DHV-1 IRES activity is modulated by downstream coding sequences
To further characterize the presence of accessory regulatory sequences downstream of the 5′-UTR of the viral genome, 4 additional dicistronic reporter plasmids were constructed, containing DHV-1 5′-UTR and 10, 40 or 60 nucleotides of the coding region adjacent to the viral 5′-UTR (DHV-1 UTR+10nt, UTR+40nt, UTR+60nt, respectively). Initiating AUG codon of the DHV-1 IRES was mutated to allow the expression of the luciferase from its own AUG. The dicistronic plasmids were in vitro transcribed and individually added to Flexi rabbit reticulocyte lysate (RRL) system (Promega). The translation of the reporter proteins was evaluated measuring 35S-methionine incorporation. Cap-dependent translation from the first ORF was determined by the level of CAT expression and internal initiation activity of the different sequences included between the two reporter genes was estimated by the accumulation of fLUC. All the constructs expressed CAT efficiently as expected but the efficiency of the DHV-1 UTR+10nt, UTR+40nt and UTR+60nt sequences was stronger than that of the DHV-1 UTR (Figure 5A). Moreover, the dicistronic plasmids were transfected into vFT7-infected BHK-21 cells and, in order to evaluate the in vivo internal initiation activity, the accumulation of fLUC protein expressed from the different constructs was determined and corrected by CAT expression levels. The experiments were performed by triplicate. As observed in the in vitro experiments, the translation initiation efficiency of DHV-1 UTR+10nt, UTR+40nt and UTR+60nt sequences was stronger than that of the DHV-1 UTR. These results indicate the presence of accessory regulatory sequences downstream the DHV-1 putative IRES, which are not required for translation initiation but positively modulate the IRES activity.
Characterization of the putative DHV-1 IRES requirement for eIF4F complex components
Translation initiation of DHV-1 IRES is not inhibited by 4E-BP1
Picornavirus IRES-mediated translation is independent of the cellular cap-binding protein eIF4E. In order to evaluate the putative DHV-1 IRES requirement for the eIF4E, the in vitro translation efficiency of the dicistronic constructs was evaluated in RRL as described upon addition of increasing amounts of purified recombinant 4E-BP1 protein. The eIF4E-binding proteins (4E-BPs) are a family of three small polypeptides that inhibit cap-dependent translation by binding to the eIF4E, obstructing its interaction with eIF4G. However, the internal translation initiation on IRES elements is not affected in these conditions. As shown in Figure 6A, whereas the recombinant 4E-BP1 efficiently inhibited the cap-dependent CAT expression in RRL, the EMCV IRES-directed and DHV-1 5′-UTR-directed translation were unaffected.
The DHV-1 IRES does not require intact eIF4G
According to the prediction of the 5′-UTR secondary structure, DHV-1 could be included among the type IV picornaviruses. Previous studies have shown that the type IV IRES elements (HCV-like IRESes) do not require the eIF4G activity to direct translation initiation and, consequently, that are insensitive to the eIF4G cleavage induced by different viral proteases, such as the foot-and-mouth disease virus (FMDV) L protease. The FMDV L protease efficiently cleaves the eukaryotic eIF4G resulting in a partial inhibition of the cap-dependent translation while the IRES-directed translation is not inhibited in these conditions and it can even be enhanced. To confirm that the translation of DHV-1 5′-UTR is cap-independent and to examine its eIF4G dependence, the effect of foot-and-mouth disease virus (FMDV) L protease addition in the RRL system was evaluated. The dicistronic RNA constructs indicated in Figure 7A were individually added to FMDV-L treated or control Flexi RRL and the translation of the reporter proteins was evaluated. Whereas the cap-dependent CAT expression was efficiently inhibited, the EMCV-directed and the DHV-1 5′-UTR-directed translation were insensitive to FMDV L protease addition. Western blot analysis was performed to control the efficiency of the eIF4G cleavage by FMDV L protease. These results indicate that the DHV IRES directs cap-independent internal initiation of translation and not requires an intact full-length eIF4G.