探討RIG-I (視黃酸誘導基因-I) 在肝細胞癌中之角色

Abstract

模式識別受體 (pattern recognition receptors, PRRs) 的基因變異已經被提出可能和多種癌症的發生相關，其中包括肝癌。視黃酸誘導基因-I (Retinoic acid-inducible gene-I, RIG-I) 在大多數的研究中被視為一個能辨認病毒RNA的PRR，能活化免疫防禦的機制。然而，已有研究指出，在沒有病毒感染的情況下，RIG-I 的表現會影響胚胎發育和細胞分化的過程；此外，在 RIG-I 基因剔除小鼠的胚胎發育過程中，會出現肝細胞大量死亡的現象。根據過去研究表示，RIG-I 在前列腺癌、乳癌、皮膚癌和星狀細胞瘤細胞中的表現量都較其正常細胞中低，但目前 RIG-I 在肝癌病人腫瘤中的表現量和其扮演的角色仍不清楚。故本篇研究欲探討 RIG-I 在肝癌中參與的生物功能及其調控機轉。我們首先確認 RIG-I mRNA 表現量在肝癌病患的腫瘤組織中顯著低於正常組織中，而 RIG-I 在腫瘤組織中的表現量也與惡性程度分級呈現負相關。接著，我們發現相較於野生型肝癌細胞株，抑制 RIG-I 表現的肝癌細胞株和 RIG-I 突變的肝癌細胞株在免疫缺陷鼠中皆有顯著較高的致腫瘤性，細胞的自我更新能力有顯著地提升，肝癌幹細胞標誌的蛋白表現量也有明顯增加。我們更發現RIG-I會影響上皮-間葉轉換的標誌蛋白表現量，且重要的是細胞轉移及侵襲的能力在 RIG-I 過度表現的細胞株中被顯著抑制；反之，細胞轉移及侵襲的能力在抑制RIG-I 表現的細胞株中則是顯著增加。而在細胞株轉染 RIG-I 不同片段的實驗中證實，只有凋亡蛋白酶募集區域 (caspase recruitment domain, CARD domain) 對細胞的自我更新及轉移侵襲有抑制的效果。我們進一步又發現CARD domain下游分子粒腺體抗病毒蛋白 (mitochondrial antiviral signaling protein, MAVS) 在細胞中被抑制表現後，細胞轉移及侵襲的能力亦顯著提升，顯示 RIG-I 很可能透過 MAVS 分子調控肝癌細胞的轉移及侵襲能力。總合上述，本篇研究證實，RIG-I 在肝癌中可能扮演抑癌基因的角色，對於肝癌細胞的致腫瘤性、自我更新及轉移侵襲能力有明顯的影響，可作為未來研究肝癌治療一個新的方向。

Inherited variation in pattern recognition receptors (PRRs) has been reported that it may be correlated to several types of cancer, including liver cancer. Retinoic acid-inducible gene-I (RIG-I), also named DDX58, is one of the PRRs and generally known as an RNA virus sensor in innate immune response. Nonetheless, recent studies have reported RIG-I may be involved in the embryogenesis and cell differentiation without any signs of viral infection. RIG-I knockout mice exhibited hepatic cell death during embryogenesis. Importantly, a lower expression level of RIG-I was observed in prostate, breast, melanocyte and astrocyte cancer cells as compared with their normal counterparts. However, its expression level and biologic effect in liver cancer remain poorly understood. Hence, we attempted to explore the function and the regulatory mechanism of RIG-I in hepatocellular carcinoma (HCC). In our study, we first demonstrated RIG-I mRNA expression level in the tumors of HCC patients were significantly lower than in their adjacent normal tissues, and the expression levels of RIG-I in tumor parts were negatively correlated to their tumor grades. In addition, RIG-I knockdown cells and RIG-I mutated cells exhibited greater tumorigenicity as compared with their parental cells in NOD/SCID mice. We also showed that even 1000 RIG-I mutated cells possessed tumorigenicity in NOD/SCID mice, and this phenomenon is one of the properties of cancer stem cell. We therefore postulated that RIG-I may play a key regulator in cancer stem cell. Our results indicated knockdown of RIG-I increased self-renewal ability, and the cancer stem cell markers expression increased as well. Furthermore, RIG-I could influence the expression level of epithelial- mesenchymal transition markers in hepatoma cells. Our data demonstrated overexpression of RIG-I suppressed invasion/migration capacity, while RIG-I knockdown cells possessed augmented invasion/migration capacity. The existence of a non-mutated caspase recruitment domain (CARD domain) in RIG-I may play an important role in suppressing of self-renewal ability and invasion/migration capacity. Due to the importance of CARD domain, we also demonstrated knockdown of mitochondrial antiviral signaling protein (MAVS), the RIG-I CARD domain downstream molecule, increased invasion/migration capacity as well. This may imply MAVS was a regulatory molecule of RIG-I mediated invasion/migration. Based on our observations, we purposed RIG-I may be a tumor suppressor, and the induction of RIG-I expression level can be a potential remedy strategy of HCC progression.