The 5'-poly(A) leader of poxvirus mRNA confers a translational advantage that can be achieved in cells with impaired cap-dependent translation

Abstract

The poly(A) leader at the 5'-untranslated region (5'-UTR) is an unusually striking feature of all poxvirus mRNAs transcribed after viral DNA replication (post-replicative mRNAs). These poly(A) leaders are non-templated and of heterogeneous lengths; and their function during poxvirus infection remains a long-standing question. Here, we discovered that a 5'-poly(A) leader conferred a selective translational advantage to mRNA in poxvirus-infected cells. A constitutive and uninterrupted 5'-poly(A) leader with 12 residues was optimal. Because the most frequent lengths of the 5'-poly(A) leaders are 8-12 residues, the result suggests that the poly(A) leader has been evolutionarily optimized to boost poxvirus protein production. A 5'-poly(A) leader also could increase protein production in the bacteriophage T7 promoter-based expression system of vaccinia virus, the prototypic member of poxviruses. Interestingly, although vaccinia virus post-replicative mRNAs do have 5'- methylated guanosine caps and can use cap-dependent translation, in vaccinia virus-infected cells, mRNA with a 5'-poly(A) leader could also be efficiently translated in cells with impaired cap-dependent translation. However, the translation was not mediated through an internal ribosome entry site (IRES). These results point to a fundamental mechanism poxvirus uses to efficiently translate its post-replicative mRNAs.

Fig 1. The 5’ poly(A) leader confers an mRNA translational advantage during the post-replicative stage of VACV replication.

(A) Schematic of experimental approach. Messenger RNA was synthesized by a T7 promoter-based in vitro transcription system. The firefly luciferase (Fluc) reporter mRNA under a 5’-UTR to be tested was transfected into cells together with a renilla luciferase (Rluc) reporter mRNA under a 5’-UTR-containing Kozak sequence. Luciferase activities were measured at 5 h post transfection using a luminometer with a dual luciferase assay system. (B) Fluc mRNA with a 5’-poly(A) leader of 20 residues was transfected into uninfected or wild-type VACV-infected HeLa cells (12 hpi) together with an Rluc mRNA. Luciferase activities were measured at 5 h post transfection. The Rluc-normalized Fluc activity was normalized as 1 in uninfected HeLa cells. (C) The relative levels of Fluc mRNA from uninfected and VACV-infected HeLa cells were quantitated at 5 h post transfection by quantitative RT-PCR (qRT-PCR). The amount of mRNA in uninfected cells was normalized as 1. (D) Fluc mRNA with a 5’-poly(A) leader of 20 residues was transfected into uninfected, or A23-deleted recombinant VACV-infected HeLa cells (12 hpi) together with an Rluc mRNA. Luciferase activities were measured at 5 h post transfection. The Rluc-normalized Fluc activity was normalized as 1 in uninfected HeLa cells. Error bars represent standard deviation (SD) of at least three experiments. P-values were obtained using the Student’s t-test; **P value < 0.01, ns = Not Significant (i.e. P value > 0.05).

Fig 2. The optimal length of 5’-poly(A) leader in conferring translational advantage is 12 residues.

(A) The Fluc reporter mRNAs with different 5’-poly(A)-leaders lengths were transfected into uninfected and VACV-infected HeLa cells, together with an Rluc mRNA. Luciferase activities were measured at 5 h post transfection. (B) Fluc reporter mRNA containing the A2L mRNA 5’-UTR with (w) or without (w/o) a poly(A) leader, was transfected into uninfected and VACV-infected HeLa cells, together with an Rluc mRNA. Luciferase activities were measured at 5 h post transfection. The Rluc-normalized Fluc activity was normalized as 1 in uninfected cells. (C) Fluc reporter mRNA containing RNF165 5’-UTR with (w) or without (w/o) an A-tract addition at the 5’-end was transfected into uninfected and VACV-infected HeLa cells together with an Rluc mRNA. Luciferase activities were measured at 5 h post transfection. Error bars represent standard deviation (SD) of at least three experiments. P- values were determined using the Student’s t-test; *P value < 0.05.

Fig 3. An uninterrupted 5’-poly(A) leader is essential for optimal translation in VACV-infected cells.

(A) Schematic of representation showing individual 5’-poly(A) residues mutated to U. (B) Fluc reporter mRNA containing each of the mutated 5’-poly(A) leaders were transfected into uninfected and VACV-infected HeLa cells, together with an Rluc mRNA. Luciferase activities were measured at 5 h post transfection. The Rluc normalized Fluc activity was normalized as 1 in uninfected HeLa cells. Error bars represent standard deviation (SD) of at least three experiments.

Fig 4. A 5’-poly(A) leader confers a translational advantage in multiple cell types and in myxoma virus infection.

(A) An Fluc reporter mRNA with a 12-residue 5’-poly(A) leader was transfected into uninfected or VACV-infected HeLa, HFF, BS-C-1 or RK13 cells, together with an Rluc mRNA, at 24 and 48 hpi. Luciferase activities were measured at 5 h post transfection. The Rluc-normalized Fluc activity was normalized as 1 in uninfected cells. (B) An Fluc reporter mRNA with a 12-residue 5’-poly(A) leader was transfected into uninfected or myxoma virus-infected RK13 cells, together with an Rluc mRNA. Luciferase activities were measured at 5 h post transfection. Error bars represent standard deviation (SD) of at least three experiments. P-values were determined using the Student’s t-test; **P value < 0.01, *P value < 0.05.

Fig 5. VACV post-replicative mRNAs are more efficiently translated than cellular mRNAs.

Boxwhisker plots of relative translation efficiency of VACV post-replicative mRNAs and host mRNAs at 4 and 8 hpi. Relative translation efficiency of mRNAs from uninfected HeLa cells is also shown. The bottom and top of the box indicate the first and third quartiles, respectively, with the line in the middle representing the median. Cell-M, cellular mRNAs from mock-infected cells. Cell-I, cellular mRNAs from VACV-infected cells. VACV-I, VACV mRNAs. The highest relative translation efficiency of VACV-I was off scale. ***indicates a P value < 0.001.

Fig 6. Messenger RNA with a 5’-poly(A) leader is more efficiently translated than mRNA without a 5’-poly(A) leader in VACV-infected cells.

(A) HeLa cells were transfected with plasmids expressing eGFP from a T7 promoter and containing an A-tract or a Kozak sequence between the ATG and the T7 promoter. Cells were infected with vT7LacOi 24 h post transfection and different concentrations of IPTG were added to the media as indicated. At 24 hpi, Western blotting tracked eGFP and GAPDH (loading control) expression levels; and eGFP intensity was normalized to GAPDH, shown below. (B) Levels of eGFP mRNA levels for (A) were measured by qRT-PCR, and normalized to 18S rRNA. (C) HeLa cells were infected with recombinant VACV vT7LacOi_Kozak-GFP or vT7LacOi_A12-GFP. Indicated concentrations of IPTG were added to the culture media. At 24 hpi Western blotting tracked eGFP and GAPDH (normalization and loading control) expression, shown below. (D) Levels of eGFP mRNA expression of (C) were measured by qRT-PCR, and normalized to 18S rRNA. (E) HeLa cells were infected with vT7LacOi_A12-GFP or vT7LacOi_Kozak-GFP. Total and ribosome/polysome-bound RNAs were isolated at 15 hpi. The GFP and F17 mRNAs were measured by qRT-PCR and normalized to 18S rRNA. The ratios of ribosome/polysome-bound mRNA to total mRNA were calculated for GFP and F17, respectively. The ratios from vT7LacOi_A12-GFP-infected cells were normalized to one. **indicates a P value < 0.01. ns = Not Significant (i.e. P value > 0.05).

Fig 7. Messenger RNA with a 5’-poly(A) leader capped by ApppG cap analog can confer the translational advantage in VACV-infected cells.

(A) Relative luciferase activity from Fluc mRNA with a 5’-poly(A) leader capped with ApppG or m⁷G was measured at 5 h after transfection into uninfected or VACV-infected HeLa cells. Fluc activity was normalized by a co-transfected Rluc mRNA. Fluc activity from m⁷G-capped mRNA in uninfected cells was normalized as 1. (B) Quantitative RT-PCR (qRT-PCR) compared the relative levels of Fluc mRNA capped by ApppG or m⁷G from uninfected and VACV-infected HeLa cells at 5 h post transfection. The amount of mRNA in uninfected cells was normalized as 1. (C) ApppG-capped Fluc reporter mRNAs, with different lengths of 5’-poly(A) leaders, were transfected into uninfected and VACV-infected HeLa cells together with an m⁷G Rluc mRNA. Luciferase activities were measured at 5 h post transfection. Error bars represent standard deviation (SD) of at least three experiments. P-values were determined using the Student’s t-test; ***P value < 0.001, **P value < 0.01, *P value < 0.05, ns = Not Significant (i.e. P value > 0.05).

Fig 8. VACV post-replicative mRNAs are efficiently translated when eIF4E is inhibited.

(A) Hypophosphorylation of 4E-BP1 was induced by LY294002 in uninfected and VACV-infected HeLa cells at 1 hpi or 8 hpi and the cells were harvested for Western blotting analysis probed using anti-4E-BP1 antibody. An asterisk indicates hyperphosphorylated 4E-BP1. The arrow indicates hypophosphorylated 4E-BP1. (B) HeLa cells were treated with LY294002 at 1 or 8 hpi. An Fluc reporter mRNA with a 12A leader was transfected into uninfected and VACV-infected cells together with an Rluc mRNA at 12 hpi. Luciferase activities were measured at 5 h post transfection. The ratios of Fluc and Rluc activities from VACV-infected to uninfected HeLa cells were calculated, respectively. (C) HeLa cells infected with VACV and treated with LY294002 at 8 hpi. Newly synthesized proteins were labeled by AHA at 20 hpi and detected using antibodies against biotin conjugated to AHA using the Click-IT chemistry-based technique. The blots of infected-cell lysates were from different lanes on the same gel. (D) HeLa cells were transfected with control (siNC) or siRNAs targeting eIF4E. The protein levels of eIF4E and GAPDH were detected using specific antibodies 48 h post transfection. (E) HeLa cells were transfected with control or eIF4E siRNAs as in (D). An Fluc reporter mRNA with a 12A leader was transfected into uninfected and VACV-infected cells together with an Rluc mRNA at 12 hpi. Luciferase activities were measured at 5 h post transfection. The ratios of Fluc and Rluc activities from VACV-infected to uninfected HeLa cells were calculated, respectively. (F) HeLa cells were transfected with control or eIF4E siRNAs as in (D). 48 h post-transfection, the cells were infected with a recombinant VACV expressing firefly luciferase gene under a viral early/late promoter. Luciferase activities were detected at 8 hpi. (G) HeLa cells were transfected with control or eIF4E siRNAs as in (D). 48 h post-transfection, the cells were infected with VACV. The viral titers were determined at 24 hpi using a plaque assay. Error bars represent standard deviation (SD) of at least three experiments. P-values were determined using the Student’s t-test; **P value < 0.01.

Fig 9. The 5’-poly(A) leader is not an internal ribosome entry site (IRES).

(A) Schematic of the bicistronic reporter mRNAs flanked by a poliovirus IRES, 20 A residues, or a Kozak-containing sequence. (B) The in vitro transcribed bicistronic mRNA was transfected into uninfected and VACV-infected HeLa cells and luciferase activities were measured 5 h post transfection. The ratios of Fluc to Rluc activities were calculated and displayed. Error bars represent standard deviation (SD) of at least three experiments. P-values were determined using the Student’s t-test; *P value < 0.05, ns = Not Significant (i.e. P value > 0.05).