trans regulation of cap-independent translation by a viral subgenomic RNA

Abstract

Many positive-strand RNA viruses generate 3'-coterminal subgenomic mRNAs to allow translation of 5'-distal open reading frames. It is unclear how viral genomic and subgenomic mRNAs compete with each other for the cellular translation machinery. Translation of the uncapped Barley yellow dwarf virus genomic RNA (gRNA) and subgenomic RNA1 (sgRNA1) is driven by the powerful cap-independent translation element (BTE) in their 3' untranslated regions (UTRs). The BTE forms a kissing stem-loop interaction with the 5' UTR to mediate translation initiation at the 5' end. Here, using reporter mRNAs that mimic gRNA and sgRNA1, we show that the abundant sgRNA2 inhibits translation of gRNA, but not sgRNA1, in vitro and in vivo. This trans inhibition requires the functional BTE in the 5' UTR of sgRNA2, but no translation of sgRNA2 itself is detectable. The efficiency of translation of the viral mRNAs in the presence of sgRNA2 is determined by proximity to the mRNA 5' end of the stem-loop that kisses the 3' BTE. Thus, the gRNA and sgRNA1 have "tuned" their expression efficiencies via the site in the 5' UTR to which the 3' BTE base pairs. We conclude that sgRNA2 is a riboregulator that switches off translation of replication genes from gRNA while permitting translation of structural genes from sgRNA1. These results reveal (i) a new level of control of subgenomic-RNA gene expression, (ii) a new role for a viral subgenomic RNA, and (iii) a new mechanism for RNA-mediated regulation of translation.

FIG. 1.

trans regulation model of BYDV gene expression. (1) In the early stage of BYDV infection, subgenomic RNAs are absent; thus, the products of ORFs 1 and 2, including the RdRp, are the only proteins produced. (2) Viral genomic-RNA replication and subgenomic-RNA transcription occur. (1, 2, and 3) Viral RNAs accumulate, and viral proteins are produced. (4) The accumulation of sgRNA2 trans inhibits translation of BYDV RdRp from gRNA. (5) However, translation of structural and movement proteins from sgRNA1 is not inhibited. Genomic RNAs are available for encapsidation in the coat protein.

FIG. 2.

Effects of sgRNA2 on translation of reporters in wheat germ translation extracts. In all cases, for each mRNA tested, the relative luciferase (Rel. Luc.) activity in the absence of sgRNA2 is defined as 100%. (A) Maps of reporter RNAs. fLUC, firefly luciferase; rLUC, Renilla luciferase. The ends of the UTRs are numbered as in the full-length BYDV genome. (B) Differential effects of sgRNA2 and sgRNA2BF in trans on translation of GfLUC or SG1rLUC in separate reactions. The error bars indicate standard deviations. (C) Effects of sgRNA2 on different reporters with the same sgRNA1 5′ UTR; relative luciferase activity of SG1rLUC or SG1fLUC in separate reactions in the presence of the indicated (molar) excess of sgRNA2. (D) Differential effects of sgRNA2 or sgRNA2BF on translation of GfLUC and SG1rLUC in the same reaction. The activities of GfLUC and SG1rLUC were plotted individually against the excess of sgRNA2 and sgRNA2BF. (E) Changes in ratios of GfLUC/SG1rLUC activity from panel D. ΔGfLUC/SG1rLUC = (GfLUC/SG1rLUC in the presence of sgRNA2 or sgRNA2BF)/(GfLUC/SG1rLUC in the absence of sgRNA2 or sgRNA2BF).

FIG. 3.

Differential effects of PAV6 and PAV6ΔSG2 replication on translation of GfLUC and SG1rLUC in oat protoplasts. (A) Diagram of the two-step electroporation method. First, oat protoplasts were inoculated with full-length infectious BYDV PAV6 or PAV6ΔSG2 transcripts. After a 24-h incubation to allow viral replication and sgRNA accumulation, the cells were electroporated again with 1 pmol GfLUC, 1 pmol SG1rLUC (C), or both (D). Luciferase activities were measured 4 h later. The error bars indicate standard deviations. (B) Northern blot hybridization showing replication of PAV6 and PAV6ΔSG2 at 24 h p.i. (C) Luciferase activities in cells first transfected with PAV6 or PAV6ΔSG2 RNAs and then reelectroporated with the indicated reporter RNA. The luciferase activity of GfLUC (or SG1rLUC) in mock-transfected cells was defined as 100%. (D) Same as in panel C, but 1 pmol GfLUC and 1 pmol SG1rLUC were coelectroporated together into the same batch of transfected protoplasts.

FIG. 4.

Differential trans inhibition by sgRNA2 alone in oat protoplasts. The two-step electroporation assay was employed, with 4 h between electroporations. The graph shows the luciferase activities of GfLUC and SG1rLUC measured 4 h after they were coelectroporated into cells previously electroporated with sgRNA2 or sgRNA2BF. The activities of GfLUC and SG1rLUC in cells that were preelectroporated with no RNA were defined as 100%. The error bars indicate standard deviations.

FIG. 5.

Mechanism of trans inhibition by sgRNA2. Shown are the differential effects of 20-fold molar excess of sgRNA2-LIII-CS, sgRNA2, and sgRNA2BF on translation of GfLUC RNA in wheat germ extract. The error bars indicate standard deviations.

FIG. 6.

Features of the 5′ UTRs of gRNA and sgRNA1 that determine the differential trans inhibition by sgRNA2. (A) The known secondary structure of the BYDV gRNA 5′ UTR (21) and the MFOLD-predicted (58) secondary structure of the sgRNA1 5′ UTR. The kissing BCL bases that participate in the long-distance interaction with the 3′ BTE are in gray. (B) Schematic diagram of the 3′ BTE-5′ UTR interactions in the indicated reporter constructs. In A-GfLUC, the 5′-proximal loop of SL-A was converted by a single C-to-A change at position 15 (italics), which made SL-A complementary to the 3′ BTE at five consecutive bases (boldface gray). The endogenous SL-D kissing bases were mutated to prevent base pairing with the 3′ BTE (GAC to CUG; black italics). In D-SG1fLUC, the loop of sg1SL-D was converted from AGUUA to CUGACAA (bases 110 to 116). The modified D-SG1LUC also contained an A-to-C change at position 10 in the loop of sg1SL-A, which prevented base pairing to the 3′ BTE. (C) Differential effects of sgRNA2 on translation of A-GfLUC, GfLUC, and SG1fLUC in wheat germ extract. The activity of GfLUC, A-GfLUC, or SG1fLUC in the absence of sgRNA2 was defined as 100%. The error bars indicate standard deviations. (D) Differential inhibition by 40-fold excess of sgRNA2 or sgRNA2BF of GfLUC, A-GfLUC, and SG1rLUC translation in oat protoplasts. The two-step electroporation assay was employed as in Fig. 4. GfLUC and SG1rLUC or A-GfLUC and SG1rLUC (2 pmol each) were coelectroporated in oat protoplasts 4 h after the indicated RNAs were electroporated into the same cells. SG1rLUC levels differed little in the presence of GfLUC or A-GfLUC, so average SG1rLUC readings are shown. For each reporter RNA, luciferase readings were normalized to the amount detected in the absence of RNA in the first electroporation. (E) Differential effects of 20-fold molar excess of sgRNA2 or sgRNA2BF on translation of GfLUC, SG1fLUC, and D-SG1fLUC in wheat germ extract. The readings were normalized to the amount detected in the absence of sgRNA2 or sgRNA2BF RNA.

FIG. 7.

Attempted detection of ORF 6 translation in vivo. (A) Maps of BYDV genomic PAV6 and PAV6-FLAG and the subgenomic sgRNA2-FLAG transcript. Nucleic acid and amino acid sequences of the FLAG tag inserted at the 3′ end of ORF 6 (black triangle) are shown. (B) Western blot using anti-FLAG antibodies on total protein from oat protoplasts inoculated with infectious BYDV PAV6, PAV6-FLAG, or nonreplicative sgRNA2-FLAG RNA 24 h posttransfection. As a positive control, the wheat germ translation product of sgRNA2-FLAG is shown in the absence (−) or presence (+) of protoplast extract, which retarded protein mobility in the gel. (C) Northern blot hybridization of total RNA from virus-infected cells showing replication of infectious BYDV PAV6 and PAV6-FLAG. (D) Translation in oat protoplasts of SG2fLUC transcript with ORF 6 fused to the firefly luciferase ORF. The presence (+) or absence (−) of a cap and/or a poly(A) tail on this transcript is indicated. Relative luciferase activity was normalized to that of a capped, polyadenylated, nonviral Renilla reporter construct.