探討癌細胞在實驗性演化壓力下惡性子群及細胞外推力

Abstract

在過往研究演化生物學中，學者大多聚焦於透過物種間相似的型態和基因體去判斷其為親緣較為接近的族群，然而對於何種因素主要去驅使演化的進展與方向則較少被深入探究。腫瘤的形成已被報導過是由上代細胞 (ancestral cell) 經一系列的基因突變，最終帶有利於生存性狀的子群體得以擴大並發展成為腫瘤主體。在此篇論文中，我們連續地採用球形成試驗 (sphere forming assay) 和體外侵略試驗 (invasion assay) 作為探究癌細胞演化的策略。 球形成試驗過去常被使用於識別癌幹細胞，我們將HCT116細胞株形成球體後再打散成一般貼盤細胞型態，以此流程連續八回合後發現篩選出的子群體具有較強的體外轉移性和體內腫瘤形成能力。除此之外，與未經篩選之族群相比，DPEP1被視為在經篩選群體間最高表現的基因其中之一，而當DPEP1表現被抑制，不論球形成能力、體外轉移能力和體內腫瘤形成能力皆有顯著性地下降。簡單來說我們成功透過此策略從人類大腸癌細胞HCT116中篩選出較為惡性的子群體並將DPEP1視為大腸癌細胞的診斷標記。 此外，CL系列和A549細胞株透過連續性體外侵略試驗成功分離出較具侵略性的群體，且侵略性高的群體 (例如：CL1-5和A549-12) 之紡錘絲有被延長的現象。相比於由同個病人體內分離出的原發性大腸癌細胞 (SW480)，繼發性大腸癌細胞 (SW620) 也同樣被證明具有較高的Aspect ratio，但透過認證更多經體內或體外侵略試驗篩選的細胞株，我們發現紡錘絲延長的特徵並非普遍存在於任何較具侵略性的細胞株。我們的結果暗示紡錘體本身存在很高的適應性，而為形成穩定的紡錘體延長現象，CL1-5可能有超過一種以上調控機制同時喪失功能。為了觀測癌細胞實際在人類組織內分裂的過程，我們製備兼具黏著性與伸縮性的水凝膠用以模擬人類細胞外間質，並發現CL1-5細胞在水凝膠中分裂時會向細胞外方向產生較強的推力，同時我們也辨認出一些微管 (microtubule)、星狀 (astral) 或皮層 (cortex) 調整因子可能會去主導紡錘絲縮短或產生由紡錘絲向細胞外推力。為了要全盤瞭解紡錘體力學演化機制，許多技術性問題與進階的實驗有待解決及進行。

In traditional evolutionary studies, researchers primarily focused on phenotype and genotype similarity to define categories of biological species. However, what drives evolutionary progression has been less investigated. It has been proposed that tumor is induced by a series of genomic mutations on an ancestral cell, and eventually the subpopulation that possesses properties stemmed from beneficial mutations develops into a tumor mass. In the present thesis, we adopted repeated sphere forming assay and continuously in vitro invasion assay as strategies to investigate cancer evolution in the laboratory. Sphere forming assay was commonly used to isolate cancer stem cells. After eight rounds of sphere forming and re-plating using HCT116, the resulting subpopulation exhibited higher metastatic capacity in vitro and increased tumorigenicity in vivo. Moreover, DPEP1 was identified as one of the most upregulated genes in the selected subpopulations compared with parental. Sphere forming capability, in vitro metastatic ability and in vivo tumorigenicity were markedly declined when knockdown of DPEP1. In sum, we have successfully isolated a malignant HCT116 subpopulation using our strategy, and DPEP1 might serve as prognostic biomarker of CRC. In addition, through repeated selection in vitro invasion assay, metastatic clones have also been isolated in CL and A549 cells. It has demonstrated that spindles were lengthened in both metastatic clones (e.g. CL1-5 and A549-12). Human secondary CRC (SW620) also exhibited higher spindle aspect compared to primary tumor (SW480) dissected from same patient. According to preliminary observations in other cancer cell lines selected in vitro an in vivo, we found that a stable spindle lengthening was not universal for all metastatic lines tested. Our data suggest that spindles may possess intrinsic adaptability, and that there may be more than one spindle regulatory mechanisms disrupted for CL series to exhibit stable lengthening phenotype. In order to mimic the process of cancer cell diving in human extracellular matrix, viscoelastic hydrogel was prepared for encapsulated cell divisions. We demonstrate that dividing CL1-5 cells generated stronger extracellular outward force in hydrogel compared to the parental. We also screened for potential microtubule or astral/cortex regulators that might modulate the transmission of outward force from spindle to extracellular space. We envision that many technical problems and further experiments are needed to be solved and carried out for the understanding of spindle force evolution in cancer.