Influenza A viral nucleoprotein interacts with cytoskeleton scaffolding protein α‐actinin‐4 for viral replication

Identification of actinin‐4 as an interacting partner of IAV NP, conserved across IAV subtypes

To identify potential interacting partners of NP of IAV (A/chicken/Hatay/2004; H5N1), we screened a human lung cDNA library using NP as the bait. A clone encoding amino acids 770–911 of actinin‐4 was obtained as a positive interactor in the screen, which was confirmed by blast analysis (Fig. 1A,B). The full‐length human cDNA of actinin‐4 was cloned in‐frame with the activation domain vector pYesTrp2 and co‐transformed with the bait vector pHLZ‐NP in the L40 strain of yeast. Furthermore, it tested positive for histidine prototrophy and β‐galactosidase activity (Fig. 1C), confirming that influenza nucleoprotein interacted with full‐length actinin‐4 protein.

The NP–actinin‐4 interaction was validated by performing a co‐immunoprecipitation assay. His‐tagged NP and HA‐tagged actinin‐4 proteins were expressed using in vitro coupled transcription–translation rabbit reticulocyte lysate system in the presence of [³⁵S]‐methionine. Co‐immunoprecipitation of these reactions, as shown in Fig. 1D, validates the interaction between NP and actinin‐4.

To confirm NP–actinin‐4 interaction in mammalian cells, HEK293 cells were transfected with His‐tagged NP (pCDNA‐NP‐His) and HA‐tagged actinin‐4 (pCMX‐actinin‐4‐HA) expressing constructs, either alone or together. His‐tagged NP and HA‐tagged actinin‐4 expression was confirmed as shown in Fig. 1E (top two panels). Immunoprecipitation analysis using anti‐His serum followed by immunoblotting with anti‐HA serm confirmed the interaction of actinin‐4 with NP (Fig. 1E) (bottom panel). Collectively, these results indicate that actinin‐4 indeed interacts with NP in yeast, cell‐free systems and mammalian cells.

To determine whether NP and actinin‐4 interact during IAV infection, lung epithelial A549 cells were infected with A/Puerto Rico/8/34 virus (H1N1; PR8) at a multiplicity of infection (MOI) of 1. Twenty‐four hours post‐infection (p.i.), cells were harvested and lysates were subjected to immunoprecipitation using antibodies specific for NP and actinin‐4. NP of PR8 virus co‐precipitated with actinin‐4 and, reciprocally, actinin‐4 was co‐precipitated with NP (Fig. 2A, panels 1 and 2). Figure 2A (panels 3 and 4) also shows immunoprecipitation of NP and western blotting with anti‐NP serum and immunoprecipitation of actinin‐4 and western blotting with anti‐actinin‐4 serum, respectively. As shown in Fig. 2A (panels 7), actinin‐4 does not interact with M2, another influenza A viral protein, in PR8 infected cells, thus authenticating the specificity of interaction between NP and actinin‐4. Significantly, the NP–actinin‐4 interaction was also confirmed in A549 cells infected with other IAV subtypes of avian and human origin, including the highly pathogenic avian influenza virus A/Vietnam/1203/04 (H5N1) and the 2009 pandemic virus A/California/08/2009 (H1N1) (Fig. 2B), suggesting that the actinin‐4‐NP interaction is conserved across multiple subtypes of influenza viruses.

Role of actinin‐4 in subcellular localization of NP during infection

To determine the site and kinetics of co‐localization of NP and actinin‐4 during infection, we performed immunofluorescence studies. Accordingly, A549 cells infected with PR8 at a MOI of 5, were harvested 5 h p.i, and subjected to immunofluorescence analysis. As shown in Fig. 3A, NP and actinin‐4 co‐localized majorly around the nucleus at 5 h p.i. and at 24 h p.i., in the cytosol (Fig. 3B). Mock‐infected cells are shown in Fig. 6E. The data collected from five different independent cells, as shown in Fig. 3C, are represented graphically, where Pearson's coefficient has been used as a quantitative measure of degree of co‐localization between NP and actinin‐4.

To explore further the significance of NP–actinin‐4 co‐localization, we examined the effect of actinin‐4 depletion on NP localization. A549 cells transfected with either nontargeting (NT) short interfering RNA (siRNA) or actinin‐4‐specific siRNA for 24 h, were infected with PR8 at a MOI of 1 for 24 h. Subsequent to subcellular fractionation, nuclear and cytoplasmic fractions were checked for the presence of actinin‐4 and NP by western blot analysis. Intriguingly, siRNA‐mediated silencing of actinin‐4 expression led to a significant reduction in the amount of NP protein in the nuclear fraction compared to control (Fig. 4A). To analyze this further, we depleted the A549 cells of actinin‐4 by treating them with actinin‐4 siRNA, followed by infection with PR8 at a MOI of 1 and observed the significantly reduced localization of NP in nucleus compared to control siRNA using an immunofluorescence assay (Fig. 4B). Viability of the actinin‐4 siRNA‐treated A549 cells was determined 24 h p.i. by measuring the cytotoxicity in the gene‐specific siRNA versus the NT siRNA. The results illustrated in Fig. 4C show that actinin‐4 specific siRNA‐treated cells were ~ 84% viable compared to transfection reagent control.

Wortmannin, an inhibitor of actinin‐4, is known to cause dissociation of actinin‐4 from the cytoskeleton and its steady translocation into the nucleus 12. To analyze the influence of actinin‐4 subcellular localization on NP localization, we treated A549 cells with either dimethylsulfoxide (DMSO) or 10 μm wortmannin for 16 h followed by infection with PR8 at a MOI of 5 and assessed NP localization using an immunofluorescence assay. Figure 5A,B shows that there was increased retention of NP in the nucleus in wortmannin‐treated cells even after 8 and 24 h p.i. compared to the dimethylsulfoxide‐treated cells, where NP showed localization in both nuclear and cytoplasmic compartments. The effect of wortmannin on the dissociation of actinin‐4 from the cytoskeleton and translocation into the nucleus is shown in Fig. 5C.

IAV enhances actinin‐4 mRNA and protein expression levels in infected cells

Because our results indicated that NP of IAV and actinin‐4 were interacting as well as co‐localizing in the cellular milieu, we next aimed to investigate the effect of IAV infection on actinin‐4 expression. Accordingly, we examined the effects of the virus dose and the duration of infection on actinin‐4 mRNA and protein expression p.i. Significant and progressive increases in the levels of actinin‐4 mRNA and protein were observed upon infecting A549 cells with PR8 virus at increasing MOIs (Fig. 6A,B). In addition, a time‐dependent increase up to 24 h was observed in the relative mRNA levels of actinin‐4 when cells were infected at a given MOI (Fig. 6C). The increase in actinin‐4 mRNA levels also corresponded with elevated levels of actinin‐4 protein expression (Fig. 6D). Collectively, these results suggest that virus dose and the duration of infection up‐regulate actinin‐4 expression, at both the protein and mRNA level.

Actinin‐4 is required for IAV replication

To understand the physiological relevance of the NP–actinin‐4 interaction, we examined the consequences of inhibiting actinin‐4 expression on viral replication. A549 cells were treated with either NT or actinin‐4 siRNA for 24 h and then infected with PR8 virus. Silencing of actinin‐4 expression was confirmed by quantitative PCR and western blot analyses (Figs 7A and 8A).

Because mRNA and vRNA levels are the hallmarks of viral gene transcription and replication, we measured these parameters by quantitative PCR. Silencing of actinin‐4 led to a significant reduction in the levels of NP mRNA and vRNA in virus‐infected A549 cells (Fig. 7B,C). Interestingly, a maximum reduction of approximately three‐ and five‐fold was observed in both mRNA and vRNA levels, respectively, at 6 h p.i., consistent with the maximum silencing of actinin‐4 mRNA levels observed at the same time point p.i. However, an absence of actinin‐4 in IAV infected cells did not have significant effect on the NS1 mRNA levels at up to 6 h p.i. (Fig. 7D). At 8 h p.i., relative NS1 mRNA levels decreased by almost two‐fold in actinin‐4 depleted A59 cells, which may be attributed to a reduction in viral titers (Fig. 8B). Thus, it is suggested that the depletion of actinin‐4 has a profound effect on NP mRNA and vRNA specifically at early time points.

In view of the significant effect of wortmannin on actinin‐4, we challenged A549 cells with 10 μm wortmannin for 16 h followed by infection with PR8 virus at a MOI of 1 for 24 h. In agreement with our previous observations, a significant five‐fold decrease was observed in NP mRNA levels in 10 μm wortmannin treated‐infected cells compared to dimethylsulfoxide treated‐infected cells (Fig. 7E).

To assess the role of actinin‐4 in virus replication in more detail, we investigated its effect on IAV replicase activity. A549 cells were mock‐treated or treated with either NT siRNA or actinin‐4 siRNA followed by co‐transfection with plasmids encoding PR8 polymerase complex genes PB2, PB1, PA and NP in conjunction with a reporter plasmid containing the UTR of the NS1 segment upstream of the luciferase gene driven by the human RNA pol I promoter. Figure 7F shows that there was a five‐fold reduction in replicase activity upon silencing actinin‐4 compared to NT siRNA treatment. Additionally, as shown in Fig. 7G, NP levels were reduced in actinin‐4 depleted cells. These results strongly suggest that actinin‐4 is essential for IAV replication.

Because actinin‐4 expression appeared to be crucial for both viral transcription and replication, we next studied the effect of actinin‐4 on the expression of viral protein levels. Western blot analysis of whole cell extracts of infected cells demonstrated a substantially lower expression of NP, NS1and M2 when actinin‐4 expression was knocked down (Fig. 8A).

To assess the effect of actinin‐4 on viral replication in greater detail, we performed a plaque assay. Accordingly, A549 cells were transfected with either NT siRNA or actinin‐4 siRNA followed by infection with influenza A PR8 virus at a MOI of 1. In agreement with our observations for the viral replicase assay, there was a significant decrease in the viral titers especially at 16 and 24 h p.i. in supernatants from cells treated with actinin‐4 siRNA compared to those treated with NT siRNA. Furthermore, the silencing of actinin‐4 expression was confirmed by western blot analysis (Fig. 8B). Complementing these observations, the titers of IAV were remarkably decreased on treating the cells with 10 μm wortmannin followed by infection with PR8 virus at a MOI of 1 (Fig. 8C).

Next, we aimed to determine the effect of actinin‐4 depletion on the infectivity of cells during virus infection at 8 and 24 h p.i. Compared to 18% and 29% of cells infected in the control at 8 and 24 h p.i., respectively, only 5.5% and 11.5% of cells were infected in the actinin‐4 siRNA treatment (Fig. 9A). Silencing of actinin‐4 at the two time points is shown in Fig. 9B,C. These findings suggest that actinin‐4 is required for viral protein synthesis and replication.

Taking all of the above findings together, the present study unravels a novel and crucial role of actinin‐4 in various stages of the IAV life cycle, including viral transcription, translation of viral proteins and viral replication.

Cells and cell lines and viruses

HEK293 cells and A549 cells were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA) and maintained in acordance with the manufacturer's instructions. A/Puerto Rico/8/34 (PR8) influenza virus strain was used at a MOI of 1 unless specified otherwise and the influenza virus strains used in the co‐immunoprecipitation experiment are listed in Table 1.

Plasmid constructs, antibodies and siRNA

The NP gene of H5N1 A/Hatay/2004 isolate was cloned into pCDNA3.1‐His plasmid to be used as bait vector for co‐immunoprecipitation studies. Full‐length human actinin‐4 gene cloned into the pCMX plasmid, which was provided by Hung‐Ying Kao, Department of Biochemistry, School of Medicine, Case Western Reserve University, Cleveland, OH, USA 41. Anti‐NP serum was obtained from the Immunology and Pathogenesis Branch, Influenza Division, Centres for Disease Control and Prevention (Atlanta, GA, USA). Anti‐actinin‐4 serum was purchased from Enzo Life Sciences Inc (Farmingdale, NY, USA). Anti‐HA and Anti‐His sera were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Anti‐β‐actin serum was purchased from Sigma‐Aldrich (St Louis, MO, USA). Pool of gene‐specific siRNAs against actinin‐4 was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The Renilla luciferase assay system kit was obtained from Promega Corp. (Madison, WI, USA).

Y2H screening

The Lex A based screening system (Hybrid hunter, version F), comprising Saccharomyces cerevisiae strain L40 [MATa his3Δ200 trp1‐901 leu2‐3112 ade2 LYS2::(4lexAop‐HIS3)URA3::(8lexAop‐lacZ) GAL4], and pHybLexA/Zeo and pYesTrp2 as binding domain and activation domain vectors, respectively, and human lung cDNA library cloned in pYesTrp2, was purchased from Invitrogen (Carlsbad, CA, USA). Screening was performed in accordance with the manufacturer's instructions. The bait plasmid was constructed by cloning IAV NP coding sequence in‐frame with the LexA DNA binding domain in pHybLexA/Zeo. PHybLexA/Zeo‐NP was co‐transformed with cDNA library in L40 and co‐transformants were selected for the activation of two reporter genes, HIS3 and LacZ. The strength of the interaction in selected co‐transformants was assessed by their ability to grow on His⁻Trp⁻ and Zeo⁺ YC media supplemented with 5 mm AT (3‐amino‐1,2,3‐trizole, competitive inhibitor of HIS3) and for positivity of a filter β‐galactosidase activity assay. Plasmids were isolated from positive co‐transformants and shuttled into Escherichia coli DH5α and sequenced. The sequence obtained was analyzed by blast to identify the insert. L40 co‐transformed with pHybLexA/Zeo‐Fos and pYesTrp2‐Jun was used as a positive control and L40 co‐transformed with pHybLexA/Zeo and pYesTrp2 was used as a negative control for the library screening 42.

Western blot analysis

Cells were treated with lysis buffer (50 mm Tris, pH 7.5, 150 mm NaCl, 1 mm EDTA, 0.1% Triton X‐100) supplemented with complete protease‐inhibitor mixture purchased from Santa Cruz Biotechnology and the lysates thus obtained were subject to SDS/PAGE.

Co‐immunoprecipitation

Cells were harvested in the above mentioned lysis buffer, and cell lysates were incubated with the primary antibody overnight followed by a 90‐min incubation with protein A Dynabeads purchased from Invitrogen. The beads were washed three times with chilled NaCl/Pi, resuspended in laemmli buffer, boiled for 10 min and the beads were spin down. Supernatants were subjected to western blotting 43.

Cell viability assay

A549 cells were seeded at 10 000 cells·well⁻¹ in a 96‐well dish. After adherence, they were treated with either NT or actinin‐4 specific siRNA for 24 h. Subsequently, 200 μg·μL⁻¹ of 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyl‐tetrazolium bromide solution/well was added and incubated at 37 °C for 30 min to allow for the formation of formazan. The medium was removed, and 200 μL of dimethylsulfoxide was added to each well to dissolve the formazan. Absorbance was measured on an ELISA plate reader with a test wavelength of 570 nm and a reference wavelength of 630 nm to obtain sample signal (A ₅₇₀–A ₆₃₀). Dimethylsulfoxide was used as a reference.

Immunofluorescence microscopy

A549 cells infected with PR8 influenza virus at a MOI of 5 were fixed at different time points in NaCl/P_i with 2% paraformaldehyde for 20 min at room temperature, permeabilized with 0.4% Triton X‐100 in NaCl/P_i for 15 min at room temperature, and blocked with NaCl/P_i containing 5% bovine serum albumin. Immunostaining was performed using mouse anti‐NP and rabbit anti‐actinin‐4 sera. Unbound‐antibodies were washed away with NaCl/P_i containing 0.5% tween‐20 and incubated with goat anti‐rabbit Alexa 488 and Goat anti‐mouse Alexa 594 conjugated sera purchased from Invitrogen. The nucleus was stained with 4′,6‐diamidino‐2‐phenylindole (DAPI). Slides were observed at × 60 magnification using a confocal microscope (A1R; Nikon, Tokyo, Japan).

Quantification of NP mRNA and vRNA by qunttaitative PCR

Total RNA from cells was extracted using the RNeasy Mini Kit from Qiagen (Hilden, Germany), and 2 μg of RNA was reverse‐transcribed using the ThermoScript RT‐PCR System (Invitrogen) in a volume of 20 μL after Dnase I treatment. The resulting cDNA was diluted 1 : 10 and 5 μL was used in a SYBR Green (SA Biosciences, Valencia, CA, USA) based real‐time PCR reaction in a volume of 25 μL using a Mx3000 real‐time PCR instrument (Stratagene, Santa Clara, CA, USA). Primers used for real‐time RT‐PCR were actinin‐4 primers, forward GTT CTC GAT CTG TGT GCC TG, actinin‐4 reverse GAC CTG CTG CTG GAC CC; the housekeeping gene was β‐actin with primers, forward ACC AAC TGG GAC GAC ATG GAG AAA, reverse TAG CAC AGC CTG GAT AGC AAC GTA; GAPDH primers, forward TCA CTG CCA CCC AGA AGA CTG, reverse GGA TGA CCT TGC CCA CAG C, and NP vRNA primer, CTC GTC GCT TAT GAC AAA GAA G, NP gene primers (for mRNA), forward CTC GTC GCT TAT GAC AAA GAA G, reverse AGA TCA TCA TGT GAG TCA GAC, and NS1 gene primers (for mRNA) forward CAG GAC ATA CTG ATG AGG ATG, reverse GTT TCA GAG ACT CGA ACT GTG were used to normalize the Ct values obtained in the real‐time PCR reactions, which were then used to calculate fold changes compared to the uninfected sample using the ΔΔCt method 44.

Silencing of actinin‐4

A549 cells were plated at a density of 10⁶ per well in a six‐well plate were transfected with 120 nm NT siRNA or gene‐specific siRNA targeted against actinin‐4 24 h before infection with A/PR/8/34 at a MOI of 1. The actinin‐4 siRNA concentration used was optimized to be 120 nm in A549 cells. Cells were harvested 24 h p.i. and analyzed by western blotting with the respective antibodies.

Plaque assay

A549 cells were treated with NT or actinin‐4 siRNA and were then infected with PR8 at a MOI of 1. Supernatants collected after 24 h were analyzed for virus growth by plaque assay using Madin–Darby canine kidney (MDCK) cells.

Luciferase reporter assay

Full‐length genomic segments of PB2, PB1, PA and NP derived from PR8 were cloned in pCDNA3.1. A549 lung epithelial cells were pre‐treated with NT siRNA and actinin‐4 siRNA. Components of influenza polymerase PB2, PB1, PA and NP were transfected along with luciferase reporter plasmid, which contains noncoding sequences from the NS1 segment of IAV and the luciferase gene driven by Pol I, 24 h post‐transfection of siRNA 45.

Protein subcellular fractionation analysis

1 × 10⁶ A549 cells were infected with IAV and then washed with NaCl/P_i, and cells were fractionated into cytosol, membranes, nuclei and cytoskeleton using the Q proteome cell compartment kit (Qiagen) in accordance with the manufacturer's instructions, and then one‐fifth of each fraction was subjected to western blotting. The fractions were analyzed to detect GAPDH and HDAC1 proteins as the controls for cross‐contamination during the cell fractionation procedure, followed by re‐probing with the respective antibodies on the same membrane.

Statistical analysis

Data are expressed as the mean ± SE. Means were compared by one‐way analysis of variance followed by Fisher's protected least significant difference test to assess specific group differences. P < 0.05 was considered statistically significant.