Long noncoding RNA CHROMR regulates antiviral immunity in humans

Significance An effective innate immune response to virus infection requires the induction of type I interferons and up-regulation of hundreds of interferon-stimulated genes (ISGs) that instruct antiviral functions and immune regulation. Deciphering the regulatory mechanisms that direct expression of the ISG network is critical for understanding the fundamental organization of the innate immune response and the development of antiviral therapies. We define a regulatory role for the primate-specific long noncoding RNA CHROMR in coordinating ISG transcription. CHROMR sequesters the interferon regulatory factor (IRF)-2/IRF2BP2 complex that restrains ISG transcription and thus is required to restrict influenza virus replication. These data identify a novel regulator of the antiviral gene program in humans and provide insights into the multilayered regulatory network that controls the innate immune response.

RNA isolation, cell fractionation and qPCR. Total RNA was isolated using TRIzol reagent (Invitrogen) and Direct-zol RNA MicroPrep columns (Zymo Research). For cell fractionation experiments RNA was isolated from separate cytoplasmic and nuclear fractions using the PARIS kit (Thermo Fisher Scientific). Upon isolation, RNA was reverse transcribed using iScript cDNA Synthesis kit (Bio-Rad Laboratories) and quantitative PCR analysis was conducted using KAPA SYBR green Supermix (KAPA Biosystems) according to the manufacturer's instructions and quantified on Quantstudio 3 (Applied Biosystems). Fold change in mRNA expression was calculated using the comparative cycle method (2 −ΔΔCt ) normalized to the housekeeping gene GAPDH. A list of primers used in this study can be found in SI Appendix, Table S6.
global impact of trimming. Reads were then aligned to the human genome (hg38) using Bowtie2 (5). Alignments were sorted and indexed using Samtools for downstream processes (6). MACS2 was then used to identify significant peaks (7). Peaks with a Q-value of less than 0.05 were retained. Custom scripts and the ChIPQC R package were used to assess ChIP-seq peak quality and reproducibility (8). Peaks present in all replicates from each condition were retained for differential enrichment analysis. Peaks were annotated using ChIPseeker package from Bioconductor (9). Peaks that overlapped a 4kb window centered at an annotated transcription start site were annotated as promoter peaks. Differential enrichment analysis was performed using DiffBind package from Bioconductor (10). Peaks with a false discovery rate (FDR-)adjusted P-value of 0.1 or less were considered differentially enriched between the knockdown and control conditions. Functional analysis of significantly enriched peaks was performed using Genomic Regions Enrichment Annotations Tool (GREAT) with default parameters (11). Motif enrichment analysis was performed on 3kb windows centered at the TSS of genes with differentially enriched promoter peaks using Hypergeometric Optimization of Motif EnRichment (HOMER) with the command 'findMotifs.pl' (12). ChIP-seq data are deposited in the GEO under the accession number GSE190413.

Chromatin Isolation by RNA Precipitation (ChIRP).
Cell harvesting, lysis, disruption, and chromatin isolation by RNA purification were performed as previously described (13) with the following modifications: 1) Cells were cross-linked in 3% formaldehyde for 30 min, followed by 0.125 M glycine quenching for 5 min; 2) Hybridization was performed for 16h; 3) For mass spectrometry (MS) experiments, lysates were pre-cleared by incubating with 30 mL washed beads per mL of lysate at 37°C for 30 min with mixing; 4) As a negative control, lysates were pooled and aliquoted into equal amounts and RNA was removed by incubating with RNase A (1 μg/mL, Sigma), and subsequent incubation at 37°C for 30 min prior to hybridization steps. RNA, DNA, protein isolation was performed as described (13) and further detailed below for ChIRP followed by DNAseq (ChIRP-seq) or Comprehensive Identification of RNA-binding Proteins by Mass Spectrometry (ChIRP-MS). RNA extraction was performed for validation of lncRNA enrichment. A list of probes used in this study can be found in SI Appendix, Table S6.
ChIRP followed by DNA-seq (ChIRP-seq). DNA was eluted from hybridized magnetic beads and subjected for Illumina sequencing. In short, beads were washed at room temperature with ChIRP wash buffer (EMD Millipore, #17-10494). Beads were subsequently captured using a DynaMag magnet (Thermo Fisher Scientific) and DNA was eluted by suspending beads in elution buffer (20 mM Tris pH 7.4, 1% SDS, 50 mM NaHCO3, 1 mM EDTA). ChIRP eluates were reverse crosslinked at 65°C for 4h, digested with Proteinase K (EMD Millipore) at 55°C followed by incubation with RNase cocktail (Ambion). ChIRP purified DNA was cleaned using PCR purification columns (Zymo Research) and subjected to Illumina sequencing. Reads were trimmed using Trimmomatic (14) and mapped to hg19 using BWA (15). Peaks were then called for each probe set and replicate using the 'callpeak' function from MACS2 (7) relative to the input from the same replicate. Peaks were then imported into the DiffBind package from Bioconductor (10) and differential peaks were called between even and odd probe sets. Only peaks with no differential binding between the probe sets were retained. Peaks were then assigned to their nearest genomic location using ChIPseeker package from Bioconductor (9). ChIRP-seq data are deposited in the GEO under the accession number GSE190413.

Comprehensive Identification of RNA-binding Proteins by Mass Spectrometry (ChIRP-MS).
Protein was isolated from magnetic beads and analyzed by MS. To elute protein beads were collected on magnetic stand, resuspended in biotin elution buffer (12.5 mM D-biotin (Thermo Fisher Scientific), 7.5 mM HEPES pH 7.5, 75 mM NaCl, 1.5 mM EDTA, 0.15% SDS, 0.075% sarkosyl, and 0.02% sodium deoxycholate). Trichloroacetic acid (25% of total volume) was added to the clean eluent and proteins were precipitated at 4°C overnight. Proteins were pelleted at 16,000 g at 4°C for 30 min, washed with cold acetone and pelleted again at 16,000 g at 4°C for 5 min. Proteins were immediately solubilized in desired volumes of Laemmli sample buffer (Invitrogen) and boiled at 95°C for 30 min with occasional mixing to reverse crosslinking. Final protein samples were sizeseparated in Bis-Tris SDS-PAGE gels (Invitrogen) and submitted for MS analysis by the Proteomics Laboratory at NYU Langone Health. Individual samples were subjected to liquid chromatography (LC) separation with MS using the autosampler of an EASY-nLC 1000 (Thermo Fisher Scientific).
Subsequently, peptides were gradient eluted from the column directly to Q Exactive mass spectrometer using a 1h gradient (Thermo Fisher Scientific). High resolution full MS spectra were acquired with a resolution of 70,000, an AGC target of 1 x 10 6 , with a maximum ion time of 120 ms, and scan range of 400 to 1,500 m/z. Following each full MS twenty data-dependent high resolution HCD MS/MS spectra were acquired. All MS/MS spectra were collected using the following instrument parameters: resolution of 17,500, AGC target of 5 x 10 4 , maximum ion time of 120 ms, one microscan, 2 m/z isolation window, fixed first mass of 150 m/z, and NCE of 27. MS/MS spectra were searched against a UniProt human database, using Sequest (16) within Proteome Discoverer (Thermo Fisher Scientific). Only high confidence peptides, based on a better than 1% FDR searched against a decoy database, were included for peptide identification. RNA Immunoprecipitation. Human histone H3, IRF2BP2, and HNRPNLL were immunoprecipitated from PMA-differentiated THP-1 macrophages. All immunoprecipitations were done using the MagnaRIP RNA-Binding Protein Immunoprecipitation Kit (EMD Millipore) according to the manufacturers' instructions. Briefly, an antibody targeting human histone H3 (Abcam, ab1791), IRF2BP2 (Abcam, ab220155), HNRNPLL (Cell Signaling, 4783), or an isotype matched control antibody (Sigma, 12-370 or 12-371) were bound to magnetic beads and incubated with lysed cells at 4°C for 24h. Beads were isolated and cleaved from the bound proteins by proteinase K, and coprecipitated RNA was purified. qPCR analysis of total RNA was performed to detect enrichment of CHROMR variants and control genes in the protein-of-interest precipitated fraction was determined as percentage of 1% input control. Mutagenesis studies. The interaction between IRF2BP2 and CHROMR3 was analysed by mutating the putative interaction site between IRF2BP2 and CHROMR3. Two GG-doublets in the sequence of a plasmid overexpressing CHROMR3 (2) were replaced with two CC-doublets creating a plasmid overexpressing CHROMR3-G4mut. All mutations were performed using the Quickchange XL kit (Stratagene) using the primers indicated in SI Appendix, Table S6.
Bioinformatics. Enrichment analysis of interferon-stimulated genes was performed using Enrichr (17), the web-based software for Gene Set Enrichment Analysis was used for ChIP Enrichment Analysis (ChEA) (18) database (2016) which contains results from transcription factor ChIP-seq studies extracted from supporting material. Results were manually curated to remove duplicate studies or studies performed with non-human transcription factors. The catRAPID algorithm (19) was used to estimate the binding propensity of IRF2BP2 and CHROMR. The interaction score is generated using the interaction propensity distribution of a reference set, as described (20). The QGRS Mapper (21) was used for recognition and mapping of putative quadruplexes in CHROMR.
RNAfold, part of The Vienna RNA Websuite (22), was used to predict the minimum free energy secondary structure of CHROMR3 and the RNA plot was created with RNArtist, developed by Fabrice Jossinet and available at https://github.com/fjossinet/RNArtist.

CHROMR expression levels in whole blood of patients infected with influenza
*Patient status at time of whole blood RNA sequencing