Depending on the environment, dendritic cells (DCs) may become active or tolerogenic, but little is known about whether heritable epigenetic modifications are involved in these processes. Here, we have found that epigenetic histone modifications can regulate the differentiation of human monocyte-derived DCs (moDCs) into either activated or tolerized DCs. The inhibition or silencing of methyltransferases or methylation-associated factors affects the expression of multiple genes. Genome mapping of transforming growth factor (TGF-β)- or lipopolysaccharide (LPS)-associated H3K4 trimethylation (H3K4me3) and H3K27 trimethylation (H3K27me3) demonstrated the presence of histone modification of gene expression in human TGF-β- or LPS-conditioned moDCs. Although the upregulated or downregulated genes were not always associated with H3K4me3 and/or H3K27me3 modifications in TGF-β-conditioned (tolerized) or LPS-conditioned (activated) moDCs, some of these genes may be regulated by the increased and/or decreased H3K4me3 or H3K27me3 levels or by the alteration of these epigenetic marks, especially in TGF-β-conditioned moDCs. Thus, our results suggested that the differentiation and function of moDCs in tumor and inflammation environments are associated with the modification of the H3K4me3 and K3K27me3 epigenetic marks.
Dendritic cells (DCs) are antigen (Ag)-presenting cells, with a unique capacity to stimulate naive T cells and initiate primary immune responses. However, DCs also have critical roles in the induction of central and peripheral immunological tolerance and the regulation of the types of T-cell immune responses.1, 2 The diverse functions of DCs in immune regulation not only depend on the heterogeneity of DC subsets but also on their functional plasticity.1, 2 DCs express a variety of pathogen recognition receptors, such as Toll-like receptors. The detection of damage- or pathogen-associated molecular patterns, such as lipopolysaccharides (LPSs), by tissue-resident DCs initiates DC maturation and migration to the regional lymph nodes.3, 4 Upon exposure to transforming growth factor (TGF)-β, DCs become tolerogenic and can thereby suppress T-cell alloreactivity5 and induce Th2 or T regulatory responses.3, 4 Studies have shown that the expression of genes coding costimulatory molecules, cytokines, chemokines and their receptors is remarkably different in the conditioned DCs; those DCs activated by LPS express higher levels of the costimulatory molecules, such as CD80 and CD86, and the Ag-presenting molecule major histocompatibility complex (MHC) II3, 4 than those conditioned with TGF-β.6
Previous reports6, 7, 8 and microarray data from the NCBI-GEO database (LPS-treated DC gene expression from the GSE10316 record and TGF-β-treated DC gene expression from the GDS2940 record) have shown that both LPS and TGF-β induce the expression of hundreds of genes in human monocyte-derived DCs (moDCs). The responses induced by treatment with LPS or TGF-β most likely employ different regulatory requirements that reflect their functions. As the expression of different classes of genes can be induced by the same receptor, differential regulation occurs via a gene-specific mechanism. Although multiple mechanisms may be involved in the regulation of gene expression in different kinds of cells,9 epigenetic modification has served as an important mechanism in regulating the cellular differentiation.10 Indeed, a genome-wide promoter of histone modifications has been identified in human moDCs.11 In LPS-conditioned macrophages, gene-specific regulation occurs at the level of chromatin and includes nucleosome remodeling and covalent histone modifications.12 Histone deacetylase (HDAC) inhibition can also affect the phenotype and function of human moDCs.13 Global mapping of H4K4me3 and H3K27 trimethylation (H3K27me3) has revealed the specificity and plasticity of lineage fate determination in differentiating CD4+ T cells10, 14 and in memory CD8+ T cells.14 Thus, the covalent modifications of histone N-terminal tails, such as methylation, can act to regulate chromatin states.15
moDCs develop from peripheral blood monocytes under the influence of granulocyte-macrophage colony-stimulating factor (GM-CSF) alone or GM-CSF with interleukin-4 (IL-4),16, 17 and are a subset of DCs that are involved in inflammatory processes and infection clearance.3, 18, 19 These moDCs could become either active or tolerogenic after exposed to different stimuli. Here, we have used chromatin immunoprecipitation (Chip) and Chip-sequencing (Chip-seq) to characterize the modification of H3K4 trimethylation (H3K4me3) and H3K27me3 in LPS- and TGF-β-conditioned moDCs. We have found that during the transition of immature moDCs into activated moDCs (LPS conditioning) and tolerized moDCs (TGF-β conditioning), changes in the modification of H3K4me3 or H3K27me3 as well as alteration in these epigenetic marks have a role in the resulting up- or downregulation of genes in these treated cells.
Histone methylation is involved in the regulation of the gene expression in human moDCs
Histone methyltransferases have a critical role in epigenetic modification during the cellular differentiation.20 To investigate the effect of methylation on the expression of TGF-β- and LPS-associated genes, we first employed the methyltransferase inhibitor adenosine dialdehyde (an adenosine analog and S-adenosylmethionine-dependent methyltransferase inhibitor) to observe the effect on the gene expression in human moDCs. As shown in Supplementary Figure S1, adenosine dialdehyde clearly promoted the expression of the surface molecules CD14, CD11C, CD80, CD86 and MHC-DR, the expression of the cytokines and chemokines IL-1A, IL-12A, interferon alpha 6 (IFNA6), tumor necrosis factor (ligand) superfamily, member 10 (TNFSF10), chemokine (C-C motif) ligand (CCL)3, CCL4, chemokine (C-C motif) receptor (CCR)7 and chemokine (C-X-C motif) ligand (CXCL)16, and also the expression of the transcription factors signal transducer and activator of transcription (STAT)3, STAT4, nuclear factor (NF)-κB1 and NF-κB2 in human moDCs. Euchromatic histone lysine N-methyltransferase 2 (EHMT2) is a H3K9- and H3K27-specific histone methyltransferase21 that is remarkably downregulated in human moDCs after exposure to LPS (LPS-treated DC gene expression, GSE10316 record and Supplementary Figure S2), and has the potential to affect the expression of multiple genes. Indeed, the silencing of EHMT2 with demonstrated transfection (Figure 1a) was found to promote the expression of surface molecules (CD14, CD11c, CD80, CD86 and MHC-DR), cytokines (IL-1A, IL-12A, IFNA and TNFSF10), chemokines (CCL3, CCL4, CCR7 and CXCL16) and transcription factors (STAT3, STAT4, NF-κB1 and NF-κB2; Figure 1b). Importantly, silencing EHMT2 also affected the Ag-presenting function of moDCs (Supplementary Figure S3). Others genes such as enhancer of zeste homolog 1 (EZH1), chromobox homolog 5 (CBX5) and SET domain-containing 6 (SETD6), which are associated with the methylation of histones,22, 23, 24 were also involved in the regulation of gene expression in moDCs. The silencing of SETD6 was found to downregulate the expression of IL-1A and upregulate that of STAT3, whereas the expression of NF-κB1 was not remarkably altered (Figures 1c and d). The silencing of EZH1 was found to downregulate IL-1A expression and upregulate both STAT3 and NF-κB1 expression, but the silencing of CBX5 downregulated both IL-1A and NF-κB1 expression and upregulated STAT3 expression (Figures 1c and d). Taken together, these results suggest that histone methylation may be involved in the gene regulation of human moDCs.
Genome-wide maps of H3K4me3 and H3K27me3 modification in TGF-β- and LPS-conditioned DCs
To reveal the epigenetic features present during the process of differentiation of moDCs in response to different environments, we generated global maps of H3K4me3 and H3K27me3 modification using the Chip-seq approach. The LPS-conditioned moDCs had typical morphology of DC; they expressed higher levels of costimulatory molecules, such as CD11C, CD40, CD80 and CD86, and Ag-presenting molecules, MHC class I and class II (Supplementary Figure S4). Consistent with previous reports,6 TGF-β-conditioned DCs exhibited an immature phenotype; they expressed lower levels of CD40, CD80, CD86, MHC I, MHC II and CD11C than did LPS-conditioned moDCs (Supplementary Figure S4). Similar to unconditioned moDCs, the Ag-presenting function of these human LPS- or TGF-β-conditioned moDCs could also be altered after silencing EHMT2, suggesting that histone methylation may be also involved in the gene regulation of human LPS- or TGF-β-conditioned moDCs (Supplementary Figure S3). Using specific antibodies, we next immunoprecipitated H3K4me3-or K3K27me3-associated genomic DNAs from TGF-β- or LPS-conditioned moDCs and control moDCs. The H3K4me3 and H3K27me3 peak (island, Chip-seq-enriched region) distribution near the transcription start site (±5 kb) of each annotated RefSeq gene was analyzed. H3K4me3 islands were enriched in the region closest to the transcription start site (from +2.5 kb to −100 bp; Figure 2a), whereas H3K27me3 peaks were enriched in the region upstream of the transcription start site sites (from +2.5 to +5 kb; Figure 2a).
To examine the overall H3K4me3 and H3K27me3 distribution, we divided the human genome into four types of regions, which included the proximal promoter (2 kb upstream and downstream of the transcription start site), the exons, introns and intergenic regions. These regions were based on the annotations of ‘known genes’ in the UCSC genome browser (http://genome.UCSC.edu). H3K4me3 islands were found in ∼20% of the proximal promoter regions of TGF-β-treated moDCs (23%) and LPS-treated moDCs (21%), whereas H3K27me3 islands were only found in 3% of the proximal promoter regions of both TGF-β-treated moDCs (3%) and LPS-treated moDCs (3%). The examination of H3K4me3 and K3K27me3 islands within gene bodies revealed that ∼1% of these was enriched for H3K4me3 or K3K27me3 islands (Figure 2b). Most of H3K4me3 and H3K27me3 modifications were located in the introns and the intergenic regions (Figure 2b). Lineage- specific H3K4me3 and H3K27me3 marks in LPS- or TGF-β-conditioned DCs were remarkably different than those in control moDCs. The number of lineage-specific H3K4me3 islands was significantly greater in TGF-β-conditioned moDCs than in LPS-conditioned moDCs, which indicated that the active genes in TGF-β-conditioned moDCs were modified by H3K4me3. However, the number of lineage-specific H3K27me3 islands was similar between LPS- and TGF-β-conditioned moDCs (Figure 2c). During the differentiation of the immature DCs into LPS-activated DCs, the proportion of genes with H3K4me3 marks alone decreased from 13 to 6.5% (Figure 2c).
We also performed gene ontology (GO) analyses for the H3K4me3- and H3K27me3-marked genes. On the basis of the genes marked by H3K4me3 and/or H3K27me3 (Supplementary Table S1), it was clear that these epigenetic modifications decorate many genes with functions related to differentiation, development and transcriptional regulation (Supplementary Figure S5 and Supplementary Table S2). As many as 30–60% of the genes in both the TGF-β-conditioned DCs and LPS-conditioned DCs were modified by H3K4me3, but less genes were modified by H3K27me3 (Supplementary Figure S5 and Supplementary Table S2).
Global analysis of H3K4me3 and H3K27me3 modification in TGF-β- and LPS-conditioned human moDCs
In addition to the microarray data on TGF-β-conditioned DCs and the global analysis of the Chip-seq in TGF-β-conditioned DCs, we next co-analyzed the modification of H3K4me3 and H3K27me3 on the upregulated, downregulated and unaltered genes, which are originated from the microarray data of TGF-β-conditioned moDCs (NCBI-GEO Database: TGF-β-treated DC gene expression, GDS2940 record), and the data are shown in Supplementary Table S3. An analysis of the more highly expressed genes in TGF-β-conditioned DCs revealed that almost half (50%) have the H3K4me3 modification on their promoter, whereas only 10% have the H3K27me3 mark (Figure 3a). There was a highly significant correlation between the H3K4me3 marks and the high expression level of genes related to TGF-β activation, such as that of NFKB2, HSP90AB1, IL-32, TGFBR1, PLCB3, PDIA3, NFKB1, PRKACA, ACVR1B, MED24 and IL-1RAP (Figure 3b, only H3K4me3 marks of NFKB2, HSP90AB1 and IL-32 are shown) were seen in TGF-β-conditioned DCs. These genes are involved in the PPARα/RXRα-activated pathway that may limit inflammatory responses. Conversely, the presence of H3K27me3 marks in TGF-β-conditioned DCs was correlated with lower levels of immune gene expression, such as that of RYR3, HDAC9, DCC, PLCB4 and PRKCH were related to H3K27me3 marks (Figure 3c, only H3K27me3 marks of RYR3 are shown). However, 40% of the upregulated genes, 50% of the downregulated genes and 20% of the unaltered genes were not modified by H3K4me3 or H3K37me3 (Figure 3a).
We also analyzed the modification of H3K4me3 and H3K27me3 on the upregulated, downregulated and unaltered genes, which are originated from the microarray data of LPS-conditioned DCs (NCBI-GEO Database, LPS-treated DC gene expression, GSE10316 record), and the data are shown in Supplementary Table S4. As shown in Figure 4, for LPS-conditioned DCs, 20% of the upregulated genes were modified by H3K4me3, and 8% of the downregulated genes were modified by H3K27me3 in LPS-conditioned DCs (Figure 4a). Similar to TGF-β-conditioned DCs, many genes were not modified by either H3K4me3 or H3K27me3. The JAK/STAT family of transcription factors is critical for DC function. We found that signaling molecules in this signal pathway were modified with H3K4me3 (Figure 4b, only H3K4me3 marks of STAT3, JAK3 and STAT5A are shown). Interestingly, the upregulation of genes such as PTPN1, STAT3, JAK3, STAT5B, STAT4, STAT5A and JAK1 in the JAK/STAT signal pathway was not significantly related to increased H3K4me3 modification in these LPS-conditioned DCs (Figure 4b), which implied the involvement of other mechanisms in the regulation of the expression of these genes. Downregulated genes such as VAV3, PAK1, PRKCA and GNB5, which are associated with phagocytosis and pinocytosis, were indeed modified by the increased H3K27me3 in LPS-conditioned DCs (Figure 4c, only H3K27me3 marks of VAV3 are shown). Additionally, in the LPS-conditioned DCs, STAT3, STAT5B and STAT5A genes were marked in their promoter regions with H3K4me3, but these genes did not have substantial H3K27me3 signals, indicating these active genes are mainly modified by H3K4me3.
Characterization of H3K4me3 and H3K27me3 modification and the expression of the surface molecules, cytokines, chemokines and their receptors
CD14 and CD83 are the critical markers of the process of the DC differentiation. Surface molecule CD14, which is highly expressed by monocytes, is downregulated during the differentiation after exposure to LPS but is regulated after exposure to TGF-β (Figure 5a). Consistent with other reporters,11 the promoter of the CD14 genes had multiple H3K4me3 marks with high CD14 expression (Figure 5a), where these marks were lost after exposure to LPS, which is in agreement with transcriptional downregulation of this surface molecule in LPS-conditioned DCs. However, TGF-β-conditioned DCs maintained these H3K4me3 marks (Figure 5a) at the CD14 promoter. These suggest that H3K4me3 modification has an important role in regulating the expression of CD14. The expression of CD83, a mature marker that is increased following DC maturation, was lower in TGF-β-conditioned DCs than LPS-conditioned DCs. However, the promoter region of CD83 did not show decreased H3K4me3 marks in TGF-β-conditioned moDCs as compared with untreated or LPS-conditioned moDCs. Interestingly, there was remarkably increased H3K27me3 modification near the CD83 locus in TGF-β-conditioned DCs (Figure 5a), which implied that the modification of H3K27me3 marks has a role in regulating the expression of CD83.
Consistent with previously published data,11 genes active in human moDCs often exhibit H3K4me3 modification. As shown in Figures 5 and 6, the genes for surface molecules CD14, CD83, CD86 and HLA-DRB1 (only H3K4me3 and H3K27me3 marks of CD40 and CD86 are shown) as well as those for the cytokines/chemokines and their receptors, such as IL-1A, IL-12A, TNFSF10, CCL3, CCL4, IFNGR2 and CCR7, were modified by H3K4me3 in their promoter regions. In addition, inconsistent with previous data,11 we found that the CD40 gene in moDCs was also remarkably modified by H3K4me3 in its promoter region. The genes for the cytokines IL-2, IL-5, IL-7, IL-15 and IL-16, which are only expressed at low levels by DCs, showed very minor H3K4me3 modification but showed strong K3H27me3 modification (Supplementary Figure S7). Thus, our results confirm that H3K4me3 modification is related to active genes, and that H3K27me3 marks are associated with repressive modification.
LPS as a potent inducer of DC maturation can promote the surface expression of costimulatory and Ag-presenting molecules, whereas TGF-β attenuates many of these processes.25 Compared with unconditioned DCs, TGF-β significantly downregulated the expression of the costimulatory molecules CD80, CD86 and CD40 and Ag-presenting molecules (Figure 5), whereas the expression of these molecules was higher in LPS-conditioned DCs. Importantly, the genes for these costimulatory molecules were modified with H3K4me3 and H3K27me3 upon exposure to LPS or TGF-β (Figure 5). TGF-β and LPS treatment not only affects the expression of costimulatory and Ag-presenting molecules but also alters the expression of cytokines, chemokines and their receptors.6, 26 As shown in Figure 6, exposure to TGF-β and LPS affected the expression of IL-1A, IL-12A, TNFSF10, CCL3, CCL4, IFNGR2 and CCR7. In a similar manner as the genes for costimulatory molecules, the loci of the genes for the cytokines/chemokines and their receptors, including IL-1A, IL-12 A and IFNGR2, which were downregulated following TGF-β treatment, were modified by increased H3K27me3. The genes that were upregulated following TGF-β treatment, such as IL-32, were modified by increased K3K4me3 (Figure 3). However, although genes such as TNFSF10, CCL3, CCL4 and CCR7 were modified by both increased H3K4me3 and H3K27me3, their expressions were not remarkably altered (Figure 6).
Interestingly, certain upregulated costimulatory molecules, cytokines/chemokines and their receptors, such as CD40, CD80, CD86 and HLA-DR, were not modified by increased H3K4me3 in LPS-conditioned DCs (Figure 5). However, the position of H3K4me3 and/or H3K27me3 marks was altered in both upregulated and downregulated genes from LPS-conditioned moDCs, such as those for CD83, IL-1A, CCL3, CCL4 and IFNGR2 (Figures 5 and 6). This suggests that exposure to LPS induces both the upregulation and downregulation of costimulatory molecules, Ag-presenting molecules, cytokines/chemokines and their receptors that may be partially dependent on the alteration of H3K4me3 and/or H3K27me3 position.
Transcription factors are modified by H3K4me3 and H3K27me3 in both LPS- and TGF-β-conditioned DCs
Transcription factors have a critical role in orchestrating the differentiation and function of DCs. In particular, the Rel/NF-κB family of transcription factors, which includes RelA, c-Rel, RelB, NF-κB1 (p50 and its precursor p105) and NF-κB2 (p52 and its precursor p100), has a central role in the immune system and determines the expression of multiple cytokine and innate immune responses.27 We finally investigated histone modifications on genes associated with transcriptional processes. The expressed genes encoding transcription factors NFKB1, NFKB2, NFATC1, STAT3, STAT4, STAT5A and STAT5B had H3K4me3 modifications at their promoters, and a few of these had H3K27me3 modification (Figures 3 and 4 and Supplementary Figure S6). TGF-β-induced expression of the NFKB2, STAT5A and STAT5B genes in conditioned moDCs was also modified by H3K4me3 marks (Figure 3 and Supplementary Figure S6). Conversely, TGF-β downregulated the expression of the NF-AC1 gene in conditioned moDCs, which was clearly modified with increased H3K27me3 marks (Supplementary Figure S6). However, similar to other genes in LPS-conditioned moDCs, upregulated transcription factors such as NFKB1, NFKB2, NFATC1, STAT3, STAT4, STAT5A and STAT5B were not modified with increased H3K4me3 marks (Figure 4 and Supplementary Figure S6).
In this study, we have shown that human moDCs can be regulated by epigenetic histone modifications during their differentiation into activated DCs or tolerized DCs. Meanwhile, we have mapped the TGF-β- and LPS-associated H3K4me3 and H3K27me3 marks across the genome. We have demonstrated that the upregulation of certain costimulatory molecules, cytokines/chemokines as well as their receptors is related to the increased H3K4me3 marks, whereas the downregulation of other genes is associated with H3K27me3 modifications in TGF-β- and LPS-conditioned DCs. However, the expression of the LPS- or TGF-β induced genes is not always dependent on the increased level of H3K4me3, and the expression of the downregulated genes is not always modified by H3K27me3. Particularly, the expression of costimulatory molecules, cytokines and chemokines and their receptors after exposure to LPS is often independent of the modifications of the H3K4me3 marks, but it is often associated with the alteration of the H3K4me3 and/or H3K27me3 marks.
DCs are extremely versatile Ag-presenting cells involved in the initiation of both innate and adaptive immunity and also in the maintenance of self-tolerance. These diverse and almost contradictory functions are dependent on the plasticity of these cells to allow them to undergo a complete genetic reprogramming in response to different stimuli.1, 2 Our results showed that methyltransferase inhibitors not only affect the expression of certain critical genes, but more importantly also alter DC functions, suggesting that epigenetic modifications are involved in the plasticity of human moDCs in different environments. Indeed, upregulation of certain costimulatory molecules, cytokines/chemokines as well as their receptors is related to the increased H3K4me3 marks, and the downregulation of other genes is associated with H3K27me3 modifications in the TGF-β- and LPS-conditioned DCs.
The basic structural unit of eukaryotic chromatin is the nucleosome, which consists of 146 bp of DNA wrapped around an octamer of four core histones (H2A, H2B, H3 and H4). Histone modifications within promoter regions have an important function in the regulation of gene expression. The majority of modifications occur at the N-terminal ends of the core histones. Studies from several model systems have determined that the methylation of lysines 4, 36 and 79 of histone H3 are typically associated with active gene expression,28, 29, 30 whereas repressive genes are associated with H3K27me3 modification. Indeed, many active genes have increased H3K4me3 marks, especially on gene promoters, whereas some repressive genes are modified by H3K27me3 marks in TGF-β- or LPS-conditioned moDCs. This is in agreement with other studies using embryonic stem cells31, 32 and moDCs,11 or CD4+ and CD8+ T cells,10, 14, 33 which have reported that H3K4me3 marks are the most characteristic modification in promoter regions. In addition, H3K4me3 and H3K27me3 marks are usually enriched at the active or inactive chromatin regions, respectively.34, 35 Our data have also shown that both H3K4me3 and H3K27me3 marks were often colocalized to certain genomic regions.
The decreased K3K4me3 marks and the increased expression of genes with or without the H3K27me3 mark in the LPS-conditioned DCs suggest that ex vivo differentiation in the presence of Toll-like receptor ligands induces a global shift toward less active chromatin in the presence of Toll-like receptor ligands. Although our results have shown that the expression of some costimulatory molecules, cytokines and chemokines and their receptors is associated with increased H3K4me3 marks in TGF-β-conditioned moDCs, the upregulation of some of these molecules was not associated with increased H3K4me3 marks in LPS-conditioned moDCs. Others have also found that transcriptionally active genes can have reduced H3K4me3 marks during DC differentiation.11 Some genes in human moDCs have been shown to lose their H3K4me3 mark independent of their gene expression,11 which suggests that both the H3K4m3 and H3K27me3 marks are unstable and are prone to either a loss of differentiation state or a gain of other permissive mark such as H3K9me3.36, 37 Additionally, an alteration of the H3K4me3 and/or H3K27me3 positions is often found in the upregulated and downregulated genes from LPS-conditioned moDCs. H3K4me3 at certain chromatin loci may also prevent aberrant gene expression or modulate transcriptional response.38 However, additional genome-wide localization and functional studies will provide important new insights into the role of other epigenetic marks, chromatin-modifying enzymes and chromatin remodelers.
Materials and methods
Preparation and isolation of human moDCs
Immature DCs were prepared from the buffy coats of healthy donor samples obtained from the blood bank of Tianjin, according to our previously described method.39 Briefly, peripheral blood mononuclear cells were separated by Ficoll-Paque gradient centrifugation, and CD14+ cells were purified using antibody-coated microbeads and magnetic separation. Selected CD14+ cells were cultured for 5 days at a concentration of 2 × 106 ml−1 in the presence of GM-CSF (500 U ml−1) and IL-4 (500 U ml−1; R&D Systems, Minneapolis, MN, USA) for the generation of immature moDCs. Immature moDCs were matured by exposure to LPS (100 ng ml−1ml−1, InvivoGen, San Diego, CA, USA) or tolerized by exposure to TGF-β (10 ng ml−1 R&D Systems) for 48 h, unless otherwise stated.
Flow cytometric analyses
For surface staining, cells were collected in ice-cold phosphate-buffered saline and incubated with the indicated phycoerythrin- or fluorescein isothiocyanate-labeled antibodies. For each analysis, isotype-matched control monoclonal antibodies were used as negative controls. The phenotypes of human moDCs were analyzed using phycoerythrin- or fluorescein isothiocyanate-labeled anti-CD14 (61D3), anti-CD83 (HB15e), anti-CD11c (3.9), anti-CD86 (IT2.2), anti-CD80 (2D10.4), anti-CD40 (HB14) and anti-CD11b (ICRF44), which were purchased from Pharmingen (San Diego, CA, USA).
For small interfering RNAs (siRNAs) and negative control oligonucleotides, cells were transfected with the indicated oligos (100 nM) using Entranster-R (Engreen Biosystem, Beijing, China) according to the manufacturer's instruction. The transfection rate of the human moDCs was as high as 85%. For validation of the siRNA(s), human moDCs were cultured in six-well plate. The following day cells were transfected in fetal calf serum-free medium. One day post transfection, the cells were treated with TRIzol (Invitrogen, San Diego, CA, USA, 15596-026) for RNA extraction.
RNA isolation and quantitative real-time PCR
Total RNA was extracted using the TRIzol reagent (Invitrogen, 15596-026), and reverse transcription was carried out with Superscript III (Invitrogen) according to the manufacturer's protocol. To measure gene expression, quantitative real-time PCR was performed using the Quantitect SYBR PCR kit with a specific set of primers according to the manufacture’s recommendations (Qiagen, Valencia, CA, USA). Amplification of glyceraldehyde-3-phosphate dehydrogenase mRNA for gene was performed for each experimental sample as an endogenous control for gene expression. Fold changes were calculated using the ΔΔCt method according to the manufacturer’s protocol (Applied Biosystems, Foster City, CA, USA). All reactions were run in triplicate, and the primer sequences are listed in Supplementary Table S5.
siRNAs for SETD6, CBX5, EZH1, EHMT2 and control oligos were purchased from Guangzhou Ribobio, Guangzhou, China, and were used to transfect cells with Entranster-R according to the manufacturer's recommendations (Engreen Biosystem). EHMT2 siRNA of the sequence 5-CATGACTGCGTGCTGTTATTC-340 was synthesized by Guangzhou Ribobio. CBX5, EZH1, JHDM1D and SETD6 siRNAs were designed and synthesized by Guangzhou Ribobio. The following sequences of these siRNAs were used: 5-TAAACCCAGGGAGAAGTCA-3 (CBX5); 5-GCCAACATATGTTAATGAG-3 (EZH1) and 5-CGCCAATCTAGAATACTCT-3 (SETD6). The nonsilencing oligo control is an irrelevant siRNA with random nucleotides (5-ACTATCTAAGTTACTACCC-3). To determine the effect of EHMT2, SETD6, EZH1 and CBX5 knockdown on the expression of genes, the cells were transfected using different siRNAs and the expression of surface markers were analyzed after 48 h. The transcriptional levels of the genes were detected with quantitative real-time PCR. The primers used in these experiments are listed in Supplementary Table S5.
Chip, Chip-seq and bioinformatics
Chip and Chip-seq, followed by high-throughput sequencing were performed by the Research and Cooperation Division of BGI-Shenzhen, according to previously described methods.41 In brief, 3 × 107 immature moDCs (control), TGF-β-conditioned DCs or LPS-conditioned DCs were digested with micrococcal nuclease (2 U) for 10 min in a 37 °C water bath to generate native chromatin template consisting primarily of mononucleosomes. The native chromatin templates were then incubated with anti-histone H3K4me3 (Kit cat. no. GAH-8208, SABiosciences, Frederick, MD, USA), H3K27me3 (Kit cat. no., GAH-9205, SABiosciences) or rabbit control IgG (SABiosciences). Antibody-bound DNA fragments were extracted, and the DNA fragments were modified to construct a sequencing library after PCR amplification according to the manufacturer's protocol (Illumina/Solexa, San Diego, CA, USA). The specificity of the immunoprecipitation was confirmed by analyzing known genes by quantitative real-time PCR, including glyceraldehyde-3-phosphate dehydrogenase for transcriptionally active euchromatin, human MYOD1 for transcriptionally inactive euchromatin and human SAT2 for heterochromatin. Primers used were from SABiosciences. DNA fragments of ∼200 bp (mononucleosomal DNA+adaptor) were selectively recovered from 2% agarose gels for further cluster generation and sequencing with the Solexa/Illumina 1G Genome Analyzer.
The image processing and base calling were performed using the Illumina pipeline. The adapter sequences were detected using the Perl program (http://www.perl.com/pub/2006/08/03/sequence-diagrams.html). Any reads with greater than 10% Ns (indicating ambiguous residues) or over 50% bases with a quality score less than 20 were trimmed. The remaining sequences were then aligned to the hg18 human reference genome using SOAP2.21 (http://soap.genomics.org.cn/soapaligner.html), and alignments with ⩽2 errors were retained. The locations of repetitive sequences in the hg18 genome (RepeatMasker) were obtained using the UCSC (the University of California, Santa Cruz) Genome Brower (http://hgdownload.cse.ucsc.edu/goldenPath/hg18/database/). The overlap of the Chip-seq reads with these repeats was obtained by intersecting the coordinates of the RepeatMasker data with the coordinates of read alignments. The UCSC annotation data were used for general reference (http://hgdownload.cse.ucsc.edu/goldenPath/hg18/database/). To obtain a set of gene regions, the overlap between reads and each gene region is used, and the reads number for each region is counted. For the peak calling process, the intensities of control rabbit immunoglobulin G (IgG), H3K4me3 and H3K27me3 were first normalized to the random genomic background and then further normalized by subtracting control IgG intensities from the corresponding modification intensities. The enriched regions were identified using the model-based analysis of Chip-seq, a P-value cutoff of 1 × 10−5 and a high-confidence fold enrichment. This method was based on the Poisson distribution. The advantage of this model was that it could capture both the mean and the variance of the distribution. Background biases were controlled by themselves. Subsequently, the peak calling results were estimated according to the peak number and the peak length distribution. The peak distributions of gene regions were analyzed as read distributions. Finally, peak-related genes were defined by the positional relationship of the islands and genes. Several figures show the read distribution of a gene region of the gene of interest. Difference analysis on peak-related genes was used to identify genes sharing partial sequences that could provide information on the regulatory strength for each gene.
GO analysis was performed using the online GO tool (http://go.princeton.edu/). The complete set of all RefSeq genes was used as a background. Complete GO analysis results are provided (Supplementary Table S2).
Liu YJ, Kanzler H, Soumelis V, Gilliet M . Dendritic cell lineage, plasticity and cross-regulation. Nat Immunol 2001; 2: 585–589.
Palucka K, Banchereau J, Mellman I . Designing vaccines based on biology of human dendritic cell subsets. Immunity 2010; 33: 464–478.
Shortman K, Naik SH . Steady-state and inflammatory dendritic-cell development. Nat Rev Immunol 2007; 7: 19–30.
Pulendran B, Palucka K, Banchereau J . Sensing pathogens and tuning immune responses. Science 2001; 293: 253–256.
Dufter C, Watzlik A, Christ C, Jung M, Wirzbach A, Opelz G et al. Suppression of T-cell alloreactivity by gene-therapeutic modulation of human dendritic stimulator cells with TGF-beta adenoviral vectors. Transplant Proc 2001; 33: 190–191.
Fainaru O, Shay T, Hantisteanu S, Goldenberg D, Domany E, Groner Y . TGFbeta-dependent gene expression profile during maturation of dendritic cells. Genes Immun 2007; 8: 239–244.
Ahn JH, Lee Y, Jeon C, Lee SJ, Lee BH, Choi KD et al. Identification of the genes differentially expressed in human dendritic cell subsets by cDNA subtraction and microarray analysis. Blood 2002; 100: 1742–1754.
Ju XS, Zenke M . Gene expression profiling of dendritic cells by DNA microarrays. Immunobiology 2004; 209: 155–161.
Liew FY, Xu D, Brint EK, O’Neill LA . Negative regulation of toll-like receptor-mediated immune responses. Nat Rev Immunol 2005; 5: 446–458.
Wei G, Wei L, Zhu J, Zang C, Hu-Li J, Yao Z et al. Global mapping of H3K4me3 and H3K27me3 reveals specificity and plasticity in lineage fate determination of differentiating CD4+ T cells. Immunity 2009; 30: 155–167.
Tserel L, Kolde R, Rebane A, Kisand K, Org T, Peterson H et al. Genome-wide promoter analysis of histone modifications in human monocyte-derived antigen presenting cells. BMC Genomics 2010; 11: 642.
Foster SL, Hargreaves DC, Medzhitov R . Gene-specific control of inflammation by TLR-induced chromatin modifications. Nature 2007; 447: 972–978.
Song W, Tai YT, Tian Z, Hideshima T, Chauhan D, Nanjappa P et al. HDAC inhibition by LBH589 affects the phenotype and function of human myeloid dendritic cells. Leukemia 2011; 25: 161–168.
Araki Y, Wang Z, Zang C, Wood III WH, Schones D, Cui K et al. Genome-wide analysis of histone methylation reveals chromatin state-based regulation of gene transcription and function of memory CD8+ T cells. Immunity 2009; 30: 912–925.
Kouzarides T . Chromatin modifications and their function. Cell 2007; 128: 693–705.
Ardavin C, Martinez del Hoyo G, Martin P, Anjuere F, Arias CF, Marin AR et al. Origin and differentiation of dendritic cells. Trends Immunol 2001; 22: 691–700.
Sallusto F, Lanzavecchia A . Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J Exp Med 1994; 179: 1109–1118.
Leon B, Lopez-Bravo M, Ardavin C . Monocyte-derived dendritic cells formed at the infection site control the induction of protective T helper 1 responses against Leishmania. Immunity 2007; 26: 519–531.
Inaba K, Inaba M, Romani N, Aya H, Deguchi M, Ikehara S et al. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J Exp Med 1992; 176: 1693–1702.
Richly H, Lange M, Simboeck E, Di Croce L . Setting and resetting of epigenetic marks in malignant transformation and development. Bioessays 2010; 32: 669–679.
Pandey RR, Mondal T, Mohammad F, Enroth S, Redrup L, Komorowski J et al. Kcnq1ot1 antisense noncoding RNA mediates lineage-specific transcriptional silencing through chromatin-level regulation. Mol Cell 2008; 32: 232–246.
Fritsch L, Robin P, Mathieu JR, Souidi M, Hinaux H, Rougeulle C et al. A subset of the histone H3 lysine 9 methyltransferases Suv39h1, G9a, GLP, and SETDB1 participate in a multimeric complex. Mol Cell 2010; 37: 46–56.
Avdic V, Zhang P, Lanouette S, Groulx A, Tremblay V, Brunzelle J et al. Structural and biochemical insights into MLL1 core complex assembly. Structure 2011; 19: 101–108.
Smith E, Lin C, Shilatifard A . The super elongation complex (SEC) and MLL in development and disease. Genes Dev 2011; 25: 661–672.
Banchereau J, Steinman RM . Dendritic cells and the control of immunity. Nature 1998; 392: 245–252.
Teicher BA . Transforming growth factor-beta and the immune response to malignant disease. Clin Cancer Res 2007; 13: 6247–6251.
Vallabhapurapu S, Karin M . Regulation and function of NF-kappaB transcription factors in the immune system. Annu Rev Immunol 2009; 27: 693–733.
Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z et al. High-resolution profiling of histone methylations in the human genome. Cell 2007; 129: 823–837.
Vakoc CR, Sachdeva MM, Wang H, Blobel GA . Profile of histone lysine methylation across transcribed mammalian chromatin. Mol Cell Biol 2006; 26: 9185–9195.
Pokholok DK, Harbison CT, Levine S, Cole M, Hannett NM, Lee TI et al. Genome-wide map of nucleosome acetylation and methylation in yeast. Cell 2005; 122: 517–527.
Guenther MG, Levine SS, Boyer LA, Jaenisch R, Young RA . A chromatin landmark and transcription initiation at most promoters in human cells. Cell 2007; 130: 77–88.
Zhao XD, Han X, Chew JL, Liu J, Chiu KP, Choo A et al. Whole-genome mapping of histone H3 Lys4 and 27 trimethylations reveals distinct genomic compartments in human embryonic stem cells. Cell Stem Cell 2007; 1: 286–298.
Roh TY, Cuddapah S, Cui K, Zhao K . The genomic landscape of histone modifications in human T cells. Proc Natl Acad Sci USA 2006; 103: 15782–15787.
Azuara V . Profiling of DNA replication timing in unsynchronized cell populations. Nat Protoc 2006; 1: 2171–2177.
Jiang H, Shukla A, Wang X, Chen WY, Bernstein BE, Roeder RG . Role for Dpy-30 in ES cell-fate specification by regulation of H3K4 methylation within bivalent domains. Cell 2011; 144: 513–525.
Qureshi IA, Mehler MF . Impact of nuclear organization and dynamics on epigenetic regulation in the central nervous system: implications for neurological disease states. Ann N Y Acad Sci 2010; 1204 (Suppl): E20–E37.
Shechter D, Nicklay JJ, Chitta RK, Shabanowitz J, Hunt DF, Allis CD . Analysis of histones in Xenopus laevis. I. A distinct index of enriched variants and modifications exists in each cell type and is remodeled during developmental transitions. J Biol Chem 2009; 284: 1064–1074.
Pinskaya M, Morillon A . Histone H3 lysine 4 di-methylation: a novel mark for transcriptional fidelity? Epigenetics 2009; 4: 302–306.
Rongcun Y, Salazar-Onfray F, Charo J, Malmberg KJ, Evrin K, Maes H et al. Identification of new HER2/neu-derived peptide epitopes that can elicit specific CTL against autologous and allogeneic carcinomas and melanomas. J Immunol 1999; 163: 1037–1044.
Watanabe H, Soejima K, Yasuda H, Kawada I, Nakachi I, Yoda S et al. Deregulation of histone lysine methyltransferases contributes to oncogenic transformation of human bronchoepithelial cells. Cancer Cell Int 2008; 8: 15.
Wang X, Elling AA, Li X, Li N, Peng Z, He G et al. Genome-wide and organ-specific landscapes of epigenetic modifications and their relationships to mRNA and small RNA transcriptomes in maize. Plant Cell 2009; 21: 1053–1069.
Zhang Z, Liu Q, Che Y, Yuan X, Dai L, Zeng B et al. Antigen presentation by dendritic cells in tumors is disrupted by altered metabolism that involves pyruvate kinase M2 and its interaction with SOCS3. Cancer Res 2010; 70: 89–98.
Zeng B, Li H, Liu Y, Zhang Z, Zhang Y, Yang R . Tumor-induced suppressor of cytokine signaling 3 inhibits toll-like receptor 3 signaling in dendritic cells via binding to tyrosine kinase 2. Cancer Res 2008; 68: 5397–5404.
This research was supported by the NSFC grants 91029736, 30771967 and 30872315; Ministry of Science and Technology grants (863 program, 2008AA02Z129); the National Key Basic Research and Development Program of China (973 Program, 2007CB914803); and the National Key Scientific Program (2011CB964902).
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on Genes and Immunity website
About this article
Cite this article
Huang, Y., Min, S., Lui, Y. et al. Global mapping of H3K4me3 and H3K27me3 reveals chromatin state-based regulation of human monocyte-derived dendritic cells in different environments. Genes Immun 13, 311–320 (2012). https://doi.org/10.1038/gene.2011.87
- dendritic cells
Fasciola gigantica excretory-secretory products (FgESPs) modulate the differentiation and immune functions of buffalo dendritic cells through a mechanism involving DNMT1 and TET1
Parasites & Vectors (2020)
Current Oral Health Reports (2019)
Alteration of histone H3K4 methylation in glomerular podocytes associated with proteinuria in patients with membranous nephropathy
BMC Nephrology (2016)
Histone demethylase Jumonji D3 (JMJD3/KDM6B) at the nexus of epigenetic regulation of inflammation and the aging process
Journal of Molecular Medicine (2014)
Acylated and unacylated ghrelin protect MC3T3-E1 cells against tert-butyl hydroperoxide-induced oxidative injury: pharmacological characterization of ghrelin receptor and possible epigenetic involvement
Amino Acids (2014)