Identification in the ancient protist Giardia lamblia of two eukaryotic translation initiation factor 4E homologues with distinctive functions

Abstract

Eukaryotic translation initiation factor 4E (eIF4E) binds to the m(7)GTP of capped mRNAs and is an essential component of the translational machinery that recruits the 40S small ribosomal subunit. We describe here the identification and characterization of two eIF4E homologues in an ancient protist, Giardia lamblia. Using m(7)GTP-Sepharose affinity column chromatography, a specific binding protein was isolated and identified as Giardia eIF4E2. The other homologue, Giardia eIF4E1, bound only to the m(2,2,7)GpppN structure. Although neither homologue can rescue the function of yeast eIF4E, a knockdown of eIF4E2 mRNA in Giardia by a virus-based antisense ribozyme decreased translation, which was shown to use m(7)GpppN-capped mRNA as a template. Thus, eIF4E2 is likely the cap-binding protein in a translation initiation complex. The same knockdown approach indicated that eIF4E1 is not required for translation in Giardia. Immunofluorescence assays showed wide distribution of both homologues in the cytoplasm. But eIF4E1 was also found concentrated and colocalized with the m(2,2,7)GpppN cap, 16S-like rRNA, and fibrillarin in the nucleolus-like structure in the nucleus. eIF4E1 depletion from Giardia did not affect mRNA splicing, but the protein was bound to Giardia small nuclear RNAs D and H known to have an m(2,2,7)GpppN cap, thus suggesting a novel function not yet observed among other eIF4Es in eukaryotes.

FIG. 1.

Alignment and phylogeny of eIF4Es from Giardia and other organisms. (A) Alignment of Giardia eIF4Es with human eIF4E-1 and S. cerevisiae eIF4E. The predicted amino acid sequences of Giardia eIF4E1 and eIF4E2 were aligned with those of human eIF4E-1 and S. cerevisiae eIF4E using the CLUSTAL W algorithm (47). Identical residues are shaded in dark gray, and related residues are in light gray. Residues with dots underneath are essential for cap binding, whereas triangles indicate residues essential for interaction with eIF4G in humans and yeast (27). The sequences predicted to be the bipartite nucleus localization signal in Giardia eIF4E1 is underlined. (B) Phylogeny of eIF4Es. The selected eIF4Es were analyzed using unweighted-pair group method using average linkages distance matrix methods from the MacVector software. The bootstrap tree was shown with 1,000 distance replicates performed. Abbreviated sequence identifiers are as follows: At, A. thaliana; Ce, C. elegans; Gl, G. lamblia; Hs, Homo sapiens; Mm, Mus musculus; Sc, S. cerevisiae; Sp, S. pombe; Xl, Xenopus laevis.

FIG. 2.

Giardia eIF4Es are incapable of complementing the function of S. cerevisiae eIF4E. S. cerevisiae yTHC strain YSC1180-7428777 was transformed with the Leu-selectable vector pGADT7′ containing cDNAs encoding yeast eIF4E, Giardia eIF4E1, and Giardia eIF4E2, respectively. Following selection on the −Leu plate under G418 selection, cells from individual colonies were serially diluted and applied to synthetic complete medium plates lacking leucine but containing doxycycline.

FIG. 3.

Purification of Giardia eIF4E2 by m⁷GTP-Sepharose affinity chromatography and identification by Western blotting. (A) Cell lysate from 2 liters of Giardia cell culture was applied to an m⁷GTP-Sepharose column and eluted. Proteins in collected fractions (lanes: 1, total protein; 2, flowthrough; 3, column wash; 4, column washed with GTP; 5, column eluted with m⁷GTP) were separated by SDS-PAGE and visualized by E-zinc staining. (B) The eluted fraction from an m⁷GTP-Sepharose affinity column separated by SDS-PAGE (Fig. 2A, lane 5) was transferred onto a polyvinylidene difluoride membrane and subjected to immunoblotting with purified rabbit polyclonal antibodies against Giardia eIF4E1 and eIF4E2, respectively. Purified recombinant Giardia eIF4Es (see Fig. S1 in the supplemental material) were included in various quantities as a control.

FIG. 4.

Binding of Giardia eIF4E1 and eIF4E2 to different cap structures. (A) HA-tagged eIF4E1 eluted through a m⁷GTP-Sepharose column was analyzed by immunoblotting (Fig. 2B), and the result demonstrated no detectable binding between the protein and the cap. (B) c-myc-tagged eIF4E2 applied to an m⁷GTP-Sepharose column showed that it was bound to the cap structure. (C) c-myc-tagged eIF4E1 was mixed with a biotinylated RNA species capped with m^2,2,7GpppG and subjected to a pulldown experiment with streptavidin-agarose beads. The results from the subsequent immunoblotting indicated that the protein was pulled down with the beads and was released only by washing with m^2,2,7GpppG. (D) In a similar experiment, c-myc-tagged eIF4E1 was not brought down by the m⁷GpppG-capped RNA.

FIG. 5.

Effects on cell growth and cap-dependent translation of knocking down the eIF4Es in Giardia. Giardia WBI trophozoites transfected with in vitro transcripts from the empty vector pC631neo (A), the Giardia eIF4E1 antisense-ribozyme construct pC631neo4E1RZ (B), and the Giardia eIF4E2 antisense-ribozyme construct pC631neo4E2RZ (C) were incubated at 37°C in culture medium containing G418. Cell growth was monitored daily for up to 5 days. The insets show the intracellular mRNA levels estimated by semiquantitative RT-PCR after 1 day of incubation. RT-PCR quantitation of NADH oxidase mRNA (NADHox) was included as a control. (D) A luciferase transcript capped with m⁷GpppG was electroporated into the ribozyme-transfected cells mentioned above. The luciferase activities thus expressed were monitored 4 h thereafter and are presented with the activity in pC631neo-transfected cells designated as 100%. RLU, relative light units.

FIG. 6.

Functional analysis of Giardia eIF4E1. (A) Firefly luciferase in vitro transcripts with different cap structures were used to transfect Giardia trophozoites. The luciferase activity thus expressed was assayed 4 h after electroporation. Each value was derived from three independent electroporation experiments. (B) Total RNA was extracted from each of the knockdown cell lines described in the legend to Fig. 5 1 day after incubation in the presence of 4 mg/ml G418. These RNA samples were used as templates for semiquantitative RT-PCR to monitor the pre-mRNA and mature mRNA of 2Fe-2S ferredoxin in the eIF4E-depleted cells. (C) Giardia total RNA was incubated with the GST-eIF4E1 fusion protein and pulled down by glutathione-Sepharose 4B beads. Results from RT-PCR amplification indicate pulldown of snRNA D and snRNA H by GST-eIF4E1, whereas the uncapped snRNA J is not.

FIG. 7.

Immunofluorescence localization of eIF4Es in Giardia. Giardia WB cells stained with DAPI were further stained with rabbit polyclonal antibodies against Giardia eIF4E1 (top panel) and eIF4E2 (bottom panel). The stained cells were examined with a fluorescence microscope. Arrows point to concentrated localizations of the protein.

FIG. 8.

Colocalization of Giardia eIF4E1 with m^2,2,7GpppN cap, fibrillarin, and 16S-like rRNA. (A) Giardia WB cells stained with rabbit polyclonal antibody against eIF4E1, mouse monoclonal antibody against m^2,2,7GpppN cap (top), and mouse monoclonal antibody against yeast fibrillarin (bottom). (B) Giardia WB trophozoites were fixed and stained with an FITC-conjugated 30-mer oligonucleotide complementary to the 3′ end of the Giardia 16S-like rRNA. Concentrated stains were visible in the nucleolus-like structures at the anterior ends of both nuclei.