探討纖維母細胞生長因子接受器在頭頸癌遷移中的交互作用

Abstract

纖維母細胞生長因子接受器（Fibroblast growth factor receptor, FGFR）是一種穿膜且帶有酪胺酸激酶活化功能的蛋白質。它會活化MAPK、PI3K-AKT、PLC-Ca2+等下游訊息傳遞路徑，在細胞的生命週期及分化扮演相當重要的角色。FGFR的突變也與許多骨骼及軟骨發育的疾病相關。許多研究指出癌症的發展與FGFR的調控異常有關，如：非小細胞肺癌、子宮頸癌、胃癌及頭頸癌。造成頭頸癌惡化的主要原因為復發轉移，與其他癌症相比之下，頭頸癌的預後明顯較差。FGFR在頭頸癌轉移過程中扮演主要的角色，能啟動上皮細胞間質的轉化（EMT）機制。因此在頭頸癌的治療策略中抑制FGFR啟動EMT是一個具發展價值的策略。 近年來發展出許多針對FGFR所設計的小分子抑制劑，但在臨床試驗上的效果有限。在我們研究頭頸癌與FGFR抑制劑BGJ398及AZD4547的過程中，發現在血清的培養基中無法抑制FGFR的下游訊息傳遞。進一步的釐清之下發現，其中EGF與 FGF會互相補償對方下游的訊息傳遞。而此情形會在使用shRNA抑制FGFR或是EGFR後消失，因此我們猜測FGFR及EGFR之間有可能存在直接的交互作用。另一方面，在過度表達FGFR3的過程中發現會抑制FGFR2的下游訊息傳遞。因此在FGFR2及FGFR3之間可能也存在著交互作用。我們的結果證實了在分子層面的治療上無法利用全面性的抑制來阻斷FGFR2的下游訊息傳遞。因此，在了解FGFR的動態訊息傳遞後，可以解釋FGFR的抑制劑在臨床試驗上所存在的問題。 綜合以上問題，我們希望使用CRISPR基因編輯技術以及螢光交叉相關光譜（Fluorescence Cross-Correlation Spectroscopy, FCCS）來釐清這些膜蛋白的交互作用。我們透過CRISPR將感興趣之FGFR2、FGFR3以及EGFR進行編輯，並標記上不同顏色的螢光蛋白。同時透過FCCS觀察兩種膜蛋白在細胞上是否具有相關聯性，此外由於標記內生性蛋白可以即時定量出不同時間點細胞膜上的表現量。在這個平台上可以排除其他研究使用過度表達所無法觀察到內生性的蛋白表現。 在細胞遷移的實驗中，為了解FGFR2對細胞遷移行為的影響，細胞過度表達或是抑制FGFR2表現時，發現皆會影響細胞遷移時的方向性。市面上研究細胞趨向性的平台通常透過產生濃度梯度來進行實驗。但此類平台有維持時間上的限制。因此，我們利用微流道設計出一個可以穩定產生濃度梯度的，可以維持長時間穩定的濃度梯度。目前已初步的利用人類淋巴球癌細胞株對趨化因子12型的趨向性進行實驗並已驗證了平台的可行性。接著，針對頭頸癌細胞進行實驗的條件測試後，也將會利用這個平台來研究FGFR2如何影響頭頸癌細胞在遷移時的方向性。

Fibroblast growth factor receptors (FGFRs) family is a transmembrane receptor protein subfamily of receptor tyrosine kinases (RTKs). FGFRs play important roles in many biological processes such as proliferation and differentiation, by means of regulating various downstream signaling pathways including MAPK, PI3K-AKT, PLC-Ca2+. Some members of FGFRs are also related to pathological conditions. For instance, achondroplasia is caused by point mutation of FGFR3. Recent studies showed that the causation of cancers is related to abnormality of FGFRs regulation. Examples include, non-small cell lung cancer, cervix cancer, stomach cancer and head and neck cancer. Specially, major cause of head and neck cancer deterioration is metastasis. Patients with recurrent or metastatic head and neck cancer have poor prognosis comparing to other cancers. FGFRs play critical roles in metastasis of head and neck cancer by activating epithelial mesenchymal transition (EMT) process. Therefore, one potential therapeutic strategy treatment is inhibition of FGFRs-EMT pathway. Recently, many small molecule inhibitors targeting to FGFRs are being developed. But the clinical efficacy of these inhibitors are not satisfactory in comparison to the effective in-use EGFRs inhibitors. During our studies on FGFRs inhibitors BGJ398 and AZD4547, we noticed that the inhibitors failed to suppress FGFRs downstream signal transduction in vitro. In our further investigation, it was found that EGF and FGF mutually regulate their downstream signal transductions. So, inhibition solely on FGF could not stop the signaling cascade. However, the mutual regulation was not observed when shRNA was used to knockdown FGFR or EGFR. So we hypothesized the existence of physical interactions between FGFR and EGFR. On the other hand, by overexpressing FGFR3, we observed that the downstream signal transduction of FGFR2 is inhibited. Thus we also conjectured the existence of antagonism between FGFR2 and FGFR3. These hypotheses can explain the low performance of FGFR inhibitors in clinical tests. By fully understanding these signaling dynamics and interactions, one can not only explain the failure of clinical FGFR inhibitors, but to also improve the future therapeutic strategy on treating head and neck cancer. To further verify the above hypotheses, we use clustered regularly interspaced short palindromic repeats (CRISPR) and fluorescence correlation spectroscopy (FCS) to study the interactions of these membrane proteins. We edited FGFR2, FGFR3 and EGFR using CRISPR to label them with different fluorescent proteins. By studying the cross correlation of spectra from two different fluorescence proteins, the detailed interaction between any two receptor can be probed. On this platform, we can also observe some phenotypes that are normally not observed by overexpression. Moreover, the expression of proteins on plasma membrane can be quantified both spatially and temporally. On the other hand, in order to understand the effects of overexpression and inhibition of FGFR2 on cell migration, corresponding experiments are also conducted to quantify migratory properties such as speed and directionality. Most commercially available devices can only create an instantaneous concentration gradient, that will eventually reach equilibrium in certain time. Some other designs create constant gradient by flow, in which fluidic shear stress induces unwanted effects to the cells. To create a persisting constant gradient, we design a microfluidic chip that allows sustainable constant gradient creation without flow. As a proof of concept, C-X-C motif chemokine 12 is used to test SAS human oral cancer cells chemotaxis. After optimizing the microfluidic platform, we then investigate the effect of FGFR2 on directionality of cell migration of human head and neck cancer cells.