血管內皮生長因子及其受器等之表現、純化及利用小角度散射進行結構分析

Abstract

血管內皮生長因子（vascular endothelial growth factors，VEGFs）可調控血管及淋巴管的形成和恆定，當血管內皮生長因子與受體結合時，會促使受體雙偶化（dimerization），活化細胞內受體酪胺酸激酶（tyrosine kinase）進行自體磷酸化，激發下游訊息分子磷酸化連鎖反應（phosphorylation cascade）。在哺乳動物體中的血管內皮生長因子共有五型：VEGF-A、B、C、D 和 P1GF，其受體有 VEGFR-1 （Flt1）、VEGFR-2（KDR/Flk1）和 VEGFR-3 （Flt4）三類，屬於第五型受體酪胺酸激酶（type V receptor tyrosine kinase, RTK5），其特色為細胞膜外含有有七個免疫球蛋白樣區域（immunoglobulin-like domains, Ig）。VEGF-C、-D 由內皮細胞分泌，主要可與 VEGF-3 結合，以調控淋巴管的新生和增生；當與 VEGFR-2 表現細胞作用時亦可調節血管新生與生成。近期研究發現 VEGF-C、-D 及 VEGFR-3 與腫瘤細胞的生長轉移息息相關。 本研究利用果蠅細胞表現系統表現多種內皮生長因子及其受體或突變型受體進行分析，含：VEGF-C、VEGF-D、VEGFR-3 細胞膜外含首三個和全七個 Ig 區域（VEGFR-3-D1~3、-D1~7）及三種突變型 mtVEGFR-3-D1~7：C1、C3、C1C3（與 VEGF-C 結合之可能重要胺基酸鏈更換為第一型所擁有之胺基酸鏈），探討血管內皮生長因子與受體之結合作用與結構研究。我們研究之初期並無任何 VEGF-C、-D 或 VEGFR-3 之結構研究報告，但最近 VEGF-C 和 VEGFR-2-D2 ~3 所形成之複合物及 VEGF-D 的晶體結構陸續被發表，目前為止VEGFR- 3 尚未有任何結構資訊的文獻報告。有鑑於此，我們以小角度X光散射（small-angle X-ray scattering, SAXS）來對 VEGFR-3 結構及與其複合體之結構進行分析，此方法以均質狀態（mono-disperse）之蛋白質溶液的分子結構為觀察目標，一般相信更能反應出蛋白之天然構型（native conformation），並利用酵素免疫分析法找出第三型受體與因子結合之重要胺基酸。 重組蛋白經由果蠅細胞表現後分泌至培養液，經濃縮、透析、親和層析及膠體層析後可得到高純度蛋白液，本實驗中可取得之各型受體皆高於10毫克/升培養液，另VEGF-C、-D 產量則約2~3毫克/升培養液。在小角度X光散射實驗中，我們利用 ATSAS 程式分析 VEGFR-3- D1~3、VEGFR-3-D1~7、VEGFR-3-D1~3/VEGF-C 三種蛋白分子數據，利用 DAMMIF 及 Situs 重建蛋白表面構造（ab-initio envelope）。依據 VEGFR-2 和/或 cKit 的晶體結構進行同源模擬，這些模擬與小角度X光散射實驗所得之表面構造進行比較，並且反計算出同源模擬之小角度X光散射曲線後與實際實驗數值比較，結果顯示同源模型與重建之表面構造符合。 酵素免疫分析結果顯示 VEGFR-3 和 VEGF-C 的結合力遠低於 VEGFR-1 和 VEGF-A 之結合力，這或許可說明之前複合物難以形成的原因；另外我們參考文獻，將 VEGFR-3 可能與 VEGF-C（或VEGF-A）結合的兩段區域置換為 VEGFR-1 序列，發現這樣的雜合受體（chimeric receptor）可增強其與 VEGF-A 的結合能力，而與 VEGF-C 之結合能力並未改變。根據上述結果我們擬設計同時可結合 VEGF-A 與 VEGF-C 之誘餌受體（decoy receptor），做為癌症治療之研究，並藉由酵素免疫分析來找出 VEGFR-3 與 VEGF-C 最佳結合條件，以利其複合物形成，做為未來結構研究所需，並進一步探討其晶體結構，直觀分析出第三型內皮生長因子受體重要的結構。

Vascular endothelial growth factors (VEGFs) regulate blood and lymphatic vessel development and homeostasis. The binding of VEGFs with their cognate receptors (VEGF receptors, VEGFRs) will induce the ligand-mediated receptor dimerization which will activate intracellular tyrosine kinase of receptors undergoing auto-phosphorylation and propagate signal-transduction cascade. In mammalians, there are five members of VEGFs (VEGF-A, B, C, D and PlGF) and three members of VEGFRs (VEGFR-1, Flt1; VEGFR-2, KDR/Flk1; and VEGFR-3, Flt4). VEGFRs belong to type V receptor tyrosine kinases which are characterized with seven immunoglobulin-like domains in their extraceullar domains. VEGF-C and VEGF-D were secreted by endothelial cells and can regulate lymphagiogenesis via binding with VEGFR-3, or regulate angiogenesis and vasculogenesis via binding with VEGFR-2. Recently studies found that VEGF-C, -D and VEGFR-3 are related with tumor growth and metastasis. In this study, recombinant proteins of VEGF-C, VEGF-D, the first three Ig-like domains of VEGFR-3 (VEGFR-3-D3), whole extracellular domain (VEGF-R3- D7) and three types mutant of whole extracellular domain were expressed by Drosophila Schneider 2 (S2) cell expression system for structural studies of ligands, receptors and ligand/receptor complex. When we started this project, no any structural information of VEGF-C, VEGF-D, or VEGFR-3 is available. However, recently the crystal structures of VEGF-C in complex with VEGFR-2-D2~D3 and VEGF-D have been published, but still no structural information of VEGFR-3 yet. For this thesis, we conducted small-angle X-ray scattering (SAXS) to investigate the structure of VEGFR-3 and its complex with VEGF-C or -D. By this method, protein structure of mono-disperse in solution is the observation target which believe can represent as the native conformation of proteins. We also tried to find the amino acids which were important for VEGFR-3 to bind VEGFs with ELISA. Recombinant proteins were purified from concentrated culture media of stable transfected S2 cells, through Ni-affinity and gel filtration chromatography. The yield of any types of VEGFR-3 domain is greater than 10 mg/liter culture media; for VEGF-C and -D is about 2-3 mg/liter culture media. However, only VEGF-C in complex with VEGFR-3-D3 has been successfully obtained. The SAXS data of VEGFR-3-D3, -D7, and VEGF-C in complex with VEGFR-D3 have been characterized by programs implant in ATSAS, ab-initio envelope of each protein was reconstructed by DAMMIF and Situs. Their homology models have been built according to crystal structures of VEGFR-2 or/and cKit and their models have been docked into average envelops and assessed. The results showed that the model of VEGFR-3-D3 can dock into SAXS envelope very well. According to ELISA data, in normal situation the effinity of VEGFR-3/VEGF-C was lower than VEGFR-1/VEGF-A. This may explain why so difficult to get the complex of VEGFR-3/VEGF-C. We have established the enzyme-linked immunosorbent assay (ELISA) for the study of VEGF ligand-receptor bindings. The result revealed that the binding affinity of VEGF-C with its coguate receptor VEGFR-3 is much lower than that of VEGF-A and VEGFR-1, and implied the difficulty to form VEGFR-3/VEGF-C complex. According to the crystal structure of VEGFR-2 (D2-D3)/VEGF-C complex and sequence alignment of VEGFRs, we constructured several chimeric VEGFR-3, by substitute C1 or/and C3 regions of VEGFR-3 with VEGFR-1 sequence. Interestingly, the chimeric VEGFR-3 receptor gains the binding affinity with VEGF-A, particularly for the C1C3 VEGFR-3 mutant, and seems do not lost their binding affinity with VEGF-C. Due to the important roles of VEGF-A and VEGF-C in turmor growth and metastasis, based on the above results, we will next design a decoy VEGF receptor which can bind both VEGF-A and VEGF-C. Mean while, according to ELISA, we will investigate the appropriate condition for the VEGF-C/VEGFR-3 complex formation, in order to form the ligand-receptor compex for structural studies by SAXS and/or X-ray crystallography in the near future.