發展針對胃癌標定分子具專一性的單株抗體

Abstract

胃癌是全世界發生率第四高的癌症，同時也是所有癌症造成的死亡率中第二高的。現今，在之前的一些研究指出胃癌是具有地區性的分布。高危險地區包括日本、中國、南韓、東歐、葡萄牙，以及部分中南美洲國家。根據美國癌症學會研究指出，胃癌之所以具有地區性分佈的可能原因是和當地具胃刺激性的飲食文化有關。 了解胃癌的臨床發展狀況對於所選擇的癌症治療及藥物治療策略相當重要。目前最常見用於癌症診斷的方式大都是一些侵入式的檢驗方法，如插入胃部的光纖維攝影機胃鏡、電腦斷層掃描及組織切片分析等方式。然而這些方式往往都需要侵入式的診斷方式、需要利用昂貴的儀器或是較複雜的採樣及分析步驟。因此有必要發展一個具專一性、準確性、快速且方便操作的胃癌診斷方法。 在各種檢驗樣品中，血液是最普遍且最容易取得的樣品。因此我們想要開發一個針對胃癌病患血液樣品做檢驗的分析方法。而胃癌的檢驗標的包括一些標的蛋白、基因表現指數，如一些胃癌誘導控制的細胞因子及趨化因子。在考量一些因素如靈敏度、專一性、方便性、偽陽性，以及參考之前的文獻後，我們選擇了 OPN、 CXCL1、 TIMP1 和SPARC 做為檢測標的。我們的策略是利用開發針對這些胃癌的標的蛋白具高度專一性的抗體來進行非侵入式的檢驗。 我們的實驗從這些標的蛋白的基因轉殖開始，接著進一步進行這些標的蛋白質的表現及純化。然後利用這些標的蛋白做為抗原進行小鼠的免疫，來製造具專一性的抗體。我們成功發展了針對 OPN、CXCL1、TIMP1 及 SPARC 具專一性的抗體。我們對這些抗體進行一些性質及專一性測試，包括酵素連結免疫吸附法及西方點墨法。利用這些抗體做為檢測探針，我們嘗試開發一個可針對胃癌進行血液樣品檢驗的方法。簡單來說，這個系統是將抗體利用硫醇基和金形成共價鍵的方式來修飾在奈米金球上。以一個三明治形式的方式進行標的蛋白的檢測。在我們的三明治形式的檢測系統裡，各個標的蛋白的偵測極限分別為OPN是2 ng/mL、CXCL1是100 pg/mL、TIMP1是300 pg/mL，及SPARC是 5 ng/mL。我們希望未來將進一步將這個系統用在病人血液樣品的檢驗上，並利用結合多個生物標的分子的策略來提高診斷的準確率。此外，未來我們也可以藉由結合四個生物標的分子在不同臨床期的胃癌病人會有不同形態的分析結果，並藉此來區分不同臨床期的病人。

Gastric cancer is the fourth frequent cancer and the second leading cause of cancer related death worldwide. Nowadays, the previous reports indicate that there is some geographical variation in the distribution of gastric cancer. High risk areas include Japan, China, Korea, Eastern Europe, Portugal, and parts of South and Central America. According to the American Cancer Society, the geographical distribution of gastric cancer is considered to have something to do with the dietary risks because of the local stomach-stimulating food culture in these countries. Understanding the progression of gastric cancer is important for the following cancer therapy and drug treatment strategies. The most common diagnosis methods used nowadays are some invasive ways include endoscopic exam with the insertion of fiber optic camera in stomach, computed tomography and histopathology test with tissue sampling. However, these methods all require various kinds of invasive ways in the diagnosis of gastric cancer, and exploiting expensive instruments and complicated steps in sample acquirement and analysis. Therefore, it is necessary to develop a specific, accurate, fast and easy-operating diagnosis method for gastric cancer. Among all kinds of samples, the blood sample is the most general kind and convenient sample to acquire. Therefore, we want to design a detection method for blood sample, and the targets for gastric cancer detection including biomarkers and genes expression, which are usually some gastric cancer-induced controlling cytokines or chemokines. After considering some effecting factors, such as sensitivity, specificity, convenience and false positive signal, we choose OPN, CXCL1, TIMP1 and SPARC as the detecting targets. The strategy of our work is to develop specific antibodies against these biomarkers of gastric cancer for non-invasive diagnosis. The processes of our work start from gene cloning of the biomarkers, further protein expression and purification. Then, exploit these biomarkers as antigens for mice immunization to produce specific antibodies. We have successfully developed the specific antibodies of OPN, CXCL1, TIMP1 and SPARC. The characterization tests including western blot and ELISA have been done for the developed antibodies. With these antibodies as detecting probes, we have developed a diagnosis system for gastric cancer detection of serum samples. The brief idea of the system is exploiting antibody, which are initially immobilized on a gold nanoparticles with the formation of thionine-gold covalent bond. A sandwich-type immunoassay format was employed for the detection of biomarkers. The detection limits of each biomarker in our sandwich immunoassay system are 2 ng/mL for OPN, 100 pg/mL for CXCL1, 300 pg/mL for TIMP1 and 5 ng/mL for SPARC, respectively. We will further exploit the system for serum sample test and exploit the strategy of combination of multiple biomarkers to increase diagnosis accuracy. Furthermore, we may develop a diagnosis system to distinguish gastric cancer in different clinical stages according to different patterns with the combination of four biomarkers.