Selective Degradation of Host RNA Polymerase II Transcripts by Influenza A Virus PA-X Host Shutoff Protein

Abstract

Influenza A viruses (IAVs) inhibit host gene expression by a process known as host shutoff. Host shutoff limits host innate immune responses and may also redirect the translation apparatus to the production of viral proteins. Multiple IAV proteins regulate host shutoff, including PA-X, a ribonuclease that remains incompletely characterized. We report that PA-X selectively targets host RNA polymerase II (Pol II) transcribed mRNAs, while sparing products of Pol I and Pol III. Interestingly, we show that PA-X can also target Pol II-transcribed RNAs in the nucleus, including non-coding RNAs that are not destined to be translated, and reporter transcripts with RNA hairpin structures that block ribosome loading. Transcript degradation likely occurs in the nucleus, as PA-X is enriched in the nucleus and its nuclear localization correlates with reduction in target RNA levels. Complete degradation of host mRNAs following PA-X-mediated endonucleolytic cleavage is dependent on the host 5'->3'-exonuclease Xrn1. IAV mRNAs are structurally similar to host mRNAs, but are synthesized and modified at the 3' end by the action of the viral RNA-dependent RNA polymerase complex. Infection of cells with wild-type IAV or a recombinant PA-X-deficient virus revealed that IAV mRNAs resist PA-X-mediated degradation during infection. At the same time, loss of PA-X resulted in changes in the synthesis of select viral mRNAs and a decrease in viral protein accumulation. Collectively, these results significantly advance our understanding of IAV host shutoff, and suggest that the PA-X causes selective degradation of host mRNAs by discriminating some aspect of Pol II-dependent RNA biogenesis in the nucleus.

Fig 1. RNA targeting by PA-X is dependent on the host RNA polymerase complex that transcribes the RNA.

(A) HEK 293T cells were transfected with a Pol II-driven RFP reporter, the indicated GFP reporters expressed from Pol I, Pol II or Pol III-driven promoters, and either PA-X-myc or an empty vector. Total cellular RNA was extracted and RFP, GFP and 18S RNA levels were analyzed by RT-qPCR. Reporter expression normalized against 18S RNA levels is plotted as fold change between PA-X-expressing vs. vector transfected cells. n.s.,**,*** = p value (Student’s t-test) > 0.05 or < 0.01, 0.001 for fold change of GFP vs. fold change of RFP. (B-C) HEK 293T (B) or A549 (C) cells expressing doxycycline-inducible wild-type PA-X-myc (“iPA-X wt”), a catalytically inactive mutant (“iPA-X D108A”) or RFP (“iRFP”) were treated with doxycycline for 18 h to induce expression of PA-X or RFP. Total cellular RNA was extracted and RT-qPCR was performed to measure levels of the indicated endogenous RNAs. Endogenous RNA levels were normalized by 18S rRNA levels and are plotted as fold change in induced (PA-X- or RFP-expressing) vs. uninduced cells. HEK293T iPA-X line #7 and A549 iPA-X wt line #10 and D108A line #8 were used for this figure. Data on additional clonal lines is presented in S1 Fig. *,**,*** = p value (Student’s t-test) < 0.05, 0.01, 0.001 vs. 293T iRFP (B, n = 5) or A549 iPA-X D108A line #8 (C, n = 3). Error bars = s.e.m.

Fig 2. Complete degradation of mRNA by PA-X is dependent on the cellular exonuclease Xrn1.

(A) Diagram showing GFP constructs with a flaviviral Xrn1-blocking element (SLII) introduced at different positions. (B) HEK 293T cells were co-transfected with GFP-SLII constructs and PA-X, SOX, or empty vector (“-”). Total cellular RNA was extracted and northern blotted using probes against the 3’UTR of the GFP reporter or against 18S (as a loading control). Arrowheads indicate RNA fragments protected from degradation by the SLII sequence. (C-E) HEK 293T shXrn1 or shNS cells were treated with doxycycline (+dox) for four days to induce shRNAs against Xrn1 or non-specific control shRNAs (shNS), or were left untreated (-dox). Three days after shRNA induction, cells were transfected with a GFP reporter plasmid and the indicated construct or empty vector (“-”). Total cellular RNA was extracted and subjected to northern blot analysis or RT-qPCR. (C) GFP mRNA levels were analyzed by RT-qPCR and normalized against levels of 18S rRNA (also detected by RT-qPCR). Normalized GFP expression is plotted as fold change in PA-X-expressing vs. vector-transfected cells. * = p value (one-sample t-test) < 0.05. (D) Northern blot probes directed against the 3’UTR of the GFP reporter or 18S (as a loading control) were used to detect RNA levels. The bracket denotes PA-X degradation fragments and the arrowhead indicates the RNA fragment generated by SOX cleavage. A representative image is shown. (E) Levels of Xrn1 for experiment in panel D were assessed by western blot, using tubulin as a loading control. This western blot is also representative of the Xrn1 knock-down observed in samples used for panel C and other repeats of the northern blot.

Fig 3. PA-X targets Pol II-transcribed mRNAs that are not translated.

(A) HEK 293T cells were transfected with Pol II-driven GFP reporters and either PA-X or an empty vector. The hp-GFP construct contains a hairpin close to the 5’ end of the GFP mRNA that blocks its translation. Total cellular RNA was extracted and GFP mRNA levels were measured by RT-qPCR. GFP expression normalized by 18S rRNA abundance is plotted as fold change in PA-X vs. vector-transfected cells. Error bars = st. dev. (B) 293T iPA-X cells were transfected with a T7 polymerase expressing construct and an EMCV IRES-containing luciferase reporter driven by the T7 promoter, which results in T7-mediated luciferase RNA transcription. PA-X expression was induced 5 h after transfection by doxycycline addition for 18 h. Total cellular RNA was extracted and the levels of the luciferase reporter and endogenous actin mRNA were measured by RT-qPCR. RNA levels were normalized by 18S rRNA abundance and are plotted as fold change in induced (PA-X-expressing) vs. uninduced cells. (C-D) HEK 293T (C) or A549 (D) cells expressing doxycycline-inducible wild-type PA-X-myc (“iPA-X wt”), a catalytically inactive mutant (“iPA-X D108A”) or RFP (“iRFP”) were treated with doxycycline for 18 h to induce expression of PA-X. Total cellular RNA was extracted and the levels of the indicated endogenous RNAs were measured by RT-qPCR. Levels of actin mRNAs (also shown in Fig 1) and Pol II-driven non-coding RNAs were normalized to 18S rRNA levels and are plotted as fold change in induced (PA-X- or RFP-expressing) vs. uninduced cells. HEK293T iPA-X line #7 and A549 iPA-X wt line #10 and D108A line #8 were used for this figure. Data on additional clonal lines is presented in S2B and S2C Fig *,**,*** = p value (Student’s t-test) < 0.05, 0.01, 0.001 vs. 293T iRFP (C, n = 5) or A549 iPA-X D108A #8 (D, n = 3). Error bars = s.e.m.

Fig 4. PA-X selectively degrades Pol II transcripts in infected cells.

(A) Schematic representation of the PA gene of IAV showing the sequence of the frameshift site in the wild-type PR8 and the substitutions made to generate PR8-PA(fs) mutant virus. (B-D) A549 cells were infected with PR8 or PR8-PA(fs) at a multiplicity of infection (MOI) of 1. The fraction of infected cells displaying nuclear PABP was quantified (B) by immunofluorescence staining (C) of cells fixed at the indicated time points. A minimum of 200 cells were quantified in each sample. (D) Total RNA was harvested at 12 hours post-infection (hpi) and RT-qPCR was performed to measure levels of the indicated endogenous RNAs. Endogenous RNA levels were normalized to 18S rRNA levels and plotted as fold change in infected vs. uninfected cells. Error bars = st. dev. from three independent biological replicates. *,*** = p value < 0.05, 0.005 for PR8-PA(fs)- vs. PR8-infected cells.

Fig 5. PA-X activity facilitates viral protein synthesis.

(A) Crystal violet staining of plaques generated by wild-type PR8 and the PR8-PA(fs) mutant viruses in Vero cells, showing a decrease in plaque size for PA-X-deficient mutant compared to wild-type virus plaques. (B-D) A549 cells were infected with PR8 or PR8-PA(fs) IAV at a MOI of 0.25. (B) At 12 hpi cells were fixed and the infected cells were identified by immunofluorescence staining for IAV PA protein (green) to confirm that infection rates were highly similar between the two virus inoculums. Cell nuclei were stained with Hoechst dye (blue). (C) Viral protein accumulation was determined by western blotting of the whole cell lysates collected at the indicated time points. (D) Band intensities from (C) were quantified using ImageJ software (NIH) from two independent experiments. Error bars = st. dev. (E) Total RNA was isolated from cells infected as in (A-C) and the mRNA and vRNA levels for a subset of IAV genes were measured by RT-qPCR. Data was normalized to 18S rRNA and values plotted relative to 9 hpi levels in wild-type virus infected cells. (F) The levels of the indicated viral mRNAs at 12 hpi from (E) were plotted as fold change in PR8-PA(fs) infected cells vs. PR8 wild-type infected cells. (G) Rates of RNA synthesis were determined for the indicated genes at 9 hpi using Click-iT nascent RNA labeling and plotted as fold change in PR8-PA(fs) infected cells vs. PR8 wild-type infected cells. 18S values were normalized to input RNA concentration, whereas viral mRNA values were normalized to 18S levels. Error bars = st.dev. from three independent biological replicates. *,** = p value < 0.05, 0.01 for PR8-PA(fs)- vs. PR8-infected cells.

Fig 6. Canonical processing of the 3’ end of Pol II transcripts confers sensitivity to PA-X.

(A) Diagram of the GFP reporter constructs used in the figure, which bear different 3’ mRNA ends: the canonical poly(A) tail added by cellular processing machinery (poly(A)), a hammerhead ribozyme (HR), the hammerhead ribozyme proceeded by a 60-nt templated poly(A) stretch (A60-HR), or the histone stem-loop structure (hisSL). The arrows indicate where the HR self-cleaves the RNA. (B-D) HEK 293Ts were transfected with indicated Pol-II driven GFP reporters and either PA-X or empty vector. Total cellular RNA was extracted and GFP mRNA levels were measured by RT-qPCR (B) or northern blotting (C). GFP expression normalized by 18S rRNA abundance is plotted (B) or reported (C, relative levels) as fold change in PA-X-expressing vs. vector-transfected cells. (D) Levels of GFP for experiments in panel B were assessed by western blot, using tubulin as a loading control. A representative image is shown. ** = p value (Student’s t-test) < 0.01 for fold change of indicated reporter vs. fold change of poly(A) GFP. (E) HEK 293Ts were transfected with indicated Pol-II driven GFP reporters and PA-X from three different IAV strains or empty vector. Total cellular RNA was extracted and GFP or actin mRNA levels were measured by RT-qPCR. GFP expression normalized by 18S rRNA abundance is plotted as fold change in PA-X-expressing vs. vector-transfected cells. Error bars = s.e.m.

Fig 7. PA-X accumulates and functions in the cell nucleus.

(A and B) A549 cells expressing doxycycline-inducible wild-type PA-X-myc (“iPA-X wt”, line #10) or a catalytically inactive mutant (“iPA-X D108A”, line #8) were treated with doxycycline for 18 h to induce expression of PA-X. Total RNA was isolated from either nuclear or cytoplasmic fraction and RT-qPCR was performed to measure levels of the indicated endogenous RNAs. RNA levels were normalized by 18S rRNA levels and are plotted as fold change in induced (PA-X-expressing) vs. uninduced cells. In (A), target level changes in nuclear (nuc) and cytoplasmic (cyto) fractions are presented. In (B), changes in total nuclear mRNA and in unspliced pre-mRNA are compared. Error bars = s.e.m, n = 3–4. (C) Schematic diagram of the GFP fusion constructs generated in this study. (D) Amino acid sequences of the X-ORF from PR8 strain (X61) aligned to a shortened 41 amino-acid construct (X41) and the mutant X-ORF with alanine substitutions at four basic residues (X61(4A)). Amino acids that are highly conserved between IAV strains are shown in bold and the basic residues that are important for PA-X function are highlighted in blue. (E and F) Fluorescence microscopy images of 293A cells transfected with the expression constructs for the indicated GFP fusion proteins listed in (C). Two fields of view are shown for each transfection experiment. Representative cells with nuclear GFP signal accumulation are highlighted with filled arrowheads and those that have even signal distribution with open arrowheads. Nuclei were stained with Hoechst dye. (G and H) Relative Firefly (FF-luc, G) and Renilla (RE-luc, H) luciferase reporter expression in 293A cells co-transfected with these reporters and the indicated wild-type and mutant PA-X constructs. Error bars = st. dev. from four independent biological replicates. *,** = p value < 0.05, 0.01 (**); ns = p value > 0.05 (Anova single factor).