Comparative structural analysis of human DEAD-box RNA helicases

Abstract

DEAD-box RNA helicases play various, often critical, roles in all processes where RNAs are involved. Members of this family of proteins are linked to human disease, including cancer and viral infections. DEAD-box proteins contain two conserved domains that both contribute to RNA and ATP binding. Despite recent advances the molecular details of how these enzymes convert chemical energy into RNA remodeling is unknown. We present crystal structures of the isolated DEAD-domains of human DDX2A/eIF4A1, DDX2B/eIF4A2, DDX5, DDX10/DBP4, DDX18/myc-regulated DEAD-box protein, DDX20, DDX47, DDX52/ROK1, and DDX53/CAGE, and of the helicase domains of DDX25 and DDX41. Together with prior knowledge this enables a family-wide comparative structural analysis. We propose a general mechanism for opening of the RNA binding site. This analysis also provides insights into the diversity of DExD/H- proteins, with implications for understanding the functions of individual family members.

Figure 1. Crystal structures of DEAD-box conserved domains-1 and -2.

(A) Superposition of the DEAD-domains of DDX2A (green), DDX2B (brown), DDX5 (red), DDX10 (turquoise), DDX18 (grey), DDX47 (dark blue), DDX52 (yellow), and DDX53 (dark yellow). The positions of conserved motifs I–III (black) are indicated. (B) Superposition of the helicase domains of DDX19 (light blue), DDX25 (grey) and DDX41 (orange). The positions of conserved motifs IV–VI (black) are indicated. (C) Cartoon representations of the DDX5 helicase domain in the same orientations as in the following two panels. (D) Conserved surface patches (green), projected onto the DDX47 DEAD-domain surface. (E) Electrostatic surface representation of DEAD-domains. Negative charges are shown in red and positive charges in blue. (F) Cartoon representations of the DDX41 helicase domain in the same orientation as in the following two panels. The RNA and AMPPNP (sticks representation) of the superposed DDX19 structure mark the RNA and nucleotide binding sites. (G) Conserved surface patches (green), projected onto the DDX25 helicase-domain surface. (H) Electrostatic surface representation of helicase domains.

Figure 2. Sequence alignments of the two RecA-like domains of the DEAD-box proteins described in this study.

Conserved sequence motifs are indicated. Secondary structural elements are given for DDX19 (PDB entry 3G0H) above the alignment. Asterisks mark the terminal aspartate of the DEAD motif and the arginine of motif V, the interaction of which is central to positioning of α-helix 8 (see also Figure 5C, D). Sequences shown are human DDX19B (gene accession number: 13177688); DDX10 (13514831); DDX18 (38327634); DDX20 (23270929); DDX25 (29792166); DDX41 (21071032); DDX47 (45786091); DDX5 (16359122); DDX52 (27697141); DDX53 (45709415); eIF4A1/DDX2A (16307020); and eIF4A2/DDX2B (45645183).

Figure 3. Details of the ATP binding sites.

(A) Superposition of multiple DEAD-domains to illustrate variability in P-loop (Motif I) conformations. P-loops in DEAD-domain structures with bound phosphate (yellow), with bound AMP (orange), with bound ADP (red), DDX19 P-loop with bound AMPPNP and Mg²⁺ (blue), DDX20 P-loop with bound AMPPNP (magenta), and P-loop in nucleotide-free eIF4A/DDX2A (green) are shown. Motifs I, II and III are indicated. (B) Two different conformations of the β- and γ-phosphates in the DDX20-AMPPNP complex. Side chains that interact with the AMPPNP are shown as balls-and-sticks. (C) DDX2B with a closed P-loop. The α-helix that follows the P-loop starts one turn earlier compared to other DEAD-domain structures shown. (D) Variability of interactions with the adenosine nucleotide. The adenosine moiety is coordinated through π-stacking interactions or hydrophobic interactions. Numbers denote the interaction surface, in Ų, between the nucleotide and the stacking side chain, as determined using the PISA server .

Figure 4. RNA binding cleft on DEAD domains.

(A) DDX19 (light blue; PDB entry 3G0H) with bound RNA (light orange). RNA-interacting side chains are shown. (B) Flexible regions in DDX2B, DDX10 and DDX53 for which the electron density was not visible. (C) Sequence conservation in the RNA binding cleft, mapped onto the DDX47 structure (red, conserved; orange, partly conserved). (D) RNA binding sites of selected DEAD-domains to illustrate their sequence variation.

Figure 5. Details of the RNA binding cleft.

(A) DDX19 closed state structure (PDB entry 3G0H). DDX19-bound RNA, Mg²⁺-ion and AMPPNP are in orange. (B) Superposition of several DEAD domain structures showing a conserved conformation of α-helix 8. (C) Interactions between the DEAD and helicase domains of DDX19. (D) “Top-down” view of the open and closed RNA binding cleft. DDX5 (red), the ATP-state of DDX19 (blue) and DDX41 (orange) are shown. RNA (superposed from the DDX19 complex structure) is shown in light orange. (E) Surface representation of the DDX19-RNA complex. Note that α-helix 8 does not come in contact with the RNA substrate. (F) Surface representation of DDX5 and the superposed RNA from the DDX19 complex structure. Note that α-helix 8 would clash with the RNA substrate.

Figure 6. Schematic model for the regulation of RNA binding by α-helix 8 of DEAD-box helicases.

(A) In the isolated domains, reflecting the open and substrate free states, the RNA binding sites are partially blocked by α-helix 8 in the DEAD-domain and motif V in the helicase domain. The aspartate indicated in the DEAD-domain is the second D of the DEAD sequence in motif II. The arginine indicated in the helicase domain is a conserved residue in motif V. Both residues are marked by asterisks in Fig 2. (B) Binding of ATP favors closure of the cleft, facilitating interaction of α-helix 8 with motif V across the cleft, thereby removing the blockage of the RNA binding site. (C) The closed cleft conformation is stabilized by RNA substrate to the competent site, allowing ATP hydrolysis to proceed.