The HRIGRXXR region of the DEAD box RNA helicase eukaryotic translation initiation factor 4A is required for RNA binding and ATP hydrolysis.

Abstract

eIF-4A is a eukaryotic translation initiation factor that is required for mRNA binding to ribosomes. It exhibits single-stranded RNA-dependent ATPase activity, and in combination with a second initiation factor, eEF-4B, it exhibits duplex RNA helicase activity. eEF-4A is the prototype of a large family of proteins termed the DEAD box protein family, whose members share nine highly conserved amino acid regions. The functions of several of these conserved regions in eIF-4A have previously been assigned to ATP binding, ATPase, and helicase activities. To define the RNA-binding region of elIF-4A, a UV-induced cross-linking assay was used to analyze binding of mutant eIF-4A proteins to RNA. Mutants carrying mutations in the ATP-binding region (AXXXXGKT), ATPase region (DEAD), helicase region (SAT), and the most carboxy-terminal conserved region of the DEAD family, HRIGRXXR, were tested for RNA cross-linking. We show that mutations, either conservative or not, in any one of the three arginines in the HRIGRXXR sequence drastically reduce eIF-4A cross-linking to RNA. In addition, all the mutations in the HRIGRXXR region abrogate RNA helicase activity.Some but not all of these mutations affect ATP binding and ATPase activity. This is consistent with the hypothesis that the HRIGRXXR region is involved in the ATP hydrolysis reaction and would explain the coupling of ATPase and RNA-binding/helicase activities. Our results show that the HRIGRXXR region, which is QRXGRXXR or QXXGRXXR in the RNA and DNA helicases of the helicase superfamily II, is involved in ATP hydrolysis-dependent RNA interaction during unwinding. We also show that mutations in other regions of eIF-4A that abolish ATPase activity sharply decrease eIF-4A cross-linking to RNA. A model is proposed in which eIF-4A first binds ATP, resulting in a change in eIF-4A conformation which allows RNA binding that is dependent on the HRIGRXXR region. Binding of RNA induces ATP hydrolysis, leading to a more stable interaction with RNA. This process is then linked to unwinding of duplex RNA in the presence of eIF-4B.Binding of 40S ribosomes to mRNA in eukaryotes is an intricate process that requires the participation of at least three eukaryotic initiation factors, eIF-4A, eIF-4B, and eIF-4F, and the hydrolysis of ATP (for reviews, see references 35, 45, and 53). eIF-4A is a 50-kDa polypeptide that exhibits single-stranded RNA-dependent ATPase activity and, in combination with eIF-4B, duplex RNA helicase activity (17,44,46). There are two eIF-4A gene products, eIF-4AI and eIF-4AII, that are almost identical throughout the coding sequence except for the amino-terminal 20 amino acids (39). Both forms exchange into the eIF-4F complex (60). eIF-4F is a three-subunit complex (12, 19, 56) comprising (i) eIF-4E, a 24-kDa polypeptide that contains the cap-binding site (54); (ii) eIF-4A, which is a mixture of eIF-4AI and eIF-4AII (4:1) (9); and (iii) p220, a 220-kDa polypeptide whose function is not clear but whose integrity is apparently important for eIF-4F function (13). It is t...

The mechanism of ATP-dependent RNA unwinding by DEAD box proteins

“…The RNA binding site of the helicase core is formed by both RecA-like domains, and binding involves contacts to motifs Ia, GG, and Ib in the N-terminal domain, and to motifs IV, QxxR, V, and VI in the C-terminal domain (Pause et al, 1993;Andersen et al, 2006;Bono et al, 2006;Sengoku et al, 2006;Nielsen et al, 2008;Collins et al, 2009;von Moeller et al, 2009) (Figures 1 and 3B). In all available structures, the N-terminal RecA-like domain binds the 39-region of the ssRNA substrate, and the C-terminal domain binds the 59-region.…”

“…The binding sites can be independent, or coupled to each other via positive or negative cooperativity. Early experiments suggested sequential binding of ATP and RNA (Abramson et al, 1987;Pause et al, 1993). This conclusion resulted from the low RNA affinity of the helicase core in the absence of nucleotides, making RNA binding difficult to detect.…”

“…These experiments are not performed under equilibrium conditions, but nevertheless they have been very valuable for comparing relative affinities of one DEAD box protein for different nucleotides, or for comparing nucleotide or RNA affinities of different DEAD box protein mutants. Nucleotide binding constants have also been derived from 'equilibrium filtration' experiments (Polach and Uhlenbeck, 2002), and RNA binding constants have been determined in electrophoretic mobility shift assays (Pause et al, 1993;Henn et al, 2001;Polach and Uhlenbeck, 2002;Cordin et al, 2004;Karginov et al, 2005;Talavera et al, 2006;Banroques et al, 2008;Liu et al, 2008). In these experiments, the equilibrium is only minimally perturbed.…”

The DEAD Box Protein eIF4A. 2. A Cycle of Nucleotide and RNA-Dependent Conformational Changes

“…It is the archetypal member of the DEAD box family of proteins, as the other members of this family are made up of core eIF4A-like domains flanked by N-and C-terminal extensions (5). eIF4A can unwind small duplex RNAs in vitro in conjunction with another factor, eIF4B (2,(6)(7)(8)(9). These and other observations have led to the proposal that the DEAD box proteins function as RNA helicases (2,6,8,(10)(11)(12)(13).…”

“…It was demonstrated that the enzyme's affinity for RNA is modulated by the presence or absence of the γ-phosphate on the bound nucleotide. In addition, UV-induced RNA cross-linking experiments suggested that the presence or absence of the γ-phosphate also alters the conformation of the enzyme (9,14).To further explore ligand-induced conformational changes in eIF4A, limited proteolysis experiments were performed. These experiments provide evidence that distinct conformations of the enzyme are stabilized upon binding of the various combinations of substrates and products.…”