Selective pharmacological targeting of a DEAD box RNA helicase - PubMed (original) (raw)

. 2008 Feb 13;3(2):e1583.

doi: 10.1371/journal.pone.0001583.

Monika Oberer, Mikhail Reibarkh, Regina Cencic, Marie-Eve Bordeleau, Emily Vogt, Assen Marintchev, Junichi Tanaka, Francois Fagotto, Michael Altmann, Gerhard Wagner, Jerry Pelletier

Affiliations

Selective pharmacological targeting of a DEAD box RNA helicase

Lisa Lindqvist et al. PLoS One. 2008.

Abstract

RNA helicases represent a large family of proteins implicated in many biological processes including ribosome biogenesis, splicing, translation and mRNA degradation. However, these proteins have little substrate specificity, making inhibition of selected helicases a challenging problem. The prototypical DEAD box RNA helicase, eIF4A, works in conjunction with other translation factors to prepare mRNA templates for ribosome recruitment during translation initiation. Herein, we provide insight into the selectivity of a small molecule inhibitor of eIF4A, hippuristanol. This coral-derived natural product binds to amino acids adjacent to, and overlapping with, two conserved motifs present in the carboxy-terminal domain of eIF4A. Mutagenesis of amino acids within this region allowed us to alter the hippuristanol-sensitivity of eIF4A and undertake structure/function studies. Our results provide an understanding into how selective targeting of RNA helicases for pharmacological intervention can be achieved.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1

Figure 1. Hippuristanol binds to eIF4AI-CTD.

(A) Chemical shift changes of 1H-15N-HSQC peaks, (Δδ(1H)+0.2 Δδ(15N), of eIF4A-I-CTD (52 µM) upon addition of hippuristanol (100 µM). Free and bound forms are in slow exchange and the resonances of eIF4AI-CTD had to be assigned in both states. The locations of secondary structures were identified by NMR and are indicated with magenta arrows (β-strands) and yellow rectangles (helices). (B) Primary amino acid sequence of eIF4AI indicating residues involved in hippuristanol binding. NOEs are highlighted in yellow, whereas those within 5Å are in grey and correspond to regions a, b, and c in A. Residues in bold denote conserved amino acids that define motifs V (ARGID) and VI (HRIGRGGRFG) of DEAD box family members . Arrows denote residues identified in Vasa that interact with ATP (red), RNA (blue), or are involved in interdomain interaction (green). (C) Surface and ribbon representations of the model for eIF4AI-CTD. The CTD is viewed from the position of the NTD. Residues of eIF4AI-CTD that show NOEs to hippuristanol are coloured yellow, those exhibiting major chemical shift changes but no NOEs are coloured blue. Residues contacting eIF4G are in red . The β-sheets (E1–E6) and α-helices (H1–H6) are labelled and refer to the locations marked in A. RNA and ADPNP are shown as sticks models. (D) Location of the hippuristanol-binding site in a model for eIF4AI complexed with RNA and ADPNP. The model is composed of the crystal structure of human eIF4AI-NTD (PDB #2G9N) and the homology model of the eIF4AI-CTD . The two domains are aligned to the structure of eIF4AIII from the EJC from which the RNA and ADPNP binding sites are adapted (PDB# 2HYI) . Color scheme of amino acid residues is as in C.

Figure 2

Figure 2. Selectivity of hippuristanol for eIF4A.

(A) Inhibition of eIF4A RNA-dependent ATPase activity by hippuristanol. ATPase assays were performed with 0.1 µg of His6-eIF4AI or His6-eIF4AII at 25°C or with 0.1 µg of His6-eIF4AIII at 37°C for 2 h with 0.1 µCi γ-32P-ATP (10 Ci/mmol). Following analysis by TLC and quantitation using a Fuji BAS 2000 phosphoimager, the percent hydrolysis was determined and set relative to the DMSO vehicle control reactions. Each value represents the average of three measurements with the error shown as the standard deviation. (B) Crosslinking of recombinant proteins to RNA in the presence of hippuristanol. 32P-labelled CAT RNA was cross-linked to 0.5–1 µg of the indicated recombinant protein in the presence or absence of hippuristanol, separated by SDS-PAGE, and visualized by autoradiography. [Note that in our hands, recombinant hDDX52 did not crosslink to RNA.] (C) Relative ATPase activity of eIF4AI, hDDX19, and hDDX52 in the presence of 50 µM hippuristanol. eIF4AI and hDDX19 where performed at 25°C for 5 minutes while hDDX52 was incubated for 60 minutes to allow for analysis to be in the linear range of ATP hydrolysis. The percent ATP hydrolysis was determined in the presence of hippuristanol and set relative to the DMSO vehicle control reactions. The results represent the average of 3 experiments with error bars signifying the standard deviation.

Figure 3

Figure 3. Characterization of eIF4A hippuristanol-resistant mutants.

(A) Consequence of mutations in the eIF4AI hippuristanol binding site on ATP hydrolysis. ATP hydrolysis was monitored using 1 µg His6-eIF4AI or His6-eIF4AIIG/T in the presence or absence of 10 µM hippuristanol. Each value represents the average of three measurements with the standard deviation presented. (B) Relative ATPase activity of eIF4A mutants in the presence of hippuristanol. The percent ATP hydrolysis was determined in the presence of hippuristanol and set relative to the values obtained in the presence of control reactions containing vehicle (DMSO). The results represent the average of 3–7 experiments with error bars signifying the standard deviation. (C) Altered hippuristanol sensitivity of eIF4AIII. ATPase assays were performed with 0.5 µg recombinant protein with 0.1 µCi γ-32P-ATP (10 Ci/mmol). Following analysis by TLC and quantitation using a Fuji BAS 2000 phosphoimager, the percent hydrolysis was determined and set relative to the DMSO vehicle control reactions. Each value represents the average of three measurements with the standard deviation shown.

Figure 4

Figure 4. Functional requirements for eIF4A activity in translation.

(A) Schematic representation of the reporter construct used in these studies is shown on top. (B) Left Panel: Rescue of hippuristanol-induced translation inhibition by eIF4AIIG/T. In vitro translations in RRL programmed with capped FF/HCV/Ren mRNA (8 µg/ml) and containing vehicle (0.1% DMSO) or 5 µM hippuristanol (in 0.1% DMSO) were supplemented with 0.5 µg recombinant protein. Protein synthesis was assessed by using 35S-methionine incorporation as well as by monitoring luciferase assays. Protein products were separated by SDS-PAGE and visualized by autoradiography. The arrow indicates the position of migration of the firefly luciferase, whereas the arrowhead denotes the position of migration of Renilla luciferase. Right panel: Relative luciferase activity obtained in the presence of recombinant eIF4A. Firefly RLU readings obtained in the presence of recombinant eIF4A and hippuristanol were standardized to Renilla RLU values and set relative to the values obtained in the presence of vehicle (DMSO). The average of 3–8 experiments is shown with the standard deviations denoted. (C) eIF4AI and eIF4AII are functionally interchangeable. In vitro translations in RRL containing vehicle (DMSO) or 5 µM hippuristanol were supplemented with 0.8 µg recombinant eIF4A where indicated, and programmed with capped FF/HCV/Ren mRNA (8 µg/ml). Left panel: Protein products were separated by SDS-PAGE and visualized by autoradiography. The arrow indicates the position of migration of the firefly luciferase, whereas the arrowhead denotes the position of migration of Renilla luciferase. Right panel: Relative luciferase activity obtained in the presence of recombinant eIF4A and hippuristanol was standardized to Renilla Luciferase levels and set relative to the values obtained in the presence of vehicle (DMSO). The average of 4–5 experiments is shown with the standard deviations denoted. (D) Translational rescue by eIF4AIIG/T is selective for hippuristanol. In vitro translations in RRL containing vehicle (DMSO), 5 µM hippuristanaol, or 0.4 µM pateamine were supplemented with 0.8 µg recombinant eIF4A where indicated, and programmed with capped FF/HCV/Ren mRNA (8 µg/ml). Protein products were separated by SDS-PAGE and visualized by autoradiography. The arrow indicates the position of migration of the firefly luciferase, whereas the arrowhead denotes the position of migration of Renilla luciferase. The figure is a representative display of one of two experiments.

Figure 5

Figure 5. Hippuristanol targets eIF4A in vivo.

(A) Inhibition of Renilla luciferase reporter in a yeast in vitro translation extract. Hippuristanol was added to a S. cerevisiae cytosolic translation extract programmed with 0.12 µg/ml capped Renilla luciferase mRNA. At various points following initiation of the translation reaction, aliquots were removed and the relative luciferase units (RLU) determined. (B) Haploinsufficiency for Tif1/2p leads to increased sensitivity to hippuristanol in vivo. Haploid wild type cells (strain CWO4) or an isogenic strain carrying the temperature-sensitive tif1V79A allele (strain SS13-3A/pSSC120) were cultivated in rich medium (YPD) at 27°C to an O.D.600 of 0.2, at which point hippuristanol (1 µM or 10 µM final concentration) or solvent (DMSO, 0.1%) was added and the growth of different cultures was monitored for several hours by measuring the O.D.600. (C) Serial dilutions of different haploid yeast strains were plated on YPD-plates containing the indicated concentrations of hippuristanol and incubated for 2–3 days at 27°C: wt (CWO4), wild type strain CWO4; 4A-ts, strain SS13-3A/pSSC120 carrying the tif1V79A allele; Δtif1, a BY4741-derivative strain carrying a tif1::kanX deletion; Δtif2, a BY4741-derivative strain carrying a tif2::kanX deletion; wt (BY4741), wild type strain BY4741.

Figure 6

Figure 6. In vivo rescue of hippuristanol-induced translation inhibition by eIF4AIIG/T.

(A) Rescue of translation in Xenopus oocytes by eIF4AIIG/T. The percent rescue was determined by normalizing the Firefly luciferase values to Renilla luciferase [to standardize for small variations in sample injection volumes], followed by dividing by the ratio obtained from the vehicle-treated samples (which was set at 100%). The data presented is the average of 9 independent sets of injections with the standard deviations denoted. (B) Western blot of extracts prepared from oocyte extracts. The equivalent of one oocyte was separated on a 10% SDS-PAGE, transferred to Immobilon-P, and probed with α-His6 (to detect recombinant His6-eIF4AI) or α-tubulin antibodies.

References

    1. Singleton MR, Dillingham MS, wigley DB. Structure and mechanism of helicases and nucleic acid translocases. Annual Review of Biochemistry. 2007;76:23–50. - PubMed
    1. Cordin O, Banroques J, Tanner NK, Linder P. The DEAD-box protein family of RNA helicases. Gene. 2006;367:17–37. - PubMed
    1. Linder P. Dead-box proteins: a family affair–active and passive players in RNP-remodeling. Nucleic Acids Res. 2006;34:4168–4180. - PMC - PubMed
    1. Fukuda T, Yamagata K, Fujiyama S, Matsumoto T, Koshida I, et al. DEAD-box RNA helicase subunits of the Drosha complex are required for processing of rRNA and a subset of microRNAs. Nat Cell Biol. 2007;9:604–611. - PubMed
    1. Grifo JA, Tahara SM, Leis JP, Morgan MA, Shatkin AJ, et al. Characterization of eukaryotic initiation factor 4A, a protein involved in ATP-dependent binding of globin mRNA. J Biol Chem. 1982;257:5246–5252. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources