差異性蛋白質抗體庫之模式建立

Abstract

柯勒 (Köhler) 與麥爾斯坦 (Milstein) 利用細胞融合技術成功地將B細胞與骨髓瘤細胞融合成融合瘤細胞 (hybridoma)，透過不斷增殖與篩選，生產對特定抗原專一性的單株抗體 (monoclonal antibody)。2006年本實驗室吳裕仁博士提出抗體庫概念，利用綠竹筍水溶性的整個蛋白質體，產出對所有蛋白質的抗體庫，但因為此方法所需的時間、資源龐大且繁瑣，因此在2008年由蔡和成提出差異性蛋白質抗體庫，以兩個不同時期的總體蛋白質免疫小鼠，分別得到融合瘤細胞後，再利用流式細胞儀篩選出具差異性的細胞，所分泌的抗體可以區別上述兩個蛋白質體。本研究的主題，母源轉胚源時期 (maternal to zygotic transition, MZT) 是早期胚胎發育的第一個重要時期，對於胚胎在後續之發育成長扮演重要角色。我們選擇斑馬魚早期胚胎MZT前期的128-cell stage和後期的sphere-cell stage當作差異性的兩組樣本。先以實驗室建立的融合瘤細胞成功自model test-1實驗中分離出來，確定融合瘤細胞膜上有抗體分子。再以斑馬魚128細胞及sphere時期之抗原分別免疫，經過融合反應後，細胞以抗原標定 (antigen labeling) 接上螢光，再以流式細胞儀負篩選 (negative selection)，去除產生相同抗體的細胞，結果無法篩得具差異性的抗體。因此，設計了model test-2實驗，以vitellogenin、六種標準蛋白質、phytochelatin synthase、六種標準蛋白質的混合物，當作兩組不同時期的蛋白質體免疫小鼠，經融合反應後先進行正篩選 (positive selection)，以保留抗體分泌強的細胞，經培養後再做負篩選。正篩選確定挑出了分泌抗體較強的細胞，但負篩選還是無差異性。再次建立差異性抗體庫時，選擇斑馬魚胚胎卵黃較少的sphere時期和75%-epiboly時期進行篩選流程，但還是無法有效篩得到差異性細胞分群的結果。所以差異性抗體庫的技術仍有待改進的空間。

Köhler and Milstein reported a novel technique by fusing B cell with myeloma cell and producing hybridoma cell lines which secreted monoclonal antibodies. Starting from this idea, we proposed a concept of massive antibody production by immunizing the total water-soluble proteins from an organism to yield antibody library against the whole proteome rather than immunizing single antigen to obtain single mAb. However, this approach consumed time, and needed huge human and material resources. Subsequently, Tsai proposed an alternative approach for differential antibody library, which focused on the different antigens of two comparable proteomes. After the immunization and cell fusion, hybridoma cells were screened using flow cytometer by the differences of the proteins samples. To validate this new approach, we focused on the maternal to zygotic transition (MZT) which is an essential period in early embryonic development. We took the earlier 128-cell stage and the sphere-cell stage of zebrafish embryo development as the two periods for comparison. In a preliminary test, an established hybridoma line was isolated successfully using this process. Then we immunized the embryonic total proteins in mice. After cell fusion, flow cytometer sorted out cells which produced antibody recognizing the fluorescent-labeled antigens shared by the two proteomes (negative screening). The remaining cells might produce antibodies which could bind to the differential proteins between the two proteomes. However, we can not harvest cell line producing useful antibody. Owing to this result, we then set up model test-2 using two sets of defined protein mixture as the antigens: (1) vitellogenin and low molecular weight markers, and (2) phytochelatin synthase and low molecular weight markers. After the cell fusion, positive selection was performed first to pick the high antibodies-secreting cell lines out, and then negative screening to remove the common cell lines subsequently. Results revealed that positive selection did identify antibodies-secreting cells, but we can not isolate differential cells in negative selection. Finally, we took zebrafish embryo proteins as the antigens to test its application in complex proteome, but still can not obtain the differential cell distribution. As a result, we should improve the whole preparation system.