Type I IFN−Inducible Downregulation of MicroRNA-27a Feedback Inhibits Antiviral Innate Response by Upregulating Siglec1/TRIM27 This information is current as of September 26, 2021. Qingliang Zheng, Jin Hou, Ye Zhou, Yingyun Yang and Xuetao Cao J Immunol 2016; 196:1317-1326; Prepublished online 23 December 2015; doi: 10.4049/jimmunol.1502134 Downloaded from http://www.jimmunol.org/content/196/3/1317

Supplementary http://www.jimmunol.org/content/suppl/2015/12/23/jimmunol.150213 http://www.jimmunol.org/ Material 4.DCSupplemental References This article cites 46 articles, 14 of which you can access for free at: http://www.jimmunol.org/content/196/3/1317.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 26, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of 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 © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Type I IFN–Inducible Downregulation of MicroRNA-27a Feedback Inhibits Antiviral Innate Response by Upregulating Siglec1/TRIM27

Qingliang Zheng,*,1 Jin Hou,†,1 Ye Zhou,† Yingyun Yang,* and Xuetao Cao*,†

Upon recognition of viral components by pattern recognition receptors, including TLRs and retinoic acid–inducible gene I–like helicases, cells are activated to produce type I IFN, which plays key roles in host antiviral innate immune response. However, excessive IFN production may induce immune disorders, and the mechanisms responsible for the regulation of type I IFN production have attracted much attention. Furthermore, type I IFN activates the downstream IFN/JAK/STAT pathway to modulate expression of a set of genes against viral infection, but whether these genes can feedback regulate type I IFN production is poorly understood. In this study, by screening the microRNAs modulated by viral infection in , we identified that Downloaded from microRNA (miR)-27a was significantly downregulated via the IFN/JAK/STAT1/runt-related transcription factor 1 pathway. Inducible downregulation of miR-27a, in turn, negatively regulated vesicular stomatitis virus–triggered type I IFN production, thus promoting vesicular stomatitis virus replication in macrophages. Mechanistically, we found that miR-27a directly targeted –binding Ig-like (Siglec)1 and E3 ubiquitin ligase tripartite motif–containing 27 (TRIM27), both of which were previously verified as negative regulators of type I IFN production. Furthermore, we constructed “Sponge” transgenic mice against miR-27a expression and found that Siglec1 and TRIM27 expression were elevated whereas type I IFN production was http://www.jimmunol.org/ inhibited and viral replication was aggregated in vivo. Therefore, type I IFN–induced downregulation of miR-27a can upregulate Siglec1 and TRIM27 expression, feedback inhibiting type I IFN production in antiviral innate response. Our study outlines a new negative way to feedback regulate type I IFN production. The Journal of Immunology, 2016, 196: 1317–1326.

ype I IFN plays key roles in the defense against viral type I IFN production in the innate response against RNA virus in- infection. Upon viral infection, pattern recognition re- fection by promoting retinoic acid–inducible gene I (RIG-I) protea- T ceptors (PRRs) will trigger TANK-binding kinase 1 somal degradation (10). We previously found that Siglec1 associated (TBK1) activation and then activate the transcription factor IFN with DNAX activation protein-12 to activate the scaffolding function regulatory factor (IRF)3 to induce type I IFN production (1, 2). of Src homology 2 domain–containing protein tyrosine phosphatase by guest on September 26, 2021 Type I IFN binds to IFN-a/b receptor (IFNAR) and initiates a 2, and then recruited E3 ligase tripartite motif–containing protein 27 signaling cascade, which has been shown to regulate .2000 IFN- (TRIM27) to degrade TBK1, thus negatively regulating type I IFN regulated genes (IRGs), including coding and noncoding RNA production (11). Hence, type I IFN production during viral infection transcripts (3), and these IRGs are important to eliminate the in- should be tightly controlled to initiate an appropriate immune re- vading pathogens (4, 5). Some IRGs encode PRRs for detecting sponse that eliminates invading pathogens but to avoid excessive viral molecules, and other IRGs encode with potential for production of type I IFN–mediated immunopathological conditions direct antiviral activity, thus constituting a positive loop for in- or immune disorders (12, 13). However, the underlying mechanisms creased IFN production and response (6). However, various types for stringent control of type I IFN production in antiviral innate re- of PRRs also tightly cross-regulate type I IFN production to en- sponse still need to be fully elucidated, and it is still unclear whether sure that the appropriate amount is produced (7–9). For example, other regulators induced by IFN effector signaling pathway play roles sialic acid–binding Ig-like lectin (Siglec)-G negatively regulates in the negative regulation of type I IFN production.

*National Key Laboratory of Medical Molecular Biology, Department of Immunol- Address correspondence and reprint requests to Dr. Xuetao Cao or Dr. Jin Hou, ogy, Institute of Basic Medical Sciences, Peking Union Medical College, Chinese National Key Laboratory of Medical Molecular Biology and Department of Immu- Academy of Medical Sciences, Beijing 100005, China; and †National Key Labora- nology, Institute of Basic Medical Sciences, Peking Union Medical College, Chinese tory of Medical Immunology and Institute of Immunology, Second Military Medical Academy of Medical Sciences, Beijing 100005, China (X.C.) or National Key Lab- University, Shanghai 200433, China oratory of Medical Immunology and Institute of Immunology, Second Military Med- ical University, Shanghai 200433, China (J.H.). E-mail addresses: caoxt@immunol. 1Q.Z. and J.H. contributed equally to this work. org (X.C.) or [email protected] (J.H.) ORCID: 0000-0002-8780-0099 (Y.Z.). The online version of this article contains supplemental material. Received for publication September 30, 2015. Accepted for publication November Abbreviations used in this article: CDS, coding sequence; IAV, influenza A PR/8/34 20, 2015. virus (H1N1 subtype); IFNAR, IFN-a/b receptor; IRF, IFN regulatory factor; IRG, This work was supported by Excellent Youth Researcher Award of National Natural IFN-regulated gene; miR, microRNA; miRNA, microRNA; NC, negative control; Science Foundation of China Grant 81422037, National Key Basic Research Program poly(I:C), polyinosinic-polycytidylic acid; pri-, primary; PRR, pattern recognition of China Grants 2012CB518900 and 2013CB530502, National Natural Science receptor; RIG-I, retinoic acid–inducible gene I; RUNX1, runt-related transcription Foundation of China Grants 81123006, 31300718, 31370864, and 31170826, Shang- factor 1; SeV, Sendai virus; Siglec, sialic acid–binding Ig-like lectin; siRNA, small hai Municipal Education Commission Chenguang Research Program Grant 12CG39, interfering RNA; TBK1, TANK-binding kinase 1; TCID50, 50% tissue culture–infec- and by an Excellent Youth Researcher Award from the Second Military Medical tive dose; TRIM27, tripartite motif–containing protein 27; UTR, untranslated region; University. VSV, vesicular stomatitis virus.

The microarray data presented in this article have been submitted to the National Ó Center for Biotechnology Information’s Gene Expression Omnibus database (http:// Copyright 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 www.ncbi.nlm.nih.gov/geo/) under accession number GSE43910. www.jimmunol.org/cgi/doi/10.4049/jimmunol.1502134 1318 DOWNREGULATION OF miR-27a INHIBITS IFN PRODUCTION

The microRNAs (miRNAs) play important roles in the regula- Reagents and pathogens tion of cellular processes, including cell proliferation and differ- The TLR ligands LPS, CpG oligodeoxynucleotide, and polyinosinic- entiation, apoptosis, cancer, and viral infections (14–17). miRNAs polycytidylic acid [poly(I:C)] were described previously (29). The STAT1 are an abundant class of highly conserved, small (18–25 nt), and inhibitor (fludarabine, 100 mg/ml), GSK inhibitor (SB216763, 10 mm/ml), noncoding RNAs that function as important posttranscriptional ERK inhibitor (PD98059, 10 mm/ml), and NF-kB inhibitor (pyrrolidine regulators to suppress gene expression by binding to the 39-un- dithiocarbamate, 200 mm/ml) were purchased from Calbiochem (San Diego, CA). rIFN-a1b was obtained from Sanyuan (Shanghai, China). Abs translated region (UTR), 59-UTR, or the protein coding sequence specific to b-actin (sc-8432) and HRP-coupled secondary Abs (sc-2749) (CDS) of target mRNAs (18, 19). Many IRGs are miRNAs, and were from Santa Cruz Biotechnology (Santa Cruz, CA). Ab specific to their important roles in regulating immune responses have Siglec1 (MAB5610) was from R&D Systems (Minneapolis, MN). Abs attracted much attention (20, 21). Several miRNAs target tran- specific to STAT1 (9172), p-IRF3 (4947), and TBK1 (3013) were from Technology (Danvers, MA). Abs specific to TRIM27 scripts encoding components at many steps of the type I IFN re- (AV34701) and RUNX1 (AV38073) were from Sigma-Aldrich (St. Louis, sponse, which introduced an additional layer of regulation for MO). VSV was propagated and amplified by infecting a monolayer of Vero antiviral pathway, and these targeting transcripts include compo- cells. The supernatant was harvested 24 h later and clarified by centrifu- nents of PRR pathways impacting IFN production, the IFN cell gation. Viral titers were determined by evaluating 50% tissue culture– surface receptors, signal transduction proteins to regulate IFN infective dose (TCID50) levels in Vero cells. Sendai virus (SeV) and HSV-1 virus (Kos strain) were obtained as described (10). Influenza A signaling, and some IRGs (22). In the activation of the RIG-I PR/8/34 virus (H1N1 subtype) (IAV) was provided by Dr. Hangping Yao antiviral pathway, a set of miRNA expressions are induced and from Zhejiang University School of Medicine (Hangzhou, China). they employ feedback regulatory mechanisms to modulate type I

IFN production and function (23, 24). Vesicular stomatitis virus Microarray assay Downloaded from (VSV) infection induces microRNA (miR)-146a expression, The cDNA microarray assay (Agilent whole mouse genome microarray 4 3 which subsequently suppresses type I IFN production by targeting 44K chips) was conducted by a service provider (Shanghai Biotechnology, IL-1R–associated kinases 1 and 2 and TNFR-associated factor 6 Shanghai, China). The raw data from the microarray were submitted to the Gene Expression Omnibus database with the accession number GSE43910 (25). Furthermore, miR-29a inhibits the antiviral response by (http://www.ncbi.nlm.nih.gov/geo/). targeting the IFN-a receptor to repress its downstream signaling

(26). However, there are limited direct studies of IFN-regulated miRNA mimics and inhibitors http://www.jimmunol.org/ miRNAs targeting IRGs, and whether other miRNAs are involved miR-27a mimics and negative control (NC) mimics as well as miR-27a in regulating the host antiviral innate response and their corre- inhibitors and NC inhibitors were designed and synthesized by Gene- sponding mechanisms still needs to be determined. Pharma (Shanghai, China). Peritoneal macrophages or RAW264.7 cells In the present study, we found that viral infection significantly were transfected with these RNAs (final concentration, 20 nM) to over- express or inhibit miR-27a activity using INTERFERin reagent (Polyplus downregulated miR-27a expression in mouse macrophages through Transfection). an IFN/JAK/STAT1/runt-related transcription factor 1 (RUNX1) pathway–dependent manner. This type I IFN–induced downregu- RNA interference lation of miR-27a feedback facilitated viral replication by sup- Thioglycollate-elicited mouse peritoneal macrophages or RAW264.7 cells pression of type I IFN production via upregulating negative were transfected with siRNA (final concentration, 20 nM) using INTER- by guest on September 26, 2021 regulators of TBK1. Hence, our findings reveal a new negative FERin reagent (Polyplus Transfection). The STAT1-specific siRNAs were regulatory mechanism for type I IFN production in the antiviral 59-GGA AAA GCA AGC GUA AUC UTT-39 (sense) and 59-AGA UUA CGC UUG CUU UUC CTT-39 (antisense). The RUNX1-specific siRNAs innate response. were 59-CUG UGA AUG CUU CUG AUU UTT-39 (sense) and 59-AAA UCA GAA GCA UUC ACA GTT-39 (antisense). The Siglec1-specific Materials and Methods siRNAs were 59-GGU GUG CAG UGU ACA AAG UTT-39 (sense) and 59-ACU UUG UAC ACU GCA CAC CTT-39 (antisense). The TRIM27- Mice specific siRNAs were 59-GAG UGA AAG ACU UGA AGA ATT-39 C57BL/6 mice (6–8 wk old) were obtained from the Joint Ventures Sipper (sense) and 59-UUC UUC AAG UCU UUC ACU CTT-39 (antisense). The BK Experimental Animal Company (Shanghai, China). IFNAR-deficient NC siRNAs were 59-UUC UCC GAA CGU GUC ACG UTT-39 (sense) and mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Mice 59-ACG UGA CAC GUU CGG AGA ATT-39 (antisense). All siRNAs were were housed and bred in specific pathogen-free conditions. All animal obtained from GenePharma. experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, with the ap- Real-time quantitative PCR proval of the Scientific Investigation Board of the Second Military Medical Total RNA was extracted with TRIzol reagent (Invitrogen) and reverse University (Shanghai, China). transcribed using the reverse transcription system from Toyobo (Osaka, Cell culture and transfection Japan). The reverse transcription products from different samples were amplified by real-time PCR and analyzed as described previously (30, 31). HEK293T, THP1 and RAW264.7 cell lines were obtained from American The primer sequences for quantitative PCR analysis were: miR-27a RT, 59- Type Culture Collection (Manassas, VA). Mouse PBMCs, myeloid con- GTC GTA TCC AGT GCA GGG TCC GAG GTA TTC GCA CTG GAT ventional dendritic cells, plasmacytoid dendritic cells, bone marrow– ACG ACG CGG AAC-39, forward, 59-GGA GCT TCA CAG TGG CTA derived macrophages, and human -derived macrophages were A-39,reverse,59-GTG CAG GGT CCG AGG T-39;U6forward,59-CTC GCT obtained and cultured as described previously (27, 28). Cells (1 3 106) TCG GCA GCA CA-39, reverse, 59-AAC GCT TCA CGA ATT TGC GT- were seeded into six-well plates and incubated overnight. jetPRIME 39; RUNX1 forward, 59-ACG ATG AAA ACT ACT CGG CAG-39, reverse, transfection reagents (Polyplus Transfection, Illkirch, France) were used 59-CTG AGG TCG TTG AAT CTC GCT-39; STAT1 forward, 59-GCT for cotransfection of plasmids and RNAs according to the manufacturer’s GCC TAT GAT GTC TCG TTT-39, reverse, 59-TGC TTT TCC GTA TGT instructions. Thioglycollate-elicited mouse peritoneal macrophages were TGT GCT-39; primary (pri)–miR-27a forward, 59-CTA TCA TGA CAA seeded (2 3 105 cells in 0.5 ml) into 24-well plates and incubated over- CTG GCC TGA G-39, reverse, 59-GAC TTT GCT GTG GAC CTT GC-39; night. Cells were then transfected with small interfering RNAs (siRNAs) VSV forward, 59-ACG GCG TAC TTC CAG ATG G-39, reverse, 59-CTC using INTERFERin (Polyplus Transfection) according to the manufac- GGT TCA AGA TCC AGG T-39; IFN-4a forward, 59-ACT CAG CAG turer’s instructions. To establish stably transfected RAW264.7 cells, G418 ACC TTG AAC CT-39, reverse, 59-CAG TCT TGG CAG CAA GTT GAC- was added (1000 mg/ml) 48 h after transfection and maintained at 800 39; IFN-b forward, 59-CAG CTC CAA GAA AGG ACG AAC-39, reverse, mg/ml for 3 wk for positive selection. For stably transfected cells, RUNX1 59-GGC AGT GTA ACT CTT CTG CAT-39; TRIM27 forward, 59-AGA expression was confirmed by Western blot. Stably transfected RAW264.7 ACC GAC TGG ACC ACC TAA-39, reverse, 59-TGC TCC TTC AAC cells were subsequently cultured in complete medium containing 500 GAG TGA TAG A-39; Siglec1 forward, 59-GCT GGT GGA CAA GCG mg/ml G418. TTT C-39, reverse, 59-TTC AAG TCT TTG AGC AAC AGG T-39; GS27 The Journal of Immunology 1319 forward, 59-CAC TAC CAG CAG AAC ACC CC-39, reverse, 59-AAG Immunoblot and immunoprecipitation TTC CGC TTA TGG ATC CG-39. Cells were lysed using cell lysis buffer (Cell Signaling Technology) Molecular cloning of related genes supplemented with mixture protease inhibitor (Calbiochem). Protein concentrations of the extracts were measured using a BCA assay Related genes were obtained from mouse by RT-PCR and (Pierce, Rockford, IL) and equalized with the extraction reagent. subsequently subcloned into pMIR or pcDNA vectors. Corresponding Equivalent amounts of extract were loaded and subjected to SDS- primers used were as follows: pMIR-Siglec1 CDS forward, 59-GGA CTA PAGE, transferred onto nitrocellulose membranes, and then blotted GTG TTT CAG GGG TCG AGC T-39, reverse, 59-CCC AAG CTT TTG as described previously (32). CCT CAG ACT TAT G-39; pMIR-Siglec1 39-UTR forward, 59-GGA CTA GTC TGC CTC AGC CTC TGC C-39, reverse, 59-CCC AAG CTT AGA Generation of GS27 mouse GAA AGG CAA AGG G-39; pMIR-TRIM27 CDS forward, 59-GGA CTA GTT TGC CCA AAA ATG TCT G-39, reverse, 59-CCC AAG CTT TCA Transgenic mice expressing the miR-27a Sponge target construct by mi- ATC CCA CTC ATA G-39; Siglec1 forward, 59-ATA AGA ATG CGG croinjection technology as described previously (33). Then, founder mice CCG CCA ATG TGT GTC CTG TTC TCC CTG CT-39, reverse, 59-CAC were identified by PCR assay using GS27 primers of genomic DNA ob- GGG AAG CTT TCA GAG AGC AGC AAC CAC TTC CT-39; TRIM27 tained from the tails of transgenic mice. Founder mice were hybridized forward, 59-CCG CTC GAG ATG GCC TCC GGG AGC GTG GC-39, with wild-type C57BL/6 mice to produce the mice identified and used for reverse, 59-GCC GGT ACC TCA CGG AGA GGT CTC CAT GGA AT-39; experiments. RUNX1 forward, 59-CCG GAA TTC GGG CTT CAG ACA GCA TTT TTG AG-39, reverse, 59-GCC GGT ACC TCA GTA GGG CCG CCA CAC Lung histology GG-39. Each construct was confirmed by sequencing. Lungs from control or virus-infected mice were dissected, fixed in 10% Luciferase reporter assays phosphate-buffered formalin, embedded into paraffin, sectioned, stained

with H&E solution, and examined by light microscopy for histological Downloaded from HEK293T cells (1 3 104) were plated in 96-well plates and transfected changes. with a mixture of the indicated luciferase reporter plasmid and the pRL- TK-Renilla luciferase plasmid together with various amounts of the fol- ELISA lowing: miR-27a mimics, miR-27a inhibitors, or the Sponge plasmid. An IFN-b in the supernatants was measured with an ELISA kit (PBL Bio- empty pcDNA3.1 vector was used to maintain equal amounts of DNA medical Laboratories, Piscataway, NJ). among wells. Cells were collected at 24 h after transfection, and luciferase activity was measured with a Dual-Luciferase assay (Promega, Madison,

Statistical analysis http://www.jimmunol.org/ WI) with a Luminoskan Ascent luminometer (Thermo Scientific, Hanover Park, IL) as described previously (30). Reporter gene activity was deter- All statistical analyses were performed by a Student t test using Prism mined by normalizing firefly luciferase activity to Renilla luciferase (version 5.0; GraphPad Software). A p value ,0.05 was considered sta- activity. tistically significant. by guest on September 26, 2021

FIGURE 1. Viral infection downregulates miR-27a expression. (A and B) Quantitative PCR analysis of miR-27a in primary peritoneal macrophages infected with VSV (A) or SeV (B) for 24 h, or infected with VSV (multiplicity of infection [MOI] of 10) (A) or SeV (MOI of 100) (B) for the indicated time. (C–G) Quantitative PCR analysis of miR-27a in peritoneal macrophages infected with HSV (C) (MOI of 10) or treated with LPS (D), CpG (E), transfection of poly(I:C) (F), and in mouse bone marrow–derived macrophages or human THP1 cells infected with IAV (MOI of 5) (G) for the indicated time. (H and I) Quantitative PCR analysis of miR-27a in mouse PBMCs (PBMNC), myeloid conventional dendritic cells (cDC), and plasmacytoid dendritic cells (pDC) (H), and in human monocyte-derived macrophages (Mo-Mf) and THP1 cells (I) infected with VSV (MOI of 10). miR-27a expression was normalized to that of the U6 internal control in each sample. Data are shown as means 6 SD. Similar results were obtained from three independent experiments. 1320 DOWNREGULATION OF miR-27a INHIBITS IFN PRODUCTION Downloaded from http://www.jimmunol.org/

FIGURE 2. Viral infection downregulates miR-27a expression through the IFN/JAK/STAT1/RUNX1 signaling pathway in macrophages. (A) Quantitative PCR analysis of miR-27a expression in macrophages from wild-type or IFNAR-deficient mice treated with VSV (multiplicity of infection [MOI] of 10) or IFN-a (200 U/ml). (B) Quantitative PCR analysis of miR-27a in macrophages treated with various doses of IFN-a for 24 h (left) or 200 U/ml IFN-a across the time points (right). (C and E) Quantitative PCR analysis of miR-27a, pri–miR-27a, or RUNX1 in macrophages pretreated for 30 min with the indicated inhibitors and stimulated with 200 U/ml IFN-a for 24 h. RUNX1 mRNA expression was normalized to that of the b-actin internal control in each sample. (D) Immunoblot analysis of RUNX1 in macrophages treated with IFN-a (200 U/ml). b-actin served as a loading control. (F) Immunoblot analysis of STAT1

(top) or RUNX1 (bottom) in macrophages transfected with STAT1 siRNA or RUNX1 siRNA. (G) Quantitative PCR analysis of RUNX1 (left) or miR-27a by guest on September 26, 2021 (right) in macrophages transfected with STAT1 siRNA or RUNX1 siRNA and then stimulated with IFN-a (200 U/ml) for 24 h. Numbers below lanes indicate densitometry of the protein presented relative to that of b-actin internal control. Data are shown as means 6 SD. Similar results were obtained from three independent experiments. *p , 0.05, **p , 0.01.

Results fection with various pathogens, especially viruses, can signifi- Viral infection significantly downregulates miR-27a expression cantly downregulate miR-27a expression in innate cells, including in macrophages macrophages. We previously reported that a panel of miRNAs was significantly IFN/JAK/STAT1 pathway is responsible for viral infection– downregulated upon VSV infection in macrophages, and miR-27a induced miR-27a downregulation by inhibiting RUNX1 was identified as one of these miRNAs (25). However, the bio- The underlying mechanism of miR-27a downregulation was logical significance of miR-27a downregulation in the antiviral further examined. Because miRNAs are known to commonly be immune response is still unknown. In this study, we further regulated by type I IFN antiviral pathway, we first determined characterized miR-27a expression in infected peritoneal macro- whether miR-27a was downregulated by signaling upstream of phages and observed that the kinetic downregulation of mature IFN production or downstream of IFN/JAK/STAT1 effector sig- miR-27a following VSV infection occurred in a time- and dose- naling. We found that IFN-a treatment markedly downregulated dependent manner, where miR-27a was most significantly down- miR-27a expression in wild-type macrophages, but miR-27a regulated 24 h after VSV challenge (Fig. 1A). Infection with SeV, another RNA virus, also induced miR-27a downregulation in downregulation was abolished in IFNAR-deficient macrophages macrophages with similar kinetics (Fig. 1B). Additionally, we (Fig. 2A), suggesting that miR-27a downregulation was dependent assessed the ability of other pathogens or pathogen-associated on signals downstream of type I IFN. IFN-a–induced miR-27a molecules to regulate miR-27a expression and found that macro- downregulation also occurred in a time- and dose-dependent phages challenged with HSV (a DNA virus), IAV, TLR agonists manner (Fig. 2B). Furthermore, inhibition of STAT1 efficiently LPS and CpG oligodeoxynucleotide, and transfection of poly(I:C) blocked IFN-a–induced miR-27a and pri–miR-27a downregula- all downregulated miR-27a expression (Fig. 1C–G). In addition to tion, but inhibition of ERK, GSK, or NF-kB had little effect miR-27a downregulation in macrophages, miR-27a expression (Fig. 2C). Taken together, these results suggest that viral infec- was also downregulated by VSV infection in mouse PBMCs, tion downregulates miR-27a expression in macrophages mainly myeloid conventional dendritic cells, and plasmacytoid dendritic through a pathway dependent on IFN/JAK/STAT1 signaling. cells, as well as in human monocyte-derived macrophages and It has been reported that the transcription factor RUNX1 THP1 cells (Fig. 1H, 1I). Therefore, these data suggest that in- positively regulates miR-27a expression (34). We found that The Journal of Immunology 1321

FIGURE 3. Inhibition of miR-27a suppresses VSV-triggered type I IFN production and promotes VSV rep- lication. (A) Quantitative PCR anal- ysis of miR-27a in macrophages transfected with miR-27a or NC mimics, or with miR-27a or NC in- hibitors, at a final concentration of 20 nM for 48 h. (B–D) Macrophages were transfected as in (A) and in- fected by VSV at a multiplicity of infection (MOI) of 10 for the indi- Downloaded from cated time. IFN-b and IFN-a4 mRNA levels were detected by quantitative PCR, and IFN-b pro- duction in supernatant was measured by ELISA. (E)VSVTCID50 was measured in supernatants from

macrophages transfected as de- http://www.jimmunol.org/ scribed in (A) and infected by VSV at an MOI of 10 for 72 h. (F and G) Quantitative PCR analysis of intra- cellular VSV RNA (F) or superna- tant VSV RNA (G) in macrophages treated as in (E). Data are shown as means 6 SD. Similar results were obtained in three independent ex- periments. **p , 0.01. by guest on September 26, 2021

IFN-a treatment downregulated RUNX1 expression (Fig. 2D, firmed that transfecting miR-27a mimics into peritoneal mac- Supplemental Fig. 1A) and RUNX1 mRNA downregulation also rophages increased miR-27a expression and that transfecting depended on the STAT1-mediated signaling pathway (Fig. 2E). miR-27a inhibitor decreased its expression (Fig. 3A, Supplemen- Furthermore, STAT1 knockdown blocked the IFN-a–induced tal Fig. 1E). Overexpression of miR-27a promoted VSV-triggered downregulation of both RUNX1 and miR-27a expression, and type I IFN production, whereas inhibition of miR-27a suppressed IFN-a treatment failed to further inhibit miR-27a expression VSV-triggered type I IFN production compared with their respec- in macrophages with RUNX1 knockdown (Fig. 2F, 2G, tive controls (Fig. 3B–D, Supplemental Fig. 1F). Therefore, miR- Supplemental Fig. 1B). Overexpression of RUNX1 blocked the 27a positively regulates type I IFN production. downregulation of miR-27a expression induced by VSV infection To investigate the biological significance of VSV-induced or IFN-a treatment (Supplemental Fig. 1C, 1D), suggesting that downregulation of miR-27a in the elimination of viral infection RUNX1 downregulation is indispensable for IFN-a–induced de- in host cells, we examined the effect of miR-27a on VSV repli- crease of miR-27a. Together, IFN/JAK/STAT1 signaling is re- cation in macrophages. By measuring the VSV TCID50 levels in sponsible for the inducible downregulation of miR-27a by the supernatant from the infected macrophages, we found that inhibiting RUNX1 expression. overexpression of miR-27a suppressed VSV replication, whereas inhibition of miR-27a facilitated VSV replication (Fig. 3E). Downregulation of miR-27a feedback inhibits virus-triggered Consistent with these data, inhibition of miR-27a increased type I IFN production and consequently promotes VSV whereas overexpression of miR-27a decreased intracellular and replication in macrophages supernatant VSV RNA replicates (Fig. 3F, 3G). These results To determine whether VSV infection–induced downregulation of suggest that lower miR-27a levels allow VSV to replicate more miR-27a could affect host antiviral responses, we investigated the effectively in host cells, likely by inhibiting type I IFN production. effect of overexpressing or inhibiting miR-27a expression on type To exclude the possibility that overexpression of miR-27a had I IFN production in response to VSV challenge. We first con- a direct effect on VSV replication, we analyzed VSV RNA by 1322 DOWNREGULATION OF miR-27a INHIBITS IFN PRODUCTION

FIGURE 4. miR-27a directly tar- gets Siglec1 and TRIM27. (A) Se- quence alignment of miR-27a and its putative target sites in the pro- tein CDS or 39-UTR of Siglec1, and in the CDS of TRIM27. (B– D) Luciferase activity in lysates of HEK293T cells cotransfected with (B) pMIR-Siglec1 CDS, (C)pMIR- Siglec1 39-UTR, or (D)pMIR- TRIM27 CDS luciferase reporter plasmids and pTK-Renilla-luciferase plasmids together with miR-27a or NC mimics, or miR-27a or NC in- hibitors, for 24 h as indicated. Lu- ciferase activity is presented relative to Renilla luciferase activity. (E–G) Downloaded from Quantitative PCR and immunoblot analysis of Siglec1 and TRIM27 in macrophages transfected with miR- 27a or NC mimics, or miR-27a or NC inhibitors, for 48 h. Numbers below lanes indicate densitometry of the protein presented relative to that http://www.jimmunol.org/ of b-actin internal control. Data are shown as means 6 SD or represen- tative images. Similar results were obtained from three independent experiments. **p , 0.01. by guest on September 26, 2021

RNA22 miRNA target detection using both VSV sense and antisense Using RNA22 miRNA target prediction (35), we found that RNA sequences (35) and found three potential target sites in VSV Siglec1 and TRIM27 contained six and one putative miR-27a RNA (data not shown). However, miR-27a is less likely to directly target sites in their respective CDS, and one putative miR-27a target VSV RNA and is instead dependent on type I IFN production target site in the 39-UTR of Siglec1 (Fig. 4A). To confirm the to affect VSV replication, as we found no significant difference in possibility that miR-27a posttranscriptionally regulated Siglec1 VSV replication when miR-27a was inhibited or overexpressed in and TRIM27 expression, we constructed luciferase reporter plas- the absence of type I IFN signaling in IFNAR-deficient macro- mids containing the CDS of Siglec1 or TRIM27, or the 39-UTR phages (Supplemental Fig. 1G). Thus, we conclude that VSV- region of Siglec1. By cotransfection with miR-27a mimics or induced downregulation of miR-27a feedback inhibits type I IFN inhibitors, we observed that miR-27a mimics markedly decreased production, thus allowing VSV replication in host macrophages. luciferase levels, whereas miR-27a inhibitors increased luciferase levels compared with their respective controls (Fig. 4B–D). Siglec1 and TRIM27 mRNAs are direct targets of miR-27a Confirming that miR-27a inhibited the expression of both genes, miRNAs function to repress gene expression mainly by targeting transfection of miR-27a mimics decreased Siglec1 and TRIM27 its mRNA. Next, we sought to identify the possible miR-27a targets expression in macrophages at both the mRNA and protein levels, that modulate VSV-triggered type I IFN production. Using gene whereas miR-27a inhibitors increased expression of both genes expression profiling by microarray analysis to detect gene ex- (Fig. 4E–G). Taken together, these results demonstrate that en- pression upon miR-27a overexpression in macrophages, we found dogenous Siglec1 and TRIM27 are direct targets of miR-27a. that many genes were downregulated (Supplemental Fig. 2A). Among them, the two with the highest differential ratio between Antiviral function of miR-27a is mainly through targeting miR-27a–overexpressing and control macrophages were Siglec1 Siglec1 and TRIM27 and TRIM27 (Supplemental Table I). Also, the pathways affected Interestingly, we previously found that Siglec1 and TRIM27 are by miR-27a overexpression were analyzed for the significantly negative regulators of VSV-triggered type I IFN production (11). regulated genes from KEGG, and the results indicated that cyto- Phenocopying the effect of miR-27a overexpression, both Siglec1 solic DNA sensing, autophagy, TLR, and RIG-I–like receptor and TRIM27 knockdown significantly increased VSV-triggered pathways were the top four of significantly regulated pathways type I IFN production and accordingly inhibited VSV replica- (Supplemental Fig. 2B). These results indicated that miR-27a may tion (11). In murine peritoneal macrophages, inhibition of type I mainly participate in regulation of host antiviral signaling, and the IFN production and increase of VSV replication induced by miR- two most downregulated genes Siglec1 and TRIM27 may be 27a inhibition was rescued by Siglec1 or TRIM27 knockdown targeted by miR-27a and responsible for the mechanisms of miR- respectively (Fig. 5A, 5B). These results suggested that Siglec1 or 27a–mediated immune regulation. TRIM27 knockdown phenocopied the effect of miR-27a over- The Journal of Immunology 1323

FIGURE 5. Antiviral function of miR- 27a is mainly through targeting Siglec1 and TRIM27. (A and B) Quantitative PCR analysis of IFN-b mRNA and VSV RNA in macrophages transfected with miR-27a or NC inhibitor plus Siglec1 siRNA or TRIM27 siRNA as indicated and infected with VSV (multiplicity of infection [MOI] of 10) at the indicated time points. (C and D) Quantitative PCR analysis of IFN-b Downloaded from mRNA and VSV RNA in RAW264.7 cells transfected with miR-27a or NC mimics plus Siglec1 or TRIM27 plasmids, as in- dicated, and infected with VSV (MOI of 10) for 12 h. (E) Immunoblot analysis of p-IRF3, TBK1, Siglec1, and TRIM27 in http://www.jimmunol.org/ macrophages transfected with miR-27a mimics or NC and infected with VSV (MOI of 10) at the indicated times. Numbers be- low lanes indicate densitometry of the pro- tein presented relative to that of b-actin internal control. Data are shown as means 6 SD or representative images. Similar results were obtained in three independent experi- ments. **p , 0.01. by guest on September 26, 2021

expression and counteracted the effect of miR-27a inhibition. Sponge target construct designed to compete with endogenous Furthermore, overexpression of Siglec1 or TRIM27 reversed the miR-27a targets. The Sponge expression vector was constructed miR-27a overexpression–mediated increase of type I IFN pro- (Fig. 6A, Supplemental Fig. 3A, 3B) and the efficacy was tested duction and inhibition of VSV replication (Fig. 5C, 5D). Taken in vitro. In RAW264.7 macrophages, transfection with the Sponge together, we conclude that miR-27a upregulates VSV-triggered vectors significantly downregulated miR-27a expression and re- type I IFN production mainly through targeting negative regula- versed the miR-27a overexpression–mediated upregulation of tors Siglec1 and TRIM27. Additionally, we observed lower type I IFN production, with 73Sponge being the most efficient Siglec1 and TRIM27 expression but higher TBK1 and phos- construct (Supplemental Fig. 3C). Siglec1 and TRIM27 mRNA phorylated IRF3 in macrophages after overexpression of miR-27a were also increased in the cells transfected with the 73Sponge compared with that in controls in response to VSV infection construct (Supplemental Fig. 3D). Thus, the miR-27a Sponge– (Fig. 5E). Taken together, type I IFN–induced downregulation of expressing vector was functional in vitro. miR-27a leads to upregulation of negative regulators Siglec1 and Next we generated transgenic mice expressing the miR-27a TRIM27 expression, consequently resulting in the suppression of 73Sponge construct using microinjection technology (GS27 type I IFN production. mice). The F1 generation from the GS27 founder mice was vali- dated by PCR using primers specific for the 73Sponge sequence In vivo inhibition of miR-27a decreases type I IFN production (Supplemental Fig. 3E). These GS27 F1 mice expressed lower and promotes viral replication via upregulation of Siglec1 and level of endogenous miR-27a but higher levels of Siglec1 and TRIM27 TRIM27 than those in their littermate controls (Fig. 6B). Impor- To assess the effect of the decreased miR-27a expression on type I tantly, higher Siglec1 and TRIM27 expression but lower TBK1 IFN production in vivo, we generated transgenic mice expressing a and phosphorylated IRF3 were detected in macrophages harvested 1324 DOWNREGULATION OF miR-27a INHIBITS IFN PRODUCTION

FIGURE 6. In vivo downregulation of miR-27a decreases type I IFN produc- tion and promotes viral replication. (A) Alignment of miR-27a and the miR-27a Sponge sequence. Superscripting repre- sents the bulge in the Sponge sequence. miR-27a Sponge expression vector con- tained a GFP gene and a 39-UTR in- cluding seven repeats of the miR-27a Sponge. (B) Quantitative PCR and im- munoblot analysis of miR-27a, Siglec1, and TRIM27 expression in mouse peri- toneal macrophages from wild-type (WT) andGS27mice.(C) Immunoblot analysis of Siglec1, TRIM27, TBK1, and p-IRF3 in macrophage lysates from GS27 mice Downloaded from infected with VSV (multiplicity of in- fection of 10) at the indicated time. (D and E)IFN-b levels were evaluated by ELISA, and IFN-b mRNA levels were evaluated by quantitative PCR, from the indicated organs in GS27 mice 12 h after i.p. injection of VSV (1 3 108 PFU/g). http://www.jimmunol.org/ (F) Quantitative PCR analysis of intracellular VSV RNA (left)and

VSV TCID50 (right) in the indicated organs from GS27 mice as described in (D). (G) H&E staining of lung sections from GS27 mice as described in (D). Scale bars, 100 mm. (H and I) Working model for the mechanism by which type

I IFN–induced downregulation of miR- by guest on September 26, 2021 27a negatively regulates type I IFN production in feedback manner via up- regulating Siglec1/TRIM27 to degrade TBK1 and impair IRF3 signaling in the initiation of type I IFN production. Numbers below lanes indicate densi- tometry of the protein presented relative to that of b-actin internal control. Data are shown as means 6 SD or repre- sentative images. Similar results were obtained from three independent exper- iments. *p , 0.05, **p , 0.01.

from GS27 mice compared with those in controls in response to which allows these two negative regulators to be upregulated and VSV infection (Fig. 6C). consequently inhibits type I IFN production in vivo (Fig. 6H, 6I). To investigate the role of miR-27a in the antiviral response in vivo, GS27 mice were infected i.p. with VSV. We observed a Discussion greater decrease of IFN-b production and increase of viral burden miR-27a was first reported by Tuschl and colleagues (36), and in GS27 mice than those found in their littermate controls earlier studies have identified miR-27a as an oncogenic miRNA, (Fig. 6D–F). Consistent with this finding, higher viral infec- with high expression in tumor (37). In macrophages, downregu- tion–mediated pathology as well as more severe infiltration of lation of miR-27a was previously reported to occur upon herpes polymorphonuclear cells and interstitial pneumonitis were found virus infection at the posttranscriptional level (38, 39). In the in the lungs of GS27 mice upon VSV challenge (Fig. 6G). These present study, we found that viral infection downregulated in vivo data convincingly demonstrate that inducible downregu- miR-27a in macrophages at the transcriptional level via an IFN/ lation of miR-27a inhibits IFN production by saving the miR-27a JAK/STAT1/RUNX1 signaling–dependent manner, and pri–miR- targets Siglec1 and TRIM27 from miR-27a–mediated suppression, 27a downregulation was also blocked by STAT1 inhibition or The Journal of Immunology 1325

RUNX1 knockdown. This is supported by the finding that RUNX1 exploit these negative regulatory mechanisms to subvert the im- can positively regulate miR-27a expression (34). Furthermore, mune responses, thus helping to promote their own replication RUNX1 overexpression blocked the IFN treatment–induced miR- and evasion in host cells. As reported that miRNAs are mediators 27a downregulation, suggesting that RUNX1 downregulation is of viral evasion of the immune system (46), we found that various indispensable for the IFN treatment–induced decrease of miR-27a. pathogens and pathogen-associated molecules can downregulate Additionally, the gap between STAT1 activation and inhibition of miR-27a, so the miR-27a downregulation may be a common RUNX1 expression still exists, and how STAT1 activation inhibits mechanism for immune evasion by different families of viruses. RUNX1 requires further investigation. Therefore, miR-27a may bear considerable potential as a new In mammals, miRNAs are well accepted to inhibit mRNA target for the development of therapies against viral infection and translation by binding to the 39-UTR of target mRNAs. In plants, infection-related diseases. however, most identified miRNAs bind the protein CDS of their Based on our findings, we propose the following working target mRNAs with high sequence complementarity and induce model to explain how inducible downregulation of miR-27a feed- RNA degradation, which is similar to the RNA interference back inhibited type I IFN production in the antiviral innate re- mechanism. Using microarray analysis, Siglec1 and TRIM27 were sponse. First, miR-27a expression was significantly downregulated identified as two of the most downregulated genes upon miR-27a upon viral infection through the IFN/JAK/STAT1/RUNX1 sig- overexpression in macrophages. Additionally, the CDS region of naling pathway in macrophages. Then this inducible downreg- both Siglec1 and TRIM27 mRNA was determined to be targeted ulation of miR-27a allowed the increased expression of its by miR-27a. Previous reports of Nanog, Oct4, Sox2, and targets Siglec1 and TRIM27, which in turn inhibited type I IFN

DNMT3b expression regulated by miRNAs also support that production. Downloaded from miRNAs can target the protein CDS of mRNAs (18, 40, 41). Thus, the concept that miRNAs can tightly regulate gene expression by Acknowledgments targeting various regions within an mRNA may facilitate the We thank Prof. Wei Pan for providing VSV, Prof. Hangping Yao for discovery of more miRNA-targeted genes and enhance our un- providing IAV, Tingting Fang, Mei Jin, and Yan Li for technical assis- derstanding of the complicated regulatory loops in mammals. tance, and Drs. Xingguang Liu, Chaofeng Han, Sheng Xu, Yanmei Han,

We previously found that inducible upregulation of Siglec1 and Taoyong Chen for valuable discussions. http://www.jimmunol.org/ associated with the adapter DNAX activation protein 12 and ac- tivated the scaffolding function of Src homology 2 domain– Disclosures containing protein tyrosine phosphatase 2, which subsequently The authors have no financial conflicts of interest. recruited TRIM27 to bind TBK1. The TRIM27 B-box domain then functions as an E3 ligase to induce K48-linked ubiquitination at both Lys251 and Lys372 residues of TBK1, consequently leading References 1. O’Neill, L. A., D. Golenbock, and A. G. Bowie. 2013. The history of Toll-like to proteasomal degradation of TBK1 and suppression of IRF3 receptors—redefining innate immunity. Nat. Rev. Immunol. 13: 453–460. phosphorylation, thus inhibiting type I IFN production (11). 2. Broz, P., and D. M. Monack. 2013. Newly described pattern recognition re- Moreover, although VSV infection or IFN treatment causes large ceptors team up against intracellular pathogens. Nat. Rev. Immunol. 13: 551–565. by guest on September 26, 2021 3. Rusinova, I., S. Forster, S. Yu, A. Kannan, M. Masse, H. Cumming, R. Chapman, increases in positive antiviral mediators and single miR-27a target and P. J. Hertzog. 2013. Interferome v2.0: an updated database of annotated (Siglec1 or TRIM27) has relatively modest negative effects (11), interferon-regulated genes. Nucleic Acids Res. 41: D1040–D1046. interestingly, miR-27a directly targets both Siglec1 and TRIM27 4. Liu, S. Y., R. Aliyari, K. Chikere, G. Li, M. D. Marsden, J. K. Smith, O. Pernet, H. Guo, R. Nusbaum, J. A. Zack, et al. 2013. Interferon-inducible cholesterol- in the same negative regulatory pathway. The cumulative nega- 25-hydroxylase broadly inhibits viral entry by production of 25-hydrox- tive effects of both Siglec1 and TRIM27 mediated by miR-27a ycholesterol. Immunity 38: 92–105. downregulation may represent a potent feedback negative regu- 5. Ivashkiv, L. B., and L. T. Donlin. 2014. Regulation of type I interferon responses. Nat. Rev. Immunol. 14: 36–49. latory mechanism for type I IFN production. Thus, miR-27a ex- 6. Yoo, J. S., H. Kato, and T. Fujita. 2014. Sensing viral invasion by RIG-I like pression in immune cells may tightly regulate major players in the receptors. Curr. Opin. Microbiol. 20: 131–138. Siglec1 regulatory pathway. Based on this observation, restoring 7. Chen, Y., J. Chen, H. Wang, J. Shi, K. Wu, S. Liu, Y. Liu, and J. Wu. 2013. HCV- induced miR-21 contributes to evasion of host immune system by targeting miR-27a may be an effective way to target and suppress the in- MyD88 and IRAK1. PLoS Pathog. 9: e1003248. hibitory Siglec1/TRIM27 pathway to enhance type I IFN pro- 8. Cui, J., L. Zhu, X. Xia, H. Y. Wang, X. Legras, J. Hong, J. Ji, P. Shen, S. Zheng, duction. In addition to Siglec1 and TRIM27 identified in the Z. J. Chen, and R. F. Wang. 2010. NLRC5 negatively regulates the NF-kB and type I interferon signaling pathways. Cell 141: 483–496. present study, it remains likely that miR-27a also targets other 9. Cui, J., Y. Li, L. Zhu, D. Liu, Z. Songyang, H. Y. Wang, and R. F. Wang. 2012. currently unidentified molecules or signaling pathways that reg- NLRP4 negatively regulates type I interferon signaling by targeting the kinase ulate the host antiviral response. This presumption may lead to TBK1 for degradation via the ubiquitin ligase DTX4. Nat. Immunol. 13: 387–395. interesting future work to reveal precisely how miR-27a functions 10. Chen, W., C. Han, B. Xie, X. Hu, Q. Yu, L. Shi, Q. Wang, D. Li, J. Wang, P. Zheng, et al. 2013. Induction of Siglec-G by RNA viruses inhibits the innate in different ways to regulate immune responses. immune response by promoting RIG-I degradation. Cell 152: 467–478. Although effective host defense against viral infection requires 11. Zheng, Q., J. Hou, Y. Zhou, Y. Yang, B. Xie, and X. Cao. 2015. Siglec1 sup- the induction of appropriate innate immune responses, excessive presses antiviral innate immune response by inducing TBK1 degradation via the ubiquitin ligase TRIM27. Cell Res. 25: 1121–1136. activation of antiviral responses may cause the development of 12. Palucka, A. K., J. P. Blanck, L. Bennett, V. Pascual, and J. Banchereau. 2005. acute or chronic inflammatory disorders (42, 43). Therefore, im- Cross-regulation of TNF and IFN-a in autoimmune diseases. Proc. Natl. Acad. mune cells need to employ various negative regulatory mecha- Sci. USA 102: 3372–3377. 13. Berggren, O., A. Alexsson, D. L. Morris, K. Tandre, G. Weber, T. J. Vyse, nisms to avoid uncontrolled immune responses when encountering A. C. Syva¨nen, L. Ro¨nnblom, and M. L. Eloranta. 2015. IFN-a production by viral invasion (44). It has been reported that miR-27a serves as an plasmacytoid dendritic cell associations with polymorphisms in gene loci related important negative feedback mechanism to regulate inflammatory to autoimmune and inflammatory diseases. Hum. Mol. Genet. 24: 3571–3581. 14. Musto, A., A. Navarra, A. Vocca, A. Gargiulo, G. Minopoli, S. Romano, response of macrophages by targeting IL-10 (45). In the present M. F. Romano, T. Russo, and S. Parisi. 2015. miR-23a, miR-24 and miR-27a study, the inducible downregulation of miR-27a leading to up- protect differentiating ESCs from BMP4-induced apoptosis. Cell Death Differ. regulation of the Siglec1/TRIM27 pathway and increase of TBK1 22: 1047–1057. 15. Yang, X. W., L. J. Zhang, X. H. Huang, L. Z. Chen, Q. Su, W. T. Zeng, W. Li, degradation may represent a new feedback regulatory mechanism and Q. Wang. 2014. miR-145 suppresses cell invasion in hepatocellular carci- within the host antiviral immune response. However, viruses may noma cells: miR-145 targets ADAM17. Hepatol. Res. 44: 551–559. 1326 DOWNREGULATION OF miR-27a INHIBITS IFN PRODUCTION

16. Tao, Z. H., J. L. Wan, L. Y. Zeng, L. Xie, H. C. Sun, L. X. Qin, L. Wang, J. Zhou, 32. Liu, J., C. Han, B. Xie, Y. Wu, S. Liu, K. Chen, M. Xia, Y. Zhang, L. Song, Z. Li, Z. G. Ren, Y. X. Li, et al. 2013. miR-612 suppresses the invasive-metastatic et al. 2014. Rhbdd3 controls autoimmunity by suppressing the production of IL- cascade in hepatocellular carcinoma. J. Exp. Med. 210: 789–803. 6 by dendritic cells via K27-linked ubiquitination of the regulator NEMO. Nat. 17. Sullivan, C. S., and D. Ganem. 2005. MicroRNAs and viral infection. Mol. Cell Immunol. 15: 612–622. 20: 3–7. 33. Ma, F., S. Xu, X. Liu, Q. Zhang, X. Xu, M. Liu, M. Hua, N. Li, H. Yao, and 18. Helwak, A., G. Kudla, T. Dudnakova, and D. Tollervey. 2013. Mapping the X. Cao. 2011. The microRNA miR-29 controls innate and adaptive immune human miRNA interactome by CLASH reveals frequent noncanonical binding. responses to intracellular bacterial infection by targeting interferon-g. Nat. Cell 153: 654–665. 19. Yates, L. A., C. J. Norbury, and R. J. Gilbert. 2013. The long and short of Immunol. 12: 861–869. microRNA. Cell 153: 516–519. 34. Ben-Ami, O., N. Pencovich, J. Lotem, D. Levanon, and Y. Groner. 2009. A 20. Li, Y., and X. Shi. 2013. MicroRNAs in the regulation of TLR and RIG-I regulatory interplay between miR-27a and Runx1 during megakaryopoiesis. pathways. Cell. Mol. Immunol. 10: 65–71. Proc. Natl. Acad. Sci. USA 106: 238–243. 21. Lu, F., A. Weidmer, C. G. Liu, S. Volinia, C. M. Croce, and P. M. Lieberman. 35. Miranda, K. C., T. Huynh, Y. Tay, Y. S. Ang, W. L. Tam, A. M. Thomson, 2008. Epstein-Barr virus-induced miR-155 attenuates NF-kB signaling and B. Lim, and I. Rigoutsos. 2006. A pattern-based method for the identification of stabilizes latent virus persistence. J. Virol. 82: 10436–10443. microRNA binding sites and their corresponding heteroduplexes. Cell 126: 22. Lee, H. M., D. T. Nguyen, and L. F. Lu. 2014. Progress and challenge of 1203–1217. microRNA research in immunity. Front. Genet. 5: 178. 36. Lagos-Quintana, M., R. Rauhut, W. Lendeckel, and T. Tuschl. 2001. Identi- 23. Wang, P., J. Hou, L. Lin, C. Wang, X. Liu, D. Li, F. Ma, Z. Wang, and X. Cao. fication of novel genes coding for small expressed RNAs. Science 294: 853– 2010. Inducible microRNA-155 feedback promotes type I IFN signaling in an- 858. tiviral innate immunity by targeting suppressor of cytokine signaling 1. J. 37. Li, Z., S. Hu, J. Wang, J. Cai, L. Xiao, L. Yu, and Z. Wang. 2010. MiR-27a Immunol. 185: 6226–6233. modulates MDR1/P-glycoprotein expression by targeting HIPK2 in human 24. Zhu, B., J. Ye, Y. Nie, U. Ashraf, A. Zohaib, X. Duan, Z. F. Fu, Y. Song, H. Chen, ovarian cancer cells. Gynecol. Oncol. 119: 125–130. and S. Cao. 2015. MicroRNA-15b modulates Japanese encephalitis virus-mediated 38. Buck, A. H., J. Perot, M. A. Chisholm, D. S. Kumar, L. Tuddenham, V. Cognat, inflammation via targeting RNF125. J. Immunol. 195: 2251–2262. 25. Hou, J., P. Wang, L. Lin, X. Liu, F. Ma, H. An, Z. Wang, and X. Cao. 2009. L. Marcinowski, L. Do¨lken, and S. Pfeffer. 2010. Post-transcriptional regulation MicroRNA-146a feedback inhibits RIG-I-dependent type I IFN production in mac- of miR-27 in murine cytomegalovirus infection. RNA 16: 307–315. Downloaded from rophages by targeting TRAF6, IRAK1, and IRAK2. J. Immunol. 183: 2150–2158. 39. Cazalla, D., T. Yario, and J. A. Steitz. 2010. Down-regulation of a host micro- 26. Papadopoulou, A. S., J. Dooley, M. A. Linterman, W. Pierson, O. Ucar, RNA by a Herpesvirus saimiri noncoding RNA. Science 328: 1563–1566. B. Kyewski, S. Zuklys, G. A. Hollander, P. Matthys, D. H. Gray, et al. 2012. The 40. Duursma, A. M., M. Kedde, M. Schrier, C. le Sage, and R. Agami. 2008. thymic epithelial microRNA network elevates the threshold for infection- miR-148 targets human DNMT3b protein coding region. RNA 14: 872– associated thymic involution via miR-29a mediated suppression of the IFN-a 877. receptor. Nat. Immunol. 13: 181–187. 41. Tay, Y., J. Zhang, A. M. Thomson, B. Lim, and I. Rigoutsos. 2008. MicroRNAs 27. Wang, P., Y. Xue, Y. Han, L. Lin, C. Wu, S. Xu, Z. Jiang, J. Xu, Q. Liu, and to Nanog, Oct4 and Sox2 coding regions modulate embryonic stem cell differ-

X. Cao. 2014. The STAT3-binding long noncoding RNA lnc-DC controls human entiation. Nature 455: 1124–1128. http://www.jimmunol.org/ dendritic cell differentiation. Science 344: 310–313. 42. Foy, E., K. Li, R. Sumpter, Jr., Y. M. Loo, C. L. Johnson, C. Wang, P. M. Fish, 28. Hou, J., L. Lin, W. Zhou, Z. Wang, G. Ding, Q. Dong, L. Qin, X. Wu, Y. Zheng, M. Yoneyama, T. Fujita, S. M. Lemon, and M. Gale, Jr. 2005. Control of antiviral Y. Yang, et al. 2011. Identification of miRNomes in human liver and hepato- defenses through hepatitis C virus disruption of retinoic acid-inducible gene-I cellular carcinoma reveals miR-199a/b-3p as therapeutic target for hepatocel- signaling. Proc. Natl. Acad. Sci. USA 102: 2986–2991. lular carcinoma. Cancer Cell 19: 232–243. 43. Forster, S. 2012. Interferon signatures in immune disorders and disease. Immu- 29. Zhang, Q., K. Zhao, Q. Shen, Y. Han, Y. Gu, X. Li, D. Zhao, Y. Liu, C. Wang, X. Zhang, et al. 2015. Tet2 is required to resolve inflammation by recruiting nol. Cell Biol. 90: 520–527. Hdac2 to specifically repress IL-6. Nature 525: 389–393. 44. Hengel, H., U. H. Koszinowski, and K. K. Conzelmann. 2005. Viruses know it 30. Hou, J., Y. Zhou, Y. Zheng, J. Fan, W. Zhou, I. O. Ng, H. Sun, L. Qin, S. Qiu, all: new insights into IFN networks. Trends Immunol. 26: 396–401. J. M. Lee, et al. 2014. Hepatic RIG-I predicts survival and interferon-a thera- 45. Xie, N., H. Cui, S. Banerjee, Z. Tan, R. Salomao, M. Fu, E. Abraham, peutic response in hepatocellular carcinoma. Cancer Cell 25: 49–63. V. J. Thannickal, and G. Liu. 2014. miR-27a regulates inflammatory response of

31. Chen, T., M. Yang, Z. Yu, S. Tang, C. Wang, X. Zhu, J. Guo, N. Li, W. Zhang, macrophages by targeting IL-10. J. Immunol. 193: 327–334. by guest on September 26, 2021 J. Hou, et al. 2014. Small GTPase RBJ mediates nuclear entrapment of MEK1/ 46. Cullen, B. R. 2013. MicroRNAs as mediators of viral evasion of the immune MEK2 in tumor progression. Cancer Cell 25: 682–696. system. Nat. Immunol. 14: 205–210.