鑑定新穎微小核醣核酸生合成及功能之調控因子

Abstract

目前已知微小核醣核酸(miRNA)是相當重要的基因表現調控因子,因其可調控超過 70%人類編碼基因之表現與否。然而,目前對於調控微小核醣核酸之生合成與活性的研究相對較少。本篇論文中,我們在線蟲中利用核醣核酸干擾作用(RNAi)抑制 DEAD/H-box 核醣核酸解旋酶家族之蛋白表現,而後觀察其是否有可能參與在微小 核醣核酸的生合成與活性之調控過程中。利用對於微小核醣核酸變化量及功能改 變較敏感的 let-7(mg279)突變株來進行實驗之後,我們發現在線蟲中抑制 DDX-23 及 DDX-17 的表現後會明顯促進 let-7 缺失時所呈現的表現型(phenotype),證明該 蛋白可能參與在 let-7 生合成或是執行功能過程當中。接著,我們亦發現抑制此兩 蛋白之表現後皆明顯造成成熟 let-7 表現量下降,但上游初期 let-7 (primary-let-7) 卻有累積的現象,指出其功能為初期 let-7 加工時所必須。我們後續檢驗了數個不 同的微小核醣核酸以及其初期微小核醣核酸(primary miRNA)之表現量後發現,除 了 let-7 以外,其他的 heterochronic 微小核醣核酸比如 lin-4,miR-48,miR-84 及 miR-241 之加工亦會受到此兩蛋白之影響。最後我們亦證明在特定組織內,非 heterochronic 微小核醣核酸 lsy-6 的功能會因為 DDX-23 及 DDX-17 表現量下降而 降低。因此,我們在此部分提出了一個 DDX-23 會參與在初期微小核醣核酸的加 工步驟過程中的新功能,而且也證明了 DDX-17 在初期微小核醣核酸加工過程中 的角色具有保守性。 在另一方面,在本篇論文中我們亦發現利用核醣核酸干擾作用降低線蟲中數個 hnRNP後只有人類細胞中hnRNP Q的同源蛋白HRP-2可以有效的減輕let-7(n2853) 突變株所表現出來的 let-7 缺失之表現型,證明 HRP-2 在 let-7 調控 lin-41 表現的過 程中扮演重要角色。此外,我們亦發現 HRP-2 可以直接跟微小核醣核酸誘導之核醣核酸靜默複合物(miRISC)形成交互作用,且這樣的交互作用是需要有核醣核酸 於其中當做橋樑來聯結。接著,我們在 lin-41 的訊息核醣核酸之 3’UTR 中位於 let-7 結合位下游但在poly-A 尾巴前找到了一個區域可以直接跟 HRP-2 結合,表示 lin-41 訊息核醣核酸可以是其中一個聯結 HRP-2 以及微小核醣核酸誘導之核醣核酸靜默 複合物的核醣核酸橋樑。然後,我們將此段會結合 HRP-2 的片段從 3’UTR 上切除, 並且接在綠螢光蛋白之核醣核酸後並送入線蟲中去表現。我們發現將此片段去除 的綠螢光蛋白表現量會明顯下降,顯示出此段蛋白對於 let-7 去調控 lin-41 是重要 的。因此我們認為在線蟲中 HRP-2 會降低 let-7 對 lin-41 的調控能力。有趣的是, 我們發現在人類細胞中,我們降低 hnRNP Q 的表現後,let-7 對於人類的 lin-41 同 源基因 TRIM71 的調控能力會有所上升。而且,如同線蟲中的 HRP-2 會結合 lin-41 訊息核醣核酸的 3’UTR 上特定位置一樣,我們在 TRIM71 的訊息核醣核酸之 3’UTR 中亦找到一個特定的區域可以跟 hnRNP Q 結合,且其同樣位於 let-7 結合位及 poly-A尾巴之間。最後我們亦將此區域從3’UTR中去除,結果發現會明顯增加let-7 對 TRIM71 調控的能力。因此,綜合以上所述,我們發現了一個新的 HRP-2 及其 同源蛋白 hnRNP Q 演化上保守的功能,也就是當其結合到 lin-41 或是其同源基因 TRIM71 的訊息核醣核酸 3’UTR 上特定的區域後,會減低 let-7 微小核醣核酸誘導 之核醣核酸靜默複合物對其調控之活性。

As post-transcriptionally gene regulators demonstrated by numerous studies, microRNAs have been predicted to control more than 70% of human coding genes expression. However, studies regarding modulators for miRNA biogenesis and/or function remain relatively fewer. Here, we performed a candidate-based RNAi screen in C. elegans to identify DEAD/H-box proteins involved in miRNA biogenesis and/or function. In a let-7(mg279) sensitized genetic background, knockdown of a homolog of yeast splicing factor Prp28p, DDX-23, or a homolog of human helicases p68 and p72, DDX-17, enhanced let-7 loss-of-function phenotypes, suggesting that these helicases play a role in let-7 processing and/or function. In both ddx-23(RNAi) and ddx-17(RNAi), levels of mature let-7 were decreased while pri-let-7 was found accumulated, indicating that the helicases likely act at the level of pri-let-7 processing. DDX-23 and DDX-17 were also required for the biogenesis of other known heterochronic miRNAs, including lin-4 and the let-7 family members miR-48, miR-84 and miR-241. Their function was not confined to the heterochronic pathway, however, since they were both necessary for down-regulation of cog-1 by the spatial patterning miRNA, lsy-6. Therefore, we present a novel function for C. elegans DDX-23 in pri-miRNA processing, and also suggest a conserved role for DDX-17 in this process. On the other hand, we also show that RNAi knockdown of C. elegans HRP-2, the homolog of mammalian hnRNP Q, relieved the heterochronic phenotypes in let-7(n2853) mutant animals, indicating an involvement of HRP-2 in let-7-lin-41 regulation. In addition, we detected an RNA-dependent interaction between HRP-2 and the Argonaute ALG-1, the core effector protein of miRNA-mediated silencing complex (miRISC). Moreover, we identified an HRP-2 response element in the lin-41 3’UTR at a position, downstream of the two let-7 complementary sites (LCSs), close to the poly(A)-tail. Deletion of this response element caused further down regulation of a GFP reporter carrying the lin-41 3’UTR in a let-7-dependent manner. Thus, we propose that HRP-2 impedes let-7/miRISC activity when binding to the lin-41 3’UTR. Interestingly, we found that depletion of human hnRNP Q also enhanced let-7-mediated down-regulation of TRIM71. Similar to the case in C. elegans, hnRNP Q binds to a response element adjacent to the poly(A)-tail in the TRIM71 3’UTR. Deleting this element from the 3’UTR significantly enhanced let-7 repression. Taken together, our findings uncover a novel evolutionarily conserved function for HRP-2/hnRNP Q to inhibit let-7/miRISC activity when they bind to specific response elements in the lin-41/TRIM71 3’UTRs.