Suppression of Rac1 Signaling by Influenza A Virus NS1 Facilitates Viral Replication

Influenza A virus (IAV) is a major human pathogen with the potential to become pandemic. IAV contains only eight RNA segments; thus, the virus must fully exploit the host cellular machinery to facilitate its own replication. In an effort to comprehensively characterize the host machinery taken over by IAV in mammalian cells, we generated stable A549 cell lines with over-expression of the viral non-structural protein (NS1) to investigate the potential host factors that might be modulated by the NS1 protein. We found that the viral NS1 protein directly interacted with cellular Rac1 and facilitated viral replication. Further research revealed that NS1 down-regulated Rac1 activity via post-translational modifications. Therefore, our results demonstrated that IAV blocked Rac1-mediated host cell signal transduction through the NS1 protein to facilitate its own replication. Our findings provide a novel insight into the mechanism of IAV replication and indicate new avenues for the development of potential therapeutic targets.

This assay provides a method to quantify Rac1 activation in the cells. The amount of activated Rac1 was determined by the western blotting of the precipitated material using a Rac1-specific antibody.
Virus generation and infection experiment. The recombinant influenza virus A/WSN/33 (H1N1) was generated using a 12-plasmid IAV reverse-genetics system and propagated in MDCK cells. An influenza virus A/WSN/33 strain that lacked the NS1 gene (NS1 − H1N1 virus) was generated using pPolI-NS2 to alter the pPolI-NS reverse genetics 6 . The viral titer was determined by plaque assay or by indirect immunofluorescence microscopy with an anti-NP antibody in MDCK cells 15,53 .
For virus infection, the cells were mock-infected or infected with virus at different multiplicities of infection (MOI). After 1 hour of incubation to allow virus adsorption, the cells were cultured in complete medium and harvested at various time points post-infection GST pull-down and co-immunoprecipitation (IP) analysis. GST and the GST fusion protein were purified from E. coli BL21 cells (DE3) using glutathione Sepharose 4B beads (Amersham Biosciences, Uppsala, Sweden). An equal amount of GST or the GST fusion protein bound to glutathione beads was incubated with the lysates from transiently transfected 293T cells in NP-40 lysis buffer for 4 hours at 4 °C. Then the beads were washed three times with PBS containing 0.1% Triton X-100. The bound proteins were eluted by boiling in 2 × SDS loading buffer and analyzed by western blotting.
For the co-immunoprecipitation experiments, total cell lysates from transfected 293T cells in lysis buffer (1% Triton X-100, 150 mM NaCl, 20 mM HEPES pH 7.5, 10% glycerol, 1 mM EDTA, and protease inhibitors) were incubated with antibody at 4 °C for 2 hours. Protein G agarose beads (Sigma, USA) were added, and the samples were incubated at 4 °C overnight. The beads were washed three times with lysis buffer and boiled in 2 × SDS loading buffer for 5 minutes. The samples were analyzed by western blotting.

Subcellular localization and immunofluorescence assay.
To determine the co-localization of the NS1 and Rac1 proteins, A549 cells were co-transfected with the pEGFP-NS1 and pcDNA4.0-myc-Rac1 plasmids. After 24 hours, the cells were washed three times with PBS, fixed in 4% paraformaldehyde for 15 minutes at room temperature, and permeabilized with 0.2% Triton X-100 for 5 minutes. After blocking with 5% BSA for 30 minutes, the cells were incubated for 1 hour with an anti-myc monoclonal antibody at room temperature. After washing with PBS, the cells were incubated for 1 hour with a TRITC-conjugated anti-mouse IgG antibody and then visualized with an Olympus confocal microscope. miRNA expression vectors. Expression vectors that encoded shRNAs specific for Rac1 or a control were constructed using psiSTRIKE TM vectors (Promega) according to the manufacturer's protocol. Primer sequences for the control shRNA and shRac1-specific shRNAs were designed according to the instructions provided by the manufacturer and are shown in Table 1. The primers were annealed to form double-stranded DNA and then inserted into the psiSTRIKE TM vectors. Bacteria were transformed with the resulting constructs, and a reasonable number of colonies were obtained. The resulting shRNA-encoding expression vectors were confirmed by digestion with the restriction enzyme PstI according to the manufacturer's protocol. C. elegans miR239b was used as a negative control.
Plaque assay. Ten-fold serial dilutions of the different virus types were used to infect confluent MDCK cells in 24-well plates for 1 h at 37 °C. The virus inocula were removed by washing with PBS. The cell monolayers were overlaid with an agar medium (DMEM supplemented with 1% low melting point agarose and 1 μ g/ml of TPCK-treated trypsin) and incubated at 37 °C for 3 days. Then, the plates were fixed with 4% paraformaldehyde for 1 h, and the agarose was carefully removed. Staining buffer (0.1% crystal violet and 20% ethanol in water) was added for 10 min, followed by washing before the virus titer was determined. All data were expressed as the mean of three independent experiments. Statistical analysis. One representative western blot is shown for three similar independent experiments.
The protein expression levels were quantified using the Image J software, and the results were presented as the mean ± SD of triplicate experiments. The data were evaluated using a t-test with the SPSS 11.5 software (SPSS, Inc., Chicago, IL, USA). P < 0.05 was considered statistically significant.

Results
NS1 alters cell morphology and inhibits cell migration. The IAV NS1 protein is a multifunctional protein that has been previously shown to interact with many host cell factors and play important roles in virus replication 54 . To better understand the roles of NS1 in influenza virus pathogenesis, we generated stable cell lines (P1, P2 and P3) that over-expressed the NS1 protein. The N1, N2 and N3 cell lines were used as the negative controls. NS1 expression was confirmed using western blotting and immunofluorescence assays (Fig. 1A,B). Then, we observed the phenotypic changes in these two stable cell lines (Fig. 1C). The spreading lamellipodia were visible in the control N1, N2 and N3 cell lines, but were retracted in the P1, P2 and P3 cell lines. Figure 1D shows the different percentages of the cell spreading area. Thus, NS1 inhibited the formation of cellullar lamellipodia. Because the lamellipodia is very important for the cell migration ability 55 , we performed wound healing and transwell assays to investigate whether NS1 over-expression inhibites cell migration. Wounds were made in a confluent cell layer with a sterile pipette tip, as shown in Fig. 1E. The control cell lines exhibited better healing abilities than the NS1 over-expression cell lines. Further, the transwell migration assay was performed to determine the differences between the two cell lines. As shown in Fig. 1F, there was a significant reduction in the migration of the P1, P2 and P3 cells from the small holes compared to the control N1, N2 and N3 cells, which indicated that over-expression of NS1 significantly inhibited the cell migration speed. Taken together, our results indicated that the NS1 protein inhibited lamellipodia formation and cell motility. NS1 down-regulates Rac1 activity. Our previous results showed that NS1 inhibited cellular lamellipodia formation and motility. Because the Rac1 protein is essential for platelet lamellipodia formation 30 , we hypothesized that NS1 might regulate the Rac1 protein functions. To test this hypothesis, we used the western blotting and GST-pull down techniques to analyze Rac1 expression and activity in the NS1 over-expressing cell lines. As shown in Fig. 2A, the Rac1 expression levels were unchanged by NS1 over-expression. Previous studies demonstrated that Rac1's function as a molecular switch was regulated by the GTP/GDP exchange 56 . Consequently, we measured the Rac1-GTP level. The Rac1-GTP level was significantly reduced in the NS1 over-expressing P1, P2 and P3 cell lines compared to the control N1, N2 and N3 cell lines (Fig. 2B). To validate this observation, the NS1 protein was transfected into 293T cells, and endogenous Rac1 activity was assessed. As shown in Fig. 2C, transient over-expression of NS1 also resulted in a significant reduction in the GTP-bound Rac1 level. Therefore, we generated an influenza virus A/WSN/33 strain that lacked the NS1 gene (NS1 − H1N1 virus) (Fig. 2D) and infected A549 cells with the wild type (wt) H1N1 or NS1 − H1N1 virus. As shown in Fig. 2E, the Rac1 activity in the wt virus-infected cells was significantly lower than the activity in the cells infected with the mutated virus. Therefore, the NS1 protein suppressed Rac1 activity and reduced the Rac1-GTP level, although the Rac1 protein expression level remained unaffected.

NS1 protein directly interacts with cellular Rac1. Our previous results demonstrated that NS1
over-expression and wt virus infection resulted in a reduced level of GTP-bound Rac1, next we investigated whether NS1 regulates Rac1 activity through a direct interaction. As shown in Fig. 3A, the use of GST fusion constructs of the NS1 protein in the pull-down assays demonstrated that the NS1 protein interacted with Rac1. We further confirmed the direct interaction of NS1 and cellular Rac1 via co-immunoprecipitation in transfected cells (Fig. 3B). Co-transfection with pEGFP-NS1 and pcDNA4.0-myc-Rac1 resulted in a complete overlay of the GFP and red anti-myc-Rac1 immunofluorescence, indicating the co-localization of NS1 and Rac1 (Fig. 3C). Moreover, we infected A549 cells with the wt H1N1 and NS1 − H1N1 viruses, immunoprecipitation of Rac1 only detected in wt virus but not NS1 − H1N1 virus infection (Fig. 3D), which indicated that the NS1 protein directly interacted with the cellular Rac1 protein. The co-localization of viral NS1 and endogenous Rac1 in the wt virus-infected A549 cells was also visualized by confocal microscopy (Fig. 3E). Collectively, our results indicated that the NS1 protein directly interacted with Rac1 in the cytoplasm and nucleus.

NS1 suppresses Rac1 activity by inhibiting its ubiquitination. Recent studies suggested that Rac1
was modulated by ubiquitination and SUMOylation. The ubiquitin pathway is the most common method used by cells to regulate protein activation, localization and function 57,58 . Therefore, to identify whether ubiquitination was involved in the NS1-mediated regulation of Rac1, we co-transfected Myc-Rac1 and HA-Ub with pcD-NA3.0-NS1 or pcDNA3.0 in 293T cells. Then, a GST-pull down assay was performed on these samples. As shown in Fig. 4A, the presence of NS1 decreased Rac1 activity. Moreover, Rac1 ubiqitination enhanced the Rac1-GTP level, whereas NS1 suppressed this process. Thus, our data indicated that Rac1 activity could be modulated by ubiquitination.
Different linkage-specific ubiquitin chain modifications have been shown to mediate distinct cellular events. K63-linked polyubiquitin chains are involved in the activation of antiviral signaling pathways and the regulation of a variety of nonproteolytic cellular functions, including gene transcription, subcellular localization, protein activity, intracellular trafficking and viral budding 59,60 . Conversely, K48-linked polyubiquitin chains generally target proteins for proteasomal degradation 61,62 . Because our results indicated that NS1 inhibited Rac1 ubiquitination, next we investigated the specific types of polyubiquitin chain linkages modulated by NS1. As shown in Fig. 4B, Rac1 ubiquitination was reduced by the co-transfection of K63-linked polyubiquitin chains, whereas no difference was noted with the co-transfection of K48-linked polyubiquitin chains, indicating that Rac1 was associated with K63-linked ubiquitination.
The C-terminus of Rac1 is crucial for binding to several regulatory proteins and the proper localization of Rac1 in the plasma membrane, and the C-terminal 21 aa of Rac1 were confirmed as the nuclear localization sequence (NLS) 58 . To confirm the location of the ubiquitin reaction in the cell, we constructed a mutant plasmid with a deletion of the C-terminal 21aa of Rac1 (called Rac1 Δ WT). The mutation plasmid is shown in Fig. 4C (Rac1 amino acid sequence) colored in yellow-green. Figure 4D showed that the Rac1 Δ WT protein was only Scientific RepoRts | 6:35041 | DOI: 10.1038/srep35041 The percentage of the cell spreading area. Image J software was used to analyze the cell spreading area. The results are presented as the mean ± SD of triplicate experiments (***p < 0.001 relative to the control group). The data were evaluated using a t-test with the SPSS 11.5 software program (SPSS, Inc., Chicago, IL, USA). (E) A wound healing assay was performed to detect the migration ability in the two different cell lines. The indicated cell lines were seeded into six-well plates. After the cells grew to confluence, wounds were made using 10 μ l sterile pipette tips. Then, the cells were washed with PBS, and fresh medium containing 10% FBS was added. After incubation for 24 hours at 37 °C, the cells were fixed and photographed. (F) The transwell migration assay showed the different migratory responses of the two stable cell lines. The cell lines were added to the upper chamber and 500 μ l of appropriate medium supplemented with 10% FBS was added to the lower chamber. After incubation for 24 hours at 37 °C, the chambers were stained and photographed. Three random fields were analyzed for each chamber. expressed in the cytoplasm, which was consistent with previous reports 63, 64 . Then, we used the wt Rac1 protein as the control to detect the degree of ubiquitination the Rac1 Δ WT protein. As shown in Fig. 4E, the degree There was no significant difference in the Rac1 protein expression level between the two different cell lines. The two cell lines were seeded into six-well plates. After 24 hours, the Rac1 protein expression levels were assessed in the whole cell lysates using the indicated antibodies as appropriate. (B) GTP-bound Rac1 levels were lower in the NS1 over-expressing cell lines than in the negative control cell lines. Lysates from the two stable cell lines were incubated with equal amounts of GST-PAK bound glutathione Sepharose 4B beads. After washing, the bound proteins were analyzed by western blotting using an anti-Rac1 antibody. The protein band intensity was measured using the ImageJ software (NIH). Data represent the mean fold change ± S.D. of three independent experiments (*p < 0.05 relative to the control group). (C) Rac1 activity was reduced in 293T cells transfected with pcDNA3.0-flag-NS1. 293T cells were co-transfected with Myc-tagged Rac1 and the indicated plasmids. After 30 hours, the whole cell lysates were subjected to the GST-PAK pull down assay, followed by western blotting analysis using an anti-Myc antibody. The protein band intensity was measured using the ImageJ software (NIH). Data represent the mean fold change ± S.D. of three independent experiments (*p < 0.05 relative to the control group). (D) Confirmation of the recombinant virus (NS1 − H1N1). A549 cells were infected with the H1N1 virus or the recombinant virus lacking the NS1 gene (NS1 − H1N1) at an MOI of 0.5 or left uninfected. After 24 hours, whole cell lysates were subjected to western blotting. (E) Rac1 activity in wild type virus-infected A549 cell was lower than the activity in cells infected with the mutated virus. A549 cells were infected with the A/WSN/33 H1N1 virus and the mutated NS1 − H1N1 virus (MOI = 0.5) for 24 hours. Then, the cell lysates were collected and subjected to the GST-PAK pull down assay followed by western blotting analysis using an anti-Rac1 antibody. The protein band intensity was measured using the ImageJ software (NIH). Data represent the mean fold change ± S.D. of three independent experiments (***p < 0.001 relative to the control group).
Scientific RepoRts | 6:35041 | DOI: 10.1038/srep35041 of ubiquitination was increased by the deletion of the C-terminal 21 aa, which indicated that the ubiquitination reactions primarily occurred in the cytoplasm. To accurately determine the Rac1 ubiquitination sites, we   (Fig. 4C). When lysine 147 was mutated to arginine (K147R), the ubiquitination nearly disappeared, which indicated that K147 was the major site in Rac1 for ubiquitin modification, although the K184 and K186 sites also underwent ubiquitination to a lesser extent (Fig. 4F). Then, 293T cells were transfected with HA-Ub, HA-K48R and HA-K63R for 24 hours and infected with the wt H1N1 and NS1 − H1N1 viruses. The anti-Rac1 antibody was used to precipitate endogenous Rac1 from the cell lysates, and western blotting of the precipitates was conducted using an HA-specific antibody. We found that the wt virus inhibited Rac1 activity through interference with its K63-linked poly-ubiquitination compared to the NS1 − H1N1 virus (Fig. 4G). These data indicated that NS1 suppressed Rac1 activity by inhibiting K63-linked ubiquitination.

NS1 inhibits Rac1 SUMOylation, resulting in reduced Rac1-GTP levels. Castillo-Lluva et al.
recently demonstrated that GTP-bound Rac1 levels could be regulated by SUMOylation, thereby stimulating cell lamellipodia, cell migration and invasion 25 . We confirmed these results in our experiments. We found that Rac1 SUMOylation was reduced following NS1 over-expression in transfected 293T cells (Fig. 5A). And with the presence of SUMO and UBC9, we observed that the NS1 protein inhibited SUMOylated GTP-bound Rac1 (Fig. 5B). Additionally, we detected the effect of SUMOylation on Rac1 activity. Figure 5C,D showed that GTP-bound Rac1 was increased by the presence of SUMO and UBC9. Next, we investigated potential SUMOylation sites in Rac1. According to the study by Castillo-Lluva and colleagues, SUMO-1 was conjugated to lysine residues 183, 184, 186 or 188 in nearly 95% of SUMOylated Rac1; therefore, we selected these amino acid residues for analysis as potential Rac1 SUMOylation sites. As shown in Fig. 5E, the K184 and K188 sites were the key residues responsible for the SUMO modifications. Furthermore, we assessed the role of NS1 on Rac1 SUMOylation. Figure 5F shows the results obtained after the precipitation of flag-tagged NS1 and probing with the HA antibody to detect the precipitated NS1. We found that NS1 itself could be SUMOylated, which confirmed a previous report 65 . Then, the Rac1 SUMOylation assay was performed following infection with the wt H1N1 and NS1 − H1N1 viruses. The Rac1 SUMOylation level in the wt virus-infected cells were lower than the level in the cells infected with the mutated virus (Fig. 5G). The results were consistent with NS1 protein over-expression. Cumulatively, these results strongly indicated that competitive inhibition of Rac1 SUMOylation by NS1 decreased the Rac1-GTP level.

Suppression of Rac1 activity results in arrested cell migration and altered cell morphology.
To determine whether Rac1 is the key regulator that mediates NS1-induced cell morphological alterations, we designed shRNA sequences specific for Rac1. Based on the depletion efficiency of the Rac1-specific shRNAs (Fig. 6A,B), we chose shRac1-1 as the Rac1 interfering sequence for the following experiments. Stable A549 cell lines expressing shRac1-1 and Rac1 were generated and confirmed by western blotting analysis (Fig. 6C,D). The 2 # and 5 # shRac1 and the 1 # and 2 # Rac1 over-expression cell lines were chosen and photographed by SEM. They showed that interference with Rac1 led to the contraction of the cell lamellipodia, whereas over-expression of Rac1 promotes lamellipodial spreading, which is concordant with the cell morphology of the stable NS1 cell lines (Fig. 1C). As shown in Fig. 6E ( the upper part of Fig. 6), the numbers of migrated cells from the shRac1 cell lines were lower than the numbers of migrated cells from the control cell lines, similar to the results obtained for the NS1 over-expressing cell lines P1, P2 and P3 shown in Fig. 1F. The wound healing assay indicated that interference with Rac1 protein expression resulted in the inhibition of cell migration, whereas over-expression of Rac1 promoted cell migration and sped cellular scratch healing (Fig. 6F,G). Thus, the above data indicated that the NS1-induced retraction of lamellipodia and inhibition of cell migration might be due to direct suppression of the normal function of Rac1.
The impact of Rac1 on H1N1 viral replication. Rac1 was reported to suppress A/PR/8/34 virus (PR8) replication 35 . Due to the different characteristics between the A/PR/8/34 and A/WSN/33 influenza viruses, we investigated whether Rac1 also inhibited A/WSN/33 viral replication. To answer this question, the A549-Rac1 2 # cell line, A549-shRac1 5 # cell line, and the control were infected with the WSN virus at an MOI of 0.5. The virus titers were determined via plaque assay. As shown in Fig. 7A, Rac1 over-expression inhibited viral replication, whereas interfering with Rac1 facilitated viral replication. The impact of Rac1 on H1N1 replication was confirmed in different stable cell lines by analyzing the viral NP protein level (Fig. 7B). Next, we observed the replication of the wt H1N1 virus in both the Rac1 over-expressing and the Rac1 interfered cell lines by SEM (Fig. 7C). The viral particles in the Rac1 over-expressing cell lines was lower than that in the Rac1 interfering cell expressed in the cytoplasm. 293T cells were co-transfected with pcDNA3.0-flag-NS1 and pcDNA4.0-myc-Rac1 WT or pcDNA4.0-myc-Rac1 Δ WT. Nuclear and cytoplasmic proteins were extracted and assayed by western blotting with the indicated antibodies. Lamin B1 and tubulin were used as loading controls for the nuclear and cytoplasmic fractions, respectively. (E) Deletion of the Rac1 C-terminal NLS sequence enhanced its ubiquitination. 293T cells were transfected with the plasmids as indicated. An anti-Myc antibody was used to precipitate the cell lysates, and a western blotting was using an anti-HA antibody. (F) The ubiquitin sites of the Rac1 protein. 293T cells were co-transfected with HA-Ub, pcDNA3.0 and mutated Rac1 plasmids. An anti-Myc antibody was used to precipitate the lysates, and western blotting was using an anti-HA antibody. (G) The wild type virus inhibited Rac1 activity through interference with its K63-linked poly-ubiquitination. 293T cells were transfected with HA-Ub, HA-K48R and HA-K63R as indicated for 24 hours. Then, the cells were infected with the wt virus and the mutated virus (MOI = 0.5) for another 24 hours. An anti-Rac1 antibody was used to precipitate endogenous Rac1 from the cell lysates, and western blotting was using an HA-specific antibody. inhibited Rac1 SUMOylation. 293T cells were co-transfected with pcDNA3.0 or pcDNA3.0-flag-NS1 with pcDNA4.0-myc-Rac1, pXJ40-HA-SUMO-1 and UBC-9 as indicated. Anti-Rac1 antibodies were used to precipitate the cell lysates, and western blotting was conducted using HA-specific antibodies. (B) NS1 decreases SUMOylated GTP-bound Rac1. 293T cells were co-transfected with pcDNA3.0 or pcDNA3.0-flag-NS1 with pcDNA4.0-myc-Rac1, pXJ40-HA-SUMO-1 and UBC-9 as indicated. After 30 hours, the whole cell lysates were subjected to the GST-PAK pull down assay followed by western blotting using a Rac1-specific antibody. (C) Rac1 SUMOylation. 293T cells were co-transfected with pcDNA4.0-myc-Rac1, pXJ40-HA-SUMO-1 and UBC-9 as indicated. Anti-Rac1 antibodies were used to precipitate the cell lysates, and western blotting was conducted using HA-specific antibodies. (D) Rac1 SUMOylation can increase Rac1 activity. 293T cells were cotransfected with pcDNA4.0-myc-Rac1, pXJ40-HA-SUMO-1 and UBC-9 as indicated. After 30 hours, the whole cell lysates were subjected to the GST-PAK pull down assay followed by western blotting using a Rac1-specific antibody. (E) The SUMOylation sites in the Rac1 protein. 293T cells were co-transfected with pcDNA3.0, pXJ40-HA-SUMO-1, UBC-9, and the mutated Rac1 plasmids as indicated. Rac1 antibodies were used for precipitation of the lysates, and western blotting was performed with HA-specific antibodies. (F) Competition for SUMOylation by the NS1 protein inhibited Rac1 SUMOylation. 293T cells were co-transfected with pcDNA3.0 or pcDNA3.0-flag-NS1 with pcDNA4.0-myc-Rac1, pXJ40-HA-SUMO-1 and UBC-9 as indicated. Anti-Rac1 antibodies were used to precipitate the cell lysates, and western blotting was performed using HAspecific antibodies. (G) The wild type virus can inhibit Rac1 SUMOylation. 293T cells were co-transfected with pcDNA3.0 or pcDNA3.0-flag-NS1 with pcDNA4.0-myc-Rac1, pXJ40-HA-SUMO-1 and UBC-9 as indicated. Then, the cells were infected with the A/WSN/33 H1N1 virus and the mutated NS1 − H1N1 virus (MOI = 0.5) for another 24 hours. Anti-Rac1 antibodies were used to precipitate the cell lysates, and western blotting was conducted using HA-specific antibodies.   Table 1) for 72 hours, and Western blotting was used to select effective shRac1 plasmids. (B) The efficacy of Rac1-specific shRac1-1 was confirmed by the immunofluorescence assay. A549 cells transfected with control shRNA and shRac1-1 for 28 hours were fixed, permeabilized, and stained for Rac1 lines, which was the same as the result obtained in the plaque assay (Fig. 7A). Thus, Rac1 can inhibit H1N1 virus A/WSN/33 replication.
The impact of Rac1 on NS1 − H1N1 viral replication. Our previous results indicated that the IAV viral NS1 protein could inhibit the normal function of the Rac1 protein, whereas interference with Rac1 resulted in increased viral replication. Next, SEM and plaque assays were performed following infection with the wt H1N1 and NS1 − H1N1 viruses. Figure 8A shows that the H1N1 viral particles in the Rac1 over-expressing cell lines was greater than the NS1 − H1N1 virus particles. This result was validated by the plaque assay. This phenomenon suggested that the NS1 protein plays an important role in enhancing viral replication by inhibiting Rac1 activity. We further tested the Rac1 interfering cell lines infected with the wild type and mutant viruses. As shown in Fig. 8B, there was no significant difference in the amount of virus particles between the wt and mutant viruses in the absence of Rac1, which indicated that the NS1 protein augmented virus replication partially by inhibiting Rac1 activity. Overall, the A/WSN/33 virus H1N1 inhibits Rac1 activity to facilitate viral replication via the viral NS1 protein. The A549-Rac1 2 # and A549-shRac1 5 # cell lines and the control cell lines were infected with the H1N1 virus at an MOI of 0.5. The cell culture supernatants were collected at the indicated times post infection, and the plaque assay was used to measure the virus titers. Data represent the mean fold change ± S.D. of three independent experiments (*p < 0.05 relative to the control group). (B) Over-expression of Rac1 decreased the NP protein content. The A549-Rac1 2 # and A549-shRac1 5 # cell lines and the control cell lines were infected with the H1N1 virus as indicated at an MOI of 0.5. After 24 hours, the cells were collected and analyzed by western blotting with the indicated antibodies. Relative NP band intensities (relative to β -actin) were measured using the ImageJ software (NIH). Data represent the mean fold change ± S.D. of three independent experiments (*p < 0.05 relative to the control group). (C) Rac1 protein over-expression inhibited virus replication, whereas interference with Rac1 promoted viral replication. The A549-Rac1 2 # and A549-shRac1 5 # cell lines were infected with the H1N1 virus for 24 and 48 hours. Both stable cell lines were examined by SEM after infection.

Discussion
IAVs rely on host factors to assist their propagation within the host. The identification of these cellular proteins will help to elucidate the mechanisms by which the viruses acquire the potential to become pandemic in the human population. In our study, we generated NS1 over-expressing cell lines, we found that NS1 over-expression induced pseudopodial retraction and inhibited cell migration. Previous studies reported that the small GTPase Rac1 was involved in cytoskeletal rearrangements and lamellipodia formation and played key roles during virus entry, replication and release 33,66,67 , thus, Rac1 inhibition leads to enhanced virus production 35 . Therefore, we measured the Rac1 protein expression levels in our NS1 over-expressing cell lines. The Rac1 expression level remained unchanged, but its activity was significantly reduced with NS1 protein over-expression. We demonstrated that Rac1 ubiquitination and SUMOylation were inhibited by NS1 and that the inhibition of these post-translational modifications led to decreased Rac1 activity and enhanced viral replication.
In addition to modulation by GEFs, GAPs and GDIs, Rac1 is subject to modifications 68 . Some studies have reported that the regulation of Rac1 drives diverse cellular processes during virus replication 69,70 , and the ubiquitination and SUMOylation of Rac1 have been suggested to be involved in pseudopodial growth and the inhibition The H1N1 virus particles were significantly higher than the NS1 − H1N1 virus particles in the Rac1 over-expressing cell lines. The A549-Rac1 2 # cell line was infected with the H1N1 virus and the NS1 − H1N1 mutant virus at an MOI of 0.5. SEM was performed on the cell lines after they were infected for 24 and 48 hours. The cell culture supernatants were collected at the indicated times post infection, and the plaque assay was used to measure the virus titers. Data represent the mean fold change ± S.D. of three independent experiments (*p < 0.05, **p < 0.005 relative to the control group). (B) The NS1 − H1N1 virus particles was approximately the same as the wild type virus particles in the Rac1-interfered cell lines. The A549-shRac1 5 # cell line was infected with the H1N1 virus and NS1 − H1N1 mutant virus at an MOI of 0.5. SEM was performed on the cell lines after infection for 24 and 48 hours. The cell culture supernatants were collected at the indicated times post infection, and the plaque assay was used to measure the virus titers. Data represent the mean fold change ± S.D. of three independent experiments. of virus replication 71,72 . Currently, little is known regarding the molecular mechanisms regulated by Rac1. Here, our studies revealed a previously unknown function of the viral NS1, which directly interacted with the Rac1 protein. Our experimental results showed that the Rac1 protein can be modified by Lys63-linked ubiquitination and SUMO1-conjugated SUMOylation. Lysine 147 of Rac1 was most likely the main acceptor site for ubiquitin, which was previously reported by Orane Visvikis 58 ; however, we demonstrated that this site of the Rac1 protein was primarily responsible for K63-linked ubiquitination during influenza virus infection. In addition to this finding, the lysine 184 and 186 sites also contributed to ubiquitination at a certain extent. Lysine 184 and 188 are the sites for Rac1 SUMOylation, which is in accordance with the data presented by Castillo-Lluva 25 . Importantly, Rac1 can inhibit viral replication, as demonstrated with the A/PR/8/34 virus by Ehrhardt et al. 35 . Our results clearly showed that the H1N1 A/WSN/33 virus inhibited Rac1 activity to facilitate viral replication via the direct interaction of the viral NS1 protein and the cellular Rac1 protein.
Influenza A virus takes advantage of host posttranslational modifications for its own benefit. The nonstructural NS1 protein is the first discovered and the most frequent SUMO target of IAV, which is SUMOylated to facilitate viral replication 73 . Our results showed that the SUMOylation of the small GTP-binding protein Rac1 was competitively inhibited by the viral NS1 protein in the host cell, indicating that IAV profoundly relied on SUMOylation to survive in host cells.
Consistent with our previous observations, the ability of NS1 to regulate viral replication by inhibiting Rac1 activity provides new insights into viral replication. This finding demonstrates the importance of the NS1 protein in regulating the host cellular response and paves the way toward new potential therapeutic targets.