得的存活分数的平均数进行分析,运用GraphPad Prism5.0软件进行单击多靶模型曲线拟合,并根据单击多靶模型求出D0、Dq及N,其中logN=Dq/DO。
9)细胞迁移能力检测：收集细胞,用5ml PBS洗涤3次,充分去除残留的血清,离心后将细胞重悬于含0.1%BSA的培养基(不含血清及细胞因子)中并计数,调整细胞密度为5×105cells/ml,吹打混匀后吸200μ1细胞悬液接种于transwell上室,并在下室加入500μ1完全培养基,置于37℃、5%C02的培养箱中培养。培养24小时后取出小室置于24孔板,加适量的PBS泡洗5次,然后用0.4%多聚甲醛固定20分钟,再用PBS泡洗2次,其后加适量的1%的结晶紫染色20分钟,最后用PBS洗涤去除未染的结晶紫并用棉签轻轻擦去上层未穿透的细胞,光学显微镜下观察并拍照,选择随机的5个视野计算小室底面的细胞。
10)制备慢病毒载体稳定过表达：niR-200c:以人基因组DNA为模板PCR扩增目的基因；表达载体酶切后回收线性化的载体片段；目的基因片段与线性化的载体片段经同源重组后转化E.coli感受态细胞；用菌落PCR鉴定转化子,阳性克隆送测序；测序无误的克隆进行质粒抽提；采用脂质体法将慢病毒包装系统中3种质粒按比例共转染293细胞,收集上清液,离心过滤后,得到慢病毒悬液。病毒滴度采用qPCR实验进行测定。将慢病毒和空病毒载体感染乳腺癌细胞。采用流式细胞仪分选出GFP(+)细胞进行扩大培养。Real time qPCR检测miR-200c表达水平。
11) Real time RT-PCR检测基因的表达水平：收集细胞,提取RNA,逆转录成cDNA,基于ABI Prism7500系列检测系统,使用ToYoBo公司的SYBR Green试剂盒进行实时荧光定量PCR。
12) Western blotting检测细胞中E-cadherin, Vimentin, CD44及CD24蛋白的表达。
13)细胞免疫荧光技术检测细胞中E-cadherin, Vimentin, CD44及CD24蛋白的表达。
14)免疫组化检测组织样本中E-cadherin, Vimentin, Fibronectin, CD44及CD24分子的表达。
15)统计方法：实验结果应用SPSS13.0软件,以两个或多个独立样本率比较的x2检验、单向方差分析(One-Way ANOVA)及Bonferroni多重比较方法进行统计分析,P<0.05为差异有统计学意义。
结果1)致瘤性细胞普遍地存在于乳腺癌细胞株中不管是CSCs群体还是non-CSCs群体,均呈现出高比例的克隆性细胞,形成的细胞克隆可以经不断扩大培养获得大量的肿瘤细胞。当达到1×107数量级时,将细胞重悬于含1:1Matrigel的无血清培养基并接种于免疫缺陷的BALB/c裸鼠体内。经这种克隆扩增法获得的细胞,不管它们来源于CSCs群体还是non-CSCs群体,均能在裸鼠体内形成肉眼可见的肿瘤。而对于永生化的正常人乳腺上皮细胞仅含少量的克隆性细胞,而且这些细胞扩增出来的细胞群体不能在免疫缺陷的小鼠体内形成任何肿瘤。
2)悬浮培养富集肿瘤干细胞与细胞的上皮-间充质转化有关分选出来的非乳腺癌干细胞经悬浮培养后同样能富集具有干细胞标记的细胞群体；悬浮培养获得的细胞具有EMT相关的基因表达模式；在高粘附培养板或者添加血清后细胞贴壁生长,细胞形态呈现间充质样外观；细胞免疫荧光及Western blot检测显示上皮标记E-cadherin表达下调,而间充质标记Vimentin的表达上调。
3)“肿瘤干细胞”是一群具有间充质表型的细胞亚群分选出来的非肿瘤干细胞在含有不同的EMT诱导因子的培养基中培养48小时后出现细胞形态由鹅卵石样向梭形或者星形样转变,这种现象在luminal型乳腺癌细胞中尤为明显。诱导培养10天后,在luminal型细胞中显著富集了CD44+细胞(在MCF7细胞中还显著富集了CD44+/CD24-/low细胞),在Basal型细胞株中显著富集了CD24-细胞,在CD44+的MCF-10A细胞中同样显著富集了CD24-细胞,这些细胞诱导产生CD44+或者CD24-细胞的结果与悬浮培养产生具有这些表型的细胞的结果一致。进一步检测发现这些诱导产生的CD44+或者CD24-细胞呈现出与EMT相关的基因表达模式及蛋白表达模式。
4)EMT并不赋予肿瘤细胞致瘤性而只是赋予肿瘤细胞其它恶性特征利用克隆扩增-成瘤法检测了经不同细胞因子诱导后产生的CD44+或者CD24-细胞的致瘤潜能。发现经EMT诱导后产生的这些细胞并没有表现出更高的克隆及成瘤能力。进一步检测这些细胞的其它生物学功能,发现经IL6和EGF+bFGF诱导产生的CD44+或CD24-细胞均具有显著增高的细胞增殖能力、阿霉素抵抗、辐射抵抗及侵袭能力。
5)细胞增殖能力的差异可能是导致CSCs的发现的主要原因当用含生长因子(40ng/mIEGF和40ng/mlbFGF)的基质胶重悬non-CSCs后接种于NOD/SCID小鼠体内时,non-CSCs能在相同的时间内形成更多的肿瘤,形成的肿瘤数甚至多于CSCs在相同时间里所形成的肿瘤数。
6)肿瘤细胞致瘤性并不依赖于间充质特性间充质型乳腺癌细胞株稳定表达miR-200c后,细胞形态发生明显的MET样改变,即从梭形的间充质形态向鹅卵石样上皮形态转化。定量RT-PCR结果显示基因呈现MET相关表达模式,即上皮表型相关基因E-cadherin表达上调,而间充质表型相关的基因(Vimentin、N-cadherin和Fibronectin)表达下调；干细胞相关的标记CD44基因表达也明显下调。流式检测显示稳定表达miR-200c组细胞具有干细胞标记的细胞比例显著减少,而非干细胞比例显著升高。Westernblot同样证实了上皮标记的上调,间充质标记的表达的抑制,同时伴随明显的CD44表达抑制。发生MET后明显减慢了细胞的生长速度,增加了细胞对阿霉素的药物敏感性及对辐射的敏感性,同时细胞的侵袭能力明显受抑制。然而MET后并没有影响克隆性及致瘤性细胞的比例。
7)乳腺癌临床样本中干细胞表型与间充质表型高度一致肿瘤干细胞阳性的样本中上皮型标记E-cadherin表达的阳性率为34.4%(11/32),而在非肿瘤干细胞样本组中E-cadherin表达的阳性率为75.0%(39/52),两组样本中E-cadherin表达的阳性率有显著统计学差异(x2=13.570,P=0.000)；另一方面,间充质标记Vimentin表达在肿瘤干细胞组的阳性率为71.9%(23/32);而其在非肿瘤干细胞组的表达阳性率仅为21.2%(11/52),两组中Vimentin的表达的阳性率同样有显著的统计学差异(x2=21.153,P=0.000)。我们还进一步研究了典型的间充质表型E-cadherin-/Vimentin+(甚至E-cadherin-/Vimentin+/Fibronectin+)在两组样本中的比例,发现这些表型在两组样本中的比例存在显著的统计学差异(E-cadherin-/Vimentin+:x2=17.376, P=0.000; E-cadherin-/Vimentin+/Fibronectin+:x2=14.114, P=0.000)。因此,在临床肿瘤样本中证实表达乳腺癌干细胞标记的样本同时表达间充质标记,从而支持了我们关于肿瘤干细胞是一群具有间充质样特性的细胞群体的结果。
结论
1.致瘤性细胞普遍地存在于乳腺癌细胞中；
2.悬浮培养富集肿瘤干细胞与细胞经历EMT有关；
3.肿瘤干细胞是一群具有间充质表型的细胞亚群；EMT并不赋予肿瘤细胞致瘤性而只是赋予肿瘤细胞其它恶性特征,即肿瘤细胞的致瘤性并不依赖于间充质特性；细胞增殖能力的差异可能是导致肿瘤干细胞被发现的主要原因；
4.乳腺癌临床样本中干细胞表型与间充质样表型一致。
BACKGROUND
There are currently two conflicting views that attempt to explain tumor formation. The classical stochastic model suggests that every cell within a tumor is a potential tumor-initiator, but that entry into the cell cycle is governed by a low probability of stochastic mutations. According to this model, it would be impossible to tell which cell initiated the tumor since each cell has an equal ability to be malignant. Cancer stem cell hypothesis have been proposed nearly for100years. However, the first conclusive evidence for cancer stem cells (CSCs) was published in1997in Nature Medicine. This new hypothesis postulates that cancer only arises from a small subpopulation of cells defined as cancer stem cells or tumor-initiating cells (TICs) which also are supposed to respond for tumor progression, metastasis, resistance to therapy, and subsequent tumor recurrence. Increasing evidence has been obtained, thus CSCs were considered as one of the most interesting discoveries. The existence of CSCs is based on important evidence that these cells have tumorigenic potential when transplanted into NOD/SCID or other highly immunocompromised mice. However, the existence of CSCs is still a subject of debate within medical research, because many studies have not been successful in discovering the similarities and differences between normal tissue stem cells and cancer stem cells. To explore whether one single cancer cell has ability to eventually develop into a tumor is critical to confirm the validity of this hypothesis. However, previous studies hardly use one-single-cell process to confirm this hypothesis. If cancer stem cell is only a wrong hypothesis, then why a large body of evidence was obtained in previous researches? We should look at the motives behind all this persuasion.
Purpose
1) To assess the frequencies of tumorigenic cells in breast cancer cells by clonal-expanded and tumor-forming analyses.
2) To explore the role of the epithelial-mesenchymal-transition (EMT) in the enrichment of CSCs in mammosphere culture system.
3) To explore the potential contribution of EMT to special biological behaviors of CSCs.
4) To explore the potential mechanism of the enhanced tumorigenicities of CSCs.
5) To evaluate the association of CSCs with cells with mesenchymal-like phenotype
in clinical samples.
Materials and Methods
1) Flow cytometric analysis Cells were trypsinized, suspended into single-cell mixtures, washed with phosphate buffered saline (PBS), and incubated on ice for30min with monoclonal antibodies specifc for human cell surface markers CD44-FITC (eBioscience, San Diego, CA, USA) or CD24-PE (eBioscience). In negative control experiments, cells were incubated with fuo-rescence-labeled isotype-matched pre-immune IgG instead. Cells were washed and analyzed using a fow cytometer (BD FACS Aria, San Jose, CA, USA)
2) One-single-cell clonal-expanded and tumor-forming analyses (OCTAs) One-single-cell suspension was seeded into96-well plates, and cultured in serum-containing medium. Wells containing only one cell were visually confirmed under a microscope and marked, and wells with no cells or more than one cell were excluded. After cell clones were formed, the clones were transferred to24-well plates and expanded in fresh serum-containing medium. Cells were similarly transferred into6-well plates when adherent cells spreaded over the bottom of well. Finally, cells were transferred to10cm x10cm Petri dish and further expanded. The total of1x107expanded clonal cells was then injected in the flank region of4-week old female BALA/C-nude mice or NOD/SCID mice.
3) Tumorigenic abilities of breast cancer cells by3D culture Coat prechilled culture surface with a thin layer of matrigel (BD Biosciences, San Jose, CA, USA) by slowly pipetting60-100μL of matrigel directly onto culture surface, spreading evenly with a pipette tip and incubating for30min at37℃to allow the Matrigel to gel. Cells were mixed with complete culture medium of5%matrigel and then seeded into precoated plates. Cells were cultured in37℃and5%CO2condition.
4) Mammosphere culture Single-cell suspensions of MCF-7cells were suspended at a density of50,000cells per milliliter in serum-free Dulbecco's modified Eagle's medium/F-12containing5μg/mL bovine insulin(Sigma),0.4%bovine serum albumin (BSA, Sigma),2%B27(Invitrogen),20ng/mL basic fibroblast growth factor (bFGF, Peprotech) and10ng/mL epidermal growth factor (EGF, Peprotech), and culture in37℃and5%CO2condition.
5) Breast cancer cells were cultured in media with EMT-induced cytokines For EMT induction, non-CSCs were cultured in media with50ng/ml IL-6,20ng/ml EGF and20ng/ml bFGF, or lOng/ml TGF-β1in37℃and5%CO2condition.
6) Assessment of proliferation and doxorubicin resistance Cell proliferation potential and the relative resistance to doxorubicin were evaluated by cell proliferation assays using MTT assays (sigma). Cells were plated at a concentration of1×103cells per well (for growth advantage assays) or1×104cells per well (for doxorubicin resistance) into96-well culture plates. For the cell proliferation potential assay,20μl MTT solution (0.5%MTT) was added at12h、36h、60h、84h、108h and132h. For the doxorubicinresistance assay,20ul of MTT solution was added to each well of the plate at4h,24h,48h,72h,96h and120h after treatment with doxorubicin at a fnal concentration of10μg/ml. After the addition of MTT solution, plates were incubated for4h. the media was then removed and150ul DMSO added to each well to dissolve formazon crystal. Absorbance was then measured at490nm using a microplate reader (SpectraMax M5, Sunnyvale, CA, USA).
7) Irradiation Cells were irradiated with using6-MV X-rays produced by a Varian2100C linear accelerator at Southern Medical University. Does rate was400cGy/min, and cells were irradiated from a vertical direction for the time required to generate a does curve of0,2,4,6,8Gy. Corresponding controls were sham irradiated.
8) Clonogenic assay For clonogenic assays, cells were trypsinized into single cell suspension, resuspended in medium supplemented with10%fetal calf serum, seeded corresponding number of cells in6-well plates, and then incubated for3hours before irradiation. Immediately following irradiation, the cells were incubated for 15-25days at37℃in a5%CO2environment to allow the colony formation. After that, the colonies were fixed with pure ethanol and stained with1%crystal violet. Colonies containing>50cells were counted as clonogenic survivors. Plating efficiency (PE)=colonies observed/number of cells plated, Surviving fraction (SF)=colonies counted/[cells seeded x (PE/100). Three independent experiments were performed, each in triplicate. Using GraphPad Prism5software, the average data were fitted into single-hit multitarget formula:S=1-(1-e-D/D0)N where S is the fraction of cells surviving a dose, Do called the "mean lethal dose", is the dose on the straight-line portion of the survival curve to decrease the survival to37%. The "quasi-threshold dose" or Dq, which is the intercept of the extrapolated high dose, was also calculated. N is referred to the extrapolation number which is a parameter to measure the width of shoulder of the survival curve.
9) Migration assay Transwell insert chambers with an8-μum porous membrane (Corning Costar, Cambridge, MA, USA) were used for the assay. Cells were washed three times with PBS and1×105cells were added to the top chamber in serum-free media. The bottom chamber was filed with complete media. Cells were incubated for24h at37℃in a5%CO2humidifed incubator. To quantify the number of invasive cells, cells on the top chamber were removed with a cotton-tipped swab, and migrated cells were fxed in methanol and stained with1%crystal violet. Five random felds were counted.
10) Construction of the pGIPZ-hsa-mir-200c plasmid The hsa-mir-200c sequence was amplified from genomic DNA of human by forward primer:"5'CAACAGAAGGCTCGAGGAAGTGTCCCCAGGGACT C3'" and reverse primer:"5'ATTCTGATCAGGATCCAACGCTCTCAGCTC AAGACG3'". The PCR products of332bp were reclaimed from agarose gel electrophoresis and cloned into lentivirus shuttle plasmid pGIPZ (Open Biosystems, Huntsville, AL, USA) between the enzyme sites Xho I and BamH I. All constructs were verified by sequencing. Lentivirus packaging followed the standard instruction. Titer of lentivirus was measured by qPCR experiment in293T cells.
11) Real time RT-PCR Cells were harvested, and RNA was extracted using Trizol following the manufacturer's protocol. One microgram of total RNA was reverse transcribed into cDNA using the SuperScript First-Strand Synthesis System. Real-time polymerase chain reactions (PCRs) using the SYBR Green PCR Master Mix were performed using an ABI PRISM7500Sequence Detection System. Data are shown after normalization to18S expression.
12) Western blot Primary antibodies included mouse anti-E-cadherin (1:5,000; BD Biosciences), mouse anti-vimentin (1:500;Clone V9, Dako, Glostrup, Denmark), and rabbit anti-CD44(1:5,000;GeneTex Inc., Irvine, CA, USA). Secondary antibodies included rabbit anti-mouse IgG-HRP (1:1,000;Santa Cruz Biotechnology, Santa Cruz, CA, USA) and goat anti-rabbit IgG-HRP (1:1,000;GE Healthcare, Chalfont St Giles, Uk). HRP-conjugated monoclonal mouse anti-GAPDH (kangchen, Shanghai, China) was used as an internal parameter. All antibodies were diluted with5%milk in PBS containing0.1%Tween-20(PBS-T) and incubated for either1h at room temperature or overnight at4℃. All Western blots were visualized with ECL Western blotting substrate (Pierce, Rockford, IL, USA).
13) Immunofuorescence A total of1×105cells per chamber were plated into Lab-Tek two-chamber slides overnight. The next day, when cells were50-70%confuent, they were washed with PBS twice, fxed in3%paraformaldehyde (Sigma) and permeabilized in0.1%Triton X-100(Sigma) in PBS buffer at4℃for30min. The cells were then washed3times with PBS and incubated with blocking solution (10%goat serum in PBS). The cells were then incubated with primary antibodies for anti-E-cadherin (BD Biosciences) or anti-vimentin V9(Clone V9, Dako) overnight at4℃. The cells were washed three times in PBS and incubated with the secondary antibody, goat anti-mouse-Alexa Fluor488(1:1,000; Molecular Probes, Invitrogen) in blocking buffer for1h at room temperature in the dark. Finally, the cells were washed three times in PBS and incubated with0.25mg/ml DAPI (Roche) for1min at room temperature in the dark. The slides were washed exten-sively with PBS and mounted with Fluoromount-G (Southern Biotech, Birmingham, AL, USA). All matched samples were photographed (controls and tests) using immunofuorescence microscope (Olympus BX51, Tokyo, Japan) with identical exposure times.
14) Immunohistochemistry The expressions of EMT-related markers (E-cadherin, vimentin and fibronectin) and BrCSC-related markers(CD44and CD24) were detected by immunohistochemistry.
15) Statistical Analysis In all experiments, differences among groups were analyzed by chi-square criterion or variance (ANOVA) method using SPSS (13.0) software. Bonferroni method was used for multiple comparisons. A probability level of0.05was chosen for statistical significance.
Results
1) Cells with tumorigenic potential are common in breast cancer cell lines CSCs and non-CSCs were isolated by flow cytometry based on CD44/CD24antigen marker profile from these6breast cancer cell lines, and then compared frequencies of tumorigenic cells between CSCs and non-CSCs based on OCTAs. CSCs and non-CSCs exhibited almost the same frequency of clonal and tumorigenic cells.
2) The enrichment of CD44+or CD24-cells in sphere culture system is associated with epithelial-mesenchymal transition
Mammosphere cells developed a gene expression pattern associated with EMT even within24h of mammosphere culture, including induction of mesenchymal markers, vimentin and Twist, along with the induction of CD44or CD24repression. Cells suspended in mammosphere media were plated on high attachment plates and allowed to grow as adherent cells. Morphological changes from a cobblestone to a spindle-like morphology, a classical marker of EMT induction, were seen even48h after exposure to mammosphere media. Consistently, the immunostaining also showed the repression of epithelial protein marks and induction of mesenchymal markers.
3) Cancer stem cells are a subpopulation of cells with mesenchymal-like properties
Non-CSCs were cultured in media with50ng/ml IL-6,20ng/ml EGF and20ng/ml bFGF, or10ng/ml TGF-β1in37℃and5%CO2condition. These cytokines, without exception, induced CD44+cells in luminal cancer lines and CD24-cells in basal cancer lines after10-days exposure, which was miraculously consistent with the enrichment of these cells in mammosphere culture. To test the resulting CD44+or CD24-cells were just a subpopulation of cells which underwent EMT in response to exposure of some EMT-inducing cytokines, CSCs (CD44+cells or CD24" cells) from cells10d-exposured by cytokines were isolated, and then gene expression patterns associated with EMT were compared with non-CSCs without any cytokine-exposure. CSCs exhibited a gene expression pattern that was consistent with EMT, compared with non-CSCs, including E-cadherin repression and the concomitant activation of mesenchymal markers, such as vimentin, N-cadherin, and fibronectin, which was accompanied by the induction of CD44or the repression of CD24. The expression changes of EMT-related markers were also confirmed by indirect immunofluorescence labeling.
4) EMT only confers other malignant characteristics but not tumorgenicities on cancer cells
CD44+or CD24-cells induced by cytokine exposure only exhibited similar frequencies of tumorigenic cells compared with cells without any exposure to cytokines. Nevertheless, CD44+or CD24-cells induced by cytokines, except for TGF-β1, displayed significantly enhanced proliferation potential (Figure2A), increased resistance to radiation (Figure2D), and reduced cell death after doxorubicin treatment compared with CD44-or CD24+cells, which was consistent with the enhanced malignant characteristics of breast cancer stem cells (BrCSCs) reported in previous studies. Additionally, the CD44+or CD24-cells induced by cytokines including TGF-β1exhibited increased invasive potential compared with CD44-or CD24+cells, as demonstrated by transwell assays.
5) Distinct speed of cell-proliferative expansion may lead to discovery of CSCs Isolated CSCs and non-CSCs were transplanted into the mammary fat pads of immunocompromised NOD/SCID mice based on limiting dilution analysis. Just as most previous reports, non-CSCs exhibited repressive tumorigenic potential compared with CSCs80days after injection. However, tumorigenic frequencies of non-CSCs approach to those of CSCs with prolonged time of observation. Non-CSCs injected with matrigel containing growth factors EGF and bFGF displayed miraculously enhanced tumorigenic potential compared with those injected with matrigel without growth factors, and more interestingly, non-CSCs injected with growth factors even produced more, at least with the same effect, tumors than CSCs.
6) Tumorigenicity is independent on mesenchymal-like feature in tumor cells
Three mesenchymal-like cell lines BT549, MDA-MB-231and MDA-MB-435S were induced into epithelial-like phenotype through mesenchymal-epithelial transition (MET) by transfecting MET-induced small noncoding regulatory RNAs miR-200c which has been established to be a potent MET inducer in mesenchymal-like cancer cells. Morphological changes consistent with the cells undergoing MET were seen in cells transfected with miR-200c. Quantitative real-time PCR analysis showed a gene expression pattern consistent with MET including induction of E-cadherin and repression of most mesenchymal markers. This induction of MET coincided with repression of CD44expression and thus led to generation of CD44-/CD24-/low cells, which was demonstrated by flow cytometry. Functionally, MET induced by overexpression of MiR-200c resulted in significantly slowed growth, reduced invasion, and increased susceptibility to radiation or doxorubicin treatment compared with control cells. Nevertheless, miR-200c did not affect the frequencies of clonal cells and tumorigenic cells, although miR-200c-treated cells exhibited slower growth clones and spend more time to form palpable tumors in immunocompromised mice than corresponding control cells.
7. CD44+/CD24" phenotype was identically associated with cells with mesenchymal-like phenotype in clinical samples
According to expression patterns of CD44and CD24, these breast carcinomas were subdivided into two main subgroups. Subgroup A was defined as samples with non-CSCs phenotype (CD44+/CD24+, CD44-/CD24+, or CD44-/CD24-/low phenotypes), and subgroup B was defined as samples with CSCs phenotype (cells with CD44+/CD24-/low phenotype). We observed that mesenchymal-like markers, expecially vimentin, were significantly more frequent in tumors with subgroup B, while epithelial-like marker E-cadherin expression in subgroup B was remarkably repressed compared with that in subgroup A.
Conclusions
1) Cells with tumorigenic potential are common in breast cancer cell lines.
2) The enrichment of CD44+or CD24-cells in sphere culture system is associated with epithelial-mesenchymal transition.
3) Cancer stem cells are a subpopulation of cells with mesenchymal-like properties.
4) EMT only confers other malignant characteristics but not tumorgenicities on cancer cells.
5) Distinct speed of cell-proliferative expansion may lead to discovery of CSCs.
6) Tumorigenicity is independent on mesenchymal-like feature in tumor cells.
7) CD44+/CD24-phenotype was identically associated with cells with mesenchymal-like phenotype in clinical samples.
克隆演变学说是最早用于解释实体肿瘤起始和进展生物学且被大部分学者接受的学说,它主要以遗传变异的多样性为理论依据,认为所有的肿瘤细胞都具有生成新的肿瘤的潜能。随着对肿瘤研究的不断深入,科学家提出了肿瘤干细胞(cancer stem cells, CSCs)学说,认为肿瘤干细胞是肿瘤中极小部分具有自我更新、无限增殖及分化潜能的细胞,肿瘤之所以具有无限增殖能力,其根本在于肿瘤细胞中存在占极少数量的肿瘤干细胞。目前已从人白血病、乳腺癌、胶质瘤、肺癌、结直肠癌、前列腺癌,黑色素瘤及胰腺癌等14种常见肿瘤中分离、鉴定出CSCs。尽管目前大多数研究结果支持肿瘤起源于某类具有“干性”特征的肿瘤细胞,然而学术界对肿瘤干细胞学说还存在很大的争议。Shipitsin等通过分析基因表达谱发现尽管乳腺癌干细胞和普通乳腺癌细胞分别是更原始和更分化的细胞群体,且它们的基因表达差异可能存在预后相关性,但是这些细胞之间的遗传学差异支持克隆进化学说参与了肿瘤内的异质性。而Kelly等通过将小鼠来源的淋巴瘤和白血病细胞移植致组织兼容的小鼠中得出所有肿瘤细胞均具有高致瘤的结果,认为肿瘤的生长在于肿瘤细胞与其周围微环境的适应能力而不在于细胞是否具有干细胞特征。这些争议牵涉到传统的克隆演变学说,直接冲击了肿瘤干细胞假说的基础一致瘤性。这一系列的争议给肿瘤干细胞的进一步研究及对这一学说的应用前景提出新的挑战,然而,这些争议着实推动了肿瘤研究的进展。另一方面,Morel等发现通过EMT过程能将人永生化的人正常乳腺上皮细胞(HMLEs)诱导成乳腺癌干细胞。这些研究结果为EMT在肿瘤细胞的播散及持续生长中的作用提供了充分的证据,更重要的是,它们建立了EMT与肿瘤干细胞起源之间的联系。总之,肿瘤干细胞研究至今,仍存在不少争议；争议的关键在于其是否具有特有的致瘤性,而分化的肿瘤细胞却不再具有成瘤能力。其实,解决这一争议的方法在于证实单个肿瘤细胞能否最终发展成肿瘤。然而,因为技术上的原因,目前仍很难准确的将单个肿瘤细胞接种于体内模型进行研究。另一方面,如果这一学说是错误的,那么之前的研究为什么会得出支持肿瘤干细胞存在的证据?科学家是否忽视了产生这些证据背后的本质呢?此外,为什么会导致EMT具有产生肿瘤干细胞的能力?寻找这些问题的答案可能需要对肿瘤干细胞的致瘤性及EMT在介导肿瘤干细胞恶性生物学行为中的本质进行探讨。
目的
1)通过单细胞克隆扩增-成瘤法探讨不同乳腺癌细胞株中致瘤性细胞所占的比例。
2)探讨EMT与悬浮培养富集乳腺癌干细胞的关系。
3)探讨EMT在肿瘤干细胞特殊生物学行为中的作用。
4)分析导致肿瘤干细胞具有高致瘤性的潜在的机制。
5)乳腺癌临床样本中分析肿瘤干细胞表型与间充质表型的一致性。
材料与方法
1) CD44+/CD24-/low细胞的流式检测及分选：将需检测的细胞,制成单细胞悬液,以PBS调整细胞密度为每管1×106/100μl。设实验组和同型对照组。实验组加入CD44-FITC和CD24-PE标记的单克隆抗体,对照组加入相应的同型对照抗体,4℃避光温育,30min后加5mlPBS漂洗去除未结合的抗体,1小时内用流式细胞仪检测或者分选。
2)单细胞克隆扩增-成瘤法：将达到80%融合状态的贴壁培养细胞,经胰酶消化成单细胞状态,按有限稀释法,将细胞稀释成10cells/ml,混匀后将细胞悬液接种于96孔板,200μl/孔,之后在倒置光学显微镜下观察仅含1个细胞的孔并标记,接种完后将细胞置于培养箱中培养。待单个细胞长成适当大小的细胞集落后,按上述消化方法将孔里的细胞集落消化并将细胞悬液相应地接种于24孔板继续培养。待细胞生长至适当融合状态后,按上述方法将细胞转移至6孔板培养；相似地,细胞最后扩大培养于1Ocm×10cm培养皿或者T75培养瓶中。收集由单个细胞扩增出来的细胞,制成单细胞悬液并计数,用5ml PBS洗涤两次以去除残留血清,再将细胞重悬于含1:matrigel的无血清的培养基中,最后细胞以1×107cells/200μl的体积皮下接种于4周龄雌性免疫缺陷BALB/c裸鼠体内；对于雌激素受体阳性的细胞(包括T47D,MCF-7和BT474),受体小鼠同时在大腿肌注0.05mg17-β雌二醇,每周注射一次。每周两次触摸观察肿瘤生长情况。
3)3D培养技术检测单个肿瘤细胞的体外成瘤能力：水泥胶包埋96孔板的每小孔,利用有限稀释法将细胞制成10cells/ml,与含胶的培养液等比例混匀,100gl/孔接种于水凝胶包埋的孔里,显微镜下观察只含单个细胞的孔并标记；在37℃、5%CO2培养箱中静止培养。
4)乳腺癌微球体的培养：将需要进行悬浮球培养的细胞重悬于不含血清的DMEM/F12培养基,并添加20ng/ml EGF、20ng/ml bFGF、0.4%BSA和2%B27,以50,000cells/ml接种于六孔板(3ml/孔),在37℃、5%CO2培养箱中静止培养,每隔一天加1ml新鲜培养基。
5)EMT诱导因子共培养乳腺癌细胞：流式细胞术分选出非肿瘤干细胞亚群重悬于分别添加有50ng/ml IL-6,20ng/ml EGF和20ng/ml b-FGF、或者10ng/mlTGF-β1的相应培养基中,置于37℃,5%CO2培养箱培养。
6)MTT细胞增殖检测及阿霉素药物敏感性检测：细胞增殖检测：消化收集细胞,重悬于相应的培养基并计数,用有限稀释法将细胞悬液稀释成5×103cells/ml,吹打均匀后接种于96孔板,200μl/孔,置于37℃,5%CO2培养箱培养。在相应的96孔板上设调零孔,即加200μl/孔不含细胞的相应培养基。分别在培养12h、36h、60h、84h、108h及132h后每孔加入20ulMTT溶液(5mg/ml,即0.5%MTT),继续培养4h。终止培养,小心吸去孔内培养液。每孔加入150u1二甲基亚砜,置摇床上低速振荡10min,使结晶物充分溶解。在酶联免疫检测仪OD490nm处测量各孔的吸光值。
阿霉素药物敏感性检测：消化收集细胞,重悬于相应的培养基并计数,用有限稀释法将细胞悬液稀释成5×104cells/ml,吹打均匀后接种于96孔板,200μ1/孔,置于37℃,5%CO2培养箱培养。在相应的96孔板上设调零孔(培养基、MTT、二甲基亚砜)及对照孔(细胞、相同浓度的药物溶解介质、培养液、MTT、二甲基亚砜),培养12h待细胞贴壁后,检测孔每孔加10μ1阿霉素工作液至终浓度为10μg/ml,继续培养。分别在加药后4h、24h、48h、72h、96h及120h后每孔加入20ulMTT溶液(5mg/ml,即0.5%MTT),继续培养4h,同上述方法检测吸光值。
7)细胞照射：采用Varian2100C直线加速器6MV X线为放射源,源皮距照射,剂量率400cGy/min, SSD=100cm, PDD=80%,照射野100mm×100mm,按0,2,4,6,8Gy的剂量分别给予照射。
8)克隆形成实验及放射生物学参数及剂量存活曲线的拟合：将照射后的细胞置于37℃,5%C02的孵育箱进行培养,连续培养15-25天。终止培养后常规以PBS缓冲液洗涤细胞2次,去除培养液中的蛋白成分,每孔加入1.5m1100%甲醇,室温固定15分钟；PBS清洗两次,每孔加入1ml1%的结晶紫染色后,室温作用20分钟；流水缓慢洗去多余染液,晾干,在显微镜下计算细胞>50个的细胞克隆数,计算克隆形成率(克隆形成率=生成的克隆数/接种细胞数×100%)和存活分数(survival fraction, SF=受照射细胞的克隆形成率/对照组细胞克隆形成率×100%)。最后,对所得的存活分数的平均数进行分析,运用GraphPad Prism5.0软件进行单击多靶模型曲线拟合,并根据单击多靶模型求出D0、Dq及N,其中logN=Dq/DO。
9)细胞迁移能力检测：收集细胞,用5ml PBS洗涤3次,充分去除残留的血清,离心后将细胞重悬于含0.1%BSA的培养基(不含血清及细胞因子)中并计数,调整细胞密度为5×105cells/ml,吹打混匀后吸200μ1细胞悬液接种于transwell上室,并在下室加入500μ1完全培养基,置于37℃、5%C02的培养箱中培养。培养24小时后取出小室置于24孔板,加适量的PBS泡洗5次,然后用0.4%多聚甲醛固定20分钟,再用PBS泡洗2次,其后加适量的1%的结晶紫染色20分钟,最后用PBS洗涤去除未染的结晶紫并用棉签轻轻擦去上层未穿透的细胞,光学显微镜下观察并拍照,选择随机的5个视野计算小室底面的细胞。
10)制备慢病毒载体稳定过表达：niR-200c:以人基因组DNA为模板PCR扩增目的基因；表达载体酶切后回收线性化的载体片段；目的基因片段与线性化的载体片段经同源重组后转化E.coli感受态细胞；用菌落PCR鉴定转化子,阳性克隆送测序；测序无误的克隆进行质粒抽提；采用脂质体法将慢病毒包装系统中3种质粒按比例共转染293细胞,收集上清液,离心过滤后,得到慢病毒悬液。病毒滴度采用qPCR实验进行测定。将慢病毒和空病毒载体感染乳腺癌细胞。采用流式细胞仪分选出GFP(+)细胞进行扩大培养。Real time qPCR检测miR-200c表达水平。
11) Real time RT-PCR检测基因的表达水平：收集细胞,提取RNA,逆转录成cDNA,基于ABI Prism7500系列检测系统,使用ToYoBo公司的SYBR Green试剂盒进行实时荧光定量PCR。
12) Western blotting检测细胞中E-cadherin, Vimentin, CD44及CD24蛋白的表达。
13)细胞免疫荧光技术检测细胞中E-cadherin, Vimentin, CD44及CD24蛋白的表达。
14)免疫组化检测组织样本中E-cadherin, Vimentin, Fibronectin, CD44及CD24分子的表达。
15)统计方法：实验结果应用SPSS13.0软件,以两个或多个独立样本率比较的x2检验、单向方差分析(One-Way ANOVA)及Bonferroni多重比较方法进行统计分析,P<0.05为差异有统计学意义。
结果1)致瘤性细胞普遍地存在于乳腺癌细胞株中不管是CSCs群体还是non-CSCs群体,均呈现出高比例的克隆性细胞,形成的细胞克隆可以经不断扩大培养获得大量的肿瘤细胞。当达到1×107数量级时,将细胞重悬于含1:1Matrigel的无血清培养基并接种于免疫缺陷的BALB/c裸鼠体内。经这种克隆扩增法获得的细胞,不管它们来源于CSCs群体还是non-CSCs群体,均能在裸鼠体内形成肉眼可见的肿瘤。而对于永生化的正常人乳腺上皮细胞仅含少量的克隆性细胞,而且这些细胞扩增出来的细胞群体不能在免疫缺陷的小鼠体内形成任何肿瘤。
2)悬浮培养富集肿瘤干细胞与细胞的上皮-间充质转化有关分选出来的非乳腺癌干细胞经悬浮培养后同样能富集具有干细胞标记的细胞群体；悬浮培养获得的细胞具有EMT相关的基因表达模式；在高粘附培养板或者添加血清后细胞贴壁生长,细胞形态呈现间充质样外观；细胞免疫荧光及Western blot检测显示上皮标记E-cadherin表达下调,而间充质标记Vimentin的表达上调。
3)“肿瘤干细胞”是一群具有间充质表型的细胞亚群分选出来的非肿瘤干细胞在含有不同的EMT诱导因子的培养基中培养48小时后出现细胞形态由鹅卵石样向梭形或者星形样转变,这种现象在luminal型乳腺癌细胞中尤为明显。诱导培养10天后,在luminal型细胞中显著富集了CD44+细胞(在MCF7细胞中还显著富集了CD44+/CD24-/low细胞),在Basal型细胞株中显著富集了CD24-细胞,在CD44+的MCF-10A细胞中同样显著富集了CD24-细胞,这些细胞诱导产生CD44+或者CD24-细胞的结果与悬浮培养产生具有这些表型的细胞的结果一致。进一步检测发现这些诱导产生的CD44+或者CD24-细胞呈现出与EMT相关的基因表达模式及蛋白表达模式。
4)EMT并不赋予肿瘤细胞致瘤性而只是赋予肿瘤细胞其它恶性特征利用克隆扩增-成瘤法检测了经不同细胞因子诱导后产生的CD44+或者CD24-细胞的致瘤潜能。发现经EMT诱导后产生的这些细胞并没有表现出更高的克隆及成瘤能力。进一步检测这些细胞的其它生物学功能,发现经IL6和EGF+bFGF诱导产生的CD44+或CD24-细胞均具有显著增高的细胞增殖能力、阿霉素抵抗、辐射抵抗及侵袭能力。
5)细胞增殖能力的差异可能是导致CSCs的发现的主要原因当用含生长因子(40ng/mIEGF和40ng/mlbFGF)的基质胶重悬non-CSCs后接种于NOD/SCID小鼠体内时,non-CSCs能在相同的时间内形成更多的肿瘤,形成的肿瘤数甚至多于CSCs在相同时间里所形成的肿瘤数。
6)肿瘤细胞致瘤性并不依赖于间充质特性间充质型乳腺癌细胞株稳定表达miR-200c后,细胞形态发生明显的MET样改变,即从梭形的间充质形态向鹅卵石样上皮形态转化。定量RT-PCR结果显示基因呈现MET相关表达模式,即上皮表型相关基因E-cadherin表达上调,而间充质表型相关的基因(Vimentin、N-cadherin和Fibronectin)表达下调；干细胞相关的标记CD44基因表达也明显下调。流式检测显示稳定表达miR-200c组细胞具有干细胞标记的细胞比例显著减少,而非干细胞比例显著升高。Westernblot同样证实了上皮标记的上调,间充质标记的表达的抑制,同时伴随明显的CD44表达抑制。发生MET后明显减慢了细胞的生长速度,增加了细胞对阿霉素的药物敏感性及对辐射的敏感性,同时细胞的侵袭能力明显受抑制。然而MET后并没有影响克隆性及致瘤性细胞的比例。
7)乳腺癌临床样本中干细胞表型与间充质表型高度一致肿瘤干细胞阳性的样本中上皮型标记E-cadherin表达的阳性率为34.4%(11/32),而在非肿瘤干细胞样本组中E-cadherin表达的阳性率为75.0%(39/52),两组样本中E-cadherin表达的阳性率有显著统计学差异(x2=13.570,P=0.000)；另一方面,间充质标记Vimentin表达在肿瘤干细胞组的阳性率为71.9%(23/32);而其在非肿瘤干细胞组的表达阳性率仅为21.2%(11/52),两组中Vimentin的表达的阳性率同样有显著的统计学差异(x2=21.153,P=0.000)。我们还进一步研究了典型的间充质表型E-cadherin-/Vimentin+(甚至E-cadherin-/Vimentin+/Fibronectin+)在两组样本中的比例,发现这些表型在两组样本中的比例存在显著的统计学差异(E-cadherin-/Vimentin+:x2=17.376, P=0.000; E-cadherin-/Vimentin+/Fibronectin+:x2=14.114, P=0.000)。因此,在临床肿瘤样本中证实表达乳腺癌干细胞标记的样本同时表达间充质标记,从而支持了我们关于肿瘤干细胞是一群具有间充质样特性的细胞群体的结果。
结论
1.致瘤性细胞普遍地存在于乳腺癌细胞中；
2.悬浮培养富集肿瘤干细胞与细胞经历EMT有关；
3.肿瘤干细胞是一群具有间充质表型的细胞亚群；EMT并不赋予肿瘤细胞致瘤性而只是赋予肿瘤细胞其它恶性特征,即肿瘤细胞的致瘤性并不依赖于间充质特性；细胞增殖能力的差异可能是导致肿瘤干细胞被发现的主要原因；
4.乳腺癌临床样本中干细胞表型与间充质样表型一致。
BACKGROUND
There are currently two conflicting views that attempt to explain tumor formation. The classical stochastic model suggests that every cell within a tumor is a potential tumor-initiator, but that entry into the cell cycle is governed by a low probability of stochastic mutations. According to this model, it would be impossible to tell which cell initiated the tumor since each cell has an equal ability to be malignant. Cancer stem cell hypothesis have been proposed nearly for100years. However, the first conclusive evidence for cancer stem cells (CSCs) was published in1997in Nature Medicine. This new hypothesis postulates that cancer only arises from a small subpopulation of cells defined as cancer stem cells or tumor-initiating cells (TICs) which also are supposed to respond for tumor progression, metastasis, resistance to therapy, and subsequent tumor recurrence. Increasing evidence has been obtained, thus CSCs were considered as one of the most interesting discoveries. The existence of CSCs is based on important evidence that these cells have tumorigenic potential when transplanted into NOD/SCID or other highly immunocompromised mice. However, the existence of CSCs is still a subject of debate within medical research, because many studies have not been successful in discovering the similarities and differences between normal tissue stem cells and cancer stem cells. To explore whether one single cancer cell has ability to eventually develop into a tumor is critical to confirm the validity of this hypothesis. However, previous studies hardly use one-single-cell process to confirm this hypothesis. If cancer stem cell is only a wrong hypothesis, then why a large body of evidence was obtained in previous researches? We should look at the motives behind all this persuasion.
Purpose
1) To assess the frequencies of tumorigenic cells in breast cancer cells by clonal-expanded and tumor-forming analyses.
2) To explore the role of the epithelial-mesenchymal-transition (EMT) in the enrichment of CSCs in mammosphere culture system.
3) To explore the potential contribution of EMT to special biological behaviors of CSCs.
4) To explore the potential mechanism of the enhanced tumorigenicities of CSCs.
5) To evaluate the association of CSCs with cells with mesenchymal-like phenotype
in clinical samples.
Materials and Methods
1) Flow cytometric analysis Cells were trypsinized, suspended into single-cell mixtures, washed with phosphate buffered saline (PBS), and incubated on ice for30min with monoclonal antibodies specifc for human cell surface markers CD44-FITC (eBioscience, San Diego, CA, USA) or CD24-PE (eBioscience). In negative control experiments, cells were incubated with fuo-rescence-labeled isotype-matched pre-immune IgG instead. Cells were washed and analyzed using a fow cytometer (BD FACS Aria, San Jose, CA, USA)
2) One-single-cell clonal-expanded and tumor-forming analyses (OCTAs) One-single-cell suspension was seeded into96-well plates, and cultured in serum-containing medium. Wells containing only one cell were visually confirmed under a microscope and marked, and wells with no cells or more than one cell were excluded. After cell clones were formed, the clones were transferred to24-well plates and expanded in fresh serum-containing medium. Cells were similarly transferred into6-well plates when adherent cells spreaded over the bottom of well. Finally, cells were transferred to10cm x10cm Petri dish and further expanded. The total of1x107expanded clonal cells was then injected in the flank region of4-week old female BALA/C-nude mice or NOD/SCID mice.
3) Tumorigenic abilities of breast cancer cells by3D culture Coat prechilled culture surface with a thin layer of matrigel (BD Biosciences, San Jose, CA, USA) by slowly pipetting60-100μL of matrigel directly onto culture surface, spreading evenly with a pipette tip and incubating for30min at37℃to allow the Matrigel to gel. Cells were mixed with complete culture medium of5%matrigel and then seeded into precoated plates. Cells were cultured in37℃and5%CO2condition.
4) Mammosphere culture Single-cell suspensions of MCF-7cells were suspended at a density of50,000cells per milliliter in serum-free Dulbecco's modified Eagle's medium/F-12containing5μg/mL bovine insulin(Sigma),0.4%bovine serum albumin (BSA, Sigma),2%B27(Invitrogen),20ng/mL basic fibroblast growth factor (bFGF, Peprotech) and10ng/mL epidermal growth factor (EGF, Peprotech), and culture in37℃and5%CO2condition.
5) Breast cancer cells were cultured in media with EMT-induced cytokines For EMT induction, non-CSCs were cultured in media with50ng/ml IL-6,20ng/ml EGF and20ng/ml bFGF, or lOng/ml TGF-β1in37℃and5%CO2condition.
6) Assessment of proliferation and doxorubicin resistance Cell proliferation potential and the relative resistance to doxorubicin were evaluated by cell proliferation assays using MTT assays (sigma). Cells were plated at a concentration of1×103cells per well (for growth advantage assays) or1×104cells per well (for doxorubicin resistance) into96-well culture plates. For the cell proliferation potential assay,20μl MTT solution (0.5%MTT) was added at12h、36h、60h、84h、108h and132h. For the doxorubicinresistance assay,20ul of MTT solution was added to each well of the plate at4h,24h,48h,72h,96h and120h after treatment with doxorubicin at a fnal concentration of10μg/ml. After the addition of MTT solution, plates were incubated for4h. the media was then removed and150ul DMSO added to each well to dissolve formazon crystal. Absorbance was then measured at490nm using a microplate reader (SpectraMax M5, Sunnyvale, CA, USA).
7) Irradiation Cells were irradiated with using6-MV X-rays produced by a Varian2100C linear accelerator at Southern Medical University. Does rate was400cGy/min, and cells were irradiated from a vertical direction for the time required to generate a does curve of0,2,4,6,8Gy. Corresponding controls were sham irradiated.
8) Clonogenic assay For clonogenic assays, cells were trypsinized into single cell suspension, resuspended in medium supplemented with10%fetal calf serum, seeded corresponding number of cells in6-well plates, and then incubated for3hours before irradiation. Immediately following irradiation, the cells were incubated for 15-25days at37℃in a5%CO2environment to allow the colony formation. After that, the colonies were fixed with pure ethanol and stained with1%crystal violet. Colonies containing>50cells were counted as clonogenic survivors. Plating efficiency (PE)=colonies observed/number of cells plated, Surviving fraction (SF)=colonies counted/[cells seeded x (PE/100). Three independent experiments were performed, each in triplicate. Using GraphPad Prism5software, the average data were fitted into single-hit multitarget formula:S=1-(1-e-D/D0)N where S is the fraction of cells surviving a dose, Do called the "mean lethal dose", is the dose on the straight-line portion of the survival curve to decrease the survival to37%. The "quasi-threshold dose" or Dq, which is the intercept of the extrapolated high dose, was also calculated. N is referred to the extrapolation number which is a parameter to measure the width of shoulder of the survival curve.
9) Migration assay Transwell insert chambers with an8-μum porous membrane (Corning Costar, Cambridge, MA, USA) were used for the assay. Cells were washed three times with PBS and1×105cells were added to the top chamber in serum-free media. The bottom chamber was filed with complete media. Cells were incubated for24h at37℃in a5%CO2humidifed incubator. To quantify the number of invasive cells, cells on the top chamber were removed with a cotton-tipped swab, and migrated cells were fxed in methanol and stained with1%crystal violet. Five random felds were counted.
10) Construction of the pGIPZ-hsa-mir-200c plasmid The hsa-mir-200c sequence was amplified from genomic DNA of human by forward primer:"5'CAACAGAAGGCTCGAGGAAGTGTCCCCAGGGACT C3'" and reverse primer:"5'ATTCTGATCAGGATCCAACGCTCTCAGCTC AAGACG3'". The PCR products of332bp were reclaimed from agarose gel electrophoresis and cloned into lentivirus shuttle plasmid pGIPZ (Open Biosystems, Huntsville, AL, USA) between the enzyme sites Xho I and BamH I. All constructs were verified by sequencing. Lentivirus packaging followed the standard instruction. Titer of lentivirus was measured by qPCR experiment in293T cells.
11) Real time RT-PCR Cells were harvested, and RNA was extracted using Trizol following the manufacturer's protocol. One microgram of total RNA was reverse transcribed into cDNA using the SuperScript First-Strand Synthesis System. Real-time polymerase chain reactions (PCRs) using the SYBR Green PCR Master Mix were performed using an ABI PRISM7500Sequence Detection System. Data are shown after normalization to18S expression.
12) Western blot Primary antibodies included mouse anti-E-cadherin (1:5,000; BD Biosciences), mouse anti-vimentin (1:500;Clone V9, Dako, Glostrup, Denmark), and rabbit anti-CD44(1:5,000;GeneTex Inc., Irvine, CA, USA). Secondary antibodies included rabbit anti-mouse IgG-HRP (1:1,000;Santa Cruz Biotechnology, Santa Cruz, CA, USA) and goat anti-rabbit IgG-HRP (1:1,000;GE Healthcare, Chalfont St Giles, Uk). HRP-conjugated monoclonal mouse anti-GAPDH (kangchen, Shanghai, China) was used as an internal parameter. All antibodies were diluted with5%milk in PBS containing0.1%Tween-20(PBS-T) and incubated for either1h at room temperature or overnight at4℃. All Western blots were visualized with ECL Western blotting substrate (Pierce, Rockford, IL, USA).
13) Immunofuorescence A total of1×105cells per chamber were plated into Lab-Tek two-chamber slides overnight. The next day, when cells were50-70%confuent, they were washed with PBS twice, fxed in3%paraformaldehyde (Sigma) and permeabilized in0.1%Triton X-100(Sigma) in PBS buffer at4℃for30min. The cells were then washed3times with PBS and incubated with blocking solution (10%goat serum in PBS). The cells were then incubated with primary antibodies for anti-E-cadherin (BD Biosciences) or anti-vimentin V9(Clone V9, Dako) overnight at4℃. The cells were washed three times in PBS and incubated with the secondary antibody, goat anti-mouse-Alexa Fluor488(1:1,000; Molecular Probes, Invitrogen) in blocking buffer for1h at room temperature in the dark. Finally, the cells were washed three times in PBS and incubated with0.25mg/ml DAPI (Roche) for1min at room temperature in the dark. The slides were washed exten-sively with PBS and mounted with Fluoromount-G (Southern Biotech, Birmingham, AL, USA). All matched samples were photographed (controls and tests) using immunofuorescence microscope (Olympus BX51, Tokyo, Japan) with identical exposure times.
14) Immunohistochemistry The expressions of EMT-related markers (E-cadherin, vimentin and fibronectin) and BrCSC-related markers(CD44and CD24) were detected by immunohistochemistry.
15) Statistical Analysis In all experiments, differences among groups were analyzed by chi-square criterion or variance (ANOVA) method using SPSS (13.0) software. Bonferroni method was used for multiple comparisons. A probability level of0.05was chosen for statistical significance.
Results
1) Cells with tumorigenic potential are common in breast cancer cell lines CSCs and non-CSCs were isolated by flow cytometry based on CD44/CD24antigen marker profile from these6breast cancer cell lines, and then compared frequencies of tumorigenic cells between CSCs and non-CSCs based on OCTAs. CSCs and non-CSCs exhibited almost the same frequency of clonal and tumorigenic cells.
2) The enrichment of CD44+or CD24-cells in sphere culture system is associated with epithelial-mesenchymal transition
Mammosphere cells developed a gene expression pattern associated with EMT even within24h of mammosphere culture, including induction of mesenchymal markers, vimentin and Twist, along with the induction of CD44or CD24repression. Cells suspended in mammosphere media were plated on high attachment plates and allowed to grow as adherent cells. Morphological changes from a cobblestone to a spindle-like morphology, a classical marker of EMT induction, were seen even48h after exposure to mammosphere media. Consistently, the immunostaining also showed the repression of epithelial protein marks and induction of mesenchymal markers.
3) Cancer stem cells are a subpopulation of cells with mesenchymal-like properties
Non-CSCs were cultured in media with50ng/ml IL-6,20ng/ml EGF and20ng/ml bFGF, or10ng/ml TGF-β1in37℃and5%CO2condition. These cytokines, without exception, induced CD44+cells in luminal cancer lines and CD24-cells in basal cancer lines after10-days exposure, which was miraculously consistent with the enrichment of these cells in mammosphere culture. To test the resulting CD44+or CD24-cells were just a subpopulation of cells which underwent EMT in response to exposure of some EMT-inducing cytokines, CSCs (CD44+cells or CD24" cells) from cells10d-exposured by cytokines were isolated, and then gene expression patterns associated with EMT were compared with non-CSCs without any cytokine-exposure. CSCs exhibited a gene expression pattern that was consistent with EMT, compared with non-CSCs, including E-cadherin repression and the concomitant activation of mesenchymal markers, such as vimentin, N-cadherin, and fibronectin, which was accompanied by the induction of CD44or the repression of CD24. The expression changes of EMT-related markers were also confirmed by indirect immunofluorescence labeling.
4) EMT only confers other malignant characteristics but not tumorgenicities on cancer cells
CD44+or CD24-cells induced by cytokine exposure only exhibited similar frequencies of tumorigenic cells compared with cells without any exposure to cytokines. Nevertheless, CD44+or CD24-cells induced by cytokines, except for TGF-β1, displayed significantly enhanced proliferation potential (Figure2A), increased resistance to radiation (Figure2D), and reduced cell death after doxorubicin treatment compared with CD44-or CD24+cells, which was consistent with the enhanced malignant characteristics of breast cancer stem cells (BrCSCs) reported in previous studies. Additionally, the CD44+or CD24-cells induced by cytokines including TGF-β1exhibited increased invasive potential compared with CD44-or CD24+cells, as demonstrated by transwell assays.
5) Distinct speed of cell-proliferative expansion may lead to discovery of CSCs Isolated CSCs and non-CSCs were transplanted into the mammary fat pads of immunocompromised NOD/SCID mice based on limiting dilution analysis. Just as most previous reports, non-CSCs exhibited repressive tumorigenic potential compared with CSCs80days after injection. However, tumorigenic frequencies of non-CSCs approach to those of CSCs with prolonged time of observation. Non-CSCs injected with matrigel containing growth factors EGF and bFGF displayed miraculously enhanced tumorigenic potential compared with those injected with matrigel without growth factors, and more interestingly, non-CSCs injected with growth factors even produced more, at least with the same effect, tumors than CSCs.
6) Tumorigenicity is independent on mesenchymal-like feature in tumor cells
Three mesenchymal-like cell lines BT549, MDA-MB-231and MDA-MB-435S were induced into epithelial-like phenotype through mesenchymal-epithelial transition (MET) by transfecting MET-induced small noncoding regulatory RNAs miR-200c which has been established to be a potent MET inducer in mesenchymal-like cancer cells. Morphological changes consistent with the cells undergoing MET were seen in cells transfected with miR-200c. Quantitative real-time PCR analysis showed a gene expression pattern consistent with MET including induction of E-cadherin and repression of most mesenchymal markers. This induction of MET coincided with repression of CD44expression and thus led to generation of CD44-/CD24-/low cells, which was demonstrated by flow cytometry. Functionally, MET induced by overexpression of MiR-200c resulted in significantly slowed growth, reduced invasion, and increased susceptibility to radiation or doxorubicin treatment compared with control cells. Nevertheless, miR-200c did not affect the frequencies of clonal cells and tumorigenic cells, although miR-200c-treated cells exhibited slower growth clones and spend more time to form palpable tumors in immunocompromised mice than corresponding control cells.
7. CD44+/CD24" phenotype was identically associated with cells with mesenchymal-like phenotype in clinical samples
According to expression patterns of CD44and CD24, these breast carcinomas were subdivided into two main subgroups. Subgroup A was defined as samples with non-CSCs phenotype (CD44+/CD24+, CD44-/CD24+, or CD44-/CD24-/low phenotypes), and subgroup B was defined as samples with CSCs phenotype (cells with CD44+/CD24-/low phenotype). We observed that mesenchymal-like markers, expecially vimentin, were significantly more frequent in tumors with subgroup B, while epithelial-like marker E-cadherin expression in subgroup B was remarkably repressed compared with that in subgroup A.
Conclusions
1) Cells with tumorigenic potential are common in breast cancer cell lines.
2) The enrichment of CD44+or CD24-cells in sphere culture system is associated with epithelial-mesenchymal transition.
3) Cancer stem cells are a subpopulation of cells with mesenchymal-like properties.
4) EMT only confers other malignant characteristics but not tumorgenicities on cancer cells.
5) Distinct speed of cell-proliferative expansion may lead to discovery of CSCs.
6) Tumorigenicity is independent on mesenchymal-like feature in tumor cells.
7) CD44+/CD24-phenotype was identically associated with cells with mesenchymal-like phenotype in clinical samples.
引文
[1]Gao C, Furge K, Koeman J, et al. Chromosome instability, chromosome transcriptome, and clonal evolution of tumor cell populations[J]. Proc Natl Acad Sci U S A,2007,104(21):8995-9000.
[2]Reya T, Morrison S J, Clarke M F, et al. Stem cells, cancer, and cancer stem cells.[J]. Nature,2001,414(6859):105-111.
[3]Lapidot T, Sirard C, Vormoor J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice.[J]. Nature,1994,367(6464):645-648.
[4]Al-Hajj M, Wicha M S, Benito-Hernandez A, et al. Prospective identification of tumorigenic breast cancer cells.[J]. Proc Natl Acad Sci U S A,2003,100(7):3983-3988.
[5]Singh S K, Hawkins C, Clarke I D, et al. Identification of human brain tumour initiating cells. [J]. Nature,2004,432(7015):396-401.
[6]Kim C F, Jackson E L, Woolfenden A E, et al. Identification of bronchioalveolar stem cells in normal lung and lung cancer.[J]. Cell,2005,121(6):823-835.
[7]O'Brien C A, Pollett A, Gallinger S, et al. A human colon cancer cell capable of initiating tumour growth in immunodeficient mice.[J]. Nature,2007,445(7123):106-110.
[8]Patrawala L, Calhoun T, Schneider-Broussard R, et al. Highly purified CD44+ prostate cancer cells from xenograft human tumors are enriched in tumorigenic and metastatic progenitor cells.[J]. Oncogene,2006,25(12):1696-1708.
[9]Fang D, Nguyen T K, Leishear K, et al. A tumorigenic subpopulation with stem cell properties in melanomas.[J]. Cancer Res,2005,65(20):9328-9337.
[10]Li C, Heidt D G, Dalerba P, et al. Identification of pancreatic cancer stem cells.[J]. Cancer Res,2007,67(3):1030-1037.
[11]Potential identity of multi-potential cancer stem-like subpopulation after radiation of cultured brain glioma.[J]. BMC Neurosci,2008,9:15.
[12]Loebinger M R, Giangreco A, Groot K R, et al. Squamous cell cancers contain a side population of stem-like cells that are made chemosensitive by ABC transporter blockade.[J]. Br J Cancer,2008,98(2):380-387.
[13]Shipitsin M, Campbell L L, Argani P, et al. Molecular definition of breast tumor heterogeneity[J]. Cancer Cell,2007,11(3):259-273.
[14]Kelly P N, Dakic A, Adams J M, et al. Tumor growth need not be driven by rare cancer stem cells[J]. Science,2007,317(5836):337.
[15]Mani S A, Guo W, Liao M J, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells[J]. Cell,2008,133(4):704-715.
[16]Huber M A, Kraut N, Beug H. Molecular requirements for epithelial-mesenchymal transition during tumor progression[J]. Curr Opin Cell Biol,2005,17(5):548-558.
[17]Morel A P, Lievre M, Thomas C, et al. Generation of breast cancer stem cells through epithelial-mesenchymal transition[J]. PLoS One,2008,3(8):e2888.
[18]Al-Hajj M, Wicha M S, Benito-Hemandez A, et al. Prospective identification of tumorigenic breast cancer cells[J]. Proc Natl Acad Sci U S A,2003,100(7):3983-3988.
[19]Ginestier C, Hur M H, Charafe-Jauffret E, et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome[J]. Cell Stem Cell,2007,1(5):555-567.
[20]Patrawala L, Calhoun T, Schneider-Broussard R, et al. Side population is enriched in tumorigenic, stem-like cancer cells, whereas ABCG2+ and ABCG2- cancer cells are similarly tumorigenic[J]. Cancer Res,2005,65(14):6207-6219.
[21]Ponti D, Costa A, Zaffaroni N, et al. Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties[J]. Cancer Res,2005,65(13):5506-5511.
[22]Li X, Lewis M T, Huang J, et al. Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy[J]. J Natl Cancer Inst,2008,100(9):672-679.
[23]Zheng X, Shen G, Yang X, et al. Most C6 cells are cancer stem cells:evidence from clonal and population analyses[J], Cancer Res,2007,67(8):3691-3697.
[24]Dontu G, Abdallah W M, Foley J M, et al. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells[J]. Genes Dev,2003,17(10):1253-1270.
[25]Sullivan N J, Sasser A K, Axel A E, et al. Interleukin-6 induces an epithelial-mesenchymal transition phenotype in human breast cancer cells[J]. Oncogene,2009,28(33):2940-2947.
[26]Xu J, Lamouille S, Derynck R. TGF-beta-induced epithelial to mesenchymal transition[J]. Cell Res,2009,19(2):156-172.
[27]Ackland M L, Newgreen D F, Fridman M, et al. Epidermal growth factor-induced epithelio-mesenchymal transition in human breast carcinoma cells[J]. Lab Invest,2003,83(3):435-448.
[28]Hugo H, Ackland M L, Blick T, et al. Epithelial--mesenchymal and mesenchymal--epithelial transitions in carcinoma progression [J]. J Cell Physiol,2007,213(2):374-383.
[29]Park S M, Gaur A B, Lengyel E, et al. The miR-200 family determines the epithelial phenotype of cancer cells by targeting the E-cadherin repressors ZEB1 and ZEB2[J]. Genes Dev,2008,22(7):894-907.
[30]Kalluri R, Weinberg R A. The basics of epithelial-mesenchymal transition[J]. J Clin Invest,2009,119(6):1420-1428.
[31]Al-Hajj M, Wicha M S, Benito-Hernandez A, et al. Prospective identification of tumorigenic breast cancer cells[J]. Proc Natl Acad Sci U S A,2003,100(7):3983-3988.
[32]Docherty N G, O'Sullivan O E, Healy D A, et al. TGF-betal-induced EMT can occur independently of its proapoptotic effects and is aided by EGF receptor activation[J]. Am J Physiol Renal Physiol,2006,290(5):F1202-F1212.
[33]Blick T, Hugo H, Widodo E, et al. Epithelial mesenchymal transition traits in human breast cancer cell lines parallel the CD44(hi/)CD24 (lo/-) stem cell phenotype in human breast cancer[J]. J Mammary Gland Biol Neoplasia,2010,15(2):235-252.
[34]Shipitsin M, Campbell L L, Argani P, et al. Molecular definition of breast tumor heterogeneity [J]. Cancer Cell,2007,11(3):259-273.
[35]Sarrio D, Rodriguez-Pinilla S M, Hardisson D, et al. Epithelial-mesenchymal transition in breast cancer relates to the basal-like phenotype[J]. Cancer Res,2008,68(4):989-997.
[36]Sorlie T, Perou C M, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications[J]. Proc Natl Acad Sci U S A,2001,98(19):10869-10874.
[37]Rodriguez-Pinilla S M, Sarrio D, Honrado E, et al. Prognostic significance of basal-like phenotype and fascin expression in node-negative invasive breast carcinomas[J]. Clin Cancer Res,2006,12(5):1533-1539.
[38]Giatromanolaki A, Sivridis E, Fiska A, et al. The CD44+/CD24-phenotype relates to 'triple-negative' state and unfavorable prognosis in breast cancer patients[J]. Med Oncol,2011,28(3):745-752.
[39]Abraham B K, Fritz P, Mcclellan M, et al. Prevalence of CD44+/CD24-/low cells in breast cancer may not be associated with clinical outcome but may favor distant metastasis[J]. Clin Cancer Res,2005,11(3):1154-1159.
[2]Reya T, Morrison S J, Clarke M F, et al. Stem cells, cancer, and cancer stem cells.[J]. Nature,2001,414(6859):105-111.
[3]Lapidot T, Sirard C, Vormoor J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice.[J]. Nature,1994,367(6464):645-648.
[4]Al-Hajj M, Wicha M S, Benito-Hernandez A, et al. Prospective identification of tumorigenic breast cancer cells.[J]. Proc Natl Acad Sci U S A,2003,100(7):3983-3988.
[5]Singh S K, Hawkins C, Clarke I D, et al. Identification of human brain tumour initiating cells. [J]. Nature,2004,432(7015):396-401.
[6]Kim C F, Jackson E L, Woolfenden A E, et al. Identification of bronchioalveolar stem cells in normal lung and lung cancer.[J]. Cell,2005,121(6):823-835.
[7]O'Brien C A, Pollett A, Gallinger S, et al. A human colon cancer cell capable of initiating tumour growth in immunodeficient mice.[J]. Nature,2007,445(7123):106-110.
[8]Patrawala L, Calhoun T, Schneider-Broussard R, et al. Highly purified CD44+ prostate cancer cells from xenograft human tumors are enriched in tumorigenic and metastatic progenitor cells.[J]. Oncogene,2006,25(12):1696-1708.
[9]Fang D, Nguyen T K, Leishear K, et al. A tumorigenic subpopulation with stem cell properties in melanomas.[J]. Cancer Res,2005,65(20):9328-9337.
[10]Li C, Heidt D G, Dalerba P, et al. Identification of pancreatic cancer stem cells.[J]. Cancer Res,2007,67(3):1030-1037.
[11]Potential identity of multi-potential cancer stem-like subpopulation after radiation of cultured brain glioma.[J]. BMC Neurosci,2008,9:15.
[12]Loebinger M R, Giangreco A, Groot K R, et al. Squamous cell cancers contain a side population of stem-like cells that are made chemosensitive by ABC transporter blockade.[J]. Br J Cancer,2008,98(2):380-387.
[13]Shipitsin M, Campbell L L, Argani P, et al. Molecular definition of breast tumor heterogeneity[J]. Cancer Cell,2007,11(3):259-273.
[14]Kelly P N, Dakic A, Adams J M, et al. Tumor growth need not be driven by rare cancer stem cells[J]. Science,2007,317(5836):337.
[15]Mani S A, Guo W, Liao M J, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells[J]. Cell,2008,133(4):704-715.
[16]Huber M A, Kraut N, Beug H. Molecular requirements for epithelial-mesenchymal transition during tumor progression[J]. Curr Opin Cell Biol,2005,17(5):548-558.
[17]Morel A P, Lievre M, Thomas C, et al. Generation of breast cancer stem cells through epithelial-mesenchymal transition[J]. PLoS One,2008,3(8):e2888.
[18]Al-Hajj M, Wicha M S, Benito-Hemandez A, et al. Prospective identification of tumorigenic breast cancer cells[J]. Proc Natl Acad Sci U S A,2003,100(7):3983-3988.
[19]Ginestier C, Hur M H, Charafe-Jauffret E, et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome[J]. Cell Stem Cell,2007,1(5):555-567.
[20]Patrawala L, Calhoun T, Schneider-Broussard R, et al. Side population is enriched in tumorigenic, stem-like cancer cells, whereas ABCG2+ and ABCG2- cancer cells are similarly tumorigenic[J]. Cancer Res,2005,65(14):6207-6219.
[21]Ponti D, Costa A, Zaffaroni N, et al. Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties[J]. Cancer Res,2005,65(13):5506-5511.
[22]Li X, Lewis M T, Huang J, et al. Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy[J]. J Natl Cancer Inst,2008,100(9):672-679.
[23]Zheng X, Shen G, Yang X, et al. Most C6 cells are cancer stem cells:evidence from clonal and population analyses[J], Cancer Res,2007,67(8):3691-3697.
[24]Dontu G, Abdallah W M, Foley J M, et al. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells[J]. Genes Dev,2003,17(10):1253-1270.
[25]Sullivan N J, Sasser A K, Axel A E, et al. Interleukin-6 induces an epithelial-mesenchymal transition phenotype in human breast cancer cells[J]. Oncogene,2009,28(33):2940-2947.
[26]Xu J, Lamouille S, Derynck R. TGF-beta-induced epithelial to mesenchymal transition[J]. Cell Res,2009,19(2):156-172.
[27]Ackland M L, Newgreen D F, Fridman M, et al. Epidermal growth factor-induced epithelio-mesenchymal transition in human breast carcinoma cells[J]. Lab Invest,2003,83(3):435-448.
[28]Hugo H, Ackland M L, Blick T, et al. Epithelial--mesenchymal and mesenchymal--epithelial transitions in carcinoma progression [J]. J Cell Physiol,2007,213(2):374-383.
[29]Park S M, Gaur A B, Lengyel E, et al. The miR-200 family determines the epithelial phenotype of cancer cells by targeting the E-cadherin repressors ZEB1 and ZEB2[J]. Genes Dev,2008,22(7):894-907.
[30]Kalluri R, Weinberg R A. The basics of epithelial-mesenchymal transition[J]. J Clin Invest,2009,119(6):1420-1428.
[31]Al-Hajj M, Wicha M S, Benito-Hernandez A, et al. Prospective identification of tumorigenic breast cancer cells[J]. Proc Natl Acad Sci U S A,2003,100(7):3983-3988.
[32]Docherty N G, O'Sullivan O E, Healy D A, et al. TGF-betal-induced EMT can occur independently of its proapoptotic effects and is aided by EGF receptor activation[J]. Am J Physiol Renal Physiol,2006,290(5):F1202-F1212.
[33]Blick T, Hugo H, Widodo E, et al. Epithelial mesenchymal transition traits in human breast cancer cell lines parallel the CD44(hi/)CD24 (lo/-) stem cell phenotype in human breast cancer[J]. J Mammary Gland Biol Neoplasia,2010,15(2):235-252.
[34]Shipitsin M, Campbell L L, Argani P, et al. Molecular definition of breast tumor heterogeneity [J]. Cancer Cell,2007,11(3):259-273.
[35]Sarrio D, Rodriguez-Pinilla S M, Hardisson D, et al. Epithelial-mesenchymal transition in breast cancer relates to the basal-like phenotype[J]. Cancer Res,2008,68(4):989-997.
[36]Sorlie T, Perou C M, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications[J]. Proc Natl Acad Sci U S A,2001,98(19):10869-10874.
[37]Rodriguez-Pinilla S M, Sarrio D, Honrado E, et al. Prognostic significance of basal-like phenotype and fascin expression in node-negative invasive breast carcinomas[J]. Clin Cancer Res,2006,12(5):1533-1539.
[38]Giatromanolaki A, Sivridis E, Fiska A, et al. The CD44+/CD24-phenotype relates to 'triple-negative' state and unfavorable prognosis in breast cancer patients[J]. Med Oncol,2011,28(3):745-752.
[39]Abraham B K, Fritz P, Mcclellan M, et al. Prevalence of CD44+/CD24-/low cells in breast cancer may not be associated with clinical outcome but may favor distant metastasis[J]. Clin Cancer Res,2005,11(3):1154-1159.