Dendritic Cell Culture and Stimulation
The cDC1 line (MutuDC1940) was procured from Prof. Hans Acha Orbea’s lab. The group has also shown through extensive studies that the cell line resembles immature splenic murine CD8α+ DCs . We cultured and maintained the cells in a humidified incubator at 37 degree celsius with 5% CO2. The MutuDC cell line carries an e-GFP reporter present at CD11c promoter.
Generation of stable KD CD8α + line
We generated NCoR1 knock down (KD) cells using Sigma mission shRNA against NCoR1 and an Empty (Control) shRNA to generate a matched control. We used a lentivirus mediated approach with plasmids having a pLKO.1 backbone.
Control and NCoR1 KD cells were stimulated with IFNγ, CpG-B, poly (I:C) (pIC) and combined CpG + pIC/CpG + pIC + IFNγ for 2h or 6h (n = 2). Cells were further taken for the whole transcriptome experiment (RNA-seq) and H3K27ac ChIP-seq. MutuDC1940 cells were either left unstimulated or stimulated for 6h for the NCoR1 ChIP experiment.
RNA-seq Library preparation and Sequencing
For RNA-seq library preparation, RNA was isolated using NEB polyA mRNA isolation kit and libraries were prepared using NEB mRNA library preparation kit. Concentrations of each sample were measured using Qubit 2.0 (Invitrogen). RNA-seq libraries were sequenced by Genotypic technology, Bangalore, India on Illumina NextSeq-500 instrument.
For RT-qPCR, 8 x 105 control cells cDC1s were stimulated with IFNγ, CpG, pIC and CpG + pIC + IFNγ for 6h. For studies on IFNγ effect we seeded 8 x 105 control cells and NCoR1 KD cDC1s and stimulated with CpG + pIC and CpG + pIC + IFNγ for 6h. The RNA were isolated using the NucleoSpin RNA Plus miniprep kit (Machery Nagel). Total RNA isolation was carried out according to the manufacturer's protocol. RNA concentration was quantified using nanodrop spectrophotometer (Thermo). This was followed by taking 500ng-1µg of total RNA for cDNA preparation using high-capacity cDNA Reverse Transcriptase kit (Applied Biosystems). Quantitative PCR was performed using SYBR Green master mix (Applied Biosystems) and PCR amplification was monitored in real-time using QuantStudio-6 instrument. Primer sequence used for Il10, Il12b, Il6, Ifnb1 has been provided in the study published previously . Ifit3 (forward: 5’-CTGAAGGGGAGCGATTGATT-3’; reverse: 5’-AACGGCACATGACCAAAGAGTAGA-3’) and Cxcl10 (forward:5’-AGTGCTGCCGTCATTTTCTG-3’; 5’-ATTCTCACTGGCCCGTCAT-3’) primer sequence was used to estimate the mRNA expression.
Flow Cytometry (FACS)
Flow cytometry analysis was carried out using a well-established intracellular (IC) staining protocol. 8 x105 cells were seeded for IC staining. Cells were either left unstimulated or were stimulated with IFNγ, CpG, pIC, CpG + pIC, and CpG + pIC + IFNγ for 6h. Brefeldin A was added 2hs post stimulation. For staining, the cells were dissociated and washed with FACS buffer (3% FBS in 1X PBS). The cells were first fixed with 2% paraformaldehyde for 20 mins followed by permeabilization using 1x permeabilization buffer (eBiosciences). The fixed and permeabilized cells were then resuspended in IC staining buffer (FACS buffer: 1x permeabilization buffer:: 1 : 1) and stained with fluorochrome conjugated antibodies for the cytokines of interest. For optimal staining, the cells were incubated for 30–45 min in dark. After incubation, the cells were washed twice with FACS buffer and then acquired for differential expression analysis using LSRII Fortessa flow cytometer (BD Biosciences). The acquired data was analyzed using FlowJo X software (Treestar).
NCoR1 Chromatin Immunoprecipitation (ChIP) and Sequencing
NCoR1 ChIP assays in pIC and CpG + pIC + IFNγ stimulation conditions were performed similarly as described in our previous study . For ChIP-seq library preparation, 30 µl ChIP-DNA was processed for library preparation using NEB ChIP-seq library preparation kit (Illumina). After library preparation and quality check, the libraries were sequenced by Genotypic technology, Bangalore, India on Illumina NextSeq-500 instrument.
H3K27ac ChIP and Sequencing
40x106 Control and NCoR1 KD cells were seeded in 15 cm2 plates and prepared for ChIP before and after 2h, 6h CpG or pIC or CpG + pIC stimulation. The cells were cross-linked using 1% formaldehyde (Sigma) for 10 min at room temperature followed by quenching the reaction using 2.5 M glycine (Sigma) for 10 min. The ChIP experiments were performed as per the Mayer’s Lab Protocol. The cells were lysed in the FARHAM lysis buffer and centrifuged at 2000rpm at 4°C for 8 min. The chromatin was fragmented using a Bioruptor (Diagenode) sonicator for 30 min using high amplitude and 30s ON & 30s OFF cycles to obtain 200–500 bp size fragments. The concentration of the chromatin was estimated using a NanoDrop (Thermo) and the chromatin was diluted with a RIPA buffer prepared without protease inhibitor to make 125µg/ml of chromatin for each IP. 30ul of Dyna Magnetic beads (Anti-rabbit) were taken and added to 1ml tube for each IP. 3ul of rabbit monoclonal anti-H3K27ac antibody (Abcam, cat no: ab-177178), were added and incubated at 4°C overnight on a rocker shaker. Next day, the beads were washed six to seven times with LiCl buffer (1% NP-40, 100mM Tris HCl (pH 7.5), 500mM LiCl, 1% Sodium Deoxycholate) followed by two washes with TE buffer (10nM Tris HCl (pH 7.5), 0.1mM EDTA (pH 8)). Samples tubes were pulse spinned and remaining buffers were discarded. After removing the wash buffer completely, protein-bound chromatin complexes were eluted from beads for 30 min using 200ul of elution buffers. The eluted chromatin was reverse-crosslinked by overnight incubation on the shaker using 8ul of 5M NaCl. Next day DNA was purified from the reverse cross-linked chromatin by proteinase-K and RNase digestion followed by purification using PCR purification kit (Qiagen). H3K27ac ChIP sample library preparation was performed using an NEB ChIP library preparation Kit and sequenced using Illumina NextSeq-550.
RNA-seq data analysis
Raw RNA-seq fastq files were processed for quality control check using FASTQC and aligned to the mouse genome (mm10 RefSeq) using tophat2 to maintain the uniformity of analysis as unstimulated and CpG stimulated samples in control and NCoR1 KD were aligned using tophat2 in previous study. We then extracted raw counts from the respective sample using featureCounts tool (v1.6.2) [63–65]. Raw counts were then analyzed for differential gene expression analysis using DESeq2 (v = 1.24) . Differentially expressed genes were then filtertered based on log2 fold change > = 1 and adjusted p-value < 0.05. Normalized count and variance stabilized transformed value were used for downstream analysis. CpG specific, pIC specific and common CpG-pIC genes were identified using comparison of control CpG and pIC samples. To identify synergy/antagonist genes among CpG/pIC specific or common genes in combined CpG + pIC + IFNγ stimulation, ratio of normalized count in CpG + pIC + IFNγ and sum of individual CpG and pIC response were calculated. Genes having ratios greater than 1.2 and less than 0.5 were defined as synergy genes and synergistic genes respectively.
NCoR1 ChIP-seq data analysis
Raw reads of ChIP-seq samples were processed for quality control analysis and aligned to the mouse reference genome (mm10) using bowtie2 (188.8.131.52) (with default parameter). Uniquely aligned reads were extracted (MAPQ > 10) using SAMtools [67, 68]. Peak calling were performed using findPeaks program of HOMER using -style as factor and p-value cut-off of 0.0001 . To visualize ChIP-seq data in IGV, BigWig files were generated using the makeUCSCfile program of HOMER. Peaks were filtered against ENCODE mm10 blacklisted regions [ref]. Merged peak files from all the conditions were generated using bedops (-m option). Differential NCoR1 binding sites in CpG, pIC and CpG + pIC + IFNγ were identified using the getDifferentialPeaks program of HOMER with fold change enrichment cut-off of 2 . Based on fold change of enrichment, peaks were categorized into four clusters. Cluster I (2 fold increase in NCoR1 enrichment in CpG, pIC and CpG + pIC + IFNγ stimulation compared to Unstimulated), Cluster II (2 fold increase in NCoR1 enrichment in CpG + pIC + IFNγ stimulation compared to pIC and CpG), Cluster III (2 fold increase in NCoR1 enrichment in pIC stimulation compared to CpG and CpG + pIC + IFNγ), Cluster IV (No significant change in NCoR1 enrichment across the stimulation condition), Cluster V (2 fold decrease in NCoR1 enrichment after CpG, pIC and CpG + pIC + IFNγ activation). Further peaks were annotated to its nearest genes using mm10 UCSC annotation bioconductor and ChIPseeker R package .
Pathway and gene set enrichment analysis
Enriched pathway terms for the gene sets from different analyses were identified using clusterProfiler R package against Reactome gene sets downloaded from MSigDB database . Adjusted p-value < 0.05 were used to filter out significantly enriched pathway terms.
Association of DEGs with NCoR1 and H3K27ac bound targets.
Association between different gene lists were performed using GeneOverlap R package and Heatmap of log odds ratio with p-value were plotted using complexHeatmap [72, 73].
H3K27ac ChIP-seq data analysis
RAW single end reads were processed for quality check using the FASTQC tool and aligned to the mouse reference genome (mm10) using bowtie2 (184.108.40.206) [63, 67]. Duplicate reads were filtered using Picard MarkDuplicates (2.18.11-SNAPSHOT) and further reads were also filtered based on MAPQ cut-off < 10 . MACS2 narrow peak calling program were used to call the peak in each sample against Input ChIP as background . Peak summits called by macs2 in each sample were extended to ± 1kb and overlapping peaks were merged. Consensus peak sets for H3K27ac ChIP were generated after merging 1kb extended peak sets from each condition using bedops. To further filter down the peaks, we performed differential acetylation analysis using getDifferentialPeaks and filtered only the regions that are having 2 fold increase or decrease in acetylation activity after CpG, pIC and CpG + pIC stimulation. Next to perform comparison of differentially enriched H3K27ac enriched regions across different condition we extracted raw counts using featureCounts function from Rsubread R package (1.34.7) and performed differential analysis using DESeq2(1.24.0) Genomic regions were filtered based on variance stabilized value (vst) using cut-off value of 100 (sum of vst value across all the conditions). Total differentially acetylated regions were then used to carry out Loglikehood ratio tests (lrt) in DESeq2 to get condition specific acetylated regions. Genomics regions from clusters were merged based on condition specific enrichment and defined into four clusters as CpG specific, pIC specific, common CpG-pIC and enhancer having decreased activity after stimulation.
Super Enhancer analysis
Super enhancer analyses were performed on macs2 called H3K27ac peaks using ROSE [48, 49]. Peaks were stitched based on the default 12kb distance between the two peaks without exclusion of TSS. SE regions were annotated to the nearest gene using mm10 UCSC annotation from bioconductor and ChIPseeker R package [49, 70].
Overlap of NCoR1 and H3K27ac genomics regions.
Differential NCoR1 binding clusters were overlapped with differential H3K27ac binding sites and significance of overlap were calculated using OLOGRAM (v1.2.1) .
Transcription factors Motif Enrichment analysis
Transcription factor motif enrichment analysis on NCoR1/H3K27ac bound genomic regions was performed using findMotifs.pl/findMotifsGenome.pl program of HOMER. Default motif lengths of 8,10,12 were selected for enrichment and vertebrate options were used as known motif sets.
Weighted gene co-expression network analysis (WGCNA)
Gene co-expression analysis of a total differentially expressed genes across comparison of samples from multiple stimulation in control and NCoR1 KD conditions were performed using WGCNA . According to the method described in the WGCNA tutorial, soft power threshold was calculated using total sample and Topological overlap map (TOM) was generated. Hierarchical clustering of genes were performed based on dissimilarity of TOM and the dendrogram was cut using following parameters (minModuleSize = 30, ds = 2, cutHeight = 0.98, dthresh = 0.15) to generate co-expression modules. Pathway enrichment analyses were carried out for each module using the Reactome database from MSigDB. We identified green and darkred two important modules enriched for immune response related pathways. Out of total known TFs and coregulators (n = 1787) in mouse 131 were found to be significantly associated with green, darkred and salmon modules. Gene-gene interaction networks were extracted for these modules. The TFs and co-regulators were ranked in each stimulation condition based on significance of association of identified target from gene-gene interaction network and the target differentially expressed in each condition. Further, known protein-protein interactions of identified gene-gene co-expression networks were validated using StringDB in Cytoscape (V.3.7.1) [77, 78].
ChIP-seq analysis of publicly available datasets.
SRA files of transcription factor PU.1, JunB, cRel, IRF3 ChIP-seq data at 0hr, 90min CpG and pIC performed in the MutuDC1940 (GSE106730) and PU.1, IRF1, IRF4, RelA, RelB, Rel, JunB, Stat1 and Stat3 ChIP-seq data at 0hr and LPS stimulation performed in bone marrow derived dendritic cells (GSE36104) were downloaded from NCBI Gene Expression Omnibus. Raw fastq files were extracted using the fastq-dump program of SRA Toolkit (2.9.2) . Reads were aligned to mouse reference genome mm10 using bowtie2 (220.127.116.11) and reads having mapping quality (MAPQ) < 10 were filtered out to carry out downstream analysis. Peak calling for mutuDC cell line ChIP data was performed using MACS2. Peaks were filtered against ENCODE mm10 blacklisted regions . Genomic regions for each TF data identified in mutuDC were overlapped with NCoR1 genomic regions overlapping with H3K27ac as well as associated with CpG/pIC or common CpG-pIC specific genes using bedtools. The bedGraph file for each ChIP-seq data was generated using the makeUCSCfile program of HOMER. TFs/H3K27ac Enrichment heatmap ± 2kb to NCoR1 peak center were generated using deepTools2 (3.5.1) . hrs
Empty and NCoR1 KD DC line were plated at 2*106 in each well of 6 well plate and treated with poly I:C at 5ug/ml (invivogen TLRL-pic-5) and CpG ODN at 1ug/ml (invivogen 1826) for 2 and 6h separately, followed by lysis in RIPA buffer (0.5 M EDTA, 1 M Tris-Cl pH7.5, 1 M NaCl, 200 mM PMSF, 10% NP-40, 10% SDS, 5% sodium deoxycholate, 1 M sodium orthovanadate and 1X Roche protease inhibitor). Cells were sonicated in Bioruptor (Diagenode) with setting of high amplitude and 30s ON & 30s OFF for 10 cycles. After complete lysis, samples were processed for protein quantification by BCA protein assay kit (Bio-Rad). We loaded the samples at 50-80ug concentration and SDS-PAGE was performed, either in 10% gel for IRF3 or 15% gel for ISG-15, at 80–100 volts. Further we transferred the gel onto a nitrocellulose membrane and probed with phospho-IRF3 (cst 29047S) or ISG-15 (sc166755) or tubulin (cst 2146S). Once p-IRF3 was developed we stripped the blot and reprobed the same blot with total-IRF3 (cst 4302s). Finally once again the blot was stripped and probed with loading control- tubulin. We developed the blot on BioRad Chemidoc. Densitometric analysis was performed using ImageJ software.
IRF3 ChIP and qPCR
ChIP for IRF3 was performed according to a well-established method used by Raghav and Deplanke’s lab . For performing ChIP assays we seeded 40*106 cells in 150 mm × 25 mm hrsdishes. The cells were either left unstimulated or stimulated with polyIC at 5µg/ml (invivogen TLRL-pic-5) for 2h. Cells were then crosslinked with 1% formaldehyde (Sigma 252549) at room temperature for 10 minutes and then the reaction was quenched using 2.5M glycine (Sigma 50046) for 5 minutes at room temperature. The pertishes were then placed on ice and cells were scraped using 1X PBS and collected in falcon tubes. The tubes were centrifuged and pellets were washed twice with chilled 1X PBS. Finally the pellets were stored in -80 degrees for future use.
On the day of performing the ChIP experiment pellets were taken out and thawed on ice. The cell pellet was then subjected to lysis by using Nuclear extraction buffer (Hepes-KOH pH7.5, NaCl, EDTA pH 8.0, glycerol, NP-40, triton-X supplemented with protease and phosphatase inhibitors) for 10 minutes at 4 degree with constant mild shaking. The cells were then centrifuged at 2500rpm for 5 minutes and pellets collected. Next the isolated nuclei were subjected to a protein extraction buffer (NaCl, EDTA, Tris-Cl pH 8.0, supplemented with protease and phosphatase inhibitors) for 10 minutes at room temperature with constant mild shaking. The tubes were centrifuged and pellets collected. Finally the nuclei were then subjected to chromatin extraction buffer (EDTA, Tris-HCl pH 8.0, triton-X supplemented with protease and phosphatase inhibitors) and incubated for 10 minutes on ice. The extract was then sonicated using bioruptor (diagenode) with the following settings: 30sec on, 30sec off, 35–40 cycles. Once the desired fragment size was obtained (200-400bp) we quantified the chromatin and 150ug chromatin was used per ChIP. The chromatin was resuspended in ChIP dilution buffer (EDTA, TriS-HCl pH 8.0, triton-X, NaCl supplemented with protease and phosphatase inhibitors). 1% input was kept separately in this step.
BSA blocked recA-sepharose beads (invitrogen 101142) was used for pull down. The pre-blocked sepharose beads were used 80ul/IP and incubated with chromatin for 2hs at 4 degree with rotation for any non-specific chromatin removal. The beads were then centrifuged and the unbound supernatant was then incubated with 5ul of total-IRF3 (cst 4302s) and mAb IgG rabbit (cst 3900s) overnight. Next day the chromatin bound antibody complex was incubated with BSA pre-blocked sepharose beads for pulling down the bound complex for 2hs. After 2h incubation the tubes were centrifuged and supernatant discarded. The pellet was then washed with the following buffers for twice each: low salt buffer (Tris-HCl pH 8.0, NaCl, EDTA, SDS, triton-X), high salt buffer (Tris-HCl, NaCl, EDTA, SDS, triton-X), lithium chloride wash buffer (Tris-HCl pH 8.0, LiCl, EDTA, NP-40, sodium deoxycholate), and TE wash buffer (Tris-HCl and EDTA). Finally the chromatin was eluted in an elution buffer (sodium bicarbonate, SDS) and eluted from beads by constant shaking at room temperature for 15 minutes. The tubes were then centrifuged and eluted supernatant was collected. The supernatant was reverse crosslinked using NaCl overnight with constant shaking at 65 degrees. Next day the reverse crosslinked chromatin was subjected to proteinase-K and RNase treatment and finally PCR purified using Qiagen PCR purification kit (Qiagen 28006).
For experimental validation of IRF ChIP, ChIP-qPCR was performed at 0h and 2h CpG activated control and NCoR1 KD DCs. Enrichment of these factors at randomly selected ChIP-seq positive genomic regions/genes was calculated in comparison to negative control genomic regions. Three independent ChIP experiments were performed for IRF3 ChIP-qPCR. Fold enrichment at positive genomic regions was calculated relative to negative control regions. The ChIP primers used are listed in the table below. The p-value for enrichment significance was calculated using two-tailed paired Student's t-test and error bars depicted SEM in the fold change error in enrichments observed in different biological replicates.
Table showing list of sequence of primer used for IRF3 qRT-PCR
ELISA was performed to estimate the IFNβ levels secreted in the cell culture supernatants according to the manufacturer's protocol (ab25263). Briefly, supernatants were collected from control and NCoR1 depleted cDC1s after 2 and 6h of pIC stimulation and stored at − 80°C in small aliquots until analysis. The supernatants were diluted to 1:4 using sample diluent and then used for the assay. All standards and samples were assayed in duplicates.
Sendai virus (SeV) infection in DCs
Control and NCoR1 KD CD8α + DCs were seeded in a flat bottom 96 well plate at a density of 4x104 cells/well. Cells were left overnight for acclimatization and proper adherence. Next day cells were stimulated with TLR3 specific synthetic ligand polyI:C (pIC) at 5µg/ml for 6h at different dilutions. After pre-incubation, SeV-tomato red infection was carried out for an additional 16h. Percent SeV infection and MFI was represented in the flow cytometer at 594nm emission wavelength.
Co-culture of DCs with CD8 + T-cells for assessing T-Cell proliferation and cytotoxicity
DC-T-cell co-culture experiments were performed according to well established protocol [82–84]. Naïve CD8 + T-cells were purified from the spleen of TCR-transgenic OT-I mice using CD8 + T-cell isolation kit. NCoR1 KD and control cDC1 were seeded at a density of 10,000 cells/well in round bottom 96 well plates followed by pulsing with OVA peptide (257–264) /OT-I was used at 5nM overnight. Further DCs were stimulated with 5µg/ml of pIC for 6h at different dilution (1:1, 1:10, 1:100). After 2h, media was aspirated and fresh media containing purified OT-I T-cells were added at the density of 100,000 cells/well. Then T-cell proliferation and cytotoxicity of T-cells were analyzed by FACS. Proliferation was measured using an amine based dye (eFluor 670). The rate of T-cell proliferation was inversely proportional to the median fluorescence intensity (MFI) measured in FACS after 72h of co-culture. For cytotoxic T-cell differentiation profiling after 96h, the co-cultured T-cells were re-stimulated with PMA (10 ng/mL), ionomycin (500 ng/mL) and Brefeldin-A (10µg/mL) for 5h. Fluorochrome conjugated antibodies specific to cytotoxic T-cell (Perforin [eBioscience:12-9392-80], IFN-γ [eBioscience :25-7311-41], Granzyme-B [Biolegend:515405]) were checked in CD3+ CD8+ CD44+ [eBioscience:48-0441-82] effector T-cells using respective fluorescence minus one (FMO) controls.