IRF7 Is Required for the Second Phase Interferon Induction During Inﬂuenza Virus Infection in Human Lung Epithelia

viruses

Article IRF7 Is Required for the Second Phase Interferon Induction during Inﬂuenza Virus Infection in Human Lung Epithelia

Wenxin Wu 1,*, Wei Zhang 1, Lili Tian 1, Brent R. Brown 1, Matthew S. Walters 1 and Jordan P. Metcalf 1,2,3,* 1 Pulmonary, Critical Care & Sleep Medicine, Department of Medicine, the University of Oklahoma Health Sciences Center, Oklahoma City, OK 73104, USA; [email protected] (W.Z.); [email protected] (L.T.); [email protected] (B.R.B.); [email protected] (M.S.W.) 2 Department of Microbiology and Immunology, the University of Oklahoma Health Sciences Center, Oklahoma City, OK 73104, USA 3 Veterans Aﬀairs Medical Center, Oklahoma City, OK 73104, USA * Correspondence: [email protected] (W.W.); [email protected] (J.P.M.)

Received: 10 January 2020; Accepted: 27 March 2020; Published: 29 March 2020

Abstract: Influenza A virus (IAV) infection is a major cause of morbidity and mortality. Retinoic acid-inducible protein I (RIG-I) plays an important role in the recognition of IAV in most cell types, and leads to the activation of interferon (IFN). We investigated mechanisms of RIG-I and IFN induction by IAV in the BCi-NS1.1 immortalized human airway basal cell line and in the A549 human alveolar epithelial cell line. We found that the basal expression levels of RIG-I and regulatory transcription factor (IRF) 7 were very low in BCi-NS1.1 cells. IAV infection induced robust RIG-I and IRF7, not IRF3, expression. siRNA against IRF7 and mitochondrial antiviral-signaling protein (MAVS), but not IRF3, significantly inhibited RIG-I mRNA expression and IFN induction by IAV infection. Most importantly, even without virus infection, IFN-β alone induced RIG-I, and siRNA against IRF7 did not inhibit RIG-I induction by IFN-β. Similar results were found in the alveolar basal epithelial A549 cell line. RIG-I and IRF7 expression in humans is highly inducible and greatly amplified by IFN produced from virus infected cells. IFN induction can be separated into two phases, that initially induced by the virus with basal RIG-I (the first phase), and that induced by the subsequent virus with amplified RIG-I from the first phase IFN (the second phase). The de novo synthesis of IRF7 is required for the second phase IFN induction during influenza virus infection in human lung bronchial and alveolar epithelial cells.

Keywords: RIG-I; IRF3; IRF7; TLR3; interferon; inﬂuenza; A549; human; lung; innate immunity

1. Introduction Influenza A virus (IAV) infection is a major cause of morbidity and mortality. In 2009, a pandemic caused by the novel H1N1 IAV infected over 300,000 individuals with at least 16,000 confirmed deaths worldwide [1]. The 2017–2018 flu season has been the worst season in the US since 2009. It is estimated that the burden of illness during the 2017–2018 season was high with an estimated 45 million people getting sick with influenza, 21 million people going to a health care provider, 810,000 hospitalizations, and 61,000 deaths from influenza (from 2017–2018 Estimated Influenza Illnesses, Medical visits, Hospitalizations, and Deaths and Estimated Influenza Illnesses, Medical visits, Hospitalizations, and Deaths Averted by Vaccination in the United States, Centers for Disease Control and Prevention). The innate immune system recognizes and responds to pathogens in a non-specific manner and provides immediate protection against infection. Induction of interferon (IFN) is a critical component

Viruses 2020, 12, 377; doi:10.3390/v12040377 www.mdpi.com/journal/viruses Viruses 2020, 12, 377 2 of 12 of the host innate immune response to influenza virus infection. IFNs are further divided into type I (mainly IFN-α and β), II (IFN-γ) and III (IFN-λ) subtypes, based in part on the differential use of unique receptors through which they mediate signal transduction to induce anti-viral activity [2]. The IFN responses to IAV are triggered by the recognition of pathogen-associated molecular patterns (PAMPs) by host internal or cell surface pattern recognition receptors (PRRs), including retinoic acid-inducible protein I (RIG-I) and Toll-like receptor 3 (TLR3) [3]. RIG-I is essential for IFN induction during RNA virus infections of non-plasmacytoid dendritic cell (pDC) cell types [4]. TLR3, a double-stranded RNA sensor, may be used by some epithelial cells to detect the viral replicative intermediate dsRNA [5]. Mitochondrial antiviral-signaling protein (MAVS), the intermediary protein of RIG-I, forms protease resistant prion-like aggregates that activate specific signaling pathways leading to activation of NF-κB and induction of the interferon regulatory transcription factors (IRF)3 and IRF7. NF-κB is crucial for inflammatory cytokine induction. IRF3 and IRF7 are then translocated into the nucleus where they act as transcription factors for the production of type I and III IFNs [6,7]. We have previously shown that both RIG-I and TLR3 are important for IFN induction by IAV in human lung alveolar epithelial cells (AEC) [3]. We also developed a human lung tissue model using precision-cut slices to study the local lung response to IAV, and we showed that RIG-I is critical for the initiation of the early antiviral cytokine response in the human lung [8]. IFN responses are the central component of the innate immune system’s control of viral infection [9]. Once released, type I and III IFNs bind to their respective receptors (IFNAR and IFNLR1/IL10R2, respectively) on the same and neighboring cells. When an IFN interacts with its cognate receptor, a signal is rapidly transmitted within the cell. The primary signal transduction cascade promoted by IFNs is mediated by the Janus family of protein tyrosine kinase 1 (JAK1) signal transducers and activators of the transcription (STAT) pathway [10]. Receptor engagement subsequently leads to the activation of the IFN-stimulated regulatory factor 3 (ISGF3) transcription complex. ISGF3 is composed of STAT1 and STAT2, both of which are activated by JAK1, and IRF9 [7]. The activation of this transcriptional activator complex leads to increased expression of interferon-stimulated genes (ISGs), including 20,50-oligoadenylate synthetase (OAS), Mx proteins, and protein kinase R (PKR), inducing an antiviral state [11]. RIG-I and IRF7 are also ISGs that are up-regulated and further amplify the entire antiviral immune system [12]. The basal expression level of RIG-I is very low in human lung. However, RIG-I expression is highly inducible and greatly amplified by first phase IFN production from the viral infected cells in an autocrine and paracrine fashion [13,14]. Therefore, there are two phases of IFN induction after IAV infection. The first phase IFN is induced when the viral genome is recognized by basal level RIG-I in the IAV-infected cells. The secreted first phase IFN proteins then bind to its cognate cellular receptor and induce vast amounts of RIG-I and IRFs in the same and neighboring cells. The significantly elevated RIG-I and IRF proteins turn the infected and uninfected neighboring cells into an “immunoactive” state primed to respond to subsequent viral RNA detection. Thus, second-phase IFN induction will be induced when IAV is detected in newly infected, or already infected, IFN-stimulated cells. / / The role of IRF3 and IRF7 in the host response to IAV infection was analyzed in IRF3− −, IRF7− −, / / and IRF3− −IRF7− − knockout mice [15]. While the absence of IRF3 had only a moderate impact on IFN expression, deletion of IRF7 completely abolished IFNα production after infection. In contrast, lack of both IRF3 and IRF7 resulted in the absence of both IFNα and IFNβ after IAV infection. Ciancanelli et al. reported that human IRF7 deficiency impaired IFN amplification in pDC, leukocytes, induced pluripotent stem cell (iPSC)-derived pulmonary epithelial cells [16]. Despite advances in the understanding of innate response to influenza in the mouse model, it is essential to conduct further studies in humans to decipher the innate immune responses to IAV, particularly at the site of infection [17]. Epithelial cells that line the conducting (bronchial) and respiratory (alveoli) airways are the primary site of IAV replication in the lung [18]. In response to viral infection, epithelial cells trigger the innate immune response to limit viral dissemination. To examine the role of IRF7 in amplified second phase induction of IFN during IAV infection of the human lung epithelium, we Viruses 2020, 12, 377 3 of 12 used immortalized cell lines representative of the bronchial (BCi-NS1.1, [19]) and alveolar (A549) lung epithelium.

2. Materials and Methods

2.1. Cell Culture The BCi-NS1.1 cell line was derived from human airway basal cells via expression of a retrovirus expressing human telomerase and was a kind gift of Dr. Ronald Crystal, Cornell Medical Center [19]. BCi-NS1.1 cells were cultured in BEBMTM Bronchial Epithelial Cell Growth Basal Medium (Lonza, cat#CC-3171) with SingleQuotsTM Supplements and Growth Factors (Lonza, cat#CC-4175). The human pulmonary epithelial cell line, A549, was obtained from the American Type Culture Collection (ATCC, Manassas, VA, CCL-185). A549 cells were propagated in Dulbecco’s modiﬁed Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS), 2 mM L-glutamine, and 80 µg of gentamicin/mL.

2.2. Preparation of Influenza Virus Stock H1N1 influenza virus, A/PR/34/8 (PR8), was passaged in Madin-Darby canine kidney (MDCK) cells. Virus was propagated in MDCK cells in DMEM/F12 with ITS+ (BD Biosciences, Franklin Lakes, NJ) and trypsin, harvested at 72 h post-infection and titered by plaque assay in MDCK cells. The virus was stored in aliquots at 80 C. There was no detectable endotoxin in the final viral preparations used − ◦ in the experiments as determined by limulus amebocyte lysate assay (Cambrex, Walkersville, MD). The lower limit of detection of this assay is 0.1 EU/mL or approximately 20 pg/mL LPS.

2.3. siRNA Transfection into BCi-NS1.1 and A549 Cells IRF7 siRNA (siIRF7) and MAVS siRNA (siMAVS) were purchased from Dharmacon (Cat# L-011810-00 and Cat# L-024237-00). IRF3 siRNA (siIRF3) and negative control siRNA were purchased from Ambion (Cat# AM16708A and Cat# 4390843). For siRNA treatment, cells were plated and cultured for 24 h before treating with siRNAs (40 nM). siRNA was diluted in 250 µL of Opti-MEM medium and mixed gently. Five µL of Lipofectamine 2000 (Invitrogen) was added to 250 µL Opti-MEM medium and incubated for 5 min. Diluted siRNA and Lipofectamine 2000 were combined and mixed gently and incubated for 20 min at RT. The siRNA-Lipofectamine 2000 complexes were added to each well and mixed gently. The siRNA ﬁnal concentration was 100 nM. The cells were incubated at 37 ◦C for 24 h and fresh media were added to replace the siRNA containing transfection media.

2.4. Lactate Dehydrogenase (LDH) Assay Staurosporine was used as a positive control (Enzo Biochem Inc., Farmingdale, NY, USA). LDH activity was measured using a coupled enzymatic reaction using a commercially available kit (BioVision Incorporated, Milpitas, CA, USA). The amount of LDH activity was assessed by detection of the reaction product, formazan, at 500 nm using a spectrophotometer (Vmax Microplate reader, Molecular Devices, Sunnyvale, CA, USA). Cytotoxicity was expressed as the percentage of LDH released (supernatant) of the total LDH present (cell + supernatant).

2.5. Measurement of mRNA Expression by Quantitative Real-Time PCR (qRT-PCR) Total RNA from cells was extracted using a modified TRIzol (Invitrogen, Carlsbad, CA, USA) protocol, spectrophometrically quantitated, and the integrity was verified by formaldehyde agarose gel electrophoresis. RNA was treated with DNAse to remove genomic DNA contamination. Equal amounts (1 µg) of RNA from each sample were reverse-transcribed using oligo (dT) as primers for production of cDNA (SuperScript II First-Strand Synthesis System for RT-PCR, Invitrogen, Carlsbad, CA, USA). Gene specific primers for PRRs, cytokines and β-actin housekeeping genes were used. qRT-PCR was performed using SYBR Green (Quanta Biosciences, Gaithersburg, MD, USA) in a Bio-Rad CFX96TM Touch Real-Time PCR Detection System. Results were calculated and graphed from the ∆CT of the Viruses 2020, 12, 377 4 of 12 target gene and normalizing housekeeping gene control. The primers’ sequences were as follows: RIG-I forward 50-TCCTTTATGAGTATGTGGGCA-30; RIG-I reverse 50-TCGGGCACAGAATATCTTTG-30; IFN-β forward 50-GCTCTCCTGTTGTGCTTCTCCAC-30; IFN-β reverse 50-CAATAGTCTCATT CCAGCCAGTGC-30; β-actin forward 50-GCCAACCGCGAGAAGATGACC-30; β-actin reverse 50-CTCCTTAATGTCACGCACGATTTC-30; TLR3 forward 50-GTCTGGGAACATTTCTCTTC-30; TLR3 reverse 50-GATTTAAACATTCCTCTTCGC-30; IRF7 forward 50-CAGATCCAGTCCCAACC AAG-30; IRF7 reverse 50-GTCTCTACTGCCCACCCGTA-30; IFN-λ1 forward 50-CGCCTTGGAAGA GTCACTCA-30; IFN-λ1 reverse 50-GAAGCCTCAGGTCCCAATTC-30; IL-6 forward 50-AGGAGCCC AGCTATGAACT-30; IL-6 reverse 50-TGAGATGCCGTCGAGGATG-30; MAVS forward 50-CAATGCCG TTTGCTGAAGAC-30; MAVS reverse 50-ATTCCTTGGGATGGCTCTGG-30.

2.6. RIG-I, TLR3 and IRF7 Protein Determination by Immunoblotting The cells were harvested and homogenized, and then lysed in 500 µL of cold lysis buffer (150 mM NaCl, 50 mM Tris, pH 8.0, 10 mM EDTA, NaF, sodium pyrophosphate, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 10 µg of leupeptin/mL). Cell homogenates were clarified by centrifugation at 10,000 g, at 4 C for 10 min, and the clarified lysates were mixed with SDS-PAGE sample buffer × ◦ (60 mM Tris, pH 6.8, 10% glycerol, 2.3% SDS) and heated to 95 ◦C for 5 min. The samples were separated using a 4–15% gradient gel and electrophoretically transferred to polyvinylidene fluoride (PVDF) membranes. For the detection of proteins, the membranes were immunoblotted with rabbit polyclonal antibody specific for RIG-I, TLR3 (both from Thermo Fisher Scientific, NY, USA), total IRF7 (Abcam, Cambridge, MA, USA) and GAPDH (R&D Systems). The membrane signals were detected using horseradish peroxidase-labeled goat anti-rabbit IgG (Cell Signaling Technology, Beverly, MA, USA) and chemiluminescent reagents (Pierce Biotechnology, Rockford, IL, USA). Blots were viewed using the Syngene G:box Bioimaging System with GeneTools software (Syngene, Frederick, MD, USA) and quantified using ImageQuant software (BD/Molecular Dynamics, Bedford, MA, USA).

2.7. Statistical Analysis Where applicable, the data have been expressed as the means standard error of the mean (SEM). ± Statistical signiﬁcance was determined by one-way ANOVA with Student–Newman–Keuls post hoc correction for multiple comparisons as appropriate. Signiﬁcance was considered as P < 0.05.

3. Results

3.1. IRF7 Knockdown Inhibited Influenza-Initiated RIG-I and IFN Induction, but not IFN-β-Mediated RIG-I Induction in Human Bronchial BCi-NS1.1 Cells In lung epithelial cells, type I and III IFN induction by IAV is dependent on RIG-I and TLR3 activation. To examine the role of IRF7 induction in this process, we first examined the effect of IRF7 knockdown by specific siRNA on IRF7 mRNA induction by IAV PR8. IAV induced IRF7 mRNA 360 fold over the levels seen in mock-infected cells. Knockdown was successful as evidenced by almost complete (99%) inhibition of IRF7 mRNA induction in specific IRF7 siRNA (siIRF7)-treated cells relative to that seen in cells treated with siRNA control, when both were infected with IAV PR8 (Figure1A). IAV infection of BCi-NS1.1 cells (Figure1A) caused a 27 and 11 fold increase in RIG-I and TLR3 mRNA levels respectively, over the levels seen in mock-infected cells. RIG-I and TLR3 mRNA induction by IAV was decreased 53% and 74% in siIRF7 treated cells compared to control (CTL) siRNA treated cells. To further confirm the inhibition of RIG-I and TLR3 by IRF7 knockdown, we also assessed downstream cytokine induction by virus. IFN-β, IFN-λ1 and IL-6 mRNA induction by IAV was 31, 120 and 33 fold over mock treated cells, respectively. IFN-β, IFN-λ1 and IL-6 mRNA induction was decreased 88%, 81% and 88%, respectively, in siIRF7-treated virus-infected cells compared to siRNA CTL-treated virus-infected cells (Figure1A). ELISA confirmed the inhibition of IFN- β, IFN-λ1 and IL-6 protein induction released in the supernatants from siIRF7-treated virus-infected cells (Figure1C). Decreased Viruses 2020, 12, 377 5 of 12

IRF7, RIG-I and IL-6 expression upon siRNA CTL-treatment was observed but the difference was not significant. This may be due to off target effects of the siRNA CTL. To control for off target effects, we Viruses 2020, 12, x FOR PEER REVIEW 5 of 12 have calculated the decreased induction in virus-infected siRNA-treated cells compared not to the IAV onlyCTL. group, To control but to for siRNA off target CTL-treated effects, cellswe have instead. calculated the decreased induction in virus-infected siRNARIG-I-treated is inducible cells compared by IFNs no releasedt to the IAV from only the group, same but and to neighboringsiRNA CTL-treated IAV infected cells instead. cells [20]. SecretedRIG IFN--I isβ proteins inducible bind by IFNs to its releasedcognate cellular from the receptor same and and neighboring activate a specific IAV infected JAK-STAT cells pathway. [20]. FollowingSecreted receptor IFN-β proteins activation bind by to IFN, its the cognate transcription cellular factors receptor STAT1 and activate and STAT2 a specific are phosphorylated JAK-STAT bypathw Janusay. protein Following tyrosine receptor kinases activation Jak1 and by Tyk2 IFN, and the released transcription from their factors docking STAT1 sites and on STAT2 the receptor. are Theyphosphorylated then associate by with Janus IRF-9 protein and tyrosine form the kinases ISGF3 Jak1 complex, and Tyk2 which and stimulates released IFN-dependentfrom their docking gene transcriptionsites on the byreceptor. binding They to the then IFN associate stimulated with response IRF-9 and element form the (ISRE) ISGF3 sequences complex, locatedwhich stimulates in the RIG-I promoter.IFN-dependent We next gene used transcription trichostatin Aby (TSA), binding an to ISGF3 the complexIFN stimulated formation response inhibitor, element to examine (ISRE) if RIG-Isequences induction located in human in the BCi-NS1.1 RIG-I promoter. cells is actuallyWe next through used trichostatin the ISGF3 A complex (TSA), anas has ISGF3 been complex reported information other model inhibitor, systems. to examine As expected, if RIG-I we induct foundion thatin human TSA significantlyBCi-NS1.1 cells inhibited is actually RIG-I, through IRF7 the and downstreamISGF3 complex cytokine as has induction been reported by IAV in in our other human model cell systems. model (Figure As expected,1B). This we confirms found that that TSA RIG-I mRNAsignificantly induction inhibited by IAV RIG requires-I, IRF7 assembly and downstream of the transcription cytokine induction factor complex by IAV ISGF3in our inhuman our human cell bronchialmodel ( epithelialFigure 1B). cells. This confirms that RIG-I mRNA induction by IAV requires assembly of the transcription factor complex ISGF3 in our human bronchial epithelial cells.

FigureFigure 1. 1.IRF7 IRF7 knockdown knockdown by by specific specific siRNAsiRNA or or chemical chemical inhibition inhibition decreased decreased influenza influenza-induced-induced innateinnate responses responses in in human human airway airway BCi-NS1.1 BCi-NS1.1 basalbasal cells. BCi BCi-NS1.1-NS1.1 cells cells were were transfected transfected with with IRF7 IRF7 siRNAsiRNA and and cultured cultured for for 72 72 h (hA ()A or) or these these cells cells were were treated treated with with TSA TSA (100 (100 ng ng/m/mL),L), an an ISGF3 ISGF3 inhibitor, inhibitor, for 16for h (B 16). h After (B). After siRNA siRNA or TSA or treatment,TSA treatment, the cells the cells were were infected infect withed with IAV IAV PR8 PR8 at the at the MOI MOI of 0.2of 0.2 for for 24 h. Total24 h. RNA Total was RNA extracted was extracted and mRNA and mRNA expression expression was assessed was assessed by qRT-PCR. by qRT- TranscriptPCR. Transcript levels levels of mRNA of weremRNA normalized were normalized relative torelative the constitutively to the constitutively expressed expressedβ-actin β- gene.actin gene. (C) Cytokine (C) Cytokine protein protein levels releasedlevels released in the supernatant in the supernatant were assessed were assessed by ELISA. by ELISA. Data wereData were expressed expressed as the as meansthe meansSEM ± SEM from ± threefrom separate three separate experiments. experiments. Statistical Statistical significance significance was determined was determined by ANOVA. by ANOVA. * denotes * significant denotes diffsignificanterence compared difference to compared data from to the data siRNA from CTLthe siRNA+ IAV infectedCTL+ IAV group, infectedp < group,0.05. p < 0.05.

Overall,Overall, the the results results demonstrate demonstrate that thatde de novonovo synthesis of of IRF7 IRF7 is is critical critical in in IFN IFN and and RIG RIG-I-I mRNA mRNA inductioninduction during during IAV IAV infection infection in in BCi-NS1.1 BCi-NS1.1 cells, and that that RIG RIG-I-I mRNA mRNA induction induction by by IAV IAV requires requires ISGF3ISGF3 assembly. assembly. The The results results demonstrate, demonstrate, forfor thethe ﬁrstfirst time, time, that that these these events events occur occur in in human human airway airway epithelialepithelial cells. cells. To determine whether IRF7 is critical in the effects of first phase IFN induction on autocrine and paracrine amplification of RIG-I and TLR3 that occurs during viral infection, it is necessary to isolate induction of these receptors by IFN in the absence of virus. Therefore, we next examined RIG-I and

Viruses 2020, 12, 377 6 of 12

To determine whether IRF7 is critical in the effects of first phase IFN induction on autocrine and paracrine amplification of RIG-I and TLR3 that occurs during viral infection, it is necessary to isolate induction of these receptors by IFN in the absence of virus. Therefore, we next examined RIG-I and TLR3 induction by IFN-β in siIRF7 treated, uninfected, BCi-NS1.1 cells. To mimic the effects of first phase IFN release only, these cells were stimulated with 500 U/ml of IFN-β for 6 h without Viruses 2020, 12, x FOR PEER REVIEW 6 of 12 IAV infection. As expected IFN-β treatment increased RIG-I mRNA 240-fold over mock treated cells (Figure2TLR3A). Itinduction also induced by IFN- TLR3β in siIRF7 mRNA treated, by uninfected, 22 fold over BCi- controls.NS1.1 cells. Furthermore,To mimic the effects IFN- ofβ firsttreatment increasedphase IRF7 IFN mRNA release 14-fold only, these over cells mock were treated stimulated cells, with but did 500 notU/ml significantly of IFN-β for augment 6 h without IFN- IAVβ mRNA infection. As expected IFN-β treatment increased RIG-I mRNA 240-fold over mock treated cells levels, showing that first phase IFN effects are isolated using this method. Surprisingly, RIG-I and (Figure 2A). It also induced TLR3 mRNA by 22 fold over controls. Furthermore, IFN-β treatment β TLR3 mRNAincreased induction IRF7 mRNA was 14 not-fold a ff overected mock by IRF7treated knockdown cells, but did during not significantly IFN- treatment augment IFN (Figure-β 2A). We alsomRNA examined levels, our showing cells withthat first a JAK-STAT phase IFN effects pathway are isolated inhibitor, using Ruxolitinib, this method. as Surprisingly, an inhibition RIG- positive controlI since and TLR3 RIG-I mRNA andIRF7 induction mRNA was inductionnot affected by by IFN-IRF7 βknockdownrequires JAK-STATduring IFN-β pathway treatment activation (Figure [21]. As expected,2A). We the also JAK-STAT examined our pathway cells with inhibitor a JAK-STAT significantly pathway inhib blockeditor, Ruxolitinib, RIG-I and as IRF7an inhibition induction by IFN-β inpositive BCi-NS1.1 control cells since (Figure RIG-I2 B). and We IRF7 also mRNA determined induction protein by IFN expression-β requires levels JAK-STAT by immunoblotting pathway activation [21]. As expected, the JAK-STAT pathway inhibitor significantly blocked RIG-I and IRF7 (Figure2C). siIRF7 successfully inhibited IRF7 protein induction by IAV. Specifically, we found that induction by IFN-β in BCi-NS1.1 cells (Figure 2B). We also determined protein expression levels by IRF7 knockdownimmunoblotting did not (Figure block 2C). RIG-I siIRF7 and successfully TLR3 protein inhibited induction IRF7 by protein IFN- β induction, but blocked by IAV. RIG-I and TLR3 proteinSpecifically, induction we found by that IAV. IRF7 knockdown did not block RIG-I and TLR3 protein induction by IFN- β, but blocked RIG-I and TLR3 protein induction by IAV.

Figure 2.FigureKnockdown 2. Knockdown of IRF7 of IRF7 did did not not inhibit inhibit IFN- IFN-β--mediatedmediated RIG RIG-I-I and and TLR3 TLR3 induction induction in BCi- inNS1.1 BCi-NS1.1 cells. BCi-NS1.1cells. BCi-NS1.1 cells cells were were transfected transfected with with IRF7 IRF7 siRNA siRNA for for three three days days (A), ( orA ), were or were treated treated with with Ruxolitinib (2 µM), a JAK inhibitor, for 16 h (B). After the siRNA or TSA treatment, the cells were Ruxolitinib (2 µM), a JAK inhibitor, for 16 h (B). After the siRNA or TSA treatment, the cells were stimulated with 500 U/mL of IFN-β for 6 h. Total RNA was extracted and mRNA expression was β stimulatedassessed with by 500 qRT U-/PCR.mL of Transcript IFN- for levels 6 h. of mRNATotal RNA were wasnormalized extracted relative and to mRNA the constitutively expression was assessedexpressed by qRT-PCR. β-actin gene. Transcript Data were levels expressed of mRNA as the means were ± normalized SEM from three relative separate to experiments. the constitutively expressedStatisticalβ-actin significance gene. Data was weredetermined expressed by ANOVA. as the * meansdenotes significantSEM from difference three compared separate to experiments. data ± Statisticalfrom significance IFN-β group, was p < determined0.05. (C) RIG- byI, TLR3 ANOVA. and IRF7 * denotes proteins significantin BCi-NS1.1 di cellsfference were compareddetected by to data from IFN-immunoblotting.β group, p < The0.05. immunoblot (C) RIG-I, shown TLR3 is and representative IRF7 proteins of three in separate BCi-NS1.1 experiments, cells were with detected quantitation of that result depicted below the image. by immunoblotting. The immunoblot shown is representative of three separate experiments, with quantitation of that result depicted below the image.

Viruses 2020, 12, 377 7 of 12 Viruses 2020, 12, x FOR PEER REVIEW 7 of 12

Taken together, the the data data suggest suggest that that IRF7 IRF7 does does not not mediate mediate paracrine paracrine induction induction of RIG of- RIG-II in non in- non-infectedinfected cells cellscaused caused by first by phase ﬁrst phase IFN-β IFN- secretedβ secreted from fromadjacent adjacent IAV infected IAV infected cells. cells.In contrast, In contrast, IRF7 IRF7is important is important for full for full induction induction of of IFN IFN in in IAV IAV-infected-infected BCi BCi-NS1.1-NS1.1 cells. cells. Th Therefore,erefore, the the siIRF7 siIRF7 knockdown resultsresults suggest IRF7 indirectly modulates RIG-IRIG-I and thusthus cytokinecytokine responsesresponses throughthrough inhibitinginhibiting generationgeneration ofof IFNIFN duringduring inﬂuenzainfluenza virusvirus infection.infection.

3.2. MAVS, but not IRF3, KnockdownKnockdown InhibitedInhibited Inﬂuenza-InitiatedInfluenza-Initiated RIG-IRIG-I andand IFNIFN InductionInduction inin HumanHuman BronchialBronchial BCi-NS1.1BCi-NS1.1 CellsCells IRF3IRF3 isis anotheranother transcriptiontranscription factorfactor implicatedimplicated inin RIG-IRIG-I mediatedmediated IFNIFN induction,induction, whichwhich couldcould potentiallypotentially induceinduce RIG-IRIG-Iin in an an autocrine autocrine or or paracrine paracrine fashion. fashion. We We next next tested tested whether whether this this occurred occurred in ourin our system system using using siRNA siRNA and and the relativethe relative roles roles of IRF3 of andIRF3 IRF7 and in IRF7 IFN in induction. IFN induction. Notably, Notably, IRF3 mRNA IRF3 inductionmRNA induction by IAV by (1.8-fold IAV (1.8 increase-fold increase over mock, over Figuremock, 3FigureA) was 3A) minimal was minimal when comparedwhen compared with IRF7 with mRNAIRF7 mRNA induction induction by IAV by (370-fold IAV (370 increase-fold over increase mock, over Figure mock,1A). Figure Furthermore, 1A). Furthermore IRF3 knockdown, IRF3 reducedknockdown RIG-I reduced and TLR3 RIG mRNA-I and inductionTLR3 mRNA by IAV induction to a lesser by extent IAV to (20% a lesser and 22 extent %, respectively) (20% and 22 than %, knockdownrespectively) of than IRF7 knockdown (59% and 73% of inhibitionIRF7 (59%of and RIG-I 73% and inhibition TLR3, respectively). of RIG-I and Therefore, TLR3, respectively). IRF3 is less involvedTherefore, in IRF3 the inductionis less involved of RIG-I in andthe TLR3induction by IAV of RIG than-I IRF7 and inTLR3 these by cells IAV (Figure than IRF73A). in these cells (Figure 3A).

Figure 3. Knockdown of MAVS, but not IRF3, inhibited influenza-mediatedinfluenza-mediated RIG-I,RIG-I, TLR3 and cytokine mRNA inductioninduction in BCi-NS1.1BCi-NS1.1 cells.cells. The cells were firstfirst transfected with (A and C) IRF3 or (B and D) MAVS siRNAsiRNA andand culturedcultured forfor 7272 hh andand thenthen infectedinfected withwith IAVIAV PR8PR8 (MOI(MOI= = 0.2) for 24 h.h. Total RNARNA waswas extractedextracted andand mRNA mRNA expression expression was was assessed assessed by by qRT-PCR qRT-PCR (A (andA andB). B Data). Data were were expressed expressed as the as means SEM from three separate experiments. Statistical significance was determined by ANOVA. the means± ± SEM from three separate experiments. Statistical significance was determined by *ANOVA. denotes significant* denotes significant difference difference compared compared to data from to data the siRNAfrom the CTL siRNA+ IAV CTL+ infected IAV group,infectedp group,< 0.05. (pC

Viruses 2020, 12, 377 8 of 12

siRNAViruses 20 (siMAVS)-treated20, 12, x FOR PEER REVIEW cells infected with IAV as compared to MAVS induction seen in IAV infected8 of 12 cells treated with siRNA control (Figure3B). siMAVS treatment reduced RIG-I (63%) and TLR3 (96%) mRNAseen in induction,IAV infected as wellcells astreated subsequent with siRNA IFN-β control(92%), IFN-(Figureλ1 (93%)3B). siMAVS and IL-6 treatment (85%) induction reduced by RIG IAV-I in(63%) the cellsand TLR3 (Figure (96%)3B). Thus,mRNA the induction, results demonstrate as well as subsequent that first phase IFN of-β IFN (92% synthesis), IFN-λ1 induced (93%) and by IAVIL-6 requires(85%) induction MAVS. We by alsoIAV confirmedin the cells the (Figure RIG-I 3B). mRNA Thus, expression the results change demonstrate affected bythat siIRF3 first phase and siMAVS of IFN atsynthesis protein induced expression by levelsIAV requires by immunoblotting MAVS. We (Figurealso confirmed3C,D). the RIG-I mRNA expression change affectedTo determineby siIRF3 and whether siMAVS the at e ffprotectsein seen expression by siRNA levels knockdown by immunoblotting could be ( dueFigure to 3 non-specific C and D). cytotoxicity,To determine we measured whether LDH the release effects as seen a percentage by siRNA of knockdown total LDH present could be (media due to+ cells) non-specific during siRNAcytotoxicity, treatment. we measured Cells were LDH exposed release to siIRF7,as a percentage siMAVS, of siIRF3 total andLDH siRNA present control (media for + 24 cells) h, followed during bysiRNA the measurement treatment. Cells of LDH were in exposed the media. to siIRF7, Staurosporine siMAVS, was siIRF3 used and as asiRNA positive control control for for 24 cytotoxicityh, followed (Figureby the 4).measurement Basal LDH release of LDH was in low the (mock). media. IAV Staurosporine infection caused was BCi-NS1.1 used as acell positive death after control 3 days, for butcytotoxicity not 1-day (Figure post-infection 4). Basal whenLDH release the samples was low were (mock). collected. IAV LDHinfection levels caused of all BCi siRNA-NS1.1 treated cell death cells andafter cells3 days, at 1-daybut not post-infection 1-day post-infection were minor when compared the samples to the were positive collected. control. LDH Thus,levels theof all eff siRNAects in siIRF7-treated andcells siMAVS-treated and cells at 1-day cells post is- notinfection due to were non-specific minor compared cytotoxicity. to the positive control. Thus, the effects in siIRF7- and siMAVS-treated cells is not due to non-specific cytotoxicity.

Figure 4.4. EEffectsffects ofof siRNA siRNA treatment treatment and and IAV IAV infection infection on cellon cell viability. viability. BCi-NS1.1 BCi-NS1.1 were were incubated incubated with siRNAwith siRNA for 3 days.for 3 days. For IAV For PR8IAV infectedPR8 infected cells, cells, BCi-NS1.1 BCi-NS1.1 were were incubated incubated with IAVwith (MOIIAV (MOI= 0.2) = for 0.2) 1 orfor 3 1 days. or 3 days. Staurosporine Staurosporine (10 µ M)(10 treatedµM) treated cells werecells usedwere asused a positive as a positive control control for cytotoxicity. for cytotoxicit LDHy. activityLDH activity of the of supernatant the supernatant was measuredwas measured by using by using the LDH-Cytotoxicitythe LDH-Cytotoxicity Assay Assay Kit. Kit. Cytotoxicity Cytotoxicity is expressedis expressed as theas the percentage percentage of supernatant of supernatant released released LDH toLDH LDH to released LDH released from Staurosporine from Staurosporine treated cells.treated Results cells. Results are shown are asshown the Mean as the MeanSEM from± SEM three from separate three separate experiments. experiments. ± 3.3. IRF7 Knockdown Inhibited Influenza-Initiated, but not IFN-β-Induced, RIG-I Induction in Human 3.3. IRF7 Knockdown Inhibited Influenza-Initiated, but not IFN-β-Induced, RIG-I Induction in Human Alveolar Epithelial Cells Alveolar Epithelial Cells We extended our work to more distal epithelial cells in the lung, human alveolar epithelial cells. We extended our work to more distal epithelial cells in the lung, human alveolar epithelial cells. We used a common model for these cells, the A549 alveolar epithelial cell line. Infection with IAV PR8 We used a common model for these cells, the A549 alveolar epithelial cell line. Infection with IAV resulted in a 40- and 15-fold increase in RIG-I and TLR3 mRNA levels, respectively, over mock-infected PR8 resulted in a 40- and 15-fold increase in RIG-I and TLR3 mRNA levels, respectively, over mock- cells. RIG-I and TLR3 mRNA induction was decreased 80% and 76% in siIRF7 treated cells compared to controlinfected siRNA cells. RIG treated-I and cells. TLR3 Downstream mRNA induction IFN-β mRNA was decreased induction 80% was decreasedand 76% in 72% siIRF7 in siIRF7-treated treated cells compared to control siRNA treated cells. Downstream IFN-β mRNA induction was decreased 72% IAV-infected cells compared to control siRNA-treated cells (Figure5A). In terms of the siIRF7 inhibition in siIRF7-treated IAV-infected cells compared to control siRNA-treated cells (Figure 5A). In terms of of its target, IRF7 mRNA induction by IAV was blocked 86% in the siIRF7-treated IAV-infected A549 the siIRF7 inhibition of its target, IRF7 mRNA induction by IAV was blocked 86% in the siIRF7- cells (Figure5A). treated IAV-infected A549 cells (Figure 5A). As was the case for BCi-NS1.1 cells, when A549 cells were treated with IFN-β only, siIRF7 did As was the case for BCi-NS1.1 cells, when A549 cells were treated with IFN-β only, siIRF7 did not inhibit IFN-β-stimulated RIG-I and TLR3 induction (Figure5B). We next examined if secreted not inhibit IFN-β-stimulated RIG-I and TLR3 induction (Figure 5B). We next examined if secreted IFN-β from IAV infected cells stimulated RIG-I/IRF7 expression in uninfected cells. Supernatants IFN-β from IAV infected cells stimulated RIG-I/IRF7 expression in uninfected cells. Supernatants were collected from IAV infected cells after 24 h poi. Then the supernatants were filtered to remove were collected from IAV infected cells after 24 h poi. Then the supernatants were filtered to remove IAV using a 100 K MW filter. Absence of IAV was confirmed using qRT-PCR for viral NP and matrix IAV using a 100 K MW filter. Absence of IAV was confirmed using qRT-PCR for viral NP and matrix genes (not shown). The filtered supernatants were added to A549 cells. After 6 h, both RIG-I and IRF7 genes (not shown). The filtered supernatants were added to A549 cells. After 6 h, both RIG-I and IRF7 mRNA were induced by the supernatants (Figure5C). We confirmed the e ffect of siRNA on RIG-I mRNA were induced by the supernatants (Figure 5C). We confirmed the effect of siRNA on RIG-I protein induction by IAV using Western blot (Figure5D). In these cells, protein levels reflected mRNA protein induction by IAV using Western blot (Figure 5D). In these cells, protein levels reflected mRNA expression levels, as there was no detectable RIG-I protein in IAV-infected A549 cells treated with siRNA for IRF7. In contrast, there was no inhibition of RIG-I protein induction by IFN-β with IRF7 siRNA treatment. The results suggested that, as we found in human bronchial BCi-NS1.1 cells,

Viruses 2020, 12, 377 9 of 12 expression levels, as there was no detectable RIG-I protein in IAV-infected A549 cells treated with siRNA for IRF7. In contrast, there was no inhibition of RIG-I protein induction by IFN-β with IRF7 siRNAViruses treatment.2020, 12, x FOR The PEER results REVIEW suggested that, as we found in human bronchial BCi-NS1.1 cells, IRF79 of 12 knockdown inhibited inﬂuenza-initiated, but not IFN-β-induced, RIG-I induction in human alveolar AECIRF7 II cells. knockdown inhibited influenza-initiated, but not IFN-β-induced, RIG-I induction in human alveolar AEC II cells.

FigureFigure 5. 5.Knockdown Knockdown of of IRF7 IRF7 inhibited inhibited influenza-mediated influenza-mediated RIG-I RIG induction-I induction and and innate innate responses responses in in humanhuman alveolar alveolar epithelial epithelial cells. cells. A549 A549 cellscells werewere transfectedtransfected with IRF7 siRNA siRNA for for three three days. days. Then, Then, the thecells cells were were infected infected with with (A ()A IAV) IAV PR8 PR8 at atthe the MOI MOI of of0.2 0.2 for for 24 24h or h ( orB) (500B) 500 U/m UL/mL of IFN of IFN--β forβ 6for h. 6Total h. TotalRNA RNA was was extracted extracted and andmRNA mRNA expression expression was wasassessed assessed by qRT by- qRT-PCR.PCR. Data Data were were expressed expressed as the asmeans the means ± SEM SEMfrom fromthree threeseparate separate experiments. experiments. Statistical Statistical significance significance was determined was determined by ANOVA. by ± ANOVA.* denotes * denotes significant significant difference diff erencecompared compared to data tofrom data the from siRNA the siRNACTL + IAV CTL infected+ IAV infected group, group,P < 0.05. p <(C0.05.) Supernatants(C) Supernatants from from IAV IAV infected infected cells cells stimulated stimulated RIG RIG-I-I/IRF7/IRF7 expression expression in in uninfected uninfected cells. cells. SupernatantsSupernatants were were collected collected from from IAV IAV infected infected cells cells after after 24 24 h poi.h poi. Then Then the the supernatants supernatants were were filtered filtered toto remove remove IAV IAV using using a 100 a 100 K MWK MW filter filter (Amicon). (Amicon). The The filtered filtered supernatants supernatants were were added added to to A549 A549 cells cells andand changes changes in in RIG-I RIG/-IRF7I/IRF7 mRNA mRNA were were assessed assessed by by qRT-PCR. qRT-PCR. (D ()D RIG-I) RIG protein-I protein in in A549 A549 cells cells was was detecteddetected by by immunoblotting. immunoblotting. The The immunoblot immunoblot shown shown is is representative representative of of three three separate separate experiments, experiments, withwith quantitation quantitation of of that that result result depicted depicted below below the the image. image. 4. Discussion 4. Discussion The airway epithelium is strategically positioned at the interface with the environment, and The airway epithelium is strategically positioned at the interface with the environment, and thus thus plays a key role in the innate immune lung response to outside stimuli or infection [22]. The plays a key role in the innate immune lung response to outside stimuli or infection [22]. The airway airway epithelium responds to IAV by increasing its production of mediators including cytokines epithelium responds to IAV by increasing its production of mediators including cytokines and and chemokines that can recruit and activate inflammatory cells [23]. Human airway basal cells (BC), chemokines that can recruit and activate inflammatory cells [23]. Human airway basal cells (BC), a a proliferating population of cells that reside in close proximity to the basement membrane, function as proliferating population of cells that reside in close proximity to the basement membrane, function stem/progenitor cells of both the mouse and human airway epithelium and differentiate into the other as stem/progenitor cells of both the mouse and human airway epithelium and differentiate into the specialized cell types during normal epithelial turnover and repair. In alveoli, the alveolar epithelia, other specialized cell types during normal epithelial turnover and repair. In alveoli, the alveolar comprised of type I alveolar epithelial cells (AEC I) and type II alveolar epithelial cells (AEC II), cover epithelia, comprised of type I alveolar epithelial cells (AEC I) and type II alveolar epithelial cells (AEC more than 99% of the internal surface area [24]. AEC II can replicate in the alveoli to replace damaged II), cover more than 99% of the internal surface area [24]. AEC II can replicate in the alveoli to replace AEC I cells, playing a similar basal role as BC in bronchiole. damaged AEC I cells, playing a similar basal role as BC in bronchiole. Despite the evidence that BCs of the bronchioles and AEC II of the alveoli are stem cells, relatively Despite the evidence that BCs of the bronchioles and AEC II of the alveoli are stem cells, little is known about their biology in innate immune responses to IAV. In this report, we examined the relatively little is known about their biology in innate immune responses to IAV. In this report, we two most important human lung epithelial progenitor/stem cells, bronchial basal cells (BCi-NS1.1) and examined the two most important human lung epithelial progenitor/stem cells, bronchial basal cells AEC II cells (A549). (BCi-NS1.1) and AEC II cells (A549). First, we found that the basal expression level, without virus stimulation, of RIG-I, TLR3 and First, we found that the basal expression level, without virus stimulation, of RIG-I, TLR3 and IRF7 were very low in BCi-NS1.1 cells, and that IAV infection induced robust RIG-I, IRF7 and IFN IRF7 were very low in BCi-NS1.1 cells, and that IAV infection induced robust RIG-I, IRF7 and IFN expression (Figure1). We then used exogenous IFN- β to examine the effects of produced IFN-β during expression (Figure 1). We then used exogenous IFN-β to examine the effects of produced IFN-β the first phase of IFN induction on non-infected cells in the absence of IAV. IFN-β alone induced during the first phase of IFN induction on non-infected cells in the absence of IAV. IFN-β alone induced significant RIG-I, TLR3 and IRF7 expression (Figure 2). This is consistent with a model whereby secreted first phase IFN protein significantly elevates RIG-I, TLR3 and IRF proteins and turns these uninfected neighboring cells into an “immunoactive” state primed to respond to subsequent viral RNA detection. The second phase IFN induction is then amplified when IAV was detected in newly infected, IFN-stimulated cells. Thus, RIG-I, TLR3 and IRF7 expression is highly

Viruses 2020, 12, 377 10 of 12 significant RIG-I, TLR3 and IRF7 expression (Figure2). This is consistent with a model whereby secreted first phase IFN protein significantly elevates RIG-I, TLR3 and IRF proteins and turns these uninfected neighboring cells into an “immunoactive” state primed to respond to subsequent viral RNA detection. The second phase IFN induction is then amplified when IAV was detected in newly infected, VirusesIFN-stimulated 2020, 12, x FOR cells. PEER Thus, REVIEW RIG-I, TLR3 and IRF7 expression is highly inducible by the first10 phase of 12 produced IFN. IRF7 is the master regulator of type-I IFN-dependent immune responses in vivo [25], but induciblethe role of by IRF7 the in first human phase airway produced epithelial IFN cell. IRF7 responses is the tomaster IAV has regulator not been of previously type-I IFN investigated.-dependent immuneOur data responses demonstrate in vivo that [25] IRF7, but controls the role the of second-phase IRF7 in human IFN airway amplification epithelial during cell responses influenza to virus IAV hasinfection not been in these previously cells. Theinvestigated. produced Our first-phase data demonstrate IFN stimulates that IRF7 RIG-I controls and IRF7 the through second-phase JAK-STAT IFN amplificationsignaling and during ISGF3 complexinfluenza formation virus infection in a positive-feedback in these cells. The amplification produced first loop-phase during IFN IAV stimula infectiontes RIGin human-I and lungIRF7 epitheliathrough JAK (Figures-STAT1 and signaling2; model and shown ISGF3 as complex Figure6 ).formation IRF7 is not in involveda positive in-feedback RIG-I amplificationinduction stimulated loop during by the IAV first-phase infection IFN- in humanβ in the lung neighboring epithelia cells, (Figure as RIG-Is 1 and induction 2; model by shown IFN-β asis Figurenot blocked 6). IRF7 by siIRF7 is not (Figure involved2). in RIG-I induction stimulated by the first-phase IFN-β in the neighboring cells, as RIG-I induction by IFN-β is not blocked by siIRF7 (Figure 2).

Figure 6. Proposed model of the RIG-I/IRF7/IFN positive-feedback amplification loop in human lung Figure 6. Proposed model of the RIG-I/IRF7/IFN positive-feedback amplification loop in human lung airway epithelial cells. airway epithelial cells. We found that knockdown of the transcription factor IRF3 only partially inhibited We found that knockdown of the transcription factor IRF3 only partially inhibited influenza- influenza-mediated RIG-I mRNA induction in BCi-NS1.1 cells. This result is consistent with the model mediated RIG-I mRNA induction in BCi-NS1.1 cells. This result is consistent with the model proposed by Sato et al. [26], in that we showed that IRF7 must first be induced in response to IAV in proposed by Sato et al. [26], in that we showed that IRF7 must first be induced in response to IAV in order for early limited IFN production to occur. In another paper, the same group showed that the order /for early limited IFN production to occur. In another paper, the same group showed that the IRF3− − mouse embryonic fibroblasts produced IRF7 and IFN-β when infected with Newcastle disease IRF3-/- mouse embryonic fibroblasts produced IRF7 and IFN-β when infected with Newcastle disease virus, indicating both IRF7 and IFN-β induction is independent of IRF3 [20]. With regards to IRF3, virus, indicating both IRF7 and IFN-β induction is independent of IRF3 [20]. With regards to IRF3, it it likely plays a role in the first phase of IFN induction, as there appeared to be modest inhibition of likely plays a role in the first phase of IFN induction, as there appeared to be modest inhibition of IFN-β by IAV (did not reach statistical significance, Figure3). It is unlikely that IRF3 plays a significant IFN-β by IAV (did not reach statistical significance, Figure 3). It is unlikely that IRF3 plays a role in the amplification of IFN-β induction, as IRF3 was minimally induced by IAV (Figure3). The significant role in the amplification of IFN-β induction, as IRF3 was minimally induced by IAV previous results from other laboratories examining the role of IRF7, IRF3 and RIG-I in IAV-induced (Figure 3). The previous results from other laboratories examining the role of IRF7, IRF3 and RIG-I in IFN production was obtained from animal models. Our data are unique in that we demonstrate that IAV-induced IFN production was obtained from animal models. Our data are unique in that we IRF7, IRF3 and RIG-I play similar roles in human lung epithelial cells during IAV infection. We also demonstrate that IRF7, IRF3 and RIG-I play similar roles in human lung epithelial cells during IAV show that MAVS is required for first-phase IFN production. Doubtless, IRF3 still plays an important infection. We also show that MAVS is required for first-phase IFN production. Doubtless, IRF3 still role in IFN induction, even though it is not critical in the amplification phase. It is demonstrated that plays an important role in IFN induction, even though it is not critical in the amplification phase. It RNA synthesis and nuclear export are required for activation of the IFN induction cascade by IAV is demonstrated that RNA synthesis and nuclear export are required for activation of the IFN using IRF3 phosphorylation as a marker [27]. induction cascade by IAV using IRF3 phosphorylation as a marker [27]. 5. Conclusions 5. Conclusion Our study demonstrates that the induction of two major PRRs, RIG-I and TLR3, in epithelial cells by IAVOur is study through demonstrates an IRF7-dependent that the positiveinduction amplification of two major loop. PRRs, IRF3 RIG plays-I and a minimum TLR3, in role epithelial in this cellsamplification by IAV is loop. through Our an current IRF7- modeldependent is that, positive after cellamplification entry, viral loop. RNA IRF3 is recognized plays a minimum by trace levelrole inRIG-I. this amplification RIG-I signaling loop. induces, Our current via MAVS model adaptor is that, protein, after cell the entry, activation viral RNA of trace is recognized level transcription by trace levelfactors RIG IRF3-I. /7. RIG Next,-I signaling the IRFs induces, induce type via MAVS I IFNs (IFN- adaptorα and protein, IFN-β the), which activation are released of trace outside level transcriptionthe cells to stimulate factors IRF3/7. the production Next, the of IRFs ISGs induce (including type I RIG-IIFNs (IFN and IRF7)-α and in IFN neighboring-β), which cells.are released In this outsideway, the the new cells produced to stimulate RIG-I the and production IRF7 establish of ISGs a positive (including feedback RIG-I loop and whenIRF7) theyin neighboring encounter morecells. In this way, the new produced RIG-I and IRF7 establish a positive feedback loop when they encounter more viruses (Figure 6). IRF7, but not IRF3, plays a central role in modulating the expression of RIG- I, TLR3 and antiviral responses during IAV infections. Whereas IRF3 is constitutively expressed, IRF3 is only important in early-phase IFN induction. IRF7 may therefore be an important molecular driver of the antiviral responses to IAV in human airway epithelial cells. Author Contributions: Conceptualization, W.W.; methodology, W.Z. and L.T.; writing, W.W., J.P.M. and M.S.W.; supervision, B.R.B. All authors have read and agreed to the published version of the manuscript.

Viruses 2020, 12, 377 11 of 12 viruses (Figure6). IRF7, but not IRF3, plays a central role in modulating the expression of RIG-I, TLR3 and antiviral responses during IAV infections. Whereas IRF3 is constitutively expressed, IRF3 is only important in early-phase IFN induction. IRF7 may therefore be an important molecular driver of the antiviral responses to IAV in human airway epithelial cells.

Author Contributions: Conceptualization, W.W.; methodology, W.Z. and L.T.; writing, W.W., J.P.M. and M.S.W.; supervision, B.R.B. All authors have read and agreed to the published version of the manuscript. Funding: The research described in this work was partially supported by Oklahoma Shared Clinical and Translational Resource (OSCTR) pilot grants, grant number U54GM104938 to W.W. and M.S.W., the Merit Review Program of the Department of Veterans Affairs, grant number I01 BX001937 to J.P.M., and the National Institute of General Medical Sciences, grant number 5P20GM103648 to J.P.M. Acknowledgments: We want to thank J. Leland Booth for his excellent technical assistance on this project. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Job, E.R.; Deng, Y.M.; Tate, M.D.; Bottazzi, B.; Crouch, E.C.; Dean, M.M.; Mantovani, A.; Brooks, A.G.; Reading, P.C. Pandemic H1N1 influenza A viruses are resistant to the antiviral activities of innate immune proteins of the collectin and pentraxin superfamilies. J. Immunol. 2010, 185, 4284–4291. [CrossRef][PubMed] 2. Secombes, C.J.; Zou, J. Evolution of Interferons and Interferon Receptors. Front. Immunol. 2017, 8, 209. [CrossRef][PubMed] 3. Wu, W.; Zhang, W.; Duggan, E.S.; Booth, J.L.; Zou, M.H.; Metcalf, J.P. RIG-I and TLR3 are both required for maximum interferon induction by influenza virus in human lung alveolar epithelial cells. Virology 2015, 482, 181–188. [CrossRef][PubMed] 4. Kato, H.; Sato, S.; Yoneyama, M.; Yamamoto, M.; Uematsu, S.; Matsui, K.; Tsujimura, T.; Takeda, K.; Fujita, T.; Takeuchi, O.; et al. Cell type-specific involvement of RIG-I in antiviral response. Immunity 2005, 23, 19–28. [CrossRef] 5. Guillot, L.; Le Goffic, R.; Bloch, S.; Escriou, N.; Akira, S.; Chignard, M.; Si-Tahar, M. Involvement of toll-like receptor 3 in the immune response of lung epithelial cells to double-stranded RNA and influenza A virus. J. Biol. Chem. 2005, 280, 5571–5580. [CrossRef] 6. Fitzgerald, K.A.; McWhirter, S.M.; Faia, K.L.; Rowe, D.C.; Latz, E.; Golenbock, D.T.; Coyle, A.J.; Liao, S.M.; Maniatis, T. IKKepsilon and TBK1 are essential components of the IRF3 signaling pathway. Nat. Immunol. 2003, 4, 491–496. [CrossRef] 7. Kotenko, S.V.; Gallagher, G.; Baurin, V.V.; Lewis-Antes, A.; Shen, M.; Shah, N.K.; Langer, J.A.; Sheikh, F.; Dickensheets, H.; Donnelly, R.P. IFN-lambdas mediate antiviral protection through a distinct class II cytokine receptor complex. Nat. Immunol. 2003, 4, 69–77. [CrossRef] 8. Wu, W.; Booth, J.L.; Duggan, E.S.; Wu, S.; Patel, K.B.; Coggeshall, K.M.; Metcalf, J.P. Innate immune response to H3N2 and H1N1 influenza virus infection in a human lung organ culture model. Virology 2010, 396, 178–188. [CrossRef] 9. Teijaro, J.R. Type I interferons in viral control and immune regulation. Curr. Opin Virol. 2016, 16, 31–40. [CrossRef] 10. Schindler, C.; Darnell, J.E., Jr. Transcriptional responses to polypeptide ligands: The JAK-STAT pathway. Annu. Rev. Biochem. 1995, 64, 621–651. [CrossRef] 11. Zhou, A.; Paranjape, J.M.; Der, S.D.; Williams, B.R.; Silverman, R.H. Interferon action in triply deficient mice reveals the existence of alternative antiviral pathways. Virology 1999, 258, 435–440. [CrossRef][PubMed] 12. Yoneyama, M.; Kikuchi, M.; Natsukawa, T.; Shinobu, N.; Imaizumi, T.; Miyagishi, M.; Taira, K.; Akira, S.; Fujita, T. The RNA helicase RIG-I has an essential function in double-stranded RNA-induced innate antiviral responses. Nat. Immunol. 2004, 5, 730–737. [CrossRef][PubMed] 13. Hui, K.P.; Lee, S.M.; Cheung, C.Y.; Mao, H.; Lai, A.K.; Chan, R.W.; Chan, M.C.; Tu, W.; Guan, Y.; Lau, Y.L.; et al. H5N1 influenza virus-induced mediators upregulate RIG-I in uninfected cells by paracrine effects contributing to amplified cytokine cascades. J. Infect. Dis. 2011, 204, 1866–1878. [CrossRef][PubMed] 14. Honda, K.; Taniguchi, T. IRFs: Master regulators of signalling by Toll-like receptors and cytosolic pattern-recognition receptors. Nat. Rev. Immunol. 2006, 6, 644–658. [CrossRef] Viruses 2020, 12, 377 12 of 12

15. Hatesuer, B.; Hoang, H.T.; Riese, P.; Trittel, S.; Gerhauser, I.; Elbahesh, H.; Geffers, R.; Wilk, E.; Schughart, K. Deletion of Irf3 and Irf7 Genes in Mice Results in Altered Interferon Pathway Activation and Granulocyte-Dominated Inflammatory Responses to Influenza A Infection. J. Innate Immun. 2017, 9, 145–161. [CrossRef] 16. Ciancanelli, M.J.; Huang, S.X.; Luthra, P.; Garner, H.; Itan, Y.; Volpi, S.; Lafaille, F.G.; Trouillet, C.; Schmolke, M.; Albrecht, R.A.; et al. Infectious disease. Life-threatening influenza and impaired interferon amplification in human IRF7 deficiency. Science 2015, 348, 448–453. [CrossRef] 17. Pulendran, B.; Maddur, M.S. Innate immune sensing and response to influenza. Curr. Top. Microbiol. Immunol. 2015, 386, 23–71. [CrossRef] 18. Manicassamy, B.; Manicassamy, S.; Belicha-Villanueva, A.; Pisanelli, G.; Pulendran, B.; Garcia-Sastre, A. Analysis of in vivo dynamics of influenza virus infection in mice using a GFP reporter virus. Proc. Natl. Acad. Sci. USA 2010, 107, 11531–11536. [CrossRef] 19. Walters, M.S.; Gomi, K.; Ashbridge, B.; Moore, M.A.; Arbelaez, V.; Heldrich, J.; Ding, B.S.; Rafii, S.; Staudt, M.R.; Crystal, R.G. Generation of a human airway epithelium derived basal cell line with multipotent differentiation capacity. Respir. Res. 2013, 14, 135. [CrossRef] 20. Sato, M.; Suemori, H.; Hata, N.; Asagiri, M.; Ogasawara, K.; Nakao, K.; Nakaya, T.; Katsuki, M.; Noguchi, S.; Tanaka, N.; et al. Distinct and essential roles of transcription factors IRF-3 and IRF-7 in response to viruses for IFN-alpha/beta gene induction. Immunity 2000, 13, 539–548. [CrossRef] 21. Harrison, C.; Kiladjian, J.J.; Al-Ali, H.K.; Gisslinger, H.; Waltzman, R.; Stalbovskaya, V.; McQuitty, M.; Hunter, D.S.; Levy, R.; Knoops, L.; et al. JAK inhibition with ruxolitinib versus best available therapy for myelofibrosis. N. Engl. J. Med. 2012, 366, 787–798. [CrossRef][PubMed] 22. Bals, R.; Hiemstra, P.S. Innate immunity in the lung: How epithelial cells fight against respiratory pathogens. Eur. Respir. J. 2004, 23, 327–333. [CrossRef][PubMed] 23. Jakiela, B.; Brockman-Schneider, R.; Amineva, S.; Lee, W.M.; Gern, J.E. Basal cells of differentiated bronchial epithelium are more susceptible to rhinovirus infection. Am. J. Respir. Cell Mol. Biol. 2008, 38, 517–523. [CrossRef][PubMed] 24. Mercer, R.R.; Russell, M.L.; Roggli, V.L.; Crapo, J.D. Cell number and distribution in human and rat airways. Am. J. Respir. Cell Mol. Biol. 1994, 10, 613–624. [CrossRef][PubMed] 25. Honda, K.; Yanai, H.; Negishi, H.; Asagiri, M.; Sato, M.; Mizutani, T.; Shimada, N.; Ohba, Y.; Takaoka, A.; Yoshida, N.; et al. IRF-7 is the master regulator of type-I interferon-dependent immune responses. Nature 2005, 434, 772–777. [CrossRef] 26. Sato, M.; Hata, N.; Asagiri, M.; Nakaya, T.; Taniguchi, T.; Tanaka, N. Positive feedback regulation of type I IFN genes by the IFN-inducible transcription factor IRF-7. FEBS Lett. 1998, 441, 106–110. [CrossRef] 27. Killip, M.J.; Smith, M.; Jackson, D.; Randall, R.E. Activation of the interferon induction cascade by influenza a viruses requires viral RNA synthesis and nuclear export. J. Virol. 2014, 88, 3942–3952. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).