胃癌與十二指腸潰瘍相關之幽門螺旋桿菌高抗原性蛋白質之鑑定與特性分析

Abstract

幽門螺旋桿菌感染是造成人類慢性活動性胃炎、十二指腸潰瘍、胃潰瘍的主要因素，並且與胃癌有很大的相關性。幽門螺旋桿菌感染胃部黏膜所引起的發炎反應被認為是造成腸胃道疾病的主因，雖然已有許多幽門螺旋桿菌的毒性因子被發表，但其對於胃部疾病的關聯性不是很確定並且通常有地理區域上的差異。由於初始感染幽門螺旋桿菌的胃部位置不同發生不同類型的慢性胃炎，其後所發展成的十二指腸潰瘍與胃癌被認定為臨床上兩種不同走向的疾病，而流行病學的研究也指出患有十二指腸潰瘍的病人將來發展為胃癌的風險相較於患有胃潰瘍、萎縮性胃炎或是伴隨腸化生現象的病人來說是很低的。 本篇論文的研究以十二指腸潰瘍與胃癌為不同走向的疾病為基礎，利用二維電泳免疫轉漬分析幽門螺旋桿菌其以甘胺酸酸性液萃取之蛋白質被胃癌及十二指腸潰瘍病人血清辨認情況。本篇論文的第一個部分其目的是要鑑定與胃癌相關的幽門螺旋桿菌抗原。利用臨床上分離出的胃癌菌株，以上述方法比較15個胃癌與15個十二指腸潰瘍血清辨認差異，找到24個胃癌血清辨認次數較多的蛋白質，其中胃癌血清辨認次數為十二指腸潰瘍血清的2倍以上的蛋白質為：threonine synthase、rod shape-determining protein、S-adenosylmethionine synthetase、peptide chain release factor 1、DNA-directed RNA polymerase a subunit、co-chaperonin GroES (包括單體及雙體)、response regulator OmpR、membrane fusion protein。在這些蛋白質中，GroES的辨認次數差異最大，並且在大量臨床檢體的篩檢下發現GroES的血清陽性率在95個胃癌檢體中為64.2%，相對高於30.9%的胃炎 (94個) 及35.5%的十二指腸潰瘍檢體 (124個)，具有統計上的差異。另外，GroES的血清陽性率在胃賁門癌中比非賁門胃癌來的高 (風險比率為2.7，95%信賴區間為1.1-6.7)。為了研究GroES在幽門螺旋桿菌致病機制上所扮演的角色，我們利用反轉錄酶-聚合酶鏈鎖反應、酵素免疫吸附試驗、西方點墨法及細胞存活率試驗來分析GroES所造成的前發炎性細胞激素分泌量以及與細胞增生相關的分子表現量。我們發現GroES會刺激週邊血液單核細胞分泌IL-8、IL-6、GM-CSF、IL-1beta、TNF-alpha、cyclooxygenase-2表現量與prostaglandin E2分泌增加，對於胃上皮細胞株 (KATO-III)則會造成IL-8分泌增加、c-jun、c-fos基因表現量上升、mutated p53、cyclin D1、p21WAF1/Cip1、survivin表現增加及抑制p27Kip1之表現。結果顯示GroES是與胃癌相關的幽門螺旋桿菌新的毒性因子，可能經由引起發炎反應及促進細胞增生在胃癌發生過程中扮演重要角色。 本篇論文的第二部分其目的是要鑑定與十二指腸潰瘍相關的幽門螺旋桿菌抗原並且發展簡單又快速的檢驗平台。利用臨床上分離出的十二指腸潰瘍菌株，利用二維電泳免疫轉漬分析比較10個十二指腸潰瘍血清與10個胃癌血清辨認幽門螺旋桿菌蛋白質情況，有11個蛋白質被十二指腸潰瘍血清強烈辨認，然而之中與胃癌血清辨認次數相比有差異的蛋白質為translation elongation factor EF-G (FusA)、catalase (KatA)與urease alpha subunit (UreA)。大量臨床檢體的篩檢結果顯示3個蛋白質在十二指腸潰瘍檢體 (124個) 的血清陽性率皆高於胃癌檢體 (95個)，分別是FusA：70.2% vs. 45.3%，KatA：50.8% vs. 41.1%，UreA：44.4% vs 27.4%，除KatA外皆具有統計上的差異。經統計分析後發現隨著辨認抗原的個數增加，能促進區別十二指腸潰瘍與胃癌檢體，其結果如下：辨認1個抗原勝算比為1.82 (95%信賴區間為0.79-4.21，P值為0.1607)，辨認2個抗原勝算比為4.95 (95%信賴區間為2.05-12.0，P值為0.0004)，辨認3個抗原勝算比為5.71 (95%信賴區間為1.86-17.6，P值為0.0024)。為落實能夠同時使用2個以上蛋白質為篩檢標記的想法，我們發展出包括FusA、KatA、UreA的蛋白質晶片能夠有效區分十二指腸潰瘍與正常人血清。此部分結果顯示：FusA、KatA與UreA為十二指腸潰瘍相關的幽門螺旋桿菌抗原，利用蛋白質晶片技術落實使用多個生物標記的概念能快速且方便地得知十二指腸潰瘍病人的抗體種類之組合。

Helicobacter pylori is a major cause of chronic active gastritis, duodenal ulcer, gastric ulcer and is strongly related to gastric cancer. Chronic inflammation induced by colonization of H. pylori is thought to contribute to pathogenesis in the gastroduodenal tract. Although many virulence factors of H. pylori have been reported, the association between the virulence factors and variant gastric diseases are unclear and not universal. As the distinct distribution and pattern of gastritis occurred, the developed duodenal ulcer and gastric cancer are considered as clinically divergent events, with much lower risk for progressing to gastric cancer among patients with duodenal ulcer than with gastric ulcer or other precancerous diseases such as atrophic gastritis and intestinal metaplasia. Owing to clinical differences between gastric cancer and duodenal ulcer, we compared 2D-immunoblot profiles of acid-glycine extracts from H. pylori by probing with sera of patients with gastric cancer (n = 15) and duodenal ulcer (n = 15). In the first part of studies, we aimed to search for immunogenic proteins related to gastric cancer and therefore extracted the proteins of the H. pylori strain isolated from a gastric cancer patient. In total, 24 protein spots were identified as antigens with better recognition in gastric cancer sera than duodenal ulcer sera. The proteins showing higher frequency of recognition in gastric cancer group are threonine synthase, rod shape-determining protein, S-adenosylmethionine synthetase, peptide chain release factor 1, DNA-directed RNA polymerase a subunit, co-chaperonin GroES (monomeric and dimeric forms), response regulator OmpR and membrane fusion protein. Of these proteins, GroES was identified as a dominant gastric cancer-related antigen with a much higher seropositivity of gastric cancer samples (64.2%, n = 95) compared with 30.9% for gastritis (n = 94) and 35.5% for duodenal ulcer (n = 124). GroES seropositivity was more commonly associated with antral gastric cancer than the non-antral gastric cancer (odds ratio = 2.7; 95% confidence interval, 1.1-6.7). For functional analysis of GroES, we monitored the expression levels of pro-inflammatory cytokines, and marker proteins related to cell proliferation by using reverse transcriptase-polymerase chain reaction (RT-PCR), an enzyme-linked immunosorbent assay (ELISA), Western blotting and a cell viability assay. In peripheral blood mononuclear cells, GroES induced expression of ilterleukin (IL)-8, IL-6, granulocyte macrophage colony-stimulating factor (GM-CSF), IL-1beta, tumor necrosis factor (TNF)-alpha, cyclooxygenase-2 (COX-2), and prostaglandin E2 (PGE2). In gastric epithelial cells (KATO-III cells), GroES up-regulated the expression of IL-8, c-jun, c-fos, mutated p53, cyclin D1, p21WAF1/Cip1 and survivin, but down-regulated the p27Kip1 level. Therefore we concluded that GroES is a novel gastric cancer-associated virulence factor and may contribute to gastric carcinogenesis via induction of inflammation and promotion of cell proliferation. In the second part of the studies, we aimed to identify H. pylori antigens showing a high seropositivity in DU and to develop a platform for rapid and easy diagnosis for DU. Since DU and gastric cancer (GC) are considered clinical divergent gastroduodenal diseases, we compared two-dimensional immunoblots of an acid-glycine extract of an H. pylori strain from a patient with DU probed with serum samples from patients with DU (n = 10) or GC (n = 10), to identify DU-related antigens. Of the 11 proteins that were strongly recognized by serum IgG from DU patients, translation elongation factor EF-G (FusA), catalase (KatA), and urease alpha subunit (UreA) were identified as DU-related antigens, showing higher seropositivity in DU samples (n = 124) than in GC samples (n = 95) (FusA: 70.2% vs. 45.3%, KatA: 50.8% vs. 41.1%, UreA: 44.4% vs 27.4%). In addition, we found that the use of multiple antigens improved the discrimination between patients with DU and those with GC, as the odds ratios increased from 1.82 (95% CI: 0.79-4.21, P = 0.1607) for seropositivity for FusA, KatA, or UreA alone to 4.95 (95% CI: 2.05-12.0, P = 0.0004) for two of the three antigens, and to 5.71 (95% CI: 1.86-17.6, P = 0.0024) for all three antigens. Moreover, a protein chip containing the 3 DU-related antigens was developed to test the idea of using multiple biomarkers in diagnosis. We conclude that FusA, KatA, and UreA are DU-related antigens of H. pylori and the combination of these proteins on a protein array provided is a rapid and convenient method for detecting serum antibody patterns of DU patients.