The NF-κB inhibitor SC75741 efficiently blocks influenza virus propagation and confers a high barrier for development of viral resistance

Authors


For correspondence. E-mail ludwigs@uni-muenster.de; Tel. (+49) (0)251 8357791; Fax (+49) (0)251 8357793.

Summary

Ongoing human infections with highly pathogenic avian H5N1 viruses and the emergence of the pandemic swine-origin influenza viruses (IV) highlight the permanent threat elicited by these pathogens. Occurrence of resistant seasonal and pandemic strains against the currently licensed antiviral medications points to the urgent need for new and amply available anti-influenza drugs. The recently identified virus-supportive function of the cellular IKK/NF-κB signalling pathway suggests this signalling module as a potential target for antiviral intervention. We characterized the NF-κB inhibitor SC75741 as a broad and efficient blocker of IV replication in non-toxic concentrations. The underlying molecular mechanism of SC75741 action involves impaired DNA binding of the NF-κB subunit p65, resulting in reduced expression of cytokines, chemokines, and pro-apoptotic factors, subsequent inhibition of caspase activation and block of caspase-mediated nuclear export of viralribonucleoproteins. SC75741 reduces viral replication and H5N1-induced IL-6 and IP-10 expression in the lung of infected mice. Besides its virustatic effect the drug suppresses virus-induced overproduction of cytokines and chemokines, suggesting that it might prevent hypercytokinemia that is discussed to be an important pathogenicity determinant of highly pathogenic IV. Importantly the drug exhibits a high barrier for development of resistant virus variants. Thus, SC75741-derived drugs may serve as broadly non-toxic anti-influenza agents.

Introduction

Influenza virus infection leads to the activation of a variety of intracellular signalling pathways that are in part exploited by the virus to support efficient replication. Among these, activation of the IKK/NF-κB module is one of the hallmark cell responses regulating antiviral cytokine expression, including expression of interferon-β (IFN-β). In unstimulated cells NF-κB is sequestered in an inactive form in the cytoplasm by the inhibitor of κB (IκB). Upon influenza virus infection the classical NF-κB signalling pathway is induced, which includes phosphorylation and degradation of IκB. Thereby, the nuclear localization signal of transcriptional active NF-κB factors such as p65 and p50 is exposed and enables the translocation into the nucleus. In the context of an influenza virus infection, accumulation of single stranded 5′-triphosphate vRNA is believed to be the major inducer of NF-κB (Pichlmair et al., 2006). However, the expression of viral proteins, such as the haemagglutinin (HA), the nucleoprotein (NP), or the matrix protein (M1) has also been shown to activate NF-κB signalling (Pahl and Baeuerle, 1995; Flory et al., 2000). NF-κB is commonly regarded as a major antiviral factor because it regulates the expression of inflammatory cytokines, chemokines and immunoreceptors. However, recently it has been shown by us and others that influenza A viruses exploit this signalling pathway for efficient replication (Nimmerjahn et al., 2004; Wurzer et al., 2004). Several mechanisms have been proposed to be involved in this virus-supportive action of NF-κB. The beneficial function was shown to be at least in part due to NF-κB-dependent expression of pro-apoptotic factors such as TNF-related apoptosis inducing ligand (TRAIL) or FasL (Wurzer et al., 2004). These ligands promoted caspase activation that in turn resulted in enhanced release of viral RNP complexes from the nucleus (Wurzer et al., 2003) presumably via controlled degradation and widening of nuclear pores (Kramer et al., 2008). Another recently described virus supporting action of NF-κB is the counteraction of the type I IFN-induced gene (ISG) expression via induction of the suppressor of cytokine signalling-3 (SOCS-3) (Pauli et al., 2008) and/or by suppression of ISG promoter regions (Wei et al., 2006). Furthermore, it was suggested that NF-κB differentially regulates viral RNA synthesis (Kumar et al., 2008). These various virus-supportive functions imply that NF-κB might be a suitable target for antiviral intervention. As a first proof-of-principle of this concept it has been shown that inhibition of NF-κB activity by chemical inhibitors or the widely used drug acetylsalicylic acid (ASA), results in an impaired influenza virus replication in vitro and in vivo (Wurzer et al., 2004; Mazur et al., 2007). However, antiviral acting concentrations of ASA are far too high to be reached in human lungs by systemic administration and aerosolic treatment may exhibit other irritative side-effects.

In the present manuscript we focused on a novel NF-κB inhibiting compound, SC75741 that has recently been described as an efficient blocker of the pathway in non-toxic concentrations (Leban et al., 2007). Here we show that micromolar concentrations of SC75741 efficiently suppress influenza virus replication without toxic side-effects and a high barrier for development of resistance.

Results

SC75741 inhibits replication of influenza A and B viruses

In a first set of experiments we investigated whether SC75741 may exhibit antiviral effects on influenza virus replication in cultured cells. Epithelial Madin–Darby canine kidney cells (MDCK) were infected with either human isolate of the highly pathogenic H5N1 A/Thailand/1(KAN-1)/2004 (Fig. 1A), the highly pathogenic avian H7N7 influenza virus A/FPV/Bratislava/79 (Fig. 1B) or the oseltamivir-resistant swine-origin influenza virus strain A/Nordrhein-Westfalen/173/09 (H1N1v) (Fig. 1C). SC75741 was added to the infection media in a concentration of 1 μM, 2 μM or 5 μM, which was left on the cells throughout the infection period. Determination of progeny virus titres 24 h post-infection (p.i.) revealed a concentration-dependent inhibition of virus propagation by SC75741 (Fig. 1A–C). Treatment of MDCK cells with 5 μM SC75741 confirmed its inhibitory potential on virus replication in mono- and multicyclic infections with highest efficiency at later stages of infection (Fig. 1D). Furthermore, SC75741 exhibited antiviral activity on several other influenza A viruses of the H1N1 and H3N2 subtype (data not shown) and on influenza B virus replication (Fig. 1E). To verify the action of SC75741 in human cells, we examined virus propagation of highly pathogenic H5N1 and H7N7 viruses in the human alveolar type II epithelial cell line A549 in the presence or absence of SC75741. Again, SC75741 provoked an efficient reduction of progeny virus titres in a concentration-dependent manner (Fig. 1F and G). These data indicate that SC75741 has broad antiviral properties towards various influenza viruses and in various cell-lines. In a further set of experiments SC75741 was removed from H7N7-infected A549 cells 6 h p.i. Subsequently, samples were left untreated or treated again with SC75741 for further 18 h. Determination of viral titres exhibits that the inhibitory effect of SC75741 is reversible upon removal of the compound (Fig. 1H). However, comparing the inhibitory effect of SC75741 on influenza A virus replication side by side with amantadine and oseltamivir, revealed a damped disability of SC75741, although the compound still is effective in inhibition of viral propagation (Fig. 1I).

Figure 1.

SC75741 efficiently blocks influenza virus propagation of influenza A and B viruses. MDCK cells were infected with (A) highly pathogenic influenza virus A/Thailand/1(KAN-1)/2004 (H5N1) (MOI = 0.001), (B) A/FPV/Bratislava/79 (H7N7) (MOI = 0.001), (C) pandemic swine-origin influenza A/Nordrhein-Westfalen/173/09 (H1N1v) (MOI = 0.1) or (E) influenza B/Maryland/59 (FluB) (MOI = 0.1). SC75741 was added to the infection media in a concentration of 1 μM, 2 μM or 5 μM. Supernatants were assayed for progeny virus yields 24 h post-infection in standard plaque titrations. Virus yields are depicted in plaque forming units per ml (pfu ml−1).

D. MDCK cells were infected with H7N7 (MOI = 0.001) and incubated in presence of 5 μM SC75741, 0.05% DMSO or were left untreated for the time periods indicated followed by virus titration.

F and G. A549 cells were infected with H5N1 (F) or H7N7 (G) (MOI = 0.001) for 24 h and SC75741 was added to the infection media in a concentration of 1 μM, 2 μM or 5 μM.

H. A549 cells infected with H7N7 (MOI = 0.001) were treated with SC75741 (5 μM) or solvent. Six-hour post-infection medium was exchanged and cells were incubated in the presence or absence of the inhibitor until determination of virus titres at 24 h post-infection.

I. A549 cells infected with H7N7 (MOI = 0.001) were treated with SC75741 (5 μM) or as reference compounds amantadine (5 μM) and oseltamivir (2 μM). Virus propagation was estimated 24 h post-infection.

Statistical significance was assessed by Student's t-test (*P < 0.05, **P < 0.01, ***P < 0.001).

SC75741 does not cause cytotoxic side-effects but affects long-term cell proliferation

Use of drugs that target cellular factors may raise concerns about side-effects on the host cell. Thus we analysed in a cytotoxicity assay whether antiviral acting concentrations of SC75741 would have an effect on cell viability. Therefore, cells were treated with SC75741 for different time periods (up to 96 h) and were stained with propidium iodide (PI) to detect the fraction of dead cells in each sample (Fig. 2A). No significant changes in the percentage of dead versus living cells were observed during a 96 h observation period. Within a time frame of about 30 h of treatment of cells with SC75741 the compound had a slight effect in MTT assays. Fifty hours and 65 h upon incubation of cells with SC75741 cell viability was reduced in a concentration-dependent manner indicating a cytostatic effect at prolonged treatment (Fig. 2B). However, in general it can be stated that there are no adverse side-effects of the compound at concentrations used and for the time periods employed during the infection experiments.

Figure 2.

SC75741 shows no toxic effects on A549 cells but results in inhibition of long-term cell proliferation.

A. A549 cells were incubated in the presence and absence of 5 μM SC75741 or DMSO for 24, 48, 72, and 96 h. Percentage of viable cells was measured by staining of dead cells with PI as described in Experimental procedures.

B. A549 cells were treated with increasing amounts of SC75741 (1, 2, 5 or 10 μM) or 0.02% DMSO for the indicated time periods. Cell viability was estimated via a colorimetric assay that followed incubation of cells with MTT. OD-values from untreated controls were set as 100%.

The SC75741 inhibitor specifically inhibits NF-κB-mediated signalling on a transcriptional level

SC75741 was previously described to belong to a novel class of potent NF-κB signalling inhibitors (Leban et al., 2007). To explore the molecular mode of action of the drug we analysed several steps of the NF-κB activation cascade in the presence or absence of SC75741 (Fig. 3A–C). While induction of IκBα degradation by tumour necrosis factor α (TNFα), a potent NF-κB activating agent, was only slightly delayed upon SC75741 treatment (Fig. 3A), no effects were observed on TNFα or influenza virus-induced p65 phosphorylation (Fig. 3B and C). Similarly, SC75741 did also not affect nuclear translocation of p65 upon TNFα stimulation (Fig. 3D). To evaluate the activity of the drug towards the transactivation properties of p65, a reporter gene assay was performed using a Gal4-driven reporter construct and a Gal4–p65 fusion protein. This assay uncouples the transactivating features from the intrinsic DNA binding properties of p65. No differences in TNFα-induced Gal4 promoter activity were observed upon stimulation in control cells and SC75741-treated cells. These results demonstrate that p65-mediated transactivation of gene expression is also not affected by the inhibitor (Fig. 3E). Finally, DNA binding of the p65 subunit to the NF-κB-dependent interleukin 6 (IL-6) promoter was assessed in chromatin immunoprecipitations (Fig. 3F). Upon treatment with SC75741 a reduction of basal and TNFα-induced DNA binding of p65 was observed. Hence, SC75741 specifically executes its NF-κB inhibitory effect on the level of DNA binding of NF-κB factors to gene promoter regions. To confirm this inhibitory potential of SC75741 the TNFα-induced NF-κB promoter activity and the induction of NF-κB-dependent genes on a transcriptional level was further analysed (Fig. 3G). A549 cells transfected with an artificial NF-κB promoter-dependent luciferase reporter gene were treated with TNFα in presence or absence of SC75741 (Fig. 3G). Additionally, TNFα-induced activation of NF-κB-dependent genuine promoter elements of IL-6 and IL-8 genes and the monocyte chemotactic protein 1 (MCP-1) (Fig. 3G) gene were examined. As a result, SC75741 treatment did not only lead to reduced activity of the artificial NF-κB promoter element but also of NF-κB-dependent gene promoters.

Figure 3.

SC75741 inhibits transcriptional activation of NF-κB-dependent genes.

A and B. A549 cells were preincubated with 5 μM SC75741 or solvent for 30 min prior to stimulation with TNFα (20 ng ml−1) in presence of the inhibitor for indicated time periods. Degradation of IκBα and phosphorylation of the NF-κB p65 subunit at Ser468 and Ser536 was detected in Western blot by specific antibodies.

C. Ser536 phosphorylation of p65 was investigated at 4, 6 and 8 h post-H7N7 infection (MOI = 5) in presence of SC75741 (5 μM) or DMSO.

D. A549 cells were preincubated with 5 μM SC75741 or solvent for 30 min and were subsequently stimulated with TNFα (20 ng ml−1) in presence of the inhibitor for 30 min or 120 min or left unstimulated. Nuclear import of p65 was visualized by immunostaining with a p65-specific antibody and an Alexa488-coupled anti-rabbit IgG.

E. For investigation of the transactivational activity of p65 in A549 cells, cells were transfected with 0.1 μg of a 5× Gal4 luciferase reporter plasmid and 0.005 μg of an expression plasmid encoding a Gal4–p65 fusion protein. Twenty-four-hour post-transfection cells were treated with 20 ng ml−1 TNFα in the presence of SC75741 or DMSO for 6 h followed by cell lysis and measurement of luciferase activities. Gal4 promoter activities are given as n-fold increase in RLU relative to untreated and unstimulated control samples. Data represent means ± SD of three independent transfections.

F. Basal and TNFα-induced binding of p65 to the IL-6 promoter region is reduced upon SC75741 treatment. A549 cells were pretreated with 5 μM SC75741 or DMSO for 30 min and thereafter stimulated with TNFα (20 ng ml−1) for 30 min. ChIP analysis was performed as described in Experimental procedures. Briefly, DNA was immunoprecipitated (IP) with p65 antibody, and PCR was performed with primers specific for the IL-6 gene promoter. A serum control was performed to exclude unspecific binding to magnetic beads. Input samples show equal loading.

G. SC75741 inhibits NF-κB-dependent gene expression on promoter level. A549 cells were transfected with NF-κB, IL-6, IL-8 or MCP-1 luciferase reporter constructs. Sixteen-hour post-transfection cells were stimulated with 20 ng ml−1 TNFα for 6 h. Luciferase activities were assayed and are given as n-fold increase in RLU relative to untreated and unstimulated control samples.

Data represent means ± SD of three independent transfections. Statistical significance was assessed by Student's t-test (*P < 0.05, **P < 0.01, ***P < 0.001).

To further examine the functionality of SC75741 in the context of viral infection NF-κB-dependent promoter activity and NF-κB-dependent gene expression was also monitored in infected cells. While the compound only slightly decreased influenza virus-induced IL-6 promoter activity in luciferase assays (Fig. 4A), qRT-PCR experiments revealed that cellular IL-6 mRNA expression (Fig. 4B) was strongly reduced in the presence of SC75741. Surprisingly, virus-induced interferon-β promoter activity was not affected by treatment of cells with SC75741 (Fig. 4C) albeit the IFN-β enhanceosome carries a NF-κB binding site (Hiscott, 2007). Consistent with these findings we could not detect any alterations of virus-induced IFN-β mRNA levels (Fig. 4D). However, it is known that the viral induction of IFN-β is mainly driven by the interferon regulatory factor-3 (IRF-3) and only needs a basal NF-κB activity. Accordingly, virus-induced transcription from an IRF-3 promoter element was also not affected (Fig. 4E). So far, these results indicate a specific inhibitory effect of SC75741 on TNFα and influenza virus-regulated NF-κB-dependent cytokine expression.

Figure 4.

SC75741 reduces virus-induced expression of the NF-κB target genes IL-6 and TRAIL but not IFN-β or IRF-3 activity. Activation of IL-6 and IFN-β expression and induction of IRF-3 driven transcription upon H1N1 infection was investigated in luciferase assays (A, C and E). For reporter gene studies A549 cells were transfected with IL-6, IFN-β or IRF-3 luciferase reporter constructs and infected 16 h post-transfection with influenza A/Puerto-Rico/8/34 (H1N1) (MOI = 5) in presence or absence of 5 μM SC75741 or 0.05% DMSO for 5 h. Luciferase activities were assayed and are given as n-fold increase in RLU relative to untreated and unstimulated control samples. Data represent means ± SD of three independent transfections.

B, D and F. For qRT-PCR RNA of H1N1-infected A549 cells (MOI = 5) treated with 5 μM SC75741, solvent or without treatment was isolated 5 h post-infection. Using the housekeeping gene GAPDH as internal standard the n-fold increase of the IL-6 (B), IFN-β (D) and TRAIL (F) mRNA amount compared with untreated and unstimulated control cells was determined.

Statistical significance was assessed by Student's t-test (**P < 0.01, ***P < 0.001).

The NF-κB signalling pathway is not only involved in regulation of cytokine expression, but also in the regulation of apoptosis, which is generally regarded as an event that limits virus spread. Some years ago, it was shown that in the context of influenza virus replication, early stage apoptosis supports viral replication (Wurzer et al., 2003) in a mechanism that was linked to the NF-κB-dependent expression of the pro-apoptotic factor TNF-related apoptosis-inducing ligand TRAIL (Wurzer et al., 2004). These findings prompted us to investigate whether SC75741 would impact virus-induced mRNA expression of TRAIL (Fig. 4F). Levels of TRAIL mRNA in infected cells were reduced in the presence of SC75741, indicating a regulatory action of SC75741 in a NF-κB-dependent manner. Reduced amounts of TRAIL would imply that there is a lower level of virus-induced apoptosis induction as reported earlier (Wurzer et al., 2003, 2004).

SC75741 prevents efficient nuclear export of vRNPs via inhibition of virus-induced apoptotic functions

To further characterize the impact of SC75741 onto influenza virus-induced apoptotic mechanisms, we monitored the activation of caspases, which are the executioners of the apoptotic process. As shown in Fig. 5A, cleavage of the effector caspase 3 is induced upon H7N7-infection, however, was strongly inhibited by SC75741 treatment (Fig. 5A). This perfectly coincides with an impaired cleavage of poly-(ADP-ribose)-polymerase (PARP), a major caspase substrate (Fig. 5A). Caspase 8 is the major initiator caspase in the receptor-induced pathway while caspase 9 serves as an initiator caspase for mitochondrial apoptosis induction (Thornberry and Lazebnik, 1998). Determining caspase 3, 8 and 9 activity in a luminescence-based activity assay revealed a strong reduction of influenza virus-induced activity of all three caspases upon SC75741 treatment (Fig. 5B). Taken together incubation of infected cells with SC75741 appears to inhibit virus propagation via suppression of NF-κB regulated cell death.

Figure 5.

SC75741 inhibits caspase activation resulting in nuclear retention of viral RNP complexes in the nucleus of infected cells.

A. MDCK cells were infected with the influenza virus strain H7N7 (MOI = 0.5) in the presence of 5 μM SC75741 or 0.05% DMSO and cell lysates prepared at 24 h post-infection were subjected to Western blot analysis. Cleavage of caspase 3 and of the caspase substrate poly (ADP-ribose) polymerase (PARP) was detected using specific antibodies. Equal protein loads on each nitrocellulose membrane were verified by detection of ERK2.

B. Activities of Caspase 3, 8 and 9 in H7N7-infected MDCK cells (MOI = 0.5) were measured 24 h post-infection using a luminescence-based CaspaseGlo® kit (Promega). Luminescence was measured and is given as n-fold increase in RLU relative to untreated and uninfected control samples. Experiments were performed three times in duplicates with similar results. Statistical significance was assessed by Student's t-test (**P < 0.01).

C and E. MDCK or A549 cells were infected with H7N7 virus (MOI = 5), treated with 5 μM SC75741 or 0.05% DMSO and protein lysates were prepared at 4, 6, and 8 h post-infection. Synthesis of the viral proteins NS1, NP and M1 was detected by use of specific antibodies. Detection of ERK2 protein served as a loading control. NS1 and M1 were detected on separate nitrocellulose membranes due to their similar molecular weight.

D and F. MDCK or A549 cells were infected with H7N7 (D: MOI = 10; F: MOI = 20) and 5 μM SC75741 or DMSO was added to the infection media. At indicated times post-infection cells were fixed and viral RNP localization was determined by immunofluorescence staining using a NP-specific goat polyclonal serum and an Alexa Fluor 594 anti-goat IgG. Localization of M1 protein was visualized by staining with an anti-M1 mouse monoclonal antibody and Alexa Fluor 488 chicken anti-mouse secondary antibody. Cell nuclei were stained with DAPI.

Earlier findings clearly indicated that inhibition of caspases resulted in efficient retention of influenza virus RNP complexes in the nucleus of infected cells without affecting accumulation of viral proteins (Wurzer et al., 2003). Interestingly, we obtained similar results upon application of SC75741. Accumulation of the viral non-structural protein 1 (NS1), NP and M1 was not significantly affected in the presence of SC75741 (Fig. 5C and E). Immunofluorescence staining of the viral NP, which is one of the constituents of the RNP complexes reveals an impairment of RNP export in the presence of the compound (Fig. 5D and F). Thus, our results confirmed a virustatic action of SC75741, which blocks the NF-κB-mediated expression of pro-apoptotic factors such as TRAIL, resulting in inhibition of viral RNP export and subsequent block of virus propagation.

SC75741 shows a high barrier for development of resistant virus variants

A most crucial concern for an anti-influenza virus compound is the emergence of resistant virus variants to the drug, an omnipresent challenge due to the high variability of influenza viruses.

To address the question whether SC75741-resistant virus variants may emerge, we performed a well-established multi-passaging experiment in the presence or absence of the drug as described earlier (Ludwig et al., 2004; Mazur et al., 2007). As shown in Fig. 6 A oseltamivir, a licensed inhibitor of the viral surface enzyme neuraminidase, rapidly resulted in the generation of resistant virus variants and had no significant inhibitory effect on virus propagation after eight passages in the presence of the drug. While the initial inhibition of virus propagation by SC75741 was less efficient than in oseltamivir-treated cells, in the presence of the NF-κB inhibitor no increase in titres was observed even after eight passages, leaving a completely drug-sensitive virus population. Direct comparison of the input H7N7 virus (wt) with the variant obtained after eight passages of oseltamivir drug pressure [H7N7 (mut)] showed full SC75741 sensitivity, while H7N7 (mut) was phenotypically resistant to oseltamivir (Fig. 6B). Comparative sequence analysis of both viruses indicated a mixed genotype in H7N7 (mut) samples. While the H7N7 (wt) possesses a codon AGG for Arginine at amino-acid position 371, the influenza A virus (mut) shows a mixed phenotype of the original codon AGG with a codon AAG for a Lysine. This mutation was already referred to as conferring neuraminidase resistance (Ferraris and Lina, 2008). Additionally, a codon change from TCT to TTT was observed at amino-acid position 370, resulting in an amino-acid replacement of Serine to Phenylalanine, probably also contributing to the resistance to neuraminidase treatment. The commonly occurring H274Y mutation (Ferraris and Lina, 2008) was not detected in the H7N7 (mut). Based on these results we conclude that SC75741 possesses a high barrier for developing resistance to the inhibitor while this rapidly occurs in oseltamivir-treated cells.

Figure 6.

SC75741 shows a high barrier for development of resistant virus variants.

A. A549 cells were infected with H7N7 (MOI = 0.001) in the presence of either 5 μM SC75741, 2 μM oseltamivir or the analogous volume of DMSO. Twenty-four-hour post-infection infectious virus titres of the supernatants were determined in plaque assay. Supernatants were used to infect fresh A549 cells (MOI = 0.001), again in the presence of SC75741, oseltamivir or DMSO. This procedure was repeated for eight passages. Virus titres are shown in percentage to the accordant DMSO-treated control as described previously (Ehrhardt et al., 2007).

B. Sensitivity of the oseltamivir-resistant H7N7 virus resulting from repeated passaging to SC75741 and oseltamivir was compared with the original H7N7 virus. Cells were infected with an H7N7 virus stock or supernatants isolated from oseltamivir-treated cells after virus passage eight (MOI = 0.001) for 24 h. Infection media were supplemented with either 5 μM SC75741 or 2 μM oseltamivir and progeny virus titres were estimated in plaque assay.

SC75741 reduces virus replication and cytokine expression in vivo

To assess whether the antiviral properties of SC75741 observed in cell culture would also be relevant in vivo, C57BL/6 mice were infected with a lethal dose of highly pathogenic avian influenza A/mallard/Bavaria/1/2006 (H5N1). SC75741 was administered intraperitoneal prior to infection in a concentration of 15 mg kg−1. Mice in the control group received placebo prior to virus infection. Forty-eight hours later the animals were killed, the lungs were taken and accumulation of virus-specific mRNA was detected as an equivalent to virus propagation. As shown in Fig. 7A, the amount of viral mRNA was reduced by 90% in SC75741-treated mice compared with placebo-treated controls, indicating that SC75741 leads to a reduced propagation of the H5N1 virus in the lungs of infected mice (Fig. 7A).

Figure 7.

SC75741 treatment influences cytokine expression in mice lungs after H5N1 influenza virus infection. C57BL/6 mice were treated with 15 mg kg−1 SC75741 and thereafter infected with 5 × 103 pfu H5N1 influenza virus. Mice of the control group received placebo prior to virus infection. Two-day post-infection mice were killed and RNA was isolated from the lung and analysed for expression of viral mRNA (A), IL-6 (B) or IP-10 mRNA (C) by qRT-PCR using the housekeeping gene GAPDH as internal standard. The bars represent the comparative delta CT value of SC75741-treated mice and untreated control mice which were set as 100%. Shown are mRNA levels of three infected mice according to Boeuf et al. (2005). This experiment was performed twice with similar results.

One of the hallmarks of H5N1 infections is a strong induction of cytokines in the lung, known as cytokine storm that contributes to the pathogenicity of these viruses. Since many of these cytokines are regulated by NF-κB we investigated whether treatment with the NF-κB inhibitor SC75741 would have an effect on the expression of NF-κB-dependent candidate genes IL-6 and IP-10. These two cytokines are known to be strongly upregulated already 48 h p.i. in the lungs of H5N1 infected mice. As shown in Fig. 7B, the amount of IL-6-specific mRNA was drastically reduced in SC75741-treated mice compared with solvent-treated controls.

Moreover, also the expression of IP-10 was altered in SC75741-treated H5N1 influenza virus-infected mice. Here, treatment resulted in roughly 90% reduction of mRNA (Fig. 7C). Thus, in addition to an antiviral effect against the virus, SC75741 treatment also reduced cytokine production in vivo that is involved in influenza virus infection-mediated hypercytokinemia.

Discussion

Infections with influenza viruses are still a major threat for human beings and control and treatment of the disease is limited. Appearance of highly pathogenic avian strains that infected and killed humans and the emergence of the 2009 pandemic H1N1 influenza point to the urgent need for efficient drugs. In consideration of the fact that our arsenal of licensed anti-influenza drugs is quite limited and taken into account that influenza viruses tend to rapid emergence of drug resistance to compounds that target viral factors, the necessity for alternative treatment strategies is obvious.

Earlier studies identified an unexpected dependence of influenza viruses on the NF-κB signalling pathway (Nimmerjahn et al., 2004; Wurzer et al., 2004). Based on these findings the pathway was discussed as potential target for antiviral intervention, although targeting of a cellular factor may raise concerns about unwanted side-effects. Inhibition of the NF-κB signalling pathway via inhibition of IKK2 can be achieved by ASA, also known as aspirin, an agent that is in frequent clinical use and considered to be very save. Indeed, ASA was shown to be an efficient inhibitor of influenza virus infections (Huang and Dietsch, 1988; Mazur et al., 2007) without causing any harmful side-effects to host cells and without any tendency to induce resistant virus variants (Mazur et al., 2007). Regarding the NF-κB-dependent mechanism of action it was shown that ASA leads to decreased expression of TRAIL and FasL, reduced caspase activity and retention of viral ribonucleoprotein complexes (RNPs) in the nucleus (Mazur et al., 2007). Another specific inhibitor of NF-κB signalling, BAY11-7085, was also shown to block influenza virus replication efficiently (Nimmerjahn et al., 2004). Similar effects were observed upon incubation of influenza virus-infected cells with the NF-κB inhibitor SC75741 indicating overlapping mechanisms of action. Efficient inhibition of virus replication was not only demonstrated for various influenza A virus strains including H5N1, H7N7 and H1N1v viruses but also for influenza B viruses indicating a broad antiviral potency. Although an effect of SC75741 on long-term cell proliferation was observed, the inhibitor was not cytotoxic and specifically decreased influenza virus-induced expression of NF-κB-dependent genes, such as IL-6, IL-8, and MCP-1 at early time points of infection. Interestingly, there was no influence on the induction of the interferon regulatory factor or the expression of IFN-β. While a decreased expression of IFN-β was previously demonstrated in cells lacking the NF-κB factors p50 and p65 (Wei et al., 2006), partial inhibition by SC75741 does not seem to be sufficient to block IFN-β. A likely explanation for this phenomenon is, that only a basal NF-κB activity is required for IFN-β expression, since the major inducing factor for the IFN-β enhanceosome is IRF-3. This basal NF-κB activity is still achieved by partial inhibition with the inhibitor but not if NF-κB signalling is completely blunted by gene knockout. However, in the context of virus infection this feature of SC75741 is a great advantage, since it leaves the type I IFN arm of the innate immune response intact.

This phenomenon may further help to manage the detrimental role of cytokines in the clinics. Although other molecules, such as statins have been reported to cope with sepsis and community acquired pneumonia (Falagas et al., 2008; Kopterides and Falagas, 2009), inhibiting effects on influenza virus replication and the overall pathogenicity of influenza virus infection have been discussed controversial, so far. While some of the studies reported reduced lung damage and inhibited viral replication (Liu et al., 2009) or reduced mortality in patients hospitalized with influenza (Vandermeer et al., 2012), in a very recent study the improved outcome in patients with sepsis and pneumonia treated with statins could not be correlated with effects on influenza virus infection (Radigan et al., 2012). Thus, regarding its antiviral and anti-inflammatory action, SC75741 exhibits great advantages over current available compounds. Another cellular factor that was controlled by the compound upon influenza virus infection is the pro-apoptotic factor TRAIL. Subsequently virus-induced PARP cleavage mediated by receptor- and mitochondria-induced caspase activation was reduced by SC75741. In accordance with a major role of pro-apoptotic ligands such as TRAIL we demonstrated an inhibitory effect of SC75741 on the receptor-mediated apoptosis-signalling pathway. Although an increased morbidity and mortality of influenza virus-infected TRAIL−/− mice has been reported (Brincks et al., 2011), the reversible and partial inhibition of TRAIL by SC75741 seems to be beneficial during the antiviral treatment of influenza virus infection. Furthermore, another study shows that inhibition of TRAIL signalling in exudate macrophages contributes to reduced apoptosis in alveolar epithelial cells, attenuated lung leakage and increased survival upon influenza A virus infection, demonstrating that reduced TRAIL activity plays a key role in reduced mortality in influenza virus pneumonia (Herold et al., 2008).

Since viral protein synthesis was not affected by SC75741, we can conclude that the compound does not interfere with early steps of the replication cycle. This observation also provides additional evidence that the drug is not toxic in an unspecific manner, otherwise all the steps of viral replication up to the synthesis of progeny viral proteins should not proceed unaffected.

Immunofluorescence studies revealed reduced export of the viral RNP complexes in the presence of the drug. With regard to the mechanism it has been shown that caspases promote diffusion of larger protein complexes due to an increase of the diffusion limit of nuclear pores (Faleiro and Lazebnik, 2000). We already proposed earlier, that by such a mechanism RNPs may be exported by passive diffusion (Wurzer et al., 2003). Data of a more recent collaborative study using atomic-force microscopy supported this assumption by demonstrating that caspase action leads to a very ordered degradation of the nuclear pore complexes (Kramer et al., 2008). This allowed diffusion of particles of the size of RNP complexes through the widened pores without disrupting general barrier function of the nuclear membrane (Kramer et al., 2008).

These results confirmed that SC75741 is a promising antiviral lead compound able to inhibit virus replication without any harming cellular side-effects, exhibiting a high barrier to cause resistant virus variants. As stated above, sequence analysis of the neuraminidase verified a R371K mutation within the H7N7 influenza A virus (mut), an amino-acid exchange already referred within context of resistance to N2 influenza A viruses (Ferraris and Lina, 2008). Additionally, an amino-acid exchange S370F was observed, probably also contributing to the resistance to neuraminidases in H7N7 influenza A virus (mut). Anyhow, a H274Y mutation of N1 influenza viruses often mentioned in context of resistance to neuraminidase inhibitors (Ferraris and Lina, 2008), was not detected in H7N7 influenza A viruses.

Additionally, our in vitro data have been confirmed and expanded to the animal model, demonstrating the potent antiviral activity of SC75741 in infected mice, without exhibition of harming effects.

Furthermore, reduction of viral replication, by activation of immune-receptors, such as PAR2 was observed to be attended by enhancement of IFNγ-production and raised IL-10 secretion in vitro and in vivo (Feld et al., 2008; Khoufache et al., 2009, 2012). Although viral replication of laboratory influenza virus strains, such as A/FPV/Bratislava/79 or A/Puerto Rico/8/34 was reduced, negative effects on infection with influenza viruses causing a cytokine storm, such as H5N1 virus strains, can not be ruled out. Interestingly, SC75741 seems to avoid such negative side-effects, since SC75741 treatment of infected mice results in reduction of H5N1-induced IP-10 mRNA production in lung (Fig. 7).

Because of the urgent need for new and amply available anti-influenza agents we conclude that SC75741 is a very promising lead compound for further efforts to develop novel antiviral drugs.

Experimental procedures

Cell lines, viruses and viral infections

A549 human lung carcinoma cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and antibiotics. MDCK cells were grown in minimal-essential medium (MEM) supplemented with 10% FBS and antibiotics. Avian influenza virus A/FPV/Bratislava/79 (H7N7) and the human prototype strain A/Puerto-Rico/8/34 (H1N1) were taken from the virus strain collection of the Institute of Virology, Giessen. The human H5N1 strain A/Thailand/1(KAN-1)/2004 (H5N1) was isolated at the Siriraj Hospital, Mahidol University, Bangkok, Thailand. Tamiflu-resistant swine-origin influenza A virus A/Nordrhein-Westfalen/173/09 (H1N1v), containing (HA D222G; NA H274Y) was isolated at the institute of Medical Virology, Münster, Germany (Seyer et al., 2012). Influenza B virus B/Maryland/59 (Flu B) was a kind of gift of T. Wolff, Robert-Koch-Institute, Berlin, Germany. Avian influenza A/mallard/Bavaria/1/2006 (H5N1) was originally obtained from the Bavarian Health and Food Safety Authority, Oberschleissheim, Germany and was further propagated in embryonated chicken eggs at the Friedrich-Loeffler-Institut, Federal Research Institute for Animal Health, Tuebingen, Germany.

For viral infection cells were washed with phosphate buffered saline supplemented with 0.2% BSA, 1 mM MgCl2, 0.9 mM CaCl2, 100 U ml−1 penicillin and 0.1 mg ml−1 streptomycin (PBS/BA) and subsequently incubated with virus at the indicated multiplicities of infection (MOI) diluted in PBS/BA for 30 min at 37°C. After the 30 min incubation period the virus dilution was aspirated, cells were washed once with PBS/BA and incubated with either DMEM or MEM containing 0.2% BSA, 1 mM MgCl2, 0.9 mM CaCl2, 100 U ml−1 penicillin and 0.1 mg ml−1 streptomycin, supplemented with inhibitor or solvent, as indicated. For multicycle infections with H1N1 influenza A virus and influenza B virus media were supplemented with 2 μg ml−1 trypsin. At given time points supernatants were collected and titrations of progeny virus in the supernatants were essentially performed as described earlier (Pleschka et al., 2001; Ehrhardt et al., 2004).

Viral infections of mice

Inbred female C57BL/6 mice at the age of 6–8 weeks were obtained from the animal breeding facilities at the Friedrich-Loeffler-Institute, Federal Research Institute for Animal Health, Tuebingen, Germany. For infection, the animals were anaesthetized by intraperitoneal injection of 200 μl ketamine/rompun. Equal amounts of a 2% rompun (Bayer) and a 10% ketamine (Sanofi) stock solution were mixed at a ratio of 1:10 with PBS. Mice were infected intranasally with adequate virus doses diluted in 50 μl BSS (sterile physiological balanced salt solution) by inoculating 25 μl into each nostril. After infection with H5N1 virus the mice were kept in individually ventilated cages (Techniplast). All animal studies were approved by the Institutional Animal Care and Use Committee of Tuebingen. SC75741 was dissolved in 10% dimethylsulfoxide (DMSO), 30% Cremophor EL (Merck) and 60% PBS. SC75741 in a volume of 200 μl was applied to the mice by intraperitoneal (i.p.) injection. Control animals were treated formulation only, containing no compound (placebo).

Inhibitors and reagents

SC75741 N-(6-benzoyl-1H-benzo[d]imidazol-2-yl)-2-(1-(thieno[3,2-d]pyrimidin-4-yl)piperidin-4-yl)thiazole-4-carboxamide (MW 565) and oseltamivir were provided by the 4SC AG (Planegg-Martinsried, Germany) and dissolved in DMSO at a 10 mM and 4 mM stock concentration respectively. TNFα was obtained from Sigma-Aldrich and was added to the cell media in a concentration of 20 ng ml−1.

Plasmids, transient transfections and reporter gene assays

The 3× NF-κB reporter plasmid was described previously (Flory et al., 2000). The 4× IRF-3 construct contains four copies of the IRF-3 binding PRDI/III motif of the IFN-β promoter in front of a luciferase reporter gene (Ehrhardt et al., 2004). The IFN-β promoter construct was a kind of gift of J. Hiscott, Lady Davis Institute for Medical Research, McGill University, Montreal, Canada. The IL-6 promoter luciferase construct was obtained from LMBO Plasmid Collection (university of Gent/Belgium). IL-8 and MCP-1 luciferase constructs were described previously (Goebeler et al., 1999; Kunz et al., 2000). A549 cells were transfected with Lipofectamine 2000 (Invitrogen) according to the protocol described in Basler et al. (2000). Sixteen hours later cells were either infected with PR8 (MOI = 5) or treated with TNFα (20 ng ml−1) for 5 or 6 h. Luciferase-reporter gene assays were carried out in triplicates as described earlier (Ludwig et al., 2001). For Gal4 promoter studies cells were cotransfected with a 4× Gal4 luciferase reporter plasmid and a p65–Gal4 fusion protein encoding expression plasmid prior to TNFα stimulation for 6 h.

Measurement of caspase activity

Activities of caspase 3/7, caspase 8 and caspase 9 were measured using commercial available Caspase-Glo kits (Promega), according to the manufacturer's protocol. Influenza A/FPV/Bratislava/79 (H7N7) infected cells were incubated for 30 min with reaction buffer (Promega) and luminescence was measured using the MicroLumatPlus LB 96 V luminometer (Berthold Technologies). For each sample two biological replicates were analysed.

Chromatin immunoprecipitation

To examine the association of the NF-κB transcription factor p65 to the IL-6 promoter, A549 cells (3 × 106 per 6 cm dish) were preincubated with 5 μM SC75741 or 0.05% DMSO for 1 h and subsequently stimulated with TNFα (20 ng ml−1) for 30 min, and ChIP assay were performed using a Magna ChIP assay kit (Millipore), according to the manufacturer's specification. Briefly, proteins/DNA were cross-linked with 1% formaldehyde for 10 min at RT, the cross-linking was blocked with 0.125 M glycine. Nuclear lysates were prepared and chromatin was sheared by sonication (10 times for 10 s each time). Immunoprecipitation was carried out overnight at 4°C using 2 μg of NF-κB p65 subunit specific antibody (Santa Cruz Biotechnology) and protein A magnetic beads. Beads were washed once with low salt wash buffer, high salt wash puffer, LiCl washing buffer and TE with rotation of samples for 5 min in between every washing step. DNA cross-links were reversed and deproteinated with proteinase K at 62°C for 2 h and DNA was purified for PCR reaction. DNA isolated from an aliquot of the total nuclear extract was used as a loading control for the PCR (input control). To detect the presence of IL-6 promoter in the processed sample, PCR reactions were carried out with 5 μl of total DNA and immunoprecipitated DNA using forward primer: 5′-CCACCCTCACCCTCCAA-3′; reverse primer: 5′-GGCTAAGGATTTCCTGCA-3′, which are specific for the human IL-6 promoter (a.n.: AF048692). The cycling parameters for PCR were: denaturation 94°C for 30 s, annealing 55°C for 30 s, and extension 72°C for 30 s (repeated 40 times). The PCR products were size-separated on a 2% agarose gel and visualized via ethidium bromide staining.

Western blot

For Western blots cells were lysed on ice with RIPA lysis buffer [1% (v/v) NP-40, 0.5% (v/v) DOC, 1% (w/v) SDS, 150 mM NaCl, 50 mM Tris pH 8, 90% H2O dest., 200 μM Pefablock, 5 μg ml−1 Aprotinin, 5 μg ml−1 Leupeptin, 1 mM Natrium-Vanadat, 5 mM Benzamidin] for 30 min. Cell lysates were cleared by centrifugation and protein yields were estimated employing a protein dye reagent (Bio-Rad Laboratories). Equal amounts of protein were separated by SDS-polyacrylamide gel electrophoresis and subsequently blotted on nitrocellulose membranes. Anti-PARP monoclonal antibody was purchased from BD Transduction Laboratories. Antisera against the influenza virus proteins NP and M1 were obtained from Serotec. An anti-NS1 antibody was provided by T. Wolff (Robert-Koch Institute, Berlin, Germany). Full-length caspase 3 and cleaved caspase 3 (Asp175) antibodies were purchased from Cell signalling technology. Detection of IκBα and the loading control ERK2 was performed with anti-IκBα and ERK2 antisera (Santa Cruz Biotechnology) respectively. Protein bands were visualized in a standard enhanced chemiluminescence reaction.

Flow cytometry analysis

A549 cells were treated with SC75741, the analogous volume of DMSO or left untreated for indicated time periods. Cells were washed with PBS, trypsinized, washed again and resuspended in PBS containing PI in a concentration of 50 μg ml−1. After incubation for 1 h at RT PI was removed by washing cells with PBS and fluorescence was determined in the FL2-channel (585 nm) using a FACScalibur cytometer (Becton Dickinson).

MTT cell proliferation assay

The MTT [3-(4,5-dimethylthaizol-2-yl)-2,5-diphenyltetrazoliumbromide] assay is based on an enzymatic reaction of the mitochondrial succinic dehydrogenase. In viable and proliferating cells these enzyme cleaves the tetrazolium rings of the pale yellow MTT which results in formation of dark blue formazan that is largely impermeable to cell membranes and therefore accumulates in healthy cells. The amount of formed formazan can be measured in a colorimetric assay at OD 562 nm and is directly proportional to the number of viable cells. A549 cells were treated with SC75741 or solvent for different time periods. Afterwards cells were washed with PBS and incubated with MTT (0.5 mg ml−1) for 1–3 h at 37°C until crystallization of formazan was visible. The MTT solution was removed and isopropanol with 0.04 N HCl was added to the cells to dissolve the blue formazan. Absorbance was measured at 570 nm.

Indirect immunofluorescence microscopy

A549 or MDCK cells were seeded onto 12 mm glass plates. Twenty-four hours later cells were infected with A/FPV/Bratislava/79 (H7N7) in the presence of either 5 μM SC75741, DMSO or cells were left untreated as described before. Four- and five/six-hour post-infection cells were washed twice with PBS and fixed for 20 min with 3.7% formaldehyde at RT. After washing with PBS, cells were permeabilized with 0.2% Triton-X, washed again and blocked with 1% BA in PBS for 30 min at RT. Cells were incubated with a 1:200 dilution of a mouse antibody against the viral M1 (Serotec) and a 1:1200 dilution of a goat anti-NP serum for 1 h. After further washes, cells were incubated with a 1:300 dilution of an Alexa Fluor 488 chicken anti-mouse IgG and a 1:400 dilution of an Alexa Fluor 594 chicken anti-goat IgG (Invitrogen) and 30 nM DAPI (4′,6-diamidino-2-phenylindol) for 30 min. For visualization of p65 localization cells were stained with anti-p65 monoclonal antibody (Upstate) and Alexa Fluor 488 chicken anti-rabbit IgG. Finally, cells were washed and mounted with fluorescence mounting medium (Dako). Fluorescence was visualized using a Zeiss Axiovert 135 fluorescence microscope.

Quantitative real-time PCR

For quantification of mRNA expression of the IL-6, IFN-β and TRAIL genes in response to virus infection A549 cells were infected with A/Puerto-Rico/8/34 (H1N1) (MOI = 5). For each sample three biological replicates were performed. After 5 h of incubation at 37°C total RNA was isolated with the RNeasy® Mini Kit (Qiagen) following the manufacturers protocol. 2.5 μg RNA was reverse transcribed using RevertAid H Minus M-MulV reverse transcriptase (Fermentas) and oligo-dT primers at 42°C for 1 h. Quantitative real-time PCR was performed using Brilliant®-SYBR Green QPCR-Mastermix (Agilent Technologies) using following primers: GAPDH sense 5′-GCAAATTTCCATGGCACCGT-3′ antisense 5′-GCCCCACTTGATTTTGGAGG-3′, TRAIL sense 5′-GTCTCTCTGTGTGGCTGTAACTTACG-3′ antisense 5′-AAACAAGCAATGCCACTTTTGG-3′, IL-6 sense 5′-AGAGGCACTGGCAGAAAACAAC-3′ antisense 5′-AGGCAAGTCTCCTCATTGAATCC-3′, IFN-β sense 5′-GGCCATGACCAACAATGTTCTCCTCC-3′ antisense 5′-GCGCTCAGTTTCGGAGGTAACCTGT-3′. The mRNA levels were determined using the 2−ΔΔCT method (Livak and Schmittgen, 2001) and normalized to GAPDH internal control.

For RNA preparations from infected mice, lungs were homogenized and incubated overnight in 1 ml TriZol® Reagent (Invitrogen) at 4°C. Total RNA isolation was performed as specified by the manufacturer (Invitrogen). RNA was solubilized in 50 μl RNase free water and diluted to a working concentration of 50 ng RNA μl−1. Reverse transcription real time PCR was performed using QuantiFast SYBR® Green RT-PCR Kit and QuantiTect Primer Assays (Qiagen). All samples were normalized to GAPDH and fold expression analysed relative to uninfected controls (Boeuf et al., 2005). Ct values were obtained with the SmartCycler® (Cepheid).

Sequencing of the NA gene of the H7N7 subtype influenza virus

Supernatants of infected MDCK cells (3 ml) were filtered (0.2 μm, Whatman) and virus was pelleted by ultracentrifugation (35.000 r.p.m., for 2 h at 4°C, SW41 Rotor, Beckman). Viral RNA (vRNA) from virus pellets was isolated using the RNeasy Mini Kit from Qiagen. To synthesize cDNA, vRNA was reverse transcribed using 0.2 μg random hexamer primer and 200 U RevertAidH Minus M-MuLV Reverse Transcriptase (Fermentas) according to manufacturer's instructions. The full-length cDNAs of the NA gene were amplified and then sequenced using NA subtype specific primers as described by (Hoffmann et al., 2001). The NA gene sequence data of A/FPV/Bratislava/79 (H7N7) virus sequenced are available upon request.

Acknowledgements

This work was supported by the fund ‘Innovative Medical Research (IMF EH12003)’ of the University of Muenster medical school, several grants of the Deutsche Forschungsgemeinschaft (DFG) and the FluResearchNet, a nationwide research network on zoonotic influenza sponsored by the German ministry of education and research (BMBF).

Furthermore, this work is part of the activities of the EUROFLU Consortium and the VIRGIL European Network of Excellence on Antiviral Drug Resistance supported by grants from the Priority 1 ‘Life Sciences, Genomics and Biotechnology for Health’ programme in the 6th Framework Program of the EU.

Ancillary