Previous Article | Next Article ![]()
Journal of Virology, February 2008, p. 1993-2003, Vol. 82, No. 4
0022-538X/08/$08.00+0 doi:10.1128/JVI.01957-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Elisabetta Groppelli,1,
Margaret M. Willcocks,1
Elizabeth Royall,1
Graham J. Belsham,2 and
Lisa O. Roberts1*
Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey GU2 7XH, United Kingdom,1 The National Veterinary Institute, Technical University of Denmark, Lindholm, DK-4771 Kalvehave, Denmark2
Received 6 September 2007/ Accepted 26 November 2007
|
|
|---|
|
|
|---|
The picornavirus IRES elements are divided into several groups which display distinct secondary structures and biological properties. One group (class I) contains IRES elements from the entero- and rhinoviruses (e.g., poliovirus [PV]), while the second group contains the cardio- and aphthovirus IRES elements (e.g., encephalomyocarditis virus [EMCV]). The cardio-/aphthovirus IRES elements function efficiently in the rabbit reticulocyte lysate (RRL) translation system. However, the PV and rhinovirus IRES elements are inefficient in this system unless the reaction mixture is supplemented with additional proteins, e.g., from HeLa cell extracts (6, 10). The IRES element from hepatitis A virus (HAV) represents a third type of IRES. It is distinct from other picornavirus IRES elements in that it requires an intact eIF4F complex, including eIF4E, for function (1, 4). In contrast, the class I and II picornavirus IRES elements can function efficiently when eIF4G has been cleaved by the expression of an entero-/rhinovirus 2A or aphthovirus L protease (5, 34). This cleavage releases the N terminus of eIF4G including its eIF4E binding site (reviewed in reference 20). The initiation factor eIF4A has also been shown to be required by group I and II IRES elements since dominant-negative mutants of this protein and inhibitors of eIF4A block their activity (9, 26, 38).
Recently, a new group of picornavirus IRES elements has been identified. This group includes the IRES elements from porcine teschovirus 1 (PTV-1) Talfan strain (8, 17, 29), simian virus 2 (SV2), and porcine enterovirus 8 (PEV-8) (9). Strikingly, these IRES elements have many similarities to those from hepatitis C virus (HCV) and classical swine fever virus (CSFV), which both belong to the Flaviviridae family. These recently characterized picornavirus IRES elements are predicted to share a very similar structure to the HCV-type elements, including a pseudoknot near the 3' end of the IRES that is critical for function (12, 18, 40). These IRES elements are generally shorter than other picornavirus elements, e.g., about 280 nt in the case of the PTV-1 IRES (8). They can also function with cleaved eIF4G (9, 29) and, unlike the group I and II IRES elements, are resistant to both dominant-negative mutants of eIF4A (8, 9) and to hippuristanol, a small molecule inhibitor of eIF4A (3). Like the HCV IRES element, the PTV-1 IRES element does not require any of the eIF4 initiation factors for assembly of 48S initiation complexes on the RNA (27, 29, 30).
Avian encephalomyelitis virus (AEV) is a picornavirus that infects young chickens, quails, pheasants, and turkeys, causing ataxia and rapid tremors, especially in the neck. AEV is a worldwide problem, and almost all flocks are susceptible unless they are vaccinated (7). Around four billion birds are vaccinated worldwide each year to protect them from infection (I. Tarpey, personal communication). The AEV genome is 7,032 nt long (smaller than that of any other picornavirus). It encodes a polyprotein of 2,134 amino acids that is processed to the individual viral proteins which are most closely related to the hepatitis A virus (HAV) proteins. AEV has therefore been assigned to the hepatovirus genus of the picornaviruses (24). The 5'UTR of the AEV RNA is 494 nt long, which is also shorter than most other picornaviruses (24). On the basis of sequence comparisons and secondary structure predictions, it has recently been suggested that the AEV genome contains an HCV-like IRES element (15).
Here, we describe the biological properties of the AEV 5'UTR. We demonstrate for the first time that this region contains an IRES element with clear functional differences from that of HAV. We also show that the AEV IRES has significant functional and structural similarities to the other HCV-like picornavirus IRES elements.
|
|
|---|
To obtain a single cDNA fragment corresponding to the AEV 5'UTR, overlap PCR was performed. Two separate AEV cDNA clones (a gift from Ian Tarpey [Intervet, United Kingdom] and Dave Cavanagh [Institute for Animal Health, Compton, United Kingdom]) were used as templates to amplify by PCR fragments corresponding to nt 1 to 238 and 238 to 494 of the AEV 5'UTR using primers AEVF1 with AEVR266 and AEVF238 with AEVR494, respectively (Table 1). The two purified products were mixed and used in a further PCR using primers AEVF1 and AEVR494 to create a single fragment corresponding to the full-length AEV 5'UTR (nt 1 to 494) flanked by BamHI sites. The PCR product was ligated into pGEM-T Easy (Promega), and from the resultant plasmid, the AEV cDNA was released by BamHI digestion and then inserted, in both orientations, into similarly digested and phosphatased pGEM-CAT/LUC between the two open reading frames (ORFs). The plasmid containing the AEV 5'UTR cDNA in the sense (genomic) orientation was designated AEVs and that containing the fragment in the antisense orientation was called AEVas (Fig. 1A). A further construct containing cDNA corresponding to the AEV 5'UTR plus 30 nt of coding sequence (AEV+30) was created in a similar way using the reverse primer AEVR524 (Table 1) rather than AEVR494 in the PCRs.
|
View this table: [in a new window] |
TABLE 1. Oligonucleotides used for analysis of the AEV IRES
|
![]() View larger version (36K): [in a new window] |
FIG. 1. The AEV 5'UTR displays IRES activity in vitro and in vivo. (A) The structures of plasmids used in this study are shown. Various fragments of the 5'UTR of the AEV genome were amplified by PCR using primers containing BamHI sites, digested, and inserted between the CAT and LUC ORFs (at the unique BamHI site) in plasmid pGEM-CAT/LUC as described in Materials and Methods. Nucleotide numbers corresponding to the fragments are shown. (B) In vitro translation reactions containing RRL and [35S]methionine were programmed with RNA transcripts derived from the dicistronic plasmids containing the indicated virus sequences. Reaction products were analyzed by SDS-PAGE and autoradiography. The positions of the CAT and LUC proteins are indicated. (C) Transient-expression assay in 293 cells. The dicistronic plasmids (2 µg) containing the indicated virus sequences were transfected into vTF7-3-infected 293 cells. After 20 h, cell lysates were prepared and analyzed for CAT and LUC expression by SDS-PAGE and immunoblotting. LUC assays were performed on cell extracts from three separate transfections, and the results were standardized to the values for LUC expression directed by the EMCV IRES, which was set at 100%. LUC activities were normalized against CAT expression determined using a quantitative CAT enzyme-linked immunosorbent assay kit (Roche). The mean values (plus standard errors of the means [error bars]) are shown.
|
100 fragment (AEVm1) was created using primers AEVF1 and AEVR394, fragment AEV3'
200 (AEVm2) was produced using primers AEVF1 and AEVR294, AEV5'
100 (AEVm3) was made using primers AEVF100 and AEVR494 while AEV5'
200 (AEVm4) was synthesized using primers AEVF200 and AEVR494 (all primer sequences are given in Table 1). The various fragments were cloned, excised with BamHI, and ligated into the pGEM-CAT/LUC dicistronic vector to produce the illustrated plasmids (Fig. 1A). The structures were confirmed by restriction enzyme analysis and sequencing of the inserts. In vitro translation reactions. The dicistronic reporter plasmids (1 µg) were assayed in the RRL coupled transcription and translation (TNT) system (Promega) using [35S]methionine as described by the manufacturer. Products were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and autoradiography. Alternatively, uncapped mRNA transcripts were produced in vitro using the Ambion MegaScript kit with T7 RNA polymerase, following linearization of the plasmid DNAs with XhoI. Each mRNA was translated in each RRL with [35S]methionine, and products were analyzed by SDS-PAGE and autoradiography.
Transient-expression assays. The dicistronic reporter plasmids (2 µg) described above were transfected into 293 or HTK-143 cells alone or with the plasmid pGEM3Z/J1 (0.2 µg), which expresses the swine vesicular disease virus (SVDV) 2A protease as previously described (35). Briefly, the plasmids were transfected into cells (35-mm dishes) previously infected with the recombinant vaccinia virus vTF7-3, which expresses T7 RNA polymerase (13), using Lipofectin (8 µl; Invitrogen) and Optimem (192 µl; Gibco BRL). Cell lysates were prepared 20 h after transfection and were analyzed by SDS-PAGE and immunoblotting to determine CAT and LUC expression and eIF4G cleavage. Detection was achieved with anti-CAT (Sigma), anti-fLUC (Promega), or anti-eIF4G (gift from Simon Morley, University of Sussex, United Kingdom) antibodies and peroxidase-labeled anti-rabbit (Amersham) or anti-goat (Dako Cytomation) antibodies, respectively, using chemiluminescence reagents (Pierce). fLUC expression was also quantified using a firefly luciferase assay kit (Promega) with a luminometer.
RNA secondary structure prediction. AEV 5'UTR sequences (EMBL accession number AJ225173) were aligned with those from HCV (EMBL accession number AB016785) and PTV-1 (EMBL accession number AB038528) using ClustalW and manually edited. Secondary structure elements (other than the pseudoknot) were generated in Mfold (42).
Mutagenesis of the AEV cDNA. Mutations were introduced into the predicted domain IIIe region in order to change the sequence in the loop region from GAUA to AAAA (nt 446 to 449). The pGEM-CAT/AEVs/LUC plasmid was used as the template for two PCRs, one with each primer set (AEVIIIeF with AEVR494 and AEVIIIeR with AEVF1 [Table 1]). After purification, the products were mixed and a further PCR was performed using AEVF1 and AEVR494 primers. The BamHI-digested product was ligated into pGEM-CAT/LUC as described above, and the resultant plasmid was named pGEM-CAT/AEVIIIemut/LUC. The plasmid was sequenced to verify the presence of the expected mutations.
Mutations within the stem sequences of the predicted pseudoknot were also created (termed S1mut and S2mut). For the S1mut plasmid, nt 273 to 275 (CUC) were changed to GGG, and for the S2mut plasmid, nt 460 to 461 (CC) were changed to GG. The pGEM-CAT/AEVs/LUC plasmid was used as the template for the primary PCRs with the specific mutagenic primers and either CATForward or LUCReverse primer (Table 1) as appropriate. Secondary PCRs used just the latter primers. The final PCR products were purified and digested with BamHI, and the ca. 500-bp fragment was ligated into BamHI-digested and dephosphorylated pGEM-CAT/LUC vector to generate pGEM-CAT/AEVmutS1/LUC and pGEM-CAT/AEVmutS2/LUC. Compensatory mutations were produced in the same way starting with the mutS1 or mutS2 plasmids as templates for the PCRs. The mutagenic primers specified the compensatory mutations (Table 1). The presence of all the expected mutations in the plasmids was confirmed by sequencing.
Translation assays in the presence of hippuristanol. The requirement of the AEV IRES element for eIF4A was investigated both in vitro and in cells using hippuristanol, a specific inhibitor of eIF4A (3). Dicistronic plasmid DNAs were expressed in the TNT RRL system with or without hippuristanol (10 µM; kind gift from Jerry Pelletier, McGill University, Canada). The products were analyzed by 10% SDS-PAGE and autoradiography. The same plasmids were also assayed in HTK-143 cells with or without the addition of 0.5 µM hippuristanol; cell lysates were prepared after 20 h, and the inhibitor was added for the final 10 h.
|
|
|---|
The AEV IRES element functions in the presence of an enterovirus 2A protease. The 2A protease from PV (plus other enteroviruses) and the foot-and-mouth disease virus (FMDV) L protease each inhibit cap-dependent translation by inducing the cleavage of eIF4G, but these proteases have different effects on the various picornavirus IRES elements. Some IRES elements function very efficiently both in the presence or absence of the 2A and L proteases, for example, the EMCV IRES. Other IRES elements, for example, those from PV and other enteroviruses, are stimulated by these proteases within certain cell types, e.g., BHK cells (34). However, the IRES from HAV, the prototype hepatovirus, is strongly inhibited under these conditions, since it requires the intact eIF4F complex (4).We first studied the effect of SVDV 2A on the AEV IRES activity in cells in order to discover any similarity with the HAV IRES. The dicistronic plasmids were transfected into HTK-143 cells either alone or with the pGEM3Z/J1 plasmid which expresses the SVDV 2A protease. After 20 h, cell extracts were prepared and analyzed by SDS-PAGE and immunoblotting to detect CAT and LUC expression. As expected, all plasmids expressed CAT efficiently when transfected into cells alone, but CAT expression was strongly inhibited in the presence of SVDV 2A protease (Fig. 2A) as expected. The AEV and EMCV IRES elements directed efficient fLUC expression in HTK-143 cells (Fig. 2A) in both the presence and absence of the 2A protease, although the AEV IRES did show some inhibition. Note that the AEV IRES displayed higher efficiency compared to the EMCV IRES in this cell type versus the 293 cells (80% versus 50%; Fig. 1). Confirmation of eIF4G cleavage in cells expressing the SVDV 2A protease was achieved by Western blot analysis for eIF4G (Fig. 2B); the C-terminal cleavage product was observed only in the presence of the protease. In addition, we compared the effect of addition of FMDV L protease on the activity of the AEV and HAV IRES elements in vitro. In the presence of FMDV L protease, both the EMCV and AEV IRES elements retained activity, but the HAV IRES was severely inhibited in the presence of the L protease (data not shown), in agreement with previous data (4). These data demonstrate that the AEV IRES directs internal initiation of translation which is cap independent, but it does appear that the intact eIF4F complex is required for optimal activity. This may be due to a direct requirement for binding of a component of eIF4F or possibly to an indirect effect such as a requirement for an eIF4F-dependent factor. These results also indicate that the AEV IRES is different from the IRES element from the other hepatovirus, HAV, as the AEV IRES functions in the presence of cleaved eIF4G.
![]() View larger version (37K): [in a new window] |
FIG. 2. The AEV IRES functions in the presence of cleaved eIF4G. (A) Dicistronic plasmid DNA of the form CAT/IRES/LUC (2 µg) containing the indicated IRES sequences was transfected into HTK-143 cells in the absence (–) or presence (+) of a plasmid encoding SVDV 2A protease (0.2 µg). After 20 h, cell extracts were prepared and analyzed for CAT and LUC expression as described in the legend to Fig. 1. LUC assays were performed on cell extracts from three separate transfections, and the results were standardized to the values for LUC expression directed by the EMCV IRES, which was set at 100%. LUC activities were normalized against CAT expression as described in the legend to Fig. 1. The mean values (plus standard errors of the means [error bars]) are shown. (B) Samples were also analyzed by immunoblotting to analyze the status of eIF4G. The position of the C-terminal cleavage product of eIF4G is indicated (Ct).
|
![]() View larger version (33K): [in a new window] |
FIG. 3. Delimitation of the AEV 5'UTR sequences required for IRES activity. Dicistronic plasmids containing the AEV 5'UTR and truncated versions of this sequence, AEVm1 (nt 1 to 294), AEVm2 (nt 1 to 394), AEVm3 (nt 100 to 494), and AEVm4 (nt 1 to 394), were transfected into vTF7-3-infected HTK-143 cells, and cell extracts were analyzed for CAT and LUC expression as described in the legend to Fig. 1. LUC expression was also measured by LUC assay, and the results are shown below the immunoblot in arbitrary units. Similar results were obtained in two independent experiments.
|
Similarity between the AEV IRES and IRES elements from HCV and PTV-1. As described above, the functional properties of the AEV IRES are clearly distinct from those of the HAV IRES. To examine the relationship of the AEV IRES to other picornavirus IRES elements, we performed sequence alignments of the AEV sequence using ClustalW. We found that the AEV IRES shares a significant level of identity with the recently characterized PTV-1 IRES (8, 17, 29). The PTV-1 IRES has been shown to resemble the IRES element from HCV, a flavivirus, and the AEV IRES also shares certain critical characteristics with them (Fig. 4A and B). Notably, the HCV domain IIIe is identical to a region of 12 nt within the AEV sequence. Overall, the AEV IRES shares 48% sequence identity with the HCV IRES and 42% identity with the PTV-1 IRES. The similarities are particularly apparent in the regions surrounding and including the pseudoknot found in the HCV and PTV-1 IRES structures (Fig. 4C). We therefore believe that the pseudoknot structure found in these IRES elements is also present in the AEV IRES.
![]() ![]() ![]() View larger version (83K): [in a new window] |
FIG. 4. Similarity between the HCV and AEV IRES elements. (A) Alignment of the HCV and AEV IRES sequences. Sequences were aligned with ClustalW and manually edited. Identical nucleotides are marked with an asterisk. Individual domains associated with the HCV and AEV IRES elements are indicated above the sequence. Thick black lines indicate regions involved in the formation of the pseudoknot structure within the HCV IRES and AEV IRES elements. The overall sequence identity from this alignment is 48.1%, but note the 100% sequence identity within the domain IIIe region. The gaps introduced to maximize alignment are indicated by dashes. (B) Proposed secondary structure of the entire AEV IRES. Domains are labeled according to corresponding domains of the HCV IRES (inset). The structure was predicted by comparative sequence analysis and using Mfold (42) to predict the most thermodynamically favorable structures. (C) Comparison of predicted secondary structures of the domain IIIe and IIIf regions of the HCV, PTV-1, and AEV IRES elements. The two stems (S1 and S2) and loop regions (L1 and L2) that form the pseudoknot are shown. A domain IV structure is also present in the HCV IRES but not in the PTV-1 or AEV IRES elements. Within the AEV sequence, the nucleotides indicated in bold type are those that were modified in the experiments shown in Fig. 5.
|
![]() View larger version (33K): [in a new window] |
FIG. 5. Mutation of the domain IIIe loop or pseudoknot structure results in loss of AEV IRES activity. (A) Dicistronic plasmids containing the wt AEV IRES or the IRES containing the loop IIIe mutation were transfected into vTF7-3-infected HTK-143 cells and analyzed for CAT and LUC expression as described in the legend to Fig. 1. LUC activities (normalized against CAT expression) are shown, and the results are the mean LUC values from three experiments. (B) Dicistronic plasmids containing the indicated mutations within the predicted pseudoknot region were transfected into HTK-143 cells as described above for panel A and analyzed for CAT and LUC expression as described in the legend to Fig. 3. The results are representative of two independent experiments.
|
Mutations in stem 2 (S2) changed nt 460 and 461 (CC) to GG, these changes were predicted to disrupt the interactions with nt 481 and 482 (GG) and hence destabilize the pseudoknot structure. The results from in vivo (Fig. 5B) and in vitro (data not shown) experiments indicated that these mutations completely abrogated IRES activity as anticipated. Furthermore, compensatory mutations that changed nt 481 and 482 (GG) to CC, predicted to restore base pair interactions, efficiently regenerated IRES activity (about 60% of wt AEV IRES activity in vivo [Fig. 5B]). These results also supported the predicted pseudoknot structure shown in Fig. 4. The fact that full restoration of IRES activity was not achieved may suggest that these nucleotides are also involved in other interactions (such as RNA-protein interactions) as well as forming the pseudoknot structure.
The AEV IRES is resistant to hippuristanol, an inhibitor of eIF4A. In previous studies, it has been found that the HCV and PTV-1 IRES elements have no requirement for eIF4A for translation initiation (3, 27, 29). As the results presented above indicated that the AEV IRES element resembles these IRES elements, the requirement for eIF4A was studied in vitro and in vivo using hippuristanol, a specific inhibitor of eIF4A. Hippuristanol inhibits cap-dependent translation and the activity of type I and II picornavirus IRES elements. In contrast, the activity of the HCV and PTV-1 IRES elements is resistant to this inhibitor (3). Selected dicistronic reporter plasmids were assayed in cells in the presence and absence of hippuristanol. As expected, the eIF4A inhibitor severely reduced translation of the upstream cistron CAT (Fig. 6A). As seen before (9), the EMCV IRES activity was also reduced to about 10% of its activity in the presence of the eIF4A inhibitor. In contrast, the AEV IRES displayed marked resistance to this inhibitor (Fig. 6A). Similar results were also observed in vitro (Fig. 6B). However, we did note that the AEV IRES was partially inhibited in the presence of this inhibitor in both systems (reduced to about 50% activity in the presence of hippuristanol, which is similar to the reduction seen in the presence of the SVDV 2A protease), although a similar effect was also observed with the HCV IRES element (Fig. 6B). In contrast, the HAV IRES was completely inhibited in the presence of hippuristanol, a previously unreported finding (Fig. 6B). This result is in agreement with the suggestion that the HAV IRES requires the whole eIF4F complex for its function (4).
![]() View larger version (38K): [in a new window] |
FIG. 6. The AEV IRES is resistant to an inhibitor of eIF4A. (A) Dicistronic plasmids containing the indicated IRES sequences were transfected into HTK-143 cells in the absence (–) or presence (+) of 0.5 µM hippuristanol (Hipp.), an inhibitor of eIF4A. Cells were harvested after 20 h, and the inhibitor was added for the last 10 h of the incubation. Cell extracts were analyzed for CAT and LUC expression as described in the legend to Fig. 1. LUC assays were performed on cell extracts from three separate transfections, and the results were standardized to the values for LUC expression directed by the EMCV IRES, which was set at 100%. The mean values (plus standard errors of the means [error bars]) are shown. (B) Dicistronic plasmids containing the indicated IRES elements were also assayed in RRL TNT in the presence of 10 µM hippuristanol. The HAV and HCV IRES elements are within a cyclin/NS dicistronic construct. The positions of all proteins are indicated. Note the slower migration of the HCV IRES-directed NS product compared to the HAV IRES-directed product (arrows) due to the inclusion of 30 nt of the HCV coding sequence in this plasmid (32).
|
|
|
|---|
We have shown that the functional AEV IRES lies within nt 100 to 494 of the 5'UTR. In contrast to the PV, EMCV, or HAV IRES elements, the AEV IRES lacks a polypyrimidine tract near the 3' end of the element. It can function when eIF4G is cleaved and is also resistant to hippuristanol, an inhibitor of eIF4A activity. These features distinguish the AEV IRES from these picornavirus IRES elements, but they are shared with the PTV-1, PEV-8, and SV2 IRES elements plus the HCV and CSFV IRES elements. Alignment of the nucleotide sequences of the HCV and AEV IRES elements confirmed that these elements share a striking similarity (about 48% overall identity). Indeed, within the HCV IIIe domain there is 100% sequence identity to a region of the AEV IRES (Fig. 4). The secondary structure models for the HCV-like IRES elements include an important pseudoknot structure. The AEV sequence is also proposed to form this structure (Fig. 4). We have obtained supporting evidence for the formation of this structure in the AEV IRES through mutational analysis of the sequences predicted to form the pseudoknot. Mutations within the S1 or S2 regions that were expected to disrupt the predicted pseudoknot structure inhibited IRES activity, but compensatory mutations designed to restore the base pairing in this structure efficiently rescued activity. Mutation of the GAUA motif within a portion of the AEV sequence that is identical to the domain IIIe of the HCV IRES also disrupted AEV IRES activity. Mutation of this loop region in the HCV and PTV-1 IRES elements also inhibited IRES activity (8, 22, 31). It is known that the IIId and IIIe regions of the HCV IRES domain III interact with the 40S ribosomal subunit (18, 23), while the IIIb region has been shown to interact with eIF3 (16, 18, 37). A recent model of initiation complex formation on the HCV IRES suggests that regions IIId and IIIe of the IRES bind to the 40S ribosomal subunit, and this is followed by the interaction with eIF3 and the ternary complex (eIF2/met-tRNAi/GTP) to form a 48S preinitiation complex (25). Recent data have shown that several proteins of the 40S ribosomal subunit, including p40, S3a, S5, and S16 are positioned close to hairpin IIIe of the HCV IRES element during the early stage of translation initiation (19). It has also recently been shown that domain II of the HCV and CSFV IRES elements plays a role in 80S ribosome assembly on RNA and promotes eIF5-induced GTP hydrolysis and eIF2/GDP release, following 48S initiation complex formation (21).
The AEV IRES does not require any viral coding sequence for function, and secondary structure predictions suggest that there is no region equivalent to HCV IRES domain IV (15; L. O. Roberts, unpublished data). There are some differences between the picornavirus "HCV-like" IRES elements in this respect, as the SV2 IRES is predicted to contain a domain IV region, whereas the PTV-1 and PEV-8 IRES elements do not (9).
The discovery of a number of distinct picornaviruses harboring an HCV-like IRES element in their 5'UTR suggests that recombination between picornavirus and flavivirus genomes has occurred. Previous work has shown that the PV IRES can be replaced with the HCV IRES (41) to produce a viable chimeric virus, suggesting that they are functionally equivalent (although mechanistically very different). However, it is important to note that there are important structural differences between the picornavirus and HCV genomes [e.g., picornaviruses possess a 3' poly(A) tail] that may be important in translation/replication. There are also differences in the sequence and predicted structures of the domain II regions of the picornavirus "HCV-like" and HCV IRES elements which may have a role in translation and/or replication. It remains to be seen whether the diverse domain II structures found in the picornavirus "HCV-like" IRES elements have the same function or are involved in replication.
AEV has been tentatively classified as being a member of the hepatovirus genus within the Picornaviridae, since it has the highest degree of protein sequence identity to HAV. However, its IRES element is clearly distinct from that of HAV in that it functions well in the presence of cleaved eIF4G, displays resistance to an inhibitor of eIF4A, and shares a striking similarity to the HCV-like IRES elements. Due to these key differences between AEV and HAV, it is suggested that the placement of AEV within the hepatovirus genus should be reconsidered.
M.B. gratefully acknowledges receipt of a scholarship from the Ministry of Science, Research and Technology and the Ministry of Jihad-Agriculture of Iran.
Published ahead of print on 12 December 2007. ![]()
Present address: Poultry Vaccines Department, Razi Vaccine & Serum Research Institute, P.O. Box 31975/148, Karaj, Iran. ![]()
Present address: Institute for Molecular and Cellular Biology, Faculty of Biological Sciences, University of Leeds, Leeds LS2 9JT, United Kingdom. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»