The human H5N1 influenza A virus polymerase complex is active in vitro over a broad range of temperatures, in contrast to the WSN complex, and this property can be attributed to the PB2 subunit

Abstract

Influenza A virus (IAV) replicates in the upper respiratory tract of humans at 33 degrees C and in the intestinal tract of birds at close to 41 degrees C. The viral RNA polymerase complex comprises three subunits (PA, PB1 and PB2) and plays an important role in host adaptation. We therefore developed an in vitro system to examine the temperature sensitivity of IAV RNA polymerase complexes from different origins. Complexes were prepared from human lung epithelial cells (A549) using a novel adenoviral expression system. Affinity-purified complexes were generated that contained either all three subunits (PA/PB1/PB2) from the A/Viet/1203/04 H5N1 virus (H/H/H) or the A/WSN/33 H1N1 strain (W/W/W). We also prepared chimeric complexes in which the PB2 subunit was exchanged (H/H/W, W/W/H) or substituted with an avian PB2 from the A/chicken/Nanchang/3-120/01 H3N2 strain (W/W/N). All complexes were functional in transcription, cap-binding and endonucleolytic activity. Complexes containing the H5N1 or Nanchang PB2 protein retained transcriptional activity over a broad temperature range (30-42 degrees C). In contrast, complexes containing the WSN PB2 protein lost activity at elevated temperatures (39 degrees C or higher). The E627K mutation in the avian PB2 was not required for this effect. Finally, the avian PB2 subunit was shown to confer enhanced stability to the WSN 3P complex. These results show that PB2 plays an important role in regulating the temperature optimum for IAV RNA polymerase activity, possibly due to effects on the functional stability of the 3P complex.

Fig. 1.

Subcellular distribution and activity of purified, adenovirus-expressed WSN polymerase. (a) A549 cells were transduced with rAd5-PA_WSN-TAP alone or with a combination of rAd5-PA_WSN-TAP, -PB1_WSN and -PB2_WSN (W/W/W). Nuclear and cytoplasmic extracts were analysed by immunoblotting using PA-, PB1- or PB2-specific antibodies. The effectiveness of the subcellular fractionation method was confirmed by probing against the cytoplasmic proteins, actin and β-tubulin as well as against RanBP5 (which is found both in the cytoplasm and nucleus). ns, non-specific protein band detected when probing against PB1/2. Cells transduced with an adenovirus vector expressing LacZ served as a negative control. Twofold more cell equivalents were loaded for the nuclear extracts than the cytoplasmic extracts. (b) IgG-purified nuclear and cytoplasmic extracts were analysed by silver staining prior to (c) comparison with total cell lysate in an ApG-primed transcription assay (1 h at 30 °C). Extracts from equivalent numbers of transduced A549 cells were used. (d) IgG-purified extracts from total cellular lysates were analysed by silver staining (upper panel) and immunoblotting against RanBP5 (lower panel). The positions of PA, PB1 and PB2 are shown on the right.

Fig. 2.

The H5N1 3P complex is more active than the WSN 3P complex in both cap-independent and cap-dependent transcription. (a) Purified 3P complexes from the VN1203 (H/H/H) and WSN (W/W/W) viruses, along with chimeric (H/H/W and W/W/H) and dimeric complexes (H/H; negative control), were detected by silver staining and immunoblotting against PB2 (top and bottom panel, respectively). PB2-normalized complexes were used in functional assays (30 °C). (b) The complexes were analysed for cap-independent (ApG) or cap-dependent transcription (top and bottom panel, respectively). The transcription products in the ApG-primed assay (14 nt long transcript) and in the globin-primed assay (triplet band; capped primer +14 nt) were detected by autoradiography. (c, d) Quantitative data for the ApG- and the cap-RNA-primed assays are shown. The data were normalized in terms of fold activity over the activity of the W/W/W complex. The results shown represent mean transcriptional activity data from at least three independent assays. Bars, sem.

Fig. 3.

The human H5N1 3P complex is active at high temperatures and the H5N1 PB2 subunit confers this same phenotype on the WSN 3P complex. Equivalent amounts of polymerase complex (as determined by polymerase activity at 30 °C) were tested for transcriptional activity at different temperatures (30, 34, 37, 39, 42 °C). (a, b) Representative autoradiograms and quantitative data (top and bottom panel, respectively) for the ApG- (a) and the cap-RNA-primed (b) assays are shown. The data were normalized in terms of percentage of the amount of fully extended product at 30 °C (as quantified by Phosphorimager analysis); the results shown represent mean transcriptional activity data from at least four independent assays. Bars, sem.

Fig. 4.

An avian PB2 subunit increases the activity of the WSN 3P complex at 30 °C. (a) Purified 3P complexes from the WSN (W/W/W) virus were prepared, along with complexes in which the PB2 subunit was omitted (W/W) or substituted by the PB2 from the avian Nanchang virus (W/W/N) or a mutated derivative (W/W/N_E627K). The purified complexes were detected by silver staining and immunoblotting against PB2 (top and bottom panel, respectively). PB2-normalized complexes were used in functional assays (30 °C). (b) The complexes were analysed for cap-independent (ApG) or cap-dependent transcription (top and bottom panel, respectively). The transcription products in the ApG primed assay (14 nt long transcript) and in the globin-primed assay (triplet band; capped primer +14 nt) were detected by autoradiography. (c, d) Quantitative data for the ApG- and the cap-RNA-primed assays are shown. The data were normalized in terms of fold activity over the activity of the W/W/W complex. The results shown represent mean transcriptional activity from at least three independent assays. Bars, sem.

Fig. 5.

An avian PB2 subunit confers broad temperature tolerance on the WSN 3P complex independently of amino acid residue 627. Purified 3P complexes were prepared from the WSN virus (W/W/W), along with complexes in which the PB2 subunit was substituted by an avian PB2 (W/W/N) or a mutated derivative (W/W/N_E627K); an extract prepared from LacZ-expressing cells (LacZ) was used as a negative control. Equivalent amounts of polymerase complex (as determined by polymerase activity at 30 °C) were tested for transcriptional activity at different temperatures (30, 34, 37, 39, 42 °C). Representative autoradiograms (a, c) and quantitative data (b, d) are shown for the ApG- (a, b) and the cap-RNA-primed (c, d) assays. The data were normalized in terms of percentage of the amount of fully extended product at 30 °C (as quantified by Phosphorimager analysis); the results shown represent mean transcriptional activity from at least four independent assays. Bars, sem.

Fig. 6.

An avian PB2 subunit confers enhanced thermostability on the WSN 3P complex. Equivalent amounts of polymerase complex (as determined by polymerase activity at 30 °C) were pre-incubated at 30 °C in the absence of vRNA for different time periods. The complexes were then tested in an ApG-primed transcription assay (30 °C, 1 h). (a) A representative autoradiogram and (b) quantitative data from replicate experiments are shown. The data were normalized to the amount of fully extended product where no pre-incubation was performed (as quantified by Phosphorimager analysis). Mean transcriptional activities from at least four independent assays are shown. Bars, sem. A nonlinear curve for a single-phase exponential decay was fitted to the data using GraphPad Prism. The r² value (goodness of fit) for the W/W/W curve was 0.81, and the half-life of the W/W/W complex was estimated at 7.3 min (with 95 % confidence intervals ranging from 5.3 to 11.8 min). The half-life of the W/W/N complex could not be determined, as it exceeded the 2 h of our analysis. (c) The purified WSN 3P complex was pre-incubated at the indicated temperatures in the presence (left panel) or absence of both 5′- and 3′-vRNA (right panel). The proteins were pulled-down with NiNTA Magnetic Agarose Beads and detected by immunoblotting with antibodies directed against PA (anti-His), PB1 and PB2.