Characterization of the Mammalian mRNA 3' Processing Complex
- Author(s): Chan, Serena Leong
- Advisor(s): Shi, Yongsheng
- et al.
mRNA 3' processing, which typically involves an endonucleoytic cleavage followed by polyadenylation (addition of a string of adenosines), is an essential step in eukaryotic gene expression and significantly impacts many other gene expression steps such as transcription, splicing, mRNA export and translation (Zhao et al. 1999) (Chan et al. 2010) (Colgan et al. 1997) (Moore et al. 2009). Additionally, 3' processing is needed for gene regulation. Recent studies revealed that approximately 70% of human genes produce multiple mRNA isoforms with different 3' processing sites (Derti et al. 2012) (Hoque et al. 2013). These mRNA isoforms may encode different proteins or produce different 3' untranslated regions. This phenomenon, called alternative polyadenylation (APA), significantly expands the coding capacity of the genome and has been increasingly recognized as a critical mechanism for eukaryotic gene regulation (Di Giammartino et al. 2011) (Shi 2012) (Proudfoot 2011) (Tian et al. 2013). In addition, aberrant mRNA 3' processing causes a wide range of diseases, including IPEX syndrome, thalassemia and has been implicated in the development of cancer (Danckwardt et al. 2008) (Mayr et al. 2009). Therefore, it is critical to understand both the mechanism of mRNA 3' processing and its regulation.
mRNA 3' processing requires specific RNA-protein and protein-protein interactions. The proteins required in mammals include four multi-subunit complexes and the poly (A) polymerase (PAP) while the main RNA sequences include the AAUAAA hexamer and U/G-rich elements (Zhao et al. 1999) (Chan et al. 2010) (Colgan et al. 1997). A central question in the mRNA 3' processing field has been to understand how the mRNA 3' processing sites, also called poly (A) sites, are specifically recognized and regulated. To shed light on this important question, I carried out three projects. First, comprehensive proteomic analyses of the mRNA 3' processing machinery were performed. These were accomplished by purifying all sixteen essential mRNA 3' processing factors by immunoprecipitation and identifying their associated proteins through high throughput mass spectrometry analyses. The results of this study not only provided a nearly comprehensive interactome map of the mRNA 3' processing machinery, but also revealed potential new regulatory mechanisms. I have experimentally validated the association between the mRNA 3' processing factors and some of the newly identified interacting proteins, including several ubiquitin E3 ligases, and have provided evidence that these factors regulate the stabilities of mRNA 3' processing factors. Second, I have characterized the mechanism by which the cleavage and polyadenylation specificity factor (CPSF) specifically recognizes the AAUAAA hexamer. In contrast to the prevalent model in which the CPSF subunit, CPSF 160, recognizes the AAUAAA by itself, my data provided direct evidence that the CPSF subunits, CPSF 30 and WDR33, directly bind to AAUAAA together. Additionally, I showed that the CPSF 30-RNA interaction is mediated by its zinc fingers two and three, which, remarkably, are directly targeted by the influenza A viral protein, NS1A, to suppress host mRNA 3' processing (Nemeroff et al. 1998) (Twu et al. 2006). Finally, I provide evidence that the specificity of the CPSF-RNA interaction is limited and that it requires additional factor(s) for proofreading. Together, the results from these three projects provide novel and significant insights into the fundamental mechanisms for mammalian poly(A) site recognition and regulation.