以B型肝炎病毒持續存在之CBA/caj小鼠模式探討其產生之肝臟腫瘤

Abstract

肝細胞癌(Hepatocellular carcinoma, HCC)是目前世界上癌症致死率排名第二名的癌症，每年超過六十萬肝癌病患因此往生。在眾多造成肝細胞癌的危險因子中，B型肝炎(Hepatitis B)慢性感染佔肝細胞癌病例總和的一半。然而，對於B型肝炎領域的研究因為缺乏理想的B型肝炎病毒(Hepatitis B virus)感染動物模式而阻礙重重。黑猩猩、樹鼩等B型肝炎病毒的自然宿主雖然可以被B型肝炎病毒感染，但遺傳背景變異大且實驗操作不易。近年來，科學家利用B型肝炎轉殖小鼠(HBV transgenic mice)進行研究，然而此小鼠模式無法探討宿主對B型肝炎病毒的免疫反應，因此也無法探討肝癌進程中免疫反應扮演的角色。由於小鼠肝臟細胞膜上缺乏B型肝炎病毒受器，本實驗室使用高壓注射(hydrodynamic injection)方式將B型肝炎病毒質體去氧核醣核酸(HBV plasmid DNA)送入免疫功能正常的CBA/caj小鼠肝臟細胞。此方法在先前研究中已證實B型肝炎病毒得以在CBA/caj小鼠肝臟細胞內長期持續地表現，藉此模擬人類感染B型肝炎後慢性帶原的現象。意外地，我們在犧牲約五十週齡的B型肝炎病毒持續表現的CBA/caj小鼠時，發現其肝臟有腫瘤產生，且產生肝臟腫瘤的小鼠佔全部犧牲小鼠的比率為百分之六十五。先前的研究顯示，在約八十週齡的其他次品系CBA小鼠自發性產生肝臟腫瘤的機率是百分之十。由此推測，遺傳背景、環境因子或B型肝炎病毒持續表現可能促進了肝臟腫瘤產生。且由於本實驗的開端是個意外的發現，不能排除高壓注射或是質體對於肝臟腫瘤生成的影響。因此，本實驗的假說為：在CBA/caj小鼠肝臟細胞中持續表現的B型肝炎病毒、高壓注射方法、質體、遺傳背景或環境因子可能促進了肝臟腫瘤的生長。 首先探討CBA/caj小鼠肝臟腫瘤是否和臨床上B型肝炎導致的肝細胞癌有相似的表徵。我們分析了血清中丙氨酸轉氨酶(Alanine aminotransferase, ALT)值，結果顯示有腫瘤的小鼠和沒有腫瘤的小鼠數值並沒有顯著差異。並且，在腫瘤產生小鼠的肝臟非腫瘤部分也沒有發現顯著的膠原蛋白堆積。接著，我們利用即時聚合酶鏈式反應分析了肝臟腫瘤細胞核酸中B型肝炎病毒X蛋白(HBx)的複製數，結果在小鼠肝臟腫瘤及非腫瘤部分數值都低於偵測極限值，指出B型肝炎病毒可能沒有嵌入宿主基因。之後，我們分析了小鼠肝臟腫瘤細胞基因是否有出現肝細胞癌病人癌細胞基因常見的突變位點。我們分析了下述三個基因：Tert promoter、Trp53以及Ctnnb1。實驗結果顯示在小鼠肝臟腫瘤細胞基因中，在Tert基因轉錄起始點上游1000個鹼基對以及Trp53外顯子四到外顯子八序列中都沒有突變。然而，在一些肝臟腫瘤細胞中，β-catenin路徑活化的指標，谷氨醯胺合成酶(glutamine synthetase, GS)會大量表現，且在一些約九十週齡的小鼠腫瘤細胞基因中有和肝細胞癌病人一致的Ctnnb1外顯子三突變位點。為了近一步探討小鼠肝臟腫瘤的特性，我們分析了一個小鼠自發性腫瘤常見的突變位點，H-ras基因突變。結果顯示在約五十週齡的小鼠肝臟腫瘤細胞中有該突變發生。 由於本研究始於意外發現，缺少控制組，因此我們設計了一組一年期追蹤實驗，其包含未注射任何物質的控制組、高壓注射載體質體去氧核醣核酸的高壓注射控制組以及高壓注射B型肝炎病毒質體去氧核醣核酸的B型肝炎病毒持續表現組，比較一年後肝臟腫瘤發生率。實驗結果顯示三組之肝臟腫瘤發生率無顯著差異。進一步測量腫瘤大小，三組中大部分肝臟腫瘤的直徑皆小於0.4公分。和先前的發現一致，我們除了在控制組小鼠腫瘤細胞中發現H-ras基因突變也在高壓注射控制組及B型肝炎病毒持續表現組的小鼠肝臟腫瘤細胞中發現谷氨醯胺合成酶過度表現。 在本研究中，我們發現小鼠肝臟腫瘤中谷氨醯胺合成酶過度表現以及Ctnnb1外顯子三突變和臨床上肝細胞癌病人表徵相似，然而在H-ras突變以及 五十二週齡小鼠肝臟腫瘤發生率比較結果顯示：在CBA/caj小鼠肝臟細胞中持續表現的B型肝炎病毒、高壓注射或質體皆不會促進肝臟腫瘤的生長。

Hepatocellular carcinoma (HCC) is the second leading cause of death from cancer worldwide, and hepatitis B virus (HBV) chronic infection accounts for about half of all HCC cases. Despite years of efforts, research on HBV has been greatly hindered due to the lack of ideal animal models. Natural hosts such as chimpanzees and tupaias face the challenge of a widely varied genetic background among individuals and the difficulty for manipulation. A popular animal model, the HBV transgenic mice, on the other hand, could not serve to study host immune respond to HBV. Here, we used hydrodynamic injection (HDI) mouse model to transfect HBV plasmid DNA in immunocompetent CBA/caj mice, a long-term persister of HBV to mimic chronically infected patients. Unexpectedly, we discovered liver tumors growth in HBV-carrying CBA/caj mice with an average age at 50 weeks and a 65% tumor incidence rate. Such phenomenon differs from previous studies, which stated a 10% spontaneous tumor incidence rate for other CBA substrain mice at the average age of 80 weeks. Thus genetic background, environment factors or HBV persistence are speculated to promote liver tumor formation in CBA/caj mice. Due to the nature of an accidental discovery, the influence of HDI or AAV vector on liver tumorigenesis could not be ruled out prior the discovery. Therefore, it is hypothesized that HBV persistence, hydrodynamic injection or AAV vector may promote liver tumor growth in HBV-carrying CBA/caj mice. In order to investigate whether the tumors of CBA/caj mice exhibit similar patterns with the clinical HBV-related HCC samples, a series of experiments was conducted. A liver inflammatory marker, the serum Alanine aminotransferase (ALT) level, was first evaluated but no significant difference between tumor-developed mice and the mice without tumor growth was indicated. In addition to the finding, the non-tumor part of liver from the tumor-developed group displayed no significant collagen deposition, implying no fibrosis. Since the above findings hinted that the limited inflammatory responses might not contribute much to tumor formation in CBA/caj mice, other possible mechanisms were explored: 1) HBV integration in host genome 2) common mutation genes in HBV-related HCC patients. First, the tumor samples were performed qPCR to evaluate the HBV x protein (HBx) gene copy number in the host genome, and the results suggested no HBx integration. Second, most common mutation genes/sites among HCC patients were inspected: TERT promoter, p53, and β-catenin genes. It was revealed that there were no mutations at neither Tert 1 kb upstream of transcription starting site nor the exons 4-8 of p53. However, glutamine synthetase (GS, reporter for β-catenin activation) was found to overexpress in some tumors and there were several mutations in β-catenin exon 3 among 90 weeks old mice. The results of β-catenin mutations were consistent to the clinical finding among HCC patients. Furthermore, to clarify the characteristics of these liver tumors, a genetic marker of mouse spontaneous liver tumor, H-ras mutation was examined. It was found that mutations occurred at codon 61 in liver tumors of HBV-carrying CBA/caj mice aged around 50 weeks. Since the discovery was accidental, an experiment including a group of naïve mice, a group of HDI control mice received vector plasmid DNA, and a group of HDI-HBV mice was required. Compared the liver tumor incidence of 52-week-old mice across the three groups, the results showed no significance. The sizes of tumors across the three groups were then further examined, but most of the tumors were visible foci. Furthermore, consistent to the previous finding, H-ras mutations were found in naïve CBA/caj mice and GS overexpression were found in HBV-carrying and HDI-vector mice. In this study, consistent to the clinical finding among HCC patients, GS overexpression in the liver tumors and the β-catenin exon 3 mutations were discovered in HBV-carrying CBA/caj mice. However, tumor incidence across the three groups are equivalent. Taken together, our data suggested HBV persistence, hydrodynamic injection and AAV vector do not promote liver tumor formation in CBA/caj mice.