Interferon Lambda Signaling in Macrophages Is Necessary for the Antiviral Response to Influenza

Interferon lambda (IFNλ) signaling is a promising therapeutic target against viral infection in murine models, yet little is known about its molecular regulation and its cognate receptor, interferon lambda receptor 1 (IFNLR1) in human lung. We hypothesized that the IFNλ signaling axis was active in human lung macrophages. In human alveolar macrophages (HAMs), we observed increased IFNLR1 expression and robust increase in interferon-stimulated gene (ISG) expression in response to IFNλ ligand. While human monocytes express minimal IFNLR1, differentiation of monocytes into macrophages with macrophage colony-stimulating factor (M-CSF) or granulocyte-macrophage colony-stimulating factor (GM-CSF) increased IFNLR1 mRNA, IFNLR1 protein expression, and cellular response to IFNλ ligation. Conversely, in mice, M-CSF or GM-CSF stimulated macrophages failed to produce ISGs in response to related ligands, IFNL2 or IFNL3, suggesting that IFNLR1 signaling in macrophages is species-specific. We next hypothesized that IFNλ signaling was critical in influenza antiviral responses. In primary human airway epithelial cells and precision-cut human lung slices, influenza infection substantially increased IFNλ levels. Pretreatment of both HAMs and differentiated human monocytes with IFNL1 significantly inhibited influenza infection. IFNLR1 knockout in the myeloid cell line, THP-1, exhibited reduced interferon responses to either direct or indirect exposure to influenza infection suggesting the indispensability of IFNLR1 for antiviral responses. These data demonstrate the presence of IFNλ - IFNLR1 signaling axis in human lung macrophages and a critical role of IFNλ signaling in combating influenza infection.


INTRODUCTION
Influenza type A and B virus infections in humans result in nearly 80,000 deaths in the United States and up to 650,000 deaths worldwide annually (1). The US economic burden is staggering; in the range of $47-$150 billion annually (2). Further, opportunistic bacterial pneumonia following influenza infection is a leading cause of death worldwide. Early in infection, interferons are released and signal through their specific receptors, conferring protection against viral infection (3,4). IFN-lambda (IFNl) is the most abundant and earliest expressed interferon during influenza infection that via its cognate receptor, interferon lambda receptor 1 (IFNLR1), induces a robust antiviral response by upregulating antiviral genes (3,5).
Influenza, an RNA virus, activates the immune system via multiple mechanisms during infection. One crucial component is the innate immune response, whereby influenza viral RNA is detected by the pattern recognition receptor RIG-I, sequentially enacting MAVS oligomerization, transcriptional activation of IRF and NF-kB, and finally type I (e.g. IFNa, IFNb) and type III interferon (IFNl) production (6,7). Interestingly, while both IFNl and IFNa are antiviral, in some systems, IFNa exclusively leads to the upregulation of pro-inflammatory genes (5). Bone marrow-derived macrophages, dendritic cells, and neutrophils secrete pro-inflammatory cytokines in response to stimulation with IFNa but not IFNl (4,5). In contrast, IFNl inhibits several inflammatory mechanisms, including ROS production, granule mobilization, and the release of neutrophil extracellular traps (NETs) (8). However, cellular tropism of IFNLR1 may mediate this differential response, and it remains to be determined whether these results are reproducible in human studies (9).
In mouse models, Ifnlr1 knockout results in widespread viral dissemination and lethality (5). In addition, mice treated with IFNl after influenza infection exhibit significantly lower mortality, decreased viral burden, with reduced inflammatory cytokines compared to untreated mice (5,10). Studies have examined interferon production in the lung to varying degrees. In human cell culture, IFNL2, IFNL3, and the human-specific IFNl isoform, IFNL1, are robustly induced after influenza infection, predicting the transcription of interferon-stimulated genes (ISGs) (11). In murine models, IFNL2 and IFNL3 are produced earlier and more abundantly than type I interferons. In humans, several studies have demonstrated the presence of interferons in the infected lung. Differentiated airway epithelial cells produce IFNl in response to rhinovirus, influenza, and polyIC treatment (12,13). ATII cells primarily produce IFNl in response to influenza (14). However, there have been no indepth studies examining IFNl production and the expression of its receptor IFNR1. Differences in the resolution of viral infection may be driven by the tropism of these receptors, underscoring the importance of understanding IFNLR1 expression in the human lung.
In this study, we observed that IFNL2, IFNL3, and the human specific IFNL1 were consistently the highest expressed interferons in human airway epithelium. We also found that IFNLR1 was enriched in human alveolar macrophages (HAMs) which led to the production of ISGs in response to IFNl treatment. IFNl restricted the infection of HAMs and differentiating primary macrophages. Further, in macrophages, IFNLR1 was necessary for a robust ISG response to direct infection and was entirely responsible for ISGs produced in response to secreted interferon. These data suggest that IFNl signaling in macrophages has an important role in the sensing and response to viral infection in the human lung.

Cells and Tissue
THP-1 and HEK-293 were purchased from the American Type Culture Collection (ATCC). THP-1 cells were cultured in RPMI supplemented with 10% FBS. HEK-293 cells were cultured in DMEM supplemented with 10% FBS. CD14 monocytes (Lonza) were either untreated for monocytes, or differentiated with M-CSF or GM-CSF (R&D systems) for 7 days to produce macrophages. HAMs were isolated via ex vivo lavage from deidentified lungs rejected for transplant and obtained from Lifeline of Ohio Organ Procurements agency (Columbus, OH) as described (15). All lungs were from subjects with no history of chronic lung disease or cancer and were non-smokers for at least 1 year. After collection, red blood cells were lysed and cells were enumerated and frozen down in FBS and 10% DMSO. Prior to experiments, HAMs were rapidly thawed, added to RPMI (10% FBS, 1% antibiotic/antimycotic). Total concentration and viability of cells were determined with trypan blue staining before seeding. Differentiated human bronchial epithelial (HBE) cultures were supplied by the Cure CF Columbus (C3) Epithelial Cell Core at Nationwide Children's Hospital as described (16). Briefly, HBE progenitors were isolated from donor airways as described previously, grown for a week to confluency and frozen for later use (17). Progenitor cells were thawed and plated on 0.4 mM pore Transwells (Corning) membranes 12 mm in diameter. Medium in both chambers was replaced with fresh medium every 2-3 days. At 7 days, when the cells were confluent and had formed tight junctions as demonstrated by electrical resistance, the apical medium was removed, and the basal medium was replaced with complete Pneumacult-ALI Medium (STEMCELL Technologies). The medium was replaced with fresh medium and the apical surface was washed with 100 mL of DMEM every 2-3 days for 3 weeks at which time they had become fully differentiated. Murine bone marrow derived macrophage (BMDM) were derived from C57Bl/6 mice. Bone marrow from the femurs and tibias were collected and following RBC lysis cells were resuspended in HEPES-buffered RPMI-1640 containing Lglutamine, penicillin/streptomycin, and 10% HI-FBS with the addition of either recombinant murine GM-CSF or M-CSF. Cells were plated on petri dishes overnight and the non-adherent were passaged to new dishes to allow for expansion and differentiation of macrophages for 7 days. Macrophages were removed from petri dishes, counted, and plated into tissue culture treated wells overnight before experiment initiation. Precision cut lung slices (PCLS) were prepared as described previously (18), Briefly, transplant-rejected lungs were filled with agarose by injecting liquid agar into a lobe via the bronchi. Approximately 1 cm 3 of human lung tissue was sliced with a Vibratome into 400µM sections. Tissue was cultured in DMEM containing antibiotic/antimycotic.

Infection Protocol
Influenza PR8 and influenza CA09 strains were propagated in MDCK cells (ATCC CCL-34) (19). HAMs, CD14 macrophages, and THP-1 macrophages were infected at the indicated MOI for 1 hour, and the media was replaced. Differentiated HBECs were infected with CA09 influenza at the indicated MOI on the apical side of the transwell. After 2 h incubation, the apical layer was washed three times with DPBS to remove unbound virus. PCLS were incubated with 1x10 5 pfu PR8 and 8x10 6 pfu CA09 virus for 2 h prior to replacing the media.

Quantitative PCR
Total cellular RNA was collected from cells using the Qiagen RNeasy Miniprep plus Kit (Qiagen), following the manufacturer's protocol. The cellular RNA was then used to create cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer's protocol. qPCR was performed using SYBR Select Master Mix (Applied Biosystems) according to the manufacturer's protocol with 20 ng cDNA as a template and primer concentration of 200 nM. Each biological replicate was performed in at least technical duplicate; data was analyzed using the DDCq method. qPCR primer sequences are available in Table 1.

Single-Cell RNA Sequencing
Normal control lungs were obtained under a protocol approved by the University of Pittsburgh, Committee for Oversight of Research and Clinical Training Involving Decedents, following rejection as candidate donors for transplant. The whole lung tissue samples were processed as described previously (22). Briefly, tissue for scRNA-seq was diced and enzymatically digested in DMEM (Thermo Fisher Scientific) containing collagenase A and 30 µg/mL DNAase I and further mechanically dispersed using the Miltenyi gentleMACS Octo Dissociator. Single cell RNA library preparation was performed utilizing the 10X Genomics Chromium instrument and its associated V2 single cell chemistry per the manufacturer's protocol. In brief, 7000 cells were mixed with reverse transcription reagents, loaded into a Single Cell A chip. Cells were separated into oil micro-droplet partitions by the Chromium instrument, containing a cell and a gel bead scaffold for an oligonucleotide composed of oligo-dT, and 10X and UMI barcodes, and reverse transcription reagents. Reverse transcription was performed, the emulsion broken and pooled fractions obtained using a recovery agent. cDNAs were amplified by 11 cycles of PCR (C1000, Bio-Rad), enzymatically sheared and DNA fragment ends were repaired, A-tailed and adaptors ligated. The library was quantified using a KAPA Universal Library Quantification Kit KK4824 (KAPA Biosystems) and further characterized for cDNA length on a Bioanalyzer using a High Sensitivity DNA kit (Agilent). Single cell RNA-seq libraries were sequenced using the Illumina NextSeq-500 through the University of Pittsburgh Genomics Core, Sequencing Facility. ScRNA-seq data can be accessed at the Gene Expression Omnibus: GSE128033.

LDH assay
LDH assay was performed with the CyQuant LDH Cytotoxicity Assay (Invitrogen) according to manufacturer's instructions.

Flow Cytometry
HAMs were plated for 24h, removed from the plates using ice cold PBS and a cell scraper. Cell were then washed in PBS and resuspended in Fish Skin Gellatin (Fisher, NC0382999) buffer (1% FSG, 0.1% NaN3). Cells were then washed 1x and resuspended in primary antibody and incubated for 30 min at 4°C. After a 2 nd wash with FSG buffer, cells were fixed with 4% paraformaldehyde followed by flow cytometry analysis. Antibodies used are available in Table 2.

Influenza Infection Induces IFNL1 in the Airways
To determine the induction of interferons after influenza infection in human airways, we exposed differentiated human airway epithelial cells to pandemic 2009 influenza virus (CA09). At 24 and 48 h post-infection, the predominant interferons induced in the primary airway cells at the mRNA level were IFNb, IFNL1, and IFNL2/IFNL3 (qRT-PCR cannot distinguish these isoforms), with IFNL1 and IFNL2/3 being the most upregulated ( Figure 1A). Consistent with the gene induction, we primarily detected the secreted interferon, IFNL1, in the supernatant of infected cells ( Figure 1B). To further assess human lung responses, we infected human precision cut lung slices (PCLS) with PR8 and CA09 for 24-72h. Lung slices were viable over the course of the experiment as demonstrated by LDH release ( Figure 1C). We confirmed that these lung slices were susceptible to infection by measuring the production of viral mRNA (M gene and NS gene) ( Figure 1D). Infection of PCLS resulted in the robust secretion of IFNg, followed by IFNL1, and IFNa2, while secreted IFNb was unchanged (IFNg>>IFNL1> IFNa2>> IFNb) ( Figures 1E-H and S1A-D).
These findings indicate that IFNl is highly induced in the infected human lung environment.

Human Alveolar Macrophages Express Functional IFNLR1
Because we observed significant IFNl induction in the infected lung, we next examined the expression of its receptor, IFNLR1, via single-cell RNA sequencing. In human lung tissue, IFNLR1 is Antibodies Reagents most highly expressed in the CD163 expressing macrophage population and to a lesser degree in the epithelium (Figure 2A). To examine the expression of IFNLR1 in alveolar macrophages in the infected lung, we isolated HAMs from transplant-rejected lungs for in vitro experiments and compared them to undifferentiated CD14 cells and M-CSF differentiated macrophages. We confirmed that HAMs expressed macrophage markers (CD68 and the MHC-II subunit HLA-DR) as well as the alveolar macrophage marker (CD206) ( Figure 2B). We also observed IFNLR1 expression in GM-CSF macrophages and HAMs, but not in CD14 monocytes demonstrating that differentiation induced IFNLR1 expression, consistent with prior reports (23). Using an IFNLR1 specific antibody, we observed IFNLR1 expression in HAMs via flow cytometry. To confirm that this expression localized to CD206 positive macrophages, we co-stained with IFNLR1 and CD206. Approximately 2/3 of the CD206+ HAM population expressed IFNLR1 ( Figure 2C). To confirm antibody specificity, we measured IFNLR1 expression in two monocyte cell lines THP-1 and HL-60. We did not observe any CD206 or IFNLR1 signal in these cells ( Figure S2A, B).

Differentiation of Monocytes to Macrophages Increases IFNLR1 Expression and Function
To examine the regulation of IFNLR1 in macrophages, we measured the response of HAMs to exogenous interferons. HAMs displayed a robust ISG induction in response to IFNL1 comparable to IFNb at 24h, except for ISG15, which was more induced with IFNL1 ( Figure 3A). Both M-CSF and GM-CSF drive the differentiation of monocytes to macrophages. GM-CSF is also known to be essential for the differentiation of alveolar macrophages and has been implicated in the expression of IFNLR1 (23)(24)(25). We therefore examined the regulation of IFNLR1 expression and signaling in our system by both factors. Compared to nearly undetectable levels of IFNLR1 in monocytes, IFNLR1 was highly upregulated by differentiation with both M-CSF and GM-CSF ( Figure 3B). We then treated differentiated macrophages with IFNL1 to measure the induction of ISGs and confirmed that GM-CSF treated macrophages showed a more robust ISG response to IFNL1 than M-CSF macrophages ( Figure 3C). Treatment of GM-CSF macrophages  with either IFNb or IFNL1 led to a similar induction of ISGs ( Figure 3D). We also determined that IFNL1-3 had a similar effect on the induction of ISGs at the same concentration ( Figure 3E). To examine the role of viral infection in IFNLR1 expression, we differentiated macrophages with GM-CSF, followed by infection with influenza PR8. IFNLR1 mRNA was unchanged in these cells ( Figure S3A). Murine macrophages have been reported to be unresponsive to IFNl (26). To examine whether M-CSF or GM-CSF differentiated mouse macrophages responded to IFNl treatment, we isolated bone marrow macrophages and incubated with either M-CSF or GM-CSF. Ifnlr1 expression was relatively similar with either differentiation protocol, and stimulation with interferons did not regulate the receptor ( Figure 3F). Although both differentiation procedures led to robust responsiveness to murine IFNb, there was little to no gene induction after treatment with murine IFNL3 (Figures 3G, H) or murine IFNL2 ( Figure S3B) (mice do not express IFNL1).

Influenza Infection of Human Macrophages Leads to the Production of Lambda Interferon
We next examined anti-viral responses of M-CSF and GM-CSF differentiated macrophages to infection with influenza. We found that infection of primary macrophages with influenza PR8 ( Figure 4A) or CA09 ( Figure S4A) led to a robust increase in both IL-6 and TNFa. Other cytokines examined via multiplex ELISA were unchanged ( Figure S4B). We next examined interferons produced in response to infection. At the gene level, a wide array of type I and type III interferons were highly induced after PR8 ( Figure 4B) and CA09 ( Figure S4C). In contrast to the airway epithelium, IFNa was highly upregulated in macrophages. Of note, when we measured interferons in the supernatant, we found that IFNL1 was the most abundant secreted interferon in both GM-CSF and M-CSF macrophages in response to PR8 ( Figure 4C) and CA09 ( Figure  S4D). Interestingly, while M-CSF vs. GM-CSF differentiation did not appear to alter the pattern of the cytokine/chemokine response, the magnitude of induction was reduced in the GM-CSF differentiated macrophages compared to the M-CSF differentiated macrophages. Importantly, the levels of viral mRNAs were not significantly altered by the type of differentiation, suggesting that these cells were infected at a similar level but failed to mount a comparable immune response. This effect held true in experiments performed at the same time in identical cell populations ( Figures S5A, B). Finally, we measured the interferon response in infected HAMs. Concordant with the in vitro differentiated macrophages, HAMs primarily produced IFNl in response to influenza infection ( Figure 4D). These results suggest that macrophages are susceptible to influenza infection and contribute to the production of lambda interferons in response to infection.

IFNl Inhibits Macrophage Infection
As both macrophages and epithelial cells produce primarily type III IFN in response to viral infection, we next asked whether IFNl could inhibit the infection of macrophages. We pre-treated macrophages with IFNL1 for 24h prior to infection with influenza PR8 and assayed mRNA copy number of the two viral replication-dependent mRNAs (M and NS gene). Indeed, IFNl pre-treatment was protective against PR8 ( Figure 5A) and CA09 ( Figure 5B) infection in GM-CSF differentiated macrophages. We also observed decreased production of the inflammatory cytokines TNFa, MIP1a, and MIP1b in cells pretreated with IFNL1 ( Figures 5C, D). All three of these cytokines have been demonstrated to contribute to the inflammatory response to influenza infection (27)(28)(29). To examine the inhibition of influenza infection in HAMs, we pre-treated with IFNL1 and IFNb for 8 or 24 hours prior to influenza infection for 24 hours. As measured by the expression of the influenza viral protein M1 and M2, interferon beta led to a robust inhibition of infection, while IFNL1 led to more modest inhibition ( Figure 5E). Due to its antiviral activity and the high amount of circulating interferon lambda in the lung, IFNLR1 likely plays a key role in the inhibition of macrophage infection.

IFNLR1 Is Necessary for the Production of ISGs in Infected and Bystander Cells
The above data suggest a key role for IFNl in conferring antiviral immunity. We next assessed how the IFNl receptor alters influenza virus infection. Thus, we knocked out IFNLR1 in the THP-1 myeloid cell line. As with CD14 monocytes, THP-1 monocytes do not express IFNLR1. However, after PMA mediated differentiation, IFNLR1 mRNA was significantly increased compared to untreated THP-1 ( Figure S6A). Differentiation also increased IFNLR1 protein ( Figure S6B). Treatment of differentiated THP-1 with IFNL1 led to the induction of the ISG IFIT3 ( Figure S6C). We confirmed IFNLR1 knockout by genomic DNA sequencing ( Figure S6D) and immunoblotting ( Figure 6A). To measure IFNLR1 activity in knockout cells, we treated differentiated wild-type (Wt), sgRNA control (sgCon), and two clonal lines of IFNLR1-KO cells (sgIFNLR1 clone #1, sgIFNLR1 clone #2) with IFNL1 and measured the phosphorylation of STAT1. Knockout of IFNLR1 inhibited IFNl signaling as expected ( Figure 6B). To examine how IFNLR1 alters the response to influenza infection, we infected IFNLR1-KO macrophages with PR8 and measured the expression of ISGs. Multiple ISGs were significantly reduced in IFNLR1-KO cells vs. control lines despite expressing moderately higher levels of viral mRNA, suggesting an increased susceptibility to infection ( Figures 6C, D). The inhibition of the ISG response was also confirmed in a second IFNLR1-KO clonal cell line ( Figure S6E). In contrast to the decreased interferon response, and consistent with the increase in viral mRNA, we observed an increased in several secreted cytokines IL1-b, TNF-a, IL-6, and MIP1-a in infected IFNLR1-KO cells.
Interestingly, MIP1-b and MCP-1 secretion was reduced in IFNLR1-KO THP-1, potentially because the interferon pathway is required for the transcription of these genes (28,30) (Figures 6E, F). Alveolar macrophages are both the direct targets of infection, and respond to secreted interferons produced by neighboring airway epithelial cells. To assess cross-talk between human airway epithelia and macrophages, we incubated sgCon and sgIFNLR1 THP-1 cells with supernatants of CA09 infected or non-infected primary differentiated human airway epithelial cells (supernatants characterized in Figure 1B). Treatment of control sgRNA THP-1 cells with virus-infected supernatant that (containing lambda interferons) resulted in the upregulation of a number of ISGs ( Figure 6G). However, IFNLR1 depleted THP-1 had a significantly abrogated influenza media-stimulated ISG response compared to sgCon THP-1 treated with the same media. Collectively, these observations strongly suggest that IFNLR1, via engagement of its IFNl ligands, is crucial for mediating antiviral immunity.

DISCUSSION
While both IFNl and IFNa induce antiviral genes, IFNl has been suggested to maintain antiviral function without the inflammatory responses observed with type I interferons (5). This has led to significant interest in the type III interferons as potential therapeutics against influenza and other viral infections (10,31,32). Compared to type I interferons, murine IFNl signaling alters barrier surfaces, such as the gastrointestinal epithelial layer (33)(34)(35)(36). However, data on IFNl and IFNLR1 in human infection is limited and the murine expression of Ifnlr1 (6, 37, 38) does not appear to match the tropism observed in human cells. In particular, in murine models, macrophages do not respond to secreted IFNl, suggesting that humans have evolved species-specific adaptive responses through the interferon network for protection against common viral pathogens (5,26,39). Here, we have described the expression, regulation, and antiviral activity of the IFNl-IFNLR1 axis in human lung. We observed abundant IFNLR1 expression in human alveolar macrophages and found that these cells are basally capable of responding to lambda interferons. In addition, IFNl was protective against influenza infection in primary alveolar macrophages. IFNLR1 was also necessary for the production of ISGs in response to secreted factors from infected cells. These data suggest that IFNl signaling in macrophages is integral to the response to influenza infection in the human lung.
In both human and murine models of influenza infection, IFNl is produced earlier and more abundantly than type I interferons (5,6). In our studies, polarized airway epithelial cells produced high amounts of IFNl with lower IFNb and very little IFNa in response to influenza. In infected lung tissue sections, IFNg was the most upregulated interferon followed by IFNl. The high amount of IFNg signaling in the lung tissue is likely the result of resident NK cells and lung resident memory T-cells. In ex vivo lung tissue sections macrophages and epithelial cells are directly infected with influenza. Neighboring NK cells in the tissue sections produced high amounts of IFNg in the infected environment (40). In addition, lung-resident memory CD8 T-cells, present in human lung tissue sections, contribute to interferon gamma (IFNg) production in response to influenza (41,42). When we directly examined infected macrophages, we found that IFNl was highly induced in response to infection. Although prior studies in murine models suggested monocytes and macrophages do not express IFNLR1 and do not respond to IFNl (26,38), more recent observations suggest that human macrophages possess the necessary machinery to respond to lambda interferons. First, CD14 monocytes differentiated with GM-CSF were responsive to IFNl treatment (23). More recently, alveolar macrophages derived from bronchoalveolar lavage produced ISGs in response to IFNl treatment (43). By examining IFNLR1 expression in a scRNA-seq dataset, we found a high degree of IFNLR1 expression in the alveolar macrophages subset. We also detected a significant ISG induction in IFNl treated HAMs. While IFNLR1 expression and IFNl signaling was induced by the differentiation of human myeloid cells to macrophages, we did not observe IFNl signaling in differentiated murine macrophages. Interestingly, porcine alveolar macrophages have also been shown to respond to IFNl (44). Claims that IFNl plays an anti-inflammatory role are, in part, based on the lack of IFNLR1 expression in monocyte and macrophages (45). Therefore, the species-dependent expression of macrophage IFNLR1 warrants further exploration and should be considered when studying interferon responses in the murine model. Experimental depletion of alveolar macrophages with clodronate prior to influenza infection resulted in higher viral titers and mortality in mouse and pig models (46,47). Genetic depletion of alveolar macrophages prior to influenza infection increases morbidity and mortality and can be rescued with the adoptive transfer of alveolar macrophages (46,48). Different strains of influenza infect myeloid cells at varying efficacies, with pandemic H5N1 strains showing a higher propensity to infect were differentiated with PMA (20 ng/mL) for 7 days prior to harvest. IFNLR1 protein was measured via immunoblotting (A). sgCon and sgIFNLR1 cells lines were differentiated as described above. At 30 minutes prior to harvest, cells were treated with 50 ng/mL IFNL1. Cells were harvested and total STAT1 and phosphorylated STAT1 induction (pSTAT1) was measured via immunoblotting (B). Representative of 2 independent experiments (A, B). sgCon and sgIFNLR1 were differentiated as above followed by PR8 infection (MOI=0.01) for 24 h. We then measured fold change in ISGs by qRT-PCR. *p < 0.05 vs. sgCon + PR8 by Students T-test (C), fold change in viral mRNA (M gene and NS gene) vs. uninfected (D), and secreted inflammatory cytokines/chemokines in influenza infected macrophages using a custom multiplex from MSD. ns, not significant. *p < 0.05, **p < 0.01, ***p < 0.001 vs. sgCon + PR8 by Multiple T-test with Holm-Sidak correction (E, F). representative of 3 independent experiment, n=3 (C-F). sgCon and sgIFNLR1 THP-1 were differentiated with PMA and treated with supernatants (culture medium) collected from human airway epithelia previously infected with (flu+) or without influenza (flu-) (CA09, MOI=1) ***p < 0.001 vs. sgCon (flu-); ### p < 0.001 vs. sgCon (flu+) by Multiple T-test with Holm-Sidak correction (G).
macrophages (49,50). Lung macrophages are both direct targets of influenza infection and respond to cytokines and chemokines produced by neighboring infected cells. Interestingly, recruited monocytes and resident macrophages are responsive to circulating interferons while viral infection blunts the ISG induction in the epithelium (51). These data strongly implicate a key role of interferon signaling in myeloid cells that may be critical in viral infections. These data also suggest resident IFNLR1 expressing lung macrophages respond differently to circulating IFNl than recruited monocytes that lack IFNLR1. To our knowledge, there have been no direct examinations of how IFNLR1 alters viral infection and the ISG response in macrophages. We found that knockout of IFNLR1 significantly reduced the production of both interferons and ISGs from directly infected cells despite moderate increase in viral mRNA. In contrast, most inflammatory cytokines were increased in these cells including TNFa, IL1b, IL-6, and MIP1a. The notable exceptions were CCL2 and CCL4, whose production have been shown to be interferon dependent (28,30). This was consistent with our observations in primary cells, where pretreatment with IFNL1 blunted the inflammatory response by limiting viral infection.
Overall, these results suggest that IFNLR1 limits viral infection and thereby dampens the inflammatory response. Although macrophages are a reservoir for viral infection, the vast majority of infected cells are epithelial. Therefore, we also examined how IFNLR1 detects and responds to secreted factors from nearby infected cells. As expected, supernatants produced from infected primary airways cells induced ISGs from macrophages. Importantly, this response was completely attenuated after IFNLR1 knockout.
Interestingly, we also found that alternative mechanisms of macrophage differentiation change the response to influenza infection. GM-CSF is necessary for the development of alveolar macrophages in both mice and humans. ATII cells produce high levels of GM-CSF that drives initial development and maintenance of alveolar macrophages (25) Differentiation with GM-CSF also results in a more robust IFNl response vs. M-CSF mediated differentiation (23). Our data suggests that GM-CSF macrophages produce fewer cytokines and interferons in response to influenza infection despite producing higher levels of viral mRNA. Although the magnitude of the chemokine response was reduced in the GM-CSF cells, the overall profile of the response was similar in both groups. Further studies are needed to determine how these factors alter the inflammatory profile of infected macrophages.
In summary, we demonstrate that IFNl is highly induced in the human lung after influenza infection and that resident human alveolar macrophages are responsive to this cytokine. We demonstrate that macrophage IFNLR1 is necessary for the induction of ISGs in response to both direct infection and in response to secreted factors from neighboring infected cells. We also show that murine macrophages are minimally responsive to IFNl, suggesting that IFNLR1 regulation is species dependent. Due to the high abundance of IFNl production early in infection, the proximity of alveolar macrophages to the site of infection, and the necessity of IFNLR1 to mount an antiviral response to these secreted factors, IFNl signaling in macrophages likely plays an important role in the response to viral infection.

STATISTICS
Statistics were performed using GraphPad Prism. Student t-tests were performed for two samples or multiple T-test with Holm-Sidak correction were performed for three or more samples.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included inthe article/Supplementary Material. Further inquiries can bedirected to the corresponding author. ScRNA-seq data can be accessed at the Gene Expression Omnibus: GSE128033.

ETHICS STATEMENT
These experiments were conducted in compliance with biosafety and laboratory biosecurity regulations, guidelines, standards, policies and procedures. The Institutional Animal Care and Use Committee at The Ohio State University reviewed and approved all animal procedures (IACUC protocol #2019A00000019). Human alveolar macrophages and lung slices were derived from de-identified samples and were not subject to IRB oversight.

AUTHOR CONTRIBUTIONS
JL and RM designed the study, performed experiments, analyzed results and wrote the manuscript. JA, AE, DF, LC, ML, MD, and VP and performed experiments and analyzed data. RM, ML, JB, AM, and MR reviewed the data and edited the manuscript. All authors contributed to the article and approved the submitted version.

FUNDING
This work was supported by NIH grants UH3HL123502, R01HL096376, R01HL097376, R01HL098174, R01HL081784, and P01HL114453 to RM. ML was supported by the Cystic Fibrosis Foundation award LONG19F5-CI.

ACKNOWLEDGMENTS
Human bronchial epithelial cultures for this work and advice and tools for working with them were supplied by the Cure CF Columbus (C3) Epithelial Cell Core at Nationwide Children's Hospital. C3 is supported by a Research Development Program Grant (MCCOY19R0) from the Cystic Fibrosis Foundation.