NLRC5 interacts with RIG-I to induce a robust antiviral response against influenza virus infection

NLRC5 expression impairs influenza A virus replication in respiratory epithelial cells

To evaluate the role of NLRC5 in influenza virus infection, we transfected the lung epithelial cell line A549 with myc-NLRC5 expression or human flag-CIITA expression vector as A549 cells express only very low levels of NLRC5 (Fig. 1). After 24 h, cells were infected with different doses of PR8 (multiplicity of infection (MOI) 0.01, 0.1, and 1.0) for a further 24 h, which resulted in a dose-dependent increase in NS1 expression (Fig. 1G). NLRC5 overexpression significantly reduced PR8 replication compared with cells transfected with the empty vector at all infection doses used as determined by viral plaque assays (Fig. 1A). Interestingly, we observed a more robust inhibition of PR8 replication at lower MOI (0.01 and 0.1) as compared with an MOI of 1.0 in NLRC5-transfected cells. In contrast, overexpression of CIITA, an NLRC5-related NLR protein, did not inhibit virus replication as viral titers were comparable to that of control vector transfected cells (Fig. 1A). Accordingly, we observed a significant decrease in nucleoprotein (NP) vRNA and NP mRNA copy numbers in infected A549 that had been transfected with NLRC5 expression vector as compared with control vector or CIITA expression vector transfected A549 cells (Fig. 1B and C). Accordingly, NS1 levels were strongly reduced in NLRC5-expressing cells (Fig. 1G). Host defense against viral infection is associated with type I IFN induction. Interestingly, NLRC5 overexpression (Fig. 1G) enhanced mRNA expression of IFN-β in A549 cells infected with PR8 (Fig. 1D). Of note, we also observed a significant increase in RIG-I expression in NLRC5-transfected cells (Fig. 1E), which we and others recently identified as an important factor to restrict influenza replication in human cells [26–30]. Increase in IFN-β and RIG-I in NLRC5-transfected cells was observed only in response to PR8 infection, which reached a maximum at 0.1 MOI infection dose. This was also observed for IFN-α, mRNA expression, another type I IFN (Fig. 1F). By contrast, PR8-induced expression of IFN-β, RIG-I, and IFN-α was not changed in A549 cells expressing CIITA (Fig. 1G).

We carried out similar studies in normal human bronchial epithelial (NHBE) cells. As shown in Supporting Information Figure 1, NHBE cells transfected with NLRC5 expression vector showed significant reduction in viral titer as compared with vector control upon PR8 infection (MOI 1.0) (Supporting Information Fig. 1A). This NLRC5-mediated inhibition of PR8 replication in NHBE cells was consistent with a reduction in NP vRNA (Supporting Information Fig. 1B). NLRC5 expression also induced IFN-β (Supporting Information Fig. 1D) and RIG-I mRNA expression (Supporting Information Fig. 1C) in PR8-infected NHBE cells. NLRs have also been reported to upregulate expression of many proinflammatory cytokines including CCL5 (RANTES) [32, 33]. Previously, we have shown that NLRC5 affects RANTES and IFN-β secretion in PMA-differentiated THP-1 cells and primary human dermal fibroblasts following Sendai virus infection and polyI:C stimulation [13]. Analysis of IFN-β and RANTES in NHBE culture supernatants by ELISA indicates that NLRC5 overexpression also resulted in increased IFN-β and RANTES levels upon PR8 infection (Supporting Information Fig. 1E and F). These observations were further supported by immunohistochemistry where we observed a significant reduction in NP-positive NHBE cells, transfected with NLRC5 expression vector (Supporting Information Fig. 1G and H).

Contrary to our findings in A549 and NHBE cells, overexpression of NLRC5 in HEK293T cells did not affect PR8 replication irrespective of the virus dose and viral titers (Supporting Information Fig. 2A and B). Immunoblot analysis of NS1 expression confirmed no change in NS1 expression level in vector- and NLRC5-transfected HEK293T cells (Supporting Information Fig. 2A). Furthermore, IFN-β promoter activity was not significantly different in NLRC5-transfected cells following PR8 infection compared with cells transfected with the empty vector (Supporting Information Fig. 2C).

These results suggest that NLRC5 plays a role in the induction of antiviral innate immune responses against influenza virus infection and that NLRC5 contributes to the restriction of influenza replication in human respiratory epithelial cells, but not in HEK293T cells.

Viral NS1 counteracts NLRC5, RIG-I, and IFN-β expression

NLRC5 expression has been shown to be induced upon viral infection in different human cell types [13, 16, 19]. Therefore, we investigated the level of endogenous NLRC5 mRNA in A549 cells in response to PR8 infection. The kinetic of virus infection was studied by quantitative RT-PCR of NP vRNA and NP mRNA, which showed a time-dependent increase in their copy numbers (Fig. 2A and B). Further, quantitative RT-PCR studies showed that PR8 infection induced NLRC5 mRNA expression in A549 cells within 3 h of infection (Fig. 2C). However, NLRC5 expression declined thereafter and reached nearly basal levels by 12 h of infection. Similar kinetics were observed for RIG-I mRNA expression (Fig. 2D). We observed only a marginal increase in IFN-β expression at 6 h following PR8 infection (Fig. 2E). Expression of many PRRs has been shown to be induced by viral PAMPs as well as cytokines [34] but viruses have devised potent strategies to counteract PRR expression, recognition, activation, and subsequent cytokine production. Influenza NS1 has been shown to suppress RIG-I activation and subsequent IFN-β induction [30, 35–38]. To determine the effect of NS1 on virus-induced NLRC5 expression PR8ΔNS1, a mutant of influenza virus PR8 that lacks functional NS1 (PR8ΔNS1) was used. Unlike WT PR8 virus, infection with PR8ΔNS1 resulted in a strong IFN-β expression in a time-dependent manner (Fig. 2E), a concordant increase in RIG-I (Fig. 2D) and NLRC5 mRNA (Fig. 2C) levels was also observed. Very low PR8ΔNS1 NP vRNA and NP mRNA expression levels confirmed replication deficiency of PR8ΔNS1 in A549 cells (Fig. 2A and B). To provide further evidence that NS1 suppresses NLRC5 induction and antiviral response, we complemented viral NS1 expression by transfecting a myc-NS1 expression vector into A549 cells prior to infection with PR8ΔNS1 (Fig. 3D). Real-time RT-PCR revealed that expression of NS1 in the host cells significantly (p < 0.05) inhibited NLRC5 (Fig. 3A) and RIG-1 mRNA (Fig. 3B) and protein (Fig. 3D) expression upon viral infection. LUC reporter assays of the same cells revealed reduced IFN-β promoter activity (Fig. 3C) in response to PR8ΔNS1 compared with cells not expressing myc-NS1.

Influenza virus mediated induction of NLRC5 and type-I IFN requires RIG-I

RIG-I has a well-established role in inhibiting influenza virus replication [26, 28, 36–38]. We observed a significant increase in RIG-I expression in response to NLRC5 overexpression in A549 as well as in NHBE cells upon influenza virus infection (Fig. 1E and Supporting Information Fig. 1C). Furthermore, infection with PR8ΔNS1 induced RIG-I expression (Fig. 2D). This likely is mediated by the IFN pathway as RIG-I is well known to be IFN inducible [39]. To delineate the specific role of NLRC5 and RIG-I in the antiviral response, we applied a knockdown approach using gene-specific siRNAs in A549 cells. A549 cells were transfected with NLRC5 or RIG-I siRNA for 24 h, and subsequently infected with PR8 or PR8ΔNS1 for another 24 h. Knockdown efficiency of NLRC5 and RIG-I was determined by quantitative RT-PCR (Fig. 4A and C) as well as by Western blot (Fig. 4D). Knockdown of RIG-I expression completely abrogated the induction of NLRC5 and IFN-β mRNA (Fig. 4A, B, and D). Analysis of RIG-I expression showed that NLRC5 knockdown did not affect RIG-I mRNA or protein expression as compared with control siRNA treated A549 cells infected with PR8ΔNS1 (Fig. 4C and D). In contrast, reduction of NLRC5 mRNA levels partially, but significantly (p < 0.03), inhibited influenza virus induced IFN-β (Fig. 4B), as shown recently for Sendai virus [13] and CMV [16] in other cell types. In any case, control siRNA did not alter the expression level of NLRC5 (Fig. 4A and D) or RIG-I (Fig. 4C and D). Release of IFN-β and RANTES also was fully blunted in RIG-I-specific siRNA-treated cells and partially reduced in cells that received NLRC5-specific siRNA (Fig. 4E and F). The RIG-I-dependent antiviral pathway also involves NFκB activation [40], therefore, we investigated NFκB activation using NFκB gene reporter LUC assays. As shown in Figure 4G, infection with PR8ΔNS1, but not PR8, resulted in increased NFκB promoter activity. Knocking down RIG-I completely inhibited NFκB activation, while NLRC5 knockdown resulted in partial but significant (p < 0.05) reduction in NFκB promoter activation. To substantiate these results, we further tested two independent sets of control and NLRC5 specific siRNAs. As shown in Supporting Information Figure 3A, both of these NLRC5-targeting siRNAs, but not the control siRNAs, effectively reduced PR8ΔNS1-induced NLRC5 protein expression below the detection limit. In contrast, RIG-I protein expression was not affected by the control or NLRC5 siRNAs (Supporting Information Fig. 3A). IFN-β promoter LUC reporter assays confirmed that the NLRC5-specific siRNAs but not the control siRNAs efficiently reduced PR8ΔNS1-induced IFN-β activation (Supporting Information Fig. 3B).

Results of the above studies suggest that NLRC5 is involved in RIG-I-dependent antiviral responses against influenza virus. Influenza virus is a natural ligand for RIG-I that induces its activation [41, 42]. In order to establish the specificity of this response, we carried out studies to investigate if NLRC5 influences other innate immune responses induced by other PAMPs such as LPS. A549 cells either transfected with NLRC5 siRNA or NLRC5 expression vector and/or IFN-β or NFκB promoter LUC reporter plasmids were treated with PBS or LPS (0.1, 1.0, 10, or 100 ng/mL) for 24 h. As shown in Figure 5, LPS treatment resulted in a dose-dependent increase in IFN-β and NFκB promoter activation in A549 cells. NLRC5 knockdown strongly reduced LPS-induced NLRC5 protein levels (Fig. 5C). However, this neither significantly altered IFN-β nor NFκB promoter activation compared with cells treated with control siRNA (Fig. 5A and B), and this trend was consistent with all the doses of LPS used in the study. Similarly, overexpression of NLRC5 in A549 cells followed by LPS treatment also did not affect IFN-β or NFκB promoter activation (Fig. 5D and E). Figure 5F shows NLRC5 and RIG-I protein expression in these cells. These results provide additional support that NLRC5 is involved specifically in RIG-I-mediated antiviral responses in A549 cells.

We further extended these studies to primary mouse normal bronchial and tracheal epithelial cells (Supporting Information Fig. 4). To this end, we knocked down mouse NLRC5 using mouse gene specific NLRC5 siRNA and infected these cells with PR8 or PR8ΔNS1 for 6 h. We observed a strong NLRC5 and RIG-I expression in response to both PR8 and PR8ΔNS1 at this time point (Supporting Information Fig. 4A). NLRC5 siRNA effectively reduced NLRC5 expression without affecting RIG-I expression (Supporting Information Fig. 4A). Analysis of IFN-β and RANTES in culture supernatants 24 h postinfection by ELISA showed that NLRC5 knockdown significantly reduced IFN-β and RANTES level (Supporting Information Fig. 4B and C). These findings confirm that our results obtained in cell lines are transferrable to primary epithelial cells.

To characterize the role of NLRC5 in antiviral response in more detail, we analyzed the expression level of RIG-I and IFN-β in infected cells over time. A549 cells treated with control or NLRC5-specific siRNA were infected with PR8ΔNS1 for the indicated time and IFN-β, RIG-I, and NLRC5 mRNA expression was measured by quantitative RT-PCR (Fig. 6). NLRC5 knockdown resulted in a significant reduction of IFN-β expression at all time points (p < 0.03, 0.01, 0.01, and 0.008) (Fig. 6A). Interestingly, RIG-I mRNA expression was not different in control siRNA versus NLRC5 siRNA treated cells at 24 h postinfection (see also Fig. 4C), however was significantly lower in the absence of NLRC5 starting from 48 h postinfection (p < 0.04, 0.04, and 0.03) (Fig. 6B). In any case, NLRC5 mRNA was efficiently reduced by the siRNA treatment (Fig. 6C). These results strongly suggest that while NLRC5 expression is RIG-I dependent, NLRC5 contributes to mount a robust IFN-β induction in response to influenza infection. The type I IFN, IFN-β, is pivotal for the antiviral host defense. It is involved in the transcriptional activation of many PRRs including NLRC5 [11, 16], a large number of so-called IFN-stimulated genes as well as its own production. To examine whether NLRC5 expression is driven directly by IFN-β also upon influenza infection, we treated A549 with recombinant human IFN-β and analyzed the expression of IFN-β, NLRC5, and RIG-I mRNA. As shown in Supporting Information Figure 5A, recombinant IFN-β treatment not only induced the expression of IFN-β, but also the expression of NLRC5 and RIG-I. Dependency of NLRC5 expression on RIG-I-induced IFN-β was further confirmed in studies using neutralizing IFN-α/β antibodies, which prevented PR8ΔNS1-induced expression of RIG-I, NLRC5, and IFN-β (Supporting Information Fig. 5B).

Taken together, these results confirmed previous findings, showing that NLRC5 mRNA is regulated by type I IFNs and suggest that it is induced upon activation of RIG-I by influenza virus. More interestingly, we revealed that NLRC5 positively affects RIG-I signaling and that NLRC5 is required to maintain a long-lasting antiviral response toward influenza virus in human host cells.

The N-terminal and NACHT domains of NLRC5 are critical for its antiviral function

Next, we examined, which domains of NLRC5 contribute to the antiviral activity. To this end, myc-tagged wtNLRC5 and deletion mutants thereof [17] were expressed in A549 cells (Fig. 7A). As expected, transfection of A549 cells with NLRC5 significantly reduced influenza viral titers following infection with PR8 (Fig. 7B). By contrast, overexpression of a Walker A mutant of NLRC5 (NLRC5-K234A) or of a deletion construct that lacks the N-terminal domain (NLRC5-ΔDD), contain only the N-terminal death-fold domain (NLRC5-DD) or of the LRR domain of NLRC5 (NLRC5-LRR), all did not significantly affect viral titers (Fig. 7B). However, overexpression of isoform 3 of NLRC5 (NLRC5-ISO3), which lacks the LRR domain [13], inhibited PR8 replication to a similar extent as wtNLRC5 (Fig. 7B). These results indicate that NLRC5-mediated restriction of influenza virus replication is dependent on a functional ATPase domain and the presence of N-terminal domain of NLRC5. Consistent with these results, only wtNLRC5 and isoform 3 but not NLRC5-K234A, NLRC5-ΔDD, NLRC5-DD, or NLRC5-LRR induced IFN-β mRNA expression upon infection (Fig. 7C). Further, increased levels of RANTES were detectable in the supernatants of A549 cells transfected with wtNLRC5 and NLRC5-ISO3 but not in the supernatants from cells transfected with the other NLRC5 constructs (Fig. 7D).

Taken together, these data suggest that the DD and NACHT ATPase domain are sufficient for NLRC5-mediated antiviral responses toward influenza. Of note, a construct comprising only these domains is unable to promote MHC class I expression [17], functionally separating the role of NLRC5 in RIG-I-mediated antiviral immunity from its contribution to MHC-I-dependent antigen presentation.

NLRC5 forms a complex with RIG-I and influenza virus protein NS1

Viruses can suppress host innate immune responses. Influenza NS1 has been shown to interact directly with RIG-I and other PRRs to dampen host innate immune activation [28, 30, 35, 38, 43–46]. Our observation that the antiviral activity of NLRC5 is dependent on RIG-I prompted us to investigate if NLRC5 might act in a physical complex with RIG-I. To this end, we carried out coimmunoprecipitation studies. A549 cells, transfected either with vector or myc-tagged wtNLRC5, were infected with PR8 24 h post-transfection. Cells were harvested at the indicated time postinfection and protein expression of RIG-I, myc-NLRC5, and viral NS1 was determined by immunoblotting (Fig. 8A). In vector-transfected A549 cells, RIG-I protein expression was detected at 3 h post-PR8 infection and rapidly declined starting 6 h postinfection, and was undetectable after 9 h. Consistent with our quantitative RT-PCR data (Fig. 1E), NLRC5 overexpression increased and prolonged RIG-I expression that was still detectable up to 12 h p.i. Notably, NLRC5 expression also delayed the expression of NS1 in PR8-infected A549 cells. In vector-transfected cells, NS1 was detectable starting at 6 h following PR8 infection, whereas in NLRC5-transfected A549 cells, NS1 appeared only after 9 h postinfection and its expression appeared to be lower than in vector-transfected A549 cells, consistent with the lower viral titer found in these cells (Fig. 8A). To examine if NLRC5 formed a complex with endogenous RIG-I or viral NS1, coimmunoprecipitations were conducted at 6, 12, and 24 h postinfection using anti-myc, anti-RIG-I, and anti-NS1 antibodies to precipitate myc-NLRC5, RIG-I, and NS1, respectively. Precipitation of myc-NLRC5 copurified RIG-I at 6 h post-PR8 infection but not at the later time points. Notably, NS1 bound to NLRC5 at 12 and 24 h postinfection but not at 6 h, suggesting that NLRC5 can form a complex with NS1 and RIG-I upon infection at the time when these proteins are most highly expressed (Fig. 8A). Precipitation of RIG-I copurified NLRC5 at 6 h postinfection. Notably, at this time point, NS1 was absent in A549 cells (Fig. 8A). NS1 coprecipitated with RIG-I at 12 h post-infection, however, we did not detect NLRC5 in coimmunoprecipitate at this time point despite its presence in the cells (Fig. 8A). These data indicate that while RIG-I is capable of interacting with NLRC5, NS1 can prevent this interaction. Precipitation of NS1 using NS1-specific antibodies copurified RIG-I in vector-transfected cells 6 h postinfection. At this time point, NS1 was absent in NLRC5 transfected cells (Fig. 8A). NS1 copurified NLRC5 at 12 and 24 h postinfection suggesting NS1 can bind to NLRC5 regardless of the presence of absence of RIG-I (Fig. 8A).

Our domain mapping studies revealed that the antiviral activity of NLRC5 depended on the DD and NACHT domain. We next investigated if the N-terminal DD and the central NACHT domain also participate in the above observed interaction with RIG-I and NS1. To this end, we overexpressed myc-tagged wtNLRC5 or the NLRC5 constructs NLRC5-K234A, NLRC5-ISO3, NLRC5-ΔDD, NLRC5-DD, or NLRC5-LRR domain in A549 cells and subsequently infected these cells with PR8 influenza virus. Cell lysates were harvested at 3 and 9 h postinfection and were analyzed for the expression of NLRC5, RIG-I, and NS1 by immunoblotting (Fig. 8B). These time points were chosen due to the high abundance of RIG-I, myc-NLRC5, and/or NS1 at these times. RIG-I expression was well detectable at 3 h post-PR8 infection irrespective of the expression of the NLRC5 constructs, whereas NS1 was not detectable at that this time. As expected RIG-I was not detectable at 9 h postinfection in mock-treated cells, however it was readily detected in cells expressing wtNLRC5 or the ISO3 construct (Fig. 8B). In any case, NS1 was detectable at 9 h postinfection (Fig. 8B). Immunoprecipitation of NLRC5 showed that wtNLRC5 and NLRC5-ISO3 but also the N-terminal DD of NLRC5 interacted with RIG-I at 3 h post-PR8 infection. However, at 9 h postinfection, we did not detect RIG-I in any coimmunoprecipitation, although RIG-I was clearly present in cells expressing wtNLRC5 or ISO3 (Fig. 8B). Interestingly, NS1 coimmunoprecipitated with all the NLRC5 constructs 9 h postinfection, however not in control immunoprecipitations in both directions (Fig. 8B). Immunoprecipitation of RIG-I copurified wtNLRC5, NLRC5-ISO3, and NLRC5-DD. Notably, NLRC5-DD did not induce antiviral effects (Fig. 7). Immunoprecipitation of NS1 copurified RIG-I in wtNLRC5 and NLRC5-ISO3 expressing A549 cells at 9 h postinfection when RIG-I was present only in wtNLRC5 or NLRC5-ISO3 expressing A549 cells (Fig. 8B).

NLRC5 overexpression enhanced RIG-I expression in PR8-infected cells, and RIG-I was detectable up to 12 h by immunoblotting. In contrast, in vector-transfected cells, PR8-induced RIG-I was detectable only up to 6 h postinfection (Fig. 8A). This could be a result of enhanced transcription and/or translation of RIG-I or decreased RIG-I degradation. To understand the role of NLRC5 in enhanced RIG-I expression, we treated NLRC5-transfected A549 cells with actinomycin D and cyclohexamide to block transcription and translation, and subsequently infected these with PR8 (MOI 1.0). Cells were harvested at 0, 3, 6, and 9 h postinfection to analyze expression of RIG-I by immunoblotting. These data show that treatment with actinomycin D and cyclohexamide did not reduce the steady-state levels of RIG-I suggesting that NLRC5 affects the half-life of the RIG-I protein (Fig. 8C).

In summary, our results suggest that NLRC5 extends and stabilizes influenza virus induced RIG-I expression and delays NS1 expression. NS1 counteracts NLRC5-mediated enhancement of RIG-I activity by competing for binding to RIG-I and/or by interacting with NLRC5 thereby preventing the binding of NLRC5 to RIG-I. RIG-I interaction with NLRC5 required the NLRC5 DD, which was sufficient for binding, though a Walker A mutant of NLRC5 failed to interact with RIG-I despite the presence of DD.

Plasmids and reagents

Myc-tagged NLRC5 full-length (NP 115582; flag-NLRC5 or myc-NLRC5) and mutants (NLRC5-K234A, NLRC5-ISO3, NLRC5-ΔDD, NLRC5-DD, and NLRC5-LRR) are described previously [13]. Flag-tagged CIITA was a kind gift from Victor Steimle, Sherbrook University, Canada. The myc-NS1 expression vector used in this study has been described previously [35]. NFκB and IFN LUC reporter plasmid was obtained from Dr. Rongtuan Lin, McGill University, Canada. Anti-RIG-I and anti-NS1 antibodies were purchased from Santa Cruz Biotechnology (CA, USA). Actinomycin D, cyclohexamide, anti-flag, anti-myc, and anti-β-actin antibodies were purchased from Sigma Aldrich (St. Louis, USA). Anti-CIITA and anti-IFN-α/β neutralizing antibodies were purchased from Millipore (MA, USA). Anti-NLRC5 monoclonal antibody 3H8 was described previously [13]. The anti-influenza A virus NP antibody and laboratory strain A/Puerto Rico/8/34 [(PR8); H1N1] were obtained from the Influenza Reagents Repository, Centers for Disease Control and Prevention (CDC), Atlanta, USA; the NS1-del mutant (PR8ΔNS1) was provided by Adolfo García-Sastre, Mount Sinai School of Medicine, NY, USA. Human and mouse IFN-β and CCL5 ELISA kits were from R&D system Inc. (MN, USA).

Cell cultures and virus infection

Human lung epithelial cell line A549, HEK293T (ATCC, VA, USA), and NHBE cells (Lonza, Switzerland) were maintained as described [13, 26]. Mouse normal bronchial and tracheal epithelial cells were purchased from (CHI Scientific, MA, USA) and maintained as per manufacturer instructions. 10⁶ cells in six-well plates were transfected with the empty vector or NLRC5 expression vectors for 24 h. Cells were then infected with PR8 or PR8ΔNS1 virus at an MOI of 1.0 or 1.0 focal forming units with trypsin supplement as described previously [26]. Twenty-four hours postinfection, cells were harvested for RNA and protein analysis and cell-culture supernatants were collected and stored at −80°C for determination of viral titer by plaque assay as described previously using MDCK cells [26]. Three independent experiments were performed at different times with each treatment carried out in duplicate cultures.

siRNA knockdown studies

Gene-specific siRNAs to silence NLRC5 [13] or RIG-I [26] in A549 or NHBE or mouse normal bronchial and tracheal epithelial cells were purchased from Thermo Fisher Scientific (Lafayette, CO, USA), and were used as described in the manufacturer’s protocol. Briefly, cells were plated at a density of 10⁶/well in a six-well plate and transfected with 75 nM each of the gene-specific siRNA 24 h prior to infection with PR8 or PR8-NS1-del virus. After 24 h, transfected cells were harvested and analyzed for NLRC5, RIG-I, IFN-β, or NS1 expression by real-time PCR or Western blot.

Real-time RT-PCR

Total RNA was isolated from cells using the RNAeasy kit (Qiagen, Valencia, CA, USA) and real-time RT-PCR was conducted using a Stratagene Mx3005P Q-PCR machine for mRNA expression of NLRC5, RIG-I, IFN-β, IFN-α NP vRNA, RANTES, and β-actin. For each sample, 2 μg of RNA was reverse-transcribed using Superscript II Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. Parallel reactions without reverse transcriptase were included as negative controls. Reverse transcription reactions were analyzed using syber green Q-PCR reagents (Sigma Aldrich, St. Louis, USA). PCR conditions were as follows: 95°C for 15 s, annealing at 57 °C for 30 s, and extension at 72 °C for 30 s for a total of 45 cycles. The threshold cycle number for cDNA was normalized to that of βactin mRNA, and the resulting value was converted to a linear scale. Data from three independent experiments were used for analysis. Primer sets used for these studies have been described previously [13, 26, 28].

Immunoblotting and coimmunoprecipitation

Cells plated in 100-mm tissues culture plates were harvested in RIPA buffer (Sigma Aldrich), and cell lysates were incubated with primary antibody overnight at 4°C followed by incubation with protein A Dynabeads (Invitrogen) for 2 h. The beads were washed three times with PBS, suspended in Laemmli buffer (62.5 mM Tris-HCl pH 6.8, 25% glycerol, 2% SDS, 0.01% Bromophenol blue), boiled for 10 min, and centrifuged at 12 000 × g for 10 min at 4°C. Supernatants were collected and analyzed by Western blotting as described previously [26].

LUC assay

A549 or NHBE cells were cotransfected with IFN-β or NFκB promoter LUC reporter plasmid and harvested at indicated time points for LUC activity using dual LUC assay kit (Promega, WI, USA) as per manufacturer’s instructions. Mean and SDs were calculated from duplicate or triplicate cultures and are representative of at least three independent experiments.

Confocal microscopy

A549 or NHBE cells were cultured on collagen-coated glass cover slips and transfected with vector or wtNLRC5 using lipofectamine as described previously [28] followed by PR8 infection. Cells on the cover slips were washed with cold PBS and fixed with 4% paraformaldehyde for 10 min. Cells were permeabilized in 0.1% Triton X-100 in PBS and blocked with 1% BSA and 5% normal goat serum in PBS. Following blocking, cover slips were incubated overnight with anti-myc mouse monoclonal antibodies (1:1000 dilution), and anti-influenza A virus NP mouse monoclonal antibody (1:2 500). Alexa 488 conjugated anti-mouse IgG (LifeTechnologies, USA) was used as the secondary antibody. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Cells weremounted on slides with Prolong Antifade Mounting Media (Molecular Probes). Images were examined using an LSM 710 inverted confocal microscope (Zeiss, Oberkochen, Germany). For calculating NP-positive cells, five independent ×20 image fields with approximately 100 cells/field from each condition were evaluated. The total number of NP-positive cells was divided by the total number of DAPI positive cells for each field to estimate the percent of NP positive cells. Data are represented as the percentage ± SD of five ×20 fields from each experimental condition.

Statistical analysis

To determine the statistical significance, we used analysis of variance (ANOVA) using GraphPad PRISM 5 and a value of- p ≤ 0.05 was considered significant when compared with respective controls.