Human ENDOU-1 and Its Ortholog Zebrafish Endouc Attenuates Human uORFchop- mediated Translational Inhibition during ER stress

• Main Authors: Yun Yu, 郁芸
• Other Authors: Huai-Jen Tsai
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/61820419766639451096

碩士 === 國立臺灣大學 === 分子與細胞生物學研究所 === 104 === In response of endoplasmic reticulum (ER) stress, C/EBP homologous protein (CHOP) is critically involved in either cell survival or apoptosis. It has been reported that the 5’UTR of CHOP mRNA contains an upstream open reading frame (uORFchop) which inhibits the translation of downstream coding sequence (CDS) of CHOP. Yet, the uORFchop-mediated translational inhibition of CDS is abolished when ER stress is encountered, resulting in generating CHOP protein. However, underlying molecular mechanisms of uORFchop -mediated translational inhibition is not fully understood. To answer this issue, we employed the zebrafish transgenic line huORFZ, which harbors the GFP reporter fused with the human uORFchop (huORFchop) and driven by a cytomegalovirus promoter. Interestingly, the downstream reporter GFP was expressed only when huORFZ embryos were treated with ER stress. Since the endonuclease poly(U)-specific C gene (endouc) of zebrafish is proven to be involved in the huORFchop-mediated translational inhibition by my colleague, here I studied whether human ENDOU gene, which is orthologue of zebrafish endouc, might also play role on the repression of huORFchop-mediated translational inhibition. When the amino acid sequences of human ENDOU and zebrafish Endouc were compared, I found that they shared a XendoU domain, which possesses an endoribonuclease ability. Furthermore, I demonstrated that human ENDOU and zebrafish Endouc were functionally conserved because (a) both of them were able to cleave the huORFchop mRNA at uridylates located within the loop structure in a Ca2+-dependent manner; and (b) injection of human ENDOU mRNAs could rescue the occurrence of phenotypes caused by knocking down of endouc in zebrafish embryos. Next, we overexpressed the human ENDOU in HEK293T cells without giving stress, resulting in the increase levels of phosphorylated eIF2α (p-eIF2α) and CHOP proteins. Additionally, the luciferase (luc) activity driven by the huORFchop motif was extremely limited in huORFZ embryos at normal condition. However, the luc was greatly expressed if human ENDOU mRNAs were injected in those embryos even they were not treated with stresses. The line of above evidence suggested that overexpression of human ENDOU did suppress the huORFchop-translational inhibition in vitro, resulting the downstream CDS such as luc cDNA was translated. Furthermore, to demonstrate that the endoribonuclease ability of ENDOU is involved in the repression of the huORFchop-translational inhibition, I constructed a mutation form, ENDOUH285A, in which its endoribonuclease activity was impaired. Interestingly, overexpression of ENDOUH285A mRNA in HEK293T cells led to decrease the protein levels of both p-eIF2α and CHOP, suggesting that the endoribonuclease activity of human ENDOU is involved in suppressing the huORFchop-translational inhibition. Finally, I used the polyribosome profiling assay to demonstrate that, at normal condition, ENDOU was bound at polysomes; while under at ER stress, ENDOU was released from polysomes and presented as a free form by taking advantage of its endoribonuclease activity. Collectively, we conclude that, like zebrafish Endouc, human ENDOU plays role on the huORFchop-mediated translational inhibition since human ENDOU and zebrafish Endouc are functionally conserved. Thus, I hypothesize that, under stress condition, ENDOU is able to recognize specific nucleotides within huORFchop structure located at 5’ end and digest thereof mRNA, resulting in a cap-independent mRNA, which in turn, the downstream CDS of CHOP cDNA is therefore reinitiated to translate through bypassing the hindrance of huORFchop structure.