The KDM5 family is required for activation of pro-proliferative cell cycle genes during adipocyte differentiation

Transcription

1 Syddansk Universitet The M family is required for activation of pro-proliferative cell cycle genes during adipocyte differentiation Brier, Ann-Sofie B; Loft, Anne; Madsen, Jesper Grud Skat; Nielsen, Thomas Rosengren; Nielsen, Ronni; Schmidt, Søren Fisker; Liu, Zongzhi; Yan, Qin; Groneyer, Hinrich; Mandrup, Susanne Published in: Nucleic Acids Research DOI:.9/nar/gkw Publication date: Docunt version Publisher's PDF, also known as Version of record Docunt license CC BY-NC Citation for pulished version (APA): Brier, A-S. B., Loft, A., Madsen, J. G. S., Nielsen, T. R., Nielsen, R., Schmidt, S. F.,... Mandrup, S. (). The M family is required for activation of pro-proliferative cell cycle genes during adipocyte differentiation. Nucleic Acids Research, (), General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirents associated with these rights. Users may download and print one copy of any publication from the public portal for the purpose of private study or research. You may not further distribute the material or use it for any profit-making activity or comrcial gain You may freely distribute the URL identifying the publication in the public portal? Take down policy If you believe that this docunt breaches copyright please contact us providing details, and we will remove access to the work imdiately and investigate your claim. Download date: 9. dec..

2 Published online November Nucleic Acids Research,, Vol., No. 9 doi:.9/nar/gkw The M family is required for activation of pro-proliferative cell cycle genes during adipocyte differentiation Ann-Sofie B. Brier,, Anne Loft,, Jesper G. S. Madsen,, Thomas Rosengren, Ronni Nielsen, Søren F. Schmidt, Zongzhi Liu,QinYan, Hinrich Groneyer and Susanne Mandrup, Departnt of Biochemistry and Molecular Biology, University of Southern Denmark, Odense M, Denmark, The Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen, Copenhagen N, Denmark, Departnt of Pathology, Yale School of Medicine, New Haven, CT, USA and Equipe Labellisée Ligue Contre le Cancer, Departnt of Functional Genomics and Cancer, Institut de Génétique et de Biologie Moléculaire et Cellulaire, Centre National de la Recherche Scientifique, UMR, Institut National de la Santé et de la Recherche Médicale, U9, Université de Strasbourg, Illkirch, France Received January, ; Revised November, ; Editorial Decision November, ; Accepted November, ABSTRACT The M family of histone dethylases removes the HK tri-thylation (HK) mark frequently found at promoter regions of actively transcribed genes and is therefore generally considered to contribute to corepression. In this study, we show that knockdown () of all expressed mbers of the M family in white and brown preadipocytes leads to deregulated gene expression and blocks differentiation to mature adipocytes. M leads to a considerable increase in HK at promoter regions; however, these changes in HK have a limited effect on gene expression per se. By contrast, geno-wide analyses demonstrate that MA is strongly enriched at M-activated promoters, which generally have high levels of HK and are associated with highly expressed genes. We show that M-activated genes include a large set of cell cycle regulators and that the Ms are necessary for mitotic clonal expansion in T-L cells, indicating that M may interfere with differentiation in part by impairing proliferation. Notably, the dethylase activity of MA is required for activation of at least a subset of pro-proliferative cell cycle genes. In conclusion, the M family acts as dual modulators of gene expression in preadipocytes and is required for early stage differentiation and activation of pro-proliferative cell cycle genes. INTRODUCTION Methylation of histone proteins constitutes important epigenomic marks involved in the dynamic regulation of the geno in response to external cues. So histone thylation marks are primarily associated with actively transcribed chromatin, whereas other marks are associated with repressed chromatin (). These marks may be modulated as a consequence of transcriptional activity, but they can also play an active role in modulating transcription. The formation and removal of these marks at histone residues is catalyzed by residue-specific histone thyltransferases and dethylases, respectively. The biological function of these has classically been tightly linked to their histone modifying catalytic activity; however, recent data indicate that both histone thyltransferases and dethylases may also modulate transcription independently of their histone modifying activities ( ). The histone lysine dethylases (M) are mbers of the family of Jumonji C (JmjC) domain-containing histone dethylases and specifically removes dithyl () and trithyl () marks from histone lysine (HK) (,). The M family is conserved among many species (), and in humans and other mammals it comprises four M paralogues, MA, MB, MC, and MD, which are very similar in structure (Supplentary Figure SA). Unlike MA and MB, MC To whom correspondence should be addressed. Susanne Mandrup, Departnt of Biochemistry and Molecular Biology, University of Southern Denmark, Campusvej, Odense M, Denmark. Tel: + ; Fax: + ; s.mandrup@bmb.sdu.dk These authors contributed equally to this work as first authors. Present addresses: Anne Loft and Søren F. Schmidt, Institute for Diabetes and Cancer, Helmholtz Zentrum München, Neuherberg, Germany C The Author(s). Published by Oxford University Press on behalf of Nucleic Acids Research. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( which permits non-comrcial re-use, distribution, and reproduction in any dium, provided the original work is properly cited. For comrcial re-use, please contact journals.permissions@oup.com

3 Nucleic Acids Research,, Vol., No. and MD are sex-chromoso-specific genes located on the X and Y chromoso, respectively (9,). Multiple studies of MA ( ) and MB ( ) have reported an involvent of these in cancer, and so studies of MC (9,) and MD (9) indicate that this could be a common feature of all M family mbers. Geno-wide mapping of binding sites of MA (,), MB (), and MC () have reported preferential binding to promoter regions, and a study of MD binding near the Engrailed gene () likewise demonstrated highest occupancy at the promoter region. More recently, so studies have reported that M family mbers may also regulate gene transcription from nonpromoter regions, i.e. putative enhancers (, ). The HK mark removed by the M family is found primarily at active promoter regions (,9,), whereas the HK mark is found at both active promoters () and enhancers (,). Both marks are considered activating (,), and consequently mbers of the M family have classically been regarded as repressors of transcription. Thus, a corepressor role through silencing of gene expression by dethylation of HK at promoters of genes has been described in diverse cellular processes such as cell cycle progression and cellular senescence (,,,,,),circadianrhythm(), and mitochondrial function (,). However, a potential coactivating function of the Ms has also been suggested by less defined chanisms such as through the interaction of MA with prb () or nuclear receptors (), by the ability of MB to prevent spreading of HK thylation into gene bodies (), and through binding of MC at enhancer regions where it has been suggested to maintain HK mono-thylation levels (). Furthermore, the M Drosophila ortholog Lid has also been implicated in transcriptional activation through a complex where dmyc masks the dethylase domain (9) through interaction with and inhibition of the deacetylase Rpd (), or by interacting with the transcription factor FOXO and preventing its ability to be recruited to promoters (). In a recent study of Drosophila knockout (KO) flies, Lid was furthermore found to be required for activation of gene expression of a set of mitochondrial genes independently of the dethylase activity but dependent on the PHD domain that recognizes HK/(). Thesereports indicatethat the Ms affect transcriptional activity by several different chanisms that may be independent of their dethylase activity. Histone dethylases have been shown to play an important role in cellular differentiation (). One of the most well studied differentiation processes is adipogenesis, i.e. the developnt of fibroblast-like preadipocytes to mature, lipid-containing adipocytes. Nurous studies over the past years, in particular studies using the murine T- L preadipocyte cell line, have carefully unraveled major transcriptional players in adipogenesis (,). Stimulation of adipocyte differentiation with a hormonal cocktail induces a cascade of transcriptional processes that is driven by at least two waves of transcription factors (TFs), the first of which includes CCAAT/enhancer-binding protein (C/EBP ) and the glucocorticoid receptor. These factors then induce the expression of the second wave of adipogenic TFs such as C/EBP and peroxiso proliferator-activated receptor (PPAR ), which switches on the adipocyte gene program. A major early event in the differentiation of T- L preadipocytes is mitotic clonal expansion, during which the cells undergo approximately two rounds of cell division. This clonal expansion is driven by C/EBP () and other first wave transcription factors and has been shown to be required for differentiation ( ). During T-L adipogenesis, histone thylation patterns are dramatically changed ( ). Several histone thyltransferases and histone dethylases have been reported to play a significant role in adipocyte differentiation (,). However, so far a role for the M family in adipocyte differentiation has not been described. In the present study, we demonstrate that the M family is required for activation of pro-proliferative cell cycle genes and differentiation of T-L as well as brown preadipocytes in response to adipogenic inducers. Using geno-wide approaches, we show that the Ms are required for balancing the levels of the activating mark HK at promoters in T-L cells. Interestingly however, the Ms appear to be exerting their main action on adipogenic gene expression as direct coactivators of a subset of promoters including promoters driving pro-proliferative cell cycle genes during adipogenesis. MATERIALS AND METHODS Cell culture T-L cells were grown in Dulbecco s modified Eagle s dium (DMEM) supplented with % calf serum. Cells were induced to differentiate two days post confluence (designated day ) in DMEM supplented with % fetal bovine serum (FBS), M dexathasone (dex),. mm - isobutyl--thylxanthine (MIX), and g/ml insulin essentially as described previously (). Mouse brown preadipocytes immortalized with the SV large T antigen were grown in DMEM supplented with % FBS. Cells were induced to differentiate at confluency (designated day ) in DMEM supplented with % FBS,. M dex,. mm MIX, nm insulin, nm T, and nm indothacin. From day, dex, MIX, and indothacin were omitted from the differentiation dium. At indicated ti points during differentiation a.mm -counting chamber was used to count the number of cells in control and samples. Oil red O staining was perford to evaluate accumulation of lipids in mature adipocytes. Cells were fixed with.% paraformaldehyde for hour and stained for hour with filtered Oil red O solution (. g Oil red O in ml isopropanol) diluted : in milliq. Cells were washed in milliq and microscopy photographs were taken at nm. Immortalized MA-/- mouse embryonic fibroblast (MEFs) with re-introduction of wild-type MA or MA-HA mutant were described previously (). Bromodeoxyuridine (BrdU) assay A BrdU assay (Millipore) was used to quantify the amount of newly synthesized DNA in control and cells in accordance with the manufacturer s instructions. Briefly, cells

4 Nucleic Acids Research,, Vol., No. were seeded in 9-well NUNC plates in five parallel replicates for each condition and : BrdU reagent was added to the dium for either h between day and h postinduction of differentiation or for hours between day and day of the differentiation. Following DNA denaturation, the amount of BrdU incorporation was assessed immunochemically as described in the manufacturer s protocol using a spectrophototer microplate reader at nm. RNA extraction, cdna synthesis, and quantitative real-ti PCR RNA extraction, cdna synthesis, and quantitative realti PCR were perford as previously described (). Data were normalized to transcription factor II B (TFIIB) and a two-tailed Student s t-test was used to determine significance. Sequences of real-ti PCR prirs can be found in Supplentary Table S. Western blotting and ECL detection Whole-cell extracts were prepared in SDS-containing buffer and subjected to western blotting as previously described (). The following antibodies were used: anti-ppar (E- ; sc-, Santa Cruz (SC) Biotechnology), anti-c/ebp (AA; sc-, SC Biotechn.), anti-c/ebp (C-9; sc-, SC Biotechn.), anti-ma (ab9, Abcam), anti- MB (ab9, Abcam), anti-mc (A-A, Bethyl lab.), HK (9, Cell Signaling Techn.), HK (, Abcam), H (; Cell Signaling Techn.), anti-rabbit IgG (P99, DAKO), and anti-mouse IgG (P, DAKO), anti-tfiib (sc-, SC Biotechn.) was included as a loading control. Lentiviral production and transduction of T-L cells For of MA, -B, and -C, shrna oligos targeting one or more of these factors were cloned into psicor PGK puro vectors (Addgene), and lentiviral particles were produced in human embryonic kidney (HEK) 9T cells as described in (). A control vector containing shrna targeting luciferase was cloned in parallel. Sequences of shrna oligos can be found in Supplentary Table S. Pre-confluent T-L and BAT-LgT cells were transduced days before induction of differentiation by incubation for h at C with dium containing lentiviral constructs and g/ml polybrene (Sigma). After h, the dium was changed to normal growth dium. Chromatin immunoprecipitation (ChIP) ChIP experints were perford in T-L preadipocytes (day ), and in T-L cells h and days post-induction of differentiation essentially as previously described (). For preparation of material for MA and MC ChIP-seq, cross-linking of the chromatin was perford with mm disuccinimidyl glutarate (DSG) (Proteochem) as described previously (). Chromatin prepared for HK ChIP-seq was cross-linked in % formaldehyde only. The antibodies used were anti-ma (ab9, Abcam), anti-mc (A-A, Bethyl lab.), and anti- HK (, Abcam). RNA-seq preparation and library construction Total RNA ( g) obtained from two independent biological experints was polya-selected and subsequently subjected to fragntation and cdna-synthesis using the TruSeq RNA Sample Prep kit v (Illumina). RNA-seq and ChIP-seq libraries were constructed according to the manufacturer s instructions (Illumina) as described () and sequenced on our in-house HiSeq platform. Data analysis Microarray analysis. The microarray data of MA- /- MEFs stably transduced with either empty vector (EV), MA WT or MA-HA mutant (MT) were used for the analysis. Raw profiling data were normalized with affy::rma and further processed using the Limma R package (9). A list of differentially expressed genes in cells overexpressing MA WT compared to control (EV) cells were called using a FDR <.. RNA-seq data analysis. Sequencing reads were aligned to the mouse reference geno (version mm9) using Bowtie (). Splice-junction reads were handled by creation of a pseudo-splice geno, similar to the strategy utilized in RSEQTools () and aligned using Bowtie with standard paraters. Reads were filtered post-alignnt for a MAPQ score greater than or equal to. The number of exon reads for all RefSeq genes was counted using Subread () and differential expression analyses were perford using DESeq (FDR <., paired analysis) (). Clustering of RNA-sequencing data was perford using Mfuzz (), only including genes with a mbership value >. for subsequent analysis. For both microarray and RNA-seq data, functional enrichnt analysis was perford with HOMER () using pathways related to tabolism from the KEGG database annotation (,) and WikiPathways (). ChIP-seq data analysis. ChIP-seq data sets were mapped to the mm9 geno with STAR (9) set to not map across splice junctions ( alignintronmax outsjfilterintronmaxvsreadn ). Tag directories for each ChIP-seq library were generated using HOMER (), where MA and MC ChIP-seq tag counts were normalized to M reads. Since HK protein levels are generally higher in M compared with control cells, we used a two-stage normalization strategy to correct for biases in the HK libraries and avoid underestimating the global increase in the HK levels in the normalization process. First, geno-scale normalization on background reads were done by counting tags in kb windows throughout the geno as described in () and normalized using a relative log expression (RLE) thod normalization (). Second, the signal in peaks was normalized for each condition over ti, but not across conditions. This two-stage normalization sche assus that the background is similar for all conditions (control and ) and ti points and that the signal for a particular condition is constant for most sites over ti, but can be different between conditions. Regions enriched for HK, MA, and MC

5 Nucleic Acids Research,, Vol., No. were called in each library with HOMER findpeaks using the -style histone and -F settings with all other paraters set at default. For HK, MA, and MC peak files, overlapping regions were rged into one master region file for each factor/mark. Subsequently, MA and HK peaks at each condition were identified as having -fold more average normalized tags over two replicates relative to the input control. Due to the lower quality and therefore higher background of the MC profiles, MC peaks were identified as having -fold more normalized tags in both MC replicates relative to the input control. For analyses of the HK libraries, high confidence promoter proximal regions were identifiedas being within ± kb of transcriptional start site (TSS) and having more than tags/kb in at least one of the conditions resulting in a rged region file consisting of, HK promoter proximal regions. MA and MC promoter proximal binding sites were defined as being within ± kb of TSS and distal binding sites between and kb away from TSS. Regions with significantly different levels of HK in compared to control at each ti point (i.e. day, h, day ) were identified using DESeq () with a fold change cut-off at.-fold averaged over the two replicate experints and a P-adjusted value <.. Clustering of the temporal profile of HK regions during differentiation was perford with Mfuzz, as described above (). HOMER was used for counting tags at identified regions, annotating binding sites to RefSeq genes, generating average profiles, and count matrices for heat maps as well as performing motif analyses in MA promoter binding sites. The University of California at Santa Cruz (UCSC) Geno Browser () was used for visualization of geno tracks. For analyses of promoter features, we retrieved, promoters containing TATA boxes from the Eukaryotic Promoter Database (EPDnew) (). We overlapped these promoter sequences with the respective gene groups and considered a gene as having a TATA box in the promoter if it was found within ± bp of the TSS of the RefSeq-annotated gene. We retrieved the position of, CpG islands from the UCSC geno browser () and considered a gene as having a CpG-island promoter when the TSS of the RefSeq-annotated gene overlapped a UCSC CpG island. A group of, RefSeq-annotated promoters were included as a control. Statistics. Two-tailed Student s t-test was used to determine significance in qpcr analyses. The P-value for the difference between data in box plots was computed using Wilcoxon rank sum test. Spearman s correlation coefficient was used to determine the linear correlation between two data sets. In the cluster analysis, 9% confidence interval around the dians was determined by bootstrapping and significance was calculated using a nonparatric Wilcoxon test. The P-value between data containing categorical variables was computed using Pearson s Chi-squared test and the two-proportion z-test. Additional data sets mcpg profiles from mature T-L adipocytes (day post-induction) () was obtained from NCBI Gene Expression Omnibus (GEO) (accession GSE). Data access The RNA-seq and ChIP-seq data sets generated in this study have been submitted to the GEO ( nlm.nih.gov/geo/) under accession number GSE. The raw microarray data is deposited in GEO under accession number GSE9. RESULTS of the M family inhibits adipogenesis In order to identify histone dethylases important for adipocyte differentiation, we constructed a lentiviral short hairpin RNA (shrna) library to individually a large number of histone dethylases in T-L preadipocytes (data not shown). Major advantages of the T-L model system are that the cells differentiate in a rather synchronous manner in response to a hormonal cocktail, and that differentiation does not require addition of PPAR agonists for differentiation (). This is important, since PPAR agonists are very powerful inducers of adipogenesis and therefore tend to override modest inhibitors of differentiation. Importantly, most of the key adipogenic regulators identified in this system have proven to hold up in vivo. The screen identified the M family among the top hits required for adipogenesis. Of the M family mbers only MA, -B, and -C are expressed in T- L preadipocytes. These subtypes are expressed at approximately equal levels and none of the family mbers display major regulation at the mrna level during adipocyte differentiation (Supplentary Figure SB). To further evaluate the role of the M family mbers in adipocyte differentiation, we knocked down MA, MB and MC in T-L preadipocytes four days prior to induction of differentiation (Figure A and B). of MB alone did not appear to result in any major effects on the differentiation as indicated by lipid accumulation (Supplentary Figure SC and D). However, the combined of MA and MC, and even more so of all three Ms, resulted in a marked inhibition of lipid accumulation compared with shluciferase control cells (Figure C and Supplentary Figure SC and D), thereby indicating redundancy between the M family mbers. Accordingly, at day of differentiation the expression levels of adipocyte characteristic genes such as fatty acid binding protein (Fabp), perilipin (Plin), perilipin (Plin), and lipoprotein lipase (Lpl) are significantly reduced in the MA/B/C (M ) cells, while genes characteristic of preadipocytes such as deltalike homolog (Drosophila) (Dlk) and thrombospondin (Thbs) are expressed to higher levels compared with control cells (Figure D). Interestingly, of the Ms does not seem to influence the initial induction of C/EBP (Figure E and F), which is part of the first wave of transcription factors induced during adipogenesis and required for

6 Nucleic Acids Research,, Vol., No. A Day - Lentiviral T-L transduction preadipocyte C Day Day Hormonal Cocktail Day Lipid-laden adipocytes Mature Adipocyte B Relative expression Day..... KdmA KdmB KdmC MA MB MC TFIIB Day D Relative expression F PPARγ C/EBPα C/EBPβ TFIIB Day Adipocyte markers Preadipocyte markers Fabp Plin Plin Lpl Relative expression Dlk Thbs Day.. p p LAP LAP E Relative expression Relative expression Relative expression 9 Cebpb D h h D D D D D Pparg D h h D D D D D Cebpa D h h D D D D D Ti point Figure. of M family mbers in T-L preadipocytes inhibits adipogenesis. Pre-confluent (day -) cells were transduced with lentivirus expressing either control luciferase shrna (control) or shrna targeting MA, MB, and MC (). Cells were induced to differentiate days post confluency (day ) using a standard differentiation cocktail. (A) Schematic representation of the experintal setup. (B) mrna (left) and protein (right) levels of MA, MB, and MC at day in control cells and M cells. (C) Representatives of phase contrast micrographs from Oil red O staining perford at day in control and M cells. (D) mrna levels of adipocyte marker genes (left), and preadipocyte marker genes (right) at day in control cells and M cells. mrna (E) and protein (F) levels of the key adipocyte transcription factors C/EBP, PPAR, and C/EBP during the course of adipocyte differentiation in control and M cells. Error bars represent standard deviation (SD), (n = ). P <., P <., and P <., Student s two-tailed t-test. mitotic clonal expansion (). However, induction of the key adipogenic second wave factors, PPAR and C/EBP, around day of differentiation is significantly impaired in the cells (Figure E and F). Importantly, around the sa ti point at this still early phase in the differentiation process, C/EBP levels decrease in compared with control cells (Figure E and F). Collectively, these data indicate that the M family is required for activation of the second wave of TFs and for proper progression of T-L adipocyte differentiation. The increase in HK promoter levels upon M has limited impact on gene expression The M family of histone dethylases is responsible for removing the activating histone mark HK primarily located at promoter regions of actively transcribed genes. We therefore hypothesized that the effect on adipocyte dif-

7 Nucleic Acids Research,, Vol., No. ferentiation by the M was the result of increased levels of HK at promoters leading to aberrant activation of genes. Using western blotting we confird that of the Ms leads to a global increase in HK that is maintained throughout adipogenesis, whereas it has modest effects on HK levels (Figure A). To be able to quantify and locate changes in HK deposition at the geno-wide level and correlate this with changes in gene expression, we perford HK ChIP-seq and RNAseq in preadipocytes (day ) and at early ti points ( h and days) during T-L differentiation. Importantly, there is a good concordance between the biological replicates for both HK ChIP-seq and RNA-seq data sets (Supplentary Figure S). In total, we identify, promoters with levels of HK significantly above background. In accordance with the total level of HK detected by western blot, the average HK level at promoters is significantly higher in M compared with control cells (Supplentary Figure SA). Using a cutoff of Padj<., >.-fold change, we identified, promoter sites with significantly increased levels of HK in preadipocytes compared with control cells, whereas only four promoters have reduced HK levels (Figure B). Similar patterns are observed at h and day of differentiation (Supplentary Figure SB). Interestingly, evaluation of the correlation between the increase in HK at promoters in relative to control cells and the increase in gene expression shows that there is only a weak correlation at all ti points (Figure C, Supplentary Figure SC). Accordingly, for the genes that are associated with increased HK signal in the cells there is a marginal average increase in gene expression relative to control cells, which is a remarkably small effect given the considerable increase in HK levels (Figure D). The majority of promoter sites gaining HK signal upon M are also marked by HK in the control cells (Figure E), and we therefore speculated that the degree to which promoters are already marked by HK could influence the transcriptional impact of an increase in HK. To evaluate this possibility, we grouped the preadipocyte promoters with gain in HK upon M into quartiles based on HK signal in control cells, with quartile having the lowest HK signal (Figure E). The largest fold increase in HK in compared with control cells is observed in quartile (Figure F), and the average fold increase in expression of associated genes is also slightly higher in quartile (Figure G). However, genes associated with quartile promoters are generally expressed at low levels (Supplentary Figure SD), and the correlation between fold increases in HK level and fold changes in gene expression is weak for all of the quartiles (Supplentary Figure SE). Examples of ChIP-seq profiles of HK promoter sites of the four quartiles as well as the gene expression of the associated genes are presented in Figure H. These analyses demonstrate a surprising degree of disconnection between the increased promoter-associated HK signal and induction of transcription by M in T-L cells. M leads to an increase in HK at dynamically regulated promoters The above analyses investigated the relationship between the increase in HK at promoters and gene expression in preadipocytes at early ti points of the differentiation process. In order to evaluate whether M disrupts the dynamic changes in the level of HK at a subset of promoters thereby resulting in aberrant gene expression of these promoters, we perford a temporal profile cluster analysis of the HK mark at promoters using Mfuzz (). First, we identified all promoter sites that exhibited differential levels of HK in control versus cells at one or more of the analyzed ti points (i.e. day, h and/or day ). Second, we clustered these promoters according to their changes in HK levels during the early ti points of the differentiation (Figure A). In accordance with the global increase in HK levels in the cells, all identified clusters have significantly higher levels of HK in compared with control cells (Figure B). However, as previously observed in the analysis of all promoters (Figure D), the substantial increase in HK levels upon at these dynamically regulated promoter regions appears to have very little effect on the expression of the associated genes (Figure C). Collectively, we find little evidence indicating that the considerable gain in promoter HK levels resulting from of the Ms is causally related to the changes in gene expression. Increased gene expression correlates with increased HK levels The above analyses indicate that an increase in promoter proximal HK per se is not a strong inducer of gene expression. We therefore addressed the reverse question, i.e. whether an increase in transcriptional activity upon M is reflected in an increase in HK at the corresponding promoters. To investigate this, we identified a set of genes that were differentially expressed (Padj <.) between and control cells at day (Figure A), or at h and days post-induction of differentiation (Supplentary Figure SA). A substantial number of dysregulated genes are detected at all ti points, with the largest number of dysregulated genes found at day (i.e. 9 regulated genes at day ; genes at h; and,9 genes at day ) (Figure A and Supplentary Figure SA). In line with this, clustering analyses based on temporal gene expression pattern (day to day ) in control cells (Supplentary Figure SB) reveal that the first wave of gene regulation initiated by the adipogenic stimuli at day is not greatly affected by the reduced M levels, whereas subsequent gene regulation at day is dysregulated upon M (Supplentary Figure SC). Thus, M leads to a profound dysregulation of gene expression at day of adipocyte differentiation. Notably, a similar number of genes are upregulated and downregulated in response to M at all ti points, supporting the notion that the Ms have both activating and repressive effects on gene expression as reported in previous studies (, ). For genes activated by the M, a strong correlation is observed between the

8 Nucleic Acids Research,, Vol., No. 9 HK promoter signal C HK vs. gene expression Gene expression Q.. -. Q Q Q Q HK signal Q HK signal H kb chr: Quartile chr:,,,, Tags/kb Quartile Kcnn,9, Quartile chr:,, Quartile chr:,,,, Plin -kb Input kb Kcnn,9, Mrgprf Quartile Hsdb,9, Hsdb Quartile Q Tags/kb Q Tags/kb HK signal Quartile.. HK promoter sites, day Quartile. Tags/kb Up All Up All HK promoter groups Log(/control) - E G HK signal..... Log(/control) Log(/control) Log(/control) F ρ=.9 HK signal, log(/control) - log(tags/kb) Gene expression.. HK signal.. D TFIIB Day Gene expression log(/control) log(tags/kb) Histone 9 HK HK B A Mrgprf Plin Figure. HK levels are globally increased at promoters upon M. (A). Western blot showing levels of total HK and HK protein during the course of differentiation in control and M cells with histone H and TFIIB included as controls. (B) Scatter plot showing the normalized log HK promoter signal (as tags per kb) (i.e. sites located within - kb to + kb of the transcription start site (TSS)) in control and M cells at day. HK sites with significantly higher tag count (Padj<., min.-fold change) in M relative to control (green) or control relative to (red) cells are shown. (C) Scatter plot showing the log fold change in HK promoter signal related to the log fold change in expression of the nearest genes in M relative to control cells at day at all identified HK sites of an expressed gene. HK promoter sites with significantly higher tag count (Padj<., min.-fold change) upon M are shown. There are no HK promoter sites with significantly lower tag count in relative to control cells of expressed genes. ρ indicates Spearman s correlation coefficient and the black line shows the linear regression between plotted values (green) of the HK sites with significantly higher tag count in cells. (D) Box plots showing log fold change in HK promoter signal (left) and gene expression (right) in relative to control cells at day promoter sites of expressed genes that significantly gain HK signal (green) or at all HK promoter sites (gray). (E) HK promoter sites were divided into four quartiles (Q Q) based upon HK signal in day control cells with Q having the weakest signal. Heat map showing input signal in control cells (left panel) and HK signal in control (middle panel) and M cells (right panel) at day in a ±-kb window around HK promoter sites that significantly gain HK upon M. (F) Box plot showing for day the log fold change in HK promoter signal in M relative to control cells in the four quartiles (Q Q) described in E. (G) Box plot showing for day the log fold change in gene expression in M relative to control cells in the four quartiles (Q Q) described in E. (H) ChIP-seq profiles of HK at representative sites from the four quartiles described in E at the Kcnn (Q), Hsdb (Q), Mrgprf (Q), and Plin (Q) loci (left). Gene expression signal (as tags per kb) of the associated genes in control (blue) and M (red) cells at day (right). D, F and G: P-value: P <., P <.e-, and P <.e-, Wilcoxon rank sum test as indicated or F) between individual quartiles. Fold change of relative to control is included above each box.

9 Nucleic Acids Research,, Vol., No. A HK signal changes B Median normalized tags C Cluster 9 promoter sites Cluster 9promoter sites Cluster promoter sites Cluster 9 promoter sites Cluster Cluster promoter sites promoter sites D h Ti D D h Ti D D h Ti D D h Ti D D h Ti D D h Ti D Cluster 9 promoter sites Cluster 9 promoter sites Cluster promoter sites Cluster 9 promoter sites Cluster Cluster promoter sites promoter sites D h D D h D D h D D h D D h D D h D Ti Ti Ti Ti Ti Ti Median normalized tags D Cluster genes h Ti D D Cluster genes h Ti D D Cluster 99 genes h Ti D D Cluster genes h Ti D D Cluster genes h Ti D D Cluster genes Figure. Cluster analysis of temporal HK occupancy at promoter sites. (A) All promoter sites with differential levels of HK in control relative to M cells at one or more of the analyzed ti points were clustered based on the temporal profile of HK during the early stage of adipocyte differentiation (day to ) in control cells. (B) HK promoter signal in the six clusters shown in A in control (blue) and M (red) cells at day. The colored region illustrates the 9% confidence interval around the dian (dashed line) as determined by bootstrapping. (C) Expression signal based on RNA-seq of genes assigned to the promoters from the six different clusters shown in A in control (blue) and M (red) cells at day. The colored region is the 9% confidence interval around the dian (dashed line) as determined by bootstrapping. The number of genes in C differs from the number of promoter sites in A-B due to a subset of the cluster promoter sites being associated with genes not expressed at detectable levels in T-L cells, which were therefore excluded from the analysis. B and C: P-value: P <., P <.e-, and P <.e-, control compared with M, Wilcoxon rank sum test. h Ti D increase in gene expression and the increase in HK at the promoter (Figure B and Supplentary Figure SD). In line with this, the average gain in HK at promoters of genes that are upregulated upon M is significantly higher than the gain in HK at promoters of downregulated and all analyzed genes (Figure C, right). Notably, genes downregulated upon M do not lose HK signal compared with control cells (Figure C, right), consistent with the notion that the HK mark is not a main determinant of gene expression. Examples of ChIP-seq profiles of HK promoter sites and the expression of associated upregulated and downregulated genes are shown in Figure D. Taken together, these results indicate that rather than promoting gene expression, the HK mark is deposited at promoter regions mainly as a result of active transcription. MA binding is strongest near M-activated genes To explore the chanisms by which the Ms are involved in gene regulation during the early stages of T-L adipogenesis, we perford MA ChIP-seq in replicates in T-L cells at different ti points (day, h and day ) of the differentiation. As previously shown (,,), a large proportion of MA binding sites (.%) are located in the promoter region (TSS ± kb) (Supplentary Figure SA), and these correlate very strongly between replicates (Supplentary Figure SB). Furthermore, MA promoter binding sites display much higher binding intensity than other genomic regions (Supplentary Figure SC). A comparable binding pattern is observed for MC (Supplentary Figure SA and SC), and there is strong correlation between the binding strength of MA and MC at their target promoters (Supplentary Figure SD), thereby indicating

10 Nucleic Acids Research,, Vol., No. A log(tags/kb) Gene expression C Log(/Luc) - Gene expression Log(/Luc) - HK signal B HK signal log(/control) - - log(tags/kb) ρ=-. p=. ρ=. p<.e- - - Gene expression, log(/control) D Up Down Up Down All Gene groups chr: chr: kb,,,, Bgalnt,,,, Exosc9 Up Down All Gene groups tags/kb tags/kb Bgalnt Exosc9 Figure. Correlation between gene expression and HK signal. (A) Scatter plot showing the normalized log gene expression signal (as tags per kb) in M relative to control cells at day. Genes significantly (Padj <.) increased (green) or reduced (red) upon M are highlighted. (B) Scatter plot showing the log fold change in gene expression signal and the log fold change in the associated HK promoter signal in M relative to control cells at day at genes that are repressed (red) or induced (green) by M. ρ indicates Spearman s correlation coefficient and the black line shows the linear regression between plotted values. The P-value of the Spearman s correlation is indicated. (C) Box plot showing the log fold change in gene expression signal (left) and the log fold change of the associated HK promoter signal (right) in M relative to control cells at day at all identified genes (grey), as well as genes that are repressed (red) or induced (green) by M. P-value: P <.e-, Wilcoxon rank sum test. (D) ChIP-seq profiles of HK at the Bgalnt and Exosc9 loci and gene expression signal (as tags per kb) (right) in control (blue) and M (red) cells at day. that these two factors display a similar mode of binding in T-L cells. Interestingly, genes with strong promoter binding of MA in preadipocytes tend to be induced and gain HK during differentiation (Figure A and B, Supplentary Figure SE), and this gene induction is impaired in the M cells (Figure C). This indicates that high MA promoter binding at day pris promoters for activation during early adipogenesis. In contrast, genes with low MA binding in preadipocytes are predominantly repressed during differentiation (Figure B), and this repression is M-dependent (Figure C). Thus, MA is associated with both induced and repressed promoters during T-L adipogenesis; however, promoters marked with low MA binding in preadipocytes are generally repressed in a M-dependent manner during differentiation, whereas promoters with high MA binding are generally activated in a M-dependent manner during differentiation. To further explore the regulatory role of M in adipocyte differentiation, we defined a set of M- activated (i.e. bound by MA and downregulated by M ) and M-repressed (i.e. bound by MA and upregulated by M ) genes. MA binding is significantly stronger at the promoter of these regulated genes compared to regions distal to these genes (Figure D). Furthermore, only minor differences in MA binding near M-activated and repressed genes are observed at these distal regions, supporting the notion that MA mainly exerts its effect through the promoter. In accordance with the analyses above, MA binds significantly less to promoters of M-repressed genes compared with M-activated promoters and promoters of all MA target genes (Figure D), and these M- repressed genes are expressed at a lower level and have lower levels of HK at their promoters (Supplentary Figure SF and G). These analyses further indicate that MA functions as a repressor at so promoters, which are generally low occupancy sites, whereas it acts as an activator of a subset of high MA occupancy promoters. Similar promoter-enriched binding to M-activated genes is observed for MC thereby further supporting a similar mode of action for MA and MC in T- L cells (Supplentary Figure SH). Further characterization of M-activated and M-repressed promoters at day revealed that M- activated promoters on average are less associated with a TATA box and are more associated with CpG islands (Figure E). Furthermore, comparison with DNA thylation profiles from a previous study () showed that M- activated promoters generally are depleted of mcpg signal in T-L adipocytes compared to M-repressed and all target sites (Figure F). Interestingly, motif analyses revealed that the EF motif is specifically enriched at

11 Nucleic Acids Research,, Vol., No. A MA log(tags/kb) D Log(tags/kb) MA binding at day Input Q Q Q Q MA binding at day MA binding at promoter and distal sites D h D Prom. Distal M-repressed B of signi cantly regulated genes from day to day Prom. Distal M-activated MA target gene expression Q Q Q Q MA binding at day Prom. Distal All target genes Up day Down day E fraction of promoters C Gene expression log (day /day ) MA target gene expression Q Q Q Day MA target promoters CpG island TATA-box Q MA binding at day M-repressed M-activated All target genes RefSeq promoters M-repressed M-activated All target genes RefSeq promoters F. mcpg at MA target promoters G Day MA target promoters Elk (ETS) motif % EF motif % Enrichnt Bp from TSS M-repressed M-activated All target genes % sites % % % % M-repressed M-activated All target genes % sites % % % % M-repressed M-activated All target genes MA Background Figure. MA binds strongly at promoters of M-activated genes. (A) MA promoter sites were divided into four quartiles (Q-Q) based on MA signal in T-L preadipocytes with Q having the weakest signal. (B) Number of genes that are significantly induced (black) or repressed (gray) from day to day during differentiation for genes with MA promoter sites in different quartiles (Q Q, defined in A) in control cells. P-value: P <., P-value: P <.e- and P <.e-. Two proportion z-test between induced and repressed genes in each quartile. (C) Box plot showing log fold change in gene expression signal (as tags per kb) from day to day of differentiation for genes with MA promoter sites in different quartiles (Q Q) in control (blue) and M (red) cells. D. Box plot showing log normalized MA signal (as tags per kb) in promoter sites (±kbfrom TSS) and distal sites ( kb from TSS) of M-repressed (green), M-activated (red), and all M target (gray) genes at day (left), h (middle) and day (right) post induction of T-L differentiation. (E) Fraction of promoter sites for each indicated group of MA-associated day promoters and annotated RefSeq promoters that harbor a CpG island (left) or TATA-box (right). P-value: P <., P <.e-, and P <.e-, Pearson s Chi-squared test. (F) Aggregate plot showing the average mcpg signal (as tags per bp per peak) at day post-induction of T-L differentiation in MA day target promoters of M-repressed (green), M-activated (red), and all M target (gray) genes. (G) The percentage of MA day promoter sites (orange) or matched genomic background sequences (gray) with the indicated motifs near M-repressed, M-activated, and all M target genes. FDR<. and FDR<.e- between MA promoter groups and background. A, C and D: P-value: P <., P <.e-, and P <.e-, Wilcoxon rank sum test, as indicated or A) between individual quartiles and D) between all groups in promoter and distal regions.

12 Nucleic Acids Research,, Vol., No. M-activated promoters at day, whereas the ETS motif is enriched at M-activated promoters and a control group of all MA-occupied promoters (Figure G, Supplentary Table S), suggesting that mbers of the M family and EF cooperate in the activation of these genes. Taken together these results indicate that the ability of M to repress or activate promoters may depend on the promoter context. The Ms are required for activation of cell cycle genes and proliferation To characterize the M target genes that are dysregulated on day in the M cells, we perford functional enrichnt analyses using KEGG and WikiPathways on M-activated, -repressed, and unchanged target genes in day cells compared with control. Interestingly, among the top five most significantly enriched pathways of the M-activated target genes are multiple cell cycle-related terms in both the KEGG and WikiPathways analyses (Figure A, Supplentary Figure SA). Genes in this group include cell cycle genes that are involved in stimulation of progression through the cell cycle such as cell division cycle (Cdc) andcdc (Figure B), as well as genes encoding important cell cycle regulators such as EF transcription factor and (Ef and Ef) (Supplentary Figure SB). These genes display transient expression profiles that peak around day during differentiation in control cells (Figure B), which is consistent withmitotic clonal expansion during the early stage of T-L differentiation ( ). Notably, M impairs the induction of cell cycle genes normally upregulated during early differentiation (Figure C, Supplentary Figure SB), whereas M does not significantly affect cell cycle genes that are repressed or unchanged between day and day (Supplentary Figure SC). Importantly, and in accordance with other M-activated genes in the above analysis (Figure D), the M-activated cell cycle genes also display enrichnt of MA and MC binding (Figure D-E and Supplentary Figure SD). Upon M, binding of MA as well as MC to the promoters of a subset of pro-proliferative cell cycle genes is significantly reduced (Figure F), indicating that both M family mbers play a direct role in the induction of these pro-proliferative cell cycle genes. Collectively, these results show that M prevents the full induction of the cell cycle gene program in response to the adipogenic cocktail and indicates that M family mbers directly coactivate these genes. Consistent with a role of M in activation of cell cycle genes, mitotic clonal expansion between day and day of differentiation (Figure A) as well as DNA synthesis (Figure B) is significantly impaired in M cells compared with control. Notably, of M at day of the differentiation, i.e. after activation of the second wave of major adipogenic transcription factors and after the post-confluent mitoses, does not interfere with adipogenesis (Supplentary Figure SE-H). Collectively, these results shows that of M interferes with early stage T-L adipogenesis, and indicates that it does so at least in part by interfering with activation of cell cycle genes. To investigate whether M plays a similar role in the differentiation of other adipocyte model systems, we knocked M down in BAT-LgT cells, which is a brown preadipocyte cell line immortalized by SV large-t antigen (Supplentary Figure SA). Interestingly and consistent with the results from T-L preadipocytes, M in brown preadipocytes also affects the expression of a subset of cell cycle genes (Supplentary Figure SB) and results in impaired proliferation during differentiation (Figure C). Furthermore, M results in reduction in lipid accumulation (Figure D and Supplentary Figure SC) as well as expression of adipocyte marker genes and UCP (FigureE). Taken together, the M family is important for post-confluent proliferation and adipogenesis of immortalized brown preadipocytes as well as T-L white preadipocytes, suggesting that these are general effects of M on ex vivo adipocyte differentiation. Both gene activation and repression is dependent on MA enzymatic activity Our results indicate that the Ms primarily promote adipogenesis by coactivation of specific promoters including promoters of cell cycle genes, whereas the general effect on reducing HK at promoters seems to be less important. To investigate whether the dethylase enzymatic activity of the Ms is required for the regulation of cell cycle genes, we used microarray data from MA KO mouse embryonic fibroblasts (MEFs) stably expressing either wild type or an enzymatically inactive mutant of MA. Similar to the results from differentiating T- L adipocytes, functional enrichnt analyses revealed that gene groups related to the cell cycle and cancer pathways are significantly enriched among the MA-activated target genes in MA KO MEFs (Supplentary Figure S). The major fraction of these MA-activated cell cycle genes have previously been reported to be pro-proliferative and/or have been shown to be overexpressed in various cancer forms ( ). Interestingly, we find that the enzymatic activity is required for both activation and repression of genes by MA (Figure A). Importantly, with the exception of the proto-oncogenes Mdm and Ccnd, there is a requirent of the dethylase activity for the activation of pro-proliferative cell cycle genes by MA, suggesting that the enzymatic activity of MA is required for the ability to induce the majority of MA-activated cell cycle genes in fibroblasts (Figure B and C). DISCUSSION In this study, we demonstrate that of the M family impairs proliferation and differentiation in white and brown preadipocytes. We show that of the M histone dethylases in T-L preadipocytes results in a global increase in HK levels at promoters but that this has little impact on gene expression per se. In contrast, ourdata indicate that M can act as an activator of a subset of cell cycle genes, and that of M abrogates the induction of these genes in response to the adipogenic cocktail (Figure D). Our approach to simultaneously all expressed M family mbers, i.e. MA, MB and