G Protein Subunit β1 Facilitates Influenza A Virus Replication by Promoting the Nuclear Import of PB2

ABSTRACT

G protein subunit β1 (GNB1), the beta subunit of the G protein family, plays an important role in regulating transmembrane signal transduction. Although a recent study has demonstrated that GNB1 can bind the matrix protein 1 (M1) to facilitate M1 transport to budding sites and promote the release of progeny influenza A virus (IAV), whether the GNB1 protein has other functions in IAV replication requires further study. Here, we found that GNB1 promoted IAV replication, as virus yield decreased in GNB1 knockdown or knockout cells. GNB1 interacted with polymerase subunits PB2, PB1, and PA. Overexpressed GNB1 facilitated PB2 binding to importin α3, α5, and α7 promoting the nuclear import of PB2, enhancing viral RNA synthesis and polymerase activity. Altogether, our results demonstrated that GNB1 positively regulates virus replication by interacting with polymerase subunits and facilitating the nuclear import of PB2, which provide novel insights into the molecular mechanism of IAV.

IMPORTANCE Until now, there has been only one article on the role of GNB1 in IAV budding. No study has investigated the role of GNB1 in IAV replication. In this study, our research demonstrated that GNB1 could increase the interaction between PB2 and the importin α isoform and mediate the nuclear import of PB2. Therefore, GNB1 could promote viral replication and transcription. Our results provide a better understanding of the molecular mechanisms of viral replication and provide potential antiviral drug targets.

INTRODUCTION

Influenza A virus (IAV), an orthomyxovirus, is a major respiratory pathogen that affects human and animal health (1–3). The viral genome comprises an 8-segmented negative-sense RNA, which can encode at least 11 proteins (4, 5). The difference between IAV and most RNA viruses is that the transcription and replication of IAV occur in the nucleus (4, 6, 7). Viral ribonucleoprotein complexes (vRNPs), which are composed of the RNA-dependent RNA polymerase complex (RdRp; composed of PB1, PB2, and PA) and multiple nucleoprotein (NP) copies, play a crucial role in viral transcription (vRNA to mRNA) and replication (vRNA to cRNA [a complementary copy of vRNA] to vRNA) (8–10). When the virus completes internalization, the vRNPs are transported into the nucleus and regulate viral transcription and replication. The newly synthesized polymerase protein subunits (PB2, PB1, and PA) must be imported into the nucleus, assembled, and form a functional trimer to become functional (11–13). Among the numerous models of nuclear import, that PB2 enters the nucleus as a monomer, and PA and PB1 enter as dimers, is accepted more widely (13–15).

Nuclear localization signals (NLSs) or nuclear export signals (NESs) of specific cargo proteins can be recognized and bound by nuclear transport receptors to transport across the nuclear pore complex (NPC) (16–18). NLSs usually comprise a single cluster basic residue or a bipartite motif, in which short linkers separate two base regions (19, 20). In the classical nuclear import pathway, the NLS of cargo proteins binds to the isoform of importin α and is then associated with importin β1 to transport to the nucleus (18, 21). Several studies have proved that all polymerase components of IAV contain an NLS of nuclear import (22–25). Although certain cellular proteins have also been demonstrated to regulate IAV replication by affecting the nuclear import of polymerase subunits, the detailed mechanism of the nuclear import of viral proteins remains unclear, and potential host factors have not yet been characterized completely.

Guanine nucleotide-binding regulatory proteins (G proteins), which mediate transmembrane signal transduction, comprise Gα, Gβ, and Gγ subunits (26–28). G proteins are dependent on binding guanine nucleotides to undergo activation-inactivation and regulate various key cellular processes (29, 30). Many studies have shown that G proteins play important roles in the IAV life cycle by directly or by indirectly binding to the viral protein (31–34). GNB1 encodes the beta subunit, which plays an important role in regulating signal transduction receptors and effectors (35, 36). However, the role of GNB1 in the life cycle of influenza virus remains unclear. In the present study, we discovered that GNB1 could interact with the polymerase subunits and facilitate IAV replication by promoting PB2 binding to the importin α family. Altogether, these data showed that GNB1, a novel host factor promoting IAV replication, could be an ideal antiviral target.

RESULTS

GNB1 facilitates influenza virus replication.

Previous studies have shown that many host proteins are packaged in the viral particles (37, 38), and among them, several proteins have been shown to play a role in viral replication (39–42). Therefore, we inferred that the host proteins in the virus particles might contribute to viral transmission or replication in the host. Thus, we aimed to discover novel host proteins in influenza virus particles and explore their functions in viral replication. The avian influenza virus A/Chicken/Hunan/8.27 YYGK3W3/2018 (H9N2) was passaged in mice, and bronchoalveolar lavage (BAL) fluid was collected at 3 days postinfection after serial passage in mice. The virus in the BAL fluid specimen was purified, the whole genome of the virus was sequenced, and proteomic spectra were analyzed. Then, we selected human homolog protein PDCDI6P, TMSB4X, IMMT, ARRB1, RPS20, GNB1, PDIA3, SLC9A3R2, SNRPD3, and S100A6 by integrative the bioinformatics analysis and detected their effects of influenza virus replication. A549 cells were transfected with the small interfering RNA (siRNAs) (Table 1) targeted to each protein or negative siRNA (siNC) for 36 h, and the efficiency of siRNA silencing was detected by reverse transcription-quantitative PCR (qRT-PCR) (Fig. 1A). At the same time, the A549 cells were infected with the H9N2-Nluc or PR8-Rluc virus 36 h after siRNA transfection, and the fluorescence value was detected 24 h after infection. The result indicated that knockdown of GNB1 dramatically decreased IAV replication compared with that in of the siNC control (Fig. 1B and C). We further validated IAV replication by knocking down GNB1 in A549 cells using siRNA. The knockdown efficiency of GNB1 in A549 cells was confirmed (Fig. 2A). GNB1 knockdown A549 cells (si-1 or si-2) and A549 control cells (siNC) were infected with H9N2 (mouse-adapted virus) or PR8 at a multiplicity of infection (MOI) of 0.01. Virus titers and NP protein expression showed that H9N2 or PR8 virus titers significantly decreased in GNB1 knockdown A549 cells compared with those in control cells (Fig. 2B and C). To confirm that GNB1 facilitates IAV replication, we transfected A549 cells with the Flag-GNB1 plasmid or Flag plasmid. Then, the cells were infected with H9N2 and PR8 virus at an MOI of 0.01. At the indicated time points, the cell supernatants were harvested, and virus titers and NP protein expression were tested. The virus titers of H9N2 and PR8 viruses significantly increased in GNB1-overexpressed A549 cells compared with those in the A549 control cells (Fig. 2D and E). We also established GNB1 knockout (GNB1-KO) A549 cells using the CRISPR-Cas9 system (Fig. 3A). Virus titers of H9N2 and PR8 were significantly decreased in GNB1-KO A549 cells compared with those in wild-type (WT) A549 cells (Fig. 3B and C). To further verify the above findings, GNB1-KO or WT A549 cells were transfected with the Flag-GNB1(T) plasmid (a synonymous mutation against a GNB1 sgRNA sequence) or pCAGGS-Flag for 24 h, and the cells were infected with the WT H9N2 or PR8 virus. The virus titers of H9N2 and PR8 viruses were significantly increased in Flag-GNB1(T)-overexpressing GNB1-KO A549 cells compared with those in the GNB1-KO A549 cells (Fig. 3D and E). Together, our results indicated that GNB1 facilitates IAV replication.

TABLE 1.

The sequence of siRNA was used for gene knockdown

siRNA targeting gene name	Sequence (5′–3′)
si-PDCD6IP-1	GAAGGATGCTTTCGATAAA
si-PDCD6IP-2	GGCACAGGCTCAAGAAGTA
si-PDCD6IP-3	GAACCTGGATAATGATGAA
si-TMSB4X-1	GTCGAAACTGAAGAAGACA
si-TMSB4X-2	CTGAGATCGAGAAATTCGA
si-TMSB4X-3	CCACTGCCTTCCAAAGAAA
si-IMMT-1	GTGGAGGCATTAAAGTACA
si-IMMT-2	CTGACAAGCTCTCTACTGA
si-IMMT-3	GTCTGAGGCTAAGGTTGTA
si-ARRB1-1	GAGCACGCTTACCCTTTCA
si-ARRB1-2	CCAGTAGATACCAATCTCA
si-ARRB1-3	GGATCATTGTTTCCTACAA
si-RPS20-1	GAGATTGTTAAGCAGATTA
si-RPS20-2	CTGAGATTGTTAAGCAGAT
si-RPS20-3	GTGTGCTGACTTGATAAGA
si-GNB1-1	CGTCTGGGATGCACTCAAA
si-GNB1-2	CCGAGCAACTTAAGAACCA
si-GNB1-3	GTTCGTCTCTGGTGCTTGT
si-PDIA3-1	GCCGTGAATTAAGTGATTT
si-PDIA3-2	CGAGTATGATGATAATGGA
si-PDIA3-3	AGACCCAAATATCGTCATA
si-SLC9A3R2-1	GTTCCGACAAGGACACTGA
si-SLC9A3R2-2	CGTGAAATCTTCAGCAACT
si-SLC9A3R2-3	TCACCCGTCACCAATGGAA
si-SNRPD3-1	GCCACATTGTGACATGTGA
si-SNRPD3-2	GAAGAACGCACCCATGTTA
si-SNRPD3-3	CGAACACCGGTGAGGTATA
si-S100A6-1	GCAGGATGCTGAAATTGCA
si-S100A6-2	TGGCCATCTTCCACAAGTA
si-S100A6-3	GGCTGATGGAAGACTTGGA

FIG 1.

Candidate protein knockdown effects on IAV replication. (A) The silencing efficiency of the siRNA target to the candidate protein. The A549 cells were transfected with siRNA for 36h, and the mRNA level of a gene was detected by real-time PCR. (B and C) Silencing candidate protein effects on virus replication. A549 cells were transfected with the individual siRNA targeted to a protein for 36 h then infected with H9N2-Nluc (B) and PR8/H1N1-Rluc (C). After 24 h of infection, luciferase activity was detected by using the Suitable kit. (*, P < 0.05; **, P < 0.01; ***, P < 0.001; two-tailed Student’s t test). Rluc, Renilla luciferase; Nluc, nanoluciferase; RLU, relative luminescence units.

FIG 2.

GNB1 promotes influenza A virus (IAV) replication. (A) The efficiency of siRNA-mediated GNB1 silencing. The protein level of GNB1 was detected by Western blot at 48 h after A549 cells were transfected with siRNAs targeted to GNB1 (si-1, si-2, and si-3) or control siRNA (siNC). GAPDH was used for the normalization of gene expression. (B and C) The effects of GNB1 siRNAs on IAV replication. A549 cells were transfected with si-1, si-2, or siNC for 36 h and then infected with the H9N2 virus (B) or PR8/H1N1 virus (C) at an MOI of 0.01. The lysates and cell supernatants were harvested at different time points postinfection. The protein levels of the viral NP protein and GNB1 were detected by Western blotting, and virus titers were determined by plaque assay on MDCK cells (mean ± SD of three independent experiments; *, P < 0.05; **, P < 0.01; ***, P < 0.001; two-tailed Student’s t test). (D and E) The effects of GNB1 overexpression on IAV replication. A549 cells were transfected with Flag-GNB1 and pCAGGS-Flag as a negative control for 24 h and then infected with the H9N2 virus (D) and PR8/H1N1 (E) virus at an MOI of 0.01. The cell lysates and supernatants were harvested at the indicated time points postinfection. The protein levels of viral NP protein and GNB1 were detected by Western blotting, and the virus titer was determined by plaque assay on MDCK cells (mean ± SD of three independent experiments; *, P < 0.05; **, P < 0.01; ***, P < 0.001; two-tailed Student’s t test).

FIG 3.

Effect of GNB1 knockout on IAV replication. (A) Knockout efficiency of GNB1-KO A549 cells. The protein level of GNB1 was determined by using Western blotting. (B and C) Virus replication was reduced in GNB1-KO A549 cells. Wild-type A549 cells (WT) and GNB1-KO A549 cells (KO) were infected with the H9N2 virus (B) or PR8/H1N1(C) virus at an MOI of 0.01. The cell supernatants were harvested at 12, 24, 36, and 48 hpi. Virus titers were determined using plaque assay on MDCK cells (mean ± SD of three independent experiments; **, P < 0.01; ***, P < 0.001; two-tailed Student’s t test). (D and E) Effect of the reconstitution of GNB1 in the KO cells on IAV replication. The GNB1-KO cells were transfected with reconstituted Flag-GNB1 or pCAGGS-Flag vector for 24 h, and the WT cells were also transfected with a pCAGGS-Flag vector for 24 h and then infected with the H9N2 virus (B) or PR8/H1N1(C) virus at an MOI of 0.01. The cell supernatants also were harvested at 12, 24, and 36 hpi. Virus titers were determined by using plaque assay on MDCK cells (*, P < 0.05; **, P < 0.01; ***, P < 0.001; two-tailed Student’s t test). (F and G) WT A549 and GNB1-KO cells were infected with VSV-GFP at MOI of 0.1 or not for 24 h. The fluorescence was examined using fluorescence microscopy (F). The GFP and GNB1 protein level were detected by Western blot (G).

To future ascertain whether GNB1 regulation of IAV replication is specific to this virus, we investigated the role of GNB1 in vesicular stomatitis virus (VSV) replication. The GNB1-KO A549 cells and WT A549 cells were mock infected or VSV-green fluorescent protein (GFP)-infected for 24 h. There was no difference observed in GFP fluorescence and GFP protein levels between the GNB1-KO and WT A549 cell groups (Fig. 3F and G), indicating that GNB1 might specifically regulate influenza virus replication.

GNB1 interacts with IAV polymerase subunits.

To determine whether GNB1 interacts with viral proteins, PB2, PB1, PA, NA, NP, HA, M1, or NS1 with hemagglutinin (HA) tags or NS2-GFP or M2-GFP were cotransfected with Flag-GNB1 in 293T cells. Coimmunoprecipitation (co-IP) was then performed with an anti-HA or anti-GFP monoclonal antibody (Mab). The results showed that PB2, PB1, and PA coprecipitated with Flag-GNB1 (Fig. 4A). In contrast, the other viral proteins, including M1, did not coprecipitate with Flag-GNB1 (Fig. 4A to C). Liu et al. indicated that the GNB1 can interact with the M1 protein of an H9N2 virus, which is not in accordance with our data (32). To determine whether the GNB1 can bind to the M1 from different IAV subtypes, co-IP was performed. The results suggested that the GNB1 could bind to M1 from H3N2 but not to M1 from H9N2 and H1N1(PR8) (Fig. 4D). Meanwhile, reverse co-IP was performed for Flag-GNB1 and PB2, PB1, and PA and Flag-GNB1 also coimmunoprecipitated with PB2, PB1, and PA (Fig. 4E). The same co-IP experiments were performed after cell lysates were treated with RNA A/T1 for 1 h to determine whether PB2, PB1, or PA interacting with GNB1 depends on RNA binding (i.e., via the viral RNA genome). The results showed that Flag-GNB1 coprecipitated with PB2, PB1, and PA after RNase treatment, indicating that GNB1 interacts with PB2, PB1, and PA in an RNA-independent manner (Fig. 4F). To further confirm this interaction, we performed a glutathione S-transferase (GST) pulldown assay. PB2, PB1, and PA were pulled down by GST-GNB1 (Fig. 4H). Moreover, A549 cells were infected with the H9N2 virus at an MOI of 2, and the cells were immunoprecipitated with an anti-GNB1 antibody or control IgG at 12 h postinfection (hpi). The results suggested that PB2, PB1, and PA interacted with GNB1 in their native state (Fig. 4G).

FIG 4.

GNB1 interacts with viral polymerase subunits, PB2, PB1, and PA. (A to C) Co-IP assay results of Flag-GNB1 and viral proteins. The 293T cells were transfected with Flag-GNB1 and HA-PB2, HA-PB1, HA-PA, HA-NP, HA-NA, HA-HA, HA-M1, HA-NS1, M2-GFP, NS2-GFP, and GFP, or the indicated combinations of plasmids. After 24 h, the cell lysates were immunoprecipitated with an anti-HA antibody (A and B) or an anti-GFP antibody (C). The levels of individual proteins were detected by Western blot. (D) Co-IP assay results of endogenous GNB1 and M1 from H9N2, H1N1(PR8), or H3N2. The A549 cells were infected with H9N2, H1N1(PR8), or H3N2 virus. After 24 h, cell lysates were immunoprecipitated using an anti-M1 antibody or rabbit IgG as a control. (E and F) The 293T cells were transfected with Flag-GNB1 and HA-PB2, HA-PB1, and HA-PA or the indicated combinations of plasmids. After 24 h, the cell lysates were treated with 100 U RNase A/T at 37°C for 1 h (F) or not (E) and were immunoprecipitated with a mouse anti-Flag MAb (E) and a mouse anti-HA MAb (F). The levels of individual proteins were detected by Western blot. (G) Identification of the interactions between endogenous GNB1 and polymerase components. A549 cells infected with the H9N2 virus (MOI of 2) were lysed at 12 hpi. Cell lysates were immunoprecipitated using an anti-GNB rabbit antibody or rabbit control IgG. (H) GST pulldown assay of GNB1 and polymerase subunit. The GNB1-GST plasmids and GST plasmids were transfected into A549 cells. Cell lysates were purified by using a Mag-Beads GST kit and then GNB1 (lane 1) and GNB1-GST (lane 2) were determined by using Coomassie blue (CB) staining. The lysates of 293T cells transfected with HA-PB2, HA-PB1, and HA-PA were mixed individually with GST-GNB1 on Mag-GST beads and then examined using Western blot with a rabbit anti-GST MAb or mouse anti-HA MAb. WCL, whole-cell lysates.

Since GNB1 can coprecipitate with PB2, PB1, and PA, we attempted to define the key binding region between GNB1 and PB2, PB1, and PA. PB2, PA, and PB1, and their truncated derivatives were constructed (Fig. 5A, B, and E). The results showed that the N terminus and C terminus of PB2 could bind with GNB1 (Fig. 5C); the C terminus of PA exhibited sufficient binding capability to GNB1 (Fig. 5D); and the N-terminal and polymerase domains of PB1 strongly bind GNB1 (Fig. 5F).

FIG 5.

GNB1 interacts with truncated polymerase subunits. (A) Schematic illustration of PB2 and its truncated derivatives. (B) Schematic diagram of PA and its truncated derivatives. (C, D, and F) Interaction between GNB1 and polymerase subunits and the truncated derivatives. 293T cells were transfected with the indicated plasmids for 24 h, then cells were lysed, and immunoprecipitation was performed using a mouse anti-HA MAb. The levels of protein were detected individually using Western blot. (E) Schematic diagram of PB1 and its truncated derivatives. (G) Colocalization of endogenous GNB1 and polymerase subunits during infection. A549 cells were infected with the H9N2 virus or not for 3 h. Samples were fixed and stained with rabbit anti-PB2, anti-PB1, anti-PA pAb, and mouse anti-GNB1 MAb, followed by incubation with CoraLite488-conjugated Affinipure goat anti-mouse IgG (H+L) (green) and CoraLite594-conjugated goat anti-rabbit IgG(H+L) (red). The nucleus was stained with DAPI (blue). NLS, nuclear localization sequence; NBM, nucleotide-binding motif.

To assess the colocalization of GNB1 with PB1, PA, or PB2, A549 cells were infected with the H9N2 virus to explore the interaction between endogenous GNB1 and PB1, PB2, or PA. GNB1 can be found in the whole cell in the infected cells, whereas it is localized mainly in the cytoplasm in the uninfected cells. PB2, PB1, PA, and GNB1 can be detected in the cytoplasm in the infected cells (Fig. 5G), and their Pearson correlation coefficients were 0.53, 0.67, 0.86, respectively, which implied that the PB2/PB1/PA and GNB1 might colocalize in the space.

GNB1 knockout reduces viral RNA synthesis.

Next, we determined whether GNB1 affects viral RNA synthesis. GNB1-KO and WT A549 cells were infected with H9N2 virus. The NP vRNA, cRNA, and mRNA were measured at 4, 6, and 8 hpi using qRT-PCR. The results showed that the levels of the three RNAs were significantly lower in GNB1-KO cells than those in WT cells at all the time points (Fig. 6A to C). To further confirm the above data, A549 cells transfected with GNB1 plasmid for 24 h were infected with H9N2, and three types of RNA at different time points were tested. As expected, the levels of the three viral RNAs were significantly increased in the GNB1-overexpressed cells (Fig. 6D to F). IAV genome replication depends on newly synthesized viral proteins, while the primary transcription does not. The cycloheximide (CHX) can inhibit the protein synthesis in the cells. To explore whether GNB1 affects viral primary transcription, the WT and GNB1-KO cells were treated with CHX or without and then infected with H9N2, and the level of viral NP mRNA was detected. The results showed that NP mRNA of GNB1-KO cells without CHX was obviously decreased compared with that in WT cells without CHX, while levels of mRNA were similar in both cells with CHX treatment (Fig. 6G and H), suggesting the GNB1 does not participate in primary transcription of H9N2 virus. In conclusion, these data demonstrated that GNB1 promotes viral RNA synthesis after primary transcription.

FIG 6.

GNB1 affects the synthesis of viral RNA, cRNA, and mRNA. (A to C) Effect of GNB1-KO on NP RNA synthesis. The GNB1-KO A549 cells and WT A549 cells were infected with the H9N2 virus at an MOI of 1. The samples were collected at 4, 6, and 8 hpi, and the virus RNA levels were normalized to the GAPDH mRNA level (*, P < 0.05; **, P < 0.01, two-tailed Student’s t test). (D to F) Effect of GNB1 overexpression on NP RNA synthesis. The GNB1-KO A549 cells and WT A549 cells were transfected with the Flag-GNB1 or pCAGGS-Flag vectors for 24 h, and cells were infected with the H9N2 virus at an MOI of 1. The followed steps were the same as described in D to F (*, P < 0.05; **, P < 0.01, two-tailed Student’s t test). (G and H) Effect of GNB1 knockout on IAV primary transcription. WT or GNB1-KO cells were treated with DMSO (G) or with 100 μg/mL CHX (H) and then infected with H9N2 virus (MOI of 1.0). The samples were collected at 4, 6, and 8 hpi. The followed steps were same as described in D to F (*, P < 0.05; **, P < 0.01; ***, P < 0.001; two-tailed Student’s t test).

GNB1 knockdown retards the nuclear import of viral PB2.

In light of the above data, we concluded that GNB1 could interact with polymerase units and affect viral RNA synthesis. Next, we explored whether GNB1 affected the distribution of vRNPs in the cells. We detected the subcellular distribution of vRNPs by NP staining at different time points. The results showed that the vRNPs localized to the nucleus at 4 hpi in WT A549 cells. At 6 hpi, NP was detected in both the nucleus and cytoplasm, indicating that the vRNPs had started to export from the nucleus. At 8 hpi, most of vRNPs were found in the cytoplasm, indicating that vRNPs had completed the nuclear export process (Fig. 7A and B). However, in the GNB1-KO cells, very little fluorescence was observed in the nucleus, suggesting that the nuclear import of vRNPs was delayed. At 6 hpi, vRNPs accumulated gradually in the nucleus. At 8 hpi, the vRNPs were localized in the nucleus and cytoplasm, similar to that in the WT cells at 6 hpi (Fig. 7A and B). These results indicate collectively that GNB1 knockout causes a delay in nuclear import of vRNPs, thus inhibiting the viral life cycle.

FIG 7.

GNB1 affects the nuclear import of PB2. (A) Localization of vRNPs in WT and GNB1-KO A549 cells during infection. WT A549 cells and GNB1-KO A549 cells were infected with the H9N2 virus at an MOI of 5. Cells were fixed and stained with rabbit anti-GNB1(green), mouse anti-NP MAb (red), and DAPI (blue). (B) Quantitative analysis results of NP localization in infected cells. At least 200 cells in each group were scored. (C) GNB1 affects the nuclear import of PB2. GNB1-KO and WT A549 cells were transfected with HA-PB2 for 24 h. The cells were fixed and stained with mouse anti-HA MAb (red), rabbit anti-GNB pAb (green), and DAPI (blue). (D) Percentage of stained cells with mainly nuclear localized PB2 among total cells was calculated in randomly selected five fields of view (**, P < 0.01; two-tailed Student’s t test). (E) Western blot analysis of the nucleocytoplasmic distribution of PB2 in WT or KO-GNB1 cells after 24 h of transfecting PB2. (F) GNB1 does not affect the nuclear import of the PB1-PA heteromer. GNB1-KO and WT A549 cells were transfected with HA-PA and PB1 for 24 h. The cells were fixed and stained with mouse anti-HA MAb (red), rabbit anti-GNB pAb (green), and DAPI (blue). (G) Percentage of stained cells with mainly nuclear localized PA-PB1 heterodimer among total cells was calculated in randomly selected five fields of view. (H) Western blot analysis of the nucleocytoplasmic distribution of PA and PB1 in WT or KO-GNB1 cells after 24 h of cotransfecting PA and PB1. (I) The nuclear import of the PA-PB1 heterodimer was unchanged in GNB1-KO A549 cells during infection. WT A549 cells and GNB1-KO A549 cells were infected with the H9N2 virus at an MOI of 5. Cells were fixed and stained with mouse anti-GNB1mAb (green), rabbit anti-PA pAb (red), and DAPI (blue). (J) Quantitative analysis results of PA localization in infected cells. At least 200 cells in each group were scored. (K) The nuclear import of PB2 was impeded in GNB1-KO A549 cells during infection. WT A549 cells and GNB1-KO A549 cells were infected with the H9N2 virus at an MOI of 5. Cells were fixed and stained with mouse anti-GNB1mAb (green), rabbit anti-PB2 pAb (red), and DAPI (blue). (L) Quantitative analysis of PB2 localization in infected cells. At least 200 cells in each group were scored. C, predominantly cytoplasmic; C+N, nuclear and cytoplasmic; N, predominantly nuclear.

When the newly translated proteins (PB2, PB1, PA, and NP) were transported to the nucleus, vRNPs were assembled and exported to the cytoplasm. The nuclear import model of IAV polymerase subunits is as follows: PB2 enters as a monomer, PA and PB1 enter as heterodimers, and the polymerase subunits form the functional trimeric polymerase in the nucleus (13–15). Our data showed that GNB1 could bind to the domain containing the NLS of PB2 and PB1 and affect the nucleocytoplasmic distribution of vRNP. To explore whether GNB1 influences the nuclear import of polymerase subunits, we investigated the effect of GNB1 on the nuclear import of PB2 and PB1-PA heterodimers. GNB1-KO and WT A549 cells were transfected with HA-PB2 or PB1 and HA-PA plasmids, and immunofluorescence and confocal microscopy were performed after 24 h. PB2 was localized mainly in the nucleus in the control cells, whereas it was distributed in both the nucleus and cytoplasm of GNB1-KO A549 cells (Fig. 7C). Compared with the WT cells, the percentage of nuclear localization of PB2 in GNB1-KO cells decreased approximately 55% (Fig. 7D). The results of nuclear and cytoplasmic fractionation showed that GNB1 knockdown retarded the nuclear import of PB2 (Fig. 7E). The distribution patterns of PB1-PA heterodimers were similar in the GNB1-KO and WT A549 cells (Fig. 7F to H). Meanwhile, we also monitored the localization of PA-PB1 at different time points after infection in GNB1-KO and WT A549 cells. The results suggested that localization patterns of PA-PB1 were similar in GNB1-KO and WT cells (Fig. 7I and J). Taken together, GNB1 cannot influence the distribution of PB1-PA heterodimers, whereas GNB1 can change the distribution of PB2.

To further determine the effect of GNB1 on the nuclear import of PB2, we monitored the localization of PB2 in both GNB1-KO and WT A549 cells at different time points. In H9N2-infected WT cells, PB2 was distributed in the nucleus at 4 hpi, while PB2 was retained in the cytoplasm of GNB1-KO cells. At 6 hpi, compared with the control, some of PB2 in GNB1-KO cells remained in the cytoplasm, indicating the incomplete nuclear import of PB2. At 8 hpi, PB2 in WT cells was detected in both the nucleus and the cytoplasm, indicating that the vRNPs had already started to export from the nucleus, while it was restricted to the nucleus in GNB1-KO cells (Fig. 7K and L). Altogether, these data indicate that GNB1 KO retards the nuclear import of viral PB2.

GNB1 enhances PB2 binding to importin proteins.

Several studies have suggested that importin α family members could mediate the nuclear import of PB2 and that PB2 could bind importin α1, α3, α5, and α7 with similar affinity in vitro (43). Our results were consistent with those of previous studies (Fig. 8A). We performed a coimmunoprecipitation experiment to explore whether GNB1 interacts with importin α isoforms. The results showed that there are robust interactions between GNB1 and importin α1, α5, and α7, whereas GNB1 and importin α3 interacted less strongly (Fig. 8B). Next, we explored whether GNB1 affects the interaction between PB2 and importin α isoforms. We cotransfected HA-PB2; each of Myc-importin α1, α3, α5, and α7; and increasing amounts of Flag-GNB1 into 293T cells, followed by co-IP. The data showed that increased importin α3, α5, and α7 were coimmunoprecipitated with increasing GNB1 (Fig. 8D to F), but importin α1 did not change (Fig. 8C). To determine the effect of GNB1 knockout on the interaction between importin α and PB2 from different subtypes, co-IP was performed in WT, GNB1-KO, and GNB1-KO cells with overexpressed GNB1 infected with H9N2 or H3N2 viruses. The results displayed that levels of importin α3, α5, and α7 immunoprecipitated by PB2 were reduced in GNB1-KO cells (Fig. 8G and H). These results suggest that GNB1 can promote the nuclear import of PB2 by increasing the interaction between PB2 and importin α3, α5, or α7.

FIG 8.

GNB1 facilitates the interaction between PB2 and importin α. (A and B) Interaction between PB2 (A) or GNB1 (B) and importin α1, importin α3, importin α5, and importin α7. 293T cells were transfected with the indicated plasmids for 24 h, and the cells were immunoprecipitated with a mouse anti-HA MAb (A) or a mouse anti-Flag MAb (B). The levels of proteins individually were detected by Western blot. (C to F) GNB1 influences the interaction between PB2 and the importin α family. The 293T cells were transfected with HA-PB2, gradually increasing amounts of Flag-GNB1, and importin α1 (C), importin α3 (D), importin α5 (E), and importin α7 (F). The cells were lysed and immunoprecipitated with a mouse anti-HA MAb (A). The proteins were detected by Western blotting with a rabbit anti-Myc pAb, a rabbit anti-Flag pAb, a rabbit anti-HA pAb to detect importin α family members, GNB1, and PB2. (G and H) Effect of GNB1 on different subtype viruses. The WT and GNB1-KO cells were infected with H9N2 or H3N2 virus. Cell lysates were immunoprecipitated using an anti-PB2 rabbit antibody or rabbit control IgG. The proteins were detected by Western blotting.

GNB1 knockdown reduces polymerase assembly and NP-PA interaction.

Considering that GNB1 facilitates the nuclear import of PB2 and IAV replication, we asked whether GNB1 can affect the formation of the virus polymerase complex (PB2, PB1, and PA). We tested the effect of GNB1 on PB1-PA heterodimer formation. We cotransfected HA-PA, PB1, and increasing amounts of Flag-GNB1 in 293T cells. The results showed that the amount of PB1 coimmunoprecipitated with PA increased with increasing GNB1 (Fig. 9A). Subsequently, we examined the effect of GNB1 on the formation of the polymerase complex. We cotransfected PB1, PB2, and HA-PA into 293T cells and performed a coimmunoprecipitation experiment. The results indicated that the levels of PB2 and PB1, immunoprecipitated by PA, increased in a dose-dependent manner with GNB1 (Fig. 9B). As NP can interact only with PA in the context of a vRNP, we used NP as a bait to immunoprecipitate the PA and investigate whether GNB1 affects NP binding with PA. The data indicated that the level of PA precipitated increased remarkably with an increase in Flag-GNB1 quantity (Fig. 9C). Conversely, we performed the same co-IP in 293T cells after GNB1 knockdown. The results showed that the levels of PB1 (immunoprecipitated by PA), PB2 and PB1 (immunoprecipitated by PA), and PA (immunoprecipitated by NP) decreased compared with those in the control cells (Fig. 9D to F). Taken together, these results provided evidence that silencing GNB1 could reduce the polymerase complex formation and NP-PA interaction.

FIG 9.

Effect of GNB1 on viral polymerase assembly. (A and D) GNB1 promotes the heterodimer formation of PB1 and PA. The 293T cells were transfected with HA-PA, PB1, and gradually increasing Flag-GNB1 (A) or si-GNB1 (D). After 24 h, the cells were lysed and immunoprecipitated with an anti-HA antibody for Western blot assay. (B and E) The effect of GNB1 on a trimeric complex of PB2, PB1, and PA. 293T cells were transfected with PB1, PB2, HA-PA, gradually increasing Flag-GNB1 (B) or si-GNB1 (E). After 24 h, the cells were lysed and immunoprecipitated with an anti-HA antibody for Western blot assay. (C and F) Effect of GNB1 on NP-PA interaction. 293T cells were transfected with Flag-GNB1 (C) or si-GNB1 (F) and then infected with the H9N2 virus (MOI of 10). After 8 h, the cells were lysed and immunoprecipitated with an anti-NP antibody for Western blot assay.

GNB1 knockdown reduces polymerase activity.

Since GNB1 can affect viral RNA synthesis and polymerase complex formation, we further examined whether GNB1 regulates polymerase activity. We cotransfected RNP reconstitution plasmids (pcDNA3.1-PB2, pcDNA3.1-PB1, pcDNA3.1-PA, pcDNA3.1-NP, and pol I-firefly), RL-TK, and Flag-GNB1 (increasing dose) into 293T cells. After 24 h, cell lysates were detected using a luciferase assay. The results indicated that as Flag-GNB1 increased, relative polymerase activity gradually increased (Fig. 10A). In contrast, relative polymerase activity was reduced when 293T cells were transfected with the GNB1 siRNA after 24 h (Fig. 10B). We also examined the levels of PB2, PB1, PA, and NP proteins and the cell viability of GNB1 siRNA 293T cells. The results indicated that protein levels were unchanged (Fig. 10A), and the viability of the cells was not significantly different (Fig. 10C), indicating that the change in polymerase activity was unrelated to protein levels. These data indicated that silencing GNB1 reduces polymerase activity.

FIG 10.

Effect of GNB1 on IAV polymerase activity. (A) Effect of overexpressed GNB1 on IAV polymerase activity. The 293T cells were transfected with plasmids containing NP, PB2, PA, PB1, PoI II, pRL-TK, and Flag-GNB1 (0.5 and 1 μg) for 24 h. Then, the luciferase activity was measured, and Renilla luciferase was used as an internal control (*, P < 0.05; **, P < 0.01; one-way analysis of variance [ANOVA]). The expression level of each protein was also detected by Western blot. (B) Effect of silencing GNB1 on polymerase activity. The 293T cells were transfected siRNA targeting GNB1 (si-1, si-2) or si-NC (negative control) for 36 h, and then cells were transfected with the indicated plasmids as above. Then luciferase activity was measured, and the GNB1 protein expression level was detected using Western blot (***, P < 0.001; one-way ANOVA). (C) The effect of si-GNB1 on 293T cell viability. Cell viability was measured by CCK-8 assay at 24 h and 48 h after being transfected with siRNA targeting GNB1 (si-1 and si-2) or si-NC.

DISCUSSION

When the vRNPs enter the nucleus in infected cells, the primary transcription of IAV begins (15). The nascent viral RNA is encapsulated into the RNPs for genome replication and transcription. Hence, the newly synthesized RNP subunits must be imported into the nucleus (12, 13). The importin α family or importin β3 plays important roles in the nuclear import of NP and polymerase subunits (13, 44, 45). To date, numerous studies have found that several host proteins can regulate the interaction between viral NP protein and importin α to mediate the nuclear import process, such as MOV10 (46), PLSCR1 (47), eEF1D (48), and Hsp40/DnaJB1 (49). However, relatively few proteins related to the nuclear import of polymerase subunits have been explored. In this study, we identified a new protein, GNB1, and found that GNB1 could facilitate interactions of PB2 with importin α3, α5, and α7 to promote the nuclear import of PB2 and dimer formation of PB1-PA. As a result, GNB1 can enhance virus genome transcription and replication and positively regulate the propagation of IAV. The model is illustrated in Fig. 11.

FIG 11.

Model of GNB1 promoting the nuclear import of influenza PB2. In the cytoplasm, GNB1 can promote newly synthesized influenza virus PB2 binding to importin α to facilitate the nuclear import of PB2.

Liu et al. demonstrated that the H9N2 virus-derived (but not H5N1 virus-derived) M1 protein can interact with the host GNB1 protein to promote avian H5N6 virus (a reassortment virus with H9N2 virus-derived internal genes) release from mammalian cell (32). Our data displayed that GNB1 can interact with M1 from H3N2 but not with M1 from an H9N2 or an H1N1 (PR8). We have compared amino acid sequence identity of the M1 in this study and M1 that Li et al. used. The amino acid homology between them is 100%. This inconsistency between our result and that of Li et al. might be caused by a variety of reasons, such as cellular factors, virus strain, and experimental conditions.

Two regions of PB2 have been shown to be necessary for the nuclear import of PB2 (23). One is located at amino acids 449 to 495. When these residues were deleted, the nuclear import of PB2 was prevented. The other region consists of residues 736 to 739 and 752 to 755, located in the C-terminal region of PB2 (23), which are supposed to interact with importin α to help the nuclear import of PB2 through the classical nuclear import pathway (25, 43, 50, 51). Our results suggested that GNB1 could bind to the C-terminal region of PB2 containing residues 736 to 739 and 752 to 755 but not 449 to 495. The domains of residues 736 to 739 and 752 to 755 are where both importin α and GNB1 bind, which can explain the finding that GNB1 promotes PB2 binding to importin α. Although previous studies and our results showed that PB2 could interact with importin α1, α3, α5, and α7 with comparable affinity in vitro (52), whether PB2 binding with the isoform of importin α is related to species is controversial. Some studies have shown that the nuclear import of PB2 from avian influenza virus and human influenza virus is dependent on importin α3 and importin α7, respectively (43, 52, 53). However, several studies have shown that the specific importin α isoform of PB2 binding does not correlate with the origin of PB2 (43, 51). In this study, our results showed that GNB1 could interact with importin α1, α3, α5, and α7, and facilitate PB2 from H9N2 and H3N2 virus binding to human importin α3, α5, or α7 to promote the nuclear import of PB2.

Among the polymerase complex subunits, PB1 contains the polymerase active site, and is necessary for synthesized RNA elongation (54, 55); while PB2 and PA employ cap snatching to initiate the transcription of viral RNA (56, 57). PB1, as the core protein of the polymerase complex, can interact with the N-terminal region of PB2 through its C-terminal region, while the N-terminal region of PB1 can bind to the C-terminal region of PA, thus forming the polymerase complex (58). Our studies showed that GNB1 could bind to the N-terminal and C-terminal domains of PB2, the N-terminal and polymerase domains of PB1, and the C-terminal domain of PA. As a result, GNB1 can promote the formation of the PB1-PA dimer and nuclear import of PB2, thereby enhancing the formation of the polymerase complex.

In summary, we demonstrated that GNB1 is a novel interacting partner of the polymerase subunit of IAV, which is essential for viral replication. Knockdown of GNB1 can impair the nuclear import of viral PB2 and formation of the PB1-PA dimer, thus retarding the life cycle of IAV. Importantly, GNB1 can facilitate PB2 binding to importin α3, α5, and α7 to promote the formation of the polymerase complex. Together, these results show that GNB1 is a novel positive host factor for influenza virus replication.

MATERIALS AND METHODS

Cells and viruses.

Human embryonic kidney 293T cells (HEK293T) and Madin-Darby canine kidney (MDCK) were cultured in Dulbecco’s modified Eagle medium (Gibco, USA) with 10% fetal bovine serum (FBS) at 37°C in a 5% CO₂ humidified incubator. Human lung epithelial cells (A549) were maintained in Ham’s F-12 medium. The IAVs used in the experiments were A/Puerto Rico/8/1934 (PR8/H1N1), mouse-adapted H9N2 avian influenza virus (A/Chicken/Hunan/8.27 YYGK3W3/2018), A/Beijing-Miyun/51/2020 (H3N2), luciferase-expressing reporter influenza virus, PR8/H1N1-Rluc, and H9N2-Nluc. All viruses were amplified in 10-day-old specific-pathogen-free (SPF) chicken eggs or MDCK cells, and virus titers were determined using plaque assays in MDCK cells.

Plasmids, small interfering RNAs, and transfection.

The human GNB1, importin α1, importin α3, importin α5, and importin α7 genes were amplified by PCR from total mRNA from A549 or 293T cells and cloned into the mammalian expression vectors pCAGGS-Flag and pCAGGS-Myc using the ClonExpress II one-step cloning kit (Vazyme, Nanjing, China). The cDNAs encoding full-length viral proteins from H9N2 were subcloned into pcDNA3.1, pCAGGS-HA, or pCAGGS-GFP. All the plasmids were confirmed by sequencing. Small interfering RNA and a nontarget siRNA (si-NC) were synthesized by RiboBio (Guangzhou, China). Transfection of plasmids and siRNAs was performed using Lipo8000 (Beyotime; China) or Lipofectamine 2000 (Invitrogen) in all cells maintained in Opti-MEM.

Antibodies.

The antibodies used in this study were as follows: anti-GST, anti-GAPDH, anti-Flag, and anti-HA mouse monoclonal antibodies (66001-2-Ig, 60004-1-Ig, 66008-3-Ig, and 66006-2-Ig, respectively; Proteintech); anti-IAV PB2, PB1, PA, NP, and M1 rabbit polyclonal antibodies (GTX125926, GTX125923, GTX125932, GTX125989, and GTX125928, respectively; GeneTex, USA); anti-NP mouse monoclonal antibody (sc-101352; Santa Cruz); anti-FLAG M2 mouse monoclonal antibody (F1804; Sigma-Aldrich, USA); anti-GNB1 rabbit polyclonal antibodies (A1867; GeneTex); GNB1 mouse monoclonal antibody (ml-K7075, Mlbio, China); antibody GNB1-FITC conjugated (CSB-PA009602LC01HU; Huamei, China); the secondary antibodies used for immunofluorescence and Western blotting (Proteintech); anti-KPNA2, -KPNA4, -KPNA1, and KPNA6 (A5012, A2026, A1742, and A7363, respectively; ABclone); and anti-PARP1 rabbit polyclonal antibody (D161071; Sangon Biotech, China).

Generation of GNB1-KO A549 cell lines.

GNB1-KO A549 cells were generated using the CRISPR-Cas9 system. The sgRNA sequence (5′-CACCGTGCACAGGTCATGACCCAGG-3′) targeting the GNB1 gene was inserted into LentiCRISPRv2. GNB1/empty LentiCRISPRv2 (8 μg), psPAX2 (4 μg), and pMD2G (2 μg) were transfected into HEK293T cells, and lentiviruses were collected 60 h posttransfection. A549 cells were seeded in 6-well plates, and lentiviruses were added. To select the positive clones, puromycin (3 μg/mL) was added at 24 hpi. The monoclonal A549 cell line was selected using the serial dilution method in 96-well plates. Finally, the knockout of GNB1 was confirmed by Western blotting at the protein level.

RNA isolation and qRT-PCR.

For most experiments, total RNA content was extracted using TRIzol reagent (Invitrogen, USA), and cDNA was reverse transcribed using the HiScript II Q RT SuperMix for qPCR (+genomic DNA [gDNA] wiper) (Vazyme, Nanjing, China). Quantitative reverse transcription-PCR (qRT-PCR) reactions were performed using the ChamQ universal SYBR qPCR master mix (Vazyme). The levels of NP genomic RNA (vRNA), cRNA, and mRNA were determined using strand-specific real-time RT-PCR as described previously (59). The strand-specific reverse primers and their sequences are as follows: vRNAtag, GGCCGTCATGGCGAATGAAAATGGAAGAAGAACAAGGGTTGC; cRNAtag, GCTAGCTTCAGCTAGGCATCAGTAGAAACAAGGGTATTTTTCTTC; and mRNAtag, CCAGATCGTTCGAGTCGTTTTTTTTTTTTTTTTTCTTCAATTGTC. qPCR primers are as follows: vRNA-F, GGCCGTCATGGTGGCGAAT; vRNA-R, CTCAGGATGAGTGCAGACCTTGCC; cRNA-F, CGATCGTGCCTTCCTTTG; cRNA-R, GCTAGCTTCAGCTAGGCATC; mRNA-F, CGATCGTGCCTTCCTTTG; and mRNA-R, CCAGATCGTTCGAGTCGT. Relative mRNA quantities were normalized to GAPDH mRNA.

Coimmunoprecipitation assay.

HEK293T cells were transfected with plasmids using Lipofectamine Lipo8000. After 24 h of transfection, the cells were washed with cold phosphate buffer saline (PBS) and lysed with IP lysis buffer containing a protease inhibitor cocktail on ice for 10 min. Subsequently, the supernatants were immunoprecipitated with the primary antibodies, and the mixture was incubated overnight at 4°C. The next day, the pretreated protein A/G agarose (Santa Cruz, USA) was added to the mixture and incubated for 1 h at room temperature. Finally, the beads were collected after washing five times with ice-cold PBS and boiled with 1× SDS loading buffer to detect protein levels by Western blotting. A549 cells were infected with the H9N2 virus at an MOI of 5 or PBS as a negative control to study the endogenous protein interactions. After 12 h of infection, the cells were lysed, and an immunoprecipitation was carried out.

Confocal microscopy.

A549 or GNB1-KO A549 cells were seeded in glass bottom plates and infected with the H9N2 virus at an MOI of 5 the following day. Indirect immunofluorescence (IFA) was performed. The process was as follows: at different times points, cells were fixed with 4% paraformaldehyde (PFA) for 20 min and permeabilized with 0.5% Triton X-100 in PBS for 15 min. Subsequently, the permeabilized cells were blocked with 5% bovine serum albumin (BSA) for 1 h and then incubated with the primary antibody at room temperature for 2 h. After being washed thrice with PBST, the cells were incubated with the appropriate Alexa Fluor-conjugated secondary antibody for 1 h. The cells were washed thrice and stained with 4′,6-diamidino-2-phenylindole (DAPI) for 15 min. The samples were examined using a confocal microscope. HeLa cells were grown on glass bottom plates and transfected with plasmids or siRNA. Subsequently, IFA and confocal microscopy were performed.

Dual-luciferase reporter assay.

The polymerase activity was determined using a luciferase reporter assay. Briefly, 293T cells were transfected with GNB1 siRNA or scrambled siRNA 24 h posttransfection and then cells were cotransfected with the four RNP plasmids (pCDNA3.1-PB2, pCDNA3.1-PB1, pCDNA3.1-PA, and pCDNA3.1-NP;0.3 μg of each, and the four genes were from an H9N2 mouse-adapted strain), pPol I-Luc (0.3 μg), and an internal control, pRL-TK (0.05 μg). At 24 h after transfection, the relative polymerase activity was measured using a dual-luciferase assay kit (Promega, USA).

GST pulldown.

Glutathione S-transferase (GST) protein and GST-fused GNB1 were purified by Mag-Beads GST fusion protein purification (Sangon Biotech, Shanghai, China) and detected by Coomassie blue (CB) staining. HEK293T cells grown in 10-cm dishes were transfected individually with 12 μg of each plasmid (HA-PB2, HA-PB1, and HA-PA) using Lipofectamine 8000. At 48 h posttransfection, the cells were solubilized with 500 μL of IP lysis buffer. The purified GST-GNB1 protein samples were added to the treated magnetic beads, and the mixture was shaken for 1 h at 4°C. After five washes with washing buffer, the lysates were added and incubated for 2 h at 4°C. After five washes, the bound proteins were analyzed using SDS-PAGE.

Nuclear and cytoplasmic fractionation.

The separation of the nucleus-cytoplasm in A549 was carried out according to the manufacturer’s instructions (P0027; Beyotime Biotechnology, China).

Cell viability assay.

Cell viability was determined using the Cell Counting kit-8 (CCK-8) (Beyotime, Shanghai, China). In brief, 293T cells or A549 cells were seeded in a 96-well plate and transfected with GNB1 siRNA or control siRNA the next day. After 24 h and 48 h posttransfection, cell viability was measured.

Statistical analysis.

All experiments were performed independently at least three times. Statistical analyses were performed using the Student’s t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Ethics statement.

All animal experimental procedures were performed in accordance with the guidance of the Institutional Animal Care and Use Committee of Wuhan Institute of Virology, Chinese Academy of Sciences (CAS). All experiments were performed under general anesthesia, and all efforts were made to minimize the number of animals used and animal suffering.

Data availability.

The GNB1-KO A549 cell line and the recombinant plasmids in this study are available on reasonable request from the corresponding author, and the material transfer agreement (MTA) was signed. The nucleotide sequences of the eight gene segments of the H9N2 virus are available from GenBank under accession numbers MW105213 to MW105220.