CHROMATIN AND TRANSCRIPTIONAL REGULATION IN MOUSE MACROPHAGES By Michael McAndrew A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Genetics-Doctor of Philosophy 2017 ABSTRACT CHROMATIN AND TRANSCRIPTIONAL REGULATION IN MOUSE MACROPHAGES By Michael McAndrew Eukaryotic genomes must be extensively packaged into a DNA-protein complex called chromatin due to their large sizes and the spatial restrictions of the nucleus. Nucleosomes, the basic repeating unit of this complex, have long been viewed as a barrier to basic cellular processes including transcription, and recent studies suggest that chromatin architecture plays a critical role in the regulation of gene expression. We have used primary bone marrow-derived macrophages (BMDMs) as a model to investigate chromatin changes associated with inducible and cell-type specific gene expression in response to bacterial lipopolysaccharide (LPS). Macrophages are specialized cells of the innate immune system that arise during differentiation from multipotent hematopoietic stem and progenitor cells (HSPCs) through the coordinated action of lineage-specific transcription factors (TFs). These cells have unique functions in response to foreign threat, and previous genome-wide studies have identified macrophage-specific distal enhancers that play a key role in the pro-inflammatory response to LPS. Using a quantitative nucleosome occupancy assay, we have shown that nucleosomes are stably evicted from these enhancers under inducing conditions in BMDMs, and this depletion correlates with signal-induced TF binding and increased gene expression. Using a knockdown approach targeting BAF/PBAF chromatin remodeling complexes, we have shown that nucleosome remodelers are recruited to regulatory elements early during differentiation by lineage-specific TFs, and that disruption of this process results in increased nucleosome occupancy at these elements and prevents nucleosome eviction and gene induction in response to LPS. In order to more precisely determine how and when enhancers might be rendered accessible during differentiation, we further investigated chromatin structure in HSPCs. This led to the surprising finding that nucleosome occupancy may be universally low in these cells. We are now using a genome-wide extension of the quantitative nucleosome occupancy (GNO-seq, Global Nucleosome Occupancy-sequencing) to analyze changes in nucleosome occupancy associated with macrophage differentiation from HSPCs genome-wide. This research will provide crucial insights into the regulation of inducible gene expression, the role of remodelers in maintaining chromatin accessibility, and may demonstrate global differences in chromatin between cell types. Copyright by MICHAEL MCANDREW 2017 To Elijah. Stay curious. v ACKNOWLEDGMENTS I would like to thank my mentor Dr. Monique Floer for the opportunity to pursue these projects. I would like to thank her for her time, insight, and patience. My graduate experience has shaped my understanding of what a scientist does and should do, as well as the questions that should be asked and which of these can be adequately addressed with evidence. I would like to thank all of the former members of the Floer laboratory for their productive scientific discussions and support: Dr. Mohita Tagore, Alison Gjidoda, Tyler Miksanek, Hunter Piegols, and Alexander Woods. I would like to thank my committee members for guidance and insight: Dr. David Arnosti, Dr. Jason Knott, and Dr. Min-Hao Kuo. I would like to thank Dr. Amy Ralston and the members of her laboratory for helpful comments and discussion during our joint lab meetings: Dr. Tristan Frum, Dr. Mike Halbisen, Dr. Alyson Lokken, and Dr. Tony Parenti. I would like to especially thank Dr. Halbisen for his assistance with principles of bioinformatics, and Dr. Parenti for providing embryonic stem cells, as well as teaching me to culture them myself. I would like to thank Dr. Louis King, Dr. Nara Parameswaran, and Michael Steury for their expertise and assistance with cell staining and flow cytometry. I would like to thank Dr. Shin-Han Shiu, Dr. Sahra Uygun, and Nicholas Panchy for their assistance with genome-wide techniques and subsequent analyses. I would like to thank Dr. Erik Martinez-Hackert for the use of his tissue culture hood. vi I would like to thank past and present members of the Genetics Graduate Program for their friendship, support, and guidance. I would like to especially thank Dr. Barbara Sears for believing in me; Dr. Cathy Ernst for listening to me; and Jeannine Lee for being a good lunch partner and an even better program secretary. Finally, I would like to thank my wife Rachel, my mother Sandra, my father Michael, my brother Josh, and my sister-in-law Lora for their unconditional love and support during these last six years. I could not have done it without you. vii TABLE OF CONTENTS LIST OF TABLES ........................................................................................................... xi LIST OF FIGURES ......................................................................................................... xii KEY TO ABBREVIATIONS .......................................................................................... xiv Chapter 1: Introduction .................................................................................................. 1 Chromatin is a barrier to transcription .................................................................. 2 Mouse macrophages as a model for the study of transcriptional regulation ........ 3 Open chromatin is a feature of lineage-specific enhancers ................................. 4 Pioneer factors in macrophage differentiation ...................................................... 5 Chromatin remodeling complexes and transcription ............................................ 6 Unique chromatin states in multipotent progenitors and stem cells ..................... 8 Clinical Significance ........................................................................................... 10 REFERENCES ................................................................................................... 11 Chapter 2: Nucleosomes are stably evicted from enhancers but not promoters upon induction of certain pro-inflammatory genes in mouse macrophages ......... 18 Abstract ................................................................................................................ 19 Introduction .......................................................................................................... 19 Experimental Procedures .................................................................................... 23 Cell isolation and culture ........................................................................... 23 mRNA determination ................................................................................. 24 Chromatin immunoprecipitation ................................................................ 24 Quantitative nucleosome occupancy assay .............................................. 25 Genomic DNA isolation ............................................................................. 26 qRT-PCR .................................................................................................. 26 Results ................................................................................................................. 26 Nucleosome occupancy at the Il12b enhancer and promoter upon LPS induction .................................................................................................... 26 Changes in nucleosome occupancy at the transcriptional regulatory regions of Il1a ........................................................................................... 31 Timing of enhancer nucleosome removal ................................................. 32 Histone modifications at the promoters and enhancers of Il12b and Il1a . 32 Binding of cis-regulatory TFs to the distal enhancers of Il12b and Il1a .... 36 Binding of the transcriptional machinery to nucleosome-free Il12b and Il1a promoters .................................................................................................. 41 Discussion ........................................................................................................... 43 Acknowledgements .............................................................................................. 49 REFERENCES .................................................................................................... 50 Chapter 3: Chromatin remodeler recruitment during macrophage differentiation facilitates transcription factor binding to enhancers in mature cells ..................... 56 viii Abstract ................................................................................................................ 57 Introduction .......................................................................................................... 57 Experimental Procedures .................................................................................... 60 Cell isolation and culture ........................................................................... 60 shRNA mediated knockdown of BRG1 and SNF5 .................................... 61 Quantitative nucleosome occupancy assay .............................................. 61 Chromatin immunoprecipitation ................................................................ 61 mRNA determination ................................................................................. 62 Chromatin fractionation and Western blotting ........................................... 62 Flow cytometry .......................................................................................... 64 Statistical analysis ..................................................................................... 65 Results ................................................................................................................. 65 BAF/PBAF is recruited to the Il12b and Il1a enhancers in BMDMs .......... 65 BAF/PBAF recruitment is a consequence of PUER translocation to the nucleus ...................................................................................................... 66 BAF/PBAF is required for Il12b and Il1a induction in BMDMs .................. 71 BRG1 KD affects nucleosome occupancy and eviction at the Il12b and Il1a enhancers .......................................................................................... 72 Knockdown of SNF5 abolishes BAF/PBAF binding at the Il12b and Il1a enhancers ................................................................................................. 78 Nucleosome occupancy at the Il12b and Il1a enhancers increases in the absence of BAF/PBAF recruitment ........................................................... 82 FACS analysis reveals effects of SNF5 KD on cytokine expression in single cells ................................................................................................ 86 Discussion ........................................................................................................... 89 Acknowledgements .............................................................................................. 93 REFERENCES .................................................................................................... 94 Chapter 4: GNO-seq quantifies nucleosome occupancy at gene promoters and enhancers in LPS stimulated macrophages .............................................................. 99 Abstract .............................................................................................................. 100 Introduction ........................................................................................................ 101 Experimental Procedures .................................................................................. 104 Cell isolation and sample preparation ..................................................... 104 Illumina library preparation and sequencing ........................................... 105 Data processing ...................................................................................... 105 Generation of lambda-normalized GNO-seq tracks ................................ 106 Random genomic sampling .................................................................... 107 GC content .............................................................................................. 107 Heatmaps and average nucleosome occupancy plots ........................... 108 Identification of nucleosome depleted regions ........................................ 110 Gene ontology analysis ........................................................................... 110 De novo motif search .............................................................................. 110 Nucleosome occupancy at super-enhancers .......................................... 111 Results ............................................................................................................... 111 GNO-seq analysis ................................................................................... 111 ix GNO-seq validation ................................................................................. 114 Nucleosome occupancy surrounding transcriptional start sites .............. 118 Promoter nucleosomes at highly induced genes .................................... 123 LPS-induced changes in nucleosome occupancy at enhancers............. 126 Identification of regions partially depleted in rM and further depleted in aM ................................................................................................................ 131 Different enhancer categories show characteristic depletion in rM and aM ................................................................................................................ 133 TF consensus-sites associated with poised-activated and poised-not activated enhancers ................................................................................ 137 Nucleosome depletion at super-enhancers ............................................ 139 Discussion ......................................................................................................... 143 Nucleosome removal at promoters ......................................................... 143 Nucleosome depletion at enhancers ....................................................... 146 Conclusion .............................................................................................. 147 Acknowledgements ............................................................................................ 148 APPENDICES .................................................................................................... 149 APPENDIX A: Supplementary Experimental Procedures ....................... 150 APPENDIX B: Supplementary Figures ................................................... 153 REFERENCES .................................................................................................. 163 Chapter 5: Global nucleosome occupancy increases during differentiation of hematopoietic stem and progenitor cells into macrophages ................................ 171 Introduction ........................................................................................................ 172 Experimental Procedures .................................................................................. 174 Cell isolation and culture ......................................................................... 174 Quantitative nucleosome occupancy assay ............................................ 175 Quantitation of total cellular chromatin protected against digestion by MNase ..................................................................................................... 175 Flow cytometry ........................................................................................ 176 Results ............................................................................................................... 176 Nucleosome occupancy at macrophage-specific enhancers increases during differentiation ............................................................................... 176 Protection of total cellular chromatin against MNase increases as HSPCs differentiate into macrophages ................................................................ 184 Nucleosome occupancy is lower in ESCs than in differentiated cells ..... 186 Discussion ......................................................................................................... 187 REFERENCES .................................................................................................. 191 Chapter 6: Conclusions and future directions ........................................................ 194 REFERENCES .................................................................................................. 199 x LIST OF TABLES Table 2.1. Statistical significance of factor binding, ChIP of Fig. 2.3 ....... 40 Table 4.1. Categories of putative enhancers identified by Ostuni et al. . 127 Table 4.S1 Threshold conditions defining partially depleted regions in rM and regions of nucleosome eviction in aM .............................. 162 xi LIST OF FIGURES Figure 2.1. Changes in nucleosome occupancy upon LPS induction at a distal enhancer and the promoter of Il12b ................................. 29 Figure 2.2. Changes in nucleosome occupancy upon LPS induction at a putative distal enhancer and promoter of Il1a, kinetics of nucleosome removal, and changes in histone modifications . 34 Figure 2.3. Binding of cis-regulatory TFs and recruitment of the transcriptional machinery to the regulatory regions of Il12b and Il1a upon LPS induction ....................................................... 38 Figure 2.4. PolII and TBP binding in the fraction of Il12b and Il1a promoters in a population of induced BMDMs that is nucleosome-free .......................................................................... 42 Figure 3.1. Recruitment of BAF/PBAF to macrophage-specific enhancers ....................................................................................................... 69 Figure 3.2. KD of the catalytic BAF/PBAF subunit BRG1 ............................ 73 Figure 3.3. Nucleosome occupancy in shLuc treated and untreated control cells .................................................................................. 77 Figure 3.4. KD of the shared BAF/PBAF core subunit SNF5 ...................... 80 Figure 3.5. Nucleosome occupancy in SNF5 KD cells ................................ 84 Figure 3.6. Cytokine expression in SNF5 KD cells ...................................... 87 Figure 3.7. Remodeler assisted competition favors TF over nucleosome binding to sites in enhancers ..................................................... 91 Figure 4.1. GNO-seq analysis ...................................................................... 113 Figure 4.2. Nucleosome occupancy at promoters ..................................... 120 Figure 4.3. Nucleosome occupancy at promoters of LPS-induced genes ..................................................................................................... 124 Figure 4.4. Nucleosome occupancy at macrophage enhancers .............. 130 Figure 4.5. Regions of nucleosome depletion and TF-motifs in putative enhancers ................................................................................... 136 xii Figure 4.6 Nucleosome occupancy at super-enhancers .......................... 141 Figure 4.S1. Correlation between fractional nucleosome occupancies obtained as the ratio of MNase over Input-fractions and occupancies obtained by curve-fitting .................................... 154 Figure 4.S2. Fragment sizes of inserts in Illumina libraries ........................ 155 Figure 4.S3. GNO-seq analysis at enhancers and promoters of three proinflammatory genes ................................................................... 156 Figure 4.S4. Nucleosome occupancy at Il12b and Il1a determined by the MNase option of deepTools bamCoverage, DANPOS and GNOseq, and compared to the qRT-PCR based assay .................. 157 Figure 4.S5. Nucleosome occupancy in the genome analyzed by GNO-seq and deepTools bamCoverage without dyad alignment .......... 158 Figure 4.S6. Expression of genes highly expressed in the presence of LPS ..................................................................................................... 159 Figure 4.S7. Nucleosome occupancy at enhancers aligned at their midpoint ..................................................................................................... 160 Figure 4.S8. Nucleosome occupancy at enhancers aligned at their PU.1 or p300 peaks .................................................................................. 161 Figure 5.1. Isolation of HSPCs ..................................................................... 177 Figure 5.2. Nucleosome occupancy in HSPCs and differentiating BMDMs .. ..................................................................................................... 181 Figure 5.3. Protection of total cellular chromatin against digestion by MNase in HSPCs and in differentiating BMDMs ...................... 185 Figure 5.4. Nucleosome occupancy in ESCs and differentiated MEFs .... 187 xiii KEY TO ABBREVIATIONS AP1 Activator protein 1 APC Allophycocyanin BAF BRG1- or HBRM-associated factors BAF155 BRG1-associated factor 155 BMDM Bone marrow derived macrophage BRG1 Brahma-related gene 1 (see also: SMARCA4) C/EBPβ CCAAT/enhancer-binding protein beta CSFR1 Colony stimulating factor 1 receptor DNA Deoxyribonucleic acid DNase I Deoxyribonuclease I EDTA Ethylenediaminetetraacetic acid ESC Embryonic stem cell FACS Fluorescence-activated cell sorting FBS Fetal bovine serum FITC Fluorescein isothiocyanate FPKM Fragments Per Kilobase of transcript per Million mapped reads H3K4me1 Histone H3 lysine 4 monomethyl H3K4me3 Histone H3 lysine 4 trimethyl H3K27ac Histone H3 lysine 27 acetyl H3K27me3 Histone H3 lysine 27 trimethyl HSPC Hematopoietic stem and progenitor cell IFNB1 Interferon B1 xiv IL1A Interleukin 1 alpha IL12B Interleukin 12B IRF Interferon regulatory factor KD Knockdown LPS Lipopolysaccharides M-CSF Macrophage colony-stimulating factor MNase Micrococcal nuclease PBAF Polybromo-associated BAF (see also: BAF) PCR Polymerase chain reaction PEG Polyethylene glycol PRC2 Polycomb repressive complex 2 PUER PU.1-estrogen receptor chimera RNA Ribonucleic acid RPL4 Ribosomal protein L4 shRNA short hairpin RNA SMARCA4 SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily A, member 4 (see also: BRG1) SMARCB1 SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily B, member 1 (see also: SNF5) SNF5 Sucrose nonfermenting 5 (see also: SMARCB1) SWI/SNF SWItch/Sucrose Non-Fermentable TF Transcription factor TLR4 Toll-like receptor 4 TSS Transcription start site xv XBP X-box binding protein xvi Chapter 1: Introduction 1 Chromatin is a barrier to transcription Eukaryotes have large genomes that must be condensed and extensively packaged into a protein-DNA complex called chromatin due to the spatial restrictions of the nucleus. The basic repeating element of this complex is the nucleosome, an octamer composed of two subunits each of the core histone proteins (H2A, H2B, H3, and H4). This octamer forms a highly basic globular core around which approximately 147 base pairs of DNA is tightly wound (1). Because of its role in compaction, the nucleosome has long been perceived as a barrier that must be overcome in order for the transcriptional machinery to bind. This perception was reinforced by early studies that identified “nucleosome free” or “nucleosome depleted” regions at specific transcriptional promoters (2). More recent genome-wide studies in the yeast S. cerevisiae have indicated that many promoter regions are indeed relatively depleted of nucleosomes compared to the surrounding regions (3,4,5), and studies at individual genes such as the yeast PHO5 and GAL1/10 loci have found that promoter nucleosome removal is required for gene induction. This process is mediated by nucleosome remodelers (e.g. the SWI/SNF complex), which are recruited to these promoters by specific transcription factors (TFs) (6,7). Nucleosomal sites at the GAL1/10 promoters are lowly occupied even before induction, allowing rapid nucleosome removal when the inducer galactose is added (8). Together, these studies suggest that transcriptional regulatory regions in yeast must be cleared of nucleosomes to allow binding of both cis-regulatory TFs and the transcriptional machinery. Genome-wide studies have suggested that active promoters are also relatively depleted of nucleosomes in mammalian systems (9,10). In higher order organisms, however, 2 promoter nucleosome depletion is often cell type-specific and limited to genes that are expressed in a particular lineage. The cKit gene promoter, for instance, is nucleosome free in mast cells, where the gene is constitutively expressed, but not in other cell types (11). Changes in nucleosome occupancy—as determined by sensitivity to micrococcal nuclease (MNase)—associated with changes in gene expression have also been detected in differentiating embryonic stem cells (ESCs) (10,12), suggesting that nucleosomes may be placed at or removed from promoters in specific cell types in order to facilitate or silence the expression of the associated gene. Mouse macrophages as a model for the study of transcriptional regulation The controlled access of transcription factors to DNA binding sites is particularly important in higher eukaryotes, where gene expression programs are often restricted to specific cell lineages. One such program is the pro-inflammatory response which may be activated in mature macrophages, a form of white blood cell whose development requires expression of the lineage-specific TFs PU.1 and C/EBPβ in order to properly differentiate from hematopoietic stem cells (HSCs) (14,15). Macrophages are responsible for the stimulation of other immune cells via the release of pro-inflammatory gene products. When macrophages are exposed to a foreign pathogen—an event which may be simulated in vitro via the addition of bacterial lipopolysaccharide (LPS) to culture media—the toll-like receptor 4 (TLR4) pathway is activated to stimulate cytokine production (for review, see (15)). The inflammation program in macrophages is thus a critical component of the healthy immune response to pathogens. Misregulation of inflammation in these cells has, however, been implicated in a number of autoimmune diseases and cancers, as well as diabetes, demonstrating the importance of tightly 3 controlled expression of pro-inflammatory genes. In addition to its role in human disease, the inducible nature of the inflammatory response makes it an ideal model for studying the role of chromatin architecture in highly regulated gene expression programs. Open chromatin is a feature of lineage-specific enhancers In addition to their promoters, mammalian genes are often regulated by more distal elements called enhancers that may be thousands of base pairs away from the associated gene (for review, see (16)). These elements may be marked by regions of “open” chromatin—as determined by their sensitivity to nucleases like DNase I and MNase—in a particular cell type in which the associated genes are active, indicating that regulatory elements may be rendered accessible for transcription factor binding in a lineage-specific manner (10,17,18). These regions were first identified in detailed studies at particular loci (19), and studies by the Smale laboratory were the first to identify a distal enhancer 10 kilobases (kb) upstream of the Il12b gene. This element was shown to be involved in Il12b expression upon LPS induction, and reporter assays that mimicked the endogenous nucleosome environment confirmed that this putative regulatory region did indeed enhance Il12b expression (20). The advent of genome-wide techniques such as chromatin immunoprecipitation sequencing (ChIP-seq) has allowed the identification of many thousands of putative enhancers of pro-inflammatory and macrophage-specific gene expression based on the presence of histone modifications (histone H3 lysine 4 monomethylation (H3K4me1) and histone H3 lysine 27 acetylation (H3K27ac)) and transcriptional co- 4 activators/histone-modifying enzymes (p300) (21,22). These elements often contain binding sites for three of the primary signal-induced TFs required for pro-inflammatory gene expression: NFκB, AP1 and IRF3/7 (23). Macrophage-specific enhancers are also typically associated with the lineage-specific TFs PU.1 and/or C/EBPβ even before induction with LPS (21), and both of these TFs have been shown to be required for proinflammatory gene expression (21,22,24). Although these studies have proven useful in identifying regions that may act as transcriptional enhancers, the data gathered does not provide detailed information about nucleosome occupancy and positioning, nor do they always provide direct evidence for the function of these elements. Therefore, it is still unknown how small differences in chromatin architecture may contribute to large differences in gene expression. Pioneer factors in macrophage differentiation In addition to their direct role in gene expression, there is mounting evidence that PU.1 and CEBP are pioneer factors, a subset of lineage-specific TFs expressed early during differentiation that have the unique ability among TFs to bind their sites on chromatinized DNA (for review, see (25)). PU.1 binding is detected at macrophage-specific enhancers even before induction with LPS, and has been shown to lead to nucleosome depletion at these sites by ourselves and others (26,27), suggesting that PU.1 may “prime” enhancers for subsequent transcription factor binding and activity even in the absence of stimuli. Coimmunoprecipitation studies have also shown that PU.1 may directly interact with members of the BAF complex, the mammalian homolog of the yeast SWI/SNF complex 5 (28, see also Chromatin remodeling complexes and transcription below). The lineagespecific TF CEBPβ also binds macrophage-specific enhancers, often in a PU.1 dependent manner (22). It has also been reported that CEBPβ contains a SWI/SNF interaction domain, which may allow the recruitment of the Brg1 subunit of BAF/PBAF in vitro (29). Elegant transdifferentiation studies by Thomas Graf’s laboratory have demonstrated that expression of CEBPβ and PU.1 is sufficient to convert both B cells (30) and fibroblasts (14) into macrophage-like cells. These results suggest that forced expression of these TFs may render previously inaccessible regulatory elements accessible in differentiated cells, providing further evidence that both of the TFs required for macrophage differentiation (i.e., PU.1 and CEBPβ) may interact directly with chromatin and/or chromatin remodelers. Chromatin remodeling complexes and transcription Nucleosome remodelers are large protein complexes capable of sliding or removing nucleosomes from DNA in vitro, and there is evidence that the SWI/SNF family of remodelers plays a direct role in facilitating TF binding and subsequent gene expression in vivo. This complex has been well studied in yeast, where it removes nucleosomes from the PHO5 and GAL1/10 promoters upon induction of those genes (6,7). The related RSC (Remodeling the Structure of Chromatin) complex has been shown to properly position nucleosomes at regulatory regions before induction (31,32) and partially unwrap nucleosomes in order to expose TF binding sites (33). The mammalian BAF and PBAF complexes are the functional homologs of the yeast SWI/SNF and RSC complexes, respectively. The mammalian complexes share the core subunits Brg1, Baf170, Baf155, and Snf5, all of which are required for full nucleosome 6 remodeling activity in vitro (34). Each complex also contains unique subunits that may contribute to differential binding and/or function in vivo (for review, see (35)). Both complexes are capable of incorporating the catalytic subunit Brg1, but BAF may also utilize the alternate catalytic subunit Brm. Knockout studies in mice have shown that Brg1 deletion is early embryonic lethal (36), but Brm-/- embryos develop normally, and it has been suggested that Brm-/- cells may compensate for the loss of Brm through the upregulation of Brg1 (37). Brg1 is also required for the differentiation of a variety of cell types, including lymphoid (38) and myeloid (29) lineages, and Brg1 appears to be recruited to cell type-specific genes during erythroid differentiation (39). Taken together, these studies suggest that Brg1 plays a key role in the differentiation of lineages derived from HSCs. Because of their role in evicting or sliding nucleosomes, chromatin remodeling complexes have long been of great interest in the study of transcriptional regulation, and studies at numerous genes in different cell types have shown an increased sensitivity to nucleases like DNase I and MNase at both promoters and enhancers upon expression of the associated genes, indicating that nucleosomes may be removed from these sites in order to allow binding of TFs or the transcriptional machinery (see for example (40,41)). At the inducible human interferon β gene, for example, the promoter is cleared of nucleosomes upon viral induction, leaving the TATA box accessible for binding (42). Genome-wide studies suggest that Brg1 is recruited to many inducible genes (43,44), and a recent study classified pro-inflammatory genes according to Brg1 dependence upon LPS induction (45) in a mouse macrophage cell line. Simultaneous knockdown of 7 both Brg1 and Brm was sufficient to impair the expression of a subset of proinflammatory genes when the cells were exposed to stimuli, while changes in the expression of other genes was minimal or unchanged. These authors suggested a role for Brg1/Brm in altering chromatin structure at non-CpG island promoters, and concluded that the expression of secondary response genes, as well as that of a subset of primary response genes, was dependent on Brg1/Brm. They did not, however, investigate the role of BAF/PBAF at pro-inflammatory enhancers, and the role that chromatin remodelers may play at enhancers therefore remains an area of intense investigation. Unique chromatin states in multipotent progenitors and stem cells Although the regulation of pro-inflammatory genes in macrophages has been an active area of study for some time, comparatively little is known about chromatin in the hematopoietic stem and progenitor cells (HSPCs) that the myeloid lineage is derived from. It thus remains unclear how lineage-specific TFs like PU.1 and CEBPβ might initially access their binding sites during differentiation. A growing number of recent studies suggest that chromatin of other multipotent stem cells may be more accessible to DNA binding proteins, however. A genome-wide MNase-seq study investigated nucleosome binding in embryonic stem cells (ESCs), mouse embryonic fibroblasts (MEFs), and neural progenitor cells (NPCs) suggested that ESC differentiation to these two lineages was associated with changes in nucleosome positioning at regulatory elements (10). Further, the authors concluded that relative nucleosome occupancy—as determined by sensitivity to MNase digestion—at various TF binding sites differed between cell types, suggesting that TFs that are active in a particular lineage may be 8 associated with nucleosome free or nucleosome depleted sites. How nucleosomes might be depleted from these sites remains unclear, but a recent study from our laboratory investigating the role of PU.1 in macrophage differentiation suggested that lineage-specific factors may bind regulatory elements early during differentiation in order to prevent heterochromatin formation, keeping these loci accessible to TF binding in mature cells (27). Total chromatin of ESCs has been shown to be more accessible to digestion by either DNase I or MNase than that of differentiated cells (46), and the number of DNase I hypersensitive sites present in ESCs decreases as cells differentiate (47). ChIP-seq studies have also shown that more of the genome is associated with “active” histone modifications (H3/H4 acetylation) in ESCs when compared to differentiated cells (48). Furthermore, less of the genome is associated with repressive histone modifications that may be associated with heterochromatin (histone H3 lysine 9 trimethylation, H3K9me3), and modest levels of transcription were detected from much of the genome in an RNA-seq study (49), suggesting that most ESC DNA may be accessible to the transcriptional machinery. This global transcription is not detected in differentiated cells, and a recent study utilizing super-resolution nanoscopy determined that chromatin becomes more compacted as ESCs differentiate (50), suggesting that increased compaction may silence much of the genome in differentiated cells. These studies suggest that chromatin of ESCs is qualitatively different from that of differentiated cells. Although chromatin of HSPCs has not been rigorously characterized, changes in chromatin compaction have long been observed in hematopoietic cells as well, and these changes have been used as a measure of a 9 cell’s differentiation state (51). This suggests that there may be analogous differences between chromatin of HSPCs and that of mature cells of the hematopoietic lineage, but this hypothesis remains to be investigated. Clinical significance Chronic inflammation is characterized by the prolonged release of pro-inflammatory gene products and repeated activation of the innate immune system, and a number of diseases, including rheumatoid arthritis, are the direct result of this aberrant inflammation. The prevalence of these diseases—classified as immune-mediated inflammatory diseases (IMIDs)—is estimated to be 5-7% in Western society (52). Furthermore, chronic or prolonged inflammation has been implicated in the onset and progression of a number of other diseases including type 2 diabetes (53), Alzheimer’s disease (54), and cancer (55). It is therefore crucial to further dissect the underlying mechanism regulating the expression of pro-inflammatory genes, and extensive characterization of the regulatory elements associated with these genes may provide insight into novel therapeutic targets. 10 REFERENCES 11 REFERENCES 1. Luger, K., Mader, A. W., Richmond, R. K., Sargent, D. F., and Richmond, T. J. (1997) Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389, 251-260. 2. Almer, A., and Horz, W. (1986) Nuclease hypersensitive regions with adjacent positioned nucleosomes mark the gene boundaries of the PHO5/PHO3 locus in yeast. The EMBO journal 5, 2681-2687. 3. Yuan GC, Liu YJ, Dion MF, Slack MD, Wu LF, et al. (2005) Genome-scale identification of nucleosome positions in S. cerevisiae. Science 309: 626–630. 4. Bernstein BE, Liu CL, Humphrey EL, Perlstein EO, Schreiber SL (2004) Global nucleosome occupancy in yeast. Genome biology 5: R62. 5. Sekinger EA, Moqtaderi Z, Struhl K (2005) Intrinsic histone-DNA interactions and low nucleosome density are important for preferential accessibility of promoter regions in yeast. Molecular cell 18: 735–748. 6. Bryant GO, Prabhu V, Floer M, Wang X, Spagna D, et al. (2008) Activator control of nucleosome occupancy in activation and repression of transcription. PLoS biology 6: 2928–2939. 7. Reinke H, Horz W (2003) Histones are first hyperacetylated and then lose contact with the activated PHO5 promoter. Molecular cell 11: 1599–1607. 8. Wang X, Bryant GO, Floer M, Spagna D, Ptashne M (2011) An effect of DNA sequence on nucleosome occupancy and removal. Nature structural & molecular biology 18: 507–509. 9. Schones DE, Cui K, Cuddapah S, Roh TY, Barski A, et al. (2008) Dynamic regulation of nucleosome positioning in the human genome. Cell 132: 887–898. 10. Teif VB, Vainshtein Y, Caudron-Herger M, Mallm JP, Marth C, Hofer T, Rippe K (2012) Genome-wide nucleosome positioning during embryonic stem cell development. Nature structural & molecular biology 19: 1185-1192. 11. Berrozpe G, Bryant GO, Warpinski K, Ptashne M (2013) Regulation of a mammalian gene bearing a CpG island promoter and a distal enhancer. Cell reports 4: 445–453. 12. Carey, T. S., Choi, I., Wilson, C. A., Floer, M., and Knott, J. G. (2014) Transcriptional reprogramming and chromatin remodeling accompanies Oct4 and 12 Nanog silencing in mouse trophoblast lineage. Stem cells and development 23, 219-229. 13. Rosenbauer, F., and Tenen, D. G. (2007) Transcription factors in myeloid development: balancing differentiation with transformation. Nature reviews immunology 7, 105-117. 14. Feng, R., Desbordes, S. C., Xie, H., Tillo, E. S., Pixley, F., Stanley, E. R., and Graf, T. (2008) PU.1 and C/EBPalpha/beta convert fibroblasts into macrophage-like cells. Proceedings of the National Academy of Sciences of the United States of America 105, 6057-6062. 15. Medzhitov R (2007) Recognition of microorganisms and activation of the immune response. Nature 449: 819–826. 16. Levine M (2010) Transcriptional enhancers in animal development and evolution. Current biology: CB 20: R754–763. 17. Song, L., Zhang, Z., Grasfeder, L. L., Boyle, A. P., Giresi, P. G., Lee, B. K., Sheffield, N. C., Graf, S., Huss, M., Keefe, D., Liu, Z., London, D., McDaniell, R. M., Shibata, Y., Showers, K. A., Simon, J. M., Vales, T., Wang, T., Winter, D., Zhang, Z., Clarke, N. D., Birney, E., Iyer, V. R., Crawford, G. E., Lieb, J. D., and Furey, T. S. (2011) Open chromatin defined by DNaseI and FAIRE identifies regulatory elements that shape cell-type identity. Genome research 21, 1757-1767. 18. Stamatoyannopoulos, J. A., Snyder, M., Hardison, R., Ren, B., Gingeras, T., Gilbert, D. M., Groudine, M., Bender, M., Kaul, R., Canfield, T., Giste, E., Johnson, A., Zhang, M., Balasundaram, G., Byron, R., Roach, V., Sabo, P. J., Sandstrom, R., Stehling, A. S., Thurman, R. E., Weissman, S. M., Cayting, P., Hariharan, M., Lian, J., Cheng, Y., Landt, S. G., Ma, Z., Wold, B. J., Dekker, J., Crawford, G. E., Keller, C. A., Wu, W., Morrissey, C., Kumar, S. A., Mishra, T., Jain, D., ByrskaBishop, M., Blankenberg, D., Lajoie, B. R., Jain, G., Sanyal, A., Chen, K. B., Denas, O., Taylor, J., Blobel, G. A., Weiss, M. J., Pimkin, M., Deng, W., Marinov, G. K., Williams, B. A., Fisher-Aylor, K. I., Desalvo, G., Kiralusha, A., Trout, D., Amrhein, H., Mortazavi, A., Edsall, L., McCleary, D., Kuan, S., Shen, Y., Yue, F., Ye, Z., Davis, C. A., Zaleski, C., Jha, S., Xue, C., Dobin, A., Lin, W., Fastuca, M., Wang, H., Guigo, R., Djebali, S., Lagarde, J., Ryba, T., Sasaki, T., Malladi, V. S., Cline, M. S., Kirkup, V. M., Learned, K., Rosenbloom, K. R., Kent, W. J., Feingold, E. A., Good, P. J., Pazin, M., Lowdon, R. F., and Adams, L. B. (2012) An encyclopedia of mouse DNA elements (Mouse ENCODE). Genome biology 13, 418. 19. McGhee, J. D., Wood, W. I., Dolan, M., Engel, J. D., and Felsenfeld, G. (1981) A 200 base pair region at the 5' end of the chicken adult beta-globin gene is accessible to nuclease digestion. Cell 27, 45-55. 13 20. Zhou L, Nazarian AA, Xu J, Tantin D, Corcoran LM, et al. (2007) An inducible enhancer required for Il12b promoter activity in an insulated chromatin environment. Molecular and cellular biology 27: 2698–2712. 21. Ghisletti S, Barozzi I, Mietton F, Polletti S, De Santa F, et al. (2010) Identification and characterization of enhancers controlling the inflammatory gene expression program in macrophages. Immunity 32: 317–328. 22. Heinz S, Benner C, Spann N, Bertolino E, Lin YC, et al. (2010) Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Molecular cell 38: 576– 589. 23. Kawai T, Akira S (2006) TLR signaling. Cell death and differentiation 13: 816– 825. 24. Grove M, Plumb M (1993) C/EBP, NF-kappa B, and c-Ets family members and transcriptional regulation of the cell-specific and inducible macrophage inflammatory protein 1 alpha immediate-early gene. Molecular and cellular biology 13: 5276–5289. 25. Zaret, K. S., and Carroll, J. S. (2011) Pioneer transcription factors: establishing competence for gene expression. Genes & development 25, 2227-2241. 26. Barozzi, I., Simonatto, M., Bonifacio, S., Yang, L., Rohs, R., Ghisletti, S., and Natoli, G. (2014) Coregulation of transcription factor binding and nucleosome occupancy through DNA features of mammalian enhancers. Molecular cell 54, 844-857. 27. Tagore, M., McAndrew, M. J., Gjidoda, A., and Floer, M. (2015) The lineagespecific transcription factor PU.1 prevents polycomb-mediated heterochromatin formation at macrophage-specific genes. Molecular and cellular biology 35, 26102625. 28. Gu, X., Hu, Z., Ebrahem, Q., Crabb, J. S., Mahfouz, R. Z., Radivoyevitch, T., Crabb, J. W., and Saunthararajah, Y. (2014) Runx1 regulation of Pu.1 corepressor/coactivator exchange identifies specific molecular targets for leukemia differentiation therapy. The Journal of biological chemistry 289, 14881-14895. 29. Kowenz-Leutz, E., and Leutz, A. (1999) A C/EBP beta isoform recruits the SWI/SNF complex to activate myeloid genes. Molecular cell 4, 735-743. 30. Xie H, Ye M, Feng R, Graf T, York N. (2004) Stepwise reprogramming of B cells into macrophages. Cell 117:663-676. 31. Badis, G., Chan, E. T., van Bakel, H., Pena-Castillo, L., Tillo, D., Tsui, K., Carlson, C. D., Gossett, A. J., Hasinoff, M. J., Warren, C. L., Gebbia, M., Talukder, S., 14 Yang, A., Mnaimneh, S., Terterov, D., Coburn, D., Li Yeo, A., Yeo, Z. X., Clarke, N. D., Lieb, J. D., Ansari, A. Z., Nislow, C., and Hughes, T. R. (2008) A library of yeast transcription factor motifs reveals a widespread function for Rsc3 in targeting nucleosome exclusion at promoters. Molecular cell 32, 878-887. 32. Hartley, P. D., and Madhani, H. D. (2009) Mechanisms that specify promoter nucleosome location and identity. Cell 137, 445-458 33. Floer, M., Wang, X., Prabhu, V., Berrozpe, G., Narayan, S., Spagna, D., Alvarez, D., Kendall, J., Krasnitz, A., Stepansky, A., Hicks, J., Bryant, G. O., and Ptashne, M. (2010) A RSC/nucleosome complex determines chromatin architecture and facilitates activator binding. Cell 141, 407-418. 34. Phelan ML, Sif S, Narlikar GJ, Kingston RE (1999) Reconstitution of a core chromatin remodeling complex from SWI/SNF subunits. Molecular cell 3: 247-253. 35. Tang, L., Nogales, E., and Ciferri, C. (2010) Structure and function of SWI/SNF chromatin remodeling complexes and mechanistic implications for transcription. Progress in biophysics and molecular biology 102, 122-128. 36. Bultman, S., Gebuhr, T., Yee, D., La Mantia, C., Nicholson, J., Gilliam, A., Randazzo, F., Metzger, D., Chambon, P., Crabtree, G., and Magnuson, T. (2000) A Brg1 null mutation in the mouse reveals functional differences among mammalian SWI/SNF complexes. Molecular cell 6, 1287-1295. 37. Reyes, J. C., Barra, J., Muchardt, C., Camus, A., Babinet, C., and Yaniv, M. (1998) Altered control of cellular proliferation in the absence of mammalian brahma (SNF2alpha). The EMBO journal 17, 6979-6991. 38. Choi, J., Ko, M., Jeon, S., Jeon, Y., Park, K., Lee, C., Lee, H., and Seong, R. H. (2012) The SWI/SNF-like BAF complex is essential for early B cell development. Journal of immunology 188, 3791-3803. 39. Hu, G., Schones, D. E., Cui, K., Ybarra, R., Northrup, D., Tang, Q., Gattinoni, L., Restifo, N. P., Huang, S., and Zhao, K. (2011) Regulation of nucleosome landscape and transcription factor targeting at tissue-specific enhancers by BRG1. Genome research 21, 1650-1658. 40. Lefevre P, Lacroix C, Tagoh H, Hoogenkamp M, Melnik S, et al. (2005) Differentiation-dependent alterations in histone methylation and chromatin architecture at the inducible chicken lysozyme gene. The Journal of biological chemistry 280: 27552–27560. 41. Bert AG, Johnson BV, Baxter EW, Cockerill PN (2007) A modular enhancer is differentially regulated by GATA and NFAT elements that direct different tissue- 15 specific patterns of nucleosome positioning and inducible chromatin remodeling. Molecular and cellular biology 27: 2870–2885. 42. Agalioti T, Lomvardas S, Parekh B, Yie J, Maniatis T, et al. (2000) Ordered recruitment of chromatin modifying and general transcription factors to the IFNbeta promoter. Cell 103: 667–678. 43. Burd, C. J., and Archer, T. K. (2013) Chromatin architecture defines the glucocorticoid response. Molecular and cellular endocrinology 380, 25-31. 44. Sena JA, Wang L, Hu CJ (2013) BRG1 and BRM chromatin-remodeling complexes regulate the hypoxia response by acting as coactivators for a subset of hypoxiainducible transcription factor target genes. Molecular and cellular biology 33: 38493863. 45. Ramirez-Carrozzi, V. R., Braas, D., Bhatt, D. M., Cheng, C. S., Hong, C., Doty, K. R., Black, J. C., Hoffmann, A., Carey, M., and Smale, S. T. (2009) A unifying model for the selective regulation of inducible transcription by CpG islands and nucleosome remodeling. Cell 138, 114-128. 46. Schaniel C, Ang YS, Ratnakumar K, Cormier C, James T, Bernstein E, Lemischka IR, Paddison PJ (2009) Smarcc1/Baf155 couples self-renewal gene repression with changes in chromatin structure in mouse embryonic stem cells. Stem cells 27: 2979-2991. 47. Stergachis AB, Neph S, Reynolds A, Humbert R, Miller B, Paige SL, Vernot B, Cheng JB, Thurman RE, Sandstrom R, Haugen E, Heimfeld S, Murry CE, Akey JM, Stamatoyannopoulos JA (2013) Developmental fate and cellular maturity encoded in human regulatory DNA landscapes. Cell 154: 888-903. 48. Meshorer E, Yellajoshula D, George E, Scambler PJ, Brown DT, Misteli T (2006) Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Developmental cell 10: 105-116. 49. Efroni S, Duttagupta R, Cheng J, Dehghani H, Hoeppner DJ, Dash C, BazettJones DP, Le Grice S, McKay RD, Buetow KH, Gingeras TR, Misteli T, Meshorer E (2008) Global transcription in pluripotent embryonic stem cells. Cell stem cell 2: 437-447. 50. Ricci MA, Manzo C, Garcia-Parajo MF, Lakadamyali M, Cosma MP (2015) Chromatin fibers are formed by heterogeneous groups of nucleosomes in vivo. Cell 160: 1145-1158. 51. Kernell AM, Bolund L, Ringertz NR (1971) Chromatin changes during erythropoiesis. Experimental cell research 65: 1-6. 16 52. Kuek, A., Hazleman, B. L., & Östör, A. J. (2007). Immune-mediated inflammatory diseases (IMIDs) and biologic therapy: a medical revolution. Postgraduate medical journal, 83(978), 251-260. 53. Shoelson, S. E., Lee, J., and Goldfine, A. B. (2006) Inflammation and insulin resistance. The Journal of clinical investigation 116, 1793-1801. 54. Heppner, F. L., Ransohoff, R. M., and Becher, B. (2015) Immune attack: the role of inflammation in Alzheimer disease. Nature Reviews Neuroscience 16, 358-372. 55. Crusz, S. M., and Balkwill, F. R. (2015) Inflammation and cancer: advances and new agents. Nature Reviews Clinical Oncology 12, 584-596. 17 Chapter 2: Nucleosomes are stably evicted from enhancers but not promoters upon induction of certain pro-inflammatory genes in mouse macrophages This chapter represents a manuscript that was published in PLoS One (2014) 9: e93971. Portions of this manuscript describing results at the IFNB1 locus have been removed. Authors who contributed to this study were: Alison Gjidoda, Mohita Tagore, Michael McAndrew, Alexander Woods, and Monique Floer. 18 Abstract Chromatin is thought to act as a barrier for binding of cis-regulatory transcription factors (TFs) to their sites on DNA and recruitment of the transcriptional machinery. Here we have analyzed changes in nucleosome occupancy at the enhancers as well as at the promoters of three pro-inflammatory genes when they are induced by bacterial lipopolysaccharides (LPS) in primary mouse macrophages. We find that nucleosomes are removed from the distal enhancers of Il12b and Il1a, as well as from the distal and proximal enhancers of Ifnb1, and that clearance of enhancers correlates with binding of various cis-regulatory TFs. We further show that for Ifnb1 the degree of nucleosome removal correlates well with the level of induction of the gene under different conditions. Surprisingly, we find that nucleosome occupancy at the promoters of Il12b and Il1a does not change significantly when the genes are induced, and that a considerably fraction of the cells is occupied by nucleosomes at any given time. We hypothesize that competing nucleosomes at the promoters of Il12b and Il1a may play a role in limiting the size of transcriptional bursts in individual cells, which may be important for controlling cytokine production in a population of immune cells. Introduction Genome-wide studies in S. cerevisiae have indicated that promoter regions are relatively depleted of nucleosomes compared to the surrounding regions (1,2,3). Where it has been analyzed, for example at the PHO5 and GAL1/10 genes of yeast, it was found that removal of promoter nucleosomes is required for gene induction and is mediated by nucleosome remodelers (e.g. the SWI/SNF complex) that are recruited to these regions by specific TFs (4,5). At the GAL1/10 promoters these nucleosomal sites 19 are only lowly occupied prior to induction and low promoter nucleosome occupancy is at least partly determined by the underlying DNA-sequence and facilitates rapid nucleosome removal when the inducer galactose is added (6). These studies have suggested that transcriptional regulatory regions have to be nucleosome-free to allow binding of cis-regulatory TFs and the transcriptional machinery. However, at least at one site of binding of a transcriptional activator, the UASg of the GAL1/10 locus, it was shown that the consensus site-containing piece of DNA is part of an, albeit unusual, nucleosome that apparently accommodates activator binding on its surface (7). Genome-wide studies in mammalian systems have similarly suggested that promoters are relatively depleted of nucleosomes (8,9) and a recent study that analyzed the constitutively expressed cKIT gene in mast cells showed that the promoter was nucleosome-free in this cell-type but not in others (10). In addition, studies at many different genes in various cell-types that used changes in sensitivity of chromatin to the enzyme micrococcal nuclease (MNase), to Dnase I or to restriction enzymes, found that chromatin architecture was altered at promoters and enhancers when these genes were expressed indicating that nucleosomes are remodeled at these sites (see for example (11,12)). In one well-studied example of an inducible gene, human interferon β, it was found that the promoter was cleared of nucleosomes upon viral induction, which led to clearing of the TATA-box (13). The interferon β gene contains a promoter proximal enhancer, which forms an enhanceosome (14), and this close proximity of TF-binding sites to the transcriptional start site (TSS) resembles the typical gene architecture in yeast where TF-binding sites are usually within 500 bp of the TSS. However, other mammalian genes are often regulated by distal enhancer elements that can be 20 thousands of base pairs away (for a recent review see (15)), and are thought to be brought in contact with the promoter by DNA-looping (for an example see (16)). This separation of enhancers and promoters at many mammalian genes prompted us to investigate the changes in nucleosome binding associated with either transcriptional regulatory element upon gene induction. We have used a quantitative assay to analyze changes in nucleosome occupancy at enhancers and promoters of three proinflammatory cytokines – Il1a, Il12b and Ifnb1 - upon their induction by LPS in primary mouse macrophages. The assay uses a wide range of MNase concentrations and detects the distinct digestion rates of the same segment of DNA, when it is naked or associated with a nucleosome, which allows us to derive the fractional occupancy of a genomic region by a nucleosome (4). Pro-inflammatory cytokines are expressed by macrophages as part of the innate immune response to various pathogens (for review see (17)) and requires the action of three main TFs, NFκB, AP1 and IRF3/7 (18). Binding sites for these TFs are found in the regulatory elements of many pro-inflammatory genes (19,20). In addition to these signal-induced TFs at least two lineage-specific TFs, PU.1 and C/EBPβ, are required for macrophage differentiation and expression of certain pro-inflammatory genes (21,22,23,24). Both of these TFs have been found to be associated with regulatory elements of many genes even prior to their induction in macrophages (19,20,25). The promoter proximal enhancer of Ifnb1 is conserved in mice (26), but mouse Ifnb1 was recently shown to also be regulated by a distal enhancer located 6 kb downstream of its TSS (27). This region was found to also bind the cis-regulatory TF XBP when Ifnb1 was induced by LPS and thapsigargin (TPG), an inducer of ER-stress that enhances 21 expression of certain pro-inflammatory cytokines through the action of XBP. Furthermore, a minimal region of 305 bp that encompasses consensus-sites for XBP and IRF3 was shown to enhance transcription of a reporter gene confirming this region as a bona fide enhancer. Similar studies of the Il12b gene performed mostly by Stephen Smale's laboratory identified a distal enhancer located 10 kb upstream of its TSS (28). This distal enhancer was shown to play a role in LPS induction of Il12b and was further found to strongly enhance Il12b expression in reporter assays that mimic the nucleosome environment found at the endogenous gene (28). The distal enhancers of Il12b and Ifnb1 were also classified as enhancers in two recent genome-wide studies (19,20) that identified thousands of putative enhancers including a region located 10 kb upstream of the Il1a gene, which we have included in our studies as a putative enhancer for Il1a. We find that nucleosomes in the distal enhancers of Il12b, Il1a and Ifnb1 are rapidly evicted when the genes are induced. Nucleosomes are also removed from the proximal enhancer of Ifnb1, which leads to clearance of the adjacent TATA-box and TSS as had been described for the human gene (13). In addition, we show that nucleosomedepletion correlates with binding of cis-regulatory TFs and the co-activator p300 to the distal enhancers of all three genes as well as to the proximal enhancer of Ifnb1. Surprisingly, we find nucleosomes at the Il12b and Il1a promoters in a large fraction of the population of cells under inducing conditions. Furthermore, we find that promoter nucleosomes around the TSSs of these genes become associated with histone modifications found at active promoters (H3K4me3 and H3K27ac). Our results indicate that promoter nucleosomes are not stably evicted but instead are bound to a fraction of 22 promoters in the population of cells at any given time. Furthermore, we find that PolII and TBP are only associated with nucleosome-free promoters and we discuss the potential role of competing nucleosomes at the promoters of these cytokine genes in limiting their expression in a population of immune cells. Experimental Procedures Cell isolation and culture Primary cells where isolated from 8–12 week old C57BL/6 mice (NCI). Bone marrow derived macrophages (BMDMs) were generated as described [29] and grown in BMDM medium (60% IMDM medium (Gibco), 30% conditioned medium from L-929 fibroblasts, 10% FBS, 0.1 mM non-essential amino acids, 1 mM sodium pyruvate and 1X penicillinstreptomycin. LPS induction was performed by adding 1 µg/ml LPS from E. coli strain EH100 (Ra mutant)(Sigma) to serum-starved BMDMs for the indicated times. Serum starvation was done by growth of cells in incomplete IMDM medium for 1 h. Other inducers were ISD (interferon stimulatory DNA) derived from Listeria monocytogenes; poly(I:C), synthetic dsRNA that acts as a TLR3 agonist; and poly(dA:dT), a synthetic analog of B-DNA (all obtained from Invivogen). 1 µg/ml of either of these inducers was given to BMDMs by transfection with Lipofectamine 2000 (Invitrogen) in an equal volume mixture [30]. Where indicated thapsigargin (Sigma) was added at 1 µM for 1 h to serum-starved cells prior to LPS addition [27]. Splenic B-cells were isolated by negative selection with CD43 antibody-coupled Dynabeads according to the instructions of the manufacturer (Life Technologies), with an additional red blood lysis step using lysis buffer (Sigma). For LPS induction B-cells were grown in B-IMDM medium (IMDM medium (Gibco), containing 55 µM 2-Mercaptoethanol and 2 mM L-glutamine) for 1.5 h 23 prior to LPS addition for the indicated times. RAW264.7 cells were grown in DMEM medium (Gibco) containing 10% FBS and 1X penicillin-streptomycin. mRNA determination Total RNA was isolated from BMDMs or B-cells using Trizol (Invitrogen/Lifetech). In brief, Trizol was added to cells growing in culture, and Trizol lysates were collected. 400 µl of chloroform was added per 1 ml Trizol lysate, the aequous phase was extracted, 170 µl isopropanol was added and the mixture was further purified on ReliaPrep RNA Cell Miniprep System columns according to the manufacturer's protocol (Promega). RNA was converted into cDNA according to the protocol described [31] except that High Capacity Reverse Transcriptase was used (Invitrogen/Lifetech) and analyzed by qRTPCR with specific primer pairs. Primers used can be given upon request. Chromatin immunoprecipitation Chromatin from 5×106 cells per antibody that had been cross-linked with 0.5% formaldehyde for 10 min was isolated by sonication with a Branson sonifier (10 pulses of 10″ at setting 4) in Lysis buffer (50 mM Hepes-KOH, pH 7.5, 1% TritonX-100, 0.1% SDS) and centrifugation for 10′ at 21,000×g. To increase the resolution of ChIP experiments when detecting histones or histone modifications, and to differentiate nucleosome binding from PolII and TBP binding, the isolated chromatin was digested with 0.5 or 1 U MNase (NEB) for 1 h 30′ in the presence of 0.15 mM CaCl2, and the digestion reaction was stopped by addition of 20 mM EDTA. Digested chromatin was diluted 3-fold with High Salt ChIP buffer (10 mM Tris-HCl, pH 8, 400 mM NaCl, 1% TritonX-100, 2 mM EDTA, Complete protease inhibitors w/o EDTA (Roche)) to yield 600 µl total volume and incubated overnight at 4°C with either 5 µl of anti-H3 (39163, Active 24 Motif, concentration is not known), 4 µg of anti-H2A.Z (ab4174; Abcam), 1 µg of antiH3K4me1 (ab8895; Abcam), 1 µg of anti-H3K4me3 (ab8580; Abcam) or 1 µg of H3K27ac (ab4729; Abcam). For all other ChIP experiments isolated chromatin was directly diluted with High Salt ChIP buffer and incubated with either 1 µg of anti-PolI antibody (sc-56767), 6 µg anti-TBP (sc-204), 4 µg anti-PU.1 (sc-352), 4 µg anti-C/EBPβ (sc-150), 6 µg anti-NFκB (sc-372), 5 µg anti-c-Jun (sc-45), 6 µg anti-p300 (sc-585) or 10 µg anti-IRF3 (sc-9082) all from Santa Cruz Biotechnologies. 20 µl of Protein A/G magnetic beads (Pierce) were added to the reaction and incubated at 4°C for 2 h. Beads were washed with 280 µl each of TSE buffer (20 mM Tris pH 8.0, 0.1% SDS, 1% TritonX-100, 2 mM EDTA), TSE250 (TSE buffer, 250 mM NaCl) and TSE500 (TSE buffer, 500 mM NaCl), Wash buffer III (10 mM Tris pH 8.5, 0.25 M LiCl, 1% NP40/Igepal, 1% deoxycholate, 1 mM EDTA) and TE (10 mM Tris-HCl pH 8.0, 1 mM EDTA) all containing Complete protease inhibitors. Antibody complexes were eluted from the beads with 2×100 µl Elution buffer (0.1 M NaHCO3, 1% SDS) by incubation for 30′ (and 10′) at 55°C. Eluates were combined and the cross-link was reversed by incubation at 65°C for 4 h. DNA was purified using a Qiagen PCR purification kit, and analyzed on a Lightcycler 480 (Roche) using primer pairs in the regions indicated. Sequences of the primers used can be given upon request. Quantitative nucleosome occupancy assay The assay was performed essentially as described in [4] with certain modifications. Cross-linked chromatin from 1 to 3×107 cells isolated as described for ChIP experiments was incubated in Lysis buffer containing 140 mM sodium chloride with 22 increasing concentrations of MNase (0.001179 U to 20 U, NEB) in the presence of 0.15 25 mM CaCl2 for 1 h 30′. DNA was purified as described and quantified using a Roche Lightcycler 480. Digestion data was analyzed using two-state exponential curve-fitting as described [4]. Data was normalized to several genomic locations, including a region in the promoter of cKIT [10] that was highly protected and a region in the ORF of Rpl4. The data was displayed in the IGV genome browser v2.3 [32] and overlays of nucleosome occupancy during a timecourse of LPS induction were created from IGV tracks using Adobe Photoshop. Genomic DNA isolation Genomic DNA was isolated from RAW264.7 macrophages as described [33] and DNA standard curves were created using a 1/3 fold dilution series with the highest concentration yielding qRT-PCR amplification at around cycle 20 for the majority of primer pairs. qRT-PCR DNA and cDNA was quantified on a Lightcycler 480 (Roche) as described [4] with the following modifications. Primers were designed using the program PCRtiler [34]. To verify that only a single amplicon was produced by each primer pair and no primer dimers were formed a Tm-curve was performed as a quality control for each primer pair at the end of each qRT-PCR run. We also found that addition of 1.5% PEG400 (Fluka) to the qRT-PCR reaction greatly enhanced performance for many primer pairs and led to a greater linear range of the qRT-PCR measurements. Results Nucleosome occupancy at the Il12b enhancer and promoter upon LPS induction Figure 2.1A and B shows an analysis of nucleosome occupancy in a 1.2 kb region 26 encompassing the 10 kb upstream enhancer of Il12b (28) at different timepoints during LPS induction of primary bone-marrow derived macrophages (BMDMs) using the assay described (4). Prior to induction (blue bars and lines) nucleosomes in the Il12b enhancer were relatively well positioned and occupied their sites in around 60% of the population of cells. 1.5 h after LPS addition (yellow) two nucleosomes in the center of the enhancer had been removed. The 5–10% occupancy we detected upon clearance of these nucleosomes is within the accuracy of our assay and we conclude that this region was largely nucleosome-free after 1.5 h. The central nucleosomal position, which remained cleared upon prolonged incubation with LPS up to 10 h (dark red), coincides with a region that was shown by Zhou et al. to become hypersensitive to Dnase I upon LPS induction (see the black bar underneath panel A (28)). We found that the flanking nucleosomes were partially re-formed as induction progressed and after 5 h of induction the nucleosome to the left was again occupied in 30% of the population (light red). The nucleosome to the right was partially removed after 1.5 h (30–40%) and regained 60% occupancy after 5 h (light red). We monitored expression of the associated Il12b gene by measuring mRNA levels during the 10 h timecourse (Figure 2.1E). Il12b mRNA was detected as early as 1.5 h after LPS addition, and levels increased for up to 5 h, after which Il12b mRNA production reached steady-state levels. Figure 2.1E also shows mRNA levels upon LPS induction of Ifnb1 and Il1a. Figure 2.1C and D shows nucleosome occupancy at the Il12b promoter including a region 600 bp upstream and 800 bp downstream of the TSS. Surprisingly, we did not find any changes in nucleosome occupancy upon LPS induction over the 10 h timecourse of LPS induction (compare blue bars and lines to increasing shades of red). The region surrounding the 27 TSS was more highly occupied by nucleosomes than the enhancer prior to induction and nucleosomes were less well positioned than in the Il12b enhancer. We found that the region directly upstream of the TSS was occupied in about 60% of the population and this region was flanked by more highly occupied nucleosomes (around 90%). A TATAA-sequence that we identified 28 bp upstream of the TSS (light blue box in C) as well as the TSS itself was found at the edge of the highly occupied nucleosome. We found that a region 400 bp downstream of the TSS that contains a TATAT-sequence was relatively lowly occupied by nucleosomes prior to induction (20–30%), which had initially suggested to us that this downstream region might function to assemble a preinitiation complex. However, a previous search for TSSs that used CAGE-analysis to detect capped mRNAs had not found any transcription starting from this downstream region, but had instead confirmed the annotated TSS for Il12b (35). We therefore conclude that the upstream TATAA-sequence is used to assemble a PIC. This conclusion was confirmed by our subsequent ChIP analysis, which detected PolII and TBP binding at this site (see Figure 2.3). 28 Figure 2.1. Changes in nucleosome occupancy upon LPS induction at a distal enhancer and the promoter of Il12b. (A) and (B) Nucleosome occupancy at an enhancer 10 kb upstream of the TSS of Il12b in BMDMs was analyzed before induction (blue bars and lines), and after 1.5 h (yellow), 3 h (orange), 5 h (light red) and 10 h (dark red) of growth of cells in the presence of 1 29 Figure 2.1. (cont’d) µg/ml LPS, using the assay described in (4) with modifications detailed in the Experimental Procedures. In brief, occupancy was measured by determining the sensitivity of cross-linked chromatin to a wide range of MNase. Digestion data for each genomic location analyzed by qRT-PCR with specific primer pairs was fitted to two-state exponential decay functions and the percentage of DNA in the population of cells found to be protected against MNase by binding of a nucleosome is indicated on the y-axis. In panel (A) each overlapping colored bar represents the length of the amplicon measured. The minimal enhancer that was shown by Zhou et al. to contain the LPS-inducible DNaseI hypersensitive site HSS1 as well as consensus-sites for Oct1/2 and C/EBPβ is indicated by the black bar (28). Consensus-sites for PU.1, NFκB, AP1 and IRF identified using ConSite are indicated. In panel (B) nucleosome occupancy at the midpoint of each amplicon measured by the experiment performed in panel (A) is indicated by a dot, with error bars showing the SEM of at least two independent measurements (10 h was measured only once). (C) and (D) BMDMs were induced as described in (A) and nucleosome occupancy in a region surrounding the TSS of Il12b was determined. The data is displayed as in panels (A) and (B) respectively. The black bar below the data in (C) indicates the TSS [35] and the light blue bars indicate putative TATA-boxes predicted by ConSite. (E) Expression of Il12b, Ifnb1 and Il1a in response to LPS. mRNA from BMDMs induced with LPS as in panel A as well as from splenic B-cells was isolated as described in the Experimental Procedures, reverse transcribed and cDNA quantified by qRT-PCR. Data was normalized to a location in the ORF of the constitutively expressed Rpl4 gene. The 30 Figure 2.1. (cont’d) SEM of at least two independent measurements is indicated (10 h timepoint was measured only once). Changes in nucleosome occupancy at the transcriptional regulatory regions of Il1a Figure 2.2 shows an analysis of nucleosome occupancy at a putative enhancer 10 kb upstream (panel A and B) and around the TSS (panel C and D) of the Il1a gene before (blue bars and lines) and upon induction of macrophages with LPS for 1.5 h (yellow) and 3 h (red). Similar to our findings at the Il12b enhancer we found that the putative Il1a enhancer was depleted of nucleosomes 1.5 h after LPS addition. This region encompasses 3–4 nucleosomes, which were occupied in 40–60% of the population prior to induction. The center of this region became essentially nucleosome free (5– 10%) and remained so even after prolonged LPS induction (3 h, red bars and lines in panels A and B). The three nucleosomes flanking this region became partially depleted upon LPS induction (20–30% occupancy after 1.5h) and occupancy of these flanking nucleosomes increased slightly upon prolonged LPS induction similar to what we had found at the Il12b enhancer (compare yellow and red bars and lines in Figure 2.2A–D). Panels C and D of Figure 2.2 show that the promoter of Il1a was not cleared of nucleosomes upon induction. We found that prior to LPS induction the Il1a promoter was less occupied by nucleosomes than the Il12b promoter. Thus, a nucleosome that incorporates the TSS and TATAA-sequence of Il1a was occupied in about 55% of the population of cells before induction. Upon LPS induction nucleosome occupancy at the TSS decreased somewhat (35% after 1.5 h, yellow bars and lines) and then increased 31 again as LPS induction progressed (45% at 3 h, red). As for Il12b, the annotated TSS was confirmed as the major TSS for Il1a by Carninci and colleagues (35) and is indicated by the black bar underneath panel C. As shown in Figure 2.1E we found that Il1a mRNA levels increased during a 10 h course of LPS induction, suggesting that Il1a transcription is sustained over this time period. We were not able to determine nucleosome occupancy in a region 100–400 bp downstream of the TSS of Il1a, since this region consists almost entirely of CTT or CCT repeats and is resistant to qPCR. Timing of enhancer nucleosome removal To determine the earliest timepoint of nucleosome removal at the distal enhancers of Il12b and Il1a we analyzed nucleosome occupancy in the centers of the two enhancers 15′, 30′, 60′ and 90′ after LPS induction. Figure 2.2E shows that the Il1a enhancer was significantly depleted 15′ after LPS induction (blue lines), whereas no nucleosomes had been removed at the Il12b enhancer at this early timepoint (red lines). Figure 2.2E indicates the fold change of nucleosome removal over the levels found before induction and nucleosome occupancy before induction was similar at the three representative locations in each enhancer. Nucleosome depletion at the Il1a enhancer was close to completion after 30′, while depletion at the Il12b enhancer had only reached 50%. After 1 h both enhancers had reached their maximal levels of nucleosome depletion. Our results show that nucleosome removal at the Il1a enhancer occurs with faster kinetics than at the Il12b enhancer. Histone modifications at the promoters and enhancers of Il12b and Il1a Figure 2.2F shows the results of ChIP experiments performed with various antibodies that detect histone H3, the histone variant H2A.Z as well as different modifications of 32 residues in H3 upon induction of BMDMs with LPS. We first confirmed that nucleosomes are evicted from the enhancers of Il12b and Il1a but not the promoters using an antibody against H3. Figure 2.2F shows that the H3 signal decreased upon LPS induction only in the regions in the enhancers where nucleosomes were evicted (compare to Figure 2.1A and 2.2A). We found similar results using an antibody against H2A.Z at the enhancers and promoters of both genes, or with an antibody detecting H3K4me1, which was previously shown to be present at the enhancers prior to and upon LPS induction (19,20). Most importantly, we detected an increase in H3K4me3 and H3K27ac at the promoters of Il12b and Il1a upon induction. Both modifications have previously been shown to be associated with actively transcribed genes (36,37) and to increase at the two genes we have investigated upon their induction (38). 33 Figure 2.2. Changes in nucleosome occupancy upon LPS induction at a putative distal enhancer and promoter of Il1a, kinetics of nucleosome removal, and changes in histone modifications. (A) and (B) Nucleosome occupancy at a putative enhancer 10 kb upstream of the TSS of Il1a was determined in BMDMs prior to (blue bars and lines) and upon 1.5 h (yellow) 34 Figure 2.2. (cont’d) or 3 h (red) induction with 1 µg/ml LPS as described in the legend of Figure 2.1. ConSite predicted consensus sites for PU.1, C/EBP, IRF, AP1 and NFκB are indicated. (C) and (D) Nucleosome occupancy at the promoter of Il1a was determined as described in panel (A) in a region surrounding the TSS of Il1a. The TSS (black bar) (35) and a putative TATA-box (blue bar) is indicated in panel (C). (E) Expression of Il12b, Ifnb1 and Il1a in response to LPS. mRNA from BMDMs induced with LPS as in panel A as well as from splenic B-cells was isolated as described in the Experimental Procedures, reverse transcribed and cDNA quantified by qRT-PCR. Data was normalized to a location in the ORF of the constitutively expressed Rpl4 gene. The SEM of at least two independent measurements is indicated (10 h timepoint was measured only once). (F) ChIP experiments with antibodies against H3 (dark blue bars), H2A.Z (light blue), H3K4me1 (green), H3K4me3 (yellow) and H3K27ac (red) were performed as described in the Experimental Procedures. For these experiments cross-linked chromatin was lightly digested with MNase before incubation with the respective antibodies to increase resolution of the ChIP signal and the data was normalized to a region in the ORF of Rpl4. Changes upon LPS induction in histone binding and histone modifications at the enhancers and promoters of Il12b and Il1a as well as at a control region in the GAPDH pseudo gene are shown as fold over levels found before induction. For H3K27ac the changes 1.5 h after LPS induction, and for all other histone variants and modifications the changes after 3 h of induction are shown. The error bars show the SEM of at least 3 independent experiments. Statistical significance of the changes in H3K4me3 and 35 Figure 2.2. (cont’d) H3K27ac upon LPS induction compared to levels found prior to induction determined by Student's T-tests is indicated (*P<0.05; **P<0.01). Binding of cis-regulatory TFs to the distal enhancers of Il12b and Il1a The minimal enhancer of Il12b was previously shown to bind C/EBPβ and Oct1/2 upon induction and consensus-sites for these TFs were identified in this region (28). We used the prediction program ConSite (39) to identify consensus-sites for other TFs involved in induction of pro-inflammatory genes in macrophages and found consensus-sites for PU.1, NFκB, AP1 and IRF3 in the region that becomes depleted upon induction (see Figure 2.1A). A similar survey of the putative enhancer of Il1a also detected consensus sites for PU.1, C/EBP, IRF3, AP1 and NFκB in the region that is depleted of nucleosomes upon LPS induction (see Figure 2.2A). To analyze binding of these TFs to the distal enhancers of Il12b and Il1a as well as recruitment of the transcriptional machinery to the enhancers and promoters we performed ChIP experiments in uninduced macrophages and cells induced for 1.5 and 3 h with LPS (Figure 2.3). We found that PolII and TBP were recruited to the TSS of both Il12b and Il1a upon induction (Figure 2.3A and B). We also found that similar amounts of PolII and TBP were recruited to the distal enhancers of both genes but not to a control region between the Il12b TSS and the distal enhancer (−7 kb). For these and all other ChIP experiments we used splenic B-cells as a control (light blue bars). The three genes we have investigated were not induced by LPS in B-cells (see Figure 2.1E, cyan bars) and no factor binding was detected (see Figure 2.3). We also determined binding of the macrophage-specific TFs PU.1 and C/EBPβ and confirmed their presence at the 36 two distal enhancers before LPS induction (Figure 2.3C and D, dark blue bars) (19, 20). Upon induction binding of both factors to the two distal enhancers increased significantly (compare yellow and orange to dark blue bars). We found similar results when we performed a ChIP experiment with an antibody for C/EBPα, indicating that both C/EBP isoforms are present (A.G. and M.F., data not shown). Furthermore, we detected binding of NFκB, c-Jun (a component of AP1) and IRF3 at the enhancers upon LPS induction (Figure 2.3E-G). The coactivator p300 was previously shown to be recruited upon LPS induction to the regions encompassing the Il12b as well as the putative Il1a enhancer (19), a finding we confirmed (Figure 2.3H). Each ChIP experiment was performed at least three times and error bars (SEM) are included. We determined the significance of the detected ChIP signals by performing Student's Ttests (Table 2.1). To obtain robust statistics we pooled all the measurements at the different loci in the enhancer or promoter regions of either gene from 3–4 independent experiments. Overall we find that binding of the cis-regulatory TFs and the co-activator p300 is significant in the enhancers, while binding of PolII and TBP is significant at both enhancers and promoters. 37 Figure 2.3. Binding of cis-regulatory TFs and recruitment of the transcriptional machinery to the regulatory regions of Il12b and Il1a upon LPS induction. (A–H) ChIP experiments in BMDMs before (dark blue bars), and upon 1.5 h (yellow) 38 Figure 2.3. (cont’d) and 3 h (orange) of LPS induction as well as in splenic B-cells (light blue) were performed as described in Experimental Procedures using antibodies that recognize (A) TBP, (B) PolII, (C) PU.1, (D) C/EBPβ, (E) NFκB, (F) c-Jun, (G) IRF3 and (H) p300. Binding data was normalized to a location in the promoter of the KIT gene, and genomic locations in relation to the TSS of Il12b or Il1a are indicated on the x-axis in each panel. Binding to a control region in the Rpl4 ORF is shown for comparison. Error bars indicate the SEM of at least three independent experiments. Statistical significance for binding in each region was determined by Student's T-tests performed for each regulatory region (see Table 2.1 for P-values). 39 Table 2.1. Statistical significance of factor binding, ChIP of Fig. 2.3. Student's tests were performed using normalized data from at least 3 (to 6) independent experiments performed with various antibodies as described in the legend of Figure 2.3. All the measurements at the 2–4 locations in each enhancer or promoter, as well as the 40 Table 2.1. (cont’d) measurements at a single location in each ORF or in the Il12b intervening region were pooled from each experiment and Student's Tests (two-tailed, equal variance) were performed on each dataset. Table 2.1 shows the P-values obtained. We compared the significance of factor binding in resting BMDMs (0 h) versus B-cells, and in BMDMs after 1.5 h or 3 h LPS induction versus binding in resting BMDMs. Binding of the transcriptional machinery to nucleosome-free Il12b and Il1a promoters To determine whether PolII and TBP might bind to the promoters of Il12b and Il1a in the presence of nucleosomes or whether the transcriptional machinery is only associated with the fraction of promoters that is nucleosome-free we performed the experiment shown in Figure 2.4. For this experiment we treated cross-linked chromatin with MNase prior to performing a ChIP experiment with antibodies detecting PolII or TBP. As seen in Figure 2.4 the PolII or TBP ChIP-signal was lost when chromatin was treated with MNase (compare solid to hatched bars). In contrast, H3, modified H3K4me3 or H3K27ac was resistant to pretreatment with MNase (see Figure 2.2F). This result indicates that only the fraction of the promoters that is nucleosome-free at any given time is associated with the transcriptional machinery. 41 Figure 2.4. PolII and TBP binding in the fraction of Il12b and Il1a promoters in a population of induced BMDMs that is nucleosome-free. (A) and (B) ChIP experiments were performed as described in the legend of Figure 2.2F with antibodies that detect (A) PolII or (B) TBP in BMDMs before (dark blue bars), and 42 Figure 2.4. (cont’d) upon 1.5 h (yellow) or 3 h (red) LPS induction. Cross-linked chromatin was either untreated (solid bars), or lightly digested with MNase (hatched bars) as described in Experimental Procedures. The data was normalized to a region in the cKIT promoter and genomic locations are indicated. The experiment was performed twice and error bars indicating the SEM are shown. Discussion Our analysis of nucleosome occupancy at the regulatory regions of three proinflammatory genes revealed that the distal enhancers of Il12b and Ifnb1 were rapidly cleared of nucleosomes when the genes were induced. The regions that became nucleosome-free include the respective minimal regions that had been shown to have bona fide enhancer activity by previous studies (see Figure 2.1A) (27,28). We found similar removal of nucleosomes in a region 10 kb upstream of Il1a, which has been suggested to be a functional enhancer of Il1a (Figure 2.2A)(19). In all three distal enhancers the nucleosome-free regions became associated with the TFs NFκB, AP1 (cJun) and IRF3 upon LPS induction, while binding of the macrophage-specific TFs PU.1 and C/EBPβ increased (see Figure 2.3). The presence of consensus-sites for these TFs was confirmed with the prediction program ConSite (39)(Figure 2.1A, 2.2A). Together our data suggest that the enhancers of these pro-inflammatory genes have to be cleared of nucleosomes to allow binding of cis-regulatory TFs, although it remains to be determined whether binding occurs only to sites that become nucleosome-free or also to putative consensus-sites found in the surrounding regions (M.F., data not shown) that remain bound by nucleosomes. Future studies will show whether removal of 43 nucleosomes from consensus-sites can be used as a criterion to distinguish functional binding-sites for specific cis-regulatory TFs in the genome from sites that remain associated with nucleosomes and may therefore not be accessible. The most surprising result of our study was the finding that the promoters of Il12b and Il1a were not cleared of nucleosomes when the genes where expressed, while nucleosomes were rapidly removed from the associated distal enhancers. Thus, we found that the TSS of Il12b was occupied in about 70% of the population prior to induction and remained essentially unchanged, while the distal enhancer became nucleosome-free in about 90% of the population (see Figure 2.1). We found similar results at the distal enhancer and promoter of Il1a (Figure 2.2). The presence of nucleosomes at the promoters before and after LPS induction was further confirmed by our histone ChIP experiments (Figure 2.2F). In these experiments, we also detected an increase in H3K4 tri-methylation and H3K27 acetylation of the highly occupied promoter nucleosomes of Il12b and Il1a in agreement with previous lower resolution studies (Figure 2.2F, yellow and red bars)(19,38). Our finding that MNase treatment abolished the PolII and TBP ChIP-signal at the Il12b and Il1a promoters (Figure 2.4) strongly suggests that the transcriptional machinery is only associated with the fraction of promoters that is nucleosome-free at any given time. We speculate that in contrast to the stable eviction of nucleosomes at enhancers, which persisted over the timecourse of our induction experiment, nucleosomes may continuously re-associate with the promoters of Il12b and Il1a. This would allow only a fraction of the cells to form a PIC at any given time. This idea is in agreement with previous findings that expression of many inducible genes, including the genes we have analyzed, is highly stochastic 44 (41,42,43,44). Another finding that supports the idea that a changing fraction of the population of cells expresses these genes at any given time, was the observation made by Smale and co-workers that expression of Il12b is not restricted to a clonal fraction of a population in a macrophage cell-line under inducing conditions (41). We hypothesize that the presence of competing nucleosomes at the promoters of these cytokines may play a role in limiting the burst size of transcription from individual cells and thus the production of cytokines in the population. We further speculate that certain histone modifications might play a role in increasing nucleosome turnover at these promoters, a hypothesis that awaits experimental confirmation. Our findings are in contrast to previous findings by Weinmann et al., which had suggested that a region about 200 to 330 bp upstream of the TSS of Il12b is nucleosome-free even prior to activation in macrophages (both in cell-lines and thioglycollate-elicited peritoneal macrophages) using sensitivity of chromatin to MNase followed by indirect end-labeling or ligation-mediated PCR to determine nucleosome binding (41). These authors had also suggested that a region downstream of the putative nucleosome-free region contained a positioned nucleosome, which they proposed to harbor putative binding sites for NFκB (Rel) and C/EBP. Upon activation they found that this region became more sensitive to various restriction enzymes as well as to Dnase I (41), and they suggested that remodeling of the positioned nucleosome might facilitate binding of cis-regulatory TFs. We did not find significant binding of NFκB or C/EBPβ to this region upon LPS induction compared to the strong binding we found at the 10 kb upstream enhancer (see Figure 2.3). Nor did we find a nucleosome-free region in the Il12b promoter prior to induction even when we extended our analysis to 45 include up to 1.5 kb upstream of the TSS of Il12b (Figure 2.1 and A.G. and M.F., unpublished data). Our quantitative MNase sensitivity assay showed that upon induction there was no significant change in the level of nucleosome occupancy at the Il12b promoter in the population of cells (Figure 2.1), which was confirmed by histone ChIP experiments (Figure 2.2F). It is possible that our assay does not detect more subtle changes in nucleosome binding that might be induced by nucleosome remodeling and which may be detected by increased sensitivity of chromatin to certain restriction enzymes or Dnase I (41). Furthermore, it is formally possible that macrophages derived from bone marrow may be different from those derived from the peritoneum or from macrophage cell-lines. Il1a contains additional regions between the 10 kb distal enhancer we have investigated and the TSS that become associated with TFs upon induction in dendritic cells (38). This might suggest that additional enhancers may also control expression of Il1a in primary macrophages, and it remains to be seen whether nucleosomes are similarly evicted from such sites. The nucleosomes that are evicted from the distal enhancers of all the genes we have analyzed are only occupied in 40–60% of a population of resting macrophages, which is lower than the occupancies we found at, for example, the TSS of Il12b and Ifnb1 (see Figure 2.1D). Our findings of moderate nucleosome occupancy at enhancers are in agreement with a previous study of an enhancer upstream of the KIT gene in mouse myeloid cells, where occupancy was found to be around 55% (10). Whether this moderate level of nucleosome occupancy allows rapid induction of these and other genes remains to be determined. We also found significant transcription factor binding at enhancers of these genes, while intervening regions (e.g. a region 7 kb 46 upstream of the TSS of Il12b) showed no binding of these factors (see Figure 2.3A and B). This finding is in agreement with the presence of the transcriptional machinery at the enhancers of other actively transcribed genes (see for example (46,47). It has been shown that DNA looping can bring distal enhancers into close proximity of promoters (16,45), and it is therefore possible that we detected PolII and TBP at the enhancers merely as a result of DNA looping. However, our experiments showed clear enrichment of signal-induced TFs and the co-activator p300 at the distal enhancers of Il12b and Il1a with very little binding at the promoters (Figure 2.3C–H). These results indicate that our ChIP assay can distinguish between genomic locations that are contacted directly by cis-regulatory TFs and the general machinery, and those that may come into proximity of these factors only indirectly as a result of DNA looping. We therefore believe that PolII and TBP are directly recruited to the distal enhancers. Our results are in agreement with previous findings that many active enhancers are transcribed and produce short eRNAs (48,49), but what the role of transcription initiating from such sites might be remains to be determined. In contrast to our findings at the Il1a and Il12b promoters we found that the TATAAsequence in the Ifnb1 promoter was cleared of nucleosomes upon induction in primary mouse macrophages as had been described for the Ifnb1 promoter in human cells (13). Ifnb1 contains a conserved proximal enhancer, which became associated with all the TFs we tested as well as with the co-activator p300 when the gene was expressed. In HeLa cells the proximal enhancer of Ifnb1 has been reported to be completely nucleosome-free prior to induction (13), but we found that in primary BMDMs the corresponding region was lowly occupied by nucleosomes prior to gene expression and 47 became completely nucleosome-free upon induction. Together, the changes in chromatin architecture at all the enhancers we have analyzed, both proximal and distal, were similar: enhancers were only moderately occupied by nucleosomes in resting macrophages and a central region was completely cleared of nucleosomes when the associated genes were induced. The size of the cleared region varied from about 1 nucleosome (at the proximal enhancer of Ifnb1) to removal of 2–3 nucleosomes in the distal enhancers of Il12b, Ifnb1 and Il1a. The small size of the nucleosome-free region in the proximal enhancer of Ifnb1 is in agreement with the assembly of an enhanceosome at this site, which forms a highly organized structure with a relatively small DNA-footprint (26). Together, our data suggest that enhancers of proinflammatory genes undergo similar changes in nucleosome occupancy regardless of their distance from a TSS, and that clearance of enhancer nucleosomes is required to allow binding of cis-regulatory TFs. Moreover, we hypothesize that removal of nucleosomes at the promoter of Ifnb1 may occur inadvertently due to its proximity to the proximal enhancer. Il1a and Ifnb1 have been classified as primary response genes while Il12b is a secondary response gene, and it has been shown that they differ in their induction kinetics as well as in their dependence on newly synthesized factors for efficient induction (50). We find that nucleosome removal at the Il1a enhancer occurs with faster kinetics than at the Il12b enhancer (see Figure 2.2E) and we hypothesize that the different kinetics may indicate the involvement of different nucleosome remodelers as has been suggested (40). While it is possible that nucleosomes may be removed from these regions by competition of signal-induced TFs for binding to their sites, the rapid 48 kinetics we have observed strongly suggest that nucleosome remodelers are involved (see Figure 2.2E). Future studies will reveal which remodelers play a role at these and other enhancers of inducible genes. Acknowledgments We would like to thank Nara Parameswaran and members of his lab for help with isolation of mouse macrophages, members of Norbert Kaminski's lab for help with isolation of B-cells, David Arnosti, Bill Henry, Rick Schwartz, John LaPres and Jason Knott for discussions. 49 REFERENCES 50 REFERENCES 1. Yuan GC, Liu YJ, Dion MF, Slack MD, Wu LF, et al. (2005) Genome-scale identification of nucleosome positions in S. cerevisiae. Science 309: 626–630. 2. Bernstein BE, Liu CL, Humphrey EL, Perlstein EO, Schreiber SL (2004) Global nucleosome occupancy in yeast. Genome biology 5: R62. 3. Sekinger EA, Moqtaderi Z, Struhl K (2005) Intrinsic histone-DNA interactions and low nucleosome density are important for preferential accessibility of promoter regions in yeast. Molecular cell 18: 735–748. 4. Bryant GO, Prabhu V, Floer M, Wang X, Spagna D, et al. (2008) Activator control of nucleosome occupancy in activation and repression of transcription. PLoS biology 6: 2928–2939. 5. Reinke H, Horz W (2003) Histones are first hyperacetylated and then lose contact with the activated PHO5 promoter. Molecular cell 11: 1599–1607. 6. Wang X, Bryant GO, Floer M, Spagna D, Ptashne M (2011) An effect of DNA sequence on nucleosome occupancy and removal. Nature structural & molecular biology 18: 507–509. 7. Floer M, Wang X, Prabhu V, Berrozpe G, Narayan S, et al. (2010) A RSC/ nucleosome complex determines chromatin architecture and facilitates activator binding. Cell 141: 407–418. 8. Schones DE, Cui K, Cuddapah S, Roh TY, Barski A, et al. (2008) Dynamic regulation of nucleosome positioning in the human genome. Cell 132: 887–898. 9. Teif VB, Vainshtein Y, Caudron-Herger M, Mallm JP, Marth C, et al. (2012) Genome-wide nucleosome positioning during embryonic stem cell development. Nature structural & molecular biology 19: 1185–1192. 10. Berrozpe G, Bryant GO, Warpinski K, Ptashne M (2013) Regulation of a mammalian gene bearing a CpG island promoter and a distal enhancer. Cell reports 4: 445–453. 11. Lefevre P, Lacroix C, Tagoh H, Hoogenkamp M, Melnik S, et al. (2005) Differentiation-dependent alterations in histone methylation and chromatin architecture at the inducible chicken lysozyme gene. The Journal of biological chemistry 280: 27552–27560. 12. Bert AG, Johnson BV, Baxter EW, Cockerill PN (2007) A modular enhancer is differentially regulated by GATA and NFAT elements that direct different tissue- 51 specific patterns of nucleosome positioning and inducible chromatin remodeling. Molecular and cellular biology 27: 2870–2885. 13. Agalioti T, Lomvardas S, Parekh B, Yie J, Maniatis T, et al. (2000) Ordered recruitment of chromatin modifying and general transcription factors to the IFNbeta promoter. Cell 103: 667–678. 14. Wathelet MG, Lin CH, Parekh BS, Ronco LV, Howley PM, et al. (1998) Virus infection induces the assembly of coordinately activated transcription factors on the IFN-beta enhancer in vivo. Molecular cell 1: 507–518. 15. Levine M (2010) Transcriptional enhancers in animal development and evolution. Current biology: CB 20: R754–763. 16. Deng W, Lee J, Wang H, Miller J, Reik A, et al. (2012) Controlling long-range genomic interactions at a native locus by targeted tethering of a looping factor. Cell 149: 1233–1244. 17. Medzhitov R (2007) Recognition of microorganisms and activation of the immune response. Nature 449: 819–826. 18. Kawai T, Akira S (2006) TLR signaling. Cell death and differentiation 13: 816– 825. 19. Ghisletti S, Barozzi I, Mietton F, Polletti S, De Santa F, et al. (2010) Identification and characterization of enhancers controlling the inflammatory gene expression program in macrophages. Immunity 32: 317–328. 20. Heinz S, Benner C, Spann N, Bertolino E, Lin YC, et al. (2010) Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Molecular cell 38: 576– 589. 21. Rahman SM, Janssen RC, Choudhury M, Baquero KC, Aikens RM, et al. (2012) CCAAT/enhancer-binding protein beta (C/EBPbeta) expression regu- lates dietaryinduced inflammation in macrophages and adipose tissue in mice. The Journal of biological chemistry 287: 34349–34360. 22. Bretz JD, Williams SC, Baer M, Johnson PF, Schwartz RC (1994) C/EBP- related protein 2 confers lipopolysaccharide-inducible expression of interleukin 6 and monocyte chemoattractant protein 1 to a lymphoblastic cell line. Proceedings of the national academy of sciences U S A 91: 7306–7310. 23. Screpanti I, Romani L, Musiani P, Modesti A, Fattori E, et al. (1995) Lymphoproliferative disorder and imbalanced T-helper response in C/EBP betadeficient mice. The EMBO journal 14: 1932–1941. 52 24. Xie H, Ye M, Feng R, Graf T (2004) Stepwise reprogramming of B cells into macrophages. Cell 117: 663–676. 25. Grove M, Plumb M (1993) C/EBP, NF-kappa B, and c-Ets family members and transcriptional regulation of the cell-specific and inducible macrophage inflammatory protein 1 alpha immediate-early gene. Molecular and cellular biology 13: 5276–5289. 26. Panne D, Maniatis T, Harrison SC (2004) Crystal structure of ATF-2/c-Jun and IRF-3 bound to the interferon-beta enhancer. The EMBO journal 23: 4384– 4393. 27. Zeng L, Liu YP, Sha H, Chen H, Qi L, et al. (2010) XBP-1 couples endoplasmic reticulum stress to augmented IFN-beta induction via a cis-acting enhancer in macrophages. Journal of immunology 185: 2324–2330. 28. Zhou L, Nazarian AA, Xu J, Tantin D, Corcoran LM, et al. (2007) An inducible enhancer required for Il12b promoter activity in an insulated chromatin environment. Molecular and cellular biology 27: 2698–2712. 29. Nociari M, Ocheretina O, Schoggins JW, Falck-Pedersen E (2007) Sensing infection by adenovirus: Toll-like receptor-independent viral DNA recognition signals activation of the interferon regulatory factor 3 master regulator. Journal of virology 81: 4145–4157. 30. Stetson DB, Medzhitov R (2006) Recognition of cytosolic DNA activates an IRF3dependent innate immune response. Immunity 24: 93–103. 31. Floer M, Bryant GO, Ptashne M (2008) HSP90/70 chaperones are required for rapid nucleosome removal upon induction of the GAL genes of yeast. Proceedings of the National Academy of Sciences of the United States of America 105: 2975– 2980. 32. Robinson JT, Thorvaldsdottir H, Winckler W, Guttman M, Lander ES, et al. (2011) Integrative genomics viewer. Nature biotechnology 29: 24–26. 33. Sharma RC, Murphy AJ, DeWald MG, Schimke RT (1993) A rapid procedure for isolation of RNA-free genomic DNA from mammalian cells. BioTechniques 14: 176–178. 34. Gervais AL, Marques M, Gaudreau L (2010) PCRTiler: automated design of tiled and specific PCR primer pairs. Nucleic acids research 38: W308–312. 35. Carninci P, Sandelin A, Lenhard B, Katayama S, Shimokawa K, et al. (2006) Genome-wide analysis of mammalian promoter architecture and evolution. Nature genetics 38: 626–635. 53 36. Heintzman ND, Stuart RK, Hon G, Fu Y, Ching CW, et al. (2007) Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nature genetics 39: 311–318. 37. Mikkelsen TS, Ku M, Jaffe DB, Issac B, Lieberman E, et al. (2007) Genome- wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448: 553–560. 38. Garber M, Yosef N, Goren A, Raychowdhury R, Thielke A, et al. (2012) A highthroughput chromatin immunoprecipitation approach reveals principles of dynamic gene regulation in mammals. Molecular cell 47: 810–822. 39. Sandelin A, Wasserman WW, Lenhard B (2004) ConSite: web-based prediction of regulatory elements using cross-species comparison. Nucleic acids research 32: W249–252. 40. Ramirez-Carrozzi VR, Braas D, Bhatt DM, Cheng CS, Hong C, et al. (2009) A unifying model for the selective regulation of inducible transcription by CpG islands and nucleosome remodeling. Cell 138: 114–128. 41. Weinmann AS, Plevy SE, Smale ST (1999) Rapid and selective remodeling of a positioned nucleosome during the induction of IL-12 p40 transcription. Immunity 11: 665–675. 42. Zhao R, Davey M, Hsu YC, Kaplanek P, Tong A, et al. (2005) Navigating the chaperone network: an integrative map of physical and genetic interactions mediated by the hsp90 chaperone. Cell 120: 715–727. 43. Shalek AK, Satija R, Adiconis X, Gertner RS, Gaublomme JT, et al. (2013) Singlecell transcriptomics reveals bimodality in expression and splicing in immune cells. Nature 498: 236–240. 44. Zhao M, Zhang J, Phatnani H, Scheu S, Maniatis T (2012) Stochastic expression of the interferon-beta gene. PLoS biology 10: e1001249. 45. Kieffer-Kwon KR, Tang Z, Mathe E, Qian J, Sung MH, et al. (2013) Interactome maps of mouse gene regulatory domains reveal basic principles of transcriptional regulation. Cell 155: 1507–1520. 46. Kagey MH, Newman JJ, Bilodeau S, Zhan Y, Orlando DA, et al. (2010) Mediator and cohesin connect gene expression and chromatin architecture. Nature 467: 430–435. 47. Wang Q, Carroll JS, Brown M (2005) Spatial and temporal recruitment of androgen receptor and its coactivators involves chromosomal looping and polymerase tracking. Molecular cell 19: 631–642. 54 48. De Santa F, Barozzi I, Mietton F, Ghisletti S, Polletti S, et al. (2010) A large fraction of extragenic RNA pol II transcription sites overlap enhancers. PLoS biology 8: e1000384. 49. Kim TK, Hemberg M, Gray JM, Costa AM, Bear DM, et al. (2010) Widespread transcription at neuronal activity-regulated enhancers. Nature 465: 182–187. 50. Ramirez-Carrozzi VR, Nazarian AA, Li CC, Gore SL, Sridharan R, et al. (2006) Selective and antagonistic functions of SWI/SNF and Mi-2beta nucleosome remodeling complexes during an inflammatory response. Genes & development 20: 282–296. 55 Chapter 3: Chromatin remodeler recruitment during macrophage differentiation facilitates transcription factor binding to enhancers in mature cells This chapter represents a manuscript that was published in The Journal of Biological Chemistry (2016) 291: 18058-71. Authors who contributed to this study were: Michael McAndrew, Alison Gjidoda, Mohita Tagore, Tyler Miksanek, and Monique Floer. 56 Abstract We show how enhancers of macrophage-specific genes are rendered accessible in differentiating macrophages to allow their induction in mature cells in response to an appropriate stimulus. Using a lentiviral knockdown approach in primary differentiating macrophages from mouse bone marrow we demonstrate that enhancers of Il12b and Il1a are kept relatively lowly occupied by nucleosomes and accessible through recruitment of the nucleosome remodeler BAF/PBAF. Our results using an inducible cell-line that expresses an estrogen receptor fusion of the macrophage-specific transcription factor PU.1 (PUER) show that BAF/PBAF recruitment to these enhancers is a consequence of translocation of PUER to the nucleus in the presence of tamoxifen, and we speculate that remodeler recruitment may be directly mediated by PU.1. In the absence of BAF/PBAF recruitment, nucleosome occupancy at the enhancer of Il12b (and to a lesser extent at Il1a) reaches high levels in bone marrow derived macrophages (BMDMs), and the enhancers are not fully cleared of nucleosomes upon LPS induction resulting in impaired gene expression. Analysis of Il12b expression in single cells suggests that recruitment of the remodeler is necessary for high levels of transcription from the same promoter and we propose that remodelers function by increasing nucleosome turnover to facilitate transcription factor over nucleosome binding in a process we have termed remodeler assisted competition. Introduction Lineage-specific transcription factors (TFs) play a crucial role in cellular differentiation. These TFs are often pioneer TFs that have been suggested to control access to cisregulatory elements — in particular gene enhancers — by other ubiquitously expressed 57 TFs (1). The idea that access to regulatory elements is controlled in a cell-type specific manner is supported by the finding that sensitivity of enhancers to nucleases like DNase I or MNase is cell-type specific (for recent studies see (2,3)), but how lineage-specific TFs render enhancers accessible during differentiation is unknown. Moreover, what constitutes accessible or “open” chromatin has remained unclear. While regulatory regions of constitutively expressed genes are often completely nucleosome-free, we recently showed that the enhancers of inducible genes are occupied by intermediate levels of nucleosomes in resting macrophages, and these nucleosomes are evicted when the genes are induced (4). Furthermore, before induction these enhancers are already bound by the macrophage-specific pioneer TF PU.1 and primed for activation as indicated by the presence of certain histone marks (i.e., H3K4me1)(5). Binding of PU.1 to enhancers was found to lead to a decrease in nucleosome binding (6,7), and we showed that in the absence of PU.1 binding macrophage-specific enhancers become associated with the polycomb repressive complex (PRC2) and with highly occupied, H3K27me3-marked nucleosomes as cells differentiate (8). These results indicated that the pioneer TF PU.1 keeps enhancers accessible and prevents heterochromatin formation at cell-type specific genes, but the underlying mechanism has remained unclear. We sought to investigate whether nucleosome remodelers are involved in priming of enhancers. Remodelers of the SWI/SNF family have been shown to facilitate gene expression in many organisms, and SWI/SNF function is best understood in the yeast S. cerevisiae, where studies showed that SWI/SNF remodelers remove nucleosomes from promoters or partially unwrap nucleosomes to expose TF binding sites (9-13). 58 Mammals have two related SWI/SNF complexes, BAF and PBAF, which share certain subunits but also contain unique subunits that are thought to play a role in recruitment of either complex to specific sites. Both BAF and PBAF use the catalytic subunit BRG1, but BAF can also use the alternate catalytic subunit BRM. BRG1 deletion results in early embryonic lethality, but BRM-/- mice develop normally and it has been suggested that upregulation of BRG1 may, in part, compensate for the loss of BRM (14,15). BRG1 is required for differentiation, including that of lymphoid and myeloid cells, and BRG1 is recruited to cell-type specific genes during differentiation of erythrocytes, suggesting that a BRG1-containing BAF/PBAF complex may prime gene regulatory regions during hematopoiesis (16-18). That BAF/PBAF may play a general role in cellular differentiation is further supported by the finding that BRG1 and other BAF/PBAF subunits are frequently mutated in diverse human cancers (19). The core subunit SNF5, for example, is mutated in malignant rhabdoid tumors, a rare aggressive cancer affecting young children, and SNF5 mutation is sufficient to induce such tumors in mice (20,21). Rhabdoid tumor cells are unable to proliferate when BRG1 is inactivated, and it has been suggested that these cells may become dependent on an altered BAF/PBAF complex that still relies on the presence of BRG1 (22). Previous studies showed that BAF/PBAF is required for induction of pro-inflammatory genes in mouse macrophages, since simultaneous knockdown of both BRG1 and BRM impaired induction of a subset of pro-inflammatory genes in a macrophage cell-line by bacterial lipopolysaccharides (LPS)(23). These investigators suggested a role for BAF/PBAF in remodeling non-CpG island promoters but did not investigate whether the remodeler creates accessible chromatin at the enhancers of these genes to prime them for later gene induction. 59 These investigators also determined whether primary and secondary response genes show differential dependence on the BAF/PBAF remodelers, and concluded that secondary and a subset of primary response genes require the remodeler, while other primary response genes are largely independent. Here, we show how regulatory regions of two representative macrophage-specific genes (i.e., Il1a, a primary and Il12b, a secondary response gene) are rendered accessible during differentiation through recruitment of BAF/PBAF, presumably as a consequence of PU.1 binding. This allows induction of these genes in mature macrophages in response to an appropriate signal. We find that both genes depend on BAF/PBAF for induction and nucleosome eviction at their enhancers, but the effects on Il1a are less pronounced. Our analysis of gene expression in single cells suggests that remodelers function by remodeler assisted competition to facilitate TF binding over nucleosome formation at cell-type specific gene enhancers. Experimental Procedures Cell isolation and culture Bone marrow cells and splenic B-cells were isolated as described from 6-8 week old C57BL/6 female mice (NCI/Charles River) with IACUC oversight (4). To obtain BMDMs, cells were differentiated into macrophages by growth in the presence of M-CSF as described (30). BMDMs were induced with LPS as described (4). The PU.1-/- and PUER cells were grown as described previously (8). HSPCs were isolated using the EasySep™ Mouse Hematopoietic Progenitor Cell Isolation Kit (Stemcell Technologies) per manufacturer’s instructions, with an additional red blood lysis step prior to progenitor isolation. Briefly, 2-3 x 107 cells were resuspended in 1 ml red cell lysis buffer (Sigma) 60 and mixed gently for 2 min. before addition of 9 ml IMDM medium (Gibco) and centrifugation at 400 x g for 5 min. The resulting cell pellet was used for HSPC isolation. shRNA mediated knockdown of BRG1 and SNF5 Lentiviral transductions were performed essentially as described (8). Briefly, lentiviral particles containing shRNAs targeting either BRG1 (Smarca4) or SNF5 (Smarcb1) selected from a pool that had been pre-validated by the Broad Consortium (TRC collection MISSION shRNA library, Sigma) or control shRNA targeting firefly luciferase were produced in HEK293T cells. For lentiviral transductions, bone marrow cells from the femur and tibia of 6-8 week old C57BL/6 female mice were infected with lentivirus after they had been grown for 48 h in BMDM medium containing L929 cell supernatant as a source of M-CSF. 4 h after viral infection the medium was replaced to remove the virus and cells were grown for 48 h. Transduced cells were then selected by growth in the presence of 5 µg/ml puromycin for 5 days. Cells were harvested for various experiments as described (4). Quantitative nucleosome occupancy assay The assay was performed essentially as described in (4) except that cross-linked chromatin from 0.5 to 1 x 107 cells was used per experiment and the MNase (NEB) concentrations were adjusted to a range from 0.0027 U to 13.3 U. Bar graphs and overlays were generated using the IGB genome browser. Primer pairs for the amplicons used can be given upon request. Chromatin immunoprecipitation ChIP experiments were performed essentially as described (4) except that sonicated chromatin from 1.5 - 2.5 x 106 cells per antibody was diluted 2.5-fold with Low Salt ChIP 61 buffer (20 mM Tris-HCl, pH 8, 200 mM NaCl, 0.5% Triton X-100, 2 mM EDTA, Halt™ Protease Inhibitor (Thermo Scientific)) to a total volume of 375 µl and incubated overnight at 4ºC with either 5 µl anti-SNF5 (ab126734; Abcam), 5 µl anti-BAF155 (D7F8S; Cell Signaling Technology) or 2 µl anti-PU.1 (sc-352; SCBT). Immunoprecipitated DNA was quantified on a Lightcycler 480 (Roche). LPS-inducible enhancers measured were identified by Ghisletti et al. (5) and have the following genomic locations: 44 kb upstream of Peli1, 64 kb upstream of IL6 and 3.9 kb upstream of Ccl5 (5). Intergenic region 1 is located 7 kb upstream of the TSS of Il12b, intergenic region 2 is located 25 kb upstream of the TSS of Il1a and intergenic region 3 is located is in the HOX cluster between Hoxd11 and Hoxd10. Primer sequences can be given upon request. ChIP data is displayed as the fold binding over average binding at control regions (i.e., the Kit promoter, Rpl4 Orf and intergenic region 1). mRNA determination RNA isolation and cDNA synthesis were performed as described (4). cDNA was analyzed by qRT-PCR on a Lightcycler 480 (Roche) using gene-specific primer pairs. Primer sequences can be given upon request. Chromatin fractionation and Western blotting Chromatin fractionation was performed essentially as described using the high salt extraction protocol of (31). Briefly, 1.5 - 2 x 106 cells that had been transduced with lentivirus bearing specific shRNAs or untreated control BMDMs were resuspended in 400 µl extraction buffer (10 mM HEPES pH 7.9, 10 mM KCl, 1.5 mM MgCl2, 0.34 M sucrose, 10% Glycerol, Halt™ Protease Inhibitor (Thermo Scientific)), which contained 0.2% NP40 but no sodium butyrate. The solution was centrifuged for 5 min. at 6,500 x 62 g. The nuclear pellet was washed in 400 µl of extraction buffer without NP40 or sodium butyrate and then centrifuged again for 5 min. at 6,500 x g. Nuclei were resuspended by vortexing in 400 µl no-salt buffer (10 mM HEPES pH 7.9, 3 mM EDTA, 0.2 mM EGTA). The solution was placed on a rotator at 4oC for 30 min. and then spun at 6,500 x g for 5 min. The pellet containing chromatin was resuspended in 160 µl high salt solubilization buffer (50 mM Tris-Cl pH 8.0, 2.5 M NaCl, 0.05% NP40), vortexed, and incubated on a rotator at 4oC for 30 min. The samples were then centrifuged for 10 min. at 16,000 x g. The supernatant containing solubilized proteins was collected as the chromatin-associated fraction. TCA was added at a final concentration of 10%, samples were incubated for 15 min. and then centrifuged at 21,000 x g for 15 min. The resulting pellet was washed with 500 µl acetone and resuspended in 40 µl of LDS Sample Buffer (106 mM Tris HCl, 141 mM Tris Base, 2% LDS, Glycerol 10%, 0.51 mM EDTA, pH 8.5). To 30 ml of each sample 3 µl 0.1% Coomassie blue, 2 µl 1 M DTT and 5 µl of 2 x LDS Sample Buffer were added, samples were incubated at 75oC for 10 min and each fraction was analyzed by SDS-PAGE on a 4-12% Bis-Tris Plus gel (Novex, Life Technologies). Western analysis was performed after protein transfer for 2 h at 90 V onto a nitrocellulose membrane and quantification of total protein by Ponceau Red staining, using antibodies against POLII (sc-56767; SCBT), BRG1 (sc-10768; SCBT), SNF5 (ab12167; Abcam), and histone H3 (ab1791; Abcam). Chemiluminescent signal after incubation with appropriate secondary antibodies was quantified on a ChemiDoc MP Imaging system (BioRad) or using ImageJ. 63 Flow cytometry Analysis was performed on a BD Biosciences LSR II flow cytometer. 1 x 105 cells were used per antibody. To determine IL12B production, Golgi inhibitor GolgiPlug™ (BD Biosciences) was added to prevent cytokine secretion as described (27). Briefly, 1 ml/ml GolgiPlugTM was added in medium without FBS and cells were incubated for 1 h at 37oC. Then 1 mg/ml LPS from E. coli strain EH100 (Ra mutant)(Sigma) was added and cells were grown for 3 h. Cells were collected using Versene (Lifetech) treatment and washed with PBS once. Cells were fixed with 1% formaldehyde for 10 min. and washed with PBS once. To block nonspecific Fc receptor binding, cells were incubated with 2.42G supernatant for 10 min., followed by a wash with PBS. Staining was performed in permeabilization buffer (PBS, 5% FBS, 0.1% sodium azide, 0.5% Triton X-100) for 30 min. in the dark with anti-IL12B-APC (554480; BD Pharmingen) and anti-SNF5AlexaFluor488 (bs-6109R; Bioss), and cells were subsequently washed twice in flow wash buffer (PBS, 5% FBS, 0.1% sodium azide). Due to differences in total count after size gating, fluorescence histograms were normalized and unit areas are shown in overlays instead of absolute cell counts. Isolation of Lin- cells was confirmed by flow cytometry using the lineage antibody cocktail provided in the EasySep™ Hematopoietic Progenitor Cell Isolation Kit probing for CD5, CD11b, CD19, CD45R/B220, Ly6G/C(Gr-1), TER119, 7-4 (19856; Stemcell Technologies) followed by secondary incubation with Streptavidin-PE (Lifetech), as well as for anti-CD117/KIT (60025; Stemcell Technologies) and anti-SCA1 (60032; Stemcell Technologies) followed by secondary incubation with anti-mouse-FITC (55499; MP 64 Biomedicals). BMDMs were also analyzed using anti-F4/80-APC (eBioscience 17-4801) and anti-CD11b-FITC (eBioscience 10-0112) antibodies. Statistical Analysis Knockdown experiments were performed at least four times for each shRNA, and mRNA results for target genes and cytokine induction are shown as the average of all experiments. Statistical significance of differences was determined by one-way ANOVA analysis and confirmed by a post-hoc Tukey HSD test. BAF155, SNF5, and PU.1 ChIP experiments were performed at least twice. Statistical significance of the observed differences was determined by one-way ANOVA and confirmed by post-hoc Tukey HSD or Fisher LSD tests. To determine binding to the enhancers, all the data from the different enhancer amplicons tested was analyzed together for statistical significance and compared to all the control amplicons. Nucleosome occupancy experiments were performed twice for each knockdown, and a full analysis including all the amplicons in each enhancer was performed once for the BRG1 KD and twice for the SNF5 KD. The error bars represent the confidence intervals of the curve-fitting analysis for a representative experiment. P-values in the figures indicate the statistical significance of differences between different conditions as determined by paired, two-tailed Student’s ttests. Results BAF/PBAF is recruited to the Il12b and Il1a enhancers in BMDMs To investigate how the enhancers of Il12b and Il1a are kept accessible and occupied only by intermediate levels of nucleosomes in BMDMs we investigated whether the BAF/PBAF complex is involved in the process. We determined binding of BAF/PBAF to 65 Il12b and Il1a by ChIP and detected the core subunits BAF155 and SNF5 at both enhancers in resting macrophages (Fig. 3.1A and B, dark blue bars). Recruitment of the remodeler to the Il12b enhancer further increased upon LPS induction (yellow bars), but the levels of remodeler at Il1a were already high in resting BMDMs and did not increase significantly upon induction. We found little binding of BAF/PBAF to the enhancers in hematopoietic stem and progenitor cells (HSPCs; isolated by Lin- selection from bone marrow) or B-cells (cyan and green bars, respectively), demonstrating that recruitment of the remodeler to these genes is macrophage-specific. Together our results indicate that BAF/PBAF is recruited to the enhancers of Il12b and Il1a at some time during macrophage differentiation, and that gene induction leads to further remodeler recruitment to Il12b. Binding of SNF5 and BAF155 to the promoters of both genes was low suggesting that the nucleosome remodeler functions predominantly at the enhancers of these genes. BAF/PBAF recruitment is a consequence of PUER translocation to the nucleus To determine how BAF/PBAF is recruited to macrophage-specific enhancers we turned to the PUER expressing cell-line that we had previously used to determine the effects of PU.1 binding on nucleosome occupancy (8). This cell-line was derived from hematopoietic progenitors of the fetal liver of a PU.1-/- mouse and expresses the pioneer TF PU.1 as an estrogen receptor fusion (PUER). Growth for prolonged times (i.e., 4 d) in the presence of tamoxifen leads to differentiation of these cells into macrophage-like cells (24). Alternatively, they can be differentiated into mast cells or erythrocyte precursors, indicating that they are multipotent progenitors. We and others previously showed that when these cells were grown in the presence of tamoxifen, PUER bound to 66 the enhancer of Il1a and other inducible genes, which led to reduced nucleosome binding at these sites (6,8). We had also shown that PUER did not bind to the enhancer of Il12b and several other inducible macrophage-specific enhancers that are bound by PU.1 in BMDMs, consistent with published results (6). Instead this subset of inducible genes became associated with the polycomb repressive complex PRC2 (i.e., Suz12) and acquired repressive histone marks (i.e., H3K27me3) when the cells were differentiated into macrophage-like cells, indicating that facultative heterochromatin was formed at these sites in the absence of PU.1 binding. To determine if PUER recruited BAF/PBAF to macrophage-specific enhancers that could bind the pioneer TF in this system, we performed a ChIP experiment probing for the BAF155 subunit and for PU.1 and found that recruitment of BAF/PBAF indeed correlated with PUER binding to the enhancers of Il1a, Peli1, Il6 and Ccl5. Statistically significant BAF155 recruitment and PUER binding was detected as early as 1 h after addition of tamoxifen at Il1a and Peli1 (Fig. 3.1C and D, orange bars) and further increased with prolonged growth in the presence of tamoxifen to reach significant levels at all four enhancers after 6 h (red bars). We had shown previously that at this time the cells still resemble progenitors and that the associated genes are not induced and signal-induced TFs are not bound (8). The rapid appearance of BAF155 binding after tamoxifen addition suggests that remodeler recruitment is a direct consequence of PUER translocation to the nucleus. We speculate that PU.1 may directly recruit BAF/PBAF to these enhancers, although further experiments will have to be performed to confirm this conclusion. We also demonstrated that in primary HSPCs from bone marrow, where BAF/PBAF was not recruited to the enhancers (Fig. 3.1A, B and D, cyan bars), PU.1 was absent as well 67 (Fig. 3.1C, cyan bars) further supporting the idea that BAF/PBAF recruitment is a consequence of PU.1 binding in primary macrophages. Together, our results suggest that upregulation of PU.1 expression during macrophage differentiation (25) induces PU.1 binding and concomitant recruitment of BAF/PBAF to enhancers of macrophagespecific genes, which primes these genes for induction in mature macrophages. 68 Figure 3.1. Recruitment of BAF/PBAF to macrophage-specific enhancers. (A) A ChIP experiment probing for BAF155 was performed in HSPCs (cyan), in BMDMs grown without (dark blue) and with LPS for 1.5 h (yellow), and in splenic B-cells (green). BAF155 binding to the enhancers, promoters and intervening sequences of Il12b and 69 Figure 3.1. (cont’d) Il1a, and at control regions is shown. ChIP experiments were performed three times and error bars indicate the SEM. One-way ANOVA shows that differences at the enhancers are statistically significant (at the p<0.05 level) between different cell-types, while differences at control locations, the promoters and the intervening regions are not statistically significant. A post-hoc Tukey HSD test confirmed that differences between uninduced BMDMs and HSPCs or B-cells at the enhancers were statistically significant. At the Il12b enhancer differences between uninduced and induced BMDMs were also statistically significant while those at the Il1a enhancer were not. (B) A SNF5 ChIP was performed in HSPCs and in BMDMs grown with and without LPS and a statistical analysis confirmed the significance of differences as for the BAF155 ChIP shown in (A). (C) A ChIP experiment using an antibody that recognizes both PU.1 and PUER was performed in HSPCs (cyan), in the PU.1-/- cell-line (magenta), and in PUER cells grown in the absence of tamoxifen (yellow), and for 1 h (orange) and 6 h (red) in the presence of tamoxifen. All cells were grown in the absence of LPS and resting BMDMs are shown as controls (blue). PU.1/PUER binding at LPS-inducible enhancers of Il1a, Peli1, Il6 and Ccl5 are shown (for genomic coordinates of the enhancers see Experimental Procedures). ChIP experiments were performed twice and error bars indicate the SEM. A one-way ANOVA shows statistically significant differences (p<0.05) between different cell-types and growth conditions. Post-hoc comparisons using a Tukey HSD test indicate that at all four enhancers growth in the presence of tamoxifen for 6 h resulted in 70 Figure 3.1. (cont’d) statistically significant binding of PUER compared to no tamoxifen, and at Il1a and Peli1 differences were already statistically significant after 1 h. (D) A BAF155 ChIP was performed with cells as in (C) and a statistical analysis confirmed significance of the differences in BAF155 recruitment as described for PU.1/PUER binding in (C). BAF/PBAF is required for Il12b and Il1a induction in BMDMs To determine whether recruitment of the BAF/PBAF complex rendered the enhancers of Il12b and Il1a accessible during macrophage differentiation we used a lentiviral shRNAmediated knockdown approach. For these experiments bone marrow cells were transduced with lentivirus containing shRNAs targeting BRG1, encoded by the Smarca4 gene, or with control shRNA targeting firefly luciferase (shLuc). The effect of BRG1 KD was then analyzed in transduced cells that had been differentiated into macrophages in the presence of M-CSF for 9 days. We identified two shRNAs from a pool of shRNAs pre-validated by the Broad Consortium (shSmarca4-3 and shSmarca4-4) that yielded 50-60% knockdown of Smarca4 as determined by mRNA analysis (Fig. 3.2A) and resulted in reduction of chromatin-associated BRG1 protein by 50% (Fig. 3.2B). This level of knockdown reduced Il12b and Il1a expression 1.5 h after LPS addition by 50% (Fig. 3.2C). Previous studies in the macrophage cell-line J774 had shown that Il12b expression was dependent on BRG1, but these investigators had classified Il1a as a BAF/PBAF-independent gene, although a small decrease in Il1a expression was reported (23). We believe that the more pronounced effect of our BRG1 KD on Il1a induction may be due to differences between the macrophage cell-line J774 and 71 primary BMDMs. The cells differentiated under these conditions still resembled macrophages and expressed the macrophage marker F4/80 (i.e., Emr1, orange bars in Fig. 3.2D). However, we found that other macrophage-specific, constitutively expressed genes were expressed at lower levels in BRG1 KD cells (e.g., Csf1r, blue bars in Fig. 3.2D). BRG1 KD affects nucleosome occupancy and eviction at the Il12b and Il1a enhancers To analyze the effect of knocking down BRG1 on nucleosome occupancy at enhancers we pooled cells transduced with lentivirus containing either of the two BRG1-specific shRNAs we had identified, and performed the quantitative nucleosome occupancy assay. We found that nucleosome occupancy over the whole Il12b enhancer was higher in BRG1 KD compared to untreated control cells (Fig. 3.2E). Nucleosome occupancy at preferred positions increased by 10-25% resulting in peak occupancies of 75-90%. Positioning of nucleosomes was largely unaffected, suggesting that other factors determine nucleosome positioning in the Il12b enhancer. Knockdown of BRG1 in hematopoietic progenitors also led to increased nucleosome occupancy at the Il1a enhancer, although the effect was less pronounced than at Il12b (Fig. 3.2F). P-values of Student’s t-tests showed that the differences found over the whole enhancer regions between BRG1 KD and control cells were statistically significant. Control regions were not affected by BRG1 KD (Fig. 3.2G). Analysis of nucleosome occupancy 1.5 h after LPS addition showed less nucleosome eviction at both enhancers in BRG1 KD compared to untreated cells (Fig. 3.2H and I). For example, occupancy at positions in the Il12b enhancer that are completely cleared of nucleosomes in response to LPS in 72 untreated cells (< 5%), remained associated with nucleosomes in 15-20% of the population when BRG1 was knocked down. Figure 3.2. KD of the catalytic BAF/PBAF subunit BRG1. (A) BRG1 was knocked down in hematopoietic progenitors using two shRNAs (shSmarca4-3 and shSmarca4-4) and cells were differentiated into BMDMs as described in Experimental Procedures. Cells transduced with control shLuc are also shown. mRNA levels of the Smarca4 gene were analyzed in untreated BMDMs, and in cells transduced with control and specific shRNAs as indicated. Cells were either grown without (blue) or with (yellow) LPS for 1.5 h. Data was normalized to mRNA levels found in untreated BMDMs grown in the absence of LPS; experiments were performed at least four times and SEMs are indicated by the error bars. One-Way ANOVA shows statistical significance between differently treated cells (p<0.05), and a post-hoc Tukey HSD test confirms statistical significance between untreated (or shLuc treated) and shSmarca4 treated cells. (B) BRG1 protein abundance was determined by Western analysis in the chromatin fraction of untreated BMDMs, and in that of cells transduced with either of the BRG1specific shRNAs identified in (A) and pooled before fractionation. SNF5, POLII and 73 Figure 3.2. (cont’d) histone H3 levels are shown as controls. Relative abundance of proteins compared to untreated BMDMs is indicated. (C) mRNA of Il12b (red) and Il1a (blue) in cells as described in (A) and grown in the presence of LPS for 1.5 h. One-Way ANOVA shows statistical significance between differently treated cells (p<0.05), and a post-hoc Tukey HSD test confirms statistical significance between untreated and shSmarca4-3 or 4 treated cells for Il12b, and shSmarca4-4 treated cells for Il1a induction. Induction data for shLuc treated cells showed higher variability, but was not statistically significantly different from untreated cells. (D) mRNA of the macrophage markers Csf1r (blue) and Emr1 (orange) is shown in cells as in (A) grown in the absence of LPS. 74 Figure 3.2. (cont’d) (E) Untreated BMDMs (blue) and BRG1 KD cells (orange) were obtained as described in (B). Nucleosome occupancy at the Il12b enhancer in cells grown without LPS is shown as a bar graph with the width of each bar corresponding to the size of each amplicon. P-value of a Student’s t-test shows significance of the differences between untreated and BRG1 KD cells. (F) Untreated BMDMs (blue) and BRG1 KD cells (orange) were obtained as described in (B). Nucleosome occupancy at the Il1a enhancer in cells grown without LPS. 75 Figure 3.2. (cont’d) (G) Untreated BMDMs (blue) and BRG1 KD cells (orange) were obtained as described in (B). Nucleosome occupancy at control regions in cells grown in the absence of LPS. (H) Untreated BMDMs (blue) and BRG1 KD cells (orange) were obtained as described in (B). Nucleosome occupancy at the Il12b enhancer in cells grown in the presence of LPS for 1.5 h. (I) Untreated BMDMs (blue) and BRG1 KD cells (orange) were obtained as described in (B). Nucleosome occupancy at the Il1a enhancer in cells grown in the presence of LPS for 1.5 h. Control experiments showed that transduction with shLuc had no effect on occupancy before or upon LPS induction (Fig. 3.3A-D). Together our results indicate that recruited BAF/PBAF prevents high levels of nucleosome binding at the Il12b and Il1a enhancers in resting macrophages and stimulates nucleosome eviction from the enhancers upon LPS induction. However, nucleosomes were still partially evicted in the absence of BRG1, suggesting that a BRM containing BAF complex may partially compensate for the loss of BRG1. 76 Figure 3.3. Nucleosome occupancy in shLuc treated and untreated control cells. (A) Nucleosome occupancy at the Il12b enhancer in BMDMs (dark blue) and cells transduced with shLuc as described in Experimental Procedures (sky blue) grown without LPS. Data is shown as line graphs with each point representing the midpoint of a single amplicon and error bars indicate the confidence interval derived from curvefitting. (B) Nucleosome occupancy at the Il12b enhancer in cells as in (A) grown with LPS for 1 h. (C) Nucleosome occupancy at the Il1a enhancer in cells grown without LPS. (D) Nucleosome occupancy at the Il1a enhancer grown in the presence of LPS for 1 h. P-values of Student’s t-tests indicate that differences between untreated and shLuc transduced cells are not statistically significant. 77 Knockdown of SNF5 abolishes BAF/PBAF binding at the Il12b and Il1a enhancers To determine whether inactivation of both BAF and PBAF has a stronger effect on nucleosome occupancy at the enhancers, we knocked down the shared core subunit SNF5 in hematopoietic progenitors using the same lentiviral approach. As shown in Fig. 3.4A we identified three shRNAs (shSmarcb1-1, 1-2 and 1-3) that knocked down Smarcb1 (the gene encoding SNF5). shSmarcb1-1 yielded better knockdown (~80%) than either of the other two shRNAs (shown as average) and we therefore selected shSmarcb1-1 for further analysis. Western blotting confirmed that KD by shSmarcb1-1 reduced the levels of chromatin-associated SNF5 protein by about 90% (Fig. 3.4B). Moreover, the catalytic subunit BRG1 was no longer detectable in the chromatin-bound fraction when SNF5 was knocked down (Fig. 3.4B). Under these conditions Il12b induction was reduced by about 75% 1.5 h after LPS addition and Il1a induction was reduced by about 50% (Fig. 3.4C). Similar to our findings in BRG1 KD cells, we found that SNF5 KD cells still resembled macrophages and expressed macrophage markers (Fig. 3.4D). However, we noted that many cells died during the timecourse of differentiation when we knocked down SNF5, suggesting that loss of SNF5 impairs differentiation and that a minimal amount of SNF5 may be necessary for cells to differentiate into macrophages. Cell survival was also impaired upon BRG1 KD, but to a lesser extent. When we analyzed recruitment of the BAF/PBAF complex to the Il12b and Il1a enhancers by ChIP, we found that recruitment of BAF155, both before and upon LPS induction, was strongly reduced in the SNF5 KD (Fig. 3.4E); as expected, SNF5 was no longer detected at the enhancers under these conditions (Fig. 3.4F). This result suggests that the SNF5 subunit is either required for recruitment of BAF/PBAF to 78 the Il12b and Il1a enhancers or for formation of a stable complex. Previous results indicated that a BAF/PBAF complex is still formed in the absence of SNF5 in rhabdoid tumor cell-lines (26), but our attempts to determine whether BAF/PBAF stability was affected when we knocked down SNF5 in BMDMs were unsuccessful, because low abundance of the complex in whole cell lysates of primary BMDMs made detection of the complex difficult (Floer, M. and Gjidoda, A., unpublished data). 79 Figure 3.4. KD of the shared BAF/PBAF core subunit SNF5. (A) mRNA levels of the Smarcb1 gene were analyzed in untreated BMDMs, and in cells transduced with control and specific shRNAs as indicated. Results in cells transduced with shSmarcb1-2 and 1-3 are shown as an average. Cells were either grown without (blue) or with (yellow) LPS for 1.5 h and data was normalized to uninduced BMDMs. 80 Figure 3.4. (cont’d) One-Way ANOVA shows statistical significance between differently treated cells (p<0.05), and a post-hoc Tukey HSD test confirms statistical significance between untreated (or shLuc treated) and shSmarcb1-1 and shSmarcb1-2 or 3 treated cells. (B) SNF5 protein was analyzed in the chromatin fractions of untreated BMDMs and of cells transduced with shSmarcb1-1. Western analysis shows loss of SNF5 and BRG1 in the SNF5 KD. POLII and histone H3 are shown as controls. Relative abundance of proteins compared to untreated BMDMs is indicated. (C) mRNA of Il12b (red) and Il1a (blue) in cells as described in (A) and grown in the presence of LPS for 1.5 h. Note that the data shown for cells transduced with shLuc is the same as in Fig. 3.2C. One-Way ANOVA shows statistical significance between differently treated cells (p<0.05), and a post-hoc Tukey HSD test confirms statistical significance between untreated and shSmarcb1-1 or 2/3 treated cells for Il12b and for shSmarcb1-1 treated cells for Il1a. (D) mRNA of the macrophage markers Csf1r (blue) and Emr1 (orange) is shown in cells as in (A) grown in the absence of LPS. (E) A BAF155 ChIP was performed in untreated BMDMs grown in the absence (blue) or presence of LPS for 1.5 h (yellow) or in cells knocked down for SNF5 (shSmarcb1-1) and grown in the absence (green) or presence of LPS (red). BAF155 binding to Il12b, Il1a and control regions is shown as described in the legend of Fig. 3.1A. One-way ANOVA shows that differences between BMDMs and SNF5 KD cells are statistically significant (p<0.05). A post-hoc Fisher LSD test confirms that differences at the enhancers are statistically significant while differences at control regions are not. 81 Figure 3.4. (cont’d) (F) A SNF5 ChIP was performed in untreated BMDMs grown in the absence (blue) or presence of LPS for 1.5 h (yellow) or in cells knocked down for SNF5 (shSmarcb1-1) and grown in the absence (green) or presence of LPS (red). BAF155 binding to Il12b, Il1a and control regions is shown as described in the legend of Fig. 3.1A. One-way ANOVA shows that differences between BMDMs and SNF5 KD cells are statistically significant (p<0.05). A post-hoc Fisher LSD test confirms that differences at the enhancers are statistically significant while differences at control regions are not. Nucleosome occupancy at the Il12b and Il1a enhancers increases in the absence of BAF/PBAF recruitment We analyzed nucleosome occupancy in BMDMs that had been transduced with shSmarcb1-1 expressing lentivirus, and found increased nucleosome occupancy at the Il12b and Il1a enhancers, both before and upon LPS induction (Fig. 3.5A-D). The increase in nucleosome occupancy at Il12b was even more pronounced than in the BRG1 KD and resulted in occupancies at preferred nucleosomal positions around 85100% before LPS induction (Fig. 3.5A), while control regions were not affected (Fig. 3.5E). 1 h after LPS addition, nucleosomes remained associated with the Il12b enhancer in 40-50% of the population (Fig. 3.5B) and occupancy did not decrease further with prolonged LPS induction for 1.5 h (see Fig. 3.6C). This result is consistent with the more pronounced effect of SNF5 KD on Il12b expression compared to KD of BRG1 (compare Fig. 3.2C to 3.4C). We also found increased nucleosome occupancy at the Il1a enhancer both before and upon LPS induction (Fig. 3.5C and D). The increase in occupancy at the Il1a enhancer before induction was similar to what we had found in 82 the BRG1 KD, while nucleosome eviction at Il1a upon LPS induction was more strongly affected by SNF5 KD. Nevertheless, we note that some level of nucleosome eviction was still seen at both enhancers in the SNF5 KD, and nucleosome occupancy at the Il1a enhancer before induction was only moderately affected. Whether this is due to the activity of residual SNF5 under the conditions of our KD, or whether other remodelers play a role in addition to BAF/PBAF at these as well as at other enhancers that may regulate these genes remains to be determined. We also analyzed nucleosome occupancy at the Il12b and Il1a promoters in the SNF5 KD, but effects on nucleosome occupancy both before and upon LPS induction at promoters were small compared to those detected at the enhancers (Fig. 3.5F-J). Together our results indicate that BAF/PBAF regulates nucleosome occupancy at the enhancers of Il12b and Il1a and less so at their promoters. This finding is consistent with our previous data showing that nucleosomes at the promoters of Il12b and Il1a were not stably evicted under inducing conditions (4), which may contribute to the highly stochastic expression of these genes (27,28). 83 Figure 3.5. Nucleosome occupancy in SNF5 KD cells. (A) Nucleosome occupancy at the Il12b enhancer is shown in untreated BMDMs (blue) and SNF5 KD (green) cells grown in the absence of LPS. P-values of Student’s t-tests indicate significance of differences. (B) Nucleosome occupancy at the Il12b enhancer in cells grown for 1 h in the presence of LPS. (C) Nucleosome occupancy at the Il1a enhancer in cells grown in the absence of LPS. 84 Figure 3.5. (cont’d) (D) Nucleosome occupancy at the Il1a enhancer in cells grown for 1.5 h in the presence of LPS. (E) Nucleosome occupancy at control regions indicated in cells grown in the absence of LPS. (F) Nucleosome occupancy at the Il12b promoter in cells grown in the absence of LPS. (G) Nucleosome occupancy at the Il12b promoter in cells grown in the presence of LPS for 1 h. (H) Nucleosome occupancy at the Il1a promoter in cells grown in the absence of LPS. (J) Nucleosome occupancy at the Il1a promoter in cells grown in the presence of LPS for 1 h. 85 FACS analysis reveals effects of SNF5 KD on cytokine expression in single cells To determine whether knockdown of SNF5 merely slowed down the rate of mRNA production in the whole macrophage population or also affected the final levels of cytokine expression, we performed a timecourse of LPS induction in SNF5 KD cells. In untreated macrophages Il12b and Il1a mRNA levels increased during the whole 6 h timecourse of LPS induction as we had shown previously (4)(Fig. 3.6A and B, blue lines). In contrast, when SNF5 was knocked down Il1a and Il12b mRNA levels reached steady-state after 90-180 min and did not increase further (green lines). Moreover, as mentioned above, we found that nucleosome eviction at the Il12b enhancer did not increase further with prolonged LPS induction and nucleosome levels 1.5 h after LPS addition were similar to levels seen after 1 h (Fig. 3.6C). These results suggested that a fraction of cells may not express Il12b or Il1a when levels of SNF5 are limiting. To further address this question we analyzed Il12b expression in single cells by FACS. We used accumulation of newly synthesized intracellular IL12B protein in cells that had been treated with the Golgi inhibitor brefeldin A to prevent protein secretion to assess Il12b expression as described (27). In control macrophages induction of Il12b by LPS for 3 h led to accumulation of significant levels of IL12B protein in about 26% of the cells (compare red to blue areas in Fig. 3.6D, and see scatterplot in Fig. 3.6E) consistent with results by others (27). When we knocked down SNF5 and monitored intracellular SNF5 protein levels, we found that KD reduced mean SNF5 levels in the population (indicated by the vertical lines in Fig. 3.6F). More significantly, the fraction of cells with high levels of SNF5 protein was reduced (compare blue to green shoulder areas in Fig. 3.6F). When we analyzed Il12b expression in SNF5 KD cells, we found that the fraction of 86 cells accumulating IL12B protein was dramatically reduced to about 9% (compare green to red area in Fig. 3.6G and see scatterplot in Fig. 3.6H). Furthermore, we found that cells that expressed Il12b in the SNF5 KD population, expressed only low levels of Il12b and accumulated less IL12B protein than control macrophages (compare the magnitude of the anti-IL12B-APC fluorescence intensity signal in Fig. 3.6E and H on the y-axis). As shown in Fig. 3.6I we found that IL12B protein accumulation correlated with residual levels of SNF5 protein present in SNF5 KD cells, further demonstrating that the remodeler is required for Il12b expression. Figure 3.6. Cytokine expression in SNF5 KD cells. (A) mRNA levels of Il12b in control BMDMs (blue) and SNF5 KD cells (green). Cells were grown in the absence of LPS, or for increasing times in the presence of LPS as indicated. mRNA levels after 1.5 h were set to 100%. Error bars represent the SEM of at least two measurements. (B) mRNA levels of Il1a in cells as in (A). (C) The average occupancy at the three peak nucleosomal positions in the Il12b enhancer is shown in cells as in (A) grown without, or with LPS for 1 h and 1.5 h. Oneway ANOVA followed by a post-hoc Tukey HSD test (p<0.05) shows that differences 87 Figure 3.6. (cont’d) between control and SNF5 KD cells are statistically significant, while occupancy levels in SNF5 KD cells after 1 h and 1.5 h are indistinguishable. (D) IL12B protein accumulation in control BMDMs grown in the absence (blue) or presence of LPS for 3 h (red) was measured by staining with anti-IL12B-APC. Normalized cell counts are displayed as Unit Areas. (E) Scatterplot representation of the data from the experiment described in (D). A threshold was set with unstained control BMDMs. 88 Figure 3.6. (cont’d) (F) SNF5 protein levels in control BMDMs (blue) and SNF5 KD cells (green) were measured by staining with anti-SNF5-AlexaFluor488. Mean fluorescence intensities of each population are indicated by lines of the respective color. (G) IL12B accumulation in control BMDMs grown in the absence (blue) or presence of LPS for 3 h (red), and in SNF5 KD cells grown in the presence of LPS for 3 h (green) was measured by staining with anti-IL12B-APC. Note that data for BMDMs is the same as in (D). (H) Scatterplot representation of the SNF5 KD data from the experiment described in (G). (I) Correlation between IL12B and SNF5 protein levels in SNF5 KD cells grown in the presence of LPS for 3 h was measured by double-staining with anti-IL12B-APC and anti-SNF5-AlexaFluor488. Quartile thresholds were set by analysis of unstained control BMDMs. Discussion Our results suggest that BAF/PBAF is recruited to macrophage-specific enhancers in response to PUER translocation to the nucleus (Fig. 3.1), and we speculate that PU.1 recruits the remodeler to these sites. Whether PU.1 directly interacts with BAF/PBAF subunits or whether the interaction is mediated by another factor remains to be determined. We and others showed previously that PU.1 binds to many enhancers together with C/EBPb, the other macrophage-lineage determining pioneer TF, and C/EBPb has been shown to directly interact with BAF/PBAF and to mediate its recruitment in other myeloid cells, suggesting that C/EBPb may recruit BAF/PBAF 89 together with PU.1 in macrophages (6,8,17). The absence of PU.1 and BAF/PBAF at macrophage-specific enhancers in HSPCs suggests that binding of the pioneer TF and recruitment of the remodeler occurs at some time during macrophage differentiation. Whether the presence of the remodeler in turn stabilizes PU.1 binding to enhancers remains to be determined. If BAF/PBAF is already recruited by PU.1 to some extent prior to gene induction in resting macrophages (Fig. 3.1), how might complete nucleosome eviction be accomplished at enhancers under inducing conditions? We propose that recruited BAF/PBAF increases nucleosome turnover (Fig. 3.7), so that fractional occupancies of enhancer nucleosomes are around 40-60% in a population of resting BMDMs (Fig. 3.2 and 3.5). 90 Figure 3.7. Remodeler assisted competition favors TF over nucleosome binding to sites in enhancers. Our model proposes that recruitment of BAF/PBAF to the distal enhancers of Il12b and Il1a by PU.1 during macrophage differentiation increases turnover of nucleosomes to prevent high occupancy in fully differentiated BMDMs. This results in fractional occupancies of 40-60% for enhancer nucleosomes in the cell population. Under inducing conditions the equilibrium is shifted towards nucleosome removal as signalinduced TFs (e.g., NFkB, AP1) bind to their sites in the enhancers. Note, that increased BAF/PBAF recruitment under inducing conditions (at some enhancers) may further shift the equilibrium towards nucleosome removal. Subsequent steps that result in assembly of a pre-initiation complex at the promoter are not shown. 91 Upon induction by LPS, signal-induced TFs such as NFkB and AP1 are activated and compete with nucleosomes for binding to their sites in the enhancers. This shifts the equilibrium towards nucleosome removal (0-5%). We call this model remodeler assisted competition between TFs and nucleosomes for binding to enhancers. In the absence of BAF/PBAF, enhancers become more highly occupied by nucleosomes, which impairs gene expression in mature cells in response to an appropriate stimulus (Fig. 3.2 and 3.4). Our model predicts that in the absence of BAF/PBAF, nucleosome turnover is low, and signal induced TFs and the transcriptional machinery are recruited only infrequently, since nucleosome formation is favored over TF binding. This prediction is borne out by our experiments in single cells, where we found that the fraction of cells expressing Il12b was reduced in the SNF5 KD (Fig. 3.6G and H). The model further predicts that in the absence of BAF/PBAF, competing nucleosomes reduce the residence times of signal-induced TFs at enhancers, which in turn may decrease the stability of a transcription complex and therefore the transcriptional output from that promoter. Our findings in single cells support this notion, since we found that the levels of IL12B protein that accumulated in individual cells were higher when BAF/PBAF was present at the Il12b enhancer than in its absence in the SNF5 KD (compare the magnitude of the IL12B-APC signal in Fig. 3.6E versus H). This finding suggests that in the absence of SNF5 a transcription complex at a promoter may only fire once before it falls apart, while in the presence of SNF5 such a complex may be stable for several rounds of transcription. Previous studies at various genes have suggested that enhancers can function either by increasing the probability that a competent transcription complex is formed at a promoter or by increasing the probability that 92 another round of transcription is initiated from the same promoter (for a review see (29)). Our results indicate that the distal enhancer of Il12b may play a role in both initiation and re-initiation and that remodeler assisted competition facilitates TF over nucleosome binding to the enhancer to stimulate both processes. Acknowledgements We thank Amy Ralston for helpful discussions; David Arnosti, Jason Knott, Min-Hao Kuo and Erik Martinez-Hackert for careful reading of the manuscript; Louis King, Nara Parameswaran and Michael Steury for help with FACS; Steve Suhr and John LaPres for help with lentiviral transductions. 93 REFERENCES 94 REFERENCES 1. Zaret, K. S., and Carroll, J. S. (2011) Pioneer transcription factors: establishing competence for gene expression. Genes & development 25, 2227-2241. 2. Stamatoyannopoulos, J. A., Snyder, M., Hardison, R., Ren, B., Gingeras, T., Gilbert, D. M., Groudine, M., Bender, M., Kaul, R., Canfield, T., Giste, E., Johnson, A., Zhang, M., Balasundaram, G., Byron, R., Roach, V., Sabo, P. J., Sandstrom, R., Stehling, A. S., Thurman, R. E., Weissman, S. M., Cayting, P., Hariharan, M., Lian, J., Cheng, Y., Landt, S. G., Ma, Z., Wold, B. J., Dekker, J., Crawford, G. E., Keller, C. A., Wu, W., Morrissey, C., Kumar, S. A., Mishra, T., Jain, D., ByrskaBishop, M., Blankenberg, D., Lajoie, B. R., Jain, G., Sanyal, A., Chen, K. B., Denas, O., Taylor, J., Blobel, G. A., Weiss, M. J., Pimkin, M., Deng, W., Marinov, G. K., Williams, B. A., Fisher-Aylor, K. I., Desalvo, G., Kiralusha, A., Trout, D., Amrhein, H., Mortazavi, A., Edsall, L., McCleary, D., Kuan, S., Shen, Y., Yue, F., Ye, Z., Davis, C. A., Zaleski, C., Jha, S., Xue, C., Dobin, A., Lin, W., Fastuca, M., Wang, H., Guigo, R., Djebali, S., Lagarde, J., Ryba, T., Sasaki, T., Malladi, V. S., Cline, M. S., Kirkup, V. M., Learned, K., Rosenbloom, K. R., Kent, W. J., Feingold, E. A., Good, P. J., Pazin, M., Lowdon, R. F., and Adams, L. B. (2012) An encyclopedia of mouse DNA elements (Mouse ENCODE). Genome biology 13, 418. 3. Teif, V. B., Vainshtein, Y., Caudron-Herger, M., Mallm, J. P., Marth, C., Hofer, T., and Rippe, K. (2012) Genome-wide nucleosome positioning during embryonic stem cell development. Nature structural & molecular biology 19, 1185-1192. 4. Gjidoda, A., Tagore, M., McAndrew, M. J., Woods, A., and Floer, M. (2014) Nucleosomes Are Stably Evicted from Enhancers but Not Promoters upon Induction of Certain Pro-Inflammatory Genes in Mouse Macrophages. PLoS One 9, e93971. 5. Ghisletti, S., Barozzi, I., Mietton, F., Polletti, S., De Santa, F., Venturini, E., Gregory, L., Lonie, L., Chew, A., Wei, C. L., Ragoussis, J., and Natoli, G. (2010) Identification and characterization of enhancers controlling the inflammatory gene expression program in macrophages. Immunity 32, 317-328. 6. Heinz, S., Benner, C., Spann, N., Bertolino, E., Lin, Y. C., Laslo, P., Cheng, J. X., Murre, C., Singh, H., and Glass, C. K. (2010) Simple combinations of lineagedetermining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Molecular cell 38, 576-589. 7. Barozzi, I., Simonatto, M., Bonifacio, S., Yang, L., Rohs, R., Ghisletti, S., and Natoli, G. (2014) Coregulation of transcription factor binding and nucleosome occupancy through DNA features of mammalian enhancers. Molecular cell 54, 844-857. 95 8. Tagore, M., McAndrew, M. J., Gjidoda, A., and Floer, M. (2015) The lineagespecific transcription factor PU.1 prevents polycomb-mediated heterochromatin formation at macrophage-specific genes. Molecular and cellular biology 35, 26102625. 9. Bryant, G. O., Prabhu, V., Floer, M., Wang, X., Spagna, D., Schreiber, D., and Ptashne, M. (2008) Activator control of nucleosome occupancy in activation and repression of transcription. PLoS biology 6, 2928-2939. 10. Reinke, H., and Horz, W. (2003) Histones are first hyperacetylated and then lose contact with the activated PHO5 promoter. Molecular cell 11, 1599-1607. 11. Floer, M., Wang, X., Prabhu, V., Berrozpe, G., Narayan, S., Spagna, D., Alvarez, D., Kendall, J., Krasnitz, A., Stepansky, A., Hicks, J., Bryant, G. O., and Ptashne, M. (2010) A RSC/nucleosome complex determines chromatin architecture and facilitates activator binding. Cell 141, 407-418. 12. Badis, G., Chan, E. T., van Bakel, H., Pena-Castillo, L., Tillo, D., Tsui, K., Carlson, C. D., Gossett, A. J., Hasinoff, M. J., Warren, C. L., Gebbia, M., Talukder, S., Yang, A., Mnaimneh, S., Terterov, D., Coburn, D., Li Yeo, A., Yeo, Z. X., Clarke, N. D., Lieb, J. D., Ansari, A. Z., Nislow, C., and Hughes, T. R. (2008) A library of yeast transcription factor motifs reveals a widespread function for Rsc3 in targeting nucleosome exclusion at promoters. Molecular cell 32, 878-887. 13. Hartley, P. D., and Madhani, H. D. (2009) Mechanisms that specify promoter nucleosome location and identity. Cell 137, 445-458. 14. Bultman, S., Gebuhr, T., Yee, D., La Mantia, C., Nicholson, J., Gilliam, A., Randazzo, F., Metzger, D., Chambon, P., Crabtree, G., and Magnuson, T. (2000) A Brg1 null mutation in the mouse reveals functional differences among mammalian SWI/SNF complexes. Molecular cell 6, 1287-1295. 15. Reyes, J. C., Barra, J., Muchardt, C., Camus, A., Babinet, C., and Yaniv, M. (1998) Altered control of cellular proliferation in the absence of mammalian brahma (SNF2alpha). The EMBO journal 17, 6979-6991. 16. Choi, J., Ko, M., Jeon, S., Jeon, Y., Park, K., Lee, C., Lee, H., and Seong, R. H. (2012) The SWI/SNF-like BAF complex is essential for early B cell development. Journal of immunology 188, 3791-3803. 17. Kowenz-Leutz, E., and Leutz, A. (1999) A C/EBP beta isoform recruits the SWI/SNF complex to activate myeloid genes. Molecular cell 4, 735-743. 18. Hu, G., Schones, D. E., Cui, K., Ybarra, R., Northrup, D., Tang, Q., Gattinoni, L., Restifo, N. P., Huang, S., and Zhao, K. (2011) Regulation of nucleosome 96 landscape and transcription factor targeting at tissue-specific enhancers by BRG1. Genome research 21, 1650-1658. 19. Shain, A. H., and Pollack, J. R. (2013) The spectrum of SWI/SNF mutations, ubiquitous in human cancers. PLoS One 8, e55119. 20. Klochendler-Yeivin, A., Fiette, L., Barra, J., Muchardt, C., Babinet, C., and Yaniv, M. (2000) The murine SNF5/INI1 chromatin remodeling factor is essential for embryonic development and tumor suppression. EMBO reports 1, 500-506. 21. Roberts, C. W., Galusha, S. A., McMenamin, M. E., Fletcher, C. D., and Orkin, S. H. (2000) Haploinsufficiency of Snf5 (integrase interactor 1) predisposes to malignant rhabdoid tumors in mice. Proceedings of the National Academy of Sciences of the United States of America 97, 13796-13800. 22. Wang, X., Sansam, C. G., Thom, C. S., Metzger, D., Evans, J. A., Nguyen, P. T., and Roberts, C. W. (2009) Oncogenesis caused by loss of the SNF5 tumor suppressor is dependent on activity of BRG1, the ATPase of the SWI/SNF chromatin remodeling complex. Cancer research 69, 8094-8101. 23. Ramirez-Carrozzi, V. R., Braas, D., Bhatt, D. M., Cheng, C. S., Hong, C., Doty, K. R., Black, J. C., Hoffmann, A., Carey, M., and Smale, S. T. (2009) A unifying model for the selective regulation of inducible transcription by CpG islands and nucleosome remodeling. Cell 138, 114-128. 24. Walsh, J. C., DeKoter, R. P., Lee, H. J., Smith, E. D., Lancki, D. W., Gurish, M. F., Friend, D. S., Stevens, R. L., Anastasi, J., and Singh, H. (2002) Cooperative and antagonistic interplay between PU.1 and GATA-2 in the specification of myeloid cell fates. Immunity 17, 665-676. 25. Laiosa, C. V., Stadtfeld, M., and Graf, T. (2006) Determinants of lymphoid-myeloid lineage diversification. Annual review of immunology 24, 705-738. 26. Doan, D. N., Veal, T. M., Yan, Z., Wang, W., Jones, S. N., and Imbalzano, A. N. (2004) Loss of the INI1 tumor suppressor does not impair the expression of multiple BRG1-dependent genes or the assembly of SWI/SNF enzymes. Oncogene 23, 3462-3473. 27. Weinmann, A. S., Plevy, S. E., and Smale, S. T. (1999) Rapid and selective remodeling of a positioned nucleosome during the induction of IL-12 p40 transcription. Immunity 11, 665-675. 28. Shalek, A. K., Satija, R., Adiconis, X., Gertner, R. S., Gaublomme, J. T., Raychowdhury, R., Schwartz, S., Yosef, N., Malboeuf, C., Lu, D., Trombetta, J. J., Gennert, D., Gnirke, A., Goren, A., Hacohen, N., Levin, J. Z., Park, H., and Regev, 97 A. (2013) Single-cell transcriptomics reveals bimodality in expression and splicing in immune cells. Nature 498, 236-240. 29. Blackwood, E. M., and Kadonaga, J. T. (1998) Going the distance: a current view of enhancer action. Science 281, 60-63. 30. Nociari, M., Ocheretina, O., Schoggins, J. W., and Falck-Pedersen, E. (2007) Sensing infection by adenovirus: Toll-like receptor-independent viral DNA recognition signals activation of the interferon regulatory factor 3 master regulator. Journal of virology 81, 4145-4157. 31. Shechter, D., Dormann, H. L., Allis, C. D., and Hake, S. B. (2007) Extraction, purification and analysis of histones. Nature protocols 2, 1445-1457. 98 Chapter 4: GNO-seq quantifies nucleosome occupancy at gene promoters and enhancers in LPS stimulated macrophages This chapter represents a manuscript under review in Genome Research (2017, manuscript ID: GENOME/2017/227504). Authors who contributed to this study were: Michael McAndrew, Alison Gjidoda, Wolfgang Resch, Kyong-Rim Kieffer-Kwon, Rafael Casellas, and Monique Floer. 99 Abstract The transcriptional response of macrophages to a variety of pathogens has been well studied, and stimulation by bacterial lipopolysaccharides (i.e., LPS) serves as a paradigm for inducible gene expression in mammalian cells. We used GNO-seq (Global Nucleosome Occupancy-sequencing) an extension of conventional MNase-seq, to quantify differences in nucleosome occupancy genome-wide between resting and LPSinduced mouse macrophages. We find that the majority of LPS-induced genes are already expressed to some extent in resting macrophages, and increased expression is associated with further nucleosome depletion at their promoters but also with partial nucleosome depletion in regions upstream of promoters and in the 5’ ORFs. In contrast, we show that the promoters of a small group of highly induced genes that are repressed in resting macrophages remain associated with nucleosomes under inducing conditions. This finding is in agreement with our previous findings at two cytokine genes (i.e., Il12b and Il1a), and we propose that tight control of promoter access by chromatin may limit expression of this group of genes. Our analysis also reveals differences in nucleosome occupancy at different types of enhancers involved in macrophage biology (i.e., constitutive, poised, latent enhancers), and we show that levels of nucleosome occupancy in resting macrophages are indicative of the response to LPS. GNO-seq therefore allows characterization of enhancers beyond histone modifications and TF binding, and we propose that incorporation of quantitative nucleosome occupancy information has the potential to facilitate identification of functional elements in other systems. 100 Introduction Nucleosomes, the basic building blocks of chromatin, are generally thought to restrict access of sequence specific transcription factors (TFs) and the transcriptional machinery to DNA. Determining nucleosome binding in the genome has therefore been the focus of intense research in the field of gene regulation. Approaches commonly used in these studies take advantage of enzymes that cut nucleosome-free DNA but leave nucleosomal DNA intact, and micrococcal nuclease (MNase) from Staphylococcus aureus has emerged as the enzyme of choice. MNase is an endo/exonuclease that preferentially cuts linker DNA, but can also digest nucleosomal DNA when present in excess or given enough time to complete the reaction. Enzymes such as DNase I and Tn5 transposase on the other hand, are more restricted by chromatin structure, possibly due to their larger sizes and approaches such as DNaseseq and ATAC-seq that use these enzymes, therefore preferentially identify regions highly accessible to sequence-specific TFs and the transcriptional machinery (1-3). In contrast, MNase-seq - performed after limited digestion of chromatin and sometimes in conjunction with histone ChIP - reveals preferred nucleosomal positions anywhere in the genome and has been used to assemble nucleosome position maps for various organisms (4-8). Variations of the assay that utilize low enzyme concentrations or digestion performed in low salt conditions have revealed “fragile” or MNase-sensitive nucleosomes at specific genomic locations that are not observed by standard protocols (9-13). What the function of these fragile nucleosomes may be remains elusive. While MNase-seq reveals preferred nucleosomal positions in the genome, the assay in its current form provides only limited information about quantitative levels of occupancy 101 of any given nucleosomal position, due to the inherent sequence bias of the enzyme and of Illumina sequencing itself. In other approaches, such as ChIP-seq sequence, bias is typically overcome by normalizing the data to input chromatin and input chromatin has been taken into account in a quantitative MNase assay developed by (14). Fractional nucleosome occupancies in a population of cells are derived in this assay by curve-fitting of digestion data from a large range of MNase concentrations to two-state exponential decay functions, which largely eliminates the underlying sequence bias in MNase digestion. Studies using this approach have provided valuable insights into the role of chromatin in inducible gene expression in different organisms (15-17). However, because this assay quantifies DNA by qRT-PCR, its use has been limited to the focused analysis of specific genomic regions of interest. Two recent studies used MNase digestion with a range of concentrations followed by Illumina sequencing and linear regression to determine differences in chromatin accessibility (11,18). However, because undigested input chromatin was not analyzed, effects of the underlying DNA sequence on MNase digestion could not be separated from differences in nucleosome occupancy in a population of cells. Furthermore, linear regression to fit data that follows exponential digestion curves - as shown by Bryant et al. (14) - is likely to introduce considerable artifacts. Most importantly, because MNase-data from different samples (i.e., different growth conditions) was not normalized to an external reference, any broader changes in nucleosome occupancy that may pertain to larger parts of the genome could not be detected. LPS induction of macrophages has become a paradigm for inducible gene expression in mammalian cells and previous studies have identified the genes whose expression 102 changes in response to LPS as well as the putative transcriptional elements involved in their regulation (19-25). LPS induction relies on two classes of transcription factors i.e., the macrophage lineage determining TFs PU.1 and C/EBPβ, and the signal-induced TFs NFκB, AP-1 and IRF. PU.1 and C/EBPβ are already bound to enhancers of many inducible genes in resting macrophages and have been shown to play a role in enhancer priming for later induction (17,20,21,26,27). We recently showed that PU.1 recruits the remodeler BAF/PBAF to the enhancers of two example pro-inflammatory genes during macrophage differentiation, which keeps these enhancers accessible and occupied only by intermediate levels of nucleosomes in mature macrophages and facilitates complete nucleosome eviction in response to LPS (28). In the absence of the remodeler the enhancers become associated with highly occupied nucleosomes, and their response to LPS is impaired. Here we have analyzed the changes in nucleosome occupancy at LPS-responsive genes genome-wide, using a novel method that combines the crucial aspects of the assay of Bryant et al. (14) with Illumina sequencing to determine fractional nucleosome occupancies anywhere in the genome. Significantly, we sequence pooled DNA isolated from chromatin digested with a range of enzyme concentrations, to capture nucleosomal DNA at most genomic regions, and we sequence DNA isolated from input chromatin, which we use for normalization. We have termed the assay GNO-seq for Global Nucleosome Occupancy sequencing and use the approach to measure quantitative changes in nucleosome occupancy upon LPS-induction of bone marrow derived mouse macrophages (BMDMs). While previous studies have compared chromatin in different cell-types, the immediate changes in nucleosome occupancy 103 upon gene induction at transcriptional regulatory regions genome-wide are less well understood in mammalian systems. Our study identified distinctive changes in nucleosome occupancy at promoters and transcriptional enhancers, and also detected broader genome-wide changes associated with macrophage activation. Our results further indicate that the levels of nucleosome occupancy at transcriptional regulatory regions are indicative of their response to an inducing stimulus. These findings highlight the importance of obtaining quantitative information on nucleosome occupancy when examining functional elements. We propose that GNO-seq has the potential to help identify regulatory elements in other cell-types and systems, and to distinguish functional elements from regions that lack regulatory activity but are nonetheless associated with histone modifications and TF-binding. Experimental Procedures Cell isolation and sample preparation Bone marrow cells were isolated from 6-8 week old female C57BL/6 mice (NCI, Charles River) under Institutional Animal Care and Use Committee oversight and bone marrow derived macrophages (BMDMs) were generated by growth in the presence of M-CSF as described (17). Cells were either grown in the absence or presence of LPS for 1.5 h (i.e., rM and aM for resting and activated macrophages, respectively). Formaldehyde cross-linked chromatin from 1.5 x 107 uninduced or induced cells was split into 24 samples, 2 samples remained undigested and 22 samples were digested with 0.002713.3 U of MNase (NEB), respectively. Digestion was analyzed by qRT-PCR and curvefitting at different locations in the genome as described (17) and primers can be given upon request. For the Input-fractions DNA isolated from the two undigested control 104 samples and a sample digested with the lowest concentration of MNase (i.e., 00027 U) were pooled. For the MNase-fractions DNA isolated from samples digested with 0.014, 0.020, and 0.030 U MNase respectively were pooled. DNA was isolated using a Qiaquick 96 well DNA purification kit (Qiagen). DNA in the Input-fractions was sonicated using a Covaris sonicator to yield fragments between 130-200 bp. Lambda DNA (Promega) was sonicated using the same conditions and 0.055 µg sonicated lambda DNA was added as a spike-in control to each Input and MNase-fraction so that the amount of lambda DNA as a fraction of total DNA was between 1-4% per sample. Illumina library preparation and sequencing DNA isolated from Input and MNase-fractions was blunt-ended with End-It DNA end repair kit (Epicenter) and polyadenylated with Taq DNA polymerase (Invitrogen) in the presence of 200 µM dATP for 40 min at 70°C. Samples were purified by column (DNA clean & concentrator kits, Zymo Research) after each reaction. Illumina compatible adaptors (Bio Scientific) were then ligated using T4 DNA ligase (Enzymatics), and the reaction was purified once with AMpure XP magnetic beads (Beckman Coulter). Samples were PCR amplified for 4 cycles with KAPA HiFi DNA polymerase mix (KAPA Biosystems) and purified by column. Paired-end sequencing data (i.e., 50 cycles) was acquired on HiSeq 2000 and 2500 sequencers (Illumina). See Suppl. Fig. 4.S2 for insert lengths and read numbers in each sample. Data processing Raw FASTA files of paired-end Illumina sequencing reads for either the Input or MNasefractions of rM and aM were trimmed using Trimmomatic 0.33 (default settings for Truseq3-PE adapters except LEADING = 20, TRAILING = 20, SLIDING WINDOW = 105 4:30)(29) and mapped to either the Mus musculus genome (UCSC mm9) or the Enterobacteria phage lambda genome (GenBank: J02459.1) using Bowtie2 2.2.6 (-phred33 --local --sensitive- local -I 0 -X 1000 --no-discordant --no-mixed --fr --nounal)(30). Merged BAM files containing all reads with MAPQ ≥ 30 for the MNase or Input-fractions from rM and aM respectively, were generated using SAMtools 0.1.19 (31). Generation of lambda normalized GNO-seq tracks To obtain GNO-seq tracks for rM and aM (i.e., rMratio_GNOseq and aMratio_GNOseq), we normalized the number of Input and MNase-seq reads to the number of lambda DNA reads per Million in each fraction as described for external reference normalization (32)(see also Suppl. Experimental Procedures) and then derived the ratio of referencenormalized MNase over Input RPM. Scaling factors for rM Input and rM MNasefractions are 2.353 and 1.200, and for aM Input and rM MNase-fractions 2.575 and 0.798, respectively. GNO-seq bigwig files were then generated using deepTools bamCompare (-b1 = MNase, -b2 = input, --ratio = ratio, bin size -bs = 1, -scaleFactors)(33). To distinguish regions without nucleosome occupancy from regions that are undersampled and lack reads in the Input-seq data, pseudocount was set to 0, so that a value of zero was assigned to regions without occupancy and a value of “Infinity” to regions lacking coverage in the Input-data. bwtool was used to remove regions lacking reads in the Input (equal “Infinity”)(34). These are the files used for all subsequent analyses. GNO-seq tracks at individual genomic locations are displayed using IGV (35). 106 Random genomic sampling To obtain random genomic regions we first generated a total of 100,000 regions of 6 kb using bedTools random from GNO-seq bigwig files for rM and aM (-l 6000, -n 100,00). These regions were shuffled into regions with sufficient GNO-seq data coverage and excluded from regions with no coverage using bedTools shuffle (-incl covered, -excl not covered). Bigwig files containing occupancy data for only these regions were produced using bwtool remove (mask random –inverse), and the resulting bigwig files were converted to bed format as described and merged using bedtools merge (-d 0, -c 5, -o mean, min, max) to produce 60,749 continuous genomic regions. The resulting regions were used for determination of average nucleosome occupancy and GC content. GC content To determine the GC content in regions without Illumina-sequencing coverage and alternatively, in regions without nucleosome occupancy, we first identified such regions in the genome. We generated bigwig coverage tracks from merged BAM files of Input and MNase-fractions from rM and aM using deepTools bamCoverage (bin size -bs = 1, -scaleFactor = 1). These files were converted from bigwig to wig format using UCSC tools bigWigToWig (36), then converted from wig to bed format using BEDOPS convert2bed (--zero-indexed)(37). To generate regions without nucleosome occupancy we used bedtools intersect (-v -a Input -b MNase)(38) to generate bed files containing regions sequenced in the Input, but not in the MNase-fractions (i.e., regions without nucleosome occupancy). To generate regions without Illumina-sequencing coverage we generated intervals that were not present in the Input or MNase-fraction bed files using bedtools complement (-g mm9). We then generated bed files containing regions not 107 sequenced in either fraction using bedtools intersect (-u -a Input -b MNase)(i.e., regions without Illumina-sequencing coverage). We used bedtools nuc (-fi mm9) to determine the GC content of each region, and the boxplot function in R to calculate the median GC content and create boxplots. We also calculated the average GC content normalized to the size of each category by dividing the number of GC base pairs by the total number of base pairs in each category. GC content was calculated similarly in the random regions generated as described above. Heatmaps and average nucleosome occupancy plots To sort genes by levels of expression in aM we used RNA-seq data of uninduced and induced (i.e., 2 h LPS) BMDMs from Mancino et al. (22). We first assigned Refseq IDs and coordinates to genes in this dataset using BioMart (39). To exclude TSSs with less than 90% coverage in the region 3 kb upstream and downstream of the TSSs, bedtools intersect was used to determine sequence coverage of the Input-fraction in these regions (-wo -a TSS +/- 3 kb -b Input). Regions with less than 90% coverage were filtered out using a custom python script available upon request. This resulted in a total of 23,265 unique TSSs, which were used for subsequent analyses. TSSs were separated into groups based on the quartiles of FPKM expression values of the associated genes in the presence of LPS. Genes were aligned at their TSSs and heatmaps and average nucleosome occupancy plots in regions 3 kb upstream and downstream were generated for each quartile using deepTools computeMatrix (-referencePoint = TSS, bin size -bs = 1) and deepTools plotHeatmap (-missingDataColor = yellow, --sortUsingSamples 2 (aM), --colorList red, white, blue, grey, black)(33). Genes in the most highly expressed quartile were further separated 108 into four clusters based on nucleosome occupancy in aM across the entire region using k- means clustering in deepTools plotHeatmap (settings as above, with --kmeans = 4). To determine nucleosome occupancy around TSSs of genes that are lowly or not expressed in rM but induced by LPS in aM we excluded genes expressed in rM (i.e., FPKM > 1) from the 23,265 Refseq genes and sorted the remaining genes into groups A-D according to levels of expression in rM and aM. Genes in each group were aligned at the TSS, and heatmaps and average plots of nucleosome occupancy were generated as described. To analyze nucleosome occupancy at enhancers we used the 69,559 macrophage enhancers identified by Ostuni et al. (23). We adhered to the enhancer classification suggested by these authors, but further split the “not steady” category into “not steadyactivated” and “not steady-repressed” enhancers based on increases and decreases in the levels of H3K27ac ChIP-seq signals, respectively, in cells grown for 4 h and/or 24 h in the presence of LPS (23)(Table 4.1). For alignment at the site of PU.1 binding we first used bedTools intersect to identify ChIP-seq peaks of PU.1-binding within enhancers in cells treated for 4 h with LPS (23). Subsequently, we aligned enhancers containing a PU.1-peak at the midpoint of each PU.1-peak and generated heatmaps and average plots of nucleosome occupancy in rM and aM in regions 3 kb upstream and downstream as described. For alignment of enhancers at their sites of p300 recruitment we first used bedTools intersect to find overlap between previously identified p300 ChIP-seq peaks in cells treated for 2 h with LPS (20) and putative enhancers of Ostuni et al. We subsequently aligned enhancers containing p300-peaks at the midpoint of the p300- 109 peaks, and generated heatmaps and average plots of nucleosome occupancy in rM and aM in the surrounding regions. Identification of nucleosome depleted regions To identify depleted regions in rM and aM, we first removed regions with occupancy values higher than a threshold (i.e., 70%, 75%, and 80% for rM; 20%, 25%, and 30% for aM) from GNO-seq data using bwtool remove (34). The resulting bigwig files contained only regions with nucleosome occupancy below the set threshold and were converted to wig using UCSC tools bigWigToWig (36), and then to bed format using BEDOPS convert2bed (--zero-indexed)(37). To produce regions of a defined length and allowing for gaps in the occupancy defined by the threshold the files were first merged using bedTools merge (-d [gap size])(38) and then filtered by size. The lengths of gaps in occupancy we allowed and the lengths of the depleted regions can be found in Suppl. Table 4.S1. To determine the fraction of the regions of interest (i.e., enhancers, promoters, super- enhancers etc.) that encompass a (partially) depleted region we used bedTools intersect (default settings using –u to return regions of interest in the –a file overlapping regions in the –b file). Scatterplots and bargraphs were generated in Microsoft Excel. Gene ontology analysis Gene ontology analysis was performed using the Gene Ontology Consortium web browser using the GO Ontology database release 2017-06-29 (40). De novo motif search De novo motif search in poised-activated and poised-not activated enhancers was performed using HOMER 4.7.2 findMotifsGenome.pl (21) with GC-content matched 110 genomic regions as the control sequence set (default settings with genome = mm9, size = given, -mask). Motif logos were generated using HOMER 4.7.2 motif2Logo.pl. Nucleosome occupancy at super-enhancers To determine nucleosome occupancy at SEs in macrophages and other cell-types we used the super-enhancers identified by (41). We aligned GNO-seq data from rM and aM at the midpoint of each superenhancer specific to ESC, BMDM, Myotubes, Pro-B and T helper cells using deepTools computeMatrix (--referencePoint = center, bin size -bs = 1). Flanking sequence was added according to the average size of SEs from different cell-types and heatmaps and average nucleosome occupancy plots were generated as described. Results GNO-seq analysis In GNO-seq we digest cross-linked chromatin with 24 different enzyme concentrations (spanning nearly four orders of magnitude i.e., 0.0027-13.3 U) and then fit the digestion data obtained by qRT-PCR at representative genomic regions to two-state exponential curves to determine the range of enzyme concentrations that captures most nucleosomal fragments in the genome as previously described (Fig. 4.1A)(14, 17). While sequencing DNA isolated from digestion with the entire range of enzyme concentrations followed by curve-fitting would ideally yield the most accurate quantitative information, such an approach is prohibitive because of the cost of sequencing 24 different samples at sufficient coverage of the mouse genome. Instead, we found in initial MNase digestion experiments using qRT-PCR that pooling DNA from digestion with a limited range of enzyme concentrations, representing the “lip” of the 111 digestion curves (i.e., MNase-fraction in Fig. 4.1B), largely eliminated the sequence bias of the enzyme and preserved nucleosomal-sized DNA fragments at most sites in the genome (Suppl. Fig. 4.S1). In GNO-seq we therefore pooled three MNase-digested samples, which we sequence as the MNase-fraction. To account for any sequence bias of Illumina-sequencing itself and any bias introduced by our chromatin extraction procedure, we also pooled undigested and lightly digested chromatin as the Inputfraction and sheared the isolated DNA by Covaris sonication to fragment sizes similar to those in the MNase-fraction before Illumina sequencing (i.e., the majority of fragments are 130-200 bp in length, see Suppl. Fig. 4.S2). Furthermore, we spiked in Covaris sonicated lambda DNA into MNase and Input- fractions before Illumina-library preparation, which we used for external reference normalization of sequencing reads obtained from different fractions and growth conditions as previously described (43,44). Specifically, we obtained reference- normalized reads per Million (RPM) from MNase and Input-fractions by multiplying the number of aligned mouse reads with a normalization factor as described (32)(see Experimental Procedures), and then calculated the fractional nucleosome occupancy (i.e., the % nucleosome occupancy) as the ratio of MNase over Input reference-normalized RPM. Initial proof-of-principle experiments using qRT-PCR measurements at many different genomic locations showed that occupancies derived as the ratio of MNase over Input correlated well with occupancies obtained by curve-fitting (R2=0.90, Suppl. Fig. 4.S1), indicating that this approach might allow determination of nucleosome occupancies anywhere in the genome at single nucleotide resolution. 112 Figure 4.1. GNO-seq analysis. C A Nucleosome sample preparation Extract cross-linked chromatin from whole cells by light sonication in detergent buffer 1.0 0.8 0.6 Digest soluble chromatin w/MNase (0U, 0.0027U-13.3U) 0.4 Determine nucleosome-protected genomic DNA by qRT-PCR at example locations and curve-fitting Input- fraction Pool 2 undigested and 1 lightly digested sample => Covaris sonication => 130-200bp 0.2 MNase-fraction Pool 3 digested samples near inflection point => mononucleosomes => ~150bp Add Covaris sonicated lambda DNA (130-200bp) to Input and MNase-fractions ea. 0 aM rM no coverage aM rM no occupancy D Prepare Illumina libraries and sequence r = 0.69, p < 10-5 Input [DNA] E MNase ln2 [MN] r = 0.72, p < 10-5 (A) Flowchart illustrating the workflow for GNO-seq sample preparation, Illumina library preparation and sequencing. 113 Figure 4.1. (cont’d) (B) Curve-fitting of MNase-digestion data measured by qRT-PCR at a location in the Il12b enhancer. Orange boxes indicate the samples pooled for the Input and MNasefractions, respectively. (C) GC content in regions without Illumina sequencing coverage and in regions without nucleosome occupancy determined in rM and aM as described in Experimental Procedures. (D) and (E) Relationship between GC content and nucleosome occupancy in rM and aM, respectively. Density plots show the GC content of ~60,000 randomly chosen regions of about 6 kb as a function of the average occupancy in these regions. Note that average occupancies include linker DNA and are distinct from occupancies at preferred nucleosomal positions in these regions. Pearson’s correlation coefficients and p-values are indicated in the figures. GNO-seq validation Our initial survey of GNO-seq data obtained from resting (rM) and activated macrophages (aM), grown for 1.5 h in the presence of LPS, indicated that levels of nucleosome occupancy at regulatory regions of three pro-inflammatory genes (i.e., Il12b, Il1a and Ifnb1) were comparable to our previous results using the qRT-PCR based approach (Suppl. Fig. 4.S3)(17). For example, GNO-seq detected complete nucleosome eviction upon LPS induction at the distal enhancers of Il12b, Il1a and Ifnb1, which were occupied by intermediate levels of nucleosomes before induction in rM. GNO-seq also showed that the promoters of Il12b and Il1a remained associated with nucleosomes under inducing conditions, while nucleosomes at the promoter of Ifnb1 114 were evicted. Together our data show that GNO-seq faithfully detects changes in nucleosome occupancy associated with LPS induction at representative genomic locations. To further validate the approach and to compare it to standard MNase-seq protocols that use only one concentration of MNase without input normalization, we also analyzed the MNase- seq fraction of our data alone without considering the input using published methods. We used either the MNase option in deepTools bamCoverage or DANPOS according to published protocols (33,42). Both programs include a dyad alignment step, which centers reads on the nucleosome dyad based on the assumption that certain nucleosome positions are preferred in a population of cells, and DANPOS further adjusts read lengths while the MNase option of bamCoverage only considers reads between 130 and 200 bp. We found that either approach detected loss of nucleosomes at the Il12b and Il1a enhancers upon LPS induction, as well as retention of nucleosomes at the Il12b promoter (Suppl. Fig. 4.S4). However, levels of nucleosome occupancy measured by the qRT-PCR based assay were best reproduced using GNOseq. We note that inclusion of Input-data in GNO-seq precluded dyad alignment, since it cannot be performed after taking the ratio of MNase over Input-data. Furthermore, we found that dyad alignment of both Input-data and MNase-data before taking the ratio greatly distorted the resulting nucleosome occupancies (Floer, M and McAndrew, M.J., unpublished results). Nevertheless, we noted that dyad alignment of MNase-data generally overemphasizes occupancy at preferred nucleosomal positions. We therefore find that nucleosome occupancies are best quantified by Input-normalization in GNOseq without dyad alignment. In addition, we found that Input-normalization in GNO-seq 115 reduced apparent differences in occupancies seen broadly in MNase-seq data between different regions of the genome that are likely an artifact of chromatin extraction and Illumina sequencing bias (Suppl. Fig. 4.S5). Nevertheless, a small decrease in nucleosome occupancy associated with macrophage activation in large parts of the genome, could be detected by GNO-seq as well as MNase-seq analysis (Suppl. Fig. 4.S5). This result is consistent with previous findings of a general loss of nucleosomes upon LPS induction of macrophages derived from fetal liver- derived monocytes (45), but what the significance of this general loss of nucleosome occupancy might be remains to be determined. We recently found that LPS activation of B-cells resulted in a similar wide-spread loss in nucleosome occupancy, and we showed that this was accompanied by a general decondensation of chromatin (Kieffer-Kwon et al, manuscript accepted). Whether macrophage activation is also associated with chromatin decondensation remains to be determined. Normalization to the input in GNO-seq also allows us to distinguish regions that are nucleosome-free from regions that are simply under-sampled and therefore lack reads also in the Input-fraction. The ability to unequivocally identify sites of nucleosome depletion is an important advance of GNO-seq over conventional MNase-seq, since transcriptional regulatory activity is usually associated with nucleosome eviction and researchers therefore often focus exclusively on nucleosome-free sites. Because previous MNase-seq analyses lack input normalization we hypothesize that many studies may erroneously have categorized some genomic regions as nucleosome-free that simply lack any sequence coverage. To determine if under-sequenced regions (i.e., no reads in MNase and Input-fractions) generally have a different sequence 116 composition than other regions in the genome, we determined their GC content. We found that the median GC content in such regions was around 37% and average occupancy after adjustment to the total number of base pairs was around 40%, which is similar to the overall GC content of the mouse genome (i.e., 42% (46))(Fig. 4.1C). This indicates that undersampling of regions in Illumina sequencing is unlikely a direct result of the underlying DNA sequence. We sequenced each sample to 300-400 million paired-end reads (see Suppl. Fig. 4.S2B), and while we found that higher levels of sequence coverage performed for samples in a parallel study using B-cells (~1,000 million paired-end reads (Kieffer-Kwon et al, manuscript accepted) allowed inclusion of a small fraction of additional genomic locations, it did not significantly alter the fractional occupancies measured and we conclude that insufficient sequencing depth is not the reason for missing data in our current study. In contrast, we found that the GC content in regions that lack nucleosomes in rM or aM (i.e., no reads in MNase, but reads in Input-fractions) was low, around 23% (Fig. 4.1C) or 28% after adjustment for the total number of base pairs. This result suggests that low nucleosome occupancy in mouse macrophages is related to low GC content, consistent with previous results from yeast and other organisms that showed a preference of nucleosome formation at GC-rich sequences (7,47-51). To determine overall correlation between nucleosome occupancy and GC content we analyzed ~60,000 random genomic regions of around 6 kb. We plotted average GC content versus average nucleosome occupancy over each region and found a strong correlation with Pearson coefficients of r = 0.69 (p < 10-5) for rM and r = 0,72 (p < 10-5) for aM, respectively (Fig. 4.1D and E). 117 Nucleosome occupancy surrounding transcriptional start sites Earlier studies of chromatin changes associated with gene induction in yeast and also in mammalian cells had suggested that nucleosomes may generally occlude promoters of silent genes and have to be removed to allow gene expression (52-54). However, our previous studies at three example genes in macrophages had indicated that some promoters remain associated with nucleosomes even under inducing conditions (17,28). To determine changes at promoters genome-wide upon LPS induction of macrophages we analyzed nucleosome occupancy around the TSSs of 23,265 mouse Refseq genes, for which published gene expression data was available in BMDMs and for which we had sufficient GNO-seq coverage (i.e., >90% coverage in the Input-sequence data) in a 6 kb window including 3 kb upstream and downstream of the TSS. We separated the genes into four groups (i.e., quartiles Q1-4,) with Q1 containing the genes most lowly expressed in aM, as inferred from mRNA levels 2 h after LPS addition (data taken from (22)). We aligned the genes in each quartile by their TSSs and generated average nucleosome occupancy plots and heatmaps of nucleosome occupancy in surrounding regions (Fig. 4.2A and B). Genes within each quartile were sorted by levels of nucleosome occupancy in aM over the whole region in the heatmaps (i.e., low to high from bottom to top). The average occupancy plots and heatmaps show that the promoter nucleosome position just upstream of the TSS was depleted at most of the highly expressed genes (i.e., genes in Q3 and Q4). The nucleosome-depleted region usually corresponded to the size of a single nucleosome and was flanked downstream by a well-positioned +1 nucleosome in the ORF. Such nucleosome arrangements have previously been described at the promoters of genes actively transcribed in yeast and 118 higher organisms (6,8,10,16,55-57). We did not detect extensive phasing of additional nucleosomes beyond the +1 position, presumably because of our omission of the dyad alignment step as discussed above. We believe that in the absence of dyad alignment nucleosome positions including that of the +1 nucleosome are less emphasized. This is consistent with our finding that peak occupancies at the +1 position were only around 50% when we averaged many different genes (Fig. 4.2A), while +1 occupancies at individual genes reached up to 100% (e.g. Fig. 4.2E). Significantly, we found that most promoters of the genes highly expressed in aM (in Q3 and Q4 in Fig. 4.2B) were already significantly depleted in rM and often became completely nucleosome-free in aM. This result is consistent with our finding that most of these genes were already expressed at some basal level in rM as shown by our analysis of the gene expression data of Mancino et al. (22). Analysis of genes in Q1-4 showed that most of the genes significantly expressed in response to LPS – setting an arbitrary threshold for significant expression to FPKM > 1 – were already significantly transcribed in rM (Fig. 4.2I Q3 and Q4, see also Suppl. Table S1). Furthermore, we found that the majority of genes that were expressed to a lesser extent in aM – setting the threshold to FPKM > 0 – were nevertheless already transcribed at a low level in rM (Fig. 4.2J, Q2, Q3 and Q4). In contrast, we found that the majority of genes in Q1 and Q2 showed very little nucleosome depletion as expected from the low levels of expression of these genes in rM and aM. 119 Figure 4.2. Nucleosome occupancy at promoters. A rM E aM 0.5 10 kb 1.0 Q1 Q2 Q3 Q4 0.3 Chr 5 0.1 0 Cxcl9 3.5 -3.0 TSS rM 3.0 -3.0 TSS 3.0kb aM aM 0 3.5 rM 0 F Q1 Chr 11 12 kb 1.0 C1 0 Q2 Irf1 3.5 C2 aM 0 3.5 rM Q3 C3 0 G Chr 1 14 kb 1.0 C4 Q4 0 Mapkapk2 3.5 aM 0 3.5 -3.0 TSS 3.0 -3.0 TSS 3. 0kb nucleosome occupancy (%) rM 0 D 0.5 rM aM C1 C2 C3 C4 0.3 Chr 11 H 11 kb 1.0 0 0.1 Peli1 3.5 -3.0 I TSS 3.0 -3.0 TSS J FPKM > 1 aM 3.0kb FPKM > 0 100 100 80 80 0 3.5 rM 0 K Chr 17 14 kb 1.0 60 60 40 0 40 rM aM 20 3.5 20 0 3.5 rM 0 0 Tnf aM 0 Fig. 2 Nucleosome occupancy was analyzed by GNO-seq in regions 3 kb upstream and Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 downstream of TSSs after aligning 23,265 Refseq genes at their TSSs. 120 Figure 4.2. (cont’d) (A) and (B) Genes were separated into quartiles (Q1-4) by gene expression levels in aM (i.e., FPKM at 2 h LPS; taken from (22)) with Q4 containing the most highly expressed genes. Average nucleosome occupancy plots and heatmaps are shown. Within each quartile genes were sorted by levels of nucleosome occupancy in aM over the whole region (i.e., low to high from bottom to top). (C) and (D) Genes in Q4 were further separated into four clusters (C1-4) by k-means analysis, and heatmaps and average plots of nucleosome occupancy in each cluster are shown. (E-H) Nucleosome occupancy in rM (blue) and aM (red) at representative genes taken from C1-4, respectively. Pol II ChIP-seq peaks from (23) in resting (grey) and activated (cyan) macrophages 4 h after LPS induction are shown underneath each panel. (I) and (J) Gene expression data from (22) is shown for Refseq genes in each quartile of (A). Gene expression in rM is shown in blue and in aM in yellow. Solid bars indicate expression using a cut-off of FPKM > 1 (I) or FPKM > 0 (J) in each quartile, and hatched bars indicate genes whose expression falls below this value. (K) Nucleosome depletion in aM over the whole Tnf gene locus is shown as in Fig. 5.2E-H. Strikingly, we found that in addition to further nucleosome eviction at promoters upon LPS induction, nucleosomes were also partially lost from the 5’ ORFs and from regions upstream of promoters of many highly expressed genes (i.e., Q3 and Q4 Fig. 4.2A and B). To further dissect the changes in nucleosome occupancy at the most highly expressed genes we separated genes in Q4 into four clusters using k-means clustering 121 (Fig. 4.2C and D). Genes in clusters C1-4 showed varying levels of additional depletion upstream and/or downstream of the completely depleted promoter nucleosome and examples of genes from each cluster (i.e., C1-4, respectively) are shown in Figs. 4.2EH. We found that genes in C1 showed only modest depletion in regions other than their promoters, and a minority of these genes even retained significant levels of promoter nucleosomes when highly expressed (e.g., Cxcl9, Fig. 4.2E). This finding is reminiscent of our previous results at the promoters of Il12b and Il1a, where nucleosomes were present even under inducing conditions (17). In contrast, we found that genes in cluster C2 were mostly associated with additional, partial depletion upstream of the promoter (e.g., IRF1 in Fig. 4.2F), while genes in cluster C3 were partially depleted in the 5’ORF and also upstream (e.g., Mapkapk2 in Fig. 4.2G). Genes in cluster C4 were most significantly depleted both upstream and downstream of the promoter (e.g., Peli1 Fig. 4.2H). However, depletion in these regions was incomplete and fractional nucleosome occupancies remained higher than at the promoters. In contrast, we identified a small group of genes that showed almost complete loss of nucleosomes over a broad region encompassing the whole gene locus (e.g., Tnf Fig. 4.2K). Other genes in this group include Ccl2, Ccl3, Ccl4 and Ccl7. This result is reminiscent of findings at heat shock genes in Drosophila and yeast, where nucleosomes are broadly lost from the gene locus (58-59) but such dramatic loss of nucleosomes has not been described for other classes of highly expressed genes. Nevertheless, our data show that complete nucleosome loss is not a prerequisite for high levels of gene expression, since genes in clusters C1-4 showed a similar range of expression in response to LPS (see Suppl. Fig. 4.S6). For the example genes in Fig. 4.2E-H the levels of Pol II binding in the absence 122 and presence of LPS (i.e., Pol II ChIP-seq peaks before and 4 h after LPS-induction taken from (23)) are indicated underneath each panel in grey and cyan, respectively. Together our data indicate that high levels of gene expression are associated with nucleosome depletion at most promoters and unexpectedly, that increased expression in response to LPS often leads to additional partial nucleosome depletion of regions surrounding promoters and extending into ORFs. Promoter nucleosomes at highly induced genes Our finding that a number of highly induced genes (e.g., Cxcl9, Il12b, Il1a) retained nucleosomes at their promoters under inducing conditions prompted us to investigate highly induced genes further. Previous studies in macrophages classified LPS-induced genes based solely on the fold induction of expression, but did not distinguish genes with different absolute levels of expression in the presence of LPS, nor did these studies distinguish genes with significant basal expression from those truly repressed in resting macrophages (19,22,25). To determine whether promoter nucleosome retention was a feature of many genes highly induced in response to LPS, we defined conditions for LPS-induction: We excluded genes that were already significantly expressed in rM (FPKM > 1 from (22)) and then further separated these genes into four groups (Fig. 4.3). Group A contained genes repressed in rM (x=0) and expressed significantly in aM (y>1). Group B contained genes with some low level of expression in rM (01). Groups C and D contained genes repressed or lowly expressed in rM respectively, but only lowly expressed in aM (x1! 0.1 22 TSS 0.1 % nuc. occ. B x=0; y>1! % nuc. occ. % nuc. occ. A TSS 3.0kb rM aM 0