A型流行性感冒病毒PB2蛋白質與細胞蛋白質hnRNP M之交互作用

Abstract

流感病毒的 PB2 蛋白質為構成其病毒特有的RNA-dependent RNA polymerase (RdRp) 的重要一員，對於病毒複製或是RNA 轉錄都是不可或缺的。本實驗室在找尋可能與PB2 進行交互作用的細胞蛋白質時，利用共免疫沈澱法(co-IP) 發現細胞中的hnRNP M 可與PB2 結合，並且在真實病毒感染之下，也能夠發現到hnRNP M 可與PB2 結合，而在前人的研究中也證實hnRNP M 可能會結合上PA-PB2- PB1-TAP tag 複合體。 在之前的研究中，我們利用GST pull-down assay觀察到PB2可能透過其中間片段(胺基酸281-511)與hnRNP M結合，而此結合片段恰包含了PB2的cap-binding位置。為了反向找出hnRNP M是利用本身哪一個位置與PB2來結合，我們將hnRNP M分為許多片段刪除組，同樣使用GST pull-down assay來觀察，發現到hnRNP M可能是透過其中間Met-Arg-Gly rich片段來結合PB2。 為了測試hnRNP M結合了PB2對於流感病毒的生長所造成之影響，利用hnRNP M shRNA分別在H1299與NPC-TW04細胞株內篩選出hnRNP M knock-down之穩定細胞株，而發現流感病毒在hnRNP M-knockdown之穩定細胞株內的複製效率會有增加的情形，顯示hnRNP M可能會抑制流感病毒的複製機轉。而在螢光酶報導系統中觀察到了hnRNP M可能不利於流感病毒的轉錄或複製。 更進一步探討hnRNP M是否參與在流感病毒內M(matrix) mRNA splicing之調控，以達到影響流感病毒生長的作用，使用流感病毒感染NPC-TW04 hnRNP M-knockdown混合穩定細胞株，並使用semi-qPCR 來偵測M1 mRNA、M2 mRNA 及mRNA3的含量，經過定量後發現hnRNP M並無影響在流感病毒M mRNA splicing的過程，但是卻觀察到在hnRNP M-knockdown穩定細胞株中的M mRNA含量較母細胞株來的更高，而在M vRNA的含量，雖然在hnRNP M-knockdown細胞中較多，但可能是由轉錄作用的二次效應而來，因此推測hnRNP M可能是透過抑制流感病毒的轉錄來達到減少病毒複製的效果。 為了測試hnRNP M是否透過阻擋了PB2的cap-binding片段去接觸mRNA來抑制病毒的轉錄作用，我們純化出流感病毒的聚合酶複合體(vRNPs)與hnRNP M蛋白質，並將兩者與含有5’-cap的微粒子一起混合，而在結果中發現到hnRNP M的確會降低PB2與5’-cap微粒子之間的結合，因此我們才會在之前的實驗中觀察到流感病毒的轉錄作用會受到hnRNP M的抑制。 為了觀察PB2對於hnRNP M本身RNA選擇性剪接功能之影響，將FGFR2 minigene 質體送入293T 細胞中，利用semi-qPCR來觀察splicing 產物，結果發現到在表現PB2時，hnRNP M的splicing功能會受到抑制。另外，我們在病毒感染後的H1299細胞內觀察到，hnRNP M的表現量在感染初期會有短暫性增加的現象，而hnRNP M的活性與表現量在病毒感染期間是如何被調控仍需要更進一步的探討。 最後我們想知道hnRNP M的MAG-rich region是否能夠因為結合了PB2的cap-binding domain而使得流感病毒的複製能力受到影響，於是我們將MAG-rich region放入細胞內，發現到流感病毒的聚合酶活性會受到抑制，使得病毒的轉錄作用有所減少，因此降低了流感病毒的生長能力，未來可以找出結合上cap-binding domain的主要胺基酸片段，利用合成peptide的方式來將此胺基酸片段成為可實際應用的流感治劑。

PB2, a component of influenza A RNA polymerase complex, plays an important role in influenza A viral transcription and replication. In the process of searching for cellular proteins that interact with PB2, we accidentally found that PB2 could interact with cellular hnRNP M protein, an mRNA splicing factor. This interaction was confirmed by co-immunoprecipitation and GST pull-down assays. More importantly, we also found that hnRNP M could interact with PB2 during influenza A virus infection of human H1299 cells. This finding is consistent with the previous report that hnRNP M can be pull-down by PA-PB2-PB1-TAP tag complex. In the previous study, our lab found that PB2 can interact with hnRNP M through a region containing amino acids 281-511 which overlaps the crucial cap-binding domain of PB2. To study which regions of hnRNP M interact with PB2, a series of truncated hnRNP M proteins were used in GST pull-down assays. We found that hnRNP M used its Met-Arg-Gly rich domain to interact with PB2. To study the effect of hnRNP M on influenza A viral replication, we generated hnRNP M knock-down cells by using lentivirus expressing shRNA against hnRNP M in H1299 and NPC-TW04 cell lines. We found that the production of influenza A virus was significantly increased in hnRNP M knock-down cells, suggesting that hnRNP M can inhibit influenza A viral replication. This conclusion was further supported by the following two results. First, overexpression of hnRNP M could inhibit influenza A virus replication. Second, Influenza replication-reporter assays showed that the viral RNA-dependent RNA polymerase activity was increased in hnRNP M knock-down cells. To study the mechanism underlying hnRNP M inhibition of influenza replication, we first tested whether hnRNP M could affect the splicing of influenza M mRNA. We infected hnRNP M-knockdown NPC-TW04 cells and the control cells with influenza A virus and quantified the splicing products of M pre-mRNA. We found that relative amounts of M splicing products were similar in hnRNP M-knockdown cells and control cells, suggesting that hnRNP M may not be involved in the regulation of M mRNA splicing. However, we did find that the level of total M mRNA was increased in hnRNP M-knockdown cells when compared to control cells. The level of NP mRNA was also higher in hnRNP M-knockdown cells than in control cells. These data suggest that hnRNP M can inhibit the transcription of influenza genes. We also tested whether hnRNP M would affect the production of influenza vRNA. Our data indicated that the expression level of M vRNA was up-regulated in hnRNP M-knockdown cells, which is in consistent with the above data that influenza viral replication was increased in hnRNP M-knockdown cells. Together, these data suggest that hnRNP M may inhibit influenza viral replication through repressing viral transcription. To study the mechanism by which hnRNP M repress influenza viral transcription, we tested whether hnRNP M could inhibit the cap-snatching activity of PB2 by performing in vitro cap-binding assays. PB2, in complex with PB1 and PA, is known to be able to bind the cap structure of mRNA. We thus tested whether purified hnRNP M would affect viral polymerase complex (vRNPs) to bind its substrate 7-methyl-GTP. Our data indicated the capability of PB2 to bind 7-methyl-GTP was decreased in the presence of hnRNP M protein, suggesting that hnRNP M can inhibit the cap-snatching activity of PB2. We also tested whether PB2, through interacting with hnRNP M, would affect hnRNP M’s regulatory function on RNA splicing. hnRNP M is known to promote alternative splicing of the FGFR2 minigene. By quantifying the splicing products of the FGFR2 minigene in the presence or absence of PB2, we found that PB2 could dose-dependently inhibit the activity of hnRNP M to promote alternative splicing of the FGFR2 minigene. Knowing that hnRNP M can inhibit PB2 cap-snatching activity and uses its Met-Arg-Gly rich domain to interact with PB2’s cap-binding domain, we next tested whether the peptide containing Met-Arg-Gly rich domain of hnRNP M could inhibit influenza A viral transcription and replication. We found that both viral transcription and replication was reduced in cells expressing the Met-Arg-Gly rich domain, suggesting that peptides that can bind to the cap-binding domain of PB2 may be used to inhibit influenza A replication. Finally, we found that the expression level of hnRNP M was up-regulated transiently during the early stage of influenza A virus infection. This data together with the above data strongly suggest that hnRNP M is a molecule host uses to defend influenza A virus infection.