NON-GENETIC CELL HETEROGENEITY IN TUMOR EVOLUTION

描述(申请人提供)：如果正常体细胞可以在不改变基因组的情况下产生独特、稳定的细胞表型，如静止的干细胞、增殖的祖细胞和多种分化的细胞类型，那么为什么通常用基因突变来解释进展性肿瘤获得的一种恶性特征，如耐药、干细胞样自我更新、侵袭等？肿瘤细胞群体具有高度的异质性，这在一定程度上反映了正常组织由干细胞和更多分化细胞组成的空间。因此，肿瘤可能使用与正常组织相同的非遗传(“表观遗传”)调节机制来产生各种稳定的表型。事实上，后生动物的基因调控网络似乎是连线的，从而要么产生一个特征(基因的表达水平)在细胞群体中的逐渐随机分散，要么产生基因表达模式的不同、稳定的变体(吸引子状态)--正如各种离散细胞类型所概括的那样。因此，这里假设，这种表观遗传细胞群体的异质性可能产生足够持久的表型变异，从而有助于肿瘤进展中细胞的经典(基于突变的)进化。具体地说，在整个细胞培养中，多药耐药的快速、全种群出现和消失支持了表观遗传适应的想法。此外，导致耐药的MDR1的表达并不是一个孤立的特殊特征，而是成人干细胞的固有属性，这表明肿瘤实际上可能会在整个稳定的细胞程序之间进行表观遗传转换，这些程序可以进行选择。为了开始解决肿瘤进化中表观遗传动力学的这一新范式，该建议的具体目标1将是量化多药耐药基因mdr1在HL60细胞中表达的群体分散动力学和表观遗传变异的稳定性。具体目标2是创造一个基于随机重组的“遗传随机数生成器”，它可以在外部初始化，并在单个细胞中可读。它将被用于特定的目标3，以确定体外耐药性的获得是通过随机突变体的克隆性扩张还是通过在多个细胞中独立地在表观遗传状态之间的随机转换来介导的。此外，在最后一个目标中，基因表达谱将揭示耐药的获得是否与干细胞表型的采用有关，如果这一过程反映了正常的表观遗传学在“预编程”细胞行为之间的转换，那么就会被预期，因为干细胞通常表达mdr1。对被忽视的动态表观遗传异质性现象的更好的、定量的认识，不仅将为后生动物细胞调控与肿瘤的研究开辟新的视角，还将有助于优化化疗计划，从而抑制耐药的选择，并更好地靶向肿瘤干细胞。

DESCRIPTION (provided by applicant): If normal somatic cells can generate distinct, stable cell phenotypes, such as quiescent stem cells, proliferating progenitors and the large variety of differentiated cell types without altering the genome, then why is the acquisition of one malignant trait by progressing tumors, such as drug-resistance, stem cell-like self-renewal, invasion, etc., usually explained by genetic mutations? Tumor cell populations are highly heterogeneous which reflects to some extent normal tissue organization into compartments of stem cells and more differentiated cells. Thus, tumors may use the same non-genetic ("epi-genetic") regulatory mechanisms to generate a variety of stable phenotypes as do normal tissues. Gene regulatory networks ol metazoan in fact appear to be wired so as to produce either a gradual stochastic dispersion of a trait (expression level of a gene) across a cell population or the distinct, stable variants of gene expression patterns (attractor states) - as epitomized by the various discrete cell types. Therefore, it is here hypothesized that this epigenetic cell population heterogeneity may produce phenotypic variants persistent enough to be selected for and hence, contribute to the classical (mutation-based) evolution of cells in tumor progression. Specifically, the rapid, population-wide appearance and disappearance of multi-drug-resistance in entire cell cultures support the idea of epigenetic adaptation. Moreover, expression of MDR1 that confers drug resistance is not an isolated idiosyncratic trait but an intrinsic property of adult stem cells, suggesting that tumors may in fact epigenetically switch between entire stable cellular programs that can be subjected to selection. To begin to address this new paradigm of epigenetic dynamics in tumor evolution, the Specific Aim 1 of this proposal will be to quantitate the dynamics of population dispersion in the expression of the multi-drug resistance gene, MDR1, in HL60 cells and the stability of epigenetic variants. Specific Aim 2 is to create a "genetic random-number generator" based on random recombination that can be externally initialized and is readable in single cells. It will be used in Specific Aim 3 to determine whether acquisition of drug resistance in vitro is mediated by clonal expansion of a random mutant or by a stochastic transitions between epigenetic states independently in multiple cells. In addition, gene expression profiling in this last Aim will reveal whether acquisition of drug resistance is linked to adoption of a stem cell phenotype, as would be expected if this process mirrors normal epigenetic switching between "preprogrammed" cellular behaviors, since stem cells normally express MDR1. A better, quantitative knowledge of the neglected phenomenon of dynamic epigenetic heterogeneity not only will open a new perspective to metazoan cell regulation and cancer, but also could help optimize scheduling of chemotherapy so as to suppress selection of resistance and better target tumor stem cells.