Crystal structure of the human eIF4AIII–CWC22 PNAS PLUS complex shows how a DEAD-box protein is inhibited by a MIF4G domain Gretel Buchwalda, Steffen Schüsslera, Claire Basquina, Hervé Le Hirb,c, and Elena Contia,1 aDepartment of Structural Cell Biology, Max Planck Institute of Biochemistry, D-82152 Martinsried/Munich, Germany; bInstitut de Biologie de l’Ecole Normale Supérieure, Centre National de la Recherche Scientifique, Unité Mixte de Recherche 8197, 75005 Paris, France; and cInstitut de Biologie de l’Ecole Normale Supérieure, Institut National de la Santé et de la Recherche Médicale U1024, 75005 Paris, France Edited by Joan A. Steitz, Howard Hughes Medical Institute, New Haven, CT, and approved October 23, 2013 (received for review August 2, 2013) DEAD-box proteins are involved in all aspects of RNA processing. (17, 18). Conformational regulation is often used to modulate the They bind RNA in an ATP-dependent manner and couple ATP ATPase activity of DEAD-box proteins, for example in the hydrolysis to structural and compositional rearrangements of ribo- activation of the translation initiation factor 4AI (eIF4AI) by nucleoprotein particles. Conformational control is a major point of the MIF4G domain of eIF4G (22) and in the inhibition of regulation for DEAD-box proteins to act on appropriate substrates eIF4AI by the MA3 domain of PDCD4 (23, 24). The con- and in a timely manner in vivo. Binding partners containing a middle formational changes of DEAD-box proteins are important not domain of translation initiation factor 4G (MIF4G) are emerging as only for catalysis, but also for protein–protein interactions. In important regulators. Well-known examples are eIF4G and Gle1, the case of the EJC, for example, the closed RNA-bound con- which bind and activate the DEAD-box proteins eIF4A and Dbp5. formation of eIF4AIII is required to form a core complex with Here, we report the mechanism of an inhibiting MIF4G domain. its binding partners MAGO, Y14, and Barentsz (also known as We determined the 2.0-Å resolution structure of the complex of Metastatic Lymph Node 51; MLN51) (8, 17, 18). human eIF4AIII and the MIF4G domain of the splicing factor Com- The EJC core is assembled in an ATP-dependent manner on plexed With Cef1 (CWC22), an essential prerequisite for exon junc- – spliced mRNAs typically, but not exclusively, 20 24 nt upstream BIOCHEMISTRY tion complex assembly by the splicing machinery. The CWC22 MIF4G of exon-exon junctions (25–27). EJC assembly is a complex, domain binds both RecA domains of eIF4AIII. The mode of RecA2 stepwise process that is tightly coupled to the splicing reaction recognition is similar to that observed in the activating complexes, (28–30). Recent studies in human cells have shown that EJC yet is specific for eIF4AIII. The way the CWC22 MIF4G domain latches assembly requires the binding of eIF4AIII to the splicing factor on the eIF4AIII RecA1 domain is markedly different from activating Complexed With Cef1 (CWC22) (31–33). The interaction of complexes. In the CWC22–eIF4AIII complex, the RNA-binding and CWC22 with eIF4AIII is incompatible with the interaction with ATP-binding motifs of the two RecA domains do not face each other, as would be required in the active state, but are in diametrically MAGO and Y14 and is necessary for eIF4AIII recruitment to opposite positions. The binding mode of CWC22 to eIF4AIII reveals the splicing machinery (31, 32). Consistent with its role in EJC a facet of how MIF4G domains use their versatile structural frame- deposition, CWC22 participates to the degradation of an en- works to activate or inhibit DEAD-box proteins. dogenous NMD target in vivo (33). In vitro, CWC22 reduces the ATPase activity of eIF4AIII, suggesting that CWC22 keeps the helicase | mRNP | NMD DEAD-box protein in an inactive state before the EJC assembles at exon ligation (31). CWC22 contains two conserved regions EAD-box proteins are a large family of RNA-dependent fi DATPases involved in many aspects of RNA metabolism, Signi cance including processing, transport, translation, and decay (reviewed in refs. 1 and 2). These proteins generally function to remodel The fate of eukaryotic mRNAs is intimately linked to the com- ribonucleoprotein particles (RNPs), by locally unwinding the plement of proteins that associate with them to form mRNA— nucleic acid or by displacing and/or recruiting other factors to the protein complexes, the so-called messenger ribonucleoprotein nucleic acid they bind to (3–8). Although DEAD-box proteins particles (mRNPs). Transitions in the architecture of an mRNP fi recognize single-stranded RNAs in a sequence-independent man- lead to speci c functional consequences. DEAD-box proteins are ner in vitro, they act with exquisite specificity in vivo. As an example, key players in orchestrating these structural rearrangements: the translation initiation factor 4AI (eIF4AI), unwinds RNA sec- They associate with RNA in response to ATP binding and disso- ondary structure at the 5′ untranslated region (UTR) of mRNAs, ciate from it upon ATP hydrolysis. In this paper, we have eluci- and promotes the recruitment of the small ribosomal subunit (9– dated the molecular mechanisms by which a DEAD-box protein, 11). In contrast, the closely related paralogue eIF4AIII binds tightly which in human cells marks spliced mRNPs for a specialized on spliced mRNAs as part of the exon junction complex (EJC) (8). surveillance pathway, is recognized by the MIF4G domain of The EJC promotes nonsense-mediated mRNA decay (NMD) in a splicing factor. This structure shows how a MIF4G domain can human cells (12–14) and the localization of oskar mRNA in the act as a negative regulator of DEAD-box ATPase activity. Drosophila embryo (14, 15). Author contributions: G.B., H.L.H., and E.C. designed research; G.B., S.S., and C.B. per- DEAD-box proteins have a common architecture based on formed research; G.B., S.S., C.B., and E.C. analyzed data; and G.B. and E.C. wrote the two RecA domains connected by a flexible linker (reviewed in paper. ref. 3). A hallmark of these proteins is the conformational The authors declare no conflict of interest. plasticity with which they cycle between the active and inactive This article is a PNAS Direct Submission. states of the ATPase reaction (reviewed in ref. 4). In the active Data deposition: The atomic coordinates have been deposited in the Protein Data Bank, state, the two RecA domains adopt a characteristic closed con- www.pdb.org (PDB ID code 4C9B). formation that positions the residues responsible for ATP hy- 1To whom correspondence should be addressed. E-mail: [email protected]. – drolysis in the appropriate geometry for catalysis (16 21). In the This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. inactive state, the two domains are in more open configurations 1073/pnas.1314684110/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1314684110 PNAS Early Edition | 1of8 Downloaded by guest on October 2, 2021 predicted to fold into MIF4G and MA3 domains. Counterintu- proteins to the RecA2 domain of eIF4AIII and to a region of itively with respect to the known activating properties of MIF4G CWC22 containing the MIF4G domain (residues 110–409) (31, domains and inhibiting properties of MA3 domains on other 32). We purified different portions of the molecules and assessed DEAD-box proteins (21–24), CWC22 uses its MIF4G domain to their relative binding affinities quantitatively by using isothermal bind eIF4AIII (32). In this work, we have elucidated the mech- titration calorimetry (ITC). anism with which the MIF4G domain of CWC22 specifically An evolutionary conserved region of CWC22 spanning the recognizes and negatively regulates eIF4AIII. MIF4G and MA3 domains (residues 116–656) bound full-length (f.l.) eIF4AIII with a Kd of 67.5 nM (Fig. 1B and Fig. S1). The Results and Discussion MA3 domain (residues 450–665) showed no detectable binding to Domain Requirements for eIF4AIII–CWC22 Complex Formation. Hu- eIF4AIII (Fig. 1B and Fig. S1). The presence of the MA3 domain man eIF4AIII (residues 1–411) contains an N-terminal RecA1 appeared to even have a small negative contribution to eIF4AIII domain and a C-terminal RecA2 domain (17, 18) (Fig. 1A). binding: A segment lacking the entire MA3 domain (residues 116– – CWC22 (residues 1 656) contains MIF4G and MA3 domains 406) resulted in a Kd of 28.4 nM (Fig. 1 B and C). Remarkably, predicted approximately between residues 145–350 and 450–656, a further 20-residue truncation to the stable structural core of the respectively (using the Phyre (34) and HHpred prediction serv- MIF4G domain identified by limited proteolysis (residues 137– ers; ref. 35) (Fig. 1A). Previous coimmunoprecipitation and pull- 406) decreased binding to eIF4AIII 100-fold (Kd 3.1 μM; Fig. 1 B down experiments have mapped the interaction between the two and C). We concluded that elements both within and outside the Fig. 1. Domain requirements for eIF4AIII–CWC22 recognition. (A) Schematic domain organization of human eIF4AIII and CWC22. The RecA domains of eIF4AIII are in yellow. In the case of CWC22, the MIF4G fold is in blue and the N- and C-terminal extension are in cyan and red, respectively. Arrows highlight the regions of the molecules used for crystallization of the complex. The residue numbers of the domain boundaries are indicated (obtained either fromthe structural analysis in this manuscript or predicted in the case of the MA3 domain). (B) Table summarizing the results of the isothermal calorimetry experiments shown in the manuscript and in Fig. S1. The dissociation constants (Kd) with SDs were calculated with the program Origin. (C) Two representative ITC experiments, showing binding of eIF4AIII full length (f.l.) with different regions of CWC22.