The IκB Kinase Is a Key Factor in Triggering Influenza A Virus-induced Inflammatory Cytokine Production in Airway Epithelial Cells*

Influenza A viruses continue to represent a severe threat worldwide, causing large epidemics and pandemics responsible for thousands of deaths every year. Excessive inflammation due to overabundant production of proinflammatory cytokines by airway epithelial cells is considered an important factor in disease pathogenesis. Here we report that influenza A virus induced IκB kinase (IKK) activity in human airway epithelial A549 cells, resulting in persistent activation of nuclear factor-κB (NF-κB), a critical regulator of the inflammatory response. Although lung epithelial cells are highly sensitive to stimulation of the IKK/NF-κB pathway by influenza virus infection, NF-κB was not activated in several non-pulmonary cells permissive to the virus, indicating a cell-specific response. Moreover, NF-κB was not essential for virus replication but triggered the expression of proinflammatory cytokines in infected lung cells and was directly responsible for production of high levels of interleukin-8, a chemokine associated with influenza-induced inflammation and airway pathology. We also report that 9-deoxy-Δ9,Δ12-13,14-dihydro-prostaglandin D2, a cyclopentenone prostanoid with therapeutic efficacy against influenza in preclinical studies, was a powerful inhibitor of influenza virus-induced IKK activity and interleukin-8 production by human pulmonary cells. The results identify IKK as an important factor in triggering influenza virus-induced inflammatory reactions in pulmonary epithelium, suggesting novel therapeutic approaches in the treatment of influenza.


Influenza A viruses continue to represent a severe threat worldwide, causing large epidemics and pandemics responsible for thousands of deaths every year. Excessive inflammation due to overabundant production of proinflammatory cytokines by airway epithelial cells is considered an important factor in disease pathogenesis. Here we report that influenza A virus induced IB kinase (IKK) activity in human airway epithelial A549 cells, resulting in persistent activation of nuclear factor-B (NF-B), a critical regulator of the inflammatory response. Although lung epithelial cells are highly sensitive to stimulation of the IKK/NF-B pathway by influenza virus infection, NF-B was not activated in several non-pulmonary cells permissive to the virus, indicating a cell-specific response. Moreover, NF-B was not essential for virus replication but triggered the expression of proinflammatory cytokines in infected lung cells and
was directly responsible for production of high levels of interleukin-8, a chemokine associated with influenzainduced inflammation and airway pathology. We also report that 9-deoxy-⌬ 9 ,⌬ 12 -13,14-dihydro-prostaglandin D 2 , a cyclopentenone prostanoid with therapeutic efficacy against influenza in preclinical studies, was a powerful inhibitor of influenza virus-induced IKK activity and interleukin-8 production by human pulmonary cells. The results identify IKK as an important factor in triggering influenza virus-induced inflammatory reactions in pulmonary epithelium, suggesting novel therapeutic approaches in the treatment of influenza.
Influenza, a highly contagious acute respiratory illness that affects all age groups, is responsible for an average of 36,000 deaths and 114,000 hospitalizations per year in the United States alone (1). The etiological agents of the disease, the influenza viruses or orthomyxoviruses, are enveloped, negative-stranded RNA viruses classified in three types (A, B, and C), of which the A type is clinically the most important. In recent history, highly pathogenic strains of influenza A virus with elevated case fatality rate have suddenly emerged (2). Influenza A virus replicates throughout the respiratory tract, where the viral antigen is found predominantly in the epithe-lial cells. The clinical responses range from mild disease to fatal viral pneumonia. Although the mechanisms underlying the expression of symptoms and the development of secondary complications that may result in respiratory failure are still not well understood, excessive inflammation due to overabundant production of proinflammatory cytokines and lung inflammatory infiltrates is considered an important factor in disease pathogenesis (2)(3)(4). Intense infiltration of inflammatory cells in the lung epithelial layer has been associated with the release of chemotactic cytokines, in particular, interleukin (IL) 1 -8, by virus-infected airway epithelial cells (5,6). Interestingly, the expression of most cytokines induced by influenza A virus is regulated by the cellular nuclear factor-B (NF-B).
NF-B is a critical regulator of the immediate early pathogen response, playing an important role in promoting inflammation and viral gene expression (7,8). NF-B normally exists as an inactive cytoplasmic complex bound to inhibitory proteins of the IB family and is induced in response to a variety of pathogenic stimuli, including proinflammatory cytokines and viral infection. In the case of proinflammatory cytokines, induction requires the activation of the IB kinase (IKK) complex containing two catalytic subunits (IKK␣ and IKK␤) and the IKK␥ regulatory subunit, which phosphorylates IBs, triggering their ubiquitination and proteasome-mediated degradation (8). Release of IBs results in nuclear translocation of NF-B and its binding to DNA at specific B sites, rapidly inducing a variety of genes encoding, among others, cell adhesion molecules, inflammatory and chemotactic cytokines, cytokine receptors, and enzymes that produce inflammatory mediators (9).
It is well known that infection with different DNA and RNA viruses, including orthomyxoviruses, can trigger NF-B activation via different signaling mechanisms (7). In the case of influenza virus infection, however, little is known regarding both the signaling pathway utilized by the virus and the role of the nuclear factor in the infection process.
In the present report, we show that influenza A virus induces a persistent (up to 48 h) activation of NF-B in human airway epithelial A549 cells. Influenza A virus mimics proinflammatory cytokines inducing NF-B by activation of the IB kinase, followed by IB␣ and IB␤ degradation. Interestingly, although lung cells are highly sensitive to influenza-induced NF-B activation, NF-B induction is not a general response of the infected cell to influenza virus and was not detected in a variety of non-pulmonary cells permissive to the virus. Furthermore, IKK-mediated NF-B activity is not essential for virus replication but triggers the expression of several proinflammatory cytokines and is directly responsible for inducing IL-8 production by infected lung cells.
Finally, we have previously shown that cyclopentenone prostaglandins (PGs) of the A and J type block TNF␣-induced NF-B activation by direct inhibition and modification of the IKK␤ subunit (10). We now show that 9-deoxy-⌬ 9 ,⌬ 12 -13,14dihydro-PGD 2 (⌬ 12 -PGJ 2 ), a natural cyclopentenone metabolite of prostaglandin D 2 that is physiologically present in human body fluids (11) and possesses therapeutic efficacy in a mouse model of lethal influenza A virus infection (12), is a powerful inhibitor of influenza A virus-induced IKK and NF-B activities and that this effect results in the block of virus-induced IL-8 production by human pulmonary cells.
Influenza Virus Preparation, Infection, and Titration-Influenza virus A/WSN/33 (H1N1) (WSN) was grown in the allantoic cavity of 8-day-old embryonated eggs. After 48 h at 37°C, the allantoic fluid was harvested and centrifuged at 5000 rpm for 30 min to remove cellular debris, and virus titers were determined by hemagglutinin titration and plaque assay, according to standard procedures (12). One hemoagglutinating unit (HAU) corresponded to 10 6 plaque-forming units in this model. Confluent cell monolayers were infected with WSN for 1 h at 37°C at a multiplicity of infection of 5 HAU/10 5 cells. After the adsorption period, the viral inoculum was removed, and cell monolayers were washed three times with phosphate-buffered saline. Cells were maintained at 37°C in culture medium containing 2% fetal calf serum. Virus yield was determined at different times post-infection by hemagglutinin titration, according to standard procedures (12).
Western Blot Analysis and Enzyme-linked Immunosorbent Assay (ELISA)-Whole-cell extracts were prepared from mock-infected or WSN-infected cells as described previously (16). Equal amounts of protein (25 g) from whole-cell extracts were separated by SDS-PAGE and blotted to nitrocellulose, and filters were incubated with polyclonal anti-IB␤ antibody (Biomol, Plymouth Meeting, PA) or monoclonal anti-IB␣ antibody (Imgenex, San Diego, CA), anti-␤-actin antibody (Santa Cruz Biotechnology, Santa Cruz, CA) or anti-IKK␣ antibody (BD Biosciences-PharMingen, San Jose, CA), followed by decoration with peroxidase-labeled anti-mouse or anti-rabbit IgG antibodies (ECL; Amersham Biosciences) (15). Immunoreactive IL-8 was quantitated in cell supernatants by ELISA (17), using a double antibody and standards obtained from R&D Systems (Minneapolis, MN).
Kinase Assay-Cell lysates were immunoprecipitated with anti-IKK␣ antibodies (Cell Signaling Technology, Beverly, MA) in the presence of 15 l of protein A-Sepharose (Sigma-Aldrich) at 4°C for 12 h. After extensive washing, endogenous IKK activity was determined using glutathione S-transferase-IB␣ (1-54) as a substrate (10). Western blot analysis for IKK␣ was performed as kinase loading control. For determination of in vitro IKK activity, A549 cells were stimulated with TNF␣ for 30 min. Endogenous IKK was immunoprecipitated with anti-IKK␣ antibody, and kinase activity was determined in the presence of different concentrations of ⌬ 12 -PGJ 2 or PGF 2␣ .
Statistical Analysis-Statistical analyses were performed using Student's t test for unpaired data. Data are expressed as the mean Ϯ S.D.; p values of Ͻ0.05 were considered significant.

Influenza A Virus Infection Induces NF-B Activity in Human Airway Epithelial Cells via Stimulation of the IB Ki-
nase-As reported for several human viral pathogens, orthomyxoviruses are also able to activate NF-B in infected cells (7, 18 -19); however, the molecular mechanisms involved in NF-B activation by influenza virus have not yet been elucidated. To investigate the effect of influenza A virus on NF-B activity in pulmonary epithelial cells, confluent monolayers of human alveolar epithelium-like A549 cells were mock-infected or infected with WSN (5 HAU/10 5 cells) for 1 h at 37°C. At different times p.i., whole-cell extracts were analyzed for NF-B activation by EMSA, and the levels of NF-B DNA binding activity were quantitated by PhosphorImager analysis. As shown in Fig. 1A (top panel), WSN infection induced NF-B activation in A549 cells starting at 12 h p.i. and reaching maximal levels (Ͼ4-fold above control) at 24 h p.i.
To investigate the molecular pathway involved in NF-B activation by influenza virus, the levels of the NF-B inhibitory proteins IB␣ and IB␤ were determined in uninfected and WSN-infected A549 cells by immunoblot analysis. As shown in Fig. 1A (bottom panels), WSN infection caused IB␣ degradation starting at 12 h p.i. The inhibitory protein appeared to be completely degraded at 16 h p.i. A similar pattern of proteolysis was observed for the IB␤ isoform. To determine whether WSN infection induced IKK activity in human pulmonary cells, extracts from mock-infected controls or A549 cells infected for 9, 12, 16, and 24 h were processed for kinase assay and immunoblot analysis to determine IKK activity and recovery, respectively. The levels of phosphorylated IB␣ were quantified by PhosphorImager analysis. As shown in Fig. 1B, influenza virus infection induces IKK in A549 cells with a pick of maximal activity (Ͼ3-fold above control) at 12 h p.i.
Several viral pathogens, including HIV and herpes simplex viruses, utilize NF-B to enhance viral transcription and replication (15,20). Recently, based on the fact that influenza virus propagation was slightly impaired in cell lines that stably express a dominant-negative mutant of IKK␤, it was suggested that NF-B signaling could be important for efficient influenza virus production (21). To investigate the possibility that influ-enza virus replication could be dependent on its ability to activate NF-B in the infected cell, confluent monolayers of human lung A549 cells, kidney 293 cells, neuroblastoma SH-SY5Y cells, and HaCaT keratinocytes were mock-infected or infected with WSN virus (5 HAU/10 5 cells) for 1 h at 37°C. In parallel, confluent monolayers of canine kidney MDCK cells, which are commonly used for studies on influenza virus replication, were treated identically. At different times p.i., wholecell extracts were analyzed for NF-B activation by EMSA, and the levels of NF-B DNA binding activity were quantitated by PhosphorImager analysis. Virus titers were determined at 24 h p.i. As expected, in A549 cells, WSN virus induced NF-B activation (Fig. 1, C and D) and caused a productive infection, with high levels of viral particles produced at 24 h p.i. (Fig. 1E). However, WSN infection was unable to induce NF-B up to 24 h p.i. in human kidney 293 cells, which are equally permissive to the virus. In addition, WSN did not induce NF-B activation in other human non-pulmonary cell lines susceptible to viral infection, including SH-SY5Y neuroblastoma cells and HaCaT keratinocytes. Finally, no activation of NF-B could be detected in MDCK cells, in which virus yield was higher than that in any of the other cell lines tested (Fig. 1, CϪE). In addition, WSN infection did not alter NF-B-dependent gene expression in MDCK cells transiently transfected with the luciferase reporter 2xNF-B-LUC (15) up to 16 h p.i. (data not shown). These results indicate that activation of NF-B is not required for influenza virus replication.
Influenza A Virus Induces NF-B-dependent Cellular Gene Transcription and IL-8 Production in Human Pulmonary Epithelial Cells-NF-B is considered to be a critical regulator of the inflammatory response because it can rapidly induce transcription of a variety of genes encoding proinflammatory proteins, including several inflammatory and chemotactic cytokines (9). To investigate whether influenza virus-induced NF-B could turn on proinflammatory gene expression in pulmonary epithelial cells, A549 cells were infected with WSN virus or mock-infected, and at 12, 24, and 48 h p.i., extracts were assayed for NF-B activity by EMSA. In parallel samples, total RNA was harvested and analyzed for transcription of the NF-B-dependent cytokines TNF␣, IL-1␤, IL-6, IL-8, and RANTES by RT-PCR. As shown in Fig. 2, activation of NF-B by influenza virus persisted at high levels for 48 h in infected cells and was associated with an increase in the level of cytokine mRNAs, which was already evident at 12 h p.i. RNA levels of all cytokines examined continued to be elevated up to 48 h p.i., a time at which the virusinduced cytopathic effect is evident and cell viability decreases.
Because the production of the chemotactic cytokine IL-8 has been associated with influenza virus-induced inflammation and airway pathology (5,6,22), we analyzed the role of NF-B in WSN-induced IL-8 transcription in A549 cells. Cells were infected with WSN virus or mock-infected and then analyzed for NF-B activity by EMSA, IL-8 transcription by RT-PCR, and IL-8 production by ELISA. As expected, NF-B DNA binding activity and accumulation of IL-8 transcript were strongly induced by WSN infection (Fig. 3A). IL-8 level in uninfected cells was Ͻ0.5 ng/ml, whereas it was greatly increased (Ͼ8 ng/ml) in the supernatant of WSN-infected cells (Fig. 3B), indicating that the IL-8 mRNA is efficiently translated.

NF-B Is Required for Induction of IL-8 during Influenza A Virus
Infection-Previous studies have identified in the 5Јflanking region of the IL-8 gene consensus binding sites for NF-B as well as for AP-1, OCT-1, and NFAT-1 transcription factors (23,24). To investigate whether influenza virus-induced IL-8 production was dependent on NF-B activation by the virus, we utilized the IB␣ super-repressor IB␣-AA, in which Ser 32/36 residues critical for phosphorylation by IKK have been replaced by alanine (14). A549 cells were transiently transfected with the IB␣-AA expression vector or with an "empty" vector. After 18 h, cells were mock-infected or infected with WSN virus. At 24 h p.i., whole-cell extracts were assayed for NF-B activity by EMSA, and the supernatant of the same samples was analyzed for IL-8 production by ELISA and virus yield by hemagglutinin titration. As shown in Fig. 3C, expression of IB␣-AA prevented WSN-induced NF-B activation. Inhibition of NF-B activation had no effect on virus production in cells expressing IB␣-AA in quadruplicate samples (empty vector-transfected cells, 64 Ϯ 0 HAU/ml; IB␣-AA vector-trans-  fected cells, 64 Ϯ 0 HAU/ml), confirming that the nuclear factor is not required for influenza virus replication in lung cells. On the other hand, expression of IB␣-AA prevented IL-8 production by infected cells, indicating that activation of NF-B is essential for WSN-mediated IL-8 induction. Inhibition of virusinduced IL-8 production was also detected after treatment of infected A549 cells with the proteasome inhibitor MG132, which prevents NF-B activation by blocking IB␣ degradation; moreover, as determined by RT-PCR, WSN infection did not induce IL-8 mRNA synthesis in 293 and SH-SY5Y cells, in which influenza virus was unable to activate NF-B (data not shown).

FIG. 3. NF-B is essential for influenza virus-induced IL-8 production by infected cells. A549 cells were infected with influenza virus (WSN) or mock-infected (U). A,
We next examined the effect of WSN on IL-8 gene promoter activity, utilizing luciferase constructs containing the 133-bp sequence upstream from the transcription start site of the human IL-8 gene (IL-8-luc) or the same sequence mutated in the NF-B site (IL-8-(NF-B-mut)-luc). A549 cells were transiently co-transfected with IL-8-luc or IL-8-(NF-B-mut)-luc reporter plasmids, together with empty or IB␣-AA vectors. After 18 h, transfected cells were infected with WSN and assayed for luciferase activity at 24 h p.i. As shown in Fig. 3D, WSN infection stimulated IL-8-luc promoter activity, whereas it had no effect on IL-8-(NF-B-mut)-luc activity. Expression of IB␣-AA resulted in a significant decrease of WSN-induced IL-8-luc promoter activity, and it had no effect in cells cotransfected with the IL-8-(NF-B-mut)-luc reporter. Taken together, these results confirm that NF-B is essential for IL-8 induction during influenza virus infection.
Effect of ⌬ 12 -PGJ 2 Treatment on IL-8 Production Induced by Influenza Virus-Cyclopentenone prostaglandins PGA 1 and 15-deoxy-⌬ 12-14 -PGJ 2 are potent inhibitors of TNF␣-induced IKK activity (10). Because the natural cyclopentenone PG ⌬ 12 -PGJ 2 was shown to possess therapeutic efficacy in a mouse model of lethal influenza A virus infection (12), we investigated the effect of ⌬ 12 -PGJ 2 treatment on influenza virus-induced NF-B activation and IL-8 production in human pulmonary cells. A549 cells were infected with WSN and then treated with ⌬ 12 -PGJ 2 (20 M) or control diluent. As previously reported in canine kidney cells (12), ⌬ 12 -PGJ 2 also inhibited influenza virus replication in human cells (control, 56 Ϯ 0; ⌬ 12 -PGJ 2treated cells, 6 Ϯ 0 HAU/ml) and partially protected A549 cells by WSN-induced cytopathic effect. At different time intervals, whole-cell extracts were analyzed for endogenous IKK and NF-B activities by kinase assay and EMSA, respectively. As expected, WSN infection induced IKK and NF-B activity starting at 12 h p.i. (Fig. 4A). Treatment with ⌬ 12 -PGJ 2 was effective in preventing WSN-induced IKK and NF-B activities, resulting in a significant decrease of virus-induced IL-8 production (Fig. 4). Under the same conditions, ⌬ 12 -PGJ 2 did not affect the activity of AP-1, another transcription factor involved in the control of IL-8 expression, which is induced during influenza virus infection (Fig. 4B). Treatment with noncyclopentenone prostanoids of the E and F type had no effect on both WSN-induced NF-B activation (Fig. 5A) and IL-8 production in A549 cells (data not shown).
We next investigated whether ⌬ 12 -PGJ 2 could inhibit IKK activity in vitro. A549 cells were stimulated with TNF␣ for 30 min. Endogenous IKK was immunoprecipitated with anti-IKK␣ antibody, and kinase activity was determined in the presence of different concentrations of ⌬ 12 -PGJ 2 or PGF 2␣ as a control (Fig. 5B). IKK activity was inhibited in a dose-dependent fashion by ⌬ 12 -PGJ 2 , whereas PGF 2␣ had no effect.

DISCUSSION
Despite large immunization programs, viral influenza remains a serious source of morbidity and mortality throughout the world and a significant cause of illness and death among people with immune deficiency associated with aging or different clinical conditions (1). Pandemics, such as the one in 1918, have been responsible for the death of millions of people, and the possibility of the sudden emergence of highly pathogenic strains of influenza A virus with elevated case fatality rate continues to represent a severe threat to human health (2,26). Antiviral chemotherapy with amantadine and rimantadine reduces the duration of symptoms of clinical influenza, but major side effects and the emergence of drug-resistant variants have been described (1,18). Novel sialic acid analogs, zanamavir and oseltamivir, have prophylactic and therapeutics effects (18). However, also due to the extremely high mutation rate of influenza viruses, the disease is by no means under control, and novel therapeutic approaches are needed, especially in view of the possible emergence of new virulent strains. Because severe lung inflammation has been associated with the devel-opment of secondary complications that may result in respiratory failure during influenza (2,4), in addition to the development of new antiviral agents, an understanding of the molecular mechanisms responsible for virus-induced inflammatory responses in the lung, could provide novel tools for therapeutic intervention during influenza virus infection.
The nuclear transcription factor NF-B is undisputedly recognized as a critical regulator of the inflammatory response (27). Once activated, NF-B rapidly triggers the expression of enzymes whose products contribute to the pathogenesis of the inflammatory process, such as cyclooxygenase 2, the inducible form of nitric-oxide synthase, cell adhesion molecules, and a variety of proinflammatory and chemotactic cytokines, including IL-8, which has been associated with lung inflammatory pathology (5,6,28). Interestingly, cytokines that are stimulated by NF-B, such as TNF␣ and IL-1␤, are also potent NF-B inducers, thus establishing a positive auto-regulatory loop that can amplify the inflammatory response and lead to chronic inflammation (27). Activation of NF-B has been reported after infection with different viruses, including the immunodeficiency virus HIV-1, hepatitis B and C viruses, herpes viruses, and influenza viruses (20, 29 -32). Different viruses have evolved different strategies to modulate the NF-B pathway, most of which converge on activation of the IB kinase IKK (7). In the case of orthomyxoviruses, whereas it has been recently reported that expression of different structural influenza viral proteins may trigger IKK activity (19), there was no direct evidence that in vivo infection with influenza virus could stimulate IKK function.
In the present report we show that influenza A virus infection strongly induces IKK activity in human pulmonary A549 cells. Stimulation of IKK function, determined by kinase assay, was maximal at 12 h after infection, leading to rapid degradation of NF-B inhibitory proteins IB␣ and IB␤, and resulting in a strong activation of NF-B. Interestingly, in lung cells NF-B DNA-binding activity was found to persist at high levels (over 4-fold above control) up to 48 h after infection, suggesting that the virus may interfere with the NF-B auto-regulatory loop.
NF-B activation during viral infection has been interpreted in some cases as a protective response of the host to viral pathogens, whereas in other cases it has been shown that the virus utilizes the factor to enhance its replication (7). Triggering of NF-B activation is particularly relevant during infection with viruses that harbor NF-B binding sites in their promoters. In this case, induction of the transcription factor results in transactivation of the B-containing viral promoter and in enhanced viral transcription. This has been documented in the case of HIV-1 infections (20,33) and, more recently, in the case of herpes simplex infections (15,34). In the case of influenza viruses, little is known thus far regarding the role of NF-B in the infection process. NF-B activation has been commonly interpreted as a marker of the cellular antiviral response to influenza infection; however, it has been recently suggested that, on the contrary, NF-B activity promotes both virus infectivity (35) and virus propagation via induction of proapoptotic factors in the infected cells (21).
The results described here indicate that viral replication is independent from the ability of the virus to activate NF-B in infected cells. In fact, the block of WSN-induced NF-B activation by expression of a dominant-negative form of the inhibitory protein IB␣ (IB␣-AA super-repressor) did not affect virus replication in pulmonary cells. Moreover, unexpectedly, whereas it potently induced NF-B activity in human lung cells, WSN virus was unable to activate NF-B in a variety of non-pulmonary human cells permissive to the virus, including kidney 293 cells, SH-SY5Y neuroblastoma cells, and HaCaT keratinocytes. In addition, induction of NF-B DNA binding activity was not detected in canine kidney MDCK cells, which are commonly used for studies on influenza virus replication. These results also indicate that lung epithelial cells are particularly sensitive to NF-B induction by influenza virus. The mechanism responsible for the highly sensitive response of the IKK/NF-B pathway to influenza virus infection in pulmonary cells is not known and needs to be investigated.
Because NF-B does not appear to be essential for influenza virus replication, we speculated that the dramatic and persistent activation of the nuclear factor observed in lung cells could induce cellular proinflammatory gene transcription. We then investigated the effect of WSN infection on the expression of NF-B-dependent cytokines TNF␣, IL-1␤, IL-6, IL-8, and RAN-TES by RT-PCR. The results indicate that activation of NF-B by influenza virus was associated with an increase in the level of all cytokine mRNAs. Cytokine RNA levels started to increase at 12 h p.i. and continued to be elevated up to 48 h p.i. In the case of IL-8, levels of the chemokine were found to be greatly increased in the supernatant of WSN-infected cells, indicating that the IL-8 mRNA is efficiently translated. Instead, IL-8 expression was not induced in other types of cells, in which NF-B was not activated following WSN infection.
As indicated above, the production of the chemotactic cytokine IL-8 has been associated with influenza virus-induced inflammation and airway pathology (5,6,22), suggesting that NF-B activation could be one of the events responsible for virus-induced pathogenicity in the lung. However, apart from B sites, consensus binding sites for different transcription factors, including AP-1, OCT-1 and NFAT-1, have been identified in the 5Ј-flanking region of the IL-8 gene (23,24). We then analyzed the role of NF-B in WSN-induced IL-8 transcription in A549 cells, inducing a functional knock-down of NF-B by using the dominant-negative form of IB␣. Expression of IB␣-AA prevented both WSN-induced NF-B activation and IL-8 production in A549 cells, indicating that activation of NF-B is essential for IL-8 induction. Similar results were obtained in experiments in which the IB␣-AA was utilized to block NF-B function in A549 cells expressing luciferase constructs containing the 133-bp sequence upstream from the transcription start site of the human IL-8 gene (IL-8-luc) or the same sequence mutated in the NF-B site. In these cells, expression of IB␣-AA resulted in a significant decrease of WSNinduced IL-8-luc promoter activity, and it had no effect in cells co-transfected with the mutated reporter. Moreover, inhibition of WSN-induced IL-8 production was also detected after treatment of infected A549 cells with the proteasome inhibitor MG132, which prevents WSN-induced NF-B activation by blocking IB␣ degradation. Taken together, these results indicate that NF-B is essential for IL-8 induction during influenza virus infection and suggest that the block of NF-B function could result in inhibition of virus-mediated inflammatory reactions. Inhibition of NF-B has also been recently suggested to decrease cell susceptibility to influenza virus infection by an as yet unknown mechanism (35).
We have shown that cyclopentenone prostanoids are potent inhibitors of NF-B activation by direct inhibition and modification of the IKK␤ subunit (10,16). In addition, we have shown that the natural cyclopentenone prostanoid ⌬ 12 -PGJ 2 possesses therapeutic efficacy in a mouse model of lethal influenza A virus infection (12). We then investigated the effect of ⌬ 12 -PGJ 2 treatment on influenza virus-induced NF-B activity and IL-8 production in human pulmonary cells. ⌬ 12 -PGJ 2 was found to be a powerful inhibitor of WSN-induced IKK and NF-B activities and caused a significant decrease of virus-induced IL-8 production, whereas treatment with non-cyclopentenone prostanoids of the E and F type had no effect. The fact that ⌬ 12 -PGJ 2 did not affect the activity of AP-1, another transcription factor involved in the control of IL-8 expression and induced during influenza virus infection (18), indicated that the NF-B pathway is the main target for the inhibitory activity of the prostanoid in influenza-infected cells. Furthermore, ⌬ 12 -PGJ 2 was found to greatly reduce the activity of the wild type IL-8 promoter, whereas it had no effect on the B-mutated reporter in WSN-infected cells. Finally, the fact that ⌬ 12 -PGJ 2 was able to directly inhibit IKK function in in vitro kinase assays identifies IKK as the target of the prostanoid in lung cells.
Cyclopentenone prostanoids are known to possess a potent antiviral activity against several RNA viruses in vitro and in vivo (12, 36 -39). The mechanism of the antiviral activity is distinct from any other known antiviral agent because cyclopentenone PGs act on the host cell by inducing a cellular defense response, which involves the synthesis of cytoprotective heat shock proteins (39 -41). We have now shown that, in the case of influenza A viruses, in addition to the previously described antiviral activity, ⌬ 12 -PGJ 2 is able to prevent IKKmediated virus-induced inflammatory reactions that could be involved in the pathogenesis of the disease. It should be pointed out that IKK inhibition could also alter normal host cell functions because the host utilizes NF-B to trigger defense mechanisms against invading viruses (7,42,43). However, the fact that a protective activity of cyclopentenone PG has been shown during influenza A virus infection in preclinical studies (12,36) argues against this hypothesis.
The results described here identify IKK as an important factor in triggering influenza A virus-induced inflammatory reactions in pulmonary epithelium, indicating a possible involvement of the IKK/NF-B pathway in the pathogenesis of influenza. They also suggest that cyclopentenone prostanoids or prostanoid-derived molecules could be good candidates for a novel class of antiviral drugs inhibiting both influenza virus replication and virus-induced inflammatory reactions.