• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Mar 29, 2011; 108(13): 5290–5295.
Published online Mar 14, 2011. doi:  10.1073/pnas.1017214108
PMCID: PMC3069155
Cell Biology

PU.1 and C/EBPα synergistically program distinct response to NF-κB activation through establishing monocyte specific enhancers

Fulai Jin,a,1 Yan Li,a,b,1 Bing Ren,a,c,2 and Rama Natarajanb,2

Abstract

Unraveling the complexity of transcriptional programs coded by different cell types has been one of the central goals of cell biology. By using genome-wide location analysis, we examined how two different cell types generate different responses to the NF-κB signaling pathway. We showed that, after TNF-α treatment, the NF-κB p65 subunit binds to distinct genome locations and subsequently induces different subsets of genes in human monocytic THP-1 cells versus HeLa cells. Interestingly, the differential p65 binding in two cell types correlates with preexisting cell type-specific enhancers before TNF-α stimulation, marked by histone modifications. We also found that two transcription factors, PU.1 and C/EBPα, appear to synergistically mediate enhancer creation and affect NF-κB target selection in THP-1 cells. In HeLa cells, coexpression of PU.1 and C/EBPα conferred TNF-α responsiveness to a subset of THP-1–specific NF-κB target genes. These results suggest that the diversity of transcriptional programs in mammalian cells arises, at least in part, from preexisting enhancers that are established by cell-specific transcription factors.

Keywords: genomics, epigenetics, chromatin immunoprecipitation, cell type-specific gene induction, reprogramming

Signaling pathways play critical roles in cellular and organ development, homeostasis, and responses to environmental changes and insults. Extracellular signals, working through membrane-bound receptor proteins and cascades of enzymatic reactions, activate transcription factors (TFs) that bind to regulatory DNA and modulate transcription of specific genes. However, what has not been fully elucidated is how a single stimulus and relatively small number of signaling pathways generate complex patterns of differential responses in diverse cell types.

One prevailing view is that combinatorial interactions between different TFs generate an unlimited number of possible TF/DNA recognition codes and diverse patterns of gene expression in different cell types (1). In support of this model, it has been shown that TF binding sites are present in clusters modularized at transcriptional promoters and enhancers (2). Additionally, many TFs are known to interact with each other, leading to the formation of multiprotein complexes at specific enhancers and under appropriate conditions (3).

Although this model explains cell-specific transcriptional programs in response to the same signaling pathways, the reported examples of synergistic interactions between TFs leading to specific activation of target genes have been limited, partly because of our still poor knowledge of transcriptional enhancers in the genome. Recently, it became possible to experimentally identify enhancers in the mammalian genome at a genome wide scale by using CBP/p300 (4, 5) or histone modification marks (6). Furthermore, chromatin modifications, especially H3K4me1, are associated with enhancers in a cell type-specific manner and correlated with cell-specific gene expression (7). A small number of lineage-determining TFs are shown to establish the H3K4me1 mark at their binding sites in hematopoietic progenitor cells. It has been postulated that such lineage-specific chromatin modifications at enhancers permit signal-dependent TF binding in a cell-specific manner (8). We decided to formally test this hypothesis by systematically investigating how these predefined cell-specific enhancers govern and respond to the binding of a common signal-dependent TF, NF-κB, in two different cell types.

NF-κB is a family of master TFs found in almost all animal cell types and plays essential roles in multiple physiological processes including inflammation and immunity (911). Binding and activation of κB sites are known to be context- and stimulus-dependent (1214). To understand how ubiquitous NF-κB induces differential transcriptional responses in different human cell types, we characterized the transcriptional programs of both HeLa and THP-1 cells in response to TNF-α, a well known inducer of NF-κB activation. We found that cell type-dependent gene induction by TNF-α is primarily governed by the differential NF-κB binding, as revealed by genome-wide location analysis of NF-κB subunit p65/RelA occupancy. Lineage-specific, preexisting chromatin modifications strongly predict the differential binding of NF-κB upon TNF-α stimulation in each cell type. We further show that a group of lineage-specific transcription TFs, including PU.1 and C/EBPα in THP-1 cells, may establish these primed enhancers and confer cell specificity to the ubiquitous NF-κB signaling. Our results suggest a general mechanism by which complexity and specificity of gene regulation by signaling pathways in different cell types can be achieved.

Results

Cell-Specific p65 Binding Correlates with Different Transcription Programs Induced by TNF-α.

We first characterized the transcription programs induced by TNF-α in HeLa and THP-1 cells. From gene expression profiling data gained with Affymetrix arrays, we identified 192 and 208 up-regulated genes in THP-1 and HeLa cells, respectively, in response to TNF-α treatment, and only 69 genes were shared between the two cell types (Fig. 1A). We also carried out Gene Ontology analysis on these induced genes and found significant differences between THP-1–specific and HeLa-specific target genes. Some function categories, such as inflammatory response and apoptosis, were enriched in TNF-α–induced genes in both cell types (Fig. 1B). However, TNF-α also induced genes with specialized cell-specific functions. For example, in THP-1 cells, many genes involved in hematopoiesis are up-regulated upon TNF-α treatment (Fig. 1B), which is consistent with the known origin of THP-1 cells and function of NF-κB in regulating hematopoietic lineage differentiation (15).

Fig. 1.
Cell-specific p65 binding correlates with distinct transcription programs induced by TNF-α. (A) Scatterplot shows gene induction in THP-1 cells versus HeLa cells. Red, brown, and green ovals represent THP-1–specific, common, and HeLa-specific ...

We further performed genome-wide location analysis with ChIP linked to whole genome arrays (ChIP-chips) to determine the genome locations of p65, the transcriptionally active subunit of NF-κB, in both cell types. Totals of 14,069 and 15,417 p65 binding sites were identified in TNF-α–treated THP-1 and HeLa cells, respectively, in agreement with previous estimates based on NF-κB binding in chromosome 22 in HeLa cells (13). Consistent with the predominantly cytoplasmic distribution of NF-κB p65 protein subunit in the absence of cell stimulation, we did not find significant p65 binding sites in untreated THP-1 cells with use of the same criteria. The in vivo location of p65 in THP-1 and HeLa cells was substantially different, with 9,837 THP-1–specific, 11,185 HeLa-specific, and only 4,232 common (Fig. 1C).

Importantly, the differential gene induction in each cell type correlated with cell type-specific p65 binding. There was strong enrichment of THP-1–unique p65 peaks around genes up-regulated in THP-1 cells (Fig. 1D), and in a similar fashion, TNF-α–induced genes in HeLa cells were most likely found around HeLa-unique p65 peaks. These data demonstrate that differential p65 binding is primarily responsible for cell type-specific p65 activity. For example, CCL3 and CCL4 genes were induced in THP-1 cells but not HeLa cells. We observed multiple p65 binding sites near these genes but not in HeLa cells (Fig. 1E). Similarly, we also observed common p65 binding sites around CCL2 (induced in both cell types) and HeLa-specific p65 bind sites around BDKRB1 (induced only in HeLa cells) genes (Fig. 1E). These data revealed a clear difference of NF-κB target selection in two cell types after TNF-α stimulation.

NF-κB Binding Sites Are Occupied by Modified Histones and Coactivator Protein p300 Before TNF-α Signaling.

As cell-specific enhancer distribution has been shown to correlate with cell-specific gene expression (5, 7), we hypothesized that cell-dependent gene induction by NF-κB may also be regulated by these cell-specific enhancers. To test this hypothesis, we mapped the genome-wide locations of the transcription coactivator protein p300, the enhancer chromatin modification mark H3K4me1, and the promoter-specific chromatin mark H3K4me3 in both HeLa and THP-1 cells before and after TNF-α treatment.

In both cell types, most p65 binding sites were also occupied by p300 and/or H3K4 methylation marks (SI Appendix, Fig. S1). Strong enrichment of H3K4me3 and H3K4me1 marks were found at proximal promoter and distal p65 binding sites, respectively (Fig. 2A). Furthermore, we also observed increased levels of these marks at p65 binding sites, which is consistent with the model that p65 binding leads to recruitment of cofactors and chromatin modifications (Fig. 2 AC). However, the overall increase of these marks, especially for H3K4me1 and H3K4me3, is modest (Fig. 2A). In fact, p300 and H3K4 methylation marks were present in the majority (83% for THP-1 cells and 84% for HeLa cells) of p65 binding sites even in the absence of TNF-α treatment, where there is very little nuclear localization of NF-κB p65 (16) (Fig. 2 A and D and SI Appendix, Fig. S2). The observation of preexisting, active chromatin modifications and p300 binding at most p65 binding sites in THP1 cells, even in the unstimulated state, suggests that the establishment of an enhancer-specific chromatin modification state at NF-κB binding sites is generally independent of NF-κB binding. Instead, these results suggest that NF-κB binding to DNA may be influenced by preexisting enhanceosomes in the genome.

Fig. 2.
Premarked enhancers and promoters predict cell-specific p65 binding. (A) Average ChIP enrichment of different factors is plotted for all promoter distal (Upper) and proximal (<2.5 kb; Lower) p65 binding sites in THP-1 cells. Blue, untreated cells; ...

Differential Binding of NF-κB to the Genome Is Predicted by Cell-Specific Premarking of Enhancers.

We next evaluated how premarking of p300 and H3K4 methylation may affect p65 binding. As shown in Fig. 2E, the cell-specific p65 binding in TNF-α–treated cells correlated well with cell-specific premarking of the enhancers by p300 or H3K4me1 in untreated cells. The promoter mark H3K4me3, however, was most enriched in the p65 binding sites shared by both cell types. These results suggest that the majority of cell-specific p65 binding sites are on enhancers, which has been shown to be marked by highly cell type-specific H3K4me1 patterns (7), whereas most promoter-associated p65 binding sites are invariant between two cell types (SI Appendix, Fig. S3). Therefore, differential binding of NF-κB to the genome in response to a stimulus could be influenced by cell-specific premarking of enhancers.

As our data showed that the majority of p65 peaks were premarked by active chromatin modifications and p300 before stimulation, we next hypothesized that the marking of enhancers by the specific histone modifications or coactivator binding might dictate NF-κB target selection along the genome in each cell type. To test this hypothesis, we investigated whether in vivo p65 binding sites could be predicted by a combination of the NF-κB recognition motif with the chromatin modification state and p300 binding in each cell type.

NF-κB binds DNA via a consensus motif 5′-GGGRNWYYCC-3′ (R, purine; W, A or T; Y, pyrimidine) known as κB sites (17, 18). We found approximately 300,000 instances of the κB motif along the human genome. However, only 3% of them were bound by p65 in each cell type. As a result, the accuracy of predictions using the motif information alone suffered from low specificity. We next combined both the κB motif and the location of preexisting p300 or H3K4 methylation marks to predict NF-κB binding sites in THP-1 cells. As shown in Fig. 2F, the specificity improved to 94%, and the sensitivity remained at 86%. Significant improvement of prediction performance was also observed in HeLa cells (SI Appendix, Fig. S4). These results suggest that preexisting active chromatin at enhancers are excellent predictors of in vivo κB site binding by NF-κB p65.

The preferential binding of p65 to active chromatin locations marked by H3K4 methylation suggested a dominant role of enhancers in priming or preparing for TF binding. One possible mechanism is that the primed enhancers may influence p65 binding through controlling the accessibility of κB sites and function to choose NF-κB targets in the genome. Indeed, recent studies have revealed existence of nucleosome-free regions in enhancers (19, 20).

However, there were still nearly 20% of p65 binding sites not premarked by active chromatin marks (Fig. 2D). One intriguing question is why some p65 binding sites need enhancer priming whereas others do not. We found that, in general, DNA sequences of the primed p65 binding sites have fewer and less optimal κB motifs compared with the p65 sites without premarking (SI Appendix, Fig. S5). These data suggested that DNA sequences of many of the cell-specific p65 binding sites may not be optimal to recruit NF-κB on their own; TF binding at these places could also be dependent on preexisting enhancers that make up an additional epigenetic layer of regulation.

Identification of Cell-Specific TFs That Govern Differential NF-κB Binding.

It was previously shown that lineage-specific TFs could establish cell-specific chromatin modification patterns at enhancers (8). To identify potential TFs governing cell-specific p65 binding in THP1 and HeLa cells, we examined TF binding motifs enriched at THP-1–specific p65 binding sites (class A, top 10 in SI Appendix, Table S1) or at HeLa-specific p65 peaks (class B, top 10 in SI Appendix, Table S2). Of the 20 corresponding human TFs, seven were differentially expressed in the two cell types and likely contributed to the establishment of cell type-specific enhancers (Fig. 3A). The class A TFs includes LMO2 and PU.1 (SPI1), which were specifically expressed in THP-1 cells and are known to be critical for hematopoietic development (21, 22). Four of five class B TFs displaying HeLa-specific expression were c-Jun, TEAD1, EVI1, and FOXQ1. Interestingly, cooperation between c-Jun and NF-κB has been well documented in the IFN-β enhanceosome (23, 24). GFI1 was the only class B TF that was highly expressed in THP-1 cells; this zinc finger protein has been shown to act as a transcription repressor through creating repressive chromatin environments (25). Therefore, GFI1 may act as negative regulator for p65 binding in THP-1 cells, and its absence in HeLa cells may allow for the premarking of enhancers in that cell type.

Fig. 3.
De novo discovery of cell-specific TFs that correlated with differential p65 binding. (A) TF binding motifs enriched in THP-1–specific (red) or HeLa-specific (blue) p65 binding sites are shown. Dashed line indicates fivefold expression difference ...

To further validate the model that cell type-specific TFs establish differential p65 occupancy, PU.1 binding was evaluated by ChIP-chip and led to identification of 7,465 peaks in untreated THP-1 cells. As shown in Fig. 3B, 1,880 of 9,837 (19%) THP-1–specific p65 peaks were bound by PU.1 before stimulation, whereas only 1.5% of HeLa-specific p65 peaks (168 of 11,185) overlapped with these PU.1 locations (12.7 fold; P < 10−300). These results suggested a possible role for PU.1 in setting up premarked enhancers before TNF-α signaling in THP-1 cells.

C/EBPα Is Predicted to Be a PU.1 Modulator Involved in Monocyte-Specific TNF-α Response.

As the PU.1 gene is not expressed in HeLa cells and is predicted to play an essential role in mediating THP-1–specific p65 binding, we hypothesized that ectopic expression of the PU.1 protein in HeLa cells might lead to the induction of at least a subset of THP-1–specific TNF-α–dependent genes. To test this hypothesis, we chose the top 10 genes specifically induced by TNF-α in THP-1 cells, but not HeLa cells, from microarray data (Fig. 1B), and verified five of them (CCL3, CCL4, IL12B, C3AR1, and GPR84) had PU.1 peaks as well as THP-1–specific p65 binding sites within 30 kb of their promoters (Fig. 4A). We posited that these genes are THP-1–specific NF-κB target genes, which can be induced by TNF-α only in THP-1 cells in which PU.1 is expressed. However, overexpressing PU.1 alone in HeLa cells did not enhance TNF-α responsiveness for any of these genes (Fig. 4B). These results suggest that PU.1 alone is not sufficient to confer TNF-α responsiveness to these THP-1–specific NF-κB–dependent genes. Therefore, other factors must be also required to generate a monocyte/macrophage-like transcription program in HeLa cells.

Fig. 4.
Ectopic expression of PU.1 and C/EBPα turns on a THP-1–specific transcription program in HeLa cells in response to TNF-α treatment. (A) Snapshots showing locations of five THP-1–specific NF-κB target genes and THP-1–specific ...

We next searched for PU.1 modulators that may help initiate monocyte-specific p65 binding sites based on the following criteria: first, colocalization of such modulators with PU.1 should increase the chance of p65 binding; and second, such a modulator must be specifically expressed in THP-1 but not HeLa cells. To identify factors satisfying the first constraint, we further examined all the PU.1 locations. As shown in Fig. 3C, we could classify all PU.1 sites into three categories based on whether they are marked by H3K4me3, H3K4me1, or neither. Strikingly, we found that PU.1 sites marked by H3K4 methylation were much more likely to be bound by p65 upon TNF-α stimulation (41% vs. 4%; Fig. 3D). These results suggest that PU.1 itself may not be sufficient to establish an active chromatin mark, as many PU.1 sites lack H3K4 methylations; additional modulators of PU.1 are necessary to facilitate the selection of THP1-specific NF-κB targets through the formation of active chromatin. We then performed motif analyses to identify candidate TFs that are enriched in the PU.1 sites with enhancer marks, and found that the motifs of CCAAT/enhancer binding proteins (C/EBPs) showed the strongest enrichment in the PU.1 enhancers with active chromatin (Fig. 3E).

C/EBPs are a family of basic leucine zipper TFs composed of six members. Interestingly, among all C/EBPs, only C/EBPα was highly expressed in THP-1 but not HeLa cells (SI Appendix, Fig. S6). ChIP-chip assay revealed 3,508 binding sites for C/EBPα in untreated THP-1 cells. We found that locations bound by both PU.1 and C/EBPα were most likely to be bound by p65 following TNF-α treatment (SI Appendix, Fig. S7) compared with locations with only PU.1 or C/EBPα. These results suggest that, in THP-1 cells, C/EBPα might be one of the factors that function synergistically with PU.1 to create active enhancers and select NF-κB targets.

Ectopic Expression of both PU.1 and C/EBPα Allows THP-1–Specific NF-κB Target Genes to Respond to TNF-α Stimulation in HeLa Cells.

To confirm our prediction that C/EBPα may act as PU.1 modulator in selecting THP-1–specific NF-κB target genes, we further evaluated the five THP-1–specific TNF-α–responsive genes mentioned earlier (Fig. 4A). Consistent with our model, we observed colocalization of THP-1–specific p65 peaks with PU.1, C/EBPα, and the enhancer mark H3K4me1 around these genes (Fig. 4A). Strikingly, we found that all these genes gained responsiveness to TNF-α stimulation only when both PU.1 and C/EBPα were coexpressed in HeLa cells (Fig. 4B). To test the chromatin events near these genes, we further identified eight THP-1–specific p65 locations (P1–P8; Fig. 4A) around these five genes and performed ChIP/quantitative PCR (qPCR) to detect p65 occupancy and active chromatin. Following TNF-α treatment, the strongest recruitment of p65 was observed at these locations when both PU.1 and C/EBPα were coexpressed in HeLa cells (Fig. 4C). Furthermore, we also observed the greatest H3K4me1 enrichment in untreated HeLa cells cotransfected with PU.1 and C/EBPα; this trend was not seen in random negative control regions (R1–R4; Fig. 4D). These results strongly support the model that combinatorial binding of THP-1–specific TFs PU.1 and C/EBPα synergistically create a distinct group of enhancers and mediate selection of a subset of monocyte/macrophage-specific NF-κB target genes for specific induction by allowing the binding of p65 to these locations for productive gene expression.

Discussion

Cells of the monocyte/macrophage lineage produce several cytokines and chemokines and play essential roles in inflammatory and immune responses. It is well known that increased inflammation and inflammatory proteins like TNF-α play important roles in the pathology of disease states such as sepsis and diabetes and its complications via NF-κB signaling. Our results provide key mechanistic information about the NF-κB–dependent inflammatory program occurring in a coordinated cell-specific manner in monocytes, which can lead to enhanced understanding of these pathologic processes.

It has been generally accepted that distinct spectra of TFs are involved in the establishment and/or maintenance of lineage-specific epigenetic marks, such as enhancer marking H3K4 monomethylation. On the contrary, the specificity of enhancer distributions in different cells also suggested the role of these marks in maintaining cell identities. Elucidating the connection between cell fate-regulating TFs and histone modifications is a very interesting topic studied in recent literature. For example, some recent genome-wide studies in mouse macrophages showed that the Ets protein PU.1 was required for maintenance of chromatin modification state at most endotoxin-inducible enhancers in mouse macrophages (19). Ectopic expression of PU.1 was sufficient to reactivate some of macrophage-specific enhancers in mouse fibroblasts (19) and myeloid progenitor cells (8). In the present work, we found that PU.1 alone was not sufficient to fully introduce monocyte-specific responses in HeLa cells; instead, another factor, C/EBPα, worked together with PU.1 to establish a monocyte-specific TNF-α–responsive transcription program.

Interestingly, a minimum of two TFs, PU.1 and C/EBPα, have previously been reported to convert mouse fibroblasts into macrophage-like cells (26). Our results, although derived from human HeLa cell lines, may still provide further insights into this transdifferentiation or reprogramming event from a genome-wide perspective. We propose that PU.1 may establish monocyte specific enhancers, but formation of active enhanceosomes at these regions may require additional factors, including C/EBPα. Our data suggest that a combination of PU.1 and C/EBPα may assist the de novo formation of active enhancers and facilitate p65 binding in conjunction with specific histone modifications. We further speculate that, during the development of cell lineages, in response to differentiation signals, a spectrum of TFs expressed in progenitor cells may predetermine the cell fates through establishing lineage-specific enhancers as well as their associated epigenetic marks; cell lineage commitment is as such tightly regulated by the sequential expression of multiple cell fate-regulating TFs.

It is worth mentioning that, although ectopic expression of both PU.1 and C/EBPα in HeLa cells could lead to significant induction of selected THP-1–specific TNF-α–responsive genes (Fig. 4B), the fold changes of these genes were still relatively modest compared with those in THP-1 cells. One likely explanation is that gene expression specificity is determined by combinatorial regulation of multiple lineage-specific TFs at different cis-regulatory elements, despite a small number of “master” transregulatory factors marking each cell type. Additionally, cell lineage identities are also governed by other epigenetic factors such as repressive histone modifications and DNA methylation (2730). Therefore, further manipulation of such additional factors may be required for more efficient reprogramming of cell-specific transcription.

In summary, our results demonstrate key functional roles for PU.1 and C/EBPα in creating monocyte specific enhancers and affecting target selection of NF-κB–dependent gene expression in response to TNF-α signaling, suggesting that it may be a general mechanism for other signaling pathways and TFs.

Methods

HeLa and THP-1 cells were obtained from ATCC and cultured under recommended conditions. For ChIP-chip assays, cells were harvested before and after TNF-α (R&D Systems) treatment (10 ng/mL, 1 h), and chromatin was used for ChIP with commercially available antibodies: p65 (sc-372; Santa Cruz), p300 (sc-585; Santa Cruz), PU.1 (sc-352; Santa Cruz), C/EBPα (sc-9314; Santa Cruz), H3K4me1 [07–436 (Millipore) or ab8895 (Abcam)], and H3K4me3 (07–473; Millipore). Chromatin preparation, ChIP, ligation-mediated PCR, and ChIP-qPCR were performed as described (31, 32). ChIP samples were labeled and hybridized to whole-genome HD2 human economy tiling arrays (hg18; Nimblegen-Roche) following the manufacturer’s protocol. We used MA2C (33) to normalize and call peaks. DNA motif models were downloaded from TRANSFAC (34) and JASPAR (35) databases, and motif scanning was performed by using the MAST program (36) with default parameters. TRANSFAC motif model V$NFKAPPAB65_01 was used to scan κB sites. For gene expression analysis, array data in HGU133 Plus 2.0 (Affymetrix) for untreated and TNF-α–treated THP-1 cells (in triplicates) were described (16); for HeLa expression data, duplicated RNA samples from both untreated and TNF-α–treated (10 ng/mL, 2 h) cells were applied to Affymetrix arrays and data were analyzed using Bioconductor R package. Raw data can be downloaded from the Gene Expression Omnibus database (accession no. GSE26994). Primers for RT-qPCR and ChIP-qPCR are listed in SI Appendix, Tables S3 and S4. Human cDNA clones were purchased from OriGene (PU.1, SC315715; and C/EBPα, SC303472). Transfections in HeLa cells were performed by using FuGENE HD transfection reagent (Roche) following the manufacturer’s instructions.

Supplementary Material

Supporting Information:

Acknowledgments

This work was supported by National Institutes of Health Grants R01DK 065073, R01 HL087864, and R01 HL106089 (to R.N.), the Juvenile Diabetes Research Foundation (R.N. and B.R.), the Ludwig Institute for Cancer Research (B.R.), the California Institute for Regenerative Medicine (B.R.), and a predoctoral fellowship from the American Heart Association Western States Affiliate (to Y.L.).

Footnotes

The authors declare no conflict of interest.

Database deposition: The data reported in this paper have been deposited in the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo (accession no. GSE26994).

*This Direct Submission article had a prearranged editor.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1017214108/-/DCSupplemental.

References

1. Tjian R, Maniatis T. Transcriptional activation: a complex puzzle with few easy pieces. Cell. 1994;77:5–8. [PubMed]
2. Berman BP, et al. Exploiting transcription factor binding site clustering to identify cis-regulatory modules involved in pattern formation in the Drosophila genome. Proc Natl Acad Sci USA. 2002;99:757–762. [PMC free article] [PubMed]
3. Levine M, Tjian R. Transcription regulation and animal diversity. Nature. 2003;424:147–151. [PubMed]
4. Kim TK, et al. Widespread transcription at neuronal activity-regulated enhancers. Nature. 2010;465:182–187. [PMC free article] [PubMed]
5. Visel A, et al. ChIP-seq accurately predicts tissue-specific activity of enhancers. Nature. 2009;457:854–858. [PMC free article] [PubMed]
6. Heintzman ND, et al. Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat Genet. 2007;39:311–318. [PubMed]
7. Heintzman ND, et al. Histone modifications at human enhancers reflect global cell-type-specific gene expression. Nature. 2009;459:108–112. [PMC free article] [PubMed]
8. Heinz S, et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010;38:576–589. [PMC free article] [PubMed]
9. Gerondakis S, et al. Unravelling the complexities of the NF-kappaB signalling pathway using mouse knockout and transgenic models. Oncogene. 2006;25:6781–6799. [PubMed]
10. Gilmore TD. Introduction to NF-kappaB: Players, pathways, perspectives. Oncogene. 2006;25:6680–6684. [PubMed]
11. Lenardo MJ, Fan CM, Maniatis T, Baltimore D. The involvement of NF-kappa B in beta-interferon gene regulation reveals its role as widely inducible mediator of signal transduction. Cell. 1989;57:287–294. [PubMed]
12. Hoffmann A, Natoli G, Ghosh G. Transcriptional regulation via the NF-kappaB signaling module. Oncogene. 2006;25:6706–6716. [PubMed]
13. Martone R, et al. Distribution of NF-kappaB-binding sites across human chromosome 22. Proc Natl Acad Sci USA. 2003;100:12247–12252. [PMC free article] [PubMed]
14. Schreiber J, et al. Coordinated binding of NF-kappaB family members in the response of human cells to lipopolysaccharide. Proc Natl Acad Sci USA. 2006;103:5899–5904. [PMC free article] [PubMed]
15. Bottero V, Withoff S, Verma IM. NF-kappaB and the regulation of hematopoiesis. Cell Death Differ. 2006;13:785–797. [PubMed]
16. Li Y, et al. Role of the histone H3 lysine 4 methyltransferase, SET7/9, in the regulation of NF-kappaB-dependent inflammatory genes. Relevance to diabetes and inflammation. J Biol Chem. 2008;283:26771–26781. [PMC free article] [PubMed]
17. Chen FE, Ghosh G. Regulation of DNA binding by Rel/NF-kappaB transcription factors: Structural views. Oncogene. 1999;18:6845–6852. [PubMed]
18. Sen R, Baltimore D. Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Cell. 1986;46:705–716. [PubMed]
19. Ghisletti S, et al. Identification and characterization of enhancers controlling the inflammatory gene expression program in macrophages. Immunity. 2010;32:317–328. [PubMed]
20. He HH, et al. Nucleosome dynamics define transcriptional enhancers. Nat Genet. 2010;42:343–347. [PMC free article] [PubMed]
21. DeKoter RP, Walsh JC, Singh H. PU.1 regulates both cytokine-dependent proliferation and differentiation of granulocyte/macrophage progenitors. EMBO J. 1998;17:4456–4468. [PMC free article] [PubMed]
22. Warren AJ, et al. The oncogenic cysteine-rich LIM domain protein rbtn2 is essential for erythroid development. Cell. 1994;78:45–57. [PubMed]
23. Maniatis T, et al. Structure and function of the interferon-beta enhanceosome. Cold Spring Harb Symp Quant Biol. 1998;63:609–620. [PubMed]
24. Wathelet MG, et al. Virus infection induces the assembly of coordinately activated transcription factors on the IFN-beta enhancer in vivo. Mol Cell. 1998;1:507–518. [PubMed]
25. Saleque S, Kim J, Rooke HM, Orkin SH. Epigenetic regulation of hematopoietic differentiation by Gfi-1 and Gfi-1b is mediated by the cofactors CoREST and LSD1. Mol Cell. 2007;27:562–572. [PubMed]
26. Feng R, et al. PU.1 and C/EBPalpha/beta convert fibroblasts into macrophage-like cells. Proc Natl Acad Sci USA. 2008;105:6057–6062. [PMC free article] [PubMed]
27. Kim K, et al. Epigenetic memory in induced pluripotent stem cells. Nature. 2010;467:285–290. [PMC free article] [PubMed]
28. Ji H, et al. Comprehensive methylome map of lineage commitment from haematopoietic progenitors. Nature. 2010;467:338–342. [PMC free article] [PubMed]
29. Wen B, Wu H, Shinkai Y, Irizarry RA, Feinberg AP. Large histone H3 lysine 9 dimethylated chromatin blocks distinguish differentiated from embryonic stem cells. Nat Genet. 2009;41:246–250. [PMC free article] [PubMed]
30. Hawkins RD, et al. Distinct epigenomic landscapes of pluripotent and lineage-committed human cells. Cell Stem Cell. 2010;6:479–491. [PMC free article] [PubMed]
31. Kim TH, Barrera LO, Ren B. ChIP-chip for genome-wide analysis of protein binding in mammalian cells. Curr Protoc Mol Biol. 2007;Chapter 21 Unit 21:13. [PubMed]
32. Kim TH, et al. A high-resolution map of active promoters in the human genome. Nature. 2005;436:876–880. [PMC free article] [PubMed]
33. Song JS, et al. Model-based analysis of two-color arrays (MA2C) Genome Biol. 2007;8:R178. [PMC free article] [PubMed]
34. Heinemeyer T, et al. Expanding the TRANSFAC database towards an expert system of regulatory molecular mechanisms. Nucleic Acids Res. 1999;27:318–322. [PMC free article] [PubMed]
35. Sandelin A, Alkema W, Engström P, Wasserman WW, Lenhard B. JASPAR: An open-access database for eukaryotic transcription factor binding profiles. Nucleic Acids Res. 2004;32(database issue):D91–D94. [PMC free article] [PubMed]
36. Bailey TL, Gribskov M. Combining evidence using p-values: Application to sequence homology searches. Bioinformatics. 1998;14:48–54. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...