Embryonic Transcription Factor Function in Human Colorectal Cancer Stem Cells

NANOGP8在干细胞中的功能：我们研究的一个主要问题是为什么位于15号染色体上的逆转录基因NANOGP8在人类癌症和白血病中被激活。它的编码区域不同于NANOG？S位于12号染色体上，由5个核苷酸组成，在305个氨基酸的蛋白质中只引起2个氨基酸的变化。如果NANOG丢失，NANOGP8是否能挽救干细胞尚不清楚。今年我们发现，当shRNA抑制NANOG和NANOGP8的表达，从而抑制人类结直肠癌细胞（CRC）形成球状体的能力时，NANOGP8的重新表达挽救了CRC细胞系形成球状体的能力，而NANOG的重新表达仅挽救了其中一种细胞系。由于细胞形成球体的能力是体外干细胞性的主要衡量指标，因此NANOGP8在这些癌症中挽救干细胞性的证明表明，NANOGP8能够取代NANOG作为核心转录因子。这些数据发表在一份手稿中，正在接受《致癌基因》杂志的最后审查。凋亡机制：今年的目标2是阐明等位基因特异性shRNA抑制NANOG或NANOGP8是否诱导细胞凋亡。慢病毒载体将shRNA传递到NANOG中，在3个人结直肠癌细胞的单层培养中，导致磷脂酰丝氨酸的出现，磷脂酰丝氨酸是细胞凋亡的标志。然而，由于NANOG或NANOGP8在单层培养中的表达较低，因此单层培养中的凋亡量较低。当在体外正常附着在底物上生长的人类结直肠癌细胞被置于悬浮培养中而没有附着在底物上的能力时，它们死于一种称为anoikis的细胞凋亡形式，这种细胞凋亡主要由Caspase 8的激活和细胞凋亡的外在途径驱动。由于悬浮培养中这两种NANOG的相对表达增加，我们研究了抑制其中一种NANOG是否会导致细胞凋亡，如果是的话，凋亡的途径是什么。我们发现，抑制NANOGP8和NANOG在较小程度上通过激活caspase 9和3的内在途径诱导细胞凋亡，从而减少肿瘤生长。通过使用Caspases 3和Caspases 9抑制剂以及NANOGP8和NANOG的过表达来阻断等位基因特异性shrna的作用，从而在悬浮培养中刺激结直肠癌细胞的生长，证实了这一观察结果。这些数据表明，诱导细胞凋亡可能是治疗的有用组成部分。与化疗的协同作用：nanog的抑制与化疗模型之间的相互作用已经得到解决。由于喜树碱是临床治疗结直肠癌的重要药物，我们评估了模型喜树碱Topotocan是否与慢病毒载体递送的shRNA具有协同作用。Topotecan是一种拓扑异构酶I抑制剂，可诱导DNA损伤，进而通过刺激细胞凋亡的外在和内在途径引起细胞毒性。拓扑替康以剂量依赖的方式诱导细胞死亡，并具有一致的动力学，在对拓扑替康敏感的结直肠癌细胞系中，通常在96小时内达到90%的杀伤率。有趣的是，两种nanog的相对转录水平在拓扑替康暴露后72小时增加。当将慢病毒等位基因特异性shRNA与拓扑替康一起加入单层培养的结直肠癌细胞系时，shRNA抑制NANOGP8导致所有3种结直肠癌细胞系的减少比单独使用拓扑替康更大。细胞凋亡的内源性和外源性途径均受到刺激。此外，阻断anoikis的浓度的caspase抑制剂并没有逆转拓扑替康和shrna引起的生长抑制。这表明慢病毒对NANOG或NANOGP8的shRNA可能与拓扑替康治疗协同作用。NEDD9的调控：虽然在存在凋亡应激的条件下，NANOGP8或NANOG的抑制可能与诱导凋亡有关，但尚不清楚是什么分子将nanog8或NANOG的抑制与诱导凋亡联系起来。为了评估这一点，我们对2个CRC细胞系进行了rna测序，这些细胞系分别在单层和悬液中培养，以及与NANOGP8过表达或与NANOGP8过表达的shRNA培养。通过NANOGP8基因改变调控的基因分析发现NEDD9是一个关键基因。NEDD9是粘附斑块中src和FAK的支架蛋白，可激活AKT，是一种抗凋亡刺激，因为它有助于FAK的组成性激活。RT-PCR证实，在CRC细胞系中，NANOG的shRNA可抑制NEDD9的表达，而NANOG的过表达可增加NEDD9的表达。此外，ChIP实验证实NANOG结合了NEDD9的启动子。随后的研究正在评估这是否是激活细胞凋亡的机制。我们目前正在评估NANOG和NEDD9在近400例原发性结肠癌中作为预后因素的作用。最后，人类NANOG的人类DNA结合位点尚未确定，我们打算在这个启动子中定义该位点。我们已经确定了启动子中响应NANOG的最小区域，并将在未来一年进一步缩小结合位点。这些研究结果目前正被整理成一份手稿，将于今年提交。体内转导：大约6个月前，我们开始了一项国防部拨款，该拨款将评估等位基因特异性慢病毒载体递送shRNA作为一种基因疗法的潜力，以抑制临床前异种移植模型中已建立的肿瘤的生长。初步实验表明，肿瘤内注射慢病毒shRNA诱导肿瘤生长适度的非特异性抑制。由于对阴性对照载体的抑制作用与特异性shrna一样大，因此慢病毒构建体诱导了先天免疫反应。通过绿色荧光蛋白（GFP）的荧光检测，肿瘤内注射慢病毒shRNA的反复评估未能揭示皮下肿瘤结节内肿瘤细胞的任何显著转导。GFP作为报告基因存在于我们所有的慢病毒shRNA结构中，并且存在于体外转导的球状体内45- 80%的细胞中。肿瘤细胞内无GFP表达。瘤内小结节（直径3mm）注射后5天。我们推测发生这种情况可能有4个原因：1)注射的病毒量过低，无法进行转导；2)间质压力阻止标准假型慢病毒颗粒与肿瘤细胞的结合；3)瘤内酸度降低病毒与肿瘤细胞膜的结合和融合，从而阻断转导；4)宿主基质抑制载体与肿瘤细胞的结合。为了评估假设1)，我们与FDA/CBER的J. Reiser博士进行了合作，他生产高质量/高滴度的慢病毒，用于选择性靶向和临床应用。为了评估假设2和3，我们与比利时的布雷克波特博士合作开发了一种更好、更有选择性的结合剂，可以克服间质压力和酸性。布鲁塞尔自由大学的副教授Karine brekpot是慢病毒先天免疫方面的权威。她从骆驼身上开发了一种天然的单链免疫球蛋白，可以识别CEA。她给了我们这种靶向质粒，我们刚刚开始将其整合到慢病毒载体中，在她和Reiser博士的指导下，我们开始在体外测试病毒结合和特异性，然后开始生产临床前测试。为了验证假设4)，我们将小鼠3T3细胞与人类CRC细胞进行共培养，以评估宿主基质是否也可能参与慢病毒转导的抑制。

NANOGP8 Function in Stemness: A major question in our research is why the retrogene NANOGP8 located on Chromosome 15 is activated in human carcinomas and leukemias. Its coding region differs from NANOG?s located on chromosome 12 by 5 nucleotides that cause only 2 amino acid changes in a protein of 305 amino acids. It has not been clear that NANOGP8 could rescue stemness if NANOG is lost. This year we have found that when shRNA inhibits the expression of both NANOG and NANOGP8 and as a consequence the ability of human colorectal carcinoma cells (CRC) to form spheroids, the re-expression of NANOGP8 rescues the ability of CRC lines to form spheroids whereas re-expression of NANOG rescues only one of the cell lines. Since the capacity of cells to form spheroids is a major in vitro measure of stemness, the demonstration that NANOGP8 rescues stemness in these carcinomas suggests that NANOGP8 is able to replace NANOG as a core transcription factor. These data are in a manuscript undergoing final review at the journal Oncogene. Mechanism of Apoptosis: A goal for this year under aim 2 was to elucidate whether allele-specific shRNA inhibition of NANOG or NANOGP8 induces apoptosis. Lentiviral vector delivered shRNA to either NANOG causes the appearance of phosphatidylserine, a marker for apoptosis, in monolayer culture in 3 human colorectal carcinoma cells. However, since NANOG or NANOGP8 expression is low in monolayer culture, the amount of apoptosis is low in monolayer cultures. When human colorectal carcinoma cells that normally grow attached to a substrate in vitro are placed in suspension culture without an ability to attach to a substrate, they die from a form of apoptosis termed anoikis that is driven mainly by the activation of Caspase 8 and the extrinsic pathway of apoptosis. Since the relative expression of both NANOGs increases in suspension culture, we investigated whether inhibition of either NANOG caused apoptosis and if so what pathway for apoptosis. We found that inhibition of NANOGP8 and to a lesser extent NANOG induces apoptosis through the intrinsic pathway by activation of Caspases 9 and 3 to reduce tumor growth. Confirmation of this observation was provided by stimulating growth of CRC cells in suspension culture by blocking the effects of allele-specific shRNAs by treatment with inhibitors of Caspases 3 and 9 as well as the overexpression of NANOGP8 and NANOG. These data suggest that the induction of apoptosis may be a useful component for therapy. Synergy with Chemotherapy: The interaction between inhibition of the NANOGs and a model of chemotherapy has been addressed. Since camptothecins are important agents for the clinical management of CRC, we have assessed whether the effects of Topotocan, a model camptothecin, might be synergistic with lentiviral vector delivered shRNA. Topotecan is a Topoisomerase I inhibitor that induces DNA damage that in turn causes cytotoxicity through stimulation of both the extrinsic and intrinsic pathways of apoptosis. Topotecan induces cell death in a dose dependent fashion and with consistent kinetics that will generally achieve a > 90% kill in CRC lines sensitive to topotecan in 96 hr. Interestingly, relative transcript levels of both NANOGs increase at 72 hr after topotecan exposure. When lentiviral allele-specific shRNAs are added to CRC lines in monolayer culture along with topotecan, shRNA inhibition of NANOGP8 led to greater reduction in all 3 CRC lines than with Topotecan alone. Both the intrinsic and extrinsic pathways of apoptosis were stimulated. Further, the caspase inhibitors at the concentrations that blocked anoikis did not reverse the inhibition of growth caused by topotecan and shRNAs. This suggests that lentiviral shRNA to NANOG or NANOGP8 may be synergistic with topotecan treatment. Regulation of NEDD9: While it is possible to link inhibition of NANOGP8 or NANOG with the induction of apoptosis under conditions in which there is apoptotic stress, it is not clear what molecules link inhibition of either NANOG with the induction of apoptosis. To assess this, RNA-seq has been done in 2 CRC lines cultured in monolayer and suspension as well as with shRNA to NANOGP8 or with overexpression with NANOGP8. Ingenuity Pathway Analysis of the genes modulated by alterations in NANOGP8 identified NEDD9 as a critical gene. NEDD9 is a scaffolding protein for src and FAK in adhesion plaques that activates AKT and is an anti-apoptotic stimulus since it contributes to constitutive activation of FAK. RT-PCR confirmed that NEDD9 expression was inhibited by shRNA to the NANOGs and increased by overexpression of either NANOG in the CRC lines tested. In addition, ChIP assays confirmed that NANOG binds the promoter of NEDD9. Subsequent research is assessing whether this is the mechanism by which apoptosis is activated. We are currently assessing the role of NANOG and NEDD9 as prognostic factors in a series of nearly 400 primary colon carcinomas. Finally, the human DNA binding site for human NANOG has not been identified and we intend to define that site within this promoter. We have identified a minimal region of the promoter that responds to NANOG and in the coming year we will further narrow the binding site. These results are currently being put into a manuscript that will be submitted this year. Transduction in Vivo: About 6 months ago we started a DOD grant that will assess the potential of allele-specific lentiviral vector delivered shRNA as a gene therapy to inhibit growth of established tumor in preclinical xenograft models. Initial experiments indicated that intratumoral injection of lentiviral shRNA induced a modest nonspecific inhibition of tumor growth. Since the inhibition was as great to the negative control vector as the specific shRNAs, the lentiviral constructs induced an innate immune response. Repeated assessment of intratumoral injections with the lentiviral shRNA failed to reveal any significant transduction of tumor cells within the subcutaneous tumor nodules as measured by the presence of Green Fluorescent Protein (GFP) fluorescence. GFP is included in all of our lentiviral shRNA constructs as a reporter and is present in 45- 80% of cells within spheroids that are transduced in vitro. There was no GFP expression within tumor cells 3 ? 5 days after intratumoral injection into small (3 mm diameter) nodules. We postulate that there may be 4 reasons why this may occur: 1) the amount of virus injected was too low for transduction, 2) interstitial pressure prevents binding of standard pseudotyped lentiviral particles to tumor cells, 3) intratumoral acidity decreases viral binding and fusion to tumor cell membranes so that transduction is blocked and 4) host stroma inhibits binding of vector to tumor cells. To assess postulate 1) we have developed a collaboration with Dr. J. Reiser at FDA/CBER who produces high quality/high titer lentivirus for selective targeting and use in clinical situations. To assess postulates 2 and 3 we have developed a collaboration with Dr. Breckpot in Belgium to develop a better and more selective binding agent that may overcome interstitial pressure and acidity. Associate Professor Karine Breckpot at the Brussels Vrje Universtat is a leading authority on innate immunity to lentivirus. She has developed a natural single chain immunoglobulin from llamas that recognizes CEA. She has given us this targeting plasmid that we are just beginning to incorporate into a lentiviral vector and with her and Dr. Reiser's guidance we are beginning to test viral binding and specificity in vitro before ramping up production for preclinical testing. To assess postulate 4) we will do co-cultures of mouse 3T3 cells with human CRC cells to assess whether host stroma also may participate in inhibition of lentiviral transduction.