IL-36Γ Protects Against Severe Influenza Infection by Promoting

Home , IL36A, IL36B, IL36G, Interleukin, Interleukin 36

IL-36γ Protects against Severe Influenza Infection by Promoting Lung Alveolar Macrophage Survival and Limiting Viral Replication This information is current as of September 25, 2021. Alexander N. Wein, Paul R. Dunbar, Sean R. McMaster, Zheng-Rong Tiger Li, Timothy L. Denning and Jacob E. Kohlmeier J Immunol published online 30 May 2018 http://www.jimmunol.org/content/early/2018/05/29/jimmun Downloaded from ol.1701796

Supplementary http://www.jimmunol.org/content/suppl/2018/05/29/jimmunol.170179 http://www.jimmunol.org/ Material 6.DCSupplemental

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 25, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2018 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published May 30, 2018, doi:10.4049/jimmunol.1701796 The Journal of Immunology

IL-36g Protects against Severe Inﬂuenza Infection by Promoting Lung Alveolar Macrophage Survival and Limiting Viral Replication

Alexander N. Wein,* Paul R. Dunbar,* Sean R. McMaster,* Zheng-Rong Tiger Li,* Timothy L. Denning,† and Jacob E. Kohlmeier*

Although influenza virus infection remains a concerning disease for public health, the roles of individual cytokines during the immune response to influenza infection are not fully understood. We have identified IL-36g as a key mediator of immune protection during both high- and low-pathogenesis influenza infection. Il36g mRNA is upregulated in the lung following influenza infection, and mice lacking IL-36g have greatly increased morbidity and mortality upon infection with either H1N1 or H3N2 influenza. The

increased severity of influenza infection in IL-36g–knockout (KO) mice is associated with increased viral titers, higher levels of Downloaded from proinflammatory cytokines early in infection, and more diffuse pathologic conditions late in the disease course. Interestingly, the increased severity of disease in IL-36g–KO mice correlates with a rapid loss of alveolar macrophages following infection. We find that the alveolar macrophages from naive IL-36g–KO mice have higher expression of M2-like surface markers compared with wild-type (WT) mice and show increased apoptosis within 24 h of infection. Finally, transfer of WT alveolar macrophages to IL- 36g–KO mice restores protection against lethal influenza challenge to levels observed in WT mice. Together, these data identify a

critical role for IL-36g in immunity against inﬂuenza virus and demonstrate the importance of IL-36g signaling for alveolar http://www.jimmunol.org/ macrophage survival during infection. The Journal of Immunology, 2018, 201: 000–000.

espite the broad availability of licensed vaccines, influ- and chemotactic factors (3–6). IFNs serve to limit protein trans- enza viral illness remains a major threat to public health, lation within infected cells and decrease the permissibility of D particularly among children and the elderly (1, 2). In uninfected cells to the virus during the mobilization and recruit- severe cases, influenza infection can cause primary viral pneu- ment of immune cells from the bone marrow and circulation, monia or sensitize the host to secondary bacterial infection and which can take 4–5 d (7–9). Several innate cell types have been can also exacerbate underlying health problems, including con- shown to be key to controlling influenza infection. NK cells can gestive heart failure, chronic obstructive pulmonary disease, and recognize infected cells via NKp46 binding to surface hemag- by guest on September 25, 2021 asthma. One reason for the continued public health burden of glutinin, leading to lysis of cells that display viral proteins (8). influenza is the imperfect antigenic match between the vaccine Neutrophils can also serve to promote viral clearance in severe and circulating influenza strains that results in only 50–80% cases of influenza, but neutrophils, along with monocytes, have protection in the best of years (2). In the absence of effective been shown to contribute to immune pathologic condition during cellular or humoral memory responses, the innate defenses of the some models of influenza infection (10, 11). In contrast, among respiratory tract serve as the only mechanisms to control the the earliest responders to influenza infection in the lung are tissue- spread of virus in the early phases of infection. resident alveolar macrophages. In mice, these cells have been Influenza infection of the respiratory epithelium leads to the shown to decrease the infection of type I alveolar epithelial cells activation of antiviral responses via the extracellular and endo- during the initial 48 h of infection, before other innate cell types somal TLR receptors or the intracellular receptors RIG-I and arrive in the lung (12). Depletion of alveolar macrophages by NLRP3, causing cells to release IFNs as well as proinflammatory either genetic knockout (KO) or liposomal clodronate treatment increases morbidity, mortality, and viral load during influenza infection (12, 13). Therefore, lung alveolar macrophages play a *Department of Microbiology and Immunology, Emory University School of Med- icine, Atlanta, GA 30322; and †Institute for Biomedical Sciences, Georgia State critical role as one of the few lung-resident innate immune cell University, Atlanta, GA 30303 types capable of responding to influenza infection during the ORCIDs: 0000-0002-8813-3523 (A.N.W.); 0000-0002-9082-6636 (P.R.D.). initial stages of infection, prior to immune recognition that leads Received for publication January 2, 2018. Accepted for publication May 8, 2018. to the initial inflammatory burst alerting the systemic immune This work was supported by National Institutes of Health Grant R01HL122559 system to a lung infection. (to J.E.K.) and Emory University. P.R.D. was supported by National Institutes of Health Severe influenza infections are characterized by viral infection Grant F31AI124611 and S.R.M. was supported by National Institutes of Health Grant F30HL118954. throughout the upper and lower respiratory tract (14). Conse- quences of increased viral titers throughout the respiratory tract Address correspondence and reprint requests to Dr. Jacob E. Kohlmeier, Emory University School of Medicine, 1510 Clifton Road, RRC, Room 3133, Atlanta, include hyperactive cytokine responses, and patients hospitalized GA 30322. E-mail address: [email protected] with severe influenza showed significant increases in proin- The online version of this article contains supplemental material. flammatory cytokines (15). Elevated expression of type I IFNs,

Abbreviations used in this article: BAL, bronchoalveolar lavage; EID50, 50% egg TNF-a, IL-1, IL-6, IL-18, and IL-33 have been observed in infectious dose; KO, knockout; NP, nuclear protein; qPCR, quantitative PCR; WT, pathogenic inﬂuenza infections, suggesting that inhibition of these wild-type. cytokines may ameliorate inﬂuenza pathologic conditions and Copyright Ó 2018 by The American Association of Immunologists, Inc. 0022-1767/18/$35.00 improve host survival (16). However, deletion or blockade of

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1701796 2 IL-36g IN INFLUENZA INFECTION proinflammatory cytokine pathways, including IFN-a,IFN-g,TNF, TGC CAT TCT GCC GCA TT-39, x31 NP reverse: 59-GCT GAT TTG IL-1, and IL-6, has shown mixed results in mice and humans (9, 17– GCC CGC AGA TG-39, probe: 59-/5HEX/TA+G T+CT +CCA TAT TTT 24). General immunosuppression with corticosteroids has also shown CAT T+G+G AA+G C/3BHQ_1/-39. Expression levels of IL-36 family members were determined by real-time quantitative PCR (qPCR) on a Bio- little to no effect on patient outcome in clinical trials (24–26). These Rad CFX96 Real-Time system and CFX Manager v3.1 using primers failures have renewed efforts to identify novel proinflammatory host against Il36a, Il36b, Il36g, and Hprt. Sequences were Il36a forward: 59- mediators that contribute to influenza pathogenesis. One such family TAG TGG GTG TAG TTC TGT AGT GTG-39, Il36a reverse: 59-GTT of proinflammatory mediators is the IL-36 family of cytokines, CGT CTC AAG AGT GTC CAG ATA T-39, Il36b forward: 59-ACA AAA AGC CTT TCT GTT CTA TCA T-39, Il36b reverse: 59-CCA TGT ATT consisting of IL-36a,-36b,and-36g, which are recently discovered TAC TTC TCA GAC T-39, Il36g forward: 59-AGA GTA ACC CCA GTC members of the IL-1 family. These proteins have been shown to be AGC GTG-39, Il36g reverse: 59-AGG GTG GTG GTA CAA ATC CAA-39, important in several different inflammatory conditions, including Hprt forward: 59-CAC AGG ACT AGA ACA CCT GC-39, Hprt reverse: asthma, colitis, and psoriasis (27–29). In addition, IL-36g adminis- 59-GCT GGT GAA AAG GAC CTC T-39. For Il36g expression in sorted tration has been shown to increase mucus production and neutrophil immune cells, cDNA was synthesized with the iScript cDNA synthesis kit before real-time qPCR (Bio-Rad). Real-time qPCR data were analyzed migration into the lungs of mice (28). A recent study showed that IL- using CFX Manager before export to Microsoft Excel. 36g–KO mice were more susceptible to pulmonary Streptococcus pneumoniae infection, demonstrating an important role for IL-36g in Lung harvest for pathological examination lung immunity (30). In addition, deletion of IL-36R has been shown Mice were euthanized as above, and the ribcage was carefully removed. to improve survival during lethal influenza challenge in mice (31). A small incision was made in the trachea, and the lungs were inflated with However, the role of individual IL-36 cytokines in immunity against 0.75 ml formalin. The lungs, heart, and thymus were removed en bloc and respiratory virus infections has not been investigated. placed into a cassette and then into formalin jars. Lungs were embedded in Downloaded from paraffin, sectioned, and stained with H&E by the Winship Research Pa- In the current study, we investigated the regulation of IL-36 cy- thology Core Lab at Emory University. Lungs were scored by a pathologist tokines during mild and severe influenza infections in mice. Il36g blinded to treatment conditions. mRNA was significantly increased in the lung during the initial stages of influenza infection, suggesting that inhibition of IL-36g Isolation of immune cells from lungs, flow cytometry, and production may prevent excessive inflammation during severe in- cytokine analysis

fluenza infection and improve host survival. Surprisingly, infection For isolation and quantification of immune cells from lungs, mice were http://www.jimmunol.org/ of IL-36g–KO mice with influenza virus showed the opposite effect, euthanized as above. Bronchoalveolar lavage (BAL) was performed using as IL-36g–KO mice had significantly increased weight loss, mortal- 1 ml of compete RPMI 1640 delivered via an 18-gauge IV catheter into a ity, and viral titers during both mild and severe influenza infection. small incision into the trachea. The airways were washed five times before the lungs were harvested into HBSS. Lungs were finely minced with Analysis of innate immune cell subsets in the lung revealed a rapid scissors before digestion (30 min, 37˚C) with warm HBSS containing 4E5 and dramatic loss of lung alveolar macrophages in IL-36g–KO mice U/ml DNase (Sigma-Aldrich) and 500 mg/L Collagenase D (Roche). compared with their wild-type (WT) counterparts. Finally, the de- Lungs were further homogenized via swishing with a 3-ml syringe before creased survival in IL-36g–KO mice could be rescued through the and after digestion and every 10 min during digestion. Following digestion, immune cells were purified via 40/80% isotonic Percoll gradient centri- transfer of WT alveolar macrophages into the lung prior to influenza fugation. Live/dead staining was performed with Zombie Yellow, Zombie infection. Overall, these data demonstrate an unexpected protective UV, or Zombie NIR (BioLegend), and Fc blocking was done with anti- by guest on September 25, 2021 role for IL-36g during influenza infectioninpromotingthesurvival CD16/32 clone 2.4G2 in FACS buffer consisting of 2% w/v BSA and of lung alveolar macrophages. 0.05% w/v sodium azide in PBS. The hybridoma expressing 2.4G2 was purchased from American Type Culture Collection and cultured in our laboratory. The Ab was purified from the supernatant by standard tech- Materials and Methods niques. Cells were stained with fluorescently labeled Abs purchased from Mice Abcam, BioLegend, BD Biosciences, or eBioscience in FACS buffer. Abs were anti–CD45.2-A488 (104; BioLegend), anti–CD45.2-A647 (104; a b C57BL/6J (WT) and B6.SJL-Ptprc Pepc /BoyJ (CD45.1) mice were pur- BioLegend), anti–CD86-PE (GL1; BioLegend) anti–CD11c-PE/CF594 chased from The Jackson Laboratory and bred in house under specific (HL3; BD Biosciences), anti–CD103-PerCP/Cy5.5 (2E7; BioLegend), pathogen-free conditions at Emory University. For some experiments, WT anti–I-A/I-E-PE/Cy7 (M5/114.15.2; BioLegend), anti–CD45R-allophyco- + + and CD45.1 mice were crossed to produce CD45.1 CD45.2 F1 offspring. cyanin (RA3-6B3; BioLegend), anti–Ly6C-allophycocyanin/Cy7 (AL-21; tm1Lex Cryopreserved sperm of the line B6;129S5-Il1f9 /Mmucd (IL-36g–KO) BD Biosciences), anti–SiglecF-BV421 (E50-2440; BD Biosciences), anti– was purchased from the Knockout Mouse Project, injected into a C57Bl6/J Ly6G-BV510, (1A8; BioLegend), anti–NK1.1-BV650 (PK136; Bio- oocyte, and backcrossed in house (32). Single nucleotide polymorphism Legend), anti–CD11b-BV711 (M1/70; BioLegend), anti–CD90.2-BV785 analysis was performed by The Jackson Laboratory to confirm backcrossing (30-H12; BioLegend), anti–CD206-A700 (C068C2; BioLegend), anti– to the C57BL/6J strain (Supplemental Table I). Intranasal infection with CD200R-PE (OX-110; BioLegend) anti–TREM2-FITC (78.18; Abcam), influenza A/HKx31 (H3N2) at 30,000 or 300,000 50% egg infectious dose anti-CCR5-PerCP/e710 (7A4; eBioscience), and anti–CD45.1-BUV395 (EID50) and PR8 (H1N1) at 450 PFU were performed under isoflurane an- (A20; BD Biosciences). Cells were sorted on a BD FACSAria II or were esthesia. For all mortality experiments, mice were monitored with daily fixed with 1% paraformaldehyde before analysis on a BD LSR II or BD weighing and were euthanized when they reached 25% weight loss. All LSRFortessa X-20. Annexin V (BioLegend) staining was performed experiments were completed in accordance with the Institutional Animal according to manufacturer’s instructions immediately before flow cyto- Care and Use Committee guidelines of Emory University. metric analysis. Cell numbers were determined by counting live cells on a hemocytometer in the presence of trypan blue (Sigma-Aldrich). Flow Whole lung mRNA isolation for viral titers and Il36 cytometry data were initially analyzed using FlowJo v10.0.8r1 before quantitative PCR being exported to Microsoft Excel for further calculations. For cytokine analysis, the first BAL pull was kept separate and spun at 700 3 g for Infected mice were euthanized by 2,2,2-Tribromoethanol overdose and 5 min to remove cells. The supernatant was stored at 280˚C, whereas the brachial exsanguination. Lungs were harvested into RNAlater (Sigma- cells were combined with the rest of the BAL pulls. Cytokine levels in the Aldrich) and stored at 4˚C overnight and then at 280˚C until processing. BAL supernatant were determined by a BioLegend LEGENDplex Mouse Lungs were homogenized in Trizol (Ambion) by either Tissue-Tearor or Inflammation Panel by manufacturer’s instructions. BeadBug according to the manufacturer’s directions. RNA was purified from the homogenized lung using Ambion Spin Columns by a modified Cell isolation and adoptive transfer manufacturer’s protocol. cDNA was synthesized from RNA using reverse transcriptase (Applied Biosystems), and viral titers were determined by For transfer of alveolar macrophages, lungs were harvested from naive B6 digital droplet and QuantaSoft software (Bio-Rad) using primers and or IL-36g–KO mice and digested (30 min, 37˚C) with Collagenase Type II probes directed at the influenza NP gene, which is conserved between x31 (625 U/ml; Worthington Biochemical) and DNase (4E5 U/ml; Sigma- and PR8. The sequences were x31 nuclear protein (NP) forward: 59-GCA Aldrich) and swished as above. Percoll purification was performed as The Journal of Immunology 3 above. Alveolar macrophages were isolated via positive selection on CD11c during influenza infection. As shown in Fig. 3A, we found by and SiglecF using MACS columns per manufacturer’s instructions (Miltenyi digital droplet PCR that IL-36g–KO mice have increased viral Biotec). The bound fraction was collected, and an aliquot was stained to load 6 d postinfection with PR8. Likewise, following infection determine purity. Five hundred thousand alveolar macrophages in 50 mlPBS were administered intranasally to recipient mice under isoflurane anesthesia. with a low dose of x31 influenza, IL-36g–KO mice have increased WT controls received PBS intranasally. For survival experiments, mice were viral load at days 6 and 9 postinfection (Fig. 3B). Mice which infected with high-dose x31 the following day and monitored daily for 14 d. were infected with the high dose of x31 had increased viral titers For next day analysis of transfer efficiency or phenotypic analysis of KO as soon as 3 d postinfection and continued to have increased titers alveolar macrophages transferred to WT mice, the mice were not infected. until 6 d postinfection, when the majority of the mice reached Statistics humane end points (Fig. 3C). We also analyzed the cytokines present in the BAL fluid of WT and IL-36g–KO mice infected Statistical analysis was performed in GraphPad Prism, and the tests used to with high dose x31 because these mice have a large difference in determine significance are listed in the appropriate figure legends. Asterisks in the figures denote significance (*p # 0.05, **p , 0.01, ***p , 0.001, viral titers early in infection. We found that compared with WT ****p , 0.0001). mice, the BAL fluid of IL-36g–KO mice had increased IFN-b and IL-6 but not IFN-g, TNF, or IL-1a (Fig. 3D). To further under- Results stand the role of IL-36g during influenza infection, formalin-fixed lungs from WT and IL-36g–KO mice infected with high-dose x31 IL-36g is induced during influenza infection and protects were evaluated for inflammation and damage by an independent against morbidity and mortality pathologist blinded to treatment conditions. On day 6 postinfec-

To determine the expression level of IL-36 family members fol- tion, IL-36g–KO mice had more diffuse immune infiltrate around Downloaded from lowing influenza infection, B6 mice were infected with influenza the bronchioles compared with WT mice (Fig. 3E). Together, A/HKx31 (H3N2), and expression of the IL-36 cytokine family these data point to a role for IL-36g in controlling the early viral genes Il36a, Il36b, and Il36g in whole lungs was measured by replication and limiting the resulting immunopathologic condition real-time qPCR (Fig. 1A–C). Il36a expression was not altered following influenza infection. following infection, and Il36b was significantly increased only on Alveolar macrophages are rapidly depleted following influenza day 3 following influenza infection. We observed that Il36g ex- http://www.jimmunol.org/ pression in the lung was significantly increased by day 3 postin- infection in IL-36g–KO mice fection, and increased expression of Il36g was maintained through The increased morbidity and viral titers observed in IL-36g–KO day 6. In addition, expression levels of Il36g in naive (day 0) micewereevidentwithin3dpostinfection,suggesting a defect whole lungs were higher than either Il36a or Il36b (Supplemental in early innate immunity. Because many different immune cell Fig. 1). Il36g expression has previously been reported in bronchial types contribute to the early response during influenza infection epithelial cells following infection with respiratory viruses or (6), we examined multiple innate and adaptive immune cell bacteria, but whether immune cells produce the cytokine during subsets in WT and IL-36g–KO mice following influenza x31 influenza infection is not known (30, 33). As Il36g expression has infection using a 13-color flow cytometry panel (Supplemental been observed in multiple immune cell types in different models Fig. 2). As shown in Fig. 4A and 4B, the number of alveolar by guest on September 25, 2021 of inflammation, we examined several types of leukocytes for macrophages was significantly decreased in IL-36g–KO mice on Il36g RNA 6 d after influenza infection. Surprisingly, among days 3 and 6 postinfection, whereas no other immune cell subsets immune cells examined, neutrophils were the highest producers of examined showed any differences across both time points. Al- Il36g mRNA at day 6 postinfection (Fig. 1D). These data show though there were increased numbers of CD11bhi dendritic cells that IL-36 family members, most notably IL-36g, are induced in in IL-36g–KO mice on day 3 postinfection, this difference was the lung following influenza infection and suggest a potential role not evident at day 6 and seemed unlikely to account for the in- of IL-36g in mediating influenza immunity. creased severity of influenza infection in IL-36g–KO mice. To Highly pathogenic influenza infections are associated with high determine whether the decrease in alveolar macrophages was the levels of proinflammatory cytokines such as TNF-a and IL-1, and result of a homeostatic defect in IL-36g–KO mice, we next some studies have shown that blocking these cytokines limited im- compared the frequency and number of lung alveolar macro- munopathologic condition and improved survival in murine models phages in WT and IL-36g–KO mice prior to, or following, in- (20, 21). Given the proinflammatory role for IL-36g in other dis- fluenza infection. As shown in Fig. 4C and 4D, alveolar eases, we examined whether deletion of Il36g could decrease mor- macrophage frequency and numbers are similar in naive WT and bidity and mortality following influenza A/PR8 (H1N1) infection. IL-36g–KO mice but rapidly decline in IL-36g–KO mice fol- AsshowninFig.2A,IL-36g–KO mice had significantly greater lowing infection. Thus, in the absence of IL-36g, lung alveolar mortality following infection, indicating that IL-36g had a protective macrophages are rapidly lost following influenza infection. rather than pathogenic role in influenza immunity. To confirm a protective role for IL-36g during influenza infection, we measured Alveolar macrophages from IL-36g–KO mice have an survival following infection with the less pathogenic influenza x31 M2-like phenotype (H3N2) virus. Compared with WT mice, IL-36g–KO mice showed We next investigated the potential mechanism governing alveolar significantly increased weight loss beginning on day 3 postinfection, macrophage loss in IL-36g–KO mice following influenza infec- as well as significantly greater mortality, with two different doses of tion. In naive mice, there was no difference in the frequency of influenza x31 (Fig. 2B, 2C). Together, these data demonstrate a live or early apoptotic alveolar macrophages by annexin V or live/ surprising protective role for IL-36g in improving survival following dead staining (Fig. 5A). However, within 1 d of influenza infec- influenza infection with different strains and doses. tion, we observed a significant decrease in live alveolar macrophages in IL-36g–KO mice compared with WT mice, with a The absence of IL-36g results in increased viral replication corresponding increase in early apoptotic cells (Fig. 5B). To better and immunopathologic condition understand this difference in apoptosis, we examined if there were We next examined whether the decreased survival in mice lacking differences in M1/M2 macrophage skewing between WT and IL- IL-36g was associated with decreased control of viral replication 36g–KO mice, as M1-like macrophages are more resistant to 4 IL-36g IN INFLUENZA INFECTION

apoptosis (34). We examined the surface expression of classical M1 proteins CD86 and MHC class II, M2 proteins CD206 and CD200R, the prosurvival proteins TREM2 and CCR5, and CD11b on alveolar macrophages from WT and IL-36g–KO mice (35–37). Alveolar macrophages isolated from naive IL-36g–KO mice had a significant increase in the expression of CD206 and CD200R, markers associated with an M2-like macrophage (Fig. 5C). To confirm this result, we sorted out alveolar macrophages from naive WT and IL-36g–KO mice and performed real- time qPCR for the hallmark M2 gene Retnla. We found that IL-36g–KO alveolar macrophages had greater expression of Retnla compared with their WT counterparts, supporting our flow cyto- metric results (Fig. 5D). We considered the possibility that IL-36g may have a role in the conditioning of alveolar macrophages and may impact macrophage polarization. Thus, we isolated and transferred alveolar macrophages from IL-36g–KO into the lungs of WT mice and allowed the cells to rest for 10 d (Fig. 5E). When the phenotype of the transferred cells was compared with the pre-

transfer phenotype, levels of surface CD200R and CD206 decreased Downloaded from over the 10 d in the presence of IL-36g (Fig. 5F). In fact, CD200R expression was equivalent to the endogenous WT alveolar macrophages, whereas CD206 expression remained slightly elevated. Together, these data show that the loss of alveolar macrophages in IL-36g–KO mice following inﬂuenza infection is associated

with a rapid increase in apoptosis, that alveolar macrophages in http://www.jimmunol.org/ IL-36g–KO mice have a phenotype consistent with a more M2-like macrophage polarization, and that IL-36g plays a role in alveolar macrophage polarization that may contribute to their decline following inﬂuenza infection. Transfer of WT alveolar macrophages restores protection against inﬂuenza infection in IL-36g–KO mice Alveolar macrophages are known to be critical for the early control

of viral replication and protection against severe influenza infection by guest on September 25, 2021 (12, 13). Therefore, we sought to determine if complementing IL- 36g–KO mice with WT alveolar macrophages prior to influenza infection would be sufficient to protect against severe influenza disease. To confirm that our method transferred alveolar macrophages in physiologically relevant numbers, WT alveolar macrophages were transferred to IL-36g–KO mice, and the BAL and lung were harvested the next day (Fig. 6A). We found that 30% of alveolar macrophages in the BAL and 15% of those in the lung were transferred WT cells, confirming that physiological numbers of cells were transferred (Fig. 6B). Next, we isolated alveolar macrophages from the lungs of naive WT or IL-36g–KO mice and transferred equal numbers intranasally to separate cohorts of IL- 36g–KO mice 1 d prior to influenza infection (Fig. 6C). As shown in Fig. 6D, IL-36g–KO mice that received WT alveolar macrophages showed a significant increase in survival compared with IL-36g–KO mice that received IL-36g–KO alveolar macrophages. The survival of IL-36g–KO mice that received WT alveolar macrophages was equivalent to WT mice. Thus, the defect in influenza immunity observed in IL-36g–KO mice resulting in severe disease can be rescued by the presence of WT alveolar macrophages.

FIGURE 1. Real-time qPCR of whole lung homogenate or sorted cells for expression of Il36 cytokines. The expression levels of (A) Il36a,(B) Expression levels of Il36g (D) from several immune cell types as indicated Il36b, and (C) Il36g were determined by real-time PCR relative to the below the x axis. Asterisks indicate significantly different expression housekeeping gene Hprt in naive C57BL/6J mice and at days 3, 6, and 9 from uninfected lungs by Student t test on logarithmically transformed postinfection with 30,000 EID50 of influenza A/HKx31. Data are combined data (A–C) or ANOVA followed by Holm–Sidak multiple comparisons test from three independent replicates of at least five mice per group. Males on log-transformed data (D). Asterisks denote significance: *p , 0.05, and females were used in equal proportions. (Figure legend continues) **p , 0.01, ***p , 0.001, ****p , 0.0001. The Journal of Immunology 5

FIGURE 2. Survival and weight loss of WT and IL-36g–KO mice following influenza infection. Age- and sex-matched C57BL/6J (solid lines) and IL- Downloaded from 36g–KO mice (dashed lines) were infected with (A) influenza PR8, (B) 30,000 EID50 x31, or (C) 300,000 EID50 x31 and monitored daily for weight loss. Data are combined from two to three independent replicate experiments of at least 10 mice per group per replicate. Male and female mice were used in equal proportions. Asterisks indicate significant differences in survival by the log-rank test or differences in weight loss by Student t test. Asterisks denote significance: *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001.

Discussion more diffuse pathologic conditions in IL-36g–KO mice. Analysis http://www.jimmunol.org/ We have shown that IL-36g is uniquely upregulated among IL-36 of innate cell subsets revealed a significant decrease in alveolar cytokines during the first 6 d of influenza infection. Mice lacking macrophages in IL-36g–KO mice as soon as 3 d postinfection that IL-36g have increased morbidity and mortality following influ- continues through day 6 postinfection. Compared with their WT enza infection, and this increase is associated with increased viral counterparts, alveolar macrophages in IL-36g–KO mice were titers, higher levels of inflammatory cytokines in the airways, and skewed toward an M2-like phenotype by surface marker and gene by guest on September 25, 2021

FIGURE 3. Viral titers, cytokine levels, and pathologic condition of WT and IL-36g–KO mice following inﬂuenza infection. Age- and sex-matched

C57BL/6J (filled bars) and IL-36g–KO mice (open bars) were infected with (A) influenza PR8, (B) 30,000 EID50 x31, or (C) 300,000 EID50 x31, and lungs were analyzed for influenza NP copy number. (D) BAL fluid was analyzed for expression of proinflammatory cytokines on day 3 postinfection in C57BL/6J (filled bars) and IL-36g–KO mice (open bars). (E) Example images of formalin-fixed, paraffin-embedded, H&E-stained lungs from C57BL/6J and IL-36g– KO mice. Inflamed bronchioles (B) and small arterioles (SA) are noted on the images. Original magnification 3100. For (A)–(D), data are combined from two to three independent replicate experiments of at least five mice per group per replicate. Males and females were used in equal proportions. Asterisks indicate significant differences by Student t test on log-transformed data. Asterisks denote significance: *p , 0.05. 6 IL-36g IN INFLUENZA INFECTION Downloaded from http://www.jimmunol.org/

FIGURE 4. Analysis of innate immune subsets in the lung by ﬂow cytometry. C57BL/6J (ﬁlled bars) and IL-36g–KO mice (open bars) were infected with

300,000 EID50 x31 influenza, and the numbers of different immune cell subsets in the lung were assessed at (A) day 3 and (B) day 6 postinfection. (C) Representative flow plots of live CD45.2+ cells are shown with gates denoting alveolar macrophages. (D) The number of alveolar macrophages in C57BL/6J (solid line) and IL-36g–KO mice (dashed line) over time following influenza infection. Data are combined from two to three independent replicates of at least five mice per group per replicate. Males and females were used in equal proportions. Asterisks indicate significant differences in cell number by Student t test on log-transformed data. Asterisks denote significance: *p , 0.05, **p , 0.01.

expression prior to influenza infection and showed increased ap- mediators, but deletion of its receptor in mice leads to increased by guest on September 25, 2021 optosis within 24 h of infection. Transfer of WT alveolar macro- IL-8 production, increased neutrophil recruitment, and greater mor- phages into the lungs of IL-36g–KO mice increased survival tality upon lethal influenza challenge (17, 38). Another example of compared with IL-36g–KO mice that received IL-36g–KO alve- the importance of maintaining a balanced cytokine response for olar macrophages, and the survival of WT-transferred IL-36g–KO proper immune function comes from studies of IL-1. Deletion of mice was equivalent to mock-transferred WT mice. the IL-1R leads to decreased pathologic condition and neutrophil Early control of viral replication in the lung is key to preventing recruitment but also delays viral clearance and increases mortality dissemination of the virus throughout the lung. Several cell types (39). IL-1 is thus key to survival of influenza infection but also have been shown to have a role in the innate immune response to contributes to the development of acute respiratory distress syn- influenza, including neutrophils, NK cells, and alveolar macro- drome in humans (40). We find that lack of IL-36g has a similar phages. However, of these, only alveolar macrophages are situated effect to deletion of IL-1R on the outcome of influenza infection, within the respiratory tract prior to infection and can respond albeit by a different mechanism. In both cases, the mice without immediately without the need for recruitment from the vasculature the cytokine pathway of interest have an extended viremia and or bone marrow (7–9). Our data and previously published studies suffer greater morbidity and mortality. However, in IL-36g–KO indicate a role for alveolar macrophages in controlling viral rep- mice, alveolar macrophages are rapidly lost from the lung because lication during the first 3 d of infection, before neutrophils and NK of increased apoptosis. This removes a key blockade to early viral cells arrive to the lungs en masse. IL-36g–KO mice, which rapidly replication and induces greater proinflammatory cytokines (simi- lose alveolar macrophages during influenza infection, have in- lar to observations in IFN-aR–KO mice), leading to increased creased viral titers, presumably leading to increased inflammation pathologic conditions as the infection progresses. We find that in and immunopathologic condition. Indeed, we detected more IL-6 naive mice, Il36g mRNA is produced at much higher levels than and IFN-b in the airways of IL-36g–KO mice at day 3 postin- either Il36a or Il36b, and other reports have found that IL-36g fection. A more robust cytokine response early during infection protein is detectable in the BAL of naive mice (30, 41). This correlated with more diffuse immune infiltrate late in the infec- baseline production of IL-36g may act on alveolar macrophages to tion, presumably leading to an acute respiratory distress syn- maintain a more M1-like phenotype in the absence of infection. drome–like state in IL-36g–KO mice and increased mortality. Following influenza infection, neutrophils have the highest ex- Cytokines function as key factors for communication between pression of Il36g among immune cells, which differs from the immune cells during influenza infection. The outcome of influenza dogmatic production of the cytokine by epithelium, dendritic infection is not only dependent on the presence or absence cells, or CD4+ T cells (42). Il36g production by neutrophils has of particular cytokines but also their levels within the tissue. For only been observed in experimental autoimmune encephalitis, a example, production of IFN-a is correlated with symptom onset mouse model of multiple sclerosis, and it is notable that full length in humans and can boost the production of other inflammatory IL-36g is cleaved to its more active form by neutrophil-derived The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 5. Apoptosis and surface marker expression of WT and IL-36g–KO mice. The apoptosis of alveolar macrophages from C57BL/6J (ﬁlled bars) and IL-36g–KO mice (open bars) was measured using live/dead and annexin V staining in (A) naive mice or (B) 24 h postinfection. Representative staining is shown gated on alveolar macrophages. (C) Phenotype of alveolar macrophages in naive C57BL/6J and IL-36g–KO mice for markers associated with M1 and M2 macrophages subsets. (D) Expression of Retnla in alveolar macrophages isolated from naive WT (black bars) and IL-36g–KO (open bars) alveolar macrophages. (E and F) Geometric mean ﬂuorescence intensity of CD206 and CD200R on input IL-36g–KO alveolar macrophages (gray bars), IL-36g–KO alveolar macrophages after 10 d in a WT mouse (open bars), or the host alveolar macrophages (black bars). (Figure legend continues) 8 IL-36g IN INFLUENZA INFECTION

neutrophils contributes to the survival of alveolar macrophages during the early stages of inﬂuenza infection. IL-36 family cytokines have been shown to have a key role in pulmonary immune responses to bacterial and viral infection as well as the development of asthma. In mice, a single nucleotide polymorphism regulating expression of IL-36g determines the asthma sensitivity or resistance of two inbred mouse lines, and house dust mite administration causes rapid expression of IL-36g in experimental asthma models (28, 45). Respiratory infection of mice with either Gram-positive or Gram-negative bacteria induces production of IL-36g, and Ab blockade or genetic deletion of Il36g causes greater morbidity and mortality postinfection with S. pneumoniae (30). In this model, mice lacking IL-36g had greater bacterial load in the lung as well as systemic dissemination. In agreement with our observed increase in surface expression of M2-like markers in naive mice, that study also found that inter- stitial and alveolar macrophages from the lungs of IL-36g–KO mice had decreased expression of classical M1-like genes. Based

on our PCR and transfer experiments, IL-36g from a macrophage- Downloaded from extrinsic source is acting on the alveolar macrophages to maintain a more M1-like proﬁle, as transfer of IL-36g–KO alveolar macrophages to WT hosts reduces their expression of M2-like markers. Although the skewing of IL-36g–KO alveolar macrophages toward an M2-like cell was not complete, it would explain the

decreased survival and increased apoptosis of the cells following http://www.jimmunol.org/ influenza infection. Collectively, these data point toward a key role for IL-36g in macrophage polarization in the lung. Deletion of IL-36R led to increased survival and decreased pathologic condition in a lethal influenza infection with PR8 (31). Mice lacking IL-36R also had an extended viremia but decreased pathologic condition and proinflammatory cytokines compared with WT mice. Analysis of immune cell subsets in the lungs of IL-36R–KO mice showed decreased neutrophils and monocytes following infection, in contrast to our findings in IL- by guest on September 25, 2021 36g–KO mice. The differences in survival and cell recruitment between IL-36R–KO and IL-36g–KO mice following influenza FIGURE 6. Transfer of WT alveolar macrophages rescues mortality in infection may indicate differential roles for the various IL-36 5 IL-36g–KO mice. (A) IL-36g–KO mice were transferred with 5 3 10 family members in the immune response. Despite similar EC50 B purified WT alveolar macrophages intranasally, and ( ) lungs were har- values in vitro, IL-36g has weaker affinity than IL-36a (KD= vested the next day for identification of the transferred cells on the con- 1800 versus 480 nM, respectively) for IL-36R, possibly af- genic marker CD45.1. Representative staining gated on live alveolar fecting how the ligands signal through their receptor (46, 47). C D macrophages and percentage WT (transferred cells) are shown. ( and ) Higgins et al. (48) showed differential effects of IL-36a and Age- and sex-matched IL-36g–KO mice were transferred with 5 3 105 IL-36b in conditioning monocyte-derived dendritic cells for purified IL-36g–KO alveolar macrophages intranasally (dashed line), and a matched cohort of IL-36g–KO mice (dotted line) received 5 3 105 purified T cell stimulation. Aoyagi et al. (41) showed that IL-36R–KO WT alveolar macrophages intranasally. C57BL/6J (solid line) mice were and IL-36g–KO, but not IL-36a–KO, mice are protected during mock transferred with PBS. Data are combined from two independent lethal pulmonary infection with Pseudomonas aeruginosa. replicate experiments of at least four mice per group. All recipients were These data together with our findings suggest differential roles male mice to prevent rejection on Y Ags. Asterisk indicates a significant for IL-36 family members during infection. It is possible that difference in survival by the log-rank test. Asterisks denote significance: IL-36g acts to oppose signaling from another IL-36 family *p , 0.05. member or has an effect in the lungs of naive mice that is only seen with genetic deletion of the cytokine and not of the receptor. Future studies are necessary to dissect the roles of in- elastase and proteinase-3 (43, 44). Because neutrophils produce dividual IL-36 family members in regulating lung immune both the cytokine and activating protease, it is possible that neu- function during homeostasis and following infection. trophils release cleaved IL-36g, but confirming that hypothesis In conclusion, we have observed that the rapid apoptosis requires further work. Although our data suggest that baseline of alveolar macrophages in IL-36g–KO mice following influ- production of IL-36g is important for macrophage polarization enza infection results in increased viral load, increased proin- in naive mice, it is also possible that the IL-36g produced by flammatory cytokines, and enhanced immunopathologic condition,

Data are combined from two to three independent replicates of at least five mice per group per replicate. Males and females were used in equal proportions. Asterisks indicate significant differences in cell number by ANOVA followed by Tukey’s multiple comparisons test (A and B), Student t test (C and D) (WT versus KO), or the one-sample t test (D) (input versus KO). Asterisks denote significance: *p , 0.05, **p , 0.01, ****p , 0.0001. The Journal of Immunology 9 leading to decreased survival. Importantly, this defect can be 21. Hussell, T., A. Pennycook, and P. J. Openshaw. 2001. Inhibition of tumor ne- crosis factor reduces the severity of virus-specific lung immunopathology. Eur. J. corrected by the transfer of WT alveolar macrophages into Immunol. 31: 2566–2573. the lungs of IL-36g–KO mice. Overall, we demonstrate a 22. Salomon, R., E. Hoffmann, and R. G. Webster. 2007. Inhibition of the cytokine critical role for IL-36g in the immune response to influenza response does not protect against lethal H5N1 influenza infection. Proc. Natl. Acad. Sci. USA 104: 12479–12481. infection. 23. Dienz, O., J. G. Rud, S. M. Eaton, P. A. Lanthier, E. Burg, A. Drew, J. Bunn, B. T. Suratt, L. Haynes, and M. Rincon. 2012. Essential role of IL-6 in protection Acknowledgments against H1N1 influenza virus by promoting neutrophil survival in the lung. Mucosal Immunol. 5: 258–266. We thank Dr. Tamas Nagy for pathological examination of mouse lungs. 24. Teijaro, J. R. 2014. The role of cytokine responses during influenza virus This research utilized shared resources at Emory University, including pathogenesis and potential therapeutic options. In. Current Topics in Microbi- the Emory+Children’s Flow Cytometry Core and the Winship Research ology and Immunology. Springer, Berlin, Heidelberg, p. 1–20. Pathology Core Lab. We thank Sarah Hayward and Emily Cartwright for 25. Brun-Buisson, C., J.-C. M. Richard, A. Mercat, A. C. M. Thie´baut, and L. Brochard, REVA-SRLF A/H1N1v 2009 Registry Group. 2011. Early corti- helpful discussion and Laurel Lawrence for invaluable help in managing costeroids in severe influenza A/H1N1 pneumonia and acute respiratory distress the animal colony. syndrome. Am. J. Respir. Crit. Care Med. 183: 1200–1206. 26. Tisoncik, J. R., M. J. Korth, C. P. Simmons, J. Farrar, T. R. Martin, and M. G. Katze. 2012. Into the eye of the cytokine storm. Microbiol. Mol. Biol. Rev. Disclosures 76: 16–32. The authors have no financial conflicts of interest. 27. Medina-Contreras, O., A. Harusato, H. Nishio, K. L. Flannigan, V. Ngo, G. Leoni, P.-A. Neumann, D. Geem, L. N. Lili, R. A. Ramadas, et al. 2016. Cutting edge: IL-36 receptor promotes resolution of intestinal damage. J. Immunol. 196: 34–38. References 28. Ramadas, R. A., S. L. Ewart, B. D. Medoff, and A. M. LeVine. 2011. 1. Thompson, W. W., D. K. Shay, E. Weintraub, L. Brammer, N. Cox, Interleukin-1 family member 9 stimulates chemokine production and neutrophil Downloaded from L. J. Anderson, and K. Fukuda. 2003. Mortality associated with influenza and influx in mouse lungs. Am. J. Respir. Cell Mol. Biol. 44: 134–145. respiratory syncytial virus in the United States. JAMA 289: 179–186. 29. Tortola, L., E. Rosenwald, B. Abel, H. Blumberg, M. Scha¨fer, A. J. Coyle, J.- 2. Dolin, R. 2012. Influenza. In Harrison’s Principles of Internal Medicine,18th C. Renauld, S. Werner, J. Kisielow, and M. Kopf. 2012. Psoriasiform dermatitis Ed., D. L. Longo, A. S. Fauci, D. L. Kasper, S. L. Hauser, J. Jameson, and is driven by IL-36-mediated DC-keratinocyte crosstalk. J. Clin. Invest. 122: J. Loscalzo, eds. McGraw-Hill, New York. 3965–3976. 3. Loo, Y.-M., J. Fornek, N. Crochet, G. Bajwa, O. Perwitasari, L. Martinez- 30. Kovach, M. A., B. Singer, G. Martinez-Colon, M. W. Newstead, X. Zeng, Sobrido, S. Akira, M. A. Gill, A. Garcıá-Sastre, M. G. Katze, and M. Gale, Jr. P. Mancuso, T. A. Moore, S. L. Kunkel, M. Peters-Golden, B. B. Moore, and

2008. Distinct RIG-I and MDA5 signaling by RNA viruses in innate immunity. J. T. J. Standiford. 2017. IL-36g is a crucial proximal component of protective http://www.jimmunol.org/ Virol. 82: 335–345. type-1-mediated lung mucosal immunity in Gram-positive and -negative bacte- 4. Ichinohe, T. 2010. Respective roles of TLR, RIG-I and NLRP3 in influenza virus rial pneumonia. Mucosal Immunol. 10: 1320–1334. infection and immunity: impact on vaccine design. Expert Rev. Vaccines 9: 31.Aoyagi,T.,M.W.Newstead,X.Zeng, S. L. Kunkel, M. Kaku, and 1315–1324. T. J. Standiford. 2017. IL-36 receptor deletion attenuates lung injury and 5. Iwasaki, A., and P. S. Pillai. 2014. Innate immunity to influenza virus infection. decreases mortality in murine influenza pneumonia. Mucosal Immunol. 10: Nat. Rev. Immunol. 14: 315–328. 1043–1055. 6. Pulendran, B., and M. S. Maddur. 2014. Innate immune sensing and response to 32. Harusato, A., H. Abo, V. L. Ngo, S. W. Yi, K. Mitsutake, S. Osuka, J. E. Kohlmeier, influenza. In Current Topics in Microbiology and Immunology. Springer, Berlin, J.D.Li,A.T.Gewirtz,A.Nusrat,andT.L.Denning.2017.IL-36g signaling Heidelberg, p. 23–71. controls the induced regulatory T cell-Th9 cell balance via NFkB activation and 7. Lim, K., Y.-M. Hyun, K. Lambert-Emo, T. Capece, S. Bae, R. Miller, STAT transcription factors. Mucosal Immunol. 10: 1455–1467. D. J. Topham, and M. Kim. 2015. Neutrophil trails guide influenza-specific 33. Bochkov, Y. A., K. M. Hanson, S. Keles, R. A. Brockman-Schneider, +

CD8 T cells in the airways. Science 349: aaa4352. N. N. Jarjour, and J. E. Gern. 2010. Rhinovirus-induced modulation of gene by guest on September 25, 2021 8. Schultz-Cherry, S. 2015. Role of NK cells in influenza infection. Curr. Top. expression in bronchial epithelial cells from subjects with asthma. Mucosal Microbiol. Immunol. 386: 109–120. Immunol. 3: 69–80. 9. Weiss, I. D., O. Wald, H. Wald, K. Beider, M. Abraham, E. Galun, A. Nagler, 34. Pachulec, E., R. B. Abdelwahed Bagga, L. Chevallier, H. O’Donnell, C. Guillas, and A. Peled. 2010. IFN-gamma treatment at early stages of influenza virus J. Jaubert, X. Montagutelli, E. Carniel, and C. E. Demeure. 2017. Enhanced infection protects mice from death in a NK cell-dependent manner. J. Interferon macrophage M1 polarization and resistance to apoptosis enable resistance to Cytokine Res. 30: 439–449. plague. J. Infect. Dis. 216: 761–770. 10. Lin, K. L., Y. Suzuki, H. Nakano, E. Ramsburg, and M. D. Gunn. 2008. CCR2 35. Roszer,} T. 2015. Understanding the mysterious M2 macrophage through acti- + monocyte-derived dendritic cells and exudate macrophages produce vation markers and effector mechanisms. Mediators Inflamm. 2015: 816460. influenza-induced pulmonary immune pathology and mortality. J. Immunol. 36. Wu, K., D. E. Byers, X. Jin, E. Agapov, J. Alexander-Brett, A. C. Patel, 180: 2562–2572. M. Cella, S. Gilfilan, M. Colonna, D. L. Kober, et al. 2015. TREM-2 promotes 11. Camp, J. V., and C. B. Jonsson. 2017. A role for neutrophils in viral respiratory macrophage survival and lung disease after respiratory viral infection. J. Exp. disease. Front. Immunol. 8: 550. Med. 212: 681–697. 12. Cardani, A., A. Boulton, T. S. Kim, and T. J. Braciale. 2017. Alveolar macro- 37. Tyner, J. W., O. Uchida, N. Kajiwara, E. Y. Kim, A. C. Patel, M. P. O’Sullivan, phages prevent lethal influenza pneumonia by inhibiting infection of type-1 al- M. J. Walter, R. A. Schwendener, D. N. Cook, T. M. Danoff, and M. J. Holtzman. veolar epithelial cells. PLoS Pathog. 13: e1006140. 2005. CCL5-CCR5 interaction provides antiapoptotic signals for macrophage 13. Schneider, C., S. P. Nobs, A. K. Heer, M. Kurrer, G. Klinke, N. van Rooijen, survival during viral infection. Nat. Med. 11: 1180–1187. J. Vogel, and M. Kopf. 2014. Alveolar macrophages are essential for protection 38. Hayden, F. G., R. Fritz, M. C. Lobo, W. Alvord, W. Strober, and S. E. Straus. from respiratory failure and associated morbidity following influenza virus in- 1998. Local and systemic cytokine responses during experimental human infection. PLoS Pathog. 10: e1004053. fluenza A virus infection. Relation to symptom formation and host defense. J. 14. Taubenberger, J. K., and D. M. Morens. 2008. The pathology of influenza virus Clin. Invest. 101: 643–649. infections. Annu. Rev. Pathol. 3: 499–522. 39. Schmitz, N., M. Kurrer, M. F. Bachmann, and M. Kopf. 2005. Interleukin-1 is 15. Beigel, J. H., J. Farrar, A. M. Han, F. G. Hayden, R. Hyer, M. D. de Jong, responsible for acute lung immunopathology but increases survival of respiratory S. Lochindarat, T. K. T. Nguyen, T. H. Nguyen, T. H. Tran, et al; Writing Com- influenza virus infection. J. Virol. 79: 6441–6448. mittee of the World Health Organization (WHO) Consultation on Human Influenza 40. Pugin, J., B. Ricou, K. P. Steinberg, P. M. Suter, and T. R. Martin. 1996. A/H5. 2005. Avian influenza A (H5N1) infection in humans. [Published erratum Proinflammatory activity in bronchoalveolar lavage fluids from patients with appears in 2006 N. Engl. J. Med. 354: 884.] N. Engl. J. Med. 353: 1374–1385. ARDS, a prominent role for interleukin-1. Am. J. Respir. Crit. Care Med. 153: 16. Guo, X. J., and P. G. Thomas. 2017. New fronts emerge in the influenza cytokine 1850–1856. storm. Semin. Immunopathol. 39: 541–550. 41. Aoyagi, T., M. W. Newstead, X. Zeng, Y. Nanjo, M. Peters-Golden, M. Kaku, 17. Seo, S.-U., H.-J. Kwon, H.-J. Ko, Y.-H. Byun, B. L. Seong, S. Uematsu, S. Akira, and T. J. Standiford. 2017. Interleukin-36g and IL-36 receptor signaling me- and M.-N. Kweon. 2011. Type I interferon signaling regulates Ly6C(hi) diate impaired host immunity and lung injury in cytotoxic Pseudomonas monocytes and neutrophils during acute viral pneumonia in mice. PLoS Pathog. aeruginosa pulmonary infection: role of prostaglandin E2. PLoS Pathog. 13: 7: e1001304. e1006737. 18. Garcıá-Sastre, A., A. Egorov, D. Matassov, S. Brandt, D. E. Levy, J. E. Durbin, 42. Gabay, C., and J. E. Towne. 2015. Regulation and function of interleukin-36 P. Palese, and T. Muster. 1998. Influenza A virus lacking the NS1 gene replicates cytokines in homeostasis and pathological conditions. J. Leukoc. Biol. 97: in interferon-deficient systems. Virology 252: 324–330. 645–652. 19. Bot, A., S. Bot, and C. A. Bona. 1998. Protective role of gamma interferon 43. Bozoyan, L., A. Dumas, A. Patenaude, and L. Vallie`res. 2015. Interleukin-36g is during the recall response to influenza virus. J. Virol. 72: 6637–6645. expressed by neutrophils and can activate microglia, but has no role in experi- 20. Perrone, L. A., K. J. Szretter, J. M. Katz, J. P. Mizgerd, and T. M. Tumpey. 2010. mental autoimmune encephalomyelitis. J. Neuroinflammation 12: 173. Mice lacking both TNF and IL-1 receptors exhibit reduced lung inflammation 44. Henry, C. M., G. P. Sullivan, D. M. Clancy, I. S. Afonina, D. Kulms, and and delay in onset of death following infection with a highly virulent H5N1 S. J. Martin. 2016. Neutrophil-derived proteases escalate inflammation through virus. J. Infect. Dis. 202: 1161–1170. activation of IL-36 family cytokines. Cell Reports 14: 708–722. 10 IL-36g IN INFLUENZA INFECTION

45. Ramadas, R. A., X. Li, D. M. Shubitowski, S. Samineni, M. Wills-Karp, and 47. Zhou, L., V. Todorovic, S. Kakavas, B. Sielaff, L. Medina, L. Wang, R. Sadhukhan, S. L. Ewart. 2006. IL-1 Receptor antagonist as a positional candidate gene in a H. Stockmann, P. L. Richardson, E. DiGiammarino, C. Sun, and V. Scott. 2017. murine model of allergic asthma. Immunogenetics 58: 851–855. Quantitative ligand and receptor binding studies reveal the mechanism of 46. Towne, J. E., B. R. Renshaw, J. Douangpanya, B. P. Lipsky, M. Shen, interleukin-36 (IL-36) pathway activation. J. Biol. Chem. 293: 403–411. C. A. Gabel, and J. E. Sims. 2011. Interleukin-36 (IL-36) ligands require pro- 48. Higgins, J., S. Mutamba, Y. Mahida, P. Barrow, and N. Foster. 2015. IL-36a in- cessing for full agonist (IL-36a, IL-36b, and IL-36g) or antagonist (IL-36Ra) duces maturation of Th1-inducing human MDDC and synergises with IFN-g to activity. J. Biol. Chem. 286: 42594–42602. induce high surface expression of CD14 and CD11c. Hum. Immunol. 76: 245–253. Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021 Supplemental Table 11

C57BL /6J 304/304 0/304 300/304 300/304 152/304 152/304

Panel Info 129S1 /SvImJ 0/304 304/304 4/304 4/304 152/304 152/304 C57BL

/6J 100.00% 0.00% 98.68% 98.68% 50.00% 50.00% 129S1

/SvImJ 0.00% 100.00% 1.32% 1.32% 50.00% 50.00%

B6J Het1 Het2 129S Sample 455187 1 455187 2

SNP_ID Expected SNPs 129S Chromosome- C57BL 1/SvI bp RS_ID /6J mJ A10 A12 B07 B08 B10 B12

02- 020551513-M rs4139548 A G A G G G G/A G/A 02- 041042651-M rs3697051 C G C G G G G/C G/C

1 SNP analysis of Il36g-/- mice. Two Il36g-/- mice from different litters were analyzed for B6 vs 129 DNA content by a 150 marker Jackson Labs analysis. Columns represent samples tested. 455187 1 and 455187 2 are Il36g-/- mice from our colony, other columns are no template, B6, 129, and B6/129 heterozygous controls. Summary statistics are listed in the rows at the top of the table. The two 129 SNPs observed in our Il36g-/- mice are listed in rows and identified by chromosome location and RS ID. B6 loci are shown in green, 129 in tan, and heterozygous in yellow. Il36g is located at chr2-24186746 thru chr2-24193567. Supplemental Figure 1

Supplemental Figure 1: Expression of Il36 family members in naïve mice as reported in

Figure 1 A-C. Expression of the three Il36 genes in naïve mice was compared to each other using ANOVA followed by Holm-Sidak multiple comparisons test.

Supplemental Figure 2

Supplemental Figure 2: Gating strategy for innate immune cell subsets in the lung.

Example flow cytometry staining for the analysis of lung immune cell subsets. Gating strategy is noted by black arrows and cell subsets analyzed are identified in red text adjacent to gates.