Sapovirus translation requires an interaction between VPg and the cap binding protein eIF4E

Abstract

Sapoviruses of the Caliciviridae family of small RNA viruses are emerging pathogens that cause gastroenteritis in humans and animals. Molecular studies on human sapovirus have been hampered due to the lack of a cell culture system. In contrast, porcine sapovirus (PSaV) can be grown in cell culture, making it a suitable model for understanding the infectious cycle of sapoviruses and the related enteric caliciviruses. Caliciviruses are known to use a novel mechanism of protein synthesis that relies on the interaction of cellular translation initiation factors with the virus genome-encoded viral protein genome (VPg) protein, which is covalently linked to the 5' end of the viral genome. Using PSaV as a representative member of the Sapovirus genus, we characterized the role of the viral VPg protein in sapovirus translation. As observed for other caliciviruses, the PSaV genome was found to be covalently linked to VPg, and this linkage was required for the translation and the infectivity of viral RNA. The PSaV VPg protein was associated with the 4F subunit of the eukaryotic translation initiation factor (eIF4F) complex in infected cells and bound directly to the eIF4E protein. As has been previously demonstrated for feline calicivirus, a member of the Vesivirus genus, PSaV translation required eIF4E and the interaction between eIF4E and eIF4G. Overall, our study provides new insights into the novel mechanism of sapovirus translation, suggesting that sapovirus VPg can hijack the cellular translation initiation mechanism by recruiting the eIF4F complex through a direct eIF4E interaction.

Importance: Sapoviruses, which are members of the Caliciviridae family, are one of the causative agents of viral gastroenteritis in humans. However, human sapovirus remains noncultivable in cell culture, hampering the ability to characterize the virus infectious cycle. Here, we show that the VPg protein from porcine sapovirus, the only cultivatable sapovirus, is essential for viral translation and functions via a direct interaction with the cellular translation initiation factor eIF4E. This work provides new insights into the novel protein-primed mechanism of calicivirus VPg-dependent translation initiation.

FIG 1

The porcine sapovirus genome is linked to the fully processed VPg at its 5′ terminus. (A) A schematic presentation of the porcine sapovirus genome showing the nonstructural (NS1 to NS6-7) and structural (VP1 and VP2) proteins. sgRNA, subgenomic RNA. A, poly(A) tail in different lengths (n). (B) Monolayers of LLC-PK cells were either mock infected or infected with PSaV at an MOI of 10 TCID₅₀s/cell. After adsorption, the medium was replaced with EMEM containing GCDCA and FBS. Infection proceeded for 30 h, and then cell lysates and purified RNA were prepared. Purified RNAs were subjected to RNase treatment, and then the cell lysates and treated RNA were subsequently analyzed by Western blotting.

FIG 2

Porcine sapovirus requires VPg linkage to confer virus infectivity and viral RNA translation. (A) Monolayers of LLC-PK cells were either mock infected or PSaV infected at an MOI of 10 TCID₅₀s/cell. Cells were harvested at 12, 24, and 36 h postinfection and lysed, and the total RNA was extracted. RNA isolated from infected cells during the time course was used to program the in vitro translation reaction using RRLs. The translation products were analyzed by SDS-PAGE, and the translation profiles were then evaluated by autoradiography. An asterisk highlights the position of a cellular protein that is translated in a cap-dependent mechanism from a highly abundant cellular mRNA present in the RNA preparations. Lane Mo, RNA from mock-infected cells. (B) Proteins produced by in vitro translation were immunoprecipitated (IP) overnight using antibodies against the murine norovirus NS7 protein, PSaV VPg, and capsid proteins. Immunoprecipitates were washed 3 times with RIPA buffer and then resolved by 12.5% SDS-PAGE. Protein precipitates were analyzed by autoradiography. (C) RNA samples prepared from in vitro-transcribed and capped cap-Rluc-FMDV IRES-Fluc dicistronic RNA (Cap-Rluc/FMDV-Fluc), mock- and PSaV-infected cells, and capped in vitro transcripts from the PSaV full-length cDNA clone pCV4A (Cap-pCV4A) were pretreated with or without Pk at 37°C for 30 min. Following RNA purification, the purified RNA samples were analyzed by nondenaturing agarose gel electrophoresis to confirm their integrity. (D) RNA samples were then used to program RRLs, and the RNA was subjected to an in vitro translation reaction. (E) RNA extracted from mock- or proteinase K-treated, PSaV-infected cells was transfected into LLC-PK cells expressing BVDV Npro. Serial 10-fold dilutions of the RNA preparations were transfected, and at 4 days posttransfection the cells were washed and incubated in 1.3% Avicel-based overlay medium containing 2.5% FBS and 0.225% sodium bicarbonate and supplemented with 200 μM GCDCA. Cells were immediately fixed and stained.

FIG 3

The porcine sapovirus VPg binds to the cellular translation initiation factors during virus infection. (A) Monolayers of LLC-PK cells transduced with BVDV Npro were either mock infected or infected with PSaV at an MOI of 10 TCID₅₀s/cell. At 36 h postinfection, cells were collected and lysed in cap-Sepharose buffer. Lysates were centrifuged and further treated with RNase. One thousand micrograms of lysates was incubated with cap-Sepharose beads overnight. After they were washed, the bound proteins were analyzed by SDS-PAGE and Western blotting. (B) A His tag pulldown assay was performed using 10 μg of His-tagged VPg or BSA immobilized on HisPur cobalt resin. Increasing amounts (0 to 50 μg) of nuclease-treated cell lysates were then incubated overnight with the bait VPg protein. Protein complexes were extensively washed, eluted, and analyzed by Western blotting. eIF4GI, eIF4 gamma 1. (C) A capture ELISA was performed using 1 μg of purified recombinant PSaV VPg or BSA as a control. Ten micrograms of nuclease-treated cytoplasmic extracts was incubated with either target, and the extracts were extensively washed prior to detection with rabbit antibodies to eIF4E, eIF4A, or eIF4G. Antibody binding was detected using a secondary antirabbit HRP-conjugated antibody, followed by incubation with ELISA substrate. Samples were analyzed in triplicate and in at least three independent experiments. One representative set of data is shown. Error bars represent standard deviations for triplicate samples. OD, optical density.

FIG 4

Porcine sapovirus VPg binds to recombinant eIF4E. (A) A capture ELISA was performed using 1 μg of purified recombinant PSaV VPg or BSA as a control. Increasing concentrations of recombinant eIF4E were then incubated with both coated proteins, and eIF4E was extensively washed prior to detection with rabbit antibodies to eIF4E. Antibody binding was detected using a secondary antirabbit HRP-conjugated antibody, followed by incubation with ELISA substrate. Samples were analyzed in triplicate in at least three independent experiments. One representative set of data is shown. (B) A pulldown assay was performed using 1 μg of recombinant His-tagged VPg immobilized on HisPur cobalt resin or resin alone. Resins with and without VPg were washed, blocked with BSA, and incubated with a similar concentration of either BSA or recombinant eIF4E. Binding to eIF4E was detected by immunoblotting with anti-eIF4E antibody. Samples were analyzed in duplicate and in at least three independent experiments, and one representative set of data is shown.

FIG 5

Porcine sapovirus translation is independent of the cap analogue but requires the eIF4E-eIF4G interaction. (A) In vitro translation was performed, as illustrated by the experimental scheme above the gels, using either VPg-linked PSaV RNA or dicistronic RNA containing a cap-dependent CAT and PTV IRES-dependent Rluc. Translation reaction mixtures were preincubated with increasing concentrations of CAP analogue, and then the RNAs were added to initiate protein synthesis. The profiles for VPg-, cap-, and IRES-dependent translations were resolved by SDS-PAGE. The gels were fixed, dried, and exposed to X-ray film. The intensity of each band was quantitated with reference to the value obtained in the absence of the cap analogue. (B) In vitro translation was performed by following the experimental scheme indicated above the gels and with increasing amounts of recombinant 4E-BP1, before addition of PSaV RNA and dicistronic RNA. RNAs were then added to initiate protein synthesis. The profiles for VPg-, cap-, and IRES-dependent translations were resolved by SDS-PAGE. The gels were fixed, dried, and exposed to X-ray film. The intensity of each band was quantitated with reference samples incubated with 4E-BP1 buffer only. The asterisk highlights the position of a cellular protein that is translated by a cap-dependent mechanism from a highly abundant cellular mRNA present in the RNA preparations. (C) In vitro translation was performed as described in the legend to panel B with the addition of recombinant eIF4E after sequestration by 4E-BP1. The effects of separation by 4E-BP1 buffer or 4E-BP1 and subsequent complementation of eIF4E in cap-, VPg-, and IRES-dependent translation were evaluated by SDS-PAGE and autoradiography. In each panel, the numbers beneath the lanes indicate the quantitated protein synthesis in percentage for each corresponding in vitro translation reaction.

FIG 6

Porcine sapovirus translation requires full-length eIF4G. (A) In vitro translation was performed by following the experimental scheme indicated above each gel using increasing amounts of FMDV Lb protease (Lb Pro) to cleave eIF4G. Subsequently, elastatinal was added to quench the protease activity before the addition of PSaV RNA or in vitro-transcribed dicistronic RNA. (B) The effect of Lb protease was confirmed by Western blotting using antibody against eIF4G. (C) The profiles of VPg-, cap-, and IRES-dependent translation were resolved by SDS-PAGE. The gels were fixed, dried, and exposed to X-ray film. The intensity of each band was quantitated with reference samples incubated without Lb protease. The asterisk highlights the position of a cellular protein that is translated by a cap-dependent mechanism from a highly abundant cellular mRNA present in the RNA preparations.

FIG 7

Porcine sapovirus translation is sensitive to eIF4E depletion. Mock-depleted or eIF4E-depleted RRLs were used for the in vitro translation reaction in which the reaction mixture was replenished with recombinant eIF4E or buffer alone. (A) Depletion of eIF4E was verified by Western blotting using 0.25, 0.5, and 1.0 μl of the RRLs. (B) The translation of VPg-, cap-, and IRES-dependent proteins was observed in mock- and eIF4E-depleted lysates with and without the addition of recombinant eIF4E.

FIG 8

eIF4E is required for PSaV replication in the cell culture. (A) LLC-PK cells were transfected with either control or eIF4E siRNA, as described in the Materials and Methods. Reduced eIF4E expression was verified by Western blotting using antibody against eIF4E. (B) PSaV was then infected at an MOI of 0.2 TCID₅₀/cell. Cells were harvested at 24 h postinfection, and RNA was extracted for quantitative reverse transcription-PCR analysis specifically targeting the PSaV protease region. Samples were analyzed in triplicate and in at least three independent experiments. Error bars represent standard errors of the means for triplicate samples.