The Viral Nucleoprotein Determines Mx Sensitivity of Influenza A Viruses

Mouse 3T3 cell lines expressing murine Mx1 (3T3-Mx1), human MxA (3T3-MxA), or Mx-negative control cells (3T3) have been described previously (22). 3T3 cells, MDCK cells, and HEK 293T cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum, 2 mM l-glutamine, and penicillin-streptomycin.

Recombinant viruses, including pH1N1 (A/Hamburg/4/09), H5N1 [A/Thailand/1(KAN-1)/04], and the NP-reassortant viruses H5N1-NP(pH1N1) and pH1N1-NP(H5N1), were generated under biosafety level (BSL) 3 conditions by the eight-plasmid reverse-genetics system as described previously (12, 20, 26). For generation of the pH1N1 viruses, the HA rescue plasmid was mutated by quick-change site-directed mutagenesis to encode the mouse-adapted mutations D131E and S186P (34). Successful introduction of these mutations was confirmed by sequencing. All recombinant viruses were plaque purified on MDCK cells in the presence of 1 μg/ml of tosylsulfonyl-phenylalanyl-chloromethyl keton (TPCK)-treated trypsin. Virus stocks were prepared on MDCK cells, and titers were determined by plaque assay.

BALB/c mice with defective Mx1 alleles and congenic BALB.A2G-Mx1 mice (designated BALB-Mx1) carrying the functional Mx1 allele (28) were bred locally. Six- to eight-week-old mice were anesthetized with a mixture of ketamine (100 μg per gram body weight) and xylazine (5 μg per gram) administered intraperitoneally (i.p.) and inoculated intranasally (i.n.) with the indicated doses of viruses in 50 μl phosphate-buffered saline (PBS) containing 0.3% bovine serum albumin (BSA). Mice were monitored daily for weight loss until 12 days postinfection (p.i.). Animals with severe symptoms or more than 25% weight loss were euthanized. For determination of lung titers and histopathological analysis, mice were euthanized at the indicated time points. All animal work was conducted under BSL 3 conditions in accordance with the guidelines of the local animal care committee.

Mice were infected i.n. with 1,000 PFU of the indicated viruses. Lungs from infected mice were collected at 48 h p.i. and homogenized using the FastPrep24 system (MP Biomedicals). Tissue debris was removed by low-speed centrifugation, and virus titers in supernatants were determined by performing 10-fold serial dilutions in PBS with 0.3% BSA followed by plaque assay on MDCK cells.

MDCK cells seeded in 6-well plates were incubated with virus at a multiplicity of infection (MOI) of 0.001 in PBS containing 0.3% BSA for 1 h at 37°C. The inoculum was removed, and 2 ml infection medium (DMEM supplemented with 0.3% BSA), additionally containing 1 μg/ml TPCK-treated trypsin for pH1N1 viruses, was added. Virus titers in cell culture supernatants were determined at the indicated time points by plaque assay and are expressed as PFU per ml.

Mouse 3T3 cells were seeded at 10⁶ cells per well in 6-well plates and infected at an MOI of 5 for 1 h at 37°C. Subsequently, the inoculum was replaced by infection medium. Cells were lysed in buffer containing 50 mM HEPES (pH 7.3), 125 mM NaCl, 1 mM EDTA, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 1 mM dithiothreitol (DTT), 25 U/ml of Benzonase, and protease inhibitors (Roche). Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and analyzed by Western blotting. The blots were probed with the monoclonal antibody M143 to detect Mx proteins (6), monoclonal mouse anti-PA antibody (G. Chase et al., unpublished data), and a rabbit antiactin antibody (Sigma). Horseradish peroxidase-labeled secondary antibodies and a chemiluminescence detection system (Pierce femto kit) were used to detect primary antibodies. Signal quantification was performed using ChemiDoc XRS equipment (Bio-Rad).

HEK 293T cells seeded in 12-well plates were transfected using the Nanofectin transfection reagent (PAA Laboratories) according to the manufacturer's protocol. Ten nanograms of pCAGGs plasmids encoding PB2, PB1, and PA and 100 ng of NP-encoding plasmid were cotransfected with 50 ng of the firefly luciferase-encoding viral minigenome construct pPolI-FFLuc-RT, which is flanked by the noncoding regions of segment 8 of influenza A virus (4). The transfection mixture also contained 25 ng of pRL-SV40, a plasmid constitutively expressing Renilla luciferase under the control of the simian virus 40 promoter, used to normalize variations in transfection efficiency. To evaluate the antiviral potential of Mx1, we cotransfected increasing amounts of Mx1-encoding plasmid. Cotransfection of the antivirally inactive mutant Mx1 (K49A) (23) was used as a control. To achieve equal amounts of transfected DNA, an empty vector plasmid was added. Twenty-four hours posttransfection, cells were lysed and firefly and Renilla luciferase activities were determined using the dual luciferase reporter assay (Promega) according to the manufacturer's protocol. Relative polymerase activity was calculated as the ratio of luciferase to Renilla luminescence.

Lungs were collected from BALB-Mx1 mice 3 and 4 days p.i. Lungs were inflated and fixed in 4% formaldehyde overnight at 4°C and subsequently embedded in paraffin. Sections (4 μm) were stained with hematoxylin and eosin (H&E) and examined under the light microscope for histopathological changes. Representative images were obtained on a Nicon Eclipse E400 microscope at magnification ×10.

Since Mx1 interferes with the transcriptional activity of vRNPs (4), we determined the polymerase activities of influenza virus strains A/Hamburg/4/09 (pH1N1) and A/Thailand/1(KAN-1)/04 (H5N1) in a minireplicon system in the presence or absence of Mx1. The H5N1 polymerase activity decreased to almost 10% at the largest amount (200 ng) of cotransfected Mx1-expressing plasmid, whereas the pH1N1 polymerase activity was reduced to only 60% (Fig. 1A). The dose of Mx1-expressing plasmid necessary to achieve 50% inhibition of the H5N1 polymerase activity (ID₅₀) was found to be 25 ng (Fig. 1A). These results demonstrate that the polymerase of the pandemic H1N1 isolate is less sensitive to Mx1 inhibition than the polymerase of the avian H5N1 strain.

To identify the viral factor determining Mx sensitivity, we exchanged each protein component of the resistant pH1N1 vRNP (i.e., the NP and the three polymerase subunits PA, PB1, and PB2) with the corresponding components of the sensitive H5N1 strain and determined the activity of the reassorted vRNPs in the minireplicon assay. The NP of H5N1 origin increased the Mx1 sensitivity of the pH1N1 polymerase (Fig. 1B and C), resulting in a drop in the ID₅₀ from over 200 ng to 36 ng of transfected Mx1 expression plasmids. In contrast, substitutions with the polymerase subunit PA, PB1, or PB2 of H5N1 did not result in enhanced Mx1 sensitivity (Fig. 1B). Conversely, the pH1N1 NP rendered the H5N1 polymerase activity relatively resistant to Mx1 inhibition (Fig. 1B and D), as demonstrated by an increase in the ID₅₀ from 25 ng to 173 ng of transfected Mx1 expression plasmids. In the absence of Mx1, the activities of both the pH1N1 and H5N1 polymerases were not affected by the source of the NP (data not shown), excluding the possibility that subunit incompatibilities were affecting Mx1 sensitivity. Together, these data indicate that the nature of the NP is the main determinant of Mx sensitivity.

To verify the importance of NP as a determinant of Mx1 sensitivity in vivo, we generated two recombinant strains, namely, wild-type strain A/Hamburg/4/09 (H1N1) (here designated pH1N1) and a reassortant A/Hamburg/4/09 strain which carries the NP gene of A/Thailand/1(KAN-1)/04 (H5N1) and is here designated pH1N1-NP(H5N1). Since pandemic H1N1 viruses are almost avirulent in BALB/c mice (14), we introduced two mutations (D131E and S186P) into the hemagglutinin (HA) protein, which have been described as increasing virulence (13, 34). Both pH1N1 and pH1N1-NP(H5N1) grew to comparable titers in MDCK cells infected with an MOI of 0.001. However, infection with the latter virus resulted in slightly higher titers 24 h p.i. (Fig. 2A). To test the replication efficiency of both viruses in Mx-expressing cells, we took advantage of stably transformed Swiss 3T3 cells that express high levels of murine Mx1 or human MxA under the control of a constitutive promoter (22) (Fig. 2B). The cells were infected with either pH1N1 or pH1N1-NP(H5N1) at an MOI of 5, and viral replication was monitored 10 or 12 h later by determining viral protein expression. The two viruses replicated with comparable efficiencies in Swiss 3T3 cells that do not express Mx proteins, as indicated by equal amounts of PA protein levels (Fig. 2C, lanes 1 to 4). The two types of viruses replicated less well in both Mx1- and MxA-expressing cells, as expected (Fig. 2C and D). However, their replication efficiencies differed significantly in the presence of either Mx1 or MxA. The PA antigen signals were much weaker following infection with the reassortant pH1N1-NP(H5N1) virus than with parental pH1N1 (Fig. 2C, lane 5 to 12). Quantification of PA signal intensities from three independent experiments confirmed the differential reduction in antigen expression levels in infected cells (Fig. 2D). These results indicate that the NP is responsible for the relative resistance of pH1N1 viruses toward the antiviral action of mouse and human Mx GTPases.

To evaluate the in vivo growth capacities of pH1N1 and pH1N1-NP(H5N1), Mx1-negative BALB/c mice were infected i.n. with 1,000 PFU. For both viruses, comparable lung titers were obtained after 48 h which were in the range of 10⁷ PFU/ml (Fig. 3A, left panel). These results indicated that the NP of H5N1 is well tolerated and does not compromise fitness of the reassortant pH1N1 virus. Next, Mx1-positive BALB-Mx1 mice were challenged with a high dose of either parental pH1N1 or reassortant pH1N1-NP(H5N1) virus. All animals infected with the parental virus showed dramatic body weight loss, and 70% of the animals succumbed within 4 days (Fig. 3B and C). In contrast, infection with the reassortant pH1N1-NP(H5N1) virus caused only moderate weight loss, and all animals survived (Fig. 3B and C). To exclude the possibility that the high virulence of pH1N1 resulted from a general suppression of the interferon response, Mx1 expression in animals was monitored 48 h after infection. Western blot analyses of lung homogenates demonstrated that both viruses induced the same amount of Mx1 protein (Fig. 3D). To determine whether the clinical signs correlated with enhanced virus replication, BALB-Mx1 mice were infected with 1,000 PFU of either pH1N1 or pH1N1-NP(H5N1), and lung titers were determined 48 h after infection. Viral titers were 10-fold higher in animals infected with pH1N1 than in animals infected with pH1N1-NP(H5N1) (Fig. 3A, right panel). To determine whether the difference in lethality between pH1N1 and pH1N1-NP(H5N1) correlated with altered lung pathology, histological tissue sections of Mx-positive mice were examined 4 days after infection (Fig. 3E). Lungs of mice infected with pH1N1 had a greater number of lesions and displayed more massive infiltration of inflammatory cells than lungs of mice infected with pH1N1-NP(H5N1). Together, these data indicate that the nature of NP determines viral replication efficiency and lung pathology in Mx1-positive mice.

The NP of the pandemic H1N1 isolate A/Hamburg/4/09 negatively affected the Mx sensitivity of the H5N1 polymerase of A/Thailand/1(KAN-1)/04, as assessed in a minireplicon system (Fig. 1C). We therefore anticipated that an A/Thailand/1(KAN-1)/04 reassortant virus equipped with the pH1N1 NP gene should gain increased Mx resistance. To prove this, we generated recombinant wild-type A/Thailand/1(KAN-1)/04, designated H5N1, and a reassortant of this parental virus, designated H5N1-NP(pH1N1), coding for the NP protein of A/Hamburg/4/09. Growth of the reassortant virus in MDCK cells infected at a low MOI of 0.001 was significantly impaired by up to 2 log₁₀ compared to that of the parental H5N1 strain (Fig. 4A). However, after infection at a high MOI of 5, the two viruses replicated with comparable efficiencies in Swiss 3T3 cells that do not express Mx proteins, as demonstrated by the accumulation of equal amounts of PA protein (Fig. 4B, lanes 1 to 4). Importantly, the replication efficiency of H5N1 and H5N1-NP(pH1N1) clearly differed in Mx-expressing cells. Compared to cells infected with H5N1-NP(pH1N1), expression of the viral PA protein was strongly reduced in H5N1-infected cells, irrespective of whether they expressed murine Mx1 or human MxA (Fig. 4B, lanes 5 to 12). Quantification of PA signal intensities from three independent experiments confirmed that the observed differences between H5N1-NP(pH1N1) and H5N1 were significant (Fig. 4C). These findings indicated that the NP of the human pH1N1 strain is able to confer partial Mx resistance to the otherwise sensitive avian H5N1 virus.

Next, we determined the growth capacity of H5N1 and H5N1-NP(pH1N1) in lungs of Mx1-negative BALB/c mice. Viral lung titers of animals infected with 1,000 PFU of reassortant H5N1-NP(pH1N1) were 100-fold lower than the titers observed for mice infected with parental H5N1 virus, in full agreement with the decreased viral replication found in cell culture (Fig. 4D, left panel). Similar differences in viral lung titers between the two virus strains were also observed in BALB-Mx1 mice (Fig. 4D, right panel), indicating that the reassortant H5N1-NP(pH1N1) virus is attenuated in both Mx1-positive and Mx1-negative mice. However, despite these differences in viral growth, the induction of Mx1 in the lungs at 48 h after infection with 10⁶ PFU was comparable (Fig. 4E). Notably, intranasal infection of BALB-Mx1 mice with 10⁶ PFU of H5N1-NP(pH1N1) virus caused rapid weight loss (Fig. 4F) and almost 100% mortality (Fig. 4G). In contrast, all H5N1-infected animals recovered rapidly (Fig. 4F) and survived (Fig. 4G). Consistent with the clinical outcome, multiple foci of extended inflammatory cell infiltrates and thickening of bronchiolar and alveolar walls were observed 3 days after infection with H5N1-NP(pH1N1) virus (Fig. 4H). On the contrary, the pulmonary pathology in mice infected with H5N1 was only moderate (Fig. 4H). Thus, despite causing considerable attenuation in mice, NP of pH1N1 rendered the H5N1 virus less sensitive to Mx1 restriction, thereby resulting in increased lung pathology and death of infected animals.