Virology 471-473 (2014) 38–48

Contents lists available at ScienceDirect

Virology journal homepage: www.elsevier.com/locate/yviro

Efficient replication and strong induction of innate immune responses by H9N2 avian influenza virus in human dendritic cells Veera Westenius a,n, Sanna M. Mäkelä a, Thedi Ziegler a,b, Ilkka Julkunen a,c, Pamela Österlund a a

Virology Unit, National Institute for Health and Welfare (THL), Mannerheimintie 166, FI-00271 Helsinki, Finland Research Center for Child Psychiatry, University of Turku, Lemminkäisenkatu 3, 20014 Turku, Finland c Department of Virology, University of Turku, Kiinamyllynkatu 13, 20520 Turku, Finland b

art ic l e i nf o

a b s t r a c t

Article history: Received 10 July 2014 Returned to author for revisions 30 July 2014 Accepted 2 October 2014

Avian influenza A (H9N2) viruses have occasionally been identified in humans with upper respiratory tract infections. The novel H7N9/2013 virus identified in China shows that a low pathogenic avian influenza (LPAI) virus can be highly pathogenic in humans. Therefore, it is important to understand virus–host cell interactions and immune responses triggered by LPAI viruses in humans. We found that LPAI A/Hong Kong/1073/99 (H9N2) virus replicated efficiently in human dendritic cells (DCs). The H9N2 virus induced strong IFN gene expression although with different kinetics than seasonal influenza A/ Beijing/353/89 (H3N2) virus. IFN inducible antiviral proteins were produced in H9N2 virus-infected cells at the same level as in H3N2 infection. The H9N2 virus was extremely sensitive to the antiviral actions of type I IFNs. These results indicate that the avian influenza H9N2 virus is inducing a strong antiviral IFN response in human DCs. & 2014 Elsevier Inc. All rights reserved.

Keywords: Avian influenza IFNs H9N2 virus Innate immunity Cytokine Gene expression

Introduction Birds, especially waterfowl, are natural reservoirs of influenza viruses. From birds influenza viruses, such as the H5, H7, H9 and H10 subtypes, can be directly transmitted to mammals, including humans (Chan, 2002). Avian influenza became of great concern in 1997 when the first outbreak of highly pathogenic avian influenza (HPAI) A H5N1 virus in humans occurred in Hong Kong (Chan, 2002). According to the World Health Organization (WHO), infections by the HPAI H5N1 virus have lead to a lethal outcome in nearly 60% of documented human cases (WHO, 2014). In 2003 HPAI H7N7 virus caused a small epidemic in humans with one death (Fouchier et al., 2004). Since then H7N7 subtype viruses have caused more than 100 cases of human infections (WHO, 2014). A totally new threat for humans emerged in March 2013 when human infections with an avian influenza A H7N9 virus were reported. This was the first time when an H7N9 virus was found in humans and it was overall the first time when a low pathogenic avian influenza virus caused severe illness in humans (WHO, 2014; Gao et al., 2013). After that, due to enhanced influenza surveillance activity, other new avian influenza viruses were found to cause infections in humans, such as H6N1 and

n

Corresponding author. E-mail address: veera.westenius@thl.fi (V. Westenius).

http://dx.doi.org/10.1016/j.virol.2014.10.002 0042-6822/& 2014 Elsevier Inc. All rights reserved.

H10N8 (WHO, 2014). These cases demonstrate how serious threats low pathogenic avian influenza viruses may represent for humans. Influenza A virus subtype H9N2 is a low pathogenic avian influenza virus (LPAI). H9N2 viruses are widespread in birds in Asia and in the Middle-East (Alexander, 2007), and occasionally H9N2 virus has been transmitted from poultry to mammalian species, including humans and pigs (WHO, 2014; Peiris et al., 1999). Since 1999, when the first human infection was detected (Peiris et al., 1999), the isolation of H9N2 viruses from humans and swine has been reported infrequently. The most recent human case of H9N2 infection was detected in China at the end of the year 2013 (WHO, 2014). All human patients with H9N2 infection have experienced a mild influenza-like illness and have fully recovered. So far there has been no evidence of human-to-human transmission of H9N2 viruses. However, some of the H9N2 viruses currently circulating in birds have receptor specificity similar to that of human H3N2 viruses, i.e. they have the ability to bind to oligosaccharides terminating with a sialic acid linked to galactose via a α2,6-linkage (SAα2–6, humantype receptor) (Matrosovich et al., 2001; Guan et al., 2000). Adaptation to human-type receptors enables the virus to infect and replicate in mammalian hosts which increases the risk of a pandemic. A/Hong Kong/1073/1999 (H9N2) virus was isolated from a 4-year-old girl in Hong Kong in 1999. The girl had a history of mild eczema and asthma from 3 years of age. She was admitted to the hospital with fever and other influenza-like symptoms. Her fever settled within 6 days and she recovered (Peiris et al., 1999).

V. Westenius et al. / Virology 471-473 (2014) 38–48

The H9N2/99 virus contains internal genes similar to those of the highly pathogenic avian influenza A/Hong Kong/156/1997 (H5N1) -like viruses which have caused fatal infections in humans. Although avian influenza viruses of the H9 subtype preferentially bind to α-2,3 linked (avian-type) sialic acid, the A/Hong Kong/1073/ 1999 virus has the ability to bind to both α-2,3 and α-2,6 linked sialic acids which may indicate a broader host range, including humans (Guan et al., 2000; Lin et al., 2000).

39

Dendritic cells (DCs) are important immune cells, which orchestrate the formation of effective innate immune responses against microbes by producing regulatory cytokines. They are the principal antigen-presenting cells inducing proliferation and activation of T cells, and thus are bridging the innate and adaptive immunity. Together with DCs, alveolar macrophages reside close to the epithelium of the respiratory tract, which all form the front barrier against influenza viruses. Together these immune cells by sensing invading viruses initiate the antiviral responses to inhibit replication and spread of the virus in inflamed tissues (Seo et al., 2004; Moltedo et al., 2009). The genome of an influenza virus is mainly recognized by retinoic acid inducible gene-I (RIG-I), which primarily detects RNAs carrying 50 -triphosphate (Pichlmair et al., 2006). Toll-like receptors (TLRs) 3 and 7 preferentially recognize double-stranded RNA (dsRNA) and single-stranded RNA (ssRNA), respectively (Akira et al., 2006). Recognition of viral genetic material activates the antiviral signaling cascades in host cells leading to the transcriptional regulation of a range of cytokine genes. NF-κB, ATF2-c-Jun, IRF3 and IRF7 are the main transcription factors that regulate the expression of type I and III interferons (IFNs) and proinflammatory cytokines (Österlund et al., 2007; Marié et al., 1998), all of which contribute to innate antiviral responses against influenza viruses (Österlund et al., 2005; Ronni et al., 1997). In addition to cytokine responses, virus infection also induces the expression of many antiviral proteins, which have a role in the progression of the infection (Haller and Kochs, 2011; Diamond and Farzan, 2013; Min and Krug, 2006; Min et al., 2007). Although there have been several human infections caused by the H9N2 virus, the innate immune responses in human cells have remained uncharacterized. Thus, in the present study we analyzed H9N2 virus-induced innate immune responses in human monocyte-derived dendritic cells (moDCs), focusing on the expression of proinflammatory cytokine and antiviral IFN genes, the activation of transcription factors regulating these genes and the expression of antiviral proteins. We observed that the H9N2 virus replicated efficiently in human moDCs, and that H9N2 virus infection induced both IFNs and proinflammatory cytokine responses. We also found out that the H9N2 virus was extremely sensitive to the antiviral actions of type I IFNs (IFN-α and IFN-β).

Results Avian influenza A (H9N2) virus replicates efficiently in human moDCs To investigate the replication of the avian H9N2 virus in human immune cells, human moDCs were infected with the A/Hong Kong/1073/1999 (H9N2/99) virus. Influenza A/Beijing/353/1989 (H3N2/89) virus was used as control virus for which innate Fig. 1. Influenza A (H9N2) virus replicates effectively in human moDCs. Human monocyte-derived DCs (moDC) are infected with human seasonal influenza A/ Beijing/353/1989 (H3N2) or avian-origin influenza A/Hong Kong/1073/1999 (H9N2) viruses at MOI of 1, cells are harvested at different time points after infection as indicated. (A) Total cellular RNA samples are isolated for qRT-PCR analysis of the viral M1-specific RNAs. The expression levels of viral M1 gene are presented in relation to that of the 30 min sample, representing increases in the viral RNA levels during the time course of infection. The results are presented as means from triplicate measurements with standard deviations. The n symbols indicate statistical significance of differences in RNA expression level in a given time point between H3N2 and H9N2 viruses, np o 0.05 (Student's t test). (B) The whole cell lysates are prepared and viral NP and M1 proteins are analyzed by Western blotting. Total cellular IRF3 is analyzed from the same blot to control for equal loading. The data is representative of 4 independent experiments. For RNA and protein lysates the cells from four different blood donors are pooled after infection experiment. (C) For analyzing the cell viability during the virus infection, cells are stained and fixed for flow cytometric analysis. Cells from four different blood donors are handled separately, and data on percentage of dead cells is presented as mean values with standard deviations. Histograms show the fluorescence intensities of the staining color in samples of one donor.

40

V. Westenius et al. / Virology 471-473 (2014) 38–48

immune responses in this cell model are well defined (Österlund et al., 2005, 2012). The plaque-forming units (PFU) and the hemagglutination (HA) titers of the H3N2/89 and H9N2/99 virus stocks were determined. Both titrations gave similar difference in the infectivity between the virus stocks, i.e. 2  107 PFU/ml and 1.6  108 PFU/ml, and 128 and 512–1024 for HA titer, respectively. Both viruses were used at an equal multiplicity of infection (MOI), which is given according to the PFU titers determined in MDCK cells.

Virus replication was analyzed from viral RNA and protein expression by qRT-PCR and immunoblotting, respectively. In virusinfected human moDCs (at MOI of 1) relative viral matrix protein 1 (M1) RNA expression increased effectively during the time of infection with both H3N2/89 and H9N2/99 viruses (Fig. 1A). Both virus strains expressed notable amounts of nucleoprotein (NP) and M1 protein in the infected cells at 8 h and 24 h time points after infection (Fig. 1B). Virus infection was also observed microscopically by detecting the cells cytopathic effect (CPE; data not shown). Although both H3N2/89 and H9N2/99 viruses resulted in explicit infection in moDCs, still H9N2/99 virus infection led to more distinct CPE than what was seen in moDCs infected with H3N2/89 virus (data not shown). However, viability analysis of the cells over the course of virus infection showed no differences in the proportions of dead cells between H3N2/89 and H9N2/99 infected cells (Fig. 1C). Furthermore, for comparing the maturation status of moDCs induced by the H9N2/99 virus infection to that induced by the H3N2/89 virus, which have been shown to inefficiently induce DC maturation (Österlund et al., 2005), the expression of costimulatory molecule CD86 and HLA-II was analyzed by FACS. Our results confirmed that neither the H9N2/99 avian-origin virus nor the H3N2/89 virus is able to induce efficient maturation of moDCs (Supplementary Fig. 1). The results showed that like the human H3N2/89 virus, also the avian-origin H9N2/99 virus was able to infect human moDCs and to replicate efficiently in these cells inducing strong cell-death already within 24 h of infection.

H9N2 virus infection induces strong IFN response Given that the IFN system is the most important host defense mechanism during virus infection by restricting the virus growth and initiating immune responses, we wanted to compare the induction of IFN responses in moDCs triggered by the avianorigin H9N2/99 and the human seasonal H3N2/89 viruses. After infecting the cells, total cellular RNA samples were collected at different time points, and qRT-PCR was carried out to quantitate cytokine mRNA levels. Both H3N2/89 and H9N2/99 viruses induced antiviral IFN-β, IFN-α and IFN-λ1 gene expression in moDCs but a clear difference in the kinetics of the IFN gene expression between these viruses was seen (Fig. 2A). Interestingly, in H9N2 virus-infected cells IFN mRNA expression took place somewhat earlier, peaking at 6–8 h after infection as compared to cells infected by the H3N2/89 virus where the peak IFN mRNA expression levels were seen at 8–24 h after infection (Fig. 2A). Furthermore, we determined IFN protein levels in culture supernatants from cells infected with serially diluted viruses with an ELISA test and we found that H3N2/89 or H9N2/99 virus-infected cells produced similar amounts of IFN-α and IFN-β during the infection experiment (24 h) (Fig. 2B). Thus, the downregulation of the IFN mRNA expression seen in the late times of H9N2/99 infection did not reflect to the accumulation of produced IFNs during the whole infection period. Fig. 2. Interferon responses induced by H3N2 and H9N2 virus infection. (A) moDCs were infected with H3N2 or H9N2 viruses at MOI of 1 for various periods of times. Cells from four different blood donors are pooled and cellular RNA is isolated. Virus-induced IFN-α, IFN-β, and IFN-λ1 gene expression is analyzed by quantitative RT-PCR. The values are normalized to 18S rRNA and presented as relative gene expression in relation to the RNA sample obtained from uninfected control moDCs. Values from triplicate measurements are presented as means and standard deviations. The experiment is repeated 4 times. The n symbols indicate significant differences between H3N2 and H9N2 viruses, npo 0.05. (B) To analyze the secreted IFN-α and IFN-β protein levels with an ELISA assay, cell culture supernatants are collected from virus-infected moDCs at 24 h post-infection. Cells were infected with different doses of influenza (MOI 0.01; 0.05; 0.25; 1.25; 6; 31; 100 from left to right). The samples are analyzed as duplicate in the assay. The results are from four different blood donors analyzed separately and presented as mean values with standard deviations.

V. Westenius et al. / Virology 471-473 (2014) 38–48

41

Fig. 3. Activation of transcription factors stimulated by H3N2 and H9N2 virus infection. (A) The expression of signaling molecules P-IRF3, IRF3, IRF1, IRF7, P-p38 MAPK and P-STAT1, and expression of IκBα, inhibitor for the transcriptionally active form of NFκB is analyzed from virus-infected moDCs at different time points as indicated by the Western blot technique. GAPDH and actin are stained for loading control. Cells from four different donors are pooled for immunoblot analysis. Western blot bands are quantitated and expression of signaling molecules in relation to the loading control is presented as fold induction over CTRL. A representative experiment out of 3 is shown.

Several transcription factors are involved in the regulation of cytokine gene expression during influenza virus infection. To reveal the transcriptional regulation of IFN responses in H9N2 infection, we analyzed the expression levels of different regulatory factors in H9N2/99 and H3N2/89 virus-infected cells. IRF3 is one of the most important transcription factors regulating the early antiviral genes after viral recognition-induced signaling. Interestingly, the phosphorylation of IRF3 was remarkably stronger in moDCs infected with H3N2/89 virus than in H9N2/99 virus-infected cells although the IRF3 phosphorylation started to become visible at the same time point (6 h) of infection by both viruses (Fig. 3). The faint activation of IRF3 by the avian-origin H9N2 virus is well in line with that seen by the novel LPAI H7N9 virus in human moDCs (Arilahti et al., 2014). The expression of other important IRFs regulating the antiviral IFN genes was also analyzed, and the expression of IRF1 and IRF7 did not differ notably between H9N2/99 or H3N2/89 virusinfected cells. Also, there was no detectable difference in the phosphorylation of p38 MAP kinase between these viruses, but the phosphorylated form of signal transducers and the activators of transcription 1 (STAT1) was expressed at two fold higher level in cells infected with H9N2/99 virus (Fig. 3). Next we studied the activation of the NFκB transcription factor by monitoring the degradation of IκBα. IκBα amount is inversely proportional to the amount of transcriptionally active NFκB since IκBα is a multifunctional inhibitor of NFκB (Jacobs and Harrison, 1998). Comparing the amount of IκBα in H3N2/89 and H9N2/99 virus-infected cells, there was no difference in the levels of IκBα in cells infected with H9N2/99 or H3N2/89 virus (Fig. 3). Altogether the activation of transcription factors regulating the antiviral IFN genes seems not to be very different in cells infected with H3N2/ 89 virus as compared with H9N2/99 virus-infected cells. Based on these results it seems clear that the H9N2/99 virus can activate the human immune system, but the activation of any specific transcription factor may not alone explain the temporal difference in IFN responses between these viruses. Proinflammatory cytokine and chemokine gene expression in moDCs infected with H3N2 and H9N2 influenza A viruses It has been shown that human infection with H5N1 avian influenza virus leads to a massive pro-inflammatory cytokine and

chemokine response, which can contribute to the severe outcome of the disease (de Jong et al., 2006). To analyze whether the avian influenza virus H9N2 can also trigger robust cytokine responses, we investigated the expression of proinflammatory cytokine TNF-α and IL-1β and chemokine CXCL10 and CCL5 genes in H3N2/89 or H9N2/ 99 virus-infected moDCs. The relative expression of TNF-α and IL1β mRNAs decreased after 6 h of infection with H9N2/99 virus but maximal expression levels were similar with both types of viruses (Fig. 4A). The expression level of CXCL10 gene was stronger in H9N2-infected cells as compared to that of H3N2/89 infection (Fig. 4A). These findings suggest that the avian-origin H9N2/99 virus and the seasonal H3N2/89 virus induce overall a very similar proinflammatory immune response in human immune cells. Since adaptive immunity, together with innate immunity, takes part in the clearance of influenza virus infection we also analyzed the expression of IL-12, which is an important regulator in the development of the Th1 and Th17 types of adaptive immunity (Vignali and Kuchroo, 2012). The gene expression of p35 and p40 subunits of IL-12 was very similar in H9N2/99 and in H3N2/89 virus-infected cells (Fig. 4B). It was clearly visible that the expression of IL-12 subunit p40 peaked already at 3–6 h time points whereas the expression of p35 subunit clearly increased at 24 h post-infection with H3N2/89 and H9N2/99 viruses. IRF3, IRF7 and NFκB transcription factors regulate innate immune responses induced by avian influenza H9N2 virus As shown in Fig. 3, there are not many differences in between transcription factor activation during H9N2 and H3N2 virus infections in moDCs. Thus, to further explore the importance of certain transcription factors in regulating the cytokine genes during human or avian influenza infections, we used mouse embryonic fibroblasts (MEFs) devoid of NFκB or IRF3 and IRF7 to analyze whether these factors are involved in H9N2 and H3N2 virus-induced cytokine responses. H3N2/89 and H9N2/99 viruses replicated efficiently in both wild type and knock-out (KO) cells as evidenced by the expression of viral M1 RNA (Fig. 5). Seasonal influenza viruses activate IRF3 and NFκB and induce the expression of IRF7 transcription factor, which are significant in regulating innate immunity during the infection (Ehrhardt et al., 2010). Here we showed that in both NFκB KO and IRF3/7 KO cells H9N2/99 virus-induced IFN-β

42

V. Westenius et al. / Virology 471-473 (2014) 38–48

Fig. 4. Human H3N2 and avian H9N2 virus-induced inflammatory cytokine responses in moDCs. Cells from four different donors are infected with H3N2 or H9N2 viruses (MOI 1) for different times, cells from different blood donors are pooled and cellular RNA is isolated. The expression of (A) pro-inflammatory cytokines CXC10, CCL5, IL-1β and TNF-α, or (B) IL-12 family subunits p35 and p40 is analyzed by quantitative RT-PCR. The values are normalized to 18S rRNA and presented as relative gene expression in relation to the uninfected mock sample. Mean values with standard deviations calculated from triplicate measurements are presented. Experiment is a representative of one out of four. The asterisks indicate statistically significant differences between the H3N2 and H9N2 viruses, np o 0.05.

and CXCL10 gene expression was significantly reduced whereas the expression of these cytokine genes readily took place in wild type cells (Fig. 5). The expression of IFN-β and CXCL10 was also inhibited in H3N2/89 infection in both types of KO cells. However, in mouse cells, the human H3N2/89 virus-induced TNF-α expression was appeared to be completely independent on IRF3 and IRF7 whereas the avian-origin H9N2/99 virus was overall unable to induce TNF-α expression at a notable level. The cytokine gene expression levels as a whole were lower in H9N2/99 virus-infected mouse cells as compared to those seen in H3N2/89 virus-infected cells (Fig. 5). Expression of regulators of innate immunity and antiviral proteins during H9N2 virus infection IFNs mediate the activation of different antiviral proteins in a paracrine and autocrine fashion that contribute to the antiviral state

in the responding cell. The development of antiviral status is regulated by various modulators at different points of the activation cascade. To analyze the regulation of antiviral status in H9N2 and H3N2 virus infections, we analyzed the expression of interferoninduced protein with tetratricopeptide repeats 3 (IFIT3) and suppressor of cytokine signaling 3 (SOCS3). SOCS3 is a negative feedback regulator of innate immune responses, especially of type I IFN signaling (Pothlichet et al., 2008), which may contribute to downregulation of robust inflammatory response during the infection. Contrary to SOCS3, IFIT3 is considered to be a positive regulator for IRF3 activation, and IFIT3 enhances the host antiviral responses (Liu et al., 2011). The expression of IFIT protein family members is induced by IFNs, and they inhibit the replication of viruses at translational level (Diamond and Farzan, 2013). In our infection experiments we observed that the expression of SOCS3 and IFIT3 mRNA was similar in H3N2 and H9N2-infected cells (Fig. 6A).

V. Westenius et al. / Virology 471-473 (2014) 38–48

43

Fig. 5. Transcription factors IRF3, IRF7 and NFκB are involved in influenza H3N2 and H9N2 virus-induced cytokine responses. Wild type (wt) mouse embryonic fibroblasts (MEF) and (A) MEFs devoid of IRF3 and IRF7 transcription factors (IRF3/7 double KO) or (B) MEFs devoid of NF-κB transcription factors (NFκB RelA  /  c-Rel  /  Nfκb1  /  triple KO) are infected with H3N2/89 and H9N2/99 viruses (MOI 5) for various periods of times. RNA samples are prepared for analysis of the relative gene expression of M1, IFN-β, CXCL10 and TNF-α genes. The data is presented as relative gene expression levels for viral M1 gene in relation to that of the 30 min or 1 h sample and for cytokine genes in relation to those of the mock sample. The results are presented as means from triplicate measurements with standard deviations. Representative experiments out of 2 are shown. The n symbols indicate significant differences between KO and wt cells, np o 0.05.

Type I and III IFNs activate the expression of numerous IFNinducible genes of antiviral proteins, like PKR, OAS, Mx and IFITM proteins (Haller and Kochs, 2011; Diamond and Farzan, 2013; Min and Krug, 2006; Min et al., 2007). The Mx proteins have been shown to exhibit antiviral activities against influenza viruses by preventing viral transcription through an interaction with viral NP (Turan et al., 2004). ISG15 is another antiviral protein, which is induced by IFNs and has an ability to inhibit influenza A and B virus infections (Hsiang et al.,

2009). In addition to MxA and ISG15, IFITM3 is also a significant antiviral protein, which has been shown to restrict infection by several enveloped viruses including influenza A virus (Brass et al., 2009). The expression of antiviral ISG15, IFITM3 and MxA proteins was analyzed in avian H9N2 and human H3N2 virus-infected moDCs, and we found out that those proteins were produced almost equally well by H9N2 and H3N2 infections (Fig. 6B). Altogether, these results suggest that the H9N2 virus is able to induce antiviral status in human cells practically

44

V. Westenius et al. / Virology 471-473 (2014) 38–48

Fig. 6. Comparison of H3N2 and H9N2 virus-induced expression of SOCS3 and antiviral genes in moDCs. Human moDCs from four different blood donors are infected with H3N2 or H9N2 viruses at MOI of 1 and cellular samples are collected at 0.5, 1, 3, 6, 8 and 24 h time points after infection, cells from different blood donors are pooled and total cellular RNA and protein samples are prepared. The data is representative of 3 independent experiments. (A) Gene expression of MxA, SOCS3 and IFIT3 mRNA is determined by quantitative RT-PCR. The values are normalized to 18 S rRNA and presented as relative gene expression in relation to the RNA sample obtained from uninfected control moDCs. The results are mean values from triplicate measurements with standard deviations. (B) From total cellular proteins the expression levels of ISG15, IFITM3, MxA and IRF3 are determined by immunoblotting with ECL system. Western blot bands are quantitated and ISG15, IFITM3 and MxA protein expression is presented in relation to IRF3 as a loading control after which the fold induction over the mock CTRL was calculated.

as efficiently as the seasonal H3N2 virus although a temporal difference in the kinetic of IFN gene expression between these viruses was seen (Fig. 2A). The H9N2 avian influenza virus is extremely sensitive to the antiviral actions of type I IFNs After the analysis of the virus-induced antiviral status in infected cells, we wanted to study whether external cytokines can prevent the infectivity and replication of the avian-origin H9N2 virus. Thus we analyzed the antiviral actions of IFN-α, IFN-β, TNF-α and IL-1β against H3N2 and H9N2 subtype. After pretreating moDCs with cytokines for 16 h, the cells were infected with these viruses at a MOI of 1 for another 20 h. RNA samples were collected before and

after infection, and the antiviral activity of different cytokines was analyzed based on inhibition of viral M1 gene expression. We also determined MxA gene expression in uninfected cytokine-primed cells prior to infection. As expected, IFN-α and IFN-β triggered a significant increase in the expression of MxA gene as compared to an untreated sample (Fig. 7). There was no detectable expression of MxA gene in TNF-α or IL-1β treated cells (Fig. 7). In IFN-pretreated and influenza virus-infected cells the relative copy number of M1 gene decreased in an inverse relation to the IFN dose indicating that IFN pretreatment efficiently inhibited the replication of both types of influenza viruses (Fig. 7). This proves that the H9N2 avian influenza virus is extremely sensitive to the antiviral actions of type I IFNs. There was no detectable reduction in the expression of M1 gene in TNF-α or IL-1β pretreated cells. On the contrary, the

V. Westenius et al. / Virology 471-473 (2014) 38–48

Fig. 7. IFN-mediated antiviral effects against human H3N2 and avian H9N2 viruses. moDCs are pretreated with 1, 10 or 100 IU/ml of IFN-α and IFN-β or with 1, 10 or 100 ng/ml of IL-1β or with 0.5, 5 or 50 ng/ml of TNF-α for 16 h prior to infection with H3N2 and H9N2 viruses (MOI 1) for 20 h. Every sample includes cells from four different blood donors that were pooled. Virus-infected samples are prepared for viral M1 gene expression analysis by qRT-PCR. Cytokine-primed uninfected control cells are harvested at the time of influenza virus infection (16 h after cytokine pretreatment) for MxA gene expression analysis. The results are mean values of three replicates with standard deviations from a representative experiment out of two. For IFN-α and IFN-β priming there is a statistically significant decrease and for TNF-α and IL-1-β priming a statistically significant increase in the M1 gene expression (np o 0.05) when compared with the unprimed, infected sample as analyzed with the Student's t test.

replication of both viruses was even increased in TNF-α and IL-1β pretreated cells. This suggests that inflammatory signals (by TNF-α or IL-1β) may enhance the replication of influenza viruses, a phenomenon that deserves further analysis.

Discussion Avian influenza A (H9N2) viruses are common in poultry and wild birds worldwide (Alexander, 2007), which increases the probability of reassortation with other influenza viruses circulating in domestic and wild birds. This has already been seen in the emergence of H7N9 and H10N8 viruses possessing internal gene cassettes originating from an avian H9N2 virus (Chen et al., 2014; Liu et al., 2013). The H9N2 viruses are low pathogenic in avian hosts and the infection may even be asymptomatic, which enables the virus and epidemics to remain unrecognized by surveillance systems.

45

Some H9N2 virus isolates are able to bind to α2,6-linked sialic acid receptors, the human-type of influenza receptor (Matrosovich et al., 2001; Choi et al., 2004), and H9N2 viruses have been occasionally isolated from humans experiencing an upper respiratory tract infection. Via reassortation with a human virus, H9N2 virus could obtain features that might lead to enhanced replication and efficient transmission in the human host (Wan et al., 2008). Since avian influenza viruses are considered to have a significant risk of transmission to humans and thus to form a serious public health threat globally, we analyzed the ability of a human isolate of avian H9N2 virus to replicate and induce innate immune responses in human immune cells, such as the dendritic cells. In mouse and chicken experiments it has been shown that LPAI H9N2 virus activates innate immune responses, including e.g. interferon, pro-inflammatory cytokine and interleukin production and induction of the expression antiviral proteins like Mx1 (Huang et al., 2013; Lee et al., 2013). Also in human A549 cells it has been shown that LPAI H9N2 virus infection stimulates type I and III interferon expression, STAT1 activation and the expression of IFN-stimulated genes (ISGs) (Sutejo et al., 2012). The present study shows that the avian-origin influenza A/Hong Kong/1073/99 (H9N2) virus can replicate efficiently in human DCs and induce both pro-inflammatory and antiviral IFN responses at significant level. Although the H9N2 virus induced early IFN responses in a similar fashion or even more efficiently than the seasonal H3N2 virus did at later time points of H9N2 infection these responses decreased (Fig. 2A). Furthermore, the reduced late IFN responses in H9N2 infection were associated with weaker activation of transcription factors IRF3 and IRF7 at these time points, which is consistent with recent observations demonstrating impaired IFN responses by the novel LPAI H7N9 virus in human DCs (Arilahti et al., 2014). One explanation for the inhibition of late IFN responses in the H9N2/99 virus-infected cells could be the viral non-structural protein 1 (NS1) which is known to have a major role in inhibiting host innate immune responses, especially IFN production (Hale et al., 2008). Xing and coworkers have shown that IFN gene expression was inhibited by the NS1 of A/pheasant/California/2373/1998 H9N2 virus whereas the expression of pro-inflammatory cytokine genes was not affected by the functions of NS1 in chicken macrophages (Xing et al., 2009). Transfer of the NS segment via reassortation from one avian influenza virus to another has been shown to have effects on the pathogenicity of the virus (Ma et al., 2010; Wang et al., 2010). The early HPAI H5N1 strains (clade 0) as well as the recent LPAI H7N9/ 2013 virus have acquired their internal protein encoding genome segments from the LPAI H9N2 virus lineage which possesses NS1 gene mutations F103Lþ M106I (Guan et al., 1999). These mutations have been shown to be associated with increased IFN antagonism, increased cytoplasmic localization of NS1 protein and altered NS1 interaction with host regulatory proteins (Dankar et al., 2013). Most of the human virus strains as well as the HPAI H5N1 strains after its reappearance in 2003 have the same consensus 103F and 106M NS1 protein signature which has been associated with increased CPSF-30 binding activity (Spesock et al., 2011). Further studies are warranted to reveal the detailed mechanisms of the viral NS1 protein of H9N2 viruses on the antiviral host responses. Different avian influenza strains express differences in the temporal activation and the induction levels of the pro-inflammatory and antiviral cytokine responses in human cell systems (Arilahti et al., 2014; Sutejo et al., 2012; Xing et al., 2011). Although the HPAI H5N1 has been reported to be able to induce strong cytokine responses, “cytokine storm” in humans (de Jong et al., 2006), in H7N9/2013 avian influenza virus-infected cells the antiviral IFN responses were largely impaired (Arilahti et al., 2014). Proinflammatory cytokine responses were also significantly weaker in H7N9 virus-infected DCs as compared to cells infected with the H5N1 or H3N2 viruses (Arilahti et al., 2014). Our present study characterizing the cytokine

46

V. Westenius et al. / Virology 471-473 (2014) 38–48

responses in H9N2 virus-infected moDCs is in line with the results by Xing and colleagues who showed that in human bronchial epithelial cells infected with the H9N2 virus IFN-β was the prominent antiviral component together with CXCL10, CCL5 and TNF-α proinflammatory responses, whereas IL-1β, IL-8 and IL-6 might play a minor role in the pathogenicity (Xing et al., 2011). These virus-specific differences in host immune responses to avian influenza viruses highlight the importance of studying in more detail the viral determinants contributing to the pathogenicity of influenza viruses in humans. Interestingly, we observed that priming the cells with proinflammatory cytokines, TNF-α and IL-1β, increased the replication of the H3N2 and especially the H9N2 influenza virus (Fig. 7). If such cytokine-induced inflammation is predisposing the cells to support a more efficient virus replication, this observation could explain the effective replication of the H5N1 influenza viruses in humans and simultaneous robust production of TNF-α and “cytokine storm” (de Jong et al., 2006). This notion is important also in virus-bacteria coinfections, since many bacteria capable of causing respiratory infections induce strong inflammatory response with increased cytokine levels (Latvala et al., 2008; Veckman et al., 2004). Bacteria-induced pro-inflammatory response could thus contribute to further increase the virus replication at the site of infection. Although the avian influenza H9N2 virus is low pathogenic and causes mainly mild disease in humans, reassortation of H9N2 virus with other influenza A virus strains can result in a new type of reassortant virus with increased virulence and pathogenicity in humans. This has already occurred in several occasions of which the most notable one was the emergence of the H7N9 virus in 2013. Thus, it is important to characterize the innate immune responses in humans induced also by low pathogenic avian influenza A viruses in order to be prepared for any new reassortant viruses and their specific properties such as the molecular determinants of pathogenesis, factors contributing to the ability to replicate in mammalian host and sensitivity to antiviral substances. Our present study clearly demonstrates that an avian-origin H9N2 virus that was initially isolated from an infected person in 1999 has the ability to efficiently replicate in human DCs. However, this virus appeared to have the ability to downregulate IFN gene expression at late times of infection. Differently from the HPAI H5N1 infection, no “cytokine storm” was associated with this LPAI H9N2 infection in human DCs, but rather the immune cells were able to respond to the avianorigin virus with a similar fashion as to the seasonal human H3N2 virus. We believe that the present study will provide valuable information for further understanding of innate immunity in humans during avian influenza infection.

Materials and methods Cell cultures Human primary monocyte-derived DCs (moDCs) were differentiated from peripheral blood monocytes following a standard procedure (Veckman et al., 2004). Buffy coats were obtained from healthy blood donors (Finnish Red Cross Blood Transfusion Service, Helsinki, Finland). Peripheral blood mononuclear cells (PBMC) were fractioned from buffy coats by Ficoll-Paque (Pharmacia Biotech) gradient centrifugation. Monocytes were purified from PBMCs by centrifugation on a Percoll gradient (Amersham Biosciences) following lymphocyte depletion with anti-CD3 and anti-CD19 magnetic beads (Dynal). Monocytes were allowed to adhere to cell culture plates in RPMI medium (Sigma Aldrich) supplemented with 0.6 mg/ml penicillin, 60 mg/ml streptomycin, 2 mM L-glutamine and 20 mM HEPES. moDCs were differentiated by culturing monocytes in RPMI medium (Sigma Aldrich) in the presence of 10% FCS (Integro), 10 ng/ml GMCSF (BioSource) and 20 ng/ml IL-4 (Invitrogen Gibcos) for 6 days to

generate cells with the characteristic phenotype of DCs (CD1asemi, CD40high, CD14  ) and showing a typical DC morphology (Veckman et al., 2004; Sallusto and Lanzavecchia, 1994). Primary wild type (wt) and Irf3  /-Irf7  /  mouse embryo fibroblasts (MEFs) and RelA  /  c-Rel  /  Nfκb1  /  (NFκB KO) 3T3 cell line were kindly provided by Dr. A. Hoffmann, Signaling Systems Lab, San Diego. Primary MEFs were immortalized by rigorous passaging protocol to obtain wt and IRF3/7 KO 3T3 cell lines. The MEFs and 3T3 cell lines were cultured in DMEM -medium supplemented with 0.6 mg/ml penicillin, 60 mg/ml streptomycin, 2 mM L-glutamine and 20 mM HEPES and 10% FCS. Influenza A viruses Human influenza A/Beijing/353/89 (H3N2) virus and human isolate of an avian influenza A/Hong Kong/1073/99 (H9N2) virus (WHO Collaborative Center for Influenza, London, UK) were grown in the allantoic cavity of 11-day-old embryonated chicken eggs at þ36 1C for 3 days. Hemagglutination assay was done with standard protocol using 0.5% turkey red blood cells. Infective virus titers were determined by a plaque assay. Viruses were serially diluted and then inoculated to confluent MDCK cells for 1 h at 37 1C. After 1 h the cells were washed and covered with an agarose overlay (1% agarose and 1 mg/ml TPCK-trypsin in Eagle-MEM). After 3–6 days incubation the plaques were counted to obtain the concentration of infective viruses as PFU/ml. The multiplicity of infection (MOI) is given according to the titers determined in MDCK cells. Virus infection experiments Cells were infected with influenza viruses for different time periods (15 min to 24 h) as indicated in the figures. The 3T3 cells were infected with A/Beijing/353/89 (H3N2) or A/Hong Kong/ 1073/99 (H9N2) viruses at multiplicity of infection (MOI) of 5 or 10 as indicated. Human monocyte-derived DCs were infected at MOI between 0.001 and 100. The propagation of the A/Hong Kong/ 1073/99 (H9N2) virus stock and all infection experiments with the H9N2 virus were carried out under biosafety level 3 (BSL-3) conditions. The H3N2 and H9N2 viruses were always handled separately to avoid unintentional reassortation. IFN priming of virus-infected cells In IFN priming experiments moDCs were treated with 1, 10 or 100 IU/ml of IFN-α or IFN-β or with 1, 10 or 100 ng/ml of IL-1β or with 0.5, 5 or 50 ng/ml of TNF-α for 24 h before the cells were infected with influenza viruses. Recombinant human IFN-β-1b and IFN-α-2b were purchased from Schering-Plough and recombinant human TNF-α and IL-1β2b were from Biosource. qRT-PCR Infected cells were harvested and cells from different blood donors were pooled (Pietilä et al., 2010) before total cellular RNA was isolated using the RNEasy Mini kit (Qiagen) including DNase digestion (RNase-free DNase kit, Qiagen). RNA isolation from H9N2/ 99 virus-infected cells was carried out in BSL-3 facilities. 0.5 or 1 mg of total cellular RNA was transcribed to cDNA using TaqMan Reverse Transcriptase kit (Applied Biosystems) with random hexamers as primers. The cDNA was amplified by PCR using TaqMan Universal PCR Mastermix and Gene Expression Assays (Applied Biosystems). The data from human cells was normalized to 18S rRNA with TaqMan Endogenous Control kit (Applied Biosystems) and for the mouse cells experiments we used mouse GAPD Endogenous Control kit (Applied Biosystems). qRT-PCR was done as multiplex (18S or

V. Westenius et al. / Virology 471-473 (2014) 38–48

GAPD and investigated gene) reactions with triplicate measurements. Gene expression data is presented as fold induction of the relative gene expression in relation to the unstimulated samples. For virus replication M1 gene expression data is presented as the relative gene expression in relation to the 30 min samples which represent the input amount of the virus. The influenza A-specific primer-probe pair for M1 detects a highly conserved sequence in the M gene of all influenza A viruses (Ward et al., 2004). The results are presented as means from triplicate measurements with standard deviations. Data from qRT-PCR was analyzed with the Student's t test to determine the statistical significance of differences between viruses or knock out/wild type cells. Values of po0.05 were considered significant. ELISA Levels of secreted IFN-α and IFN-β were analyzed from cell culture supernatants using VeriKine™ human interferon alpha Multi-subtype ELISA kit (PBL Assay Science) or VeriKine™ human IFN beta ELISA kit (PBL Assay Science). The cytokine levels in cell culture supernatants from different blood donors were analyzed separately. Western blot analysis For Western blot analysis moDCs were harvested and the cells from four different blood donors were pooled. Cell pellets were lysed with the passive lysis buffer of the Dual Luciferase Assay Kit (Promega) containing 1 mM Na3VO4. MEF cells were harvested and cell pellets lysed with a RIPA buffer (1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 0.15 M NaCl, 0.01 M sodium phosphate, pH 7.2) containing 0.5 mM DTT, 1 mM Na3VO4 and protease inhibitor mixture (Complete, Roche). Both the moDCs and MEF cell lysates were boiled for at least 10 min in the Laemmli sample buffer before taking the samples out of the BSL-3 facilities. Equal volumes of samples were separated on SDS-PAGE by using the Laemmli buffer system. After the gel electrophoresis proteins were transferred to Hybond-P polyvinylidene difluoride (PVDF) membranes (Amersham Biosciences). The membranes were blocked with 5% milk protein in PBS or TBS. The primary rabbit antibodies against IRF1, IRF3, IRF7, MxA, influenza A NP and M1 were prepared as described previously (Österlund et al., 2005; Melén et al., 2007; Matikainen et al., 1996; Ronni et al., 1993). The staining was done in blocking buffer at RT for 1 h. Antibodies for both phosphorylated IRF3 (P-IRF3) and p38 MAPK (P-p38 MAPK), ISG15, GAPDH and IκBα were commercial antibodies from Cell Signaling Technology. The antibody against IFITM3 was from Abgent and the antibody against actin and phosphorylated STAT1 (P-STAT1) was from Santa Cruz Biotechnology. Staining with commercial antibodies was done following the manufacturer's instructions. For the secondary antibody anti-rabbit HRP-conjugated antibodies (Dako) were used. Secondary antibody staining was done at RT for 1 h. Protein bands were visualized on HyperMax films using an ECL plus system (GE Healthcare). When indicated, the protein band intensities were quantitated with ImageJ software (http://imagej.nih.gov/ij/) to turn pixel intensity into optical density (OD). Protein expression data was normalized to the loading controls and then presented as fold induction of the relative protein expression in relation to the unstimulated samples. Flow cytometry Flow cytometry was used to analyze the viability and maturation status of moDCs during virus infection. MoDCs were infected with A/Beijing/353/89 (H3N2) or A/Hong Kong/1073/99 (H9N2) viruses at multiplicity of infection (MOI) of 1 and cells were harvested after different time periods (3 h, 8 h and 24 h). The cells from four

47

different blood donors were collected and handled separately. The viability of infected cells was determined with LIVE/DEADs Fixable Far Red Dead Cell Stain Kit (Life Technologies) and staining was done following the manufacturer's instructions. Cells were fixed with 90% methanol on ice for 20 min, washed and nonspecific binding of antibodies was prevented by keeping the cells in 1% BSA in PBS. H9N2 virus-infected cells were not released from BSL-3 containment before the methanol fixation. The maturation of moDCs was analyzed by staining cells with FITC–anti-HLA-II and PE–anti-CD86 (Caltag Laboratories). The fluorescence-labeled cells were analyzed with a FACSCanto II flow cytometer using FACSDiva software (Becton Dickinson).

Acknowledgments The study was supported by the Medical Research Council of the Academy of Finland (Grant numbers 252252 and 255780). This study was also funded by the Sigrid Jusélius and the Finnish Cultural Foundation and Integrative Life Science Doctoral Program. We thank Hanna Valtonen, Teija Aalto, Riitta Santanen and Esa Rönkkö for excellent technical assistance and we appreciate the help from Susanna Sissonen on biosafety aspects. We are grateful to Dr. Alexander Hoffmann for kindly providing the MEF cells and to Dr. Riku Fagerlund for the excellent guidelines cultivating the MEF cells.

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.virol.2014.10.002.

References Akira, S., Uematsu, S., Takeuchi, O., 2006. Pathogen recognition and innate immunity. Cell 124, 783–801. http://dx.doi.org/10.1016/j.cell.2006.02.015. Alexander, D.J., 2007. An overview of the epidemiology of avian influenza. Vaccine 25, 5637–5644. http://dx.doi.org/10.1016/j.vaccine.2006.10.051. Arilahti, V., Mäkelä, S., Tynell, J., Julkunen, I., Osterlund, P., 2014. Novel avian influenza A (H7N9) virus induces impaired interferon responses in human dendritic cells. PLoS One 9, e96350. http://dx.doi.org/10.1371/journal. pone.0096350. Brass, A.L., Huang, I., Benita, Y., John, S.P., Krishnan, M.N., Feeley, E.M., Ryan, B.J., Weyer, J.L., van der Weyden, L., Fikrig, E., Adams, D.J., Xavier, R.J., Farzan, M., Elledge, S.J., 2009. The IFITM proteins mediate cellular resistance to influenza A H1N1 virus, West Nile virus, and dengue virus. Cell 139, 1243–1254. http://dx. doi.org/10.1016/j.cell.2009.12.017. Chan, P.K.S., 2002. Outbreak of avian influenza A(H5N1) virus infection in Hong Kong in 1997. Clin. Infect. Dis. 34, S58–S64. http://dx.doi.org/10.1086/338820. Chen, H., Yuan, H., Gao, R., Zhang, J., Wang, D., Xiong, Y., Fan, G., Yang, F., Li, X., Zhou, J., Zou, S., Yang, L., Chen, T., Dong, L., Bo, H., Zhao, X., Zhang, Y., Lan, Y., Bai, T., Dong, J., Li, Q., Wang, S., Zhang, Y., Li, H., Gong, T., Shi, Y., Ni, X., Li, J., Zhou, J., Fan, J., Wu, J., Zhou, X., Hu, M., Wan, J., Yang, W., Li, D., Wu, G., Feng, Z., Gao, G., Wang, Y., Jin, Q., Liu, M., Shu, Y., 2014. Clinical and epidemiological characteristics of a fatal case of avian influenza A H10N8 virus infection: a descriptive study. Lancet 383, 714–721. http://dx.doi.org/10.1016/S0140-6736(14)60111-2. Choi, Y.K., Ozaki, H., Webby, R.J., Webster, R.G., Peiris, J.S., Poon, L., Butt, C., Leung, Y. H.C., Guan, Y., 2004. Continuing evolution of H9N2 influenza viruses in Southeastern ChinaJ. Virol. 78, 8609–8614. http://dx.doi.org/10.1128/JVI.78.16.86098614.2004. Dankar, S., Miranda, E., Forbes, N., Pelchat, M., Tavassoli, A., Selman, M., Ping, J., Jia, J., Brown, E., 2013. Influenza A/Hong Kong/156/1997(H5N1) virus NS1 gene mutations F103L and M106I both increase IFN antagonism, virulence and cytoplasmic localization but differ in binding to RIG-I and CPSF30. Virol. J. 10, 243. de Jong, M., Simmons, C., Thanh, T., Hien, V., Smith, G., Chau, T., Hoang, D., Chau, N., Khanh, T., Dong, V., Qui, P., Cam, B., Ha, Q., do, Y., Guan, J., Peiris, N., Chinh, T., Hien, Farrar, J., 2006. Fatal outcome of human influenza A (H5N1) is associated with high viral load and hypercytokinemia. Nat. Med. 12, 1203–1207. Diamond, M.S., Farzan, M., 2013. The broad-spectrum antiviral functions of IFIT and IFITM proteins. Nat. Rev. Immunol. 13, 46–57. http://dx.doi.org/10.1038/ nri3344.

48

V. Westenius et al. / Virology 471-473 (2014) 38–48

Ehrhardt, C., Seyer, R., Hrincius, E.R., Eierhoff, T., Wolff, T., Ludwig, S., 2010. Interplay between influenza A virus and the innate immune signaling. Microb. Infect. 12, 81–87. http://dx.doi.org/10.1016/j.micinf.2009.09.007. Fouchier, R., Schneeberger, P., Rozendaal, F., Broekman, J., Kemink, S., Munster, V., Kuiken, T., Rimmelzwaan, G., Schutten, M., Van Doornum, G., Koch, G., Bosman, A., Koopmans, M., Osterhaus, A., 2004. Avian influenza A virus (H7N7) associated with human conjunctivitis and a fatal case of acute respiratory distress syndrome. Proc. Natl. Acad. Sci. USA 101, 1356–1361. Gao, R., Cao, B., Hu, Y., Feng, Z., Wang, D., Hu, W., Chen, J., Jie, Z., Qiu, H., Xu, K., Xu, X., Lu, H., Zhu, W., Gao, Z., Xiang, N., Shen, Y., He, Z., Gu, Y., Zhang, Z., Yang, Y., Zhao, X., Zhou, L., Li, X., Zou, S., Zhang, Y., Li, X., Yang, L., Guo, J., Dong, J., Li, Q., Dong, L., Zhu, Y., Bai, T., Wang, S., Hao, P., Yang, W., Zhang, Y., Han, J., Yu, H., Li, D., Gao, G.F., Wu, G., Wang, Y., Yuan, Z., Shu, Y., 2013. Human infection with a novel avian-origin influenza A (H7N9) virus. N. Engl. J. Med. 368, 1888–1897. http://dx.doi.org/10.1056/NEJMoa1304459. Guan, Y., Shortridge, K., Krauss, S., Webster, R., 1999. Molecular characterization of H9N2 influenza viruses: were they the donors of the internal genes of H5N1 viruses in Hong Kong. Proc. Natl. Acad. Sci. USA 96, 9363–9367. Guan, Y., Shortridge, K., Krauss, S., Chin, P., Dyrting, K., Ellis, T., Webster, R., Peiris, M., 2000. H9N2 influenza viruses possessing H5N1-like internal genomes continue to circulate in poultry in southeastern China. J. Virol. 74, 9372–9380. http://dx.doi.org/10.1128/JVI.74.20.9372-9380.2000. Hale, B., Randall, R., Ortín, J., Jackson, D., 2008. The multifunctional NS1 protein of influenza A viruses. J. Gen. Virol. 89, 2359. Haller, O., Kochs, G., 2011. Human MxA protein: an interferon-induced dynaminlike GTPase with broad antiviral activity. J Interferon Cytokine Res. 31, 79–87. http://dx.doi.org/10.1089/jir.2010.0076. Hsiang, T., Zhao, C., Krug, R.M., 2009. Interferon-induced ISG15 conjugation inhibits influenza A virus gene expression and replication in human cells. J. Virol. 83, 5971–5977. http://dx.doi.org/10.1128/JVI.01667-08. Huang, R., Liu, J., Liang, W., Wang, A., Liu, Z., Yang, Y., Lv, J., Bao, Y., Gao, Y., Miao, Z., Chai, T., 2013. Response profiles of cytokines and chemokines against avian H9N2 influenza virus within the mouse lung. Med. Microbiol. Immunol. , http: //dx.doi.org/10.1007/s00430-013-0317-y. Jacobs, M.D., Harrison, S.C., 1998. Structure of an IκBα/NF-κB complex. Cell 95, 749–758. http://dx.doi.org/10.1016/S0092-8674(00)81698-0. Latvala, S., Pietila, T., Veckman, V., Kekkonen, R., Tynkkynen, S., Korpela, R., Julkunen, I., 2008. Potentially probiotic bacteria induce efficient maturation but differential cytokine production in human monocyte-derived dendritic cells. World J. Gastroenterol. 14, 5570–5583. Lee, D., Yuk, S., Park, J., Kwon, J., Erdene-Ochir, T., Noh, J., Yu, S., Hwang, S., Lee, S., Song, C., 2013. Innate immune response gene expression profiles in specific pathogen-free chickens infected with avian influenza virus subtype H9N2. BioChip J. 7, 393–398. http://dx.doi.org/10.1007/s13206-013-7411-5. Lin, Y., Shaw, M., Gregory, V., Cameron, K., Lim, W., Klimov, A., Subbarao, K., Guan, Y., Krauss, S., Shortridge, K., Webster, R., Cox, N., Hay, A., 2000. Avian-to-human transmission of H9N2 subtype influenza A viruses: relationship between H9N2 and H5N1 human isolates. Proc. Natl. Acad. Sci. USA 97, 9654–9658. Liu, D., Shi, W., Shi, Y., Wang, D., Xiao, H., Li, W., Bi, Y., Wu, Y., Li, X., Yan, J., Liu, W., Zhao, G., Yang, W., Wang, Y., Ma, J., Shu, Y., Lei, F., Gao, G., 2013. Origin and diversity of novel avian influenza A H7N9 viruses causing human infection: phylogenetic, structural, and coalescent analyses. Lancet 381, 1926–1932. http: //dx.doi.org/10.1016/S0140-6736(13)60938-1. Liu, X., Chen, W., Wei, B., Shan, Y., Wang, C., 2011. IFN-induced TPR protein IFIT3 potentiates antiviral signaling by bridging MAVS and TBK1. J. Immunol. 187, 2559–2568. http://dx.doi.org/10.4049/jimmunol.1100963. Ma, W., Brenner, D., Wang, Z., Dauber, B., Ehrhardt, C., Högner, K., Herold, S., Ludwig, S., Wolff, T., Yu, K., Richt, J., Planz, O., Pleschka, S., 2010. The NS segment of an H5N1 highly pathogenic avian influenza virus (HPAIV) is sufficient to alter replication efficiency, cell tropism, and host range of an H7N1 HPAIV. J. Virol. 84, 2122–2133. http://dx.doi.org/10.1128/JVI.01668-09. Marié, I., Durbin, J., Levy, D., 1998. Differential viral induction of distinct interferonalpha genes by positive feedback through interferon regulatory factor-7. EMBO J. 17, 6660–6669. http://dx.doi.org/10.1093/emboj/17.22.6660. Matikainen, S., Ronni, T., Hurme, M., Pine, R., Julkunen, I., 1996. Retinoic acid activates interferon regulatory factor-1 gene expression in myeloid cells. Blood 88, 114–123. Matrosovich, M.N., Krauss, S., Webster, R.G., 2001. H9N2 influenza A viruses from poultry in Asia have human virus-like receptor specificity. Virology 281, 156–162. http://dx.doi.org/10.1006/viro.2000.0799. Melén, K., Kinnunen, L., Fagerlund, R., Ikonen, N., Twu, K., Krug, R., Julkunen, I., 2007. Nuclear and nucleolar targeting of influenza A virus NS1 protein: striking differences between different virus subtypes. J. Virol. 81, 5995–6006. http://dx. doi.org/10.1128/JVI.01714-06. Min, J., Krug, R., 2006. The primary function of RNA binding by the influenza A virus NS1 protein in infected cells: inhibiting the 20 –50 oligo (A) synthetase/RNase L pathway. Proc. Natl. Acad. Sci. USA 103, 7100–7105. Min, J., Li, S., Sen, G.C., Krug, R.M., 2007. A site on the influenza A virus NS1 protein mediates both inhibition of PKR activation and temporal regulation of viral RNA synthesis. Virology 363, 236–243. http://dx.doi.org/10.1016/j.virol.2007.01.038.

Moltedo, B., López, C., Pazos, M., Becker, M., Hermesh, T., Moran, T., 2009. Cutting edge: stealth influenza virus replication precedes the initiation of adaptive immunity. J. Immunol. 183, 3569–3573. http://dx.doi.org/10.4049/jimmunol.0900091. Österlund, P., Veckman, V., Sirén, J., Klucher, K., Hiscott, J., Matikainen, S., Julkunen, I., 2005. Gene expression and antiviral activity of alpha/beta interferons and interleukin-29 in virus-infected human myeloid dendritic cells. J. Virol. 79, 9608–9617. http://dx.doi.org/10.1128/JVI.79.15.9608-9617.2005. Österlund, P., Strengell, M., Sarin, L., Poranen, M., Fagerlund, R., Melén, K., Julkunen, I., 2012. Incoming influenza A virus evades early host recognition, while influenza B virus induces interferon expression directly upon entry. J. Virol. 86, 11183–11193. Österlund, P.I., Pietilä, T.E., Veckman, V., Kotenko, S.V., Julkunen, I., 2007. IFN regulatory factor family members differentially regulate the expression of type III IFN (IFN-λ) genes. J. Immunol. 179, 3434–3442. Peiris, M., Yuen, K., Leung, C., Chan, K., Ip, P., Lai, R., Orr, W., Shortridge, K., 1999. Human infection with influenza H9N2. Lancet 354, 916–917. http://dx.doi.org/ 10.1016/S0140-6736(99)03311-5. Pichlmair, A., Schulz, O., Tan, C.P., Näslund, T.I., Liljeström, P., Weber, F., Reis e Sousa, C., 2006. RIG-I-mediated antiviral responses to single-stranded RNA bearing 50 phosphates. Science 314, 997–1001. http://dx.doi.org/10.1126/science.1132998. Pietilä, T., Latvala, S., Österlund, P., Julkunen, I., 2010. Inhibition of dynamindependent endocytosis interferes with type III IFN expression in bacteriainfected human monocyte-derived DCs. J. Leukoc. Biol. 88, 665–674. http://dx. doi.org/10.1189/jlb.1009651. Pothlichet, J., Chignard, M., Si-Tahar, M., 2008. Cutting edge: innate immune response triggered by influenza A virus is negatively regulated by SOCS1 and SOCS3 through a RIG-I/IFNAR1-dependent pathway. J. Immunol. 180, 2034–2038. Ronni, T., Melén, K., Malygin, A., Julkunen, I., 1993. Control of IFN-inducible MxA gene expression in human cells. J. Immunol. 150, 1715–1726. Ronni, T., Matikainen, S., Sareneva, T., Melén, K., Pirhonen, J., Keskinen, P., Julkunen, I., 1997. Regulation of IFN-alpha/beta, MxA, 20 ,50 -oligoadenylate synthetase, and HLA gene expression in influenza A-infected human lung epithelial cells. J. Immunol. 158, 2363–2374. Sallusto, F., Lanzavecchia, A., 1994. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J. Exp. Med. 179, 1109–1118. http://dx.doi.org/10.1084/ jem.179.4.1109. Seo, S.H., Webby, R., Webster, R.G., 2004. No apoptotic deaths and different levels of inductions of inflammatory cytokines in alveolar macrophages infected with influenza viruses. Virology 329, 270–279. http://dx.doi.org/10.1016/j. virol.2004.08.019. Spesock, A., Malur, M., Hossain, M., Chen, L., Njaa, B., Davis, C., Lipatov, A., York, I., Krug, R., Donis, R., 2011. The virulence of 1997 H5N1 influenza viruses in the mouse model is increased by correcting a defect in their NS1 proteins. J. Virol. 85, 7048–7058. http://dx.doi.org/10.1128/JVI.00417-11. Sutejo, R., Yeo, D., Myaing, M., Hui, C., Xia, J., Ko, D., Cheung, P., Tan, B., Sugrue, R., 2012. Activation of type I and III interferon signalling pathways occurs in lung epithelial cells infected with low pathogenic avian influenza viruses. PLoS One 7, e33732. http://dx.doi.org/10.1371/journal.pone.0033732. Turan, K., Mibayashi, M., Sugiyama, K., Saito, S., Numajiri, A., Nagata, K., 2004. Nuclear MxA proteins form a complex with influenza virus NP and inhibit the transcription of the engineered influenza virus genome. Nucleic Acids Res. 32, 643–652. http://dx.doi.org/10.1093/nar/gkh192. Veckman, V., Miettinen, M., Pirhonen, J., Sirén, J., Matikainen, S., Julkunen, I., 2004. Streptococcus pyogenes and Lactobacillus rhamnosus differentially induce maturation and production of Th1-type cytokines and chemokines in human monocyte-derived dendritic cells. J. Leukoc. Biol. 75, 764–771. Vignali, D.A., Kuchroo, V.K., 2012. IL-12 family cytokines: immunological playmakers. Nat. Immunol. 13, 722–728. http://dx.doi.org/10.1038/ni.2366. WHO. 2014. Influenza at the human-animal interface (HAI). 〈http://www.who.int/ influenza/human_animal_interface/en/〉. Wan, H., Sorrell, E., Song, H., Hossain, M.J., Ramirez-Nieto, G., 2008. Replication and transmission of H9N2 influenza viruses in ferrets: evaluation of pandemic potential. PLoS ONE 3, e2923. http://dx.doi.org/10.1371/journal.pone.0002923. Wang, Z., Robb, N., Lenz, E., Wolff, T., Fodor, E., Pleschka, S., 2010. NS reassortment of an H7-type highly pathogenic avian influenza virus affects its propagation by altering the regulation of viral RNA production and antiviral host response. J. Virol. 84, 11323–11335. http://dx.doi.org/10.1128/JVI.01034-10. Ward, C.L., Dempsey, M.H., Ring, C.J.A., Kempson, R.E., Zhang, L., Gor, D., Snowden, B.W., Tisdale, M., 2004. Design and performance testing of quantitative real time PCR assays for influenza A and B viral load measurement. J. Clin. Virol. 29, 179–188. http://dx.doi.org/10.1016/S1386-6532(03)00122-7. Xing, Z., Cardona, C., Adams, S., Yang, Z., Li, J., Perez, D., Woolcock, P., 2009. Differential regulation of antiviral and proinflammatory cytokines and suppression of Fasmediated apoptosis by NS1 of H9N2 avian influenza virus in chicken macrophages. J. Gen. Virol. 90, 1109–1118. http://dx.doi.org/10.1099/vir.0.007518-0. Xing, Z., Harper, R., Anunciacion, J., Yang, Z., Gao, W., Qu, B., Guan, Y., Cardona, C., 2011. Host immune and apoptotic responses to avian influenza virus H9N2 in human tracheobronchial epithelial cells. Am. J. Respir. Cell Mol. Biol. 44, 23–33. http://dx.doi.org/10.1165/rcmb.2009-0120OC.