Juggling Key Players in NMD Initiation

In this issue of Structure, Melero and colleagues use electron microscopy combined with biochemistry to provide structural insight into the complex between SMG1, SMG8, SMG9, UPF1, and UPF2, elucidating how key players in nonsense-mediated mRNA decay assemble at the initial steps of the process. In this issue of Structure, Melero and colleagues use electron microscopy combined with biochemistry to provide structural insight into the complex between SMG1, SMG8, SMG9, UPF1, and UPF2, elucidating how key players in nonsense-mediated mRNA decay assemble at the initial steps of the process. Nonsense-mediated mRNA decay (NMD) is a translation-dependent surveillance mechanism that degrades transcripts with a premature termination codon (PTC), preventing the accumulation of truncated proteins in the cell. NMD acts to degrade mRNAs with mutations introduced either at the DNA level or in the RNA itself during transcription or processing. Because about a third of all disease mutations in humans yield mRNAs with a PTC, NMD is also a clinically relevant pathway (Schweingruber et al., 2013Schweingruber C. Rufener S.C. Zünd D. Yamashita A. Mühlemann O. Biochim. Biophys. Acta. 2013; 1829: 612-623Crossref PubMed Scopus (247) Google Scholar). Furthermore, NMD has been shown to affect gene expression by regulating the abundance of physiological transcripts that, for example, acquire an apparent PTC through alternative splicing mechanisms or inclusion of upstream ORFs (Schweingruber et al., 2013Schweingruber C. Rufener S.C. Zünd D. Yamashita A. Mühlemann O. Biochim. Biophys. Acta. 2013; 1829: 612-623Crossref PubMed Scopus (247) Google Scholar). Triggering NMD requires the formation of a series of protein complexes that recognize the PTC and subsequently recruit the degradation machinery. In humans, PTCs are often recognized by their position relative to the exon junction complex (EJC) (Le Hir et al., 2001Le Hir H. Gatfield D. Izaurralde E. Moore M.J. EMBO J. 2001; 20: 4987-4997Crossref PubMed Scopus (603) Google Scholar). The core NMD factors are UPF proteins (UPF1, UPF2, and UPF3) that interact to sequentially assemble a “surveillance complex” (Serin et al., 2001Serin G. Gersappe A. Black J.D. Aronoff R. Maquat L.E. Mol. Cell. Biol. 2001; 21: 209-223Crossref PubMed Scopus (200) Google Scholar, Chamieh et al., 2008Chamieh H. Ballut L. Bonneau F. Le Hir H. Nat. Struct. Mol. Biol. 2008; 15: 85-93Crossref PubMed Scopus (222) Google Scholar). UPF1, a conserved ATP-dependent RNA helicase, interacts directly with the release factors eRF1 and eRF2 on the ribosome together with SMG1 forming the so-called SURF complex that links NMD to active translation (Kashima et al., 2006Kashima I. Yamashita A. Izumi N. Kataoka N. Morishita R. Hoshino S. Ohno M. Dreyfuss G. Ohno S. Genes Dev. 2006; 20: 355-367Crossref PubMed Scopus (438) Google Scholar). A key event during the initial steps of NMD is the phosphorylation of UPF1 by the SMG1 kinase (Chamieh et al., 2008Chamieh H. Ballut L. Bonneau F. Le Hir H. Nat. Struct. Mol. Biol. 2008; 15: 85-93Crossref PubMed Scopus (222) Google Scholar), which is in turn regulated by UPF2 and UPF3. SMG1 is a large (3,657 amino acid) phosphatidylinositol 3-kinase-related kinase, consisting of a globular C-terminal region with the catalytic site and an elongated N-terminal region of α-helical HEAT repeats. SMG1 activity is finely regulated by several interactions, not only with the NMD factors UPF2 and UPF3, but also with the kinase inhibitors SMG8 and SMG9 (Yamashita et al., 2009Yamashita A. Izumi N. Kashima I. Ohnishi T. Saari B. Katsuhata Y. Muramatsu R. Morita T. Iwamatsu A. Hachiya T. et al.Genes Dev. 2009; 23: 1091-1105Crossref PubMed Scopus (177) Google Scholar, Arias-Palomo et al., 2011Arias-Palomo E. Yamashita A. Fernández I.S. Núñez-Ramírez R. Bamba Y. Izumi N. Ohno S. Llorca O. Genes Dev. 2011; 25: 153-164Crossref PubMed Scopus (62) Google Scholar). At present, structural information is available that provides detailed mechanistic insight on EJC binding to RNA, on the interaction that assembles the UPF complex on the EJC and on UPF1 function in NMD (Schweingruber et al., 2013Schweingruber C. Rufener S.C. Zünd D. Yamashita A. Mühlemann O. Biochim. Biophys. Acta. 2013; 1829: 612-623Crossref PubMed Scopus (247) Google Scholar, Llorca, 2013Llorca O. Curr. Opin. Struct. Biol. 2013; 23: 161-167Crossref PubMed Scopus (4) Google Scholar). However, given that NMD is tightly regulated with numerous dynamic interactions and remodeling of complexes, several open questions remain. In particular, we still have limited mechanistic information on the link between PTC recognition by the ribosome and the steps leading to the recruitment of UPF1 to NMD substrates and to UPF1 phosphorylation. A mechanistic understanding of this process requires the combination of structural studies providing static snapshots, with biochemical and functional assays. The article by Melero et al., 2014Melero R. Uchiyama A. Castaño R. Kataoka N. Kurosawa H. Ohno S. Yamashita A. Llorca O. Structure. 2014; 22 (this issue): 1105-1119Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar in this issue of Structure is a beautiful example of how such a combination can provide insights into the dynamics of a multiprotein complex. Melero et al., 2014Melero R. Uchiyama A. Castaño R. Kataoka N. Kurosawa H. Ohno S. Yamashita A. Llorca O. Structure. 2014; 22 (this issue): 1105-1119Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar present structural information for a series of complexes formed by SMG1, SMG8, SMG9, UPF1, and UPF2 that gives insight into conformational changes and remodeling steps that might occur during NMD in vivo. SMG1 is a core scaffolding component of NMD complexes, but how it organizes the other components was previously not appreciated. The current view is that UPF2 binds to SMG1 via UPF1. Now Melero et al., 2014Melero R. Uchiyama A. Castaño R. Kataoka N. Kurosawa H. Ohno S. Yamashita A. Llorca O. Structure. 2014; 22 (this issue): 1105-1119Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar present evidence that UPF1 and UPF2 can associate independently and simultaneously to the kinase. Mutant forms of UPF1 and UPF2 unable to interact with each other are still able to bind to SMG1, changing our understanding of the recruitment of UPF proteins to the kinase. Melero et al., 2014Melero R. Uchiyama A. Castaño R. Kataoka N. Kurosawa H. Ohno S. Yamashita A. Llorca O. Structure. 2014; 22 (this issue): 1105-1119Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar used electron microscopy (EM) of negatively stained samples and were able to obtain low-resolution EM surfaces for five different structures, including SMG1 and four subcomplexes of SMG1 with UPF1, UPF2, SMG8, and SMG9. They combined EM with biochemical assays, revealing complex interactions and also temporal and spatial constrains in the assembly of the complex. The work also presents further details about the inhibition of SMG1 by SMG8/9. The EM structures also reveal a large conformational change of the HEAT-repeat region upon SMG8 and SMG9 binding, complementing previous work by the authors (Arias-Palomo et al., 2011Arias-Palomo E. Yamashita A. Fernández I.S. Núñez-Ramírez R. Bamba Y. Izumi N. Ohno S. Llorca O. Genes Dev. 2011; 25: 153-164Crossref PubMed Scopus (62) Google Scholar) (Figure 1). Moreover, they show that although UPF2 and UPF1 alone dock at distinct sites on the SMG1 surface, the order of the interactions is important to assemble a stable complex with all components and influences the final conformation of the assembled pentameric complex. In the pentameric complex, UPF2 does not contact SMG1 directly, but via UPF1 that appears to be in an open, uninhibited conformation (Figure 1). This complex may form when a preassembled UPF1-UPF2 complex binds to the SMG1C (SMG1-SMG8-SMG9) assembly. Indeed, in vitro competition experiments showed that the order of interactions is important for stability. UPF1 can associate when UPF2 is already bound to SMG1C, whereas UPF2 joining displaces a prebound UPF1 (Figure 1). This is supported by the observation that a UPF2 mutant, impaired in UPF1 binding, no longer displaces prebound UPF1, nor can it associate with SMG1. This competition reveals a hitherto unknown complexity in the regulation of the assembly of SMG1 complexes. In the future, it will be interesting to see which of the SMG1/8/9-UPF1/2 complexes is competent to interact with downstream effectors during NMD. This will also give insights on how alternative NMD pathways that have been proposed to act independently of the EJC, of UPF3 or UPF2, might function at the onset of NMD. Another interesting question is how UPF3 binds to the complex and how this contributes to the activation of SMG1. Is UPF3 involved in the release of SMG8 and SMG9? The approach taken by Melero et al., 2014Melero R. Uchiyama A. Castaño R. Kataoka N. Kurosawa H. Ohno S. Yamashita A. Llorca O. Structure. 2014; 22 (this issue): 1105-1119Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar will be useful to answer these questions in the future. The combination of EM and various biochemical assays, including crosslinking to stabilize labile interactions, is a powerful means to obtain structure-function information from low amounts of large and dynamic multiprotein assemblies operating during NMD. I thank Atlanta Cook and Gáspár Jékely for critical reading of the manuscript. This work was supported by the Max Planck Gesellschaft and European Research Council under the European Union's Seventh Framework Programme (FP7/2007-2013), ERC grant agreement no. 310957. Structures of SMG1-UPFs Complexes: SMG1 Contributes to Regulate UPF2-Dependent Activation of UPF1 in NMDMelero et al.StructureJuly 4, 2014In BriefSMG1 phosphorylates UPF1, an essential step in nonsense-mediated mRNA decay in mammals. Melero et al. describe the 3D architecture of SMG1-SMG8-SMG9 bound to UPF1 and UPF2 (a SMG1 and UPF1 activator), revealing that SMG1C recruits UPF1 and UPF2 to distinct sites and that UPF2 can bind and activate UPF1 within SMG1C. Full-Text PDF Open Archive