APCs such as myeloid dendritic cells (DCs) are key sentinels of the innate immune system. In response to pathogen recognition and innate immune stimulation, DCs transition from an immature to a mature state that is characterized by widespread changes in host gene expression, which include the upregulation of cytokines, chemokines, and costimulatory factors to protect against infection. Several transcription factors are known to drive these gene expression changes, but the mechanisms that negatively regulate DC maturation are less well understood. In this study, we identify the transcription factor IL enhancer binding factor 3 (ILF3) as a negative regulator of innate immune responses and DC maturation. Depletion of ILF3 in primary human monocyte-derived DCs led to increased expression of maturation markers and potentiated innate responses during stimulation with viral mimetics or classic innate agonists. Conversely, overexpression of short or long ILF3 isoforms (NF90 and NF110) suppressed DC maturation and innate immune responses. Through mutagenesis experiments, we found that a nuclear localization sequence in ILF3, and not its dual dsRNA-binding domains, was required for this function. Mutation of the domain associated with zinc finger motif of ILF3’s NF110 isoform blocked its ability to suppress DC maturation. Moreover, RNA-sequencing analysis indicated that ILF3 regulates genes associated with cholesterol homeostasis in addition to genes associated with DC maturation. Together, our data establish ILF3 as a transcriptional regulator that restrains DC maturation and limits innate immune responses through a mechanism that may intersect with lipid metabolism.

Visual Abstract

This article is featured in Top Reads, p.2773

Engagement of the innate immune system in response to an invading pathogen is a central element of host defense. Viral, bacterial, and fungal pathogens can be detected by a wide variety of innate immune cells. Myeloid dendritic cells (DCs) play key roles in this response by initiating local innate responses and programing subsequent adaptive immune responses (1). Immature DCs are poised to respond to pathogen components and inflammatory cytokines, which trigger morphological and functional changes that facilitate Ag presentation, cytokine/chemokine secretion, and expression of costimulatory molecules. These functional changes are characteristic of mature DCs, which have the capacity to prime naive T cells and program adaptive immunity (2). Many aspects of DC biology are governed by transcription factors that regulate gene expression signatures and influence cell behavior, including DC maturation. Although efficient DC maturation may be beneficial for controlling an active infection, sustained or excessive innate responses following maturation can have deleterious pathologic effects (3). Thus, it is important to decipher the complex molecular mechanisms that regulate innate immune responses and DC maturation and understand their function in health and disease.

DCs express an array of pattern recognition receptors that serve as “sensors” positioned to detect extracellular pathogens (e.g., TLR3 and TLR4) and intracellular pathogens (e.g., cGAS, IFI16, and RIG-I). Viral nucleic acids within infected cells are detected by DNA and RNA sensors such as cGAS and RIG-I, respectively. These sensors activate signaling cascades that lead to phosphorylation of transcription factors such as IRF3 and NF-κB that drive induction of inflammatory cytokines, including a potent class of antiviral signaling molecules known as type I IFNs. Autocrine and paracrine signaling of IFNs through IFNR and the JAK–STAT pathway induce a battery of IFN-stimulated genes (ISGs) that can act to directly restrict viral replication or modify cellular processes to establish an antiviral state. During chronic infection with viruses such as HIV-1, sustained production of type I IFNs and expression of ISGs can exacerbate nonspecific inflammation, increase target cell susceptibility to the virus, and contribute to pathogenesis (4, 5). Given the pivotal role of DCs in linking innate and adaptive immunity, a deeper understanding of the mechanisms that restrain IFN responses or negatively regulate DC maturation during infection could lead to treatments that engage protective antiviral immune responses during viral transmission and may guide the development of therapies for late-stage disease.

Numerous transcription factors essential for DC differentiation and maturation have been identified, yet the factors that restrain and fine-tune this response remain poorly understood (6). In this report, we have identified IL enhancer binding factor 3 (ILF3) as one such factor. ILF3 was originally discovered as a positive regulator of IL2 transcription as part of the NFAT–AP1–NF-κB enhanceosome and a positive regulator of IL2 mRNA stabilization via 3′ untranslated region binding (711). Subsequent studies have uncovered diverse functions for ILF3 that include a role in host RNA decay as a component of a messenger ribonucleoprotein complex and a role as a transcriptional modulator for ILs in addition to IL-2, such as IL-13 (12) and the oncogene uPA (13). Additionally, ILF3 can interact directly with viral nucleic acids to modulate replication of dengue virus (14), hepatitis C virus (15), bovine viral diarrhea virus (16), human rhinovirus, and Zaire ebolavirus (17). Despite these known roles in regulating transcription, mRNA stabilization, interaction with viral nucleic acids, and regulation of pathogen fitness, ILF3 has not previously been described to regulate the function of myeloid immune cells.

In this study, we demonstrate that both major isoforms of ILF3, NF90 and NF110, restrain myeloid DC maturation and type I IFN responses. Analysis of deletion mutants reveals that the nuclear localization sequences of both NF90 and NF110 are required for this function. Interestingly, mutation of the domain associated with zinc finger (DZF) domain ablated the ability of NF110 to suppress DC maturation but not NF90. Furthermore, through RNA-sequencing (RNA-seq) analysis of DCs expressing mutant or wild type (wt) ILF3, we demonstrate that ILF3-dependent genes are strongly enriched for genes associated with cholesterol homeostasis. These data establish ILF3 as a regulator of DC responses to innate immune stimuli and therefore a potential target for host-directed therapies to fine-tune inflammation and innate immunity.

We generated immature monocyte-derived DCs (MDDCs) as previously described (18). Briefly, leukocytes from anonymized healthy human donors were acquired under the Bloodworks Donor Products for Research and Test Development/Standardization External Investigators Protocol (Western Institutional Review Board protocol 20150119), and informed consent was obtained from all subjects. CD14+ monocytes from PBMC buffy coats were isolated with anti-human CD14 magnetic beads (Miltenyi Biotec; catalog [Cat] no. 130-050-201) and cultured in RPMI 1640 (Thermo Fisher Scientific) containing 10% heat-inactivated FBS (Peak Serum), 50 U/ml penicillin and 50 μg/ml streptomycin (pen/strep, Thermo Fisher Scientific), 10 mM HEPES (Sigma), 2-ME (Thermo Fisher Scientific), and 2 mM l-glutamine (Thermo Fisher Scientific) in the presence of recombinant human GM-CSF at 10 ng/ml and IL-4 at 50 ng/ml (PeproTech). The day following isolation and transduction with lentiviral vectors, fresh media and cytokines were added to cells (50% by volume). On the fourth day postisolation, cells were resuspended in fresh media and cytokines for subsequent experiments. 293FT cells (Life Technologies; Cat no. R70007, Research Resource Identifiers: CVCL_6911) were cultured in DMEM (Thermo Fisher Scientific) supplemented with 10% FBS, pen/strep, and 10 mM HEPES, and with 0.1 mM MEM nonessential amino acids (Thermo Fisher Scientific), 6 mM glutamine, and 1 mM sodium pyruvate (Thermo Fisher Scientific). HL116 cells were cultured similar to 293FT cells, except with the addition of hypoxanthine–aminopterin–thymidine medium (Thermo Fisher Scientific; Cat 21060-017). THP-1 cells (American Type Culture Collection; Cat no. TIB-202, Research Resource Identifiers: CVCL_0006) were cultured in RPMI with 10% heat-inactivated FBS, pen/strep, 10 mM HEPES, 2-ME, and 2 mM glutamine and kept at a density between 250,000 and 1 million cells/ml. All cells were maintained at 37°C and 5% CO2 and used at early passage numbers (<20 passages for 293FT and 8 wk for HL116 and THP-1). Mycoplasma contaminant checks were performed every 6 mo. THP-1 experiments were performed on biological replicates from independent cultures, and MDDC experiments were performed using individual donors as biological replicates.

HIV-1–GFP is env- vpu- vpr- vif- nef-, with the GFP open reading frame in place of nef (19). Vpx-containing virus-like particles were generated from the plasmid pSIV3+ (20). Overexpression vectors for wt isoforms of ILF3, dual dsRNA-binding domain (dsRBD) deletion mutants, DZF deletion mutants, and nuclear localization signal deletion mutants were generated in-house using overlap extension mutagenesis to modify NF90 cDNA (GE Dharmacon) or directly synthesized (Thermo Fisher Scientific GeneArt) (NF90b mutants and NF110b forms or GE Dharmacon) (NF90b cDNA) in a pLKO.1 vector backbone. lentiCRISPR single guide RNAs (sgRNAs) were designed using the E-CRISP algorithm (http://www.e-crisp.org/E-CRISP/), selecting for the highest scoring guides in target specificity and efficiency for ILF3. Oligos spanning the sgRNA sequence were annealed and ligated into the lentiCRISPRv2 backbone (21) (Addgene plasmid no. 52961, gift from Feng Zhang). The sequences are as follows: ILF3 target sequence no. 1 GCTGGAGGCAGTCCAGAACA and ILF3 target sequence no. 2: GCCTCCAGCTCCTCTTGTGT. The parental control vector (LCV2) was created from lentiCRISPRv2 digested with BsmBI and religated, removing the 2-kb stuffer. All lentiviral constructs were transformed into Stbl3 bacteria (Thermo Fisher Scientific; Cat C737303) for propagation of plasmid DNA. All plasmids were prepared using a NucleoBond Xtra Maxi Kit (Takara Bio; Cat740414.100). Coding sequences of overexpression constructs, short hairpin RNA (shRNA) hairpins, and sgRNAs were confirmed by automated sequencing (Genewiz).

As described (18), lentivirus stocks were produced by PEI-mediated transfection into 293FT cells. For lentiviral vectors, plasmid amounts were 3.4 μg CMV–vesicular stomatitis virus (VSV)–G (Addgene; plasmid no. 8454), 9 μg psPax2 (Addgene; plasmid no. 12260), and 10.1 μg transgene (LKO.1 control [Addgene; Cat 10878], LKO ILF3 overexpression vector, shRNA control [Sigma; Cat SHC002], ILF3 shRNA, or lentiCRISPRv2 constructs). For HIV-1–GFP, plasmid amounts were 3.4 μg CMV–VSV-G and 19.1 μg HIV-1–GFP cassette. Virus-like particles containing Vpx were produced using 3.4 μg CMV–VSV-G and 19.1 μg pSIV3+. Medium was washed and refreshed the morning after transfection, and virus supernatants were harvested 32 h later. Sufficient p24 levels were verified using Lenti-Go Stix Plus (Takara Bio; Cat 631280). Supernatants were passed through 0.45-μm syringe filters (Corning), transferred to thin-wall Conical tubes (Beckman Coulter), and concentrated by ultracentrifugation at 24,000 rpm for 2 h at 4°C in a SW28 swing-bucket rotor (Beckman Coulter). Pellets were resuspended in RPMI–DC medium without cytokines, and insoluble material was clarified by centrifuging at 700 rcf for 4 min. Then, 50× concentrated viral stocks were frozen at −80°C and titrated on 293FT and THP-1 cells.

MDDCs were modified by lentiviral shRNA and overexpression constructs similar to previously described protocols (18). Isolated CD14+ monocytes were resuspended in medium with cytokines and polybrene (Sigma, 1 μg/ml) and aliquoted to 96-well U-bottom plates with 200,000 cells in 150 μl per well. Supernatant containing virus-like particles packaging Vpx was added to overcome the block to reverse transcription ∼30 min prior to adding lentiviral vectors. Then, 10 μl of concentrated lentiviral stocks were used to transduce 200,000 CD14+ cells. shRNA clones for targeting ILF3 were used independently (Sigma; ILF3 shRNA1 [sh1]: TRCN0000329787, ILF3 shRNA2 [sh2]: TRCN0000329786). THP-1 monocytic cells were transduced with shRNA or lentiCRISPR constructs in six-well cluster plates using 1 million cells per well in 2 ml of medium with polybrene (2 μg/ml) and concentrated viral stocks (150 μl of shRNA or 250 μl of lentiCRISPR per well). Cells were placed under selection with puromycin (1 μg/ml, InvivoGen) 2 d after transduction for 1 wk, and puromycin-resistant populations were allowed to expand. lentiCRISPR-transduced cells were used at day 8 for infection. shRNA-transduced cells were used for experiments beginning 2 wk after selection. Independent transductions were performed for biological replicates, unless otherwise indicated. Perturbation of ILF3 expression was confirmed by quantitative PCR (qPCR) or immunoblot.

MDDCs were infected with HIV-1–GFP for 48 h beginning on day 4 after differentiation. DCs were spun down on day 4 and resuspended in fresh medium with GM-CSF, IL-4, and polybrene (1 μg/ml). For most assays, DCs were plated in round-bottom 96-well plates in 75 μl. Infections and stimulations were performed by diluting virus in MDDC medium (without cytokines or polybrene) to a final volume normalized to control (150 μl per well). Antiretroviral drugs (National Institutes of Health AIDS Reagent Program or Selleck Chemicals; Cat S2005) were added prior to virus infection at the following concentrations: efavirenz (EFV) (20 nM) and raltegravir (RAL) (25 μM). Innate and inflammatory stimuli, immunostimulatory DNA (ISD), or control DNA complexed with LyoVec (InvivoGen; Cat tlrl-isdc/tlrl-isdcc), 2′3′-cGAMP (InvivoGen; Cat tlrl-cga23 s), and R848 (InvivoGen; Cat tlrl-r848) were used as indicated in (Fig. 3. THP-1 monocytic cells were infected in the absence of Vpx at a density of 70,000 cell per 150 μl in complete RPMI 1640 medium with polybrene (2 μg/ml). B18R (R&D Systems; Cat 8185-BR-025) for IFN neutralization was used for flow cytometry analysis of CD86 by adding B18R at a concentration of 100 ng/ml after the medium refresh on day 4 and once again 24 h later.

Infected or stimulated MDDCs were washed with PBS (Corning); cell pellets were incubated for 15 min at 4°C with 2 μl of Fc block (BD Biosciences; Cat 564219) and then exposed to LIVE/DEAD Violet (Thermo Fisher Scientific; Cat L34955) in PBS for 15 min at 4°C in the dark. Cells were either simultaneously stained for surface markers (CD40 Thermo Fisher Scientific, Cat CD4004; CD80 eBioscience, Cat 15-0809-42; CD86 eBioscience, Cat 15-0869-42; HLA-DR BioLegend, Cat 307607 or Cat 307619; CD1c eBioscience, Cat 331505; and CD83 eBioscience, Cat 305307) or were then washed with PBS and fixed with 0.4% paraformaldehyde (Electron Microscopy Sciences) diluted in PBS. Cells were analyzed on an LSR II flow cytometer (BD Biosciences). For intracellular staining using anti-human ISG15 (R&D Systems; Cat: IC8044P) in MDDCs and THP-1 monocytic cells, cells were first exposed to LIVE/DEAD Violet and surface markers as described above, washed in PBS, fixed and permeabilized using a Cytofix/Cytoperm Kit (BD Biosciences; Cat 554714), blocked with 2 μl Fc block for 10 min at room temperature, and stained according to the manufacturer’s instructions. Cells were washed and resuspended in PBS with 1% BSA, and data were acquired on an LSR II flow cytometer (BD Biosciences) and analyzed using FlowJo software (FlowJo). ArC Amine Reactive Compensation Bead Kit (Thermo Fisher Scientific; Cat A10346), UltraComp eBeads Plus Compensation Beads (Thermo Fisher Scientific; Cat 01-3333-41), and GFP BrightComp eBeads (Thermo Fisher Scientific; Cat A10514) were used for compensation.

Approximately 200,000 DCs were lysed in TRIzol reagent (Thermo Fisher Scientific; Cat 15596026), and RNA was isolated according to the manufacturer’s instructions with the following modifications: two sequential chloroform extractions were performed, and GlycoBlue (Thermo Fisher Scientific; Cat AM9516) was added as a carrier prior to precipitation. cDNA was converted using Superscript IV VILO with ezDNase treatment (Thermo Fisher Scientific; Cat 11766050). qPCRs were carried out using TaqMan primer probes (Thermo Fisher Scientific) and TaqMan Fast Universal PCR Master Mix (Thermo Fisher Scientific) in a CFX96 thermocycler (Bio-Rad Laboratories) or QuantStudio3 (Thermo Fisher Scientific) in a volume of 10 μl according to the following cycling conditions: 50°C for 2 min, 95°C for 2 min, and 50 cycles each of 95°C for 3 s, to 60°C for 30 s, followed by 95°C for 5 s. For total ILF3 SYBR Green qPCR, PowerUp SYBR Green Master Mix (Thermo Fisher Scientific; Cat A25742) was used in a total volume of 10 μl with the following cycling protocol: 50°C for 2 min, 95°C for 2 min, 95°C for 15 s, 58°C for 15 s, and 72°C for 15 s, repeated for 40 cycles. Data were plotted as 2−(ΔCt) × 1000 relative to GAPDH. For experiments with a larger number of samples, Direct-Zol 96-Well RNA Isolation Kits (Zymo Research; Cat R2056) were used as per the manufacturer’s instructions, using the optional DNase step and forgoing the ezDNase step with the SuperScript IV VILO master mix.

MDDCs from three unique donors were transduced with either a control shRNA, ILF3 sh1, or ILF3 sh2 in the presence of Vpx. RNA was extracted using TRIzol on day 5. Purified RNA was labeled and hybridized to SurePrint G3 8X60K Microarrays (Agilent Technologies), and data were acquired at the Institute for Systems Biology. Probe sequences were mapped against the Ensembl transcript database (ensembl.org, GRCh37.74), and sequences that mapped to more than one gene or had more than five mismatches from the database sequence were removed for a total of 37,623 unique probes. Probe-specific logarithmically transformed expression was quantile normalized. Duplicate probe sequences were averaged. Gene-specific expression was computed by using the probe that showed the highest average expression across all samples in cases in which multiple probes mapped to a single gene for a total of 26,319 gene-specific probes. Statistical significance of the coefficients were computed with the LIMMA R package (https://bioconductor.org/packages/release/bioc/html/limma.html). The p values were adjusted for multiple hypothesis testing with the Benjamini–Hochberg method for controlling the false discovery rate (FDR). These data have been deposited in National Center for Biotechnology Information’s Gene Expression Omnibus (GEO) and are accessible through GEO Series accession number GSE159458 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE159458). Data from our previous publications (18, 22) that were reanalyzed as indicated in Supplemental Fig. 1 can be accessed through accession number GSE100374 (microarray) and GSE125918 (transposase-accessible chromatin using sequencing).

FIGURE 1.

ILF3 restrains MDDC maturation. (A) Schematic illustration for transcriptomic analyses of MDDCs transduced with shRNAs targeting ILF3. Cells from three unique donors were transduced with a control shRNA or one of two shRNAs targeting ILF3. After 4 d, mRNA was isolated and analyzed by microarray. (B) Representative Western blot of MDDC whole cell lysates prepared as in (A), 4 d after transduction with the indicated shRNAs. (C) GSEA plots for the Lindstedt DC maturation gene sets, in each case comparing the mean expression values from ILF3 sh1 and sh2 conditions to the control sh condition. Gene sets A–C (light purple) contain genes upregulated during DC maturation, and gene set D (dark purple) contains genes downregulated during DC maturation (D) Scatter plot of expression changes for genes significantly differentially expressed following transduction with at least one of two shRNAs targeting ILF3 plotted on a log2 scale (|FC| > 1.5, FDR < 0.05). (E) Flow cytometry analysis of MDDCs after ILF3 knockdown. Plots show a representative histogram of CD86, HLA-DR, and CD40 expression from one donor and corresponding graphs showing pooled data from 13 donors across four individual experiments for CD86, 4 donors from one experiment for HLA-DR, and 4 donors from one experiment for CD40, gated on forward scatter versus side scatter, singlets, and live cells. Statistics were calculated by matching each donor in a mixed-effects model using Dunnett test for multiple comparisons. Mean and SEM for each marker in a given lentiviral treatment across all represented donors indicated on each histogram used throughout the paper. (F) Flow cytometry plots of CD86 versus CD1c or HLA-DR of one representative donor, gated as in (E). *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 1.

ILF3 restrains MDDC maturation. (A) Schematic illustration for transcriptomic analyses of MDDCs transduced with shRNAs targeting ILF3. Cells from three unique donors were transduced with a control shRNA or one of two shRNAs targeting ILF3. After 4 d, mRNA was isolated and analyzed by microarray. (B) Representative Western blot of MDDC whole cell lysates prepared as in (A), 4 d after transduction with the indicated shRNAs. (C) GSEA plots for the Lindstedt DC maturation gene sets, in each case comparing the mean expression values from ILF3 sh1 and sh2 conditions to the control sh condition. Gene sets A–C (light purple) contain genes upregulated during DC maturation, and gene set D (dark purple) contains genes downregulated during DC maturation (D) Scatter plot of expression changes for genes significantly differentially expressed following transduction with at least one of two shRNAs targeting ILF3 plotted on a log2 scale (|FC| > 1.5, FDR < 0.05). (E) Flow cytometry analysis of MDDCs after ILF3 knockdown. Plots show a representative histogram of CD86, HLA-DR, and CD40 expression from one donor and corresponding graphs showing pooled data from 13 donors across four individual experiments for CD86, 4 donors from one experiment for HLA-DR, and 4 donors from one experiment for CD40, gated on forward scatter versus side scatter, singlets, and live cells. Statistics were calculated by matching each donor in a mixed-effects model using Dunnett test for multiple comparisons. Mean and SEM for each marker in a given lentiviral treatment across all represented donors indicated on each histogram used throughout the paper. (F) Flow cytometry plots of CD86 versus CD1c or HLA-DR of one representative donor, gated as in (E). *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

For ILF3 shRNA knockdown microarray, 37,913 gene features for each construct were ranked by the t test metric of gene set enrichment analysis (GSEA) using the mean values computed for ILF3 sh1 and sh2 compared with the control shRNA. The analysis was performed using the weighted enrichment statistic, ranking genes using the t test metric, against the C2 curated gene sets containing x number of genes (with 15 > x > 200) within the Molecular Signature Database. The normalized enrichment score was calculated using 1000 gene set permutations. For NF110 overexpression RNA-seq, a preranked list of fold changes (FC) of NF110 wt compared with LKO control was ranked by t test metric containing 13,125 gene features. The analysis was performed using the standard weighted enrichment statistic against the C1 hallmark gene sets with 15 > x > 200 gene membership within the Molecular Signature Database. The normalized enrichment score was calculated using 1000 permutations.

RNA were isolated from TRIzol lysates as described above and converted to cDNA libraries using the Illumina TruSeq stranded mRNA Kit per the manufacturer’s instructions. Libraries were amplified and then sequenced on an Illumina NovaSeq (2 × 150, paired end). Reads with more that 67% identical bases were discarded prior to alignment. The remaining read pairs were aligned to the human genome (hg19, GRCh37 Genome Reference Consortium human reference 37 [GCA_000001405.1]) using the gsnap aligner (v. 2016-08-24) allowing for novel splicing. Concordantly mapping read pairs (average 17 million per sample) that aligned uniquely were assigned to exons using the subRead program and gene definitions from GRCh38.87. Genes with low expression were filtered using the filterByExpr function in the edgeR package from Bioconductor.org, resulting in a total of 13,125 genes in the final dataset. Differential expression was calculated using the edgeR package. The data discussed in this publication have been deposited in National Center for Biotechnology Information’s GEO and are accessible through GEO Series accession number GSE159143 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE159143).

Samples were prepared as previously described (18). Blots were incubated with primary ILF3 Ab ILF3 (Abcam; Cat 92355) (1:3000). After incubating blots overnight at 4°C or for at least 1 h at room temp, they were washed with TBS/Tween and then incubated with the corresponding HRP-conjugated anti-rabbit secondary Ab (1:10,000) for 1 h at room temperature. To confirm equal protein loading, blots were incubated with an anti-actin–HRP Ab (directly conjugated) (Abcam; Cat AB20272) (1:100,000) for 30 min at room temperature. Alternatively, blots were incubated with primary GAPDH Ab (Cell Signaling Technology; Cat 5174T) (1:1000) to confirm equal protein loading in samples taken over the course of differentiation from monocytes to MDDCs, as actin expression is not constant during differentiation. After incubation with primary Abs, blots were washed with TBS/Tween and then incubated with the corresponding HRP-conjugated anti-rabbit secondary Ab (1:10,000) for 1 h at room temperature. Blots were then washed in TBS/Tween, reacted with West Femto ECL kit (Thermo Fisher Scientific; Cat 34095), and developed using a FluorChem E Imager (ProteinSimple).

On day 5 after transduction, 75 μl of MDDCs were plated into a chamber of an eight-chamber mounting slide (MatTek). Cells were collected onto the coverglass by centrifugation for 7 min at 1500 rpm at room temperature. The supernatant was aspirated, and cells were fixed with 2% paraformaldehyde for 30 min at room temperature. After washing the cells with PBS, the cells were then permeabilized and blocked with a solution of PBS + 0.1% Triton X-100 + 5% FBS (Perm/Block, 0.2 μm filtered) for 1 h at room temperature. The Perm/Block solution was then aspirated, and anti-ILF3 Ab was added at a concentration of 1:1000 in Perm/Block solution. Cells were washed 3× for 5 min each with Perm/Block and then incubated with Alexa Fluor 594 goat anti-rabbit secondary Ab (Thermo Fisher Scientific; Cat A-11012) at 1:750 in Perm/Block for 1 h at room temperature. Cells were washed twice for 5 min each at room temperature in the dark with Perm/Block. Then, 1.25 μl of phalloidin (Thermo Fisher Scientific; Cat: R415) was added to 200 μl of PBS per chamber and incubated for 20 min at room temperature in the dark, except for where indicated. Cells were washed twice with PBS and then mounted with ProLong antifade diamond with DAPI (Thermo Fisher Scientific; Cat P36962) and left to cure for 24 h before imaging on a DeltaVision Elite (Cytiva) widefield microscope. Images were collected with a 100× 1.4 numerical aperture objective (Olympus) on a CoolSnapHQ2 charge-coupled device camera (Photometrics). The sides of each pixel are 6.45 µm. Images were deconvolved using algorithms provided by Huygens software (Scientific Volume Imaging, Hilversum, the Netherlands). For deconvolution, three-dimensional datasets were processed to remove noise and reassign blur by an iterative classic maximum likelihood estimation widefield algorithm using an experimentally derived point spread function. Image processing was performed using Imaris (Bitplane). Cells were segmented using the Cells command, which identifies cells and their nuclei. For each cell, the ratio between nuclear and cytoplasmic ILF3 fluorescence signal intensity was determined.

IFN-β in DC supernatants was determined by ELISA (R&D Systems; Cat DIFNB0), according to the manufacturer’s instructions. Supernatants (50 μl) from MDDCs infected with HIV-1–GFP for 32 h were mixed with 50 μl assay diluent reagent per well and measured in duplicate compared with a standard curve of IFN-β ranging from 7.81 pg/ml to 500 pg/ml. IFN-β was also measured in supernatants of ILF3 knockdown MDDC supernatants using high-sensitivity ELISA IFN-β (PBL Assay Science; Cat 41435) on mock, 2′3′-cGAMP, and B18R-treated MDDCs (R&D Systems; Cat 8185-BR-025) (added at 100 ng/ml 1 h before 2′3′-cGAMP treatment) 7 h posttreatment, measured in duplicate compared with a standard curve of IFN-β ranging from 150 pg/ml to 2.34 pg/ml. CXCL10 and IL-6 were measured in the supernatants of mock or ILF3 knockdown MDDC supernatants using ELISAs for CXCL10 (Abcam; Cat 173194) and IL-6 (Abcam; Cat178013) on mock, 2′3′-cGAMP, and B18R-treated MDDCs for 7 h, measured in duplicate compared with a standard curve ranging from 800 pg/ml to 12.5 pg/ml for CXCL10 and 500 pg/ml to 7.8 pg/ml for IL-6. CCL23 was measured in the supernatants of either ILF3 knockdown or NF90/NF110-overexpressing MDDCs compared with relevant controls 48 h post–medium refresh on day 4 using a CCL23 ELISA (Abcam; Cat 216169), measured in duplicate compared with a standard curve of CCL23 ranging from 450 pg/ml to 7.03 pg/ml. All duplicates were averaged per donor, represented as the mean.

IFN activity in HL116 cells was measured as previously described (18). Briefly, 20,000 HL116 cells were incubated with supernatants from MDDC cultures for 7 h before passively lysing the cells and scoring firefly luciferase activity in the presence of luciferin.

Statistical tests were performed as indicated in the figure legends or otherwise using Prism 8.0.1 (GraphPad) to calculate a mixed model two-tailed t test using paired samples and setting an α value of 0.05. In this study, n is defined in the figure legends and represents the number of biological replicates performed of unique donors for MDDC experiments or the number of independent, nontechnical replicates for THP-1 experiments, unless otherwise indicated.

All graphical illustrations were done using BioRender.com.

To probe the DC response to innate immune stimuli, we employed HIV-1–GFP, a VSV-pseudotyped, single-cycle, HIV-derived reporter virus that lacks all accessory proteins and expresses GFP in place of Nef. Normally, HIV-1 infection is severely limited in DCs because of expression of the restriction factor SAMHD1, which prevents reverse transcription of viral RNA (23, 24). Providing the SIV accessory protein Vpx, in trans, leads to SAMHD1 degradation, allows for reverse transcription to proceed, and enables efficient, productive infection (19, 20). This system is a well-established model for examining DC maturation and type I IFN responses (18, 19, 25). Infection of MDDCs with HIV-1–GFP robustly induces type I IFN through the DNA sensing pathway cGAS–STING, a key initiator of DC maturation via the transcription factor IRF3 and antiviral immunity through the induction of ISGs.

To identify molecules that restrain maturation and IFN responses in myeloid DCs, we reanalyzed existing datasets that were generated from an assay for transposase-accessible chromatin using sequencing, together with microarray datasets from MDDCs infected with HIV-1–GFP–infected MDDCs (18, 22). We compared changes in genome-wide chromatin accessibility and the transcriptome over time to search for genes that had increased chromatin accessibility near their transcription start sites and decreased gene expression, profiles that could suggest negative transcriptional regulation (Supplemental Fig. 1A). As expected, chromatin accessibility was increased at the promoters of numerous genes with established roles in DC maturation (CD40, CIITA, CD80, and CD86) and IFN responses (IFNB1, ISG15, OASL, and IFIT1) at 24 h postinfection (Supplemental Fig. 1B) (22), and these increases were associated with altered expression of these genes at later timepoints (Supplemental Fig. 1C). We defined a set of 85 genes that had both large increases in chromatin accessibility (FC > 10, p < 0.01) and significantly reduced expression (FC < 1, FDR < 0.05) at 24 h following infection with HIV-1–GFP. Of those 85 genes, only five were transcription factors, as defined by the Transcriptional Regulatory Relationships Unraveled by Sentence-based Text mining list of human transcription factors: CDK2AP2, ERCC2, HSF1, ILF3, and KLF2. We were particularly intrigued by the transcription factor ILF3, as it has been reported to negatively regulate IFN production during influenza infection in primary human bronchial epithelial cells (26). Additionally, under different experimental conditions in HeLa and A549 cells, ILF3 has been shown to promote the IFN response upon dsRNA stimulation (27, 28). The gene encoding ILF3 produces two major isoforms, NF90 and NF110 (collectively named ILF3 for the purposes of this study), and both are known to act as transcriptional regulators (1113, 2931). We found that human CD14+ monocytes isolated from whole blood express low levels of ILF3 isoforms, but these are dramatically upregulated over the course of differentiation into immature MDDC in the presence of IL-4 and GM-CSF (Supplemental Fig. 1D). Because our data in MDDCs suggested that accessibility of the ILF3 promoter is altered during innate immune stimulation and the transition from immature to mature MDDC coincides with a decrease in ILF3 expression (Supplemental Fig. 1E, 1F), we pursued ILF3 as a candidate regulator of innate immune function in human DCs to better clarify its role.

To determine the set of genes regulated by ILF3 in unstimulated myeloid cells, we transduced MDDCs with two independent shRNAs targeting all isoforms of ILF3 and analyzed their transcriptomes by microarray (Fig. 1A). Knockdown of ILF3 in resting MDDCs (Fig. 1B) altered the expression of 106 genes (FDR < 0.05; |FC| > 1.5). GSEA revealed that this set of genes was enriched for members of curated gene sets associated with DC maturation in response to inflammatory stimuli. Upregulated genes were significantly enriched in three of the four “Lindstedt DC Maturation” gene sets (A–C) (32) (Fig. 1C, Supplemental Fig. 2A). Genes downregulated during DC maturation showed enrichment in the remaining set (set D), demonstrating the concordance of the effects. (Fig. 1D). The set of 106 genes whose expression was affected by ILF3 knockdown also contained additional genes that are not in the Lindstedt gene sets but have established associations with innate immune activation and myeloid DC maturation (CHI3L1, PPARG, TLR3, WFDC21P [Lnc-DC], CCR2, ST6GAL1, VENTX, and CCL23) (Fig. 1D) (3340). Together, these transcriptomic analyses suggested that ILF3 functions as a negative regulator of MDDC maturation and innate immune responses. To determine whether phenotypic changes to DC maturation accompanied these transcriptional changes, we measured surface expression of the costimulatory factors CD86, CD40, and the MHC class II cell surface receptor HLA-DR by flow cytometry following shRNA-mediated knockdown of ILF3. In agreement with the transcriptomic data, knockdown of ILF3 led to significantly elevated surface expression of CD86, HLA-DR, and CD40 (Figs. 1E, Supplemental Fig. 2B) while having a minimal effect on cell viability (Supplemental Fig. 2C). Upregulation of CD86 correlated with expression HLA-DR in a subset of cells and was associated with reduced expression of CD1c, which are characteristic expression patterns of mature, heterogeneous MDDC cultures (Fig. 1F) (41, 42). In agreement with these findings, we also observed upregulation of CD83 (an additional marker of MDDC maturation) following ILF3 knockdown, although the sh2 condition did not reach statistical significance (Supplemental Fig. 2D). Nearly all cells in these cultures expressed high levels of CD1c, suggesting that ILF3 knockdown does not alter differentiation of monocytes into MDDCs but rather affects MDDC maturation (Supplemental Fig. 2E, 2F).

FIGURE 2.

ILF3 restrains MDDC maturation and IFN signaling in response to HIV-1–GFP viral challenge. (A) Flow cytometry analysis of CD86 expression in ILF3 knockdown MDDCs that were either mock treated or infected with HIV-1–GFP at the indicated multiplicities of infection (MOIs), gated on forward scatter versus side scatter, singlets, and live cells. n = 6 donors quantified over two individual experiments. (B) qPCR of CCL23 and CCL23 ELISA data of mock-treated MDDCs. n = 4 donors. (C) qPCR of CIITA expression in MDDCs infected with HIV-1–GFP for 32 h. n = 3 donors. (D) MFI of ISG15 expression in MDDCs infected with HIV-1–GFP for 48 h. n = 4 donors over two experiments. (E) Major protein isoforms of ILF3 and their domains: Nuclear export signal (NES), DZF, dsRBD, RGG-repeat motif (RGG = RGG-repeat), GQSY-repeat motif (GQSY-repeat), nuclear localization signal (NLS), and NVKQ motif (NVKQ). Western blot of NF90 and NF110 expression in MDDC whole cell lysates after transduction with LKO, NF90, or NF110 overexpression vectors. (F) Flow cytometry analysis of CD86 surface expression in MDDCs overexpressing NF90 or NF110 that were infected with HIV-1–GFP for 48 h. n = 8 donors from two individual experiments. (G) qPCR of CCL23 and CCL23 ELISA data of mock-treated MDDCs. n = 4 donors. (H) Expression of CIITA, CCL22, and IFNB1 in MDDCs infected with HIV-1–GFP (MOI = 1 for CIITA, MOI = 0.5 for CCL22 and IFNB1) for 32 h. n = 8 donors over two experiments for CIITA, and n = 4 donors for CCL22 and IFNB1. (I) MFI of ISG15 expression under the conditions shown in (E). For (A–D, F–I), statistics were calculated by matching each donor in a mixed-effects model using Dunnett test for multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 2.

ILF3 restrains MDDC maturation and IFN signaling in response to HIV-1–GFP viral challenge. (A) Flow cytometry analysis of CD86 expression in ILF3 knockdown MDDCs that were either mock treated or infected with HIV-1–GFP at the indicated multiplicities of infection (MOIs), gated on forward scatter versus side scatter, singlets, and live cells. n = 6 donors quantified over two individual experiments. (B) qPCR of CCL23 and CCL23 ELISA data of mock-treated MDDCs. n = 4 donors. (C) qPCR of CIITA expression in MDDCs infected with HIV-1–GFP for 32 h. n = 3 donors. (D) MFI of ISG15 expression in MDDCs infected with HIV-1–GFP for 48 h. n = 4 donors over two experiments. (E) Major protein isoforms of ILF3 and their domains: Nuclear export signal (NES), DZF, dsRBD, RGG-repeat motif (RGG = RGG-repeat), GQSY-repeat motif (GQSY-repeat), nuclear localization signal (NLS), and NVKQ motif (NVKQ). Western blot of NF90 and NF110 expression in MDDC whole cell lysates after transduction with LKO, NF90, or NF110 overexpression vectors. (F) Flow cytometry analysis of CD86 surface expression in MDDCs overexpressing NF90 or NF110 that were infected with HIV-1–GFP for 48 h. n = 8 donors from two individual experiments. (G) qPCR of CCL23 and CCL23 ELISA data of mock-treated MDDCs. n = 4 donors. (H) Expression of CIITA, CCL22, and IFNB1 in MDDCs infected with HIV-1–GFP (MOI = 1 for CIITA, MOI = 0.5 for CCL22 and IFNB1) for 32 h. n = 8 donors over two experiments for CIITA, and n = 4 donors for CCL22 and IFNB1. (I) MFI of ISG15 expression under the conditions shown in (E). For (A–D, F–I), statistics were calculated by matching each donor in a mixed-effects model using Dunnett test for multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

As various IFNs and proinflammatory cytokines can influence DC maturation, we examined whether ILF3’s effect on maturation is solely cell intrinsic or mediated by secreted factors. We prepared immature DCs transduced with either a control shRNA or ILF3-targeting shRNAs, and after 4 d, we replaced the culture media. Twenty-four hours later, the supernatants of the cultures with control or ILF3-targeting shRNAs were exchanged and incubated for another 24 h before measuring CD86 levels by flow cytometry (Supplemental Fig. 2G). Supernatants from cultures transduced with ILF3-targeting shRNAs induced baseline maturation in control DCs to levels similar to those in DCs transduced with ILF3 shRNAs (Supplemental Fig. 2H). Supernatants from control cells applied to ILF3 knockdown cells had no additional effect on CD86 expression (Supplemental Fig. 2H). Together, these data indicate that loss of ILF3 promotes steady-state maturation of MDDCs and that secreted factors contribute to this effect.

To further decipher the molecular mechanisms involved in ILF3-mediated regulation of myeloid cell biology, we perturbed ILF3 expression using two complementary approaches in THP-1 cells, a myeloid leukemia-derived monocytic cell line. Knockdown of ILF3 with shRNAs led to elevated expression of the maturation marker CD80, the differentiation markers CD209 and SAMHD1, and the IFN-related markers CXCL10 and ISG15 (Supplemental Fig. 2I). Additionally, we generated ILF3 knockout THP-1 cells using CRISPR–Cas9 (Supplemental Fig. 2J). We found that expression of the maturation marker CD86 and the differentiation markers CD209, SAMHD1, and CD14 were all significantly elevated after ILF3 was knocked out using either of two independent guide RNA sequences 9 d posttransduction (Supplemental Fig. 2K). These data further support the role of ILF3 as a negative regulator of IFN-related genes and myeloid cell maturation. Given that THP-1 monocytic cell lines do not fully recapitulate primary cell behavior, we chose to focus on primary MDDCs for further studies.

Our transcriptomic analysis of unstimulated MDDCs indicated a role for ILF3 in regulating myeloid cell maturation, so we also examined the impact of ILF3 knockdown or overexpression on MDDC responses to infection with HIV-1–GFP. Knockdown of ILF3 resulted in significantly elevated expression of CD86 (Fig. 2A), CD80, and HLA-DR (Supplemental Fig. 3A, 3B) in HIV-1–GFP–infected cells compared with controls. In contrast to markers of mature MDDCs, expression of CCL23, a chemokine marker of immature MDDCs, was suppressed by knockdown of ILF3, as measured at the levels of mRNA and protein (Fig. 2B). Similarly, expression of CIITA, which typically decreases during DC maturation (43), was further decreased by ILF3 knockdown during HIV-1–GFP infection (Fig. 2C). In addition, knockdown of ILF3 increased expression and secretion of type I IFN (Supplemental Fig. 3C, 3D) and expression of the hallmark ISG15 in response to HIV-1–GFP infection (Fig. 2D).

FIGURE 3.

ILF3 dampens innate sensing of HIV-1 and other pathogen-associated molecular patterns. (A) Illustration of the first half of the HIV-1 lifecycle and the point of action of inhibitors that target reverse transcription (EFV) and integration (RAL). (B) Flow cytometry analysis of CD86 expression on MDDCs infected with HIV-1–GFP for 48 h that were treated or untreated with RAL. n = 6 donors from two individual experiments. (C) Illustration of ILF3 knockdown in MDDCs and stratification of stimulation with either ISD (1.5 μg/ml), 2′3′-cGAMP (1 μg/ml), or R848 (3 μg/ml). (D) Illustration of signaling pathways downstream of cGAS and TLR7/8. (E) qPCR analysis of the indicated targets in MDDCs that were mock treated or stimulated with the agonists depicted in (C). n = 4 donors. (F) Flow cytometry analysis of HLA-DR expression on MDDCs that were mock treated or stimulated with 2′3′-cGAMP (1 μg/ml) for 24 h. n = 4 donors. (G) Flow cytometry analysis of CD86 expression on MDDCs that were mock treated or stimulated with R848 (3 μg/ml). n = 6 donors from two individual experiments. For (B), (E), (F), and (G), statistics were calculated by matching each donor in a mixed-effects model using Dunnett test for multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

ILF3 dampens innate sensing of HIV-1 and other pathogen-associated molecular patterns. (A) Illustration of the first half of the HIV-1 lifecycle and the point of action of inhibitors that target reverse transcription (EFV) and integration (RAL). (B) Flow cytometry analysis of CD86 expression on MDDCs infected with HIV-1–GFP for 48 h that were treated or untreated with RAL. n = 6 donors from two individual experiments. (C) Illustration of ILF3 knockdown in MDDCs and stratification of stimulation with either ISD (1.5 μg/ml), 2′3′-cGAMP (1 μg/ml), or R848 (3 μg/ml). (D) Illustration of signaling pathways downstream of cGAS and TLR7/8. (E) qPCR analysis of the indicated targets in MDDCs that were mock treated or stimulated with the agonists depicted in (C). n = 4 donors. (F) Flow cytometry analysis of HLA-DR expression on MDDCs that were mock treated or stimulated with 2′3′-cGAMP (1 μg/ml) for 24 h. n = 4 donors. (G) Flow cytometry analysis of CD86 expression on MDDCs that were mock treated or stimulated with R848 (3 μg/ml). n = 6 donors from two individual experiments. For (B), (E), (F), and (G), statistics were calculated by matching each donor in a mixed-effects model using Dunnett test for multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To confirm that the negative regulation of DC maturation and IFN responses was specific to modulation of ILF3, we overexpressed each of its two major isoforms, NF90 and NF110 (Fig. 2E), and tested responses to HIV-1–GFP infection. We observed inverse effects compared with knockdown, as overexpression of either isoform suppressed surface expression of CD86 in resting MDDCs relative to controls (Fig. 2F). In HIV-1–GFP–infected MDDCs, both isoforms suppressed surface expression of CD86 (Fig. 2F). Similarly, overexpression of either ILF3 isoform potentiated CCL23 mRNA and secreted protein (Fig. 2G), enhanced expression of CIITA in response to HIV-1–GFP infection, and suppressed expression of CCL22, an additional chemokine marker of mature DCs (Fig. 2H). We also found that overexpression of NF90 or NF110 suppressed IFNB1 in HIV-1–GFP–infected MDDCs and trended toward suppression of ISG15, with only the NF110 isoform reaching statistical significance (Fig. 2H, 2I). Taken together, these experiments demonstrate that ILF3 negatively regulates myeloid DC maturation and innate immune responses driven by HIV-1–GFP.

Under permissive conditions, HIV-1 is detected in myeloid cells primarily through the cGAS–STING pathway (25, 44). These responses depend on reverse transcription of the incoming HIV-1 RNA, which in MDDCs is allowed to proceed by providing Vpx in trans to disable SAMHD1 (19). They are also facilitated by capsid destabilization (45), are limited by the exonuclease TREX1 (46), and can occur both before and after integration (18). Detection of HIV-1 components has also been reported to occur through TLR7/TLR8 (47, 48), the MAVS pathway (49, 50), NONO (51), and other cellular sensors (52). To determine whether ILF3 impacts sensing of HIV-1–GFP at different stages of the virus life cycle, we inhibited reverse transcription using EFV or blocked integration using RAL and tested innate responses to HIV-1–GFP in ILF3 knockdown cells (Fig. 3A). RAL treatment did not affect ILF3’s ability to restrain MDDC maturation upon HIV-1–GFP infection as measured by the percentage of live, CD86+ cells (Fig. 3B) or CD86 mean fluorescence intensity (MFI) (Supplemental Fig. 3E). We noted that blocking reverse transcription inhibited responses to HIV-1–GFP as expected (18, 19) and did not affect baseline increases in maturation in ILF3 knockdown conditions as measured by CD86 (Supplemental Fig. 3F, 3G). These data suggested that loss of ILF3 potentiates baseline maturation and innate immune signaling in MDDCs and that these responses persist during stimulation with HIV-1 when reverse transcription is allowed to proceed.

We next sought to determine whether loss of ILF3 potentiated responses to stimulation specifically through the cGAS–STING and TLR–MyD88 pathways, which are known to act through several shared downstream kinases and transcription factors (Fig. 3C). cGAS can be directly activated by transfection of ISD. Alternatively, robust cGAS activation can by phenocopied by exogenous delivery of its enzymatic product, 2′3′-cGAMP, a secondary messenger that binds to and activates the adapter protein STING. We also tested the ssRNA mimetic R848, which is detected through TLR8 and relays innate immune activation of inflammatory genes downstream of the adapter protein MyD88 (Fig. 3D). Stimulation with 2′3′-cGAMP in ILF3 knockdown MDDCs led to increased expression IFNB1, the maturation factor WFDC21P, the ISGs CXCL10, and ISG15 and the proinflammatory cytokine IL6, and decreased expression of CIITA compared with controls (Fig. 3E). Stimulation with ISD similarly potentiated expression of IFNB1, CXCL10, ISG15, and IL6. This result supports a role for ILF3 as a negative regulator of DC responses to cGAS–STING stimuli. Following stimulation with R848, knockdown of ILF3 increased expression of IL1B and IL6, proinflammatory cytokines downstream of MyD88 and NF-κB, and ISG15 and WFDC21P while also decreasing CIITA (Fig. 3E). Knockdown of ILF3 also significantly increased surface expression of HLA-DR and CD86 in response to 2′3′-cGAMP and R848 stimulation, respectively, compared with controls (Fig. 3F, 3G). In THP-1 cells lacking ILF3, expression of CD80, IFNB1, CXCL10, and ISG15 were also significantly increased at baseline and when stimulated with 2′3′-cGAMP (Supplemental Fig. 3H). Knockdown of ILF3 in MDDCs also led to elevated expression of CCL22 and a reciprocal suppression of CCL23, both at baseline and following 2′3′-cGAMP stimulation (Supplemental Fig. 3I). These data indicate that ILF3 dampens responses through both STING and MyD88 pathways, suggesting that ILF3 might function broadly as a negative regulator of innate responses.

To verify that these changes in expression of IFN and maturation-related genes in ILF3 knockdown MDDCs correspond to increases in protein expression, we measured secretion of IFN-β, CXCL10, and IL-6 under mock or 2′3′-cGAMP conditions (Supplemental Fig. 3J). Knockdown of ILF3 elaborated secretion of all three cytokines following stimulation with 2′3′-cGAMP and a trend toward increased secretion at baseline. Because we observed significant potentiation of IFN-β upon ILF3 knockdown, we sought to neutralize IFN produced in culture using the vaccinia virus IFNR decoy protein B18R to evaluate IFN’s contribution to ILF3-dependent gene expression. Interestingly, treatment with B18R only partially suppressed CD86 induction in unstimulated ILF3 knockdown MDDCs, which suggested that IFNs might contribute to but are not solely responsible for MDDC maturation under these conditions (Supplemental Fig. 3K). Neutralization of IFN was effective, as B18R treatment efficiently blocked induction of ISG15, MX1, and STAT1 and reduced expression of CXCL10 and OASL in cells stimulated with 2′3′-cGAMP (Supplemental Fig. 3L). In ILF3 knockdown conditions, B18R treatment reduced expression of these ISGs, and yet, similar to observations for CD86, differences between control and ILF3 shRNA conditions persisted, particularly for STAT1, CXCL10, and OASL. Because B18R did not completely block elevated expression of ISGs and CD86 in ILF3 knockdown conditions, we can interpret these results in one of two ways: 1) B18R neutralization was insufficient to block all IFN-driven signaling, or 2) IFN-dependent and IFN-independent signaling pathways contribute to the ILF3 phenotype. The latter interpretation should be strongly considered, given that induction of ISGs and MDDC maturation can occur through IFN-independent signals (22, 53). This would suggest that ILF3 negatively regulates innate immune responses and MDDC maturation through IFN-dependent and -independent mechanisms.

Having established a role for NF90 and NF110 in restraining MDDC maturation and innate responses to various stimuli, we sought to determine the specific domains of each protein required for their function. Both NF90 and NF110 forms of ILF3 have been reported to exist predominantly in the nucleus (54, 55). To test whether nuclear localization is required for the suppression of DC maturation and type I IFN responses, we overexpressed versions of NF90 and NF110 lacking the bipartite nuclear localization sequence (aa 370–394: KRPMEEDGEEKSPSKKKKKIQKKE) (NF90/NF110ΔNLS) in MDDCs (Fig. 4A) and confirmed via immunofluorescence microscopy that NF90/NF110ΔNLS mutants were localized to the cytosol (Fig. 4B), as indicated by the shift in the ratio of total nuclear to cytosolic ILF3 (Fig. 4C). We also confirmed overexpression of wt and ΔNLS forms of NF90 and NF110 by qPCR (Fig. 4D). In MDDCs challenged with HIV-1–GFP, full-length NF90 and NF110 were able to suppress CD86, WFDC21P, and ISG15 induction, and deletion of the NLS from either isoform prevented suppression of these markers (Fig. 4D). By flow cytometry, we saw concordant effects on protein expression of CD86 and ISG15 (Fig. 4E, 4F). NF90 and NF110 have been previously found to translocate to the cytosol during influenza infection of epithelial cells (56). To determine whether HIV infection could influence ILF3 localization, we compared the nuclear versus cytosolic ratios of endogenous NF90 and NF110 in uninfected or HIV-1–GFP+ MDDCs, and we found no significant shift in their localization (Fig. 4G). These experiments suggest that nuclear localization is essential to the function of both of ILF3’s isoforms in regulating myeloid maturation and IFN responses, and in this context, infection with HIV-1–GFP does not impact ILF3 localization.

FIGURE 4.

Nuclear localization of NF90 and N110 are required for the suppression of innate responses to HIV-1–GFP. (A) Illustration of the bipartite nuclear localization sequence in NF90 and NF110. (B) Immunofluorescence of mock-treated MDDCs that were transduced with the indicated constructs and stained for actin (magenta), ILF3 (yellow, staining endogenous and overexpression constructs), and DNA (cyan, DAPI). (C) Quantification of three-dimensional immunofluorescence microscopy of total ILF3 Ab staining represented as the intensity of signal over voxel space, presented as a ratio between nuclear and cytosolic compartment. Data are from one representative donor using 50 cells per condition. Statistics were calculated using one-way ANOVA with Sidak test for multiple comparisons. (D) qPCR quantification of NF90 and NF110 expression using SYBR Green probes to detect endogenous and overexpressed isoforms of ILF3. (E) Flow cytometry analysis of CD86 expression in MDDCs that were transduced with the indicated overexpression constructs and either mock treated or infected with HIV-1–GFP for 48 h. n = 4 donors. (F) MFI quantification of ISG15 expression. n = 4 donors. Donor 2 was excluded from the HIV-1–GFP condition for aberrant activation. For (D), (E), and (F), statistics were calculated using a paired mixed-effects model with Dunnett test for multiple comparisons. (G) Representative three-dimensional immunofluorescence microscopy quantification of total NF90 and NF110 from one donor, demonstrating the intensity of signal over voxel space as a ratio of nuclear versus cytosolic space. 75 cells were quantified from each condition: mock-infected or HIV-1–GFP+ infected MDDCs at 27-h postinfection (selecting GFP+ cells). Image channels depict total ILF3 (yellow), DNA (cyan, DAPI), and GFP. An unpaired, two-tailed t test was used for statistical analysis. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 4.

Nuclear localization of NF90 and N110 are required for the suppression of innate responses to HIV-1–GFP. (A) Illustration of the bipartite nuclear localization sequence in NF90 and NF110. (B) Immunofluorescence of mock-treated MDDCs that were transduced with the indicated constructs and stained for actin (magenta), ILF3 (yellow, staining endogenous and overexpression constructs), and DNA (cyan, DAPI). (C) Quantification of three-dimensional immunofluorescence microscopy of total ILF3 Ab staining represented as the intensity of signal over voxel space, presented as a ratio between nuclear and cytosolic compartment. Data are from one representative donor using 50 cells per condition. Statistics were calculated using one-way ANOVA with Sidak test for multiple comparisons. (D) qPCR quantification of NF90 and NF110 expression using SYBR Green probes to detect endogenous and overexpressed isoforms of ILF3. (E) Flow cytometry analysis of CD86 expression in MDDCs that were transduced with the indicated overexpression constructs and either mock treated or infected with HIV-1–GFP for 48 h. n = 4 donors. (F) MFI quantification of ISG15 expression. n = 4 donors. Donor 2 was excluded from the HIV-1–GFP condition for aberrant activation. For (D), (E), and (F), statistics were calculated using a paired mixed-effects model with Dunnett test for multiple comparisons. (G) Representative three-dimensional immunofluorescence microscopy quantification of total NF90 and NF110 from one donor, demonstrating the intensity of signal over voxel space as a ratio of nuclear versus cytosolic space. 75 cells were quantified from each condition: mock-infected or HIV-1–GFP+ infected MDDCs at 27-h postinfection (selecting GFP+ cells). Image channels depict total ILF3 (yellow), DNA (cyan, DAPI), and GFP. An unpaired, two-tailed t test was used for statistical analysis. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To determine the domains of NF90 and NF110 that are required for regulation of DC responses, we designed constructs that lacked the dual dsRBDs (ΔdsRBD1: aa 402–465, ΔdsRBD2: aa 531–576) or lacked the DZF domain (ΔDZF: aa 89–342) (Fig. 5A). These constructs were robustly overexpressed at the mRNA level in MDDCs (Fig. 5B) and were translated to significantly higher levels than the corresponding native proteins (Fig. 5C). Deletion of the dsRBDs in either NF90 or NF110 had no effect on levels of the maturation transcripts CD86, CD80, or CCL23 or surface expression of CD86 following HIV-1–GFP infection (Fig. 5D, 5E). Surprisingly, deletion of the DZF domain eliminated the ability of NF110 to suppress MDDC maturation, whereas deletion of the same domain in NF90 had no effect (Fig. 5D, 5E). We also quantified IFN-β secretion during HIV-1–GFP infection in MDDCs that overexpressed wt or mutant ILF3 constructs. Although none of these experimental conditions reached statistical significance, all ILF3 constructs except for the NF110 DZF deletion mutant trended toward IFN-β suppression (Fig. 5F).

FIGURE 5.

The DZF domain of NF110 suppresses MDDC maturation. (A) Illustration of wt NF90 and NF110 constructs and the corresponding domain mutants with deletions in either the dsRBD or the DZF domains. (B) Representative qPCR quantification of both endogenous and overexpressed ILF3. n = 4 donors. (C) Representative Western blot of MDDC whole cell lysates depicting overexpression of NF90 and NF110 constructs. Red arrowheads indicate overexpressed mutant constructs. (D) qPCR quantification of CD86, CD80, and CCL23 expression in MDDCs that were transduced with the indicated overexpression constructs and either mock treated or infected with HIV-1–GFP (multiplicity of infection [MOI] = 0.5) for 32 h. n = 8 donors from two independent experiments. (E) Flow cytometry quantification of CD86 expression in MDDCs infected with HIV-1–GFP (MOI = 0.5) for 48 h. n = 8 donors from two independent experiments. (F) ELISA of IFN-β in supernatants from MDDCs infected with HIV-1–GFP (MOI = 0.5) for 32 h. n = 4 donors. Shapes represent individual donors. For (B), (D), and (E), statistics were calculated using a paired mixed-effects model with Dunnett test for multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 5.

The DZF domain of NF110 suppresses MDDC maturation. (A) Illustration of wt NF90 and NF110 constructs and the corresponding domain mutants with deletions in either the dsRBD or the DZF domains. (B) Representative qPCR quantification of both endogenous and overexpressed ILF3. n = 4 donors. (C) Representative Western blot of MDDC whole cell lysates depicting overexpression of NF90 and NF110 constructs. Red arrowheads indicate overexpressed mutant constructs. (D) qPCR quantification of CD86, CD80, and CCL23 expression in MDDCs that were transduced with the indicated overexpression constructs and either mock treated or infected with HIV-1–GFP (multiplicity of infection [MOI] = 0.5) for 32 h. n = 8 donors from two independent experiments. (E) Flow cytometry quantification of CD86 expression in MDDCs infected with HIV-1–GFP (MOI = 0.5) for 48 h. n = 8 donors from two independent experiments. (F) ELISA of IFN-β in supernatants from MDDCs infected with HIV-1–GFP (MOI = 0.5) for 32 h. n = 4 donors. Shapes represent individual donors. For (B), (D), and (E), statistics were calculated using a paired mixed-effects model with Dunnett test for multiple comparisons. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To more comprehensively define the molecular pathways regulated by ILF3 in DCs, we performed RNA-seq analysis of mock MDDCs transduced with the LKO control vector, a vector expressing full-length NF110, or the NF110 DZF-deletion mutant (ΔDZF). The 355 genes were significantly differentially expressed in cells overexpressing wt NF110 compared with the LKO control [|log2(NF110/LKO)| > log2(1.5), FDR < 0.05, for genes with an average log2(cpm) > 1]. For 97 of these genes, more than 50% of the expression change induced by overexpression of NF110 was retained when NF110 ΔDZF was overexpressed. We termed these genes “DZF independent” with respect to the ILF3-mediated effect (Supplemental Fig. 4A). Many of the most strongly downregulated, DZF-independent genes were found to be noncoding RNAs (Fig. 6A), consistent with ILF3’s known role as a binder and modulator of long noncoding RNAs (5762). Interestingly, genes that were upregulated by NF110 overexpression in a DZF-independent manner were generally protein coding (Supplemental Fig. 4A). Of note, NF110 overexpression led to upregulation of AQP7, CD163L1, CD14, C3AR1, CD300a, ABCA9, MAFB, FMN1, STEAP4, CD163, IL10, and CFH (Fig. 6B), which are all associated with immature or tolerogenic myeloid/DC phenotypes (6373). Conversely, DCSTAMP, CHI3L1, CH25H, ALDH1A2, BHLHE41, and CCL22, which are characteristic genes expressed in mature DCs (33, 7478), were downregulated in a DZF-dependent manner (Fig. 6B). Together, these data demonstrate that NF110 restrains maturation in uninfected DCs and that this activity is dependent on the DZF domain.

FIGURE 6.

RNA-seq analysis of NF110 overexpression reveals DZF-dependent genes associated with DC maturation and metabolic pathways. (A) Plot of 355 differentially expressed genes on log2 scale, averaged across n = 4 donors, with a |FC| > 1.5, average Log2 cpm > 1, and FDR < 0.05 of NF110 and NF110 ΔDZF constructs under unstimulated conditions. Blue points indicate noncoding RNAs downregulated by NF110 and NF110 ΔDZF. Red point indicates overexpression of NF110 (ILF3), as expected. (B) Genes represented in (A), |FC| > 2, ordered by magnitude of FC by group. (C) GSEA plot of top-ranking cholesterol homeostasis hallmark gene set from analysis of MDDCs overexpressing NF110 compared with the LKO control. (D) Log2 FC of core enrichment genes (FDR < 0.05) from (C), preranked by t test. (E) Average log2 FC for genes from the Lindstedt DC maturation gene sets (groups A–C) in MDDCs overexpressing NF110 (closed circles) or NF110 ΔDZF (open circles) compared with the LKO control. Highlighted genes represent genes with a p value <0.05.

FIGURE 6.

RNA-seq analysis of NF110 overexpression reveals DZF-dependent genes associated with DC maturation and metabolic pathways. (A) Plot of 355 differentially expressed genes on log2 scale, averaged across n = 4 donors, with a |FC| > 1.5, average Log2 cpm > 1, and FDR < 0.05 of NF110 and NF110 ΔDZF constructs under unstimulated conditions. Blue points indicate noncoding RNAs downregulated by NF110 and NF110 ΔDZF. Red point indicates overexpression of NF110 (ILF3), as expected. (B) Genes represented in (A), |FC| > 2, ordered by magnitude of FC by group. (C) GSEA plot of top-ranking cholesterol homeostasis hallmark gene set from analysis of MDDCs overexpressing NF110 compared with the LKO control. (D) Log2 FC of core enrichment genes (FDR < 0.05) from (C), preranked by t test. (E) Average log2 FC for genes from the Lindstedt DC maturation gene sets (groups A–C) in MDDCs overexpressing NF110 (closed circles) or NF110 ΔDZF (open circles) compared with the LKO control. Highlighted genes represent genes with a p value <0.05.

Close modal

We also examined the set of genes that were differentially expressed following overexpression of NF110 wt compared with the LKO control by GSEA (79). Of note, the expression of genes belonging to the cholesterol homeostasis (32/74) and oxidative phosphorylation (111/200) gene sets was suppressed under conditions of NF110 overexpression. These were the only statistically significantly enriched gene sets out of all tested hallmark gene sets in the Molecular Signatures Database (80) containing between 15 and 200 genes (Fig. 6C, Supplemental Fig. 4B). Inspection of individual genes belonging to the oxidative phosphorylation gene set revealed they had subtle changes in magnitude compared with the control (Supplemental Fig. 4C). Genes in the cholesterol homeostasis gene set were suppressed to a greater degree by NF110 and included PPARG, ATF3, LGALS3, LPL, and TNFRSF21 (FDR < 0.05) (Fig. 6D). The top DZF-dependent gene in the cholesterol homeostasis group, PPARG (the gene that encodes PPARγ), was also significantly upregulated in ILF3 knockout THP-1 populations (Supplemental Fig. 4D).

To determine if our RNA-seq data from NF110 overexpression reflected known biological perturbations, we compared our set of NF110-regulated DZF-dependent genes with public datasets using EnrichR (81). We found significant overlap with genes affected by the PPARγ agonists rosiglitazone and atorvastatin in MDDCs and monocytes, respectively (Supplemental Fig. 4E). This overlap is noteworthy, considering that PPARγ is known to control oxidation of fatty acids and regulate NF-κB–mediated proinflammatory responses (82). Taken together, these data suggest that ILF3 shapes the transcriptome of myeloid DCs through pathways that may intersect, at least in part, with PPARγ and lipid metabolism.

We also observed that many genes in the Lindstedt DC maturation gene sets were moderately suppressed by NF110 overexpression (Fig. 6E), although these gene sets did not exhibit high GSEA scores when analyzed in full. This is likely due to the fact that we tested MDDCs in an immature state, with expression of these genes already being low, thereby making further gene suppression difficult to detect. Nevertheless, the majority of genes (75%) in gene sets A–C (elevated in mature DCs in (Fig. 1) were downregulated by overexpression of NF110, predominantly in a DZF-dependent manner (Fig. 6E). Apart from the significantly downregulated genes by NF110, the majority of genes downregulated by the wt form of NF110 had consistently lower p values than those downregulated by NF110 ΔDZF. These data are in agreement with what we observed in ILF3 knockdown experiments (as expression of these genes associated with DC maturation was increased) and support a specific role for NF110 in negatively regulating myeloid DC maturation.

Delineating the mechanisms that regulate DC maturation and innate responses is critical to understanding antiviral immunity and can inform the design of new therapeutic agents to modulate inflammation. In this study, we have identified that the transcription factor ILF3 acts as a negative regulator of DC maturation and innate immune responses. We discovered that CD14+ monocytes express minimal levels of ILF3 and then increase its expression upon derivation into immature DCs. As DCs are exquisitely sensitive to IFN signaling compared with circulating monocytes (22), it is plausible that DCs change expression levels of ILF3 to deploy or withdraw an additional regulator of innate immune responses. Although several previous studies have examined the role of ILF3 in regulating innate immune responses during stimulation, they have not reached a consensus on the role of ILF3, as these functions are likely context dependent. Our findings are consistent with the only other study of ILF3 in primary human cells during virus infection, which demonstrated that knockdown of ILF3 in primary human bronchial epithelial cells led to an increase in IFN-β production in response to infection with influenza virus (26). However, this effect is not observed in all experimental systems because small interfering RNA knockdown of ILF3 in HeLa and mouse embryonic fibroblast cells did not affect levels of type I IFNs following stimulation with the dsRNA-mimetic polyinosinic:polycytidylic acid (83). Other examples suggest that ILF3 may positively regulate IFN responses under certain conditions. In one case, A549 cells infected with Sendai virus exhibited decreased IFN upon small interfering RNA knockdown of ILF3 (28). A more recent study in HeLa cells found that NF110 enhanced translation of IFNB1 mRNA and a subset of ISGs (27). Given that the mechanisms regulating innate immune sensing and IFN production are known to be different between cell lines of immune and nonimmune origin (84), it is not particularly surprising that ILF3’s function is context dependent. Our data, based on knockdown, knockout, and overexpression studies, demonstrate that ILF3 acts to restrain IFN and downstream ISGs in MDDCs and myeloid cell lines and is a significant contributing factor in regulating the DC maturation state.

More broadly, the role of ILF3 in regulating innate immune responses has been evaluated primarily during RNA virus infection (26, 28). Previous studies have also established that NF90 and NF110 isoforms inhibit replication of RNA viruses by physical association with the ISG PKR via their dsRBDs to block viral translation (85, 86). Until now, it has not been established whether ILF3 also has a role in regulating DNA sensing pathways. In this study, we have used HIV-1–GFP to model retrovirus infection and innate immune stimulation through the cGAS–STING pathway and have used antiretroviral drugs to separate stages of the virus life cycle. We found that in ILF3 knockdown conditions, the elevated innate immune responses and DC maturation phenotype persisted during infection with HIV-1–GFP. Responses to HIV-1–GFP were dependent on reverse transcription, as previously described (18, 19, 25, 44). Our observation that the ILF3 phenotype was sustained during stimulation with ISD, 2′3′-cGAMP, or R848 suggest that ILF3 acts broadly to negatively regulate pathways in the innate immune response. Placed in context with earlier publications, our data emphasize the varied roles of ILF3, not only across different cell types but across different pathogen sensing pathways.

ILF3 has been described to affect several aspects of HIV-1 infection. Many of these studies focused on a variant of ILF3, NF90ctv, that contains a two base pair CT insertion that results in a frameshift and translation of a highly acidic C terminus (87). This variant does not appear to be expressed in the human transcriptome (29, 87) but has been implicated in the positive regulation of ISGs (87), interactions with HIV-1 Rev and the Rev-responsive element in HIV-1 RNA (88), and binding to HIV-1 TAR RNA (89). In light of our results demonstrating that loss of ILF3 is associated with heightened innate responses and DC maturation, we speculate that the NF90ctv variant may behave as a dominant negative to influence ILF3-dependent gene transcription. Interestingly, NF90 has been shown to enhance HIV gene expression through cyclin T1 regulation (90). Testing these roles for ILF3 and ILF3 variants is beyond the scope of our study, but future work will likely uncover whether these effects dovetail with our findings that ILF3 modulates myeloid cell activation.

Another unresolved question that has emerged from our studies is why the DZF domain in NF110, but not in NF90, is required for the suppression of DC maturation. Of the two major isoforms of ILF3, NF110 has been shown to be more effective than NF90 at stimulating transcription through a proliferating cell nuclear Ag promoter in a transient reporter assay, whereas NF90 appears to have a greater capacity to bind RNA (29, 56). We note that the dsRBDs have been found to be dispensable for transcriptional coregulatory activity in other experimental systems (30). One hypothesis for why NF110 is a more effective transcriptional regulator is that its GQSY repeat–containing C terminus sterically interferes with the dsRBD, thereby reducing its activity relative to the DZF domain, which is required for interactions with DNA and for protein–protein interactions with binding partners like NF45 or NF90 (56), and increasing the DZF domain’s relative importance in transcriptional regulation. Thus, it is possible that the DZF domain in NF110 plays a key role in regulating transcription in MDDCs, likely through interactions with its binding partners that are not impacted by mutation of the DZF domain in NF90. We speculate that ILF3 transcriptional phenotypes may arise from disruptions in the balance of NF90 and NF110 interactions with other protein partners because of the following observations: 1) overexpression of either full-length NF90 or an NF90 mutant lacking the DZF domain suppressed MDDC maturation, and 2) NF90 and NF110 are known to form large heterodimeric complexes (91, 92), the stoichiometry of which will be impacted by isoform expression and availability. Future studies are required to disentangle the functional role of the DZF domain of each isoform.

The largest effects resulting from manipulation of ILF3 levels in uninfected MDDCs occurred in gene expression pathways related to cholesterol homeostasis. Although the oxidative phosphorylation gene set members were affected, DCs have been found to exhibit unique plasticity with respect to their ability to generate ATP from either glycolysis or oxidative phosphorylation during stimulation (93). GSEA analysis of MDDCs overexpressing NF110 identified the cholesterol homeostasis pathway as the most significantly enriched hallmark gene set (FDR < 0.05). PPARG, ATF3, LGALS3, LPL, TNFRSF12A, and other genes in this pathway were suppressed by NF110. Expression of PPARG was also significantly elevated in unstimulated THP-1 ILF3 knockout cells that displayed elevated maturation and IFN responses at baseline. Decreased expression of genes such as PPARG and CH25H would be expected to result in a buildup of cholesterol through suppressed efflux and conversion of lipid products, which could impact IFN signaling, inflammatory responses, and myeloid cell maturation as others have shown (94, 95).

PPARγ is known to heterodimerize with retinoid X receptor family members (RXR), and the resulting transcriptional complex has an important function in regulating energy balance, including roles in triglyceride metabolism, fatty acid processing and storage, and glucose homeostasis (96). RXRs can partner with retinoic acid receptor family members (RARs) as well as PPARs (97). We recently reported that RAR α (RARA) functions as negative regulator of DC maturation (22) and we speculate that this phenotype intersects with our observations reported in this study regarding ILF3. Along these lines, activation of PPARγ has been found to result in retinoid synthesis in DCs (98), and RAR/RXR signaling can be activated through certain retinoids in DCs to suppress maturation (99, 100). Additionally, natural agonists of PPARγ (such as 15d-PGJ2) and synthetic agonists (such as troglitazone and ciglitizone) have been shown to inhibit NF-κB and mitogen-activated protein kinase inflammatory pathways, resulting in the decreased surface expression of DC maturation markers (101, 102). PPARγ has been shown in murine DCs to be important for sustained expression of Aldh1a2, which promotes tolerogenic CD103+ DCs and for suppressing the Th17-skewing cytokines IL-6 and IL-23p19 in all CD11c+ DCs (103). Given that PPARγ was recently found to interact with both ILF3 and RXRA (RARA’s heterodimeric partner) (104), it is likely that ILF3 isoforms can also impact PPARγ/RXR and RARA/RXR transcriptional control of lipid metabolism by altering these heteromeric transcriptional complexes and consequently influence innate immune responses and DC maturation. Additional studies are required to determine whether there is a direct link between ILF3, PPARγ, and RARA as a transcriptional coregulatory complex in the context of lipid metabolism, myeloid cell inflammatory responses, and DC maturation.

Importantly, there is evidence that ILF3 could have a critical role in inflammatory pathophysiology in vivo. Small nucleotide polymorphisms within the ILF3 locus are correlated with more frequent cardiac events in individuals with high and low high-density lipoprotein cholesterol profiles (105). These findings are congruent with a role for ILF3 in regulating cholesterol/lipid metabolism and inflammation. A separate small nucleotide polymorphism within the ILF3 locus is associated with increased susceptibility to rheumatoid arthritis (106), further reinforcing the principal findings from our study, which indicate that ILF3 perturbations are linked with spontaneous induction of inflammatory gene expression. Taken together, our data identify ILF3 as an important negative regulator of inflammation and myeloid cell maturation, which has broad implications for how innate immune responses are governed during viral infection and inflammatory disease.

We thank Pamela Troisch at the Institute for Systems Biology for processing samples for microarray and thank members of the Aderem laboratory and the Aitchison laboratory for helpful discussions and for critically reading the manuscript.

This project was supported by National Institute of Allergy and Infectious Diseases T32 Immunology Training Grant AI106677 to R.N., P50 AI150464 to J.S.J., R01 AI032972 to A.A., and U19 AI100627 to A.A.

The sequences presented in this article have been submitted to ImmPort (https://www.immport.org) under accession number SDY1722.

The online version of this article contains supplemental material.

Abbreviations used in this article

Cat

catalog

DC

dendritic cell

dsRBD

dsRNA-binding domain

DZF

domain associated with zinc finger

EFV

efavirenz

FC

fold change

FDR

false discovery rate

GEO

Gene Expression Omnibus

GSEA

gene set enrichment analysis

ILF3

IL enhancer binding factor 3

ISD

immunostimulatory DNA

ISG

IFN-stimulated gene

MDDC

monocyte-derived DC

MFI

mean fluorescence intensity

pen/strep

50 U/ml penicillin and 50 μg/ml streptomycin

qPCR

quantitative PCR

RAL

raltegravir

RAR

retinoic acid receptor family member

RARA

RAR α

RNA-seq

RNA-sequencing

RXR

retinoid X receptor family member

sgRNA

single guide RNA

sh1

shRNA1

sh2

shRNA2

shRNA

short hairpin RNA

wt

wild type

VSV

vesicular stomatitis virus

1.
Banchereau
J.
,
R. M.
Steinman
.
1998
.
Dendritic cells and the control of immunity.
Nature.
392
:
245
252
.
2.
Iwasaki
A.
,
R.
Medzhitov
.
2015
.
Control of adaptive immunity by the innate immune system.
Nat. Immunol.
16
:
343
353
.
3.
Banchereau
J.
,
V.
Pascual
.
2006
.
Type I interferon in systemic lupus erythematosus and other autoimmune diseases.
Immunity.
25
:
383
392
.
4.
Sandler
N. G.
,
S. E.
Bosinger
,
J. D.
Estes
,
R. T.
Zhu
,
G. K.
Tharp
,
E.
Boritz
,
D.
Levin
,
S.
Wijeyesinghe
,
K. N.
Makamdop
,
G. Q.
del Prete
, et al
2014
.
Type I interferon responses in rhesus macaques prevent SIV infection and slow disease progression.
Nature.
511
:
601
605
.
5.
Fernandez
S.
,
S.
Tanaskovic
,
K.
Helbig
,
R.
Rajasuriar
,
M.
Kramski
,
J. M.
Murray
,
M.
Beard
,
D.
Purcell
,
S. R.
Lewin
,
P.
Price
,
M. A.
French
.
2011
.
CD4+ T-cell deficiency in HIV patients responding to antiretroviral therapy is associated with increased expression of interferon-stimulated genes in CD4+ T cells.
J. Infect. Dis.
204
:
1927
1935
.
6.
Belz
G. T.
,
S. L.
Nutt
.
2012
.
Transcriptional programming of the dendritic cell network. [Published erraturm appears in 2013 Nat. Rev. Immunol. 13: 149.]
Nat. Rev. Immunol.
12
:
101
113
.
7.
Shi
L.
,
W. R.
Godfrey
,
J.
Lin
,
G.
Zhao
,
P. N.
Kao
.
2007
.
NF90 regulates inducible IL-2 gene expression in T cells.
J. Exp. Med.
204
:
971
977
.
8.
Shi
L.
,
D.
Qiu
,
G.
Zhao
,
B.
Corthesy
,
S.
Lees-Miller
,
W. H.
Reeves
,
P. N.
Kao
.
2007
.
Dynamic binding of Ku80, Ku70 and NF90 to the IL-2 promoter in vivo in activated T-cells.
Nucleic Acids Res.
35
:
2302
2310
.
9.
Shim
J.
,
H.
Lim
,
J. R.
Yates
,
M.
Karin
.
2002
.
Nuclear export of NF90 is required for interleukin-2 mRNA stabilization.
Mol. Cell
10
:
1331
1344
.
10.
Corthésy
B.
,
P. N.
Kao
.
1994
.
Purification by DNA affinity chromatography of two polypeptides that contact the NF-AT DNA binding site in the interleukin 2 promoter.
J. Biol. Chem.
269
:
20682
20690
.
11.
Kao
P. N.
,
L.
Chen
,
G.
Brock
,
J.
Ng
,
J.
Kenny
,
A. J.
Smith
,
B.
Corthésy
.
1994
.
Cloning and expression of cyclosporin A- and FK506-sensitive nuclear factor of activated T-cells: NF45 and NF90.
J. Biol. Chem.
269
:
20691
20699
.
12.
Kiesler
P.
,
P. A.
Haynes
,
L.
Shi
,
P. N.
Kao
,
V. H.
Wysocki
,
D.
Vercelli
.
2010
.
NF45 and NF90 regulate HS4-dependent interleukin-13 transcription in T cells.
J. Biol. Chem.
285
:
8256
8267
.
13.
Hu
Q.
,
Y. Y.
Lu
,
H.
Noh
,
S.
Hong
,
Z.
Dong
,
H. F.
Ding
,
S. B.
Su
,
S.
Huang
.
2013
.
Interleukin enhancer-binding factor 3 promotes breast tumor progression by regulating sustained urokinase-type plasminogen activator expression.
Oncogene.
32
:
3933
3943
.
14.
Gomila
R. C.
,
G. W.
Martin
,
L.
Gehrke
.
2011
.
NF90 binds the dengue virus RNA 3′ terminus and is a positive regulator of dengue virus replication.
PLoS One.
6
:
e16687
.
15.
Li
Y.
,
T.
Masaki
,
T.
Shimakami
,
S. M.
Lemon
.
2014
.
hnRNP L and NF90 interact with hepatitis C virus 5′-terminal untranslated RNA and promote efficient replication.
J. Virol.
88
:
7199
7209
.
16.
Isken
O.
,
C. W.
Grassmann
,
R. T.
Sarisky
,
M.
Kann
,
S.
Zhang
,
F.
Grosse
,
P. N.
Kao
,
S. E.
Behrens
.
2003
.
Members of the NF90/NFAR protein group are involved in the life cycle of a positive-strand RNA virus.
EMBO J.
22
:
5655
5665
.
17.
Shabman
R. S.
,
D. W.
Leung
,
J.
Johnson
,
N.
Glennon
,
E. E.
Gulcicek
,
K. L.
Stone
,
L.
Leung
,
L.
Hensley
,
G. K.
Amarasinghe
,
C. F.
Basler
.
2011
.
DRBP76 associates with Ebola virus VP35 and suppresses viral polymerase function.
J. Infect. Dis.
204
(
Suppl 3
):
S911
S918
.
18.
Johnson
J. S.
,
S. Y.
Lucas
,
L. M.
Amon
,
S.
Skelton
,
R.
Nazitto
,
S.
Carbonetti
,
D. N.
Sather
,
D. R.
Littman
,
A.
Aderem
.
2018
.
Reshaping of the dendritic cell chromatin landscape and interferon pathways during HIV infection.
Cell Host Microbe.
23
:
366
381.e9
.
19.
Manel
N.
,
B.
Hogstad
,
Y.
Wang
,
D. E.
Levy
,
D.
Unutmaz
,
D. R.
Littman
.
2010
.
A cryptic sensor for HIV-1 activates antiviral innate immunity in dendritic cells. [Published erratum appears in 2011 Nature. 470: 424.]
Nature.
467
:
214
217
.
20.
Mangeot
P. E.
,
K.
Duperrier
,
D.
Nègre
,
B.
Boson
,
D.
Rigal
,
F. L.
Cosset
,
J. L.
Darlix
.
2002
.
High levels of transduction of human dendritic cells with optimized SIV vectors.
Mol. Ther.
5
:
283
290
.
21.
Sanjana
N. E.
,
O.
Shalem
,
F.
Zhang
.
2014
.
Improved vectors and genome-wide libraries for CRISPR screening.
Nat. Methods
11
:
783
784
.
22.
Johnson
J. S.
,
N.
De Veaux
,
A. W.
Rives
,
X.
Lahaye
,
S. Y.
Lucas
,
B. P.
Perot
,
M.
Luka
,
V.
Garcia-Paredes
,
L. M.
Amon
,
A.
Watters
, et al
2020
.
A comprehensive map of the monocyte-derived dendritic cell transcriptional network engaged upon innate sensing of HIV.
Cell Rep.
30
:
914
931.e9
.
23.
Hrecka
K.
,
C.
Hao
,
M.
Gierszewska
,
S. K.
Swanson
,
M.
Kesik-Brodacka
,
S.
Srivastava
,
L.
Florens
,
M. P.
Washburn
,
J.
Skowronski
.
2011
.
Vpx relieves inhibition of HIV-1 infection of macrophages mediated by the SAMHD1 protein.
Nature.
474
:
658
661
.
24.
Laguette
N.
,
B.
Sobhian
,
N.
Casartelli
,
M.
Ringeard
,
C.
Chable-Bessia
,
E.
Ségéral
,
A.
Yatim
,
S.
Emiliani
,
O.
Schwartz
,
M.
Benkirane
.
2011
.
SAMHD1 is the dendritic- and myeloid-cell-specific HIV-1 restriction factor counteracted by Vpx.
Nature.
474
:
654
657
.
25.
Gao
D.
,
J.
Wu
,
Y. T.
Wu
,
F.
Du
,
C.
Aroh
,
N.
Yan
,
L.
Sun
,
Z. J.
Chen
.
2013
.
Cyclic GMP-AMP synthase is an innate immune sensor of HIV and other retroviruses.
Science.
341
:
903
906
.
26.
Shapira
S. D.
,
I.
Gat-Viks
,
B. O.
Shum
,
A.
Dricot
,
M. M.
de Grace
,
L.
Wu
,
P. B.
Gupta
,
T.
Hao
,
S. J.
Silver
,
D. E.
Root
, et al
2009
.
A physical and regulatory map of host-influenza interactions reveals pathways in H1N1 infection.
Cell.
139
:
1255
1267
.
27.
Watson
S. F.
,
N.
Bellora
,
S.
Macias
.
2020
.
ILF3 contributes to the establishment of the antiviral type I interferon program.
Nucleic Acids Res.
48
:
116
129
.
28.
Patzina
C.
,
C. H.
Botting
,
A.
García-Sastre
,
R. E.
Randall
,
B. G.
Hale
.
2017
.
Human interactome of the influenza B virus NS1 protein.
J. Gen. Virol.
98
:
2267
2273
.
29.
Reichman
T. W.
,
A. M.
Parrott
,
I.
Fierro-Monti
,
D. J.
Caron
,
P. N.
Kao
,
C. G.
Lee
,
H.
Li
,
M. B.
Mathews
.
2003
.
Selective regulation of gene expression by nuclear factor 110, a member of the NF90 family of double-stranded RNA-binding proteins.
J. Mol. Biol.
332
:
85
98
.
30.
Nakadai
T.
,
A.
Fukuda
,
M.
Shimada
,
K.
Nishimura
,
K.
Hisatake
.
2015
.
The RNA binding complexes NF45-NF90 and NF45-NF110 associate dynamically with the c-fos gene and function as transcriptional coactivators.
J. Biol. Chem.
290
:
26832
26845
.
31.
Wu
T. H.
,
L.
Shi
,
A. W.
Lowe
,
M. R.
Nicolls
,
P. N.
Kao
.
2019
.
Inducible expression of immediate early genes is regulated through dynamic chromatin association by NF45/ILF2 and NF90/NF110/ILF3.
PLoS One.
14
:
e0216042
.
32.
Lindstedt
M.
,
B.
Johansson-Lindbom
,
C. A.
Borrebaeck
.
2002
.
Global reprogramming of dendritic cells in response to a concerted action of inflammatory mediators.
Int. Immunol.
14
:
1203
1213
.
33.
Di Rosa
M.
,
D.
Tibullo
,
S.
Saccone
,
G.
Distefano
,
M. S.
Basile
,
F.
Di Raimondo
,
L.
Malaguarnera
.
2016
.
CHI3L1 nuclear localization in monocyte derived dendritic cells.
Immunobiology.
221
:
347
356
.
34.
Nencioni
A.
,
F.
Grünebach
,
A.
Zobywlaski
,
C.
Denzlinger
,
W.
Brugger
,
P.
Brossart
.
2002
.
Dendritic cell immunogenicity is regulated by peroxisome proliferator-activated receptor gamma.
J. Immunol.
169
:
1228
1235
.
35.
Muzio
M.
,
D.
Bosisio
,
N.
Polentarutti
,
G.
D’amico
,
A.
Stoppacciaro
,
R.
Mancinelli
,
C.
van’t Veer
,
G.
Penton-Rol
,
L. P.
Ruco
,
P.
Allavena
,
A.
Mantovani
.
2000
.
Differential expression and regulation of toll-like receptors (TLR) in human leukocytes: selective expression of TLR3 in dendritic cells.
J. Immunol.
164
:
5998
6004
.
36.
Wang
P.
,
Y.
Xue
,
Y.
Han
,
L.
Lin
,
C.
Wu
,
S.
Xu
,
Z.
Jiang
,
J.
Xu
,
Q.
Liu
,
X.
Cao
.
2014
.
The STAT3-binding long noncoding RNA lnc-DC controls human dendritic cell differentiation.
Science.
344
:
310
313
.
37.
D’Amico
G.
,
G.
Frascaroli
,
G.
Bianchi
,
P.
Transidico
,
A.
Doni
,
A.
Vecchi
,
S.
Sozzani
,
P.
Allavena
,
A.
Mantovani
.
2000
.
Uncoupling of inflammatory chemokine receptors by IL-10: generation of functional decoys.
Nat. Immunol.
1
:
387
391
.
38.
Videira
P. A.
,
I. F.
Amado
,
H. J.
Crespo
,
M. C.
Algueró
,
F.
Dall’Olio
,
M. G.
Cabral
,
H.
Trindade
.
2008
.
Surface alpha 2-3- and alpha 2-6-sialylation of human monocytes and derived dendritic cells and its influence on endocytosis.
Glycoconj. J.
25
:
259
268
.
39.
Wu
X.
,
H.
Gao
,
R.
Bleday
,
Z.
Zhu
.
2014
.
Homeobox transcription factor VentX regulates differentiation and maturation of human dendritic cells.
J. Biol. Chem.
289
:
14633
14643
.
40.
Skelton
L.
,
M.
Cooper
,
M.
Murphy
,
A.
Platt
.
2003
.
Human immature monocyte-derived dendritic cells express the G protein-coupled receptor GPR105 (KIAA0001, P2Y14) and increase intracellular calcium in response to its agonist, uridine diphosphoglucose.
J. Immunol.
171
:
1941
1949
.
41.
Jin
P.
,
T. H.
Han
,
J.
Ren
,
S.
Saunders
,
E.
Wang
,
F. M.
Marincola
,
D. F.
Stroncek
.
2010
.
Molecular signatures of maturing dendritic cells: implications for testing the quality of dendritic cell therapies.
J. Transl. Med.
8
:
4
.
42.
Sander
J.
,
S. V.
Schmidt
,
B.
Cirovic
,
N.
McGovern
,
O.
Papantonopoulou
,
A. L.
Hardt
,
A. C.
Aschenbrenner
,
C.
Kreer
,
T.
Quast
,
A. M.
Xu
, et al
2017
.
Cellular differentiation of human monocytes is regulated by time-dependent interleukin-4 signaling and the transcriptional regulator NCOR2.
Immunity.
47
:
1051
1066.e12
.
43.
Landmann
S.
,
A.
Mühlethaler-Mottet
,
L.
Bernasconi
,
T.
Suter
,
J. M.
Waldburger
,
K.
Masternak
,
J. F.
Arrighi
,
C.
Hauser
,
A.
Fontana
,
W.
Reith
.
2001
.
Maturation of dendritic cells is accompanied by rapid transcriptional silencing of class II transactivator (Ciita) expression.
J Exp Med.
194
:
379
391
.
44.
Lahaye
X.
,
T.
Satoh
,
M.
Gentili
,
S.
Cerboni
,
C.
Conrad
,
I.
Hurbain
,
A.
El Marjou
,
C.
Lacabaratz
,
J. D.
Lelièvre
,
N.
Manel
.
2013
.
The capsids of HIV-1 and HIV-2 determine immune detection of the viral cDNA by the innate sensor cGAS in dendritic cells.
Immunity.
39
:
1132
1142
.
45.
Rasaiyaah
J.
,
C. P.
Tan
,
A. J.
Fletcher
,
A. J.
Price
,
C.
Blondeau
,
L.
Hilditch
,
D. A.
Jacques
,
D. L.
Selwood
,
L. C.
James
,
M.
Noursadeghi
,
G. J.
Towers
.
2013
.
HIV-1 evades innate immune recognition through specific cofactor recruitment.
Nature.
503
:
402
405
.
46.
Yan
N.
,
A. D.
Regalado-Magdos
,
B.
Stiggelbout
,
M. A.
Lee-Kirsch
,
J.
Lieberman
.
2010
.
The cytosolic exonuclease TREX1 inhibits the innate immune response to human immunodeficiency virus type 1.
Nat. Immunol.
11
:
1005
1013
.
47.
Lepelley
A.
,
S.
Louis
,
M.
Sourisseau
,
H. K.
Law
,
J.
Pothlichet
,
C.
Schilte
,
L.
Chaperot
,
J.
Plumas
,
R. E.
Randall
,
M.
Si-Tahar
, et al
2011
.
Innate sensing of HIV-infected cells.
PLoS Pathog.
7
:
e1001284
.
48.
Gringhuis
S. I.
,
M.
van der Vlist
,
L. M.
van den Berg
,
J.
den Dunnen
,
M.
Litjens
,
T. B.
Geijtenbeek
.
2010
.
HIV-1 exploits innate signaling by TLR8 and DC-SIGN for productive infection of dendritic cells.
Nat. Immunol.
11
:
419
426
.
49.
Akiyama
H.
,
C. M.
Miller
,
C. R.
Ettinger
,
A. C.
Belkina
,
J. E.
Snyder-Cappione
,
S.
Gummuluru
.
2018
.
HIV-1 intron-containing RNA expression induces innate immune activation and T cell dysfunction.
Nat. Commun.
9
:
3450
.
50.
McCauley
S. M.
,
K.
Kim
,
A.
Nowosielska
,
A.
Dauphin
,
L.
Yurkovetskiy
,
W. E.
Diehl
,
J.
Luban
.
2018
.
Intron-containing RNA from the HIV-1 provirus activates type I interferon and inflammatory cytokines.
Nat. Commun.
9
:
5305
.
51.
Lahaye
X.
,
M.
Gentili
,
A.
Silvin
,
C.
Conrad
,
L.
Picard
,
M.
Jouve
,
E.
Zueva
,
M.
Maurin
,
F.
Nadalin
,
G. J.
Knott
, et al
2018
.
NONO detects the nuclear HIV capsid to promote cGAS-mediated innate immune activation.
Cell.
175
:
488
501.e22
.
52.
Yin
X.
,
S.
Langer
,
Z.
Zhang
,
K. M.
Herbert
,
S.
Yoh
,
R.
König
,
S. K.
Chanda
.
2020
.
Sensor sensibility-HIV-1 and the innate immune response.
Cells.
9
:
254
.
53.
Chevrier
N.
,
P.
Mertins
,
M. N.
Artyomov
,
A. K.
Shalek
,
M.
Iannacone
,
M. F.
Ciaccio
,
I.
Gat-Viks
,
E.
Tonti
,
M. M.
DeGrace
,
K. R.
Clauser
, et al
2011
.
Systematic discovery of TLR signaling components delineates viral-sensing circuits.
Cell.
147
:
853
867
.
54.
Parker
L. M.
,
I.
Fierro-Monti
,
M. B.
Mathews
.
2001
.
Nuclear factor 90 is a substrate and regulator of the eukaryotic initiation factor 2 kinase double-stranded RNA-activated protein kinase.
J. Biol. Chem.
276
:
32522
32530
.
55.
Parrott
A. M.
,
M. R.
Walsh
,
T. W.
Reichman
,
M. B.
Mathews
.
2005
.
RNA binding and phosphorylation determine the intracellular distribution of nuclear factors 90 and 110.
J. Mol. Biol.
348
:
281
293
.
56.
Reichman
T. W.
,
M. B.
Mathews
.
2003
.
RNA binding and intramolecular interactions modulate the regulation of gene expression by nuclear factor 110.
RNA.
9
:
543
554
.
57.
Jiang
Z.
,
C. M.
Slater
,
Y.
Zhou
,
K.
Devarajan
,
K. J.
Ruth
,
Y.
Li
,
K. Q.
Cai
,
M.
Daly
,
X.
Chen
.
2017
.
LincIN, a novel NF90-binding long non-coding RNA, is overexpressed in advanced breast tumors and involved in metastasis.
Breast Cancer Res.
19
:
62
.
58.
Parrott
A. M.
,
M. B.
Mathews
.
2007
.
Novel rapidly evolving hominid RNAs bind nuclear factor 90 and display tissue-restricted distribution.
Nucleic Acids Res.
35
:
6249
6258
.
59.
Huang
W.
,
J.
Liu
,
J.
Yan
,
Z.
Huang
,
X.
Zhang
,
Y.
Mao
,
X.
Huang
.
2020
.
LncRNA LINC00470 promotes proliferation through association with NF45/NF90 complex in hepatocellular carcinoma.
Hum. Cell
33
:
131
139
.
60.
Zhang
Y.
,
J.
Sun
,
Y.
Qi
,
Y.
Wang
,
Y.
Ding
,
K.
Wang
,
Q.
Zhou
,
J.
Wang
,
F.
Ma
,
J.
Zhang
,
B.
Guo
.
2020
.
Long non-coding RNA TPT1-AS1 promotes angiogenesis and metastasis of colorectal cancer through TPT1-AS1/NF90/VEGFA signaling pathway.
Aging (Albany NY).
12
:
6191
6205
.
61.
Zhang
H.
,
Y.
Liu
,
L.
Yan
,
M.
Zhang
,
X.
Yu
,
W.
Du
,
S.
Wang
,
Q.
Li
,
H.
Chen
,
Y.
Zhang
, et al
2018
.
Increased levels of the long noncoding RNA, HOXA-AS3, promote proliferation of A549 cells.
Cell Death Dis.
9
:
707
.
62.
Sur
S.
,
H.
Nakanishi
,
R.
Steele
,
D.
Zhang
,
M. A.
Varvares
,
R. B.
Ray
.
2020
.
Long non-coding RNA ELDR enhances oral cancer growth by promoting ILF3-cyclin E1 signaling.
EMBO Rep.
21
:
e51042
.
63.
de Baey
A.
,
A.
Lanzavecchia
.
2000
.
The role of aquaporins in dendritic cell macropinocytosis.
J. Exp. Med.
191
:
743
748
.
64.
Wacleche
V. S.
,
A.
Cattin
,
J. P.
Goulet
,
D.
Gauchat
,
A.
Gosselin
,
A.
Cleret-Buhot
,
Y.
Zhang
,
C. L.
Tremblay
,
J. P.
Routy
,
P.
Ancuta
.
2018
.
CD16+ monocytes give rise to CD103+RALDH2+TCF4+ dendritic cells with unique transcriptional and immunological features.
Blood Adv.
2
:
2862
2878
.
65.
Pickl
W. F.
,
O.
Majdic
,
P.
Kohl
,
J.
Stöckl
,
E.
Riedl
,
C.
Scheinecker
,
C.
Bello-Fernandez
,
W.
Knapp
.
1996
.
Molecular and functional characteristics of dendritic cells generated from highly purified CD14+ peripheral blood monocytes.
J. Immunol.
157
:
3850
3859
.
66.
Li
K.
,
H.
Fazekasova
,
N.
Wang
,
Q.
Peng
,
S. H.
Sacks
,
G.
Lombardi
,
W.
Zhou
.
2012
.
Functional modulation of human monocytes derived DCs by anaphylatoxins C3a and C5a.
Immunobiology.
217
:
65
73
.
67.
Ju
X.
,
M.
Zenke
,
D. N.
Hart
,
G. J.
Clark
.
2008
.
CD300a/c regulate type I interferon and TNF-alpha secretion by human plasmacytoid dendritic cells stimulated with TLR7 and TLR9 ligands.
Blood.
112
:
1184
1194
.
68.
Van der Borght
K.
,
C. L.
Scott
,
V.
Nindl
,
A.
Bouché
,
L.
Martens
,
D.
Sichien
,
J.
Van Moorleghem
,
M.
Vanheerswynghels
,
S.
De Prijck
,
Y.
Saeys
, et al
2017
.
Myocardial infarction primes autoreactive T cells through activation of dendritic cells.
Cell Rep.
18
:
3005
3017
.
69.
Liu
T. M.
,
H.
Wang
,
D. N.
Zhang
,
G. Z.
Zhu
.
2019
.
Transcription factor MafB suppresses type I interferon production by CD14+ monocytes in patients with chronic hepatitis C.
Front. Microbiol.
10
:
1814
.
70.
Navarro-Barriuso
J.
,
M. J.
Mansilla
,
M.
Naranjo-Gómez
,
A.
Sánchez-Pla
,
B.
Quirant-Sánchez
,
A.
Teniente-Serra
,
C.
Ramo-Tello
,
E. M.
Martínez-Cáceres
.
2018
.
Comparative transcriptomic profile of tolerogenic dendritic cells differentiated with vitamin D3, dexamethasone and rapamycin.
Sci. Rep.
8
:
14985
.
71.
Maniecki
M. B.
,
H. J.
Møller
,
S. K.
Moestrup
,
B. K.
Møller
.
2006
.
CD163 positive subsets of blood dendritic cells: the scavenging macrophage receptors CD163 and CD91 are coexpressed on human dendritic cells and monocytes.
Immunobiology.
211
:
407
417
.
72.
Corinti
S.
,
C.
Albanesi
,
A.
la Sala
,
S.
Pastore
,
G.
Girolomoni
.
2001
.
Regulatory activity of autocrine IL-10 on dendritic cell functions.
J. Immunol.
166
:
4312
4318
.
73.
Olivar
R.
,
A.
Luque
,
S.
Cárdenas-Brito
,
M.
Naranjo-Gómez
,
A. M.
Blom
,
F. E.
Borràs
,
S.
Rodriguez de Córdoba
,
P. F.
Zipfel
,
J. M.
Aran
.
2016
.
The complement inhibitor factor H generates an anti-inflammatory and tolerogenic state in monocyte-derived dendritic cells.
J. Immunol.
196
:
4274
4290
.
74.
Sanecka
A.
,
M.
Ansems
,
A. C.
Prosser
,
K.
Danielski
,
K.
Warner
,
M. H.
den Brok
,
B. J.
Jansen
,
D.
Eleveld-Trancikova
,
G. J.
Adema
.
2011
.
DC-STAMP knock-down deregulates cytokine production and T-cell stimulatory capacity of LPS-matured dendritic cells.
BMC Immunol.
12
:
57
.
75.
Yokota
A.
,
H.
Takeuchi
,
N.
Maeda
,
Y.
Ohoka
,
C.
Kato
,
S. Y.
Song
,
M.
Iwata
.
2009
.
GM-CSF and IL-4 synergistically trigger dendritic cells to acquire retinoic acid-producing capacity.
Int. Immunol.
21
:
361
377
.
76.
Goudot
C.
,
A.
Coillard
,
A. C.
Villani
,
P.
Gueguen
,
A.
Cros
,
S.
Sarkizova
,
T. L.
Tang-Huau
,
M.
Bohec
,
S.
Baulande
,
N.
Hacohen
, et al
2017
.
Aryl hydrocarbon receptor controls monocyte differentiation into dendritic cells versus macrophages.
Immunity.
47
:
582
596.e6
.
77.
Vulcano
M.
,
C.
Albanesi
,
A.
Stoppacciaro
,
R.
Bagnati
,
G.
D’Amico
,
S.
Struyf
,
P.
Transidico
,
R.
Bonecchi
,
A.
Del Prete
,
P.
Allavena
, et al
2001
.
Dendritic cells as a major source of macrophage-derived chemokine/CCL22 in vitro and in vivo.
Eur. J. Immunol.
31
:
812
822
.
78.
Park
K.
,
A. L.
Scott
.
2010
.
Cholesterol 25-hydroxylase production by dendritic cells and macrophages is regulated by type I interferons.
J. Leukoc. Biol.
88
:
1081
1087
.
79.
Subramanian
A.
,
P.
Tamayo
,
V. K.
Mootha
,
S.
Mukherjee
,
B. L.
Ebert
,
M. A.
Gillette
,
A.
Paulovich
,
S. L.
Pomeroy
,
T. R.
Golub
,
E. S.
Lander
,
J. P.
Mesirov
.
2005
.
Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles.
Proc. Natl. Acad. Sci. USA.
102
:
15545
15550
.
80.
Liberzon
A.
,
C.
Birger
,
H.
Thorvaldsdóttir
,
M.
Ghandi
,
J. P.
Mesirov
,
P.
Tamayo
.
2015
.
The Molecular Signatures Database (MSigDB) hallmark gene set collection.
Cell Syst.
1
:
417
425
.
81.
Chen
E. Y.
,
C. M.
Tan
,
Y.
Kou
,
Q.
Duan
,
Z.
Wang
,
G. V.
Meirelles
,
N. R.
Clark
,
A.
Ma’ayan
.
2013
.
Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool.
BMC Bioinformatics.
14
:
128
.
82.
Croasdell
A.
,
P. F.
Duffney
,
N.
Kim
,
S. H.
Lacy
,
P. J.
Sime
,
R. P.
Phipps
.
2015
.
PPARγ and the Innate Immune System Mediate the Resolution of Inflammation.
PPAR Res.
2015
:
549691
.
83.
Pfeifer
I.
,
R.
Elsby
,
M.
Fernandez
,
P. A.
Faria
,
D. R.
Nussenzveig
,
I. S.
Lossos
,
B. M.
Fontoura
,
W. D.
Martin
,
G. N.
Barber
.
2008
.
NFAR-1 and -2 modulate translation and are required for efficient host defense.
Proc. Natl. Acad. Sci. USA.
105
:
4173
4178
.
84.
Sun
L.
,
J.
Wu
,
F.
Du
,
X.
Chen
,
Z. J.
Chen
.
2013
.
Cyclic GMP-AMP synthase is a cytosolic DNA sensor that activates the type I interferon pathway.
Science.
339
:
786
791
.
85.
Saunders
L. R.
,
D. J.
Perkins
,
S.
Balachandran
,
R.
Michaels
,
R.
Ford
,
A.
Mayeda
,
G. N.
Barber
.
2001
.
Characterization of two evolutionarily conserved, alternatively spliced nuclear phosphoproteins, NFAR-1 and -2, that function in mRNA processing and interact with the double-stranded RNA-dependent protein kinase, PKR.
J. Biol. Chem.
276
:
32300
32312
.
86.
Harashima
A.
,
T.
Guettouche
,
G. N.
Barber
.
2010
.
Phosphorylation of the NFAR proteins by the dsRNA-dependent protein kinase PKR constitutes a novel mechanism of translational regulation and cellular defense.
Genes Dev.
24
:
2640
2653
.
87.
Krasnoselskaya-Riz
I.
,
A.
Spruill
,
Y. W.
Chen
,
D.
Schuster
,
T.
Teslovich
,
C.
Baker
,
A.
Kumar
,
D. A.
Stephan
.
2002
.
Nuclear factor 90 mediates activation of the cellular antiviral expression cascade.
AIDS Res. Hum. Retroviruses.
18
:
591
604
.
88.
Urcuqui-Inchima
S.
,
M. E.
Castaño
,
D.
Hernandez-Verdun
,
G.
St-Laurent
III
,
A.
Kumar
.
2006
.
Nuclear Factor 90, a cellular dsRNA binding protein inhibits the HIV Rev-export function.
Retrovirology.
3
:
83
.
89.
Agbottah
E. T.
,
C.
Traviss
,
J.
McArdle
,
S.
Karki
,
G. C.
St Laurent
III
,
A.
Kumar
.
2007
.
Nuclear factor 90(NF90) targeted to TAR RNA inhibits transcriptional activation of HIV-1.
Retrovirology.
4
:
41
.
90.
Hoque
M.
,
R. A.
Shamanna
,
D.
Guan
,
T.
Pe’ery
,
M. B.
Mathews
.
2011
.
HIV-1 replication and latency are regulated by translational control of cyclin T1.
J. Mol. Biol.
410
:
917
932
.
91.
Guan
D.
,
N.
Altan-Bonnet
,
A. M.
Parrott
,
C. J.
Arrigo
,
Q.
Li
,
M.
Khaleduzzaman
,
H.
Li
,
C. G.
Lee
,
T.
Pe’ery
,
M. B.
Mathews
.
2008
.
Nuclear factor 45 (NF45) is a regulatory subunit of complexes with NF90/110 involved in mitotic control.
Mol. Cell. Biol.
28
:
4629
4641
.
92.
Chaumet
A.
,
S.
Castella
,
L.
Gasmi
,
A.
Fradin
,
G.
Clodic
,
G.
Bolbach
,
R.
Poulhe
,
P.
Denoulet
,
J. C.
Larcher
.
2013
.
Proteomic analysis of interleukin enhancer binding factor 3 (Ilf3) and nuclear factor 90 (NF90) interactome.
Biochimie.
95
:
1146
1157
.
93.
Everts
B.
,
E.
Amiel
,
S. C.
Huang
,
A. M.
Smith
,
C. H.
Chang
,
W. Y.
Lam
,
V.
Redmann
,
T. C.
Freitas
,
J.
Blagih
,
G. J.
van der Windt
, et al
2014
.
TLR-driven early glycolytic reprogramming via the kinases TBK1-IKKE supports the anabolic demands of dendritic cell activation.
Nat. Immunol.
15
:
323
332
.
94.
Baker
A. D.
,
A.
Malur
,
B. P.
Barna
,
M. S.
Kavuru
,
A. G.
Malur
,
M. J.
Thomassen
.
2010
.
PPARgamma regulates the expression of cholesterol metabolism genes in alveolar macrophages.
Biochem. Biophys. Res. Commun.
393
:
682
687
.
95.
York
A. G.
,
K. J.
Williams
,
J. P.
Argus
,
Q. D.
Zhou
,
G.
Brar
,
L.
Vergnes
,
E. E.
Gray
,
A.
Zhen
,
N. C.
Wu
,
D. H.
Yamada
, et al
2015
.
Limiting cholesterol biosynthetic flux spontaneously engages type I IFN signaling.
Cell.
163
:
1716
1729
.
96.
Plutzky
J.
2011
.
The PPAR-RXR transcriptional complex in the vasculature: energy in the balance.
Circ. Res.
108
:
1002
1016
.
97.
Larange
A.
,
H.
Cheroutre
.
2016
.
Retinoic acid and retinoic acid receptors as pleiotropic modulators of the immune system.
Annu. Rev. Immunol.
34
:
369
394
.
98.
Szatmari
I.
,
A.
Pap
,
R.
Rühl
,
J. X.
Ma
,
P. A.
Illarionov
,
G. S.
Besra
,
E.
Rajnavolgyi
,
B.
Dezso
,
L.
Nagy
.
2006
.
PPARgamma controls CD1d expression by turning on retinoic acid synthesis in developing human dendritic cells.
J. Exp. Med.
203
:
2351
2362
.
99.
Geissmann
F.
,
P.
Revy
,
N.
Brousse
,
Y.
Lepelletier
,
C.
Folli
,
A.
Durandy
,
P.
Chambon
,
M.
Dy
.
2003
.
Retinoids regulate survival and antigen presentation by immature dendritic cells.
J. Exp. Med.
198
:
623
634
.
100.
Szatmari
I.
,
L.
Nagy
.
2008
.
Nuclear receptor signalling in dendritic cells connects lipids, the genome and immune function.
EMBO J.
27
:
2353
2362
.
101.
Appel
S.
,
V.
Mirakaj
,
A.
Bringmann
,
M. M.
Weck
,
F.
Grünebach
,
P.
Brossart
.
2005
.
PPAR-gamma agonists inhibit toll-like receptor-mediated activation of dendritic cells via the MAP kinase and NF-kappaB pathways.
Blood.
106
:
3888
3894
.
102.
Coutant
F.
,
S.
Agaugué
,
L.
Perrin-Cocon
,
P.
André
,
V.
Lotteau
.
2004
.
Sensing environmental lipids by dendritic cell modulates its function.
J. Immunol.
172
:
54
60
.
103.
Khare
A.
,
K.
Chakraborty
,
M.
Raundhal
,
P.
Ray
,
A.
Ray
.
2015
.
Cutting edge: dual function of PPARγ in CD11c+ cells ensures immune tolerance in the airways.
J. Immunol.
195
:
431
435
.
104.
Li
C. G.
,
C.
Mahon
,
N. M.
Sweeney
,
E.
Verschueren
,
V.
Kantamani
,
D.
Li
,
J. K.
Hennigs
,
D. P.
Marciano
,
I.
Diebold
,
O.
Abu-Halawa
, et al
2019
.
PPARγ interaction with UBR5/ATMIN promotes DNA repair to maintain endothelial homeostasis.
Cell Rep.
26
:
1333
1343.e7
.
105.
Yoshida
T.
,
K.
Kato
,
M.
Oguri
,
H.
Horibe
,
T.
Kawamiya
,
K.
Yokoi
,
T.
Fujimaki
,
S.
Watanabe
,
K.
Satoh
,
Y.
Aoyagi
, et al
2011
.
Association of polymorphisms of BTN2A1 and ILF3 with myocardial infarction in Japanese individuals with different lipid profiles.
Mol. Med. Rep.
4
:
511
518
.
106.
Knevel
R.
,
T. W. J.
Huizinga
,
F.
Kurreeman
.
2017
.
Genomic influences on susceptibility and severity of rheumatoid arthritis.
Rheum. Dis. Clin. North Am.
43
:
347
361
.

The authors have no financial conflicts of interest.

Supplementary data