Regulatory hotspot on the influenza A virus polymerase revealed through the structure of the NEP-polymerase complex

Abstract

Influenza A virus (IAV) transcribes and replicates its segmented RNA genome in the host nucleus within viral ribonucleoproteins (vRNPs), which are exported for virion assembly. The nuclear export protein (NEP) is essential for this process and also regulates viral RNA synthesis, implicating a direct interaction with the viral RNA polymerase. Here, we present a 2.5-Å cryogenic electron microscopy structure of NEP bound to the IAV polymerase and demonstrate that NEP alone is sufficient to promote vRNP export, with the viral matrix protein 1 enhancing export efficiency. NEP forms a four-helix bundle that binds at the interface of the PA C-terminal domain and PB1 N terminus of the polymerase. The NEP binding site at this interface overlaps with those for the host ANP32 and the C-terminal domain of RNA polymerase II, indicating that it functions as a regulatory hotspot coordinating transitions of the viral polymerase between RNA synthesis and nuclear export, revealing a critical layer of control in the IAV replication cycle.

Fig. 1.. NEP is required for vRNP nuclear export.

(A) Schematic of the vRNP reconstitution assay. Cells were transfected with plasmids expressing PB2, PB1, PA, NP, and NA vRNA, in the presence or absence of GFP-NEP and/or M1. PA was omitted as negative control (−PA). vRNP localization was assessed using smFISH probes targeting NA vRNA. Cellular and nuclear boundaries were delineated using GFP fluorescence and DAPI staining, respectively. The cytoplasmic-to-nuclear (C/N) ratio was calculated by dividing the mean fluorescence intensity in the cytoplasm by that in the nucleus. (B) Representative images of Vero E6 cells transfected with vRNP reconstitution components and fixed 24 hours post-transfection. NA vRNA signals are shown in the upper panels; DAPI staining of nuclei is shown in the lower panels. Scale bars, 10 μm. (C) Quantification of the C/N ratio from individual cells across two independent biological replicates: vRNP only (n = 100), M1 (n = 105), NEP (n = 132), and NEP + M1 (n = 109). The black lines represent medians. Statistical analysis was performed using nonparametric one-way analysis of variance (ANOVA) with multiple comparisons relative to the vRNP-only condition, ****P ≤ 0.0001.

Fig. 2.. Structure of the pandemic A/Brevig Mission/1/18 (H1N1) IAV NEP-polymerase complex.

(A) Cryo-EM map of the NEP-polymerase complex. (B) Cartoon model of the NEP-polymerase complex. Dashed rectangle denotes the close-up view shown in (C). (C) Close-up view of the NEP-polymerase interface demonstrating four pairs of interacting amino acid residues.

Fig. 3.. NEP interacts with the viral polymerase in cells via the NEP-polymerase interface observed by cryo-EM.

(A) Split-luciferase complementation assay schematic. (B) HEK-293T cells were cotransfected with plasmids expressing NEP-Luc1 and the indicated luciferase-tagged polymerase subunit in the presence or absence of the remaining two polymerase subunits. Luminescence was measured 24 hours post-transfection and quantified as the normalized luminescence ratio (NLR). (C) NEP-polymerase interactions measured by split-luciferase assay using NEP-Luc1 and PA-Luc2 in the presence or absence of the full polymerase complex. Relative NLR was calculated by normalizing the NLR to the value in condition containing the full polymerase complex, set as 100%. (D) NEP-polymerase complex cryo-EM density map with superimposition of polymerase structures bound by Nb8191 (PDB 7NIR), Nb8196 (PDB 7NJ3), and Nb8210 (PDB 7NKR). (E) Effect of nanobodies on NEP-polymerase interactions measured by split-luciferase assay. Indicated nanobodies were coexpressed in the split-luciferase system. NLR values were normalized to vector control, set as 100%. (F) Effect of NEP interface mutations on NEP-polymerase interactions determined by split-luciferase assay. The 4A mutant was analyzed in a separate experiment with wild-type (WT) NEP. NLR values were normalized to the value of WT NEP, set as 100% within each experiment. The 4A mutant was then plotted alongside the individual interface mutants for the clarity of visualization. Western blots show expression levels of WT and interface mutants NEP-Luc1 and PA-Luc2, with GAPDH as a loading control. Graphs represent the NLR or relative NLR of three biological replicates (mean ± SEM, n = 3). Significance was determined using either two-tailed, unpaired Student’s t test (B) or ordinary one-way ANOVA with multiple comparisons relative to (C) full polymerase complex, (E) Vector, (F) WT, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001.

Fig. 4.. Effect of NEP mutants on vRNP localization.

(A) Vero E6 cells were transfected with vRNP reconstitution components in the presence of GFP-tagged wild-type (WT) NEP or interface mutant NEPs or the absence of NEP (−NEP) and fixed 24 hours post-transfection. vRNP localization was assessed using smFISH probes targeting NA vRNA. Nuclear boundaries were delineated by DAPI staining (dashed lines; see fig. S8 for corresponding DAPI staining and GFP signal). Scale bars, 10 μm. (B) Quantification of cytoplasmic-to-nuclear (C/N) vRNA signal intensity ratios corresponding to (A). The black lines represent medians. (C) Same as in (A), with the addition of M1 expression. (D) Quantification of cytoplasmic-to-nuclear (C/N) vRNA signal intensity ratios corresponding to (C). Cell counts per condition for (B) [NEP only]: WT (n = 132), R15A (n = 114), R42A (n = 119), Q96A (n = 122), Q101A (n = 116), 4A (n = 117), 121A (n = 112), and vRNP only (n = 100). For (D) [NEP + M1]: WT (n = 109), R15A (n = 117), R42A (n = 114), Q96A (n = 123), Q101A (n = 123), 4A (n = 133), 121A (n = 126), and vRNP only (n = 105). Quantitation was performed from two independent biological replicates. Statistical analysis was performed using nonparametric one-way ANOVA with multiple comparisons to WT NEP condition, *P ≤ 0.05, **P ≤ 0.01, and ****P ≤ 0.0001.

Fig. 5.. Effect of NEP mutants on viral RNA synthesis.

(A) Primer extension analysis of viral RNAs from vRNP reconstitution assays in the presence of increasing concentrations of wild-type (WT) NEP or vector only (Vec). Omission of PB1 (−PB1) served as a negative control and 5S rRNA was used as a loading control. Western blotting using antibodies against NEP and GAPDH was conducted to assess protein levels of NEP and GAPDH (loading control). (B) Quantification of the viral RNA species from (A). Error bars represent the SEM (mean ± SEM, n = 4). Statistical significance was determined using ordinary one-way ANOVA with multiple comparisons to Vec condition, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. (C) Primer extension analysis of viral RNAs from vRNP reconstitution assays in the presence of low (25 ng) or high concentrations (1600 ng) of wild-type or indicated interface mutant NEPs or vector only (Vec). 5S rRNA was used as a loading control. Western blotting using antibodies against NEP and GAPDH was conducted to assess protein levels of NEP and GAPDH (loading control). (D) Quantification of the viral RNA species from (C). Error bars represent the standard error of the mean from three independent biological replicates (mean ± SEM, n = 3). Statistical significance was determined using ordinary one-way ANOVA with multiple comparisons to Vec condition, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001. (E) Schematic representation of NEP protein depicting interface point mutations. (F) Cryo-EM model of the NEP-polymerase complex with the polymerase in surface representation and the NEP modeled in cartoon representation with superimposition of cartoon models ANP32B (PDB 8R1L) and Pol II CTD peptide mimic (PDB 8R60).

Fig. 6.. Roles of NEP in vRNP nuclear export and regulation of viral RNA synthesis.

NEP, by competing with Pol II CTD and ANP32 for polymerase binding, inhibits viral transcription and genome replication, respectively, and promotes nuclear export. Without NEP, the vRNP cannot export, and with M1, export is more efficient.