IRF2BP2 Modulates the Crosstalk between Glucocorticoid and TNF Signaling

Kokoteksti

(1)UEF//eRepository DSpace Rinnakkaistallenteet. https://erepo.uef.fi Terveystieteiden tiedekunta. 2019. IRF2BP2 Modulates the Crosstalk between Glucocorticoid and TNF Signaling Kaiser Manjur, ABM Elsevier BV Tieteelliset aikakauslehtiartikkelit © Elsevier Ltd CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ http://dx.doi.org/10.1016/j.jsbmb.2019.105382 https://erepo.uef.fi/handle/123456789/7658 Downloaded from University of Eastern Finland's eRepository.

(2) Accepted Manuscript Title: IRF2BP2 Modulates the Crosstalk between Glucocorticoid and TNF Signaling Authors: A.B.M. Kaiser Manjur, Joanna K. Lempiäinen, Marjo Malinen, Jorma J. Palvimo, Einari A. Niskanen PII: DOI: Article Number:. S0960-0760(19)30295-X https://doi.org/10.1016/j.jsbmb.2019.105382 105382. Reference:. SBMB 105382. To appear in:. Journal of Steroid Biochemistry & Molecular Biology. Please cite this article as: Kaiser Manjur ABM, Lempiäinen JK, Malinen M, Palvimo JJ, Niskanen EA, IRF2BP2 Modulates the Crosstalk between Glucocorticoid and TNF Signaling, Journal of Steroid Biochemistry and Molecular Biology (2019), https://doi.org/10.1016/j.jsbmb.2019.105382 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain..

(3) IRF2BP2 Modulates the Crosstalk between Glucocorticoid and TNF Signaling. A.B.M. Kaiser Manjur1, Joanna K. Lempiäinen1, Marjo Malinen1,2, Jorma J. Palvimo1,*, and Einari A. Niskanen1,*. Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland.. 2. IP T. 1. Department of Environmental and Biological Sciences, University of Eastern Finland, Joensuu, Finland.. SC R. *Address correspondence to Email: jorma.palvimo@uef.fi or einari.niskanen@uef.fi Postal address:. M. A. N. U. Einari Niskanen, PhD University researcher Institute of Biomedicine Room 3195 (Snellmania) University of Eastern Finland P.O.Box 1627 (Yliopistonranta 1E) FI-70211 Kuopio Finland. ED. Abstract. IRF2BP2 (interferon regulatory factor-2 binding protein-2) is an uncharacterized interaction partner of. PT. glucocorticoid (GC) receptor (GR), an anti-inflammatory and metabolic transcription factor. Here, we show that GC changes the chromatin binding of IRF2BP2 in natural chromatin milieu. The GC-induced. CC E. IRF2BP2-binding sites co-occur with GR binding sites and are associated with GC-induced genes. Moreover, the depletion of IRF2BP2 modulates transcription of GC-regulated genes, represses cell proliferation and increases cell movement in HEK293 cells. In A549 cells, the depletion extensively. A. alters the responses to GC and tumor necrosis factor  (TNF), including metabolic and inflammatory pathways. Taken together, our data supports the role of IRF2BP2 as a coregulator of both GR and NFB, potentially modulating the crosstalk between GC and TNF signaling.. 1.

(4) Key Words: Glucocorticoid receptor (GR), Interferon regulatory factor-2 binding protein-2 (IRF2BP2), Nuclear factor-B (NF-B), Dexamethasone, Tumor necrosis factor α (TNF). Introduction Glucocorticoids (GCs) are among the most widely prescribed drugs due to their anti-inflammatory and immunosuppressive effects1,2. The actions of both natural and synthetic GCs are mediated through. IP T. glucocorticoid receptor (GR)3, a hormone-controlled transcription factor (TF), belonging to the nuclear receptor superfamily4. The GR is ubiquitously expressed in the human body and it exerts pleiotropic. SC R. effects on fundamental biological processes, for example, on cell proliferation, development,. inflammation and immune responses5. Upon GC induction, GR translocates to the nucleus where it. U. binds to GC response element (GRE)-containing enhancers and, together with co-regulators and other. N. TFs, regulates gene transcription6.. A. The anti-inflammatory effects of GCs result largely from GR’s ability to repress genes. M. activated by pro-inflammatory TFs, including nuclear factor-B (NF-B)7-9. NF-B is a family of TFs that are homo- or heterodimers of five structurally related proteins: p65 (aka RELA), RELB, C-REL, p52/p100. ED. or p50/p10510. The heterodimer of p50 and p65 is the predominant NF-B complex in most cell types11. Variety of pro-inflammatory cytokines, including tumor necrosis factor α (TNF), rapidly. PT. activate NF-B, leading to its translocation to the nucleus where it regulates the transcription of a. CC E. plethora of inflammatory genes.. Transcriptional regulation by GR and other TFs requires coregulator proteins, which. modulate TF’s actions usually as a part of multisubunit complexes12. However, the functional roles of. A. any specific coregulator proteins in the modulation of GR function are still poorly defined. We have recently identified interferon regulatory factor-2 binding protein-2 (IRF2BP2) as a putative coregulator of GR13. IRF2BP2 was originally identified as a transcriptional repressor of interferon regulatory factor 2 (IRF2)14 and has been shown to repress the transactivation by TFs p5315 and LDB-116. IRF2BP2 belongs to a family of three evolutionary conserved proteins: IRF2BP1, IRF2BP2 (two splicing isoforms, A and 2.

(5) B) and IRF2BPL. IRF2BP2 (in this study IRF2BP2 refers to isoform IRF2BP2A) has conserved N-terminal zinc finger and C-terminal RING finger domains. Mechanistically, IRF2BP2 forms homo or heterodimers with other members of the family and it has been proposed to interact with NCOR1 complex to repress gene expression16,17. IRF2BP2 is important for the regulation of inflammatory responses in various cell types. For instance, IRF2BP2 deficient microglia express increased amounts of inflammatory cytokines. IP T. in response to lipopolysaccharide challenge, while showing impaired activation of anti-inflammatory markers in response to interleukin-4 stimulation18. Furthermore, IRF2BP2 regulates lipid metabolism. SC R. by inducing the expression of anti-inflammatory TF Krüppel-like factor 2 in mouse macrophages,. which, besides the inflammatory signaling, regulates the cholesterol homeostasis thus reducing the susceptibility to atherosclerosis19. Moreover, IRF2BP2 regulates anti-apoptotic signaling. The. U. depletion of the IRF2BP2 increases the apoptosis induced by actinomycin D and doxorubicin in U2OS. N. cells15, and apoptosis induced by ITGB3BP (aka NRIF2) and a protein kinase A inhibitor H89 in breast. A. cancer cells20.. M. Here, we analyzed the role of IRF2BP2 in the regulation of GC and TNF signaling. First,. ED. the reporter assays showed that IRF2BP2 is a potent coactivator of GR and NFB. Next, the ChIP-seq analysis in HEK293 cells revealed that, the GC-induced chromatin binding sites of IRF2BP2 co-occur. PT. with GR binding sites and associate with GC-induced genes. In agreement with the notion that IRF2BP2 is a part of the GR transcription coregulatory complex at GR-bound enhancers, the silencing of IRF2BP2. CC E. modulated transcription of approximately half of GC-responsive genes and lead to a reduction of cell proliferation and an increase in cell migration of HEK293 cells. In A549 cells, the silencing altered the. A. transcriptional responses to GC and TNF, leading to attenuation of inflammatory pathways. Collectively, these results show that IRF2BP2 can act as a modulator of GC and TNF signaling, and suggest that it has a role in regulation of the crosstalk between GR and NF-B in inflammatory signaling.. 3.

(6) Materials and Methods Cell culture, transfection and treatments Isogenic HEK293 cells that constitutively express GR (HEK293-GR)21, African Green monkey kidney (COS-1, ATCC) cells and Adenocarcinoma human alveolar basal epithelial cells (A549, ATCC) were maintained in Dulbecco’s modified Eagle’s medium (Gibco® Invitrogen) supplemented with 10% (v/v). IP T. fetal bovine serum (FBS), 25 U/ml penicillin and 25 µg/ml streptomycin. In addition, 100 µg/ml. hygromycin-B was used for HEK293-GR cells. For transfection, medium was changed to steroid-. SC R. depleted medium (DMEM containing 2.5% (v/v) charcoal-stripped FBS). Transfection was performed. using TransIT ®-LT1 (Mirus Bio LLC) reagent for reporter gene assay and Lipofectamine RNAiMAX. U. transfection reagent (Invitrogen) for silencing experiments. For treatments, 100 nM dexamethasone. N. (dex), 10 ng/ml Tumor Necrosis Factor-α (TNF), or vehicle (veh, equal amounts of ethanol) were used.. A. Plasmids. M. cDNA clone of IRF2BP2 was transferred to the destination vector pDEST-C1-FLAG-GFP11-GW (a kind gift from Dr. Maria Vartiainen, University of Helsinki) using gateway cloning (Invitrogen). C4 zinc finger. ED. (mZnF, C37S, C40S) and RING finger (mRING, C506S, C509S) mutants of IRF2BP2 were constructed by changing two cysteine codons to serine codons by site-directed mutagenesis (Quick Change II Site-. PT. Directed Mutagenesis Kit, Agilent). IRF2BP2 mutants devoid of the N-terminal zinc finger domain. CC E. (ZnF, amino acid residues 1-324 deleted) or RING finger (RING, amino acid residues 372-587 deleted) and cloned to the same expression vector as the full length IRF2BP2. All the mutants were verified by DNA sequencing. The primer sequences used to construct the mutants are available upon. A. request.. RNAi, RNA isolation and real-time quantitative PCR (RT-qPCR) For silencing experiments, cells were seeded onto 6-well plates using regular growth medium. After 24 h, medium was changed to steroid-depleted medium and cells were transfected with 20 nM siRNAs (Dharmacon, ON-TARGETplus SMARTpool) against IRF2BP2 (L-007177-02-0005), or control siRNA. 4.

(7) (Dharmacon, non-targeting pool) using Lipofectamine RNAiMAX transfection reagent (Invitrogen). Cells were subsequently exposed to veh, dex, TNF or both dex and TNF (DT) for 6 h before harvesting. Total RNA was extracted from four biological replicates using TriPure isolation reagent (Roche) and converted to cDNA using Transcriptor First strand cDNA synthesis Kit (Roche). RT-qPCR and fold change calculations were done as previously described21 using glyceraldehyde-3-phosphate. IP T. dehydrogenase (GAPDH) mRNA levels for normalization. RT-qPCR primers sequences are available upon request.. SC R. Antibodies. We used following polyclonal rabbit antibodies for immunoblotting and ChIP-seq: anti-IRF2BP2 (Bethyl laboratories, A303-190A), anti-GR (Santa Cruz Biotechnology, sc-1003), anti-p65 (Santa Cruz. U. Biotechnology, sc-372), anti-GAPDH (Santa Cruz Biotechnology, sc-25778). The appropriate secondary. N. antibodies were from Invitrogen and chemiluminescence reagents were from Pierce and Thermo. M. A. Fisher Scientific.. Reporter gene assay. ED. Reporter gene assays were done similarly as described before13,22,23 with the following modifications. COS-1 cells were seeded on 12-well plates. After 24 h, the medium was changed to steroid-depleted. PT. medium 4 h before the transfection. For GR-IRF2BP2 study, cells were transfected with p(GRE)4-tkLUC24,25 (200 ng) along with expression vectors encoding GR (20 ng) and increasing amounts (1, 5, 25. CC E. or 100 ng) of full length IRF2BP2 or the mutants. Beta-galactosidase expressing pCMVβ (20 ng, Clontech) was cotransfected as an internal control for transfection efficacy. Empty vector, pFLAG-. A. CMV-2 (Sigma-Aldrich), was used to balance the total amount of DNA in each transfection reaction. After 24 h of transfection, cells were exposed to dex or vehicle for 16 h. For p65-IRF2BP2 studies, cells were transfected with pĸB6-tk-LUC reporter26 (200 ng) along with expression vectors encoding p65 (20 ng) and increasing amounts (1, 5, 25 or 100 ng) of full-length IRF2BP2 or the mutants. All the experiments were performed in triplicates. Passive lysis buffer (Promega) was used to lyse the cells. Luciferase and β-galactosidase activities were measured as described before22 and the luciferase 5.

(8) activity was calculated by dividing the background-corrected luciferase values by the corresponding β-galactosidase values.. ChIP-seq and ChIP-qPCR ChIP-seq was performed according to a published protocol21 with following modifications and two. IP T. biological replicates. Briefly, the HEK293-GR cells were grown on 10-cm dishes in charcoal-stripped medium for 2 days, treated with vehicle or dex (100 nM, 1 h), and fixed with 1% (v/v) formaldehyde. SC R. 10 min at room temperature. Chromatin was fragmented to ∼200–400 bp using sonication (Bioruptor UCD-300, Diagenode). Antibodies were coupled to protein-A- beads (Millipore), fragmented. chromatin was incubated with antibody-coupled beads overnight, washed, eluted and de-crosslinked. U. in the presence of proteinase K (Fermentas). Chromatin fragments were purified using MiniElute. N. columns (Qiagen), ChIP-seq libraries were prepared using NEBNext kit (New England Biolabs) and. A. sequenced with HiSeq2000 at EMBL GeneCore (Heidelberg, Germany). Fragmented de-crosslinked. M. chromatin was used as an input control. Previously published ChIP-seq data for GR binding in HEK293-. ED. GR cells was used for overlap analysis of IRF2BP2 chromatin-binding21. Publically available ChIP-seq data of histone marks H3K27ac and H3K9me3 from HEK293 cells were used in heatmap and line profile. PT. analysis27. Sequenced raw reads were quality controlled using FastQC and quality filtered using FASTXtoolkit as described28 before reads were mapped to human genome (hg19) using bowtie and keeping. CC E. only uniquely mapping reads. Initial binding sites were defined for both biological replicates against input control using findPeaks program of HOMER package29 using options: -F 3, -tagTreshold 20, minDist 350. Binding sites found in both biological replicates were considered representative for the. A. given condition and used in the analysis. Signal matrixes for heat maps and line profiles were done in HOMER and visualized using imageJ30 and R (www.R-project.org). DNA motif discovery was performed with findMotifsGenome tool of the HOMER package. ChIP-qPCR samples were prepared from A549 cells using the ChIP-seq sample preparation protocol as described above. To find overlapping binding sites, the HEK293-GR IRF2BP2 ChIP-seq data were 6.

(9) mapped to hg38 and compared with similarly processed A549 data sets of GR in dex treatment27 (GSE91327) and p65 in TNF treatment31 (GSE34329). Figures of chromatin ChIP-seq signals were produced using Integrative Genomics Viewer32. qPCR analysis was carried out with LightCycler 480 SYBR Green I Master (Roche Diagnostics Gmbh). Results were calculated using the formula 2-DCt x E, where DCt = Ct(output) - Ct(intput) and E = coefficiency factor. The ChIP-qPCR primers are available on. IP T. request.. SC R. RNA-seq. Silencing of IRF2BP2 in HEK293-GR cells and in A549 cells was performed as described above. In case of HEK293–GR cells, they were exposed to dex or vehicle for 6 h before harvesting. A549 cells were. U. treated with vehicle, dex, TNF or co-treated with dex and TNF (DT) for 6 h. Total RNAs were extracted. N. using TriPure reagent (Roche) and RNeasy kit (Qiagen). RNA-seq libraries were prepared from three. A. biological replicates per condition using NEBNext Poly(A) mRNA Magnetic isolation module (New. M. England Biolabs) and sequenced with Illumina HiSeq2000 at EMBL GeneCore (Heidelberg, Germany).. ED. Sequencing reads were trimmed using Trimmomatic33 (minQual5 = minQual3 = 5, minLen = 36), polyA tails were removed using homerTools of HOMER package29, contaminants were removed by. PT. mapping reads against the hg19 abundant sequences, and reads were mapped against hg19 transcriptome and genome using tophat234. Read counts were analyzed at transcript level using. CC E. HOMER package and edgeR35, and differential expression was called using following criteria: RPKM (reads per kilobase per million mapped reads) > 0.5 for A549 and >5 for HEK293-GR, FDR (false discovery rate) < 0.01 and absolute value of log2FC (fold change) > 0.5 for HEK293-GR or > 1 for A549. A. cells. Pathway enrichment analysis of differentially expressed genes in RNA-seq data was done using Ingenuity pathway analysis (IPA) tool (October 2018 release). Effect of IRF2BP2 silencing was analyzed by comparing siNON and siIRF2BP2 samples within each treatment (veh, dex, TNF, DT). Effect of treatments (dex, TNF, DT) was analyzed by comparing treatment induced expression changes to. 7.

(10) vehicle in siNON and siIRF2BP2. Hierarchical clustering of canonical pathways was done in IPA using cut-offs of Z-score > 2 or < -2 and p-value < 0.05.. Cell proliferation and migration assays. IP T. For real time cell proliferation, HEK293-GR cells were seeded onto 96 well plates. IRF2BP2 was silenced by reverse transfection in four biological replicates. After 24 h, the cells were treated with 100nM dex. or vehicle (equal amounts of EtOH). The cell confluence was monitored for 72 h after the treatment. SC R. by live cell imaging (IncyCyte S3, Sartorius). Migration assay was done following the IncuCyte Scratch Wound Assay protocol. Briefly, cells were seeded onto 96 well image lock plates (Essen Bioscience). U. and IRF2BP2 was silenced by reverse transfection in four biological replicates. After 24 h, when the. N. cell mono-layer was near to 100% confluence, the plate was removed from the incubator and wound. A. was created using the Wound Maker (Sartorius). After wounding, the media was aspirated and. M. replaced with fresh media containing the treatment (100 nM dex or equal amount vehicle). Subsequently, the plate was placed into the IncuCyte live-cell analysis system (IncyCyte S3, Sartorius). ED. and was monitored for 36 h. Cell confluence and migration was determined by automatic counting. CC E. Results. PT. using IncuCyte S3 with cell proliferation and scratch wound cell migration software modules.. IRF2BP2 is a potent coactivator of GR- and NF-B-mediated transcription in vitro To study whether IRF2BP2 shows coregulator, coactivator or corepressor activity on the GR mediated. A. transactivation, we used luciferase reporter gene assays in COS-1 cells, which are devoid of endogenous GR. The cells were co-transfected with a GR expression vector, a reporter plasmid where the luciferase gene is driven by four copies of GRE (pGRE4-tk-LUC), and increasing amounts of a vector expressing either a full-length IRF2BP2 or an IRF2BP2 with mutated (mZnF) or deleted (ZnF) zinc finger or RING finger (mRING, RING) domain (Fig. 1A). GR, but not IRF2BP2 or empty vector, clearly. 8.

(11) increased the luciferase activity in response to dexamethasone (dex; a synthetic glucocorticoid; 16 h, 100 nM) treatment compared to vehicle (Fig. 1 B). Interestingly, the dex-induced GR luciferase activity showed consistently a dose-dependent and strong increase in response to increasing amounts of the full length IRF2BP2 co-transfected with GR. In comparison, IRF2BP2 with point mutations in either ZnF (mZnF) or RING (mRING) domains had severely compromised GR-activating effect, and the deletion of. IP T. either domain (ZnF, RING) completely nullified the effect of IRF2BP2 on GR-mediated transactivation (Fig. 1B). Using the same setup, we found that IRF2BP2 was similarly capable of. SC R. activating androgen receptor, a nuclear receptor closely related to GR (supplementary fig 1).. Next, we determined whether IRF2BP2 plays a similar coactivator activity on NF-B-mediated transactivation. COS-1 cells were co-transfected with a p65 expression vector26 and a luciferase. N. U. reporter plasmid driven by six B-binding sites. The p65, but not IRF2BP2, was able to induce the. A. luciferase activity in comparison to empty vector (Fig. 1C). In addition, co-transfections with increasing. M. amounts of full length IRF2BP2 enhanced the p65-mediated transactivation consistently in a dosedependent manner (Fig 1C). On the contrary, the mZnF form of IRF2BP2 showed severely reduced. ED. ability to coactivate p65, and the mRING, ZnF and RING mutants were devoid of this ability (Fig 1C). Taken together, our results indicate that IRF2BP2 is a potent coactivator of both GR- and NFB-. CC E. PT. mediated transcription in vitro.. A. IRF2BP2 and GR co-occur on chromatin upon glucocorticoid treatment GR interacts with several coregulator proteins, many of which modulate its function on enhancers. Next, we studied if IRF2BP2 is a part of GR’s transcription regulation machinery at the GC-regulated enhancers. Genome-wide chromatin-binding of IRF2BP2 was analyzed using ChIP-seq in vehicle (veh) and dex (100 nM, 1 h)- exposed GR-expressing HEK293 (HEK293-GR) cells. The genome-wide analyses. 9.

(12) revealed ~1100 IRF2BP2 chromatin-binding sites (BS) in the presence of veh (Fig 2A). Interestingly, IRF2BP2 responded strongly to dex treatment by dissociating from ~600 chromatin BS (veh unique) and gaining ~900 new BS (dex unique, Fig 2A). Next, we compared the IRF2BP2-BS with GR ChIP-seq data from HEK293-GR cells21 and ChIP-seq data of active (H3K27ac) and repressive (H3K9me3) histone marks from HEK293 cells27. Interestingly, the veh unique IRF2BP2-BS were largely devoid of GR ChIP-. IP T. seq signal, while shared and especially dex unique IRF2BP2-BS showed a strong GR binding, suggesting that the dex induces the binding of IRF2BP2 at GR-bound enhancers (Fig 2A). The active histone mark. SC R. was prominently flanking the veh unique and shared IRF2BP2-BS and less the dex unique IRF2BP2-BS. (Fig. 2A) whereas the histone mark for repressed chromatin (H3K9me3) was not enriched at IRF2BP2BS (Fig. 2A). Line profile analysis further showed that GR co-occurs strongly with dex unique IRF2BP2-. U. BS and to a lesser degree with shared IRF2BP2-BS, while the H3K27ac signal was enriched at veh. N. unique and shared IRF2BP2-BS (Fig 2B). A strong dex-induced GR binding and relatively weak H3K27ac. A. signal in the absence of GC treatment suggests that dex unique IRF2BP2-BS are GR binding-activated. M. enhancers36. Furthermore, genome-wide comparison of IRF2BP2-BS and GR-BS revealed that ~80%. ED. (722 of 916) of dex unique IRF2BP2-BS were co-occupied by GR, whereas ~44% (212 of 484) of shared and only ~12% (72 of 624) of veh unique IRF2BP2-BS co-localized with the GR-BS (Fig 2C). The majority. PT. of GR-BS did not co-occur with IRF2BP2, suggesting that additional factors regulate the chromatin cooccurrence of IRF2BP2 with the GR. Peak distribution analysis showed that, irrespectively of the. CC E. treatment, IRF2BP2-BS often reside at intronic and intergenic regions, which is a typical binding pattern for TFs and associated coregulators (Fig 2D). Analysis of DNA-binding motifs at different IRF2BP2-BS categories displayed an enrichment of the nuclear receptor motifs at the dex-induced. A. IRF2BP2-BS (Fig 2E). In addition, a homeobox motif (HOXB13) was most significantly associated with the dex-repressed IRF2BP2-BS and several basic Leucine Zipper motifs (bZIP) motifs with the shared IRF2BP2-BS (Fig 2E). IRF2BP2-BS analysis using GREAT37 in turn showed that genes enriched with TF regulation and developmental processes associated with IRF2BP2-BS in vehicle treatment (Fig 2F). Interestingly, the genes associated with IRF2BP2 in dex treatment were enriched with GR binding 10.

(13) regulation, apoptotic signaling, and negative regulation of protein kinase activity (Fig 2F). Collectively, our result show that IRF2BP2 co-occurs at a subset of GR-bound enhancers in dex-induced cells, suggesting that IRF2BP2 is a part of the transcription regulatory complex of the GR at specific enhancers.. IP T. IRF2BP2 changes glucocorticoid-induced transcriptional responses, cell proliferation and cell migration. After confirming the co-occurrence of IRF2BP2 and GR at GR-bound enhancers, we were interested to. SC R. determine whether IRF2BP2 regulates GC-responsive genes. To that end, we silenced IRF2BP2 in the. HEK293-GR cells and analyzed the gene expression upon dex exposure (100 nM, 6 h) using RNA-seq.. U. The IRF2BP2 siRNA effectively depleted the IRF2BP2 protein as assessed by immunoblotting (Fig 3A).. N. The analysis of RNA-seq data showed over 1100 dex-regulated genes (FDR < 0.01, |logFC|> 0.5, RPKM. A. > 5) in control samples (siNON, nontargeting siRNAs) and ~800 in IRF2BP2-silenced samples. M. (siIRF2BP2; Fig 3B). Next, we grouped all dex-regulated genes based on the response in siNON and siIRF2BP2 conditions, which resulted in six different groups (Fig 3B, group 1-6). Approximately half of. ED. the genes (661 of 1305) responded similarly to dex in both siNON and siIRF2BP2 (group 1: dex-induced and group 3: dex-repressed). The remaining half (644 of 1305) of dex-responsive genes were. PT. differentially regulated in siNON and siIRF2BP2 conditions (groups 2, 4, 5, and 6). IRF2BP2 depletion rendered 207 dex-induced and 291 dex-repressed genes unresponsive to dex (Fig 3B, group 2 and 4).. CC E. In addition, 65 and 81 genes were significantly induced (group 5) and repressed (group 6) in dex treatment, respectively, only when IRF2BP2 was silenced (Fig 3B).. A. To analyze whether IRF2BP2 chromatin binding correlates with the dex-regulated genes, we associated IRF2BP2-BS with genes. Intriguingly, dex unique IRF2BP2-BS were strongly associated with dex-induced genes, while veh unique and shared IRF2BP2-BS were associated with dex-repressed genes (Fig 3C). IRF2BP2 depletion changed the dex-regulation of IRF2BP2-BS-associated genes and over 25% of genes associated with dex-induced IRF2BP2-BS (183 genes in total) were dex-regulated. 11.

(14) only in siNON (26 genes) or siIRF2BP2 (24 genes). However, the overall dex-response of IRF2BP2-BSassociated genes was not affected by the IRF2BP2 depletion (Fig 3C). This suggests that the coregulatory function of IRF2BP2 is restricted to a specific set of GR target genes. To understand the IRF2BP2 regulated cellular processes, we utilized the upstream regulator analysis and the diseases and functions modules of the Ingenuity Pathway Analysis (IPA) tool to analyze the. IP T. dex-regulated genes (|logFC| > 1, FDR < 0.01) in the RNA-seq data. As expected, the upstream regulator analysis showed a strong dex-exposure-triggered activation of GR-related (e.g.. SC R. dexamethasone, NR3C1) pathways in siNON and siIRF2BP2 treatments (Fig 3D; see supplementary. table 1 for all pathways). In addition, the upstream regulator analysis predicted that dex induces. U. activation and repression of several other signaling pathways. This can be explained by the overlap of. N. target genes of dex and other signaling pathways. These overlapping genes include e.g. NFKBIA that is a dex-induced inhibitor of NF-B, but also induced by NF-B as a part of a negative feedback. A. regulation38. In comparison to siNON conditions, the depletion of IRF2BP2 augmented the dex-. M. mediated activation of the anti-inflammatory IL4 pathway and attenuated the dex-induction of several. ED. pathways, including the pro-inflammatory IL6, NF-B and protein kinase pathways (Fig. 3D). The analysis of diseases and functions pathways (Fig. 3E) revealed that the IRF2BP2 depletion reduced the. PT. dex-mediated induction of development and migration related pathways, e.g. development of epithelial tissue and migration of cells, while the cell clustering pathways, for example aggregation of. CC E. cells, became dex-inducible.. We next used live-cell imaging to investigate the predicted effect of IRF2BP2 silencing on HEK293-GR. A. proliferation and migration. The dex markedly reduced the proliferation of HEK293-GR cells, when compared to vehicle (Fig 3F). The silencing of IRF2BP2 further inhibited the cell proliferation, especially in vehicle treatment (Fig. 3F). In contrast to the pathway prediction (Fig. 3E), the dex treatment strongly inhibited the cell migration in wound healing assay, irrespectively of IRF2BP2 silencing (Fig 3G). Moreover, the silencing of IRF2BP2 increased the cell migration compared to siNON control in. 12.

(15) dex and veh treatments (Fig 3G). A closer inspection revealed that the prediction of dex-induced activation of migration pathways was based on several contradicting changes in gene expression, i.e. changes that promote and repress migration (supplementary figure 2), which is likely to the difference between the predicted and the observed changes in cell migration. These data show that dex-induced IRF2BP2-BS associate with dex-induced genes and silencing of. IP T. IRF2BP2 alters dex-regulated transcription, suggesting that the IRF2BP2 functions as a coregulator of the GR at a subset of GC-regulated genes. Moreover, the silencing of IRF2BP2 affected the cell. U. involved in the regulation of GR´s anti-inflammatory functions.. SC R. proliferation and the cell mobility. Intriguingly, the pathway analysis implied that the IRF2BP2 is. N. Glucocorticoid and TNF signaling alter chromatin binding of IRF2BP2 To investigate the connection of IRF2BP2 with NF-B signaling, we analyzed whether dex (100nM, 1h),. M. A. pro-inflammatory cytokine tumor necrosis factor  (TNF, 10ng/ml, 1h) or co-treatment with dex and TNF (DT) alter IRF2BP2 chromatin binding in lung epithelial A549 cells. To find putative dex or TNF-. ED. responsive IRF2BP2-BS in A549 cells, we first identified chromatin regions with co-occurring binding of IRF2BP2 in vehicle and dex-treated HEK293-GR cells, GR in dex-treated A549 cells27 and p65 in TNF-. PT. treated A549 cells31. In general, the overlap of IRF2BP2-BS was greater with GR-BS (14.7 % of IRF2BP2BS in vehicle and 36.0 % for IRF2BP2-BS in dex) than with p65-BS (1.2% of IRF2BP2-BS in vehicle and. CC E. 1.1% of IRF2BP2-BS in dex).. Next, we used ChIP-qPCR to analyze how chromatin binding of IRF2BP2 changes in vehicle, dex, TNF,. A. or DT on seven selected chromatin regions with co-occurring IRF2BP2-BS, GR-BS and p65-BS. On three of the analyzed regions, we did not observe treatment-responsive changes in the chromatin binding of IRF2BP2 (data not shown). On three of the regions, we observed dex-induced chromatin binding of IRF2BP2 (Fig 4 A-C), and on one chromatin region we observed TNF induced binding of IRF2BP2 (Fig. 4D). Interestingly, on all regions with dex (Fig. 4 A-C) or TNF (Fig. 4D) -induced IRF2BP2 binding, the. 13.

(16) DT co-treatment reduced the binding of IRF2BP2. Moreover, on none of the analyzed regions the chromatin binding of IRF2BP2 was induced by both dex and TNF, suggesting that GR and NF-B compete for chromatin recruitment of IRF2BP2.. IP T. IRF2BP2 modulates glucocorticoid and TNF-induced gene programs To analyze the role of IRF2BP2 in GR´s anti-inflammatory actions, we silenced IRF2BP2 in A549 cells,. treated the cells for 6 h with vehicle (veh), dex (100 nM), TNF (10 ng/ml), or both dex and TNF (DT),. SC R. and analyzed the changes in gene expression using RNA-seq. Immunoblotting confirmed that the IRF2BP2 was efficiently silenced with siRNA (Fig 5A). In addition, the silencing of IRF2BP2 slightly. U. increased the amounts of GR and p65 in TNF and DT treatments compared to control transfection. N. (Fig 5A).. A. Hierarchical clustering of differentially expressed genes (RPKM > 0.5, FDR < 0.01, |log2FC| > 1. M. compared to vehicle) grouped similarly treated samples together, indicating that treatments (dex, TNF, or DT) had a larger effect on the gene expression than the silencing of IRF2BP2 (Fig 5B). Compared. ED. to control transfection (siNON), the silencing of IRF2BP2 (siIRF2BP2) rendered 60 to 75 genes nonresponsive in treatments (dex, TNF, DT), while ~180 to 360 new genes became treatment-responsive. PT. (Fig 5C). To assess the role of IRF2BP2 in the crosstalk between GC and TNF signaling, we sought for. CC E. genes that were differentially regulated in DT compared to single treatments (dex or TNF). Specifically, we looked for genes that were (i) under competitive regulation, i.e. their gene regulation by dex and TNF was nullified in co-treatment, or (ii) under synergistic regulation, i.e. genes that were differentially. A. expressed only in co-treatment. The silencing of IRF2BP2 modulated the expression of multiple genes that were differentially regulated by DT co-treatment compared to either of the single treatments. We found ~100 genes that were differentially regulated by co-treatment in siNON, when compared to dex or TNF, but not in siIRF2BP2 (supplementary figure 3A). This group of genes included e.g. FOS that was induced by dex and repressed by TNF in siNON and siIRF2BP2 conditions, but induced by DT co-. 14.

(17) treatment only in siIRF2BP2 conditions, i.e. losing its repression by TNF upon silencing of IRF2BP2. In addition, ~200 genes were differentially regulated by co-treatment only after silencing of the IRF2BP2 (supplementary figure 3B). This group of genes included STAT3 that was not regulated by dex or TNF alone, but was significantly induced by DT co-treatment upon silencing of IRF2BP2. We used RT-qPCR to verify the effect of siIRF2BP2 in the regulation of five inflammation or. IP T. development-related genes that responded to IRF2BP2 depletion in RNA-seq analysis. Expression of TNF and DT-induced genes SOD2 (encoding superoxide dismutase 2) and CXCL8 (encoding interleukin-. SC R. 8 protein) was increased and decreased, respectively, in siIRF2BP2 compared to siNON conditions (Fig 5D). In addition, depletion of IRF2BP2 inhibited the induction of PTGS2 (prostaglandin-endoperoxidase. U. synthase 2, encoding COX2 protein) by TNF, the induction of FSTL3 (encoding follistatin-related protein 3) by dex and DT and that of S100P by dex when compared to siNON conditions. On the other. N. hand, the over-expression of IRF2BP2 by transfecting an IRF2BP2 expression plasmid augmented the. A. TNF-induction of PTGS2 and DT-induction of FSTL3 (supplementary figure 4). In agreement with the. M. RNA-seq data, the RT-qPCR results confirm that the IRF2BP2 modulates dex, TNF and DT responses in. ED. A549 cells.. Next, we analyzed how siIRF2BP2 alone affects the gene expression in A549 cells. Depending on the. PT. treatment, there were ~400 differentially expressed genes (veh: 381, dex: 429, TNF:456, DT:379) in. CC E. siIRF2BP2 compared to control siRNA conditions (RPKM > 0.5, |log2FC| > 1, FDR < 0.01; see supplementary table 2 for RNA-seq analysis). Of note, compared to siNON conditions, the silencing of IRF2BP2 led to reduced expression of over 70% of the affected genes regardless of the treatment,. A. indicating that IRF2BP2 was predominantly an activator for their expression. To gain a comprehensive view on the biologically meaningful changes in the RNA-seq data, we used IPA to analyze how differentially expressed genes enrich to known functional pathways. The upstream analysis module of IPA indicated that IRF2BP2 silencing-induced changes in the gene expression are similar to those induced by various protein kinase inhibitors (Fig 5E for the 5 most induced or repressed pathways in. 15.

(18) siIRF2BP2; see supplementary table 3 of all IPA pathways). In accordance with its role as a repressor of interferon signaling, the silencing of IRF2BP2 activated interferon-response-related pathways, especially in response to TNF and DT (interferon, IFNB2, IFNA1 and IRF7 in Fig 5E). In the diseases and functions module, the inflammation and cell death pathways were induced, while those of angiogenesis and cell movement were repressed in siIRF2BP2 (supplementary table 3). These data. IP T. show that the level of IRF2BP2 affects the expression of hundreds of genes in A549 cells and suggests that IRF2BP2 is a modulator of several protein kinase, anti-inflammatory and anti-apoptotic pathways.. SC R. Next, we used IPA to analyze how siIRF2BP2 affects dex, TNF or DT-regulated cellular pathways. The upstream regulator analysis predicted activation of GC and TNF pathways in corresponding treatments. U. and activation of both pathways in DT treatment, confirming that the A549 cells were responsive to treatments (supplementary table 3). In canonical pathway analysis that aims to predict the activity of. N. curated cellular signaling pathways, the depletion of IRF2BP2 enhanced the dex-induced repression. A. of inflammatory and immunity pathways, e.g. NFB and TREM1 signaling (selected pathways in Fig.. M. 5F, see supplementary table 3 for all pathways). The TNF and DT-regulated genes converged with. ED. canonical pathways more often than dex-regulated genes (Fig 5F). Silencing of the IRF2BP2 dampened the activation of several inflammatory pathways by TNF and DT, including several interleukin signaling. PT. pathways and osteoarthritis pathway (Fig 5F). On the contrary, the depletion of IRF2BP2 increased the TNF and DT-mediated activation of e.g. TNFR1/2 signaling, hypoxia signaling and RIG-1 anti-viral. CC E. pathways. Silencing of IRF2BP2 also repressed DT-mediated activation of e.g. insulin receptor signaling and IL-9 pathways.. A. These data suggest that IRF2BP2 is important for the transcriptional regulation of a subset of GC- and TNF-regulated genes. The effect of IRF2BP2 depletion on gene expression is gene selective and can either augment or attenuate the effects of GC and TNF signaling. Moreover, the IRF2BP2-sensitive genes converge to inflammatory, metabolic and cell signaling pathways.. 16.

(19) Discussion Coregulators are thought to bridge TFs to components of basal transcription machinery and/or alter the chromatin structure and the accessibility of the chromatin for TFs, thus having a fundamental impact in the regulation of gene expression. In fact, dysfunction of coregulators may lead to severe pathological states and the coregulators are emerging as potential new drug targets in various. IP T. diseases12,39,40. However, the exact mechanism how TFs, including GR and NF-B, cooperate with coregulators to create a regulatory environment facilitating a rapid modulation of gene transcription,. SC R. is not known. Therefore, identification and characterization of GR coregulators are of crucial importance and could e.g. provide new approaches for the treatment of inflammatory diseases.. U. We recently identified IRF2BP2, that was originally reported as a IRF2 corepresor14, as a novel GRinteracting protein in an unbiased proximity mapping of proteins residing within 10-nm radius of. N. agonist-bound GR13. Since GR is a potent anti-inflammatory TF and the actions of IRF2BP2 are. A. associated with the regulation of inflammatory processes, we hypothesized that the IRF2BP2 could be. M. a coregulator of the crosstalk between GC and TNF signaling. IRF2BP2 contains a conserved N-terminal. ED. C4 zinc finger and a C-terminal C3HC4 RING finger domain. The zinc fingers often recognize specific DNA sequences41, while the RING fingers mediate protein-protein interactions and formation of multi. PT. protein complexes42. In the case of IRF2BP2, the zinc finger domain is important for its dimerization and interactions with IRF2BP1 and IRF2BPL, while the RING finger mediates its interaction with the. CC E. IRF214,17. Many RING finger-containing proteins are E3 ubiquitin or SUMO ligases that assist in ubiquitination or SUMOylation in their RING finger domain-dependent fashion43,44. We tested the. A. potential ubiquitin and SUMO ligase activity of IRF2BP2 in COS-1 cell-based coexpression assays, but did not observe E3 ligase activity for IRF2BP2 towards GR or cellular proteins in general (not shown), which is in agreement with data obtained with other substrates16,20. However, our results show that the ability of IRF2BP2 to function as a potent coactivator of for GR and NF-B in reporter gene assays is dependent on RING finger and zinc finger domains. The observed coactivation function of IRF2BP2 contradicts its previously reported role as a corepressor of the IRF2 and nuclear factor of activated T17.

(20) 1 (NFAT-1), a function which is also dependent on the zinc-coordinating domains14,45. However, IRF2BP2 has also been reported to activate gene expression, being an activator of vascular endothelial growth factor-A gene46 and coactivator of Krüppel-like factor 2, an anti-inflammatory TF19. Moreover, a transcriptome-wide analysis in mouse erythroleukemia cells showed that the depletion of IRF2BP2 represses and induces a comparable number of genes16.. IP T. IRF2BP2 is important for erythroid cell differentiation, as IRF2BP2 knockout mice died at birth due to. ineffective liver erythropoiesis16. In erythroid cells, IRF2BP2 functions as a transcriptional corepressor. SC R. of the LDB1 complex, preventing its activation on chromatin and thus premature differentiation of the cells16. In mouse erythroleukemia cells, IRF2BP2 binds to chromatin as a part of ETO2-repressive coregulator complex, which is suggested to repress transcription by recruiting an NCOR1 co-repressor. U. complex to the LDB1 complex-bound enhancers16. By using ChIP-seq, we found that the chromatin. N. binding of IRF2BP2 responds strongly to the GC treatment. Upon GC stimulus, the majority of IRF2BP2-. A. binding sites co-occur with GR at GRE-containing enhancers and associate with the GC-activated. M. genes. In the absence of GC stimulus, IRF2BP2 binds to chromatin sites that are devoid of GR and. ED. associate with GC-repressed genes. As suggested by the motif enrichment analysis, IRF2BP2 binds to GC-repressed sites together with TFs of homeobox and bZip families, e.g. activator protein 1 (AP-1).. PT. These data suggest that IRF2BP2 is a part of the transcription regulatory complex of GR and possibly. CC E. other TFs, including AP-1.. Depletion of IRF2BP2 in GC- and TNF-responsive A549 cells altered the gene expression in response to anti-inflammatory and pro-inflammatory signaling. The effect of IRF2BP2 silencing was most. A. prominent in conditions that mimic the GC effects in inflamed tissue, i.e. when both GC and TNF signaling pathways were activated. Irrespectively of the presence of GC, TNF or GC plus TNF, the depletion of IRF2BP2 resulted in induction and repression of a comparable number of genes. Based on ChIP-qPCR, the chromatin binding of IRF2BP2 may be induced by GC and TNF at GR- and NF-B bound enhancers in A549 cells. Based on the current data, we cannot however deduce whether the. 18.

(21) function of GR, NF-B, or both, is modulated by IRF2BP2 in co-treatment conditions. Expression of a number of inflammation-related genes is sensitive for the IRF2BP2 level. For example, the depletion of IRF2BP2 potentiated TNF-triggered induction of SOD2 and attenuated that of PTGS2. The depletion of IRF2BP2 also activated genes of protein kinase inhibitor-associated pathways, suggesting that the IRF2BP2 can modulate protein kinase signaling. This observation is also interesting, as mitogen-. IP T. activated protein kinases (MAPKs), such as pro-inflammatory kinases p38 and JNKs (c-Jun N-terminal kinases), are activated in response to various inflammatory-stimuli and regulate the inflammatory. SC R. responses, e.g. by activating pro-inflammatory TFs47. The activity of MAPKs is in turn repressed by a GC-induced dual-specificity phosphatase 1 (DUSP1)48,49. The role of IRF2BP2 in the modulation of. U. signaling by MAPKs could in part explain its role in the inflammatory signaling.. N. The anti-inflammatory effects of GCs often result from their ability to repress genes activated by pro-. A. inflammatory TFs such as NF-B6. However, the functional interplay between GR and the proinflammatory TFs, including NF-B, is complex and can also lead to a synergistic regulation of genes,. M. which is important for the dampening of the inflammatory signal via regulatory feedback loops38,50.. ED. Our data link IRF2BP2 with GC-mediated regulation of NF-B pathway. Expression of a large number of genes that were repressed by GC and induced by TNF (or vise versa) or induced only in the presence. PT. of GC plus TNF were regulated by GC and TNF, were sensitive to IRF2BP2 depletion in A549 cells. This suggests that IRF2BP2 is involved in the regulation of the crosstalk between GR and NF-B. Differential. CC E. recruitment of IRF2BP2 to enhancers co-bound by GR and p65 subunit of NF-B in response to GC, TNF and co-treatment supports this idea.. A. Conclusion. Whether a coregulator functions as a coactivator or a corepressor can be TF- and cell type-specific or depend on the chromatin context of the regulatory region51. Interestingly, our reporter gene assays show that IRF2BP2 is a potent coactivator of the GR- and the NF-B-mediated transactivation, while the effects of IRF2BP2 depletion on transcriptome-wide analysis were more equally divided between 19.

(22) activation and repression of genes. Although we cannot confirm the source for these contradictory results, they likely result from the differences in the enhancer environment (plasmid vs. chromatin) and in the binding of other regulatory proteins (TFs and coregulators). In conclusion, the IRF2BP2 is a. IP T. novel coregulator of a subset of target genes on glucocorticoid and TNF signaling pathways.. Data sets. SC R. ChIP-seq and RNA-seq data is available through GEO (https://www.ncbi.nlm.nih.gov/geo/) with accession number GSE124636.. Acknowledgements. M. A. N. U. We thank Merja Räsänen and Eija Korhonen for their assistance with cell culture, Kirsi Ketola for proofreading the manuscript, and Professor Claude Libert for the TNF. We also thank EMBL GeneCore sequencing team for deep sequencing and UEF Cell and Tissue Imaging Unit for help with live-cell imaging.. Funding. ED. This study was supported by the Academy of Finland, the Finnish Cancer Organisations, the Sigrid Jusélius Foundation, and the UEF Doctoral Programme in Molecular Medicine .. PT. Competing financial interests. A. CC E. We declare no competing financial interests.. 20.

(23) References 1. Barnes PJ. Anti-inflammatory actions of glucocorticoids: Molecular mechanisms. Clin Sci (Lond). 1998;94(6):557-572. 2. Stahn C, Löwenberg M, Hommes DW, Buttgereit F. Molecular mechanisms of glucocorticoid action and selective glucocorticoid receptor agonists. Mol Cell Endocrinol. 2007;275(1-2):71-78.. IP T. 3. Silverman MN, Sternberg EM. Glucocorticoid regulation of inflammation and its functional correlates: From HPA axis to glucocorticoid receptor dysfunction. Ann N Y Acad Sci. 2012;1261(1):5563. 4. Heitzer MD, Wolf IM, Sanchez ER, Witchel SF, DeFranco DB. Glucocorticoid receptor physiology. Reviews in endocrine and metabolic disorders. 2007;8(4):321-330.. SC R. 5. Vandevyver S, Dejager L, Libert C. Comprehensive overview of the structure and regulation of the glucocorticoid receptor. Endocr Rev. 2014;35(4):671-693.. U. 6. Kadmiel M, Cidlowski JA. Glucocorticoid receptor signaling in health and disease. Trends Pharmacol Sci. 2013;34(9):518-530.. A. N. 7. De Bosscher K, Vanden Berghe W, Haegeman G. The interplay between the glucocorticoid receptor and nuclear factor-κB or activator protein-1: Molecular mechanisms for gene repression. Endocr Rev. 2003;24(4):488-522.. M. 8. Smoak KA, Cidlowski JA. Mechanisms of glucocorticoid receptor signaling during inflammation. Mech Ageing Dev. 2004;125(10-11):697-706.. ED. 9. Vallabhapurapu S, Karin M. Regulation and function of NF-κB transcription factors in the immune system. Annu Rev Immunol. 2009;27:693-733.. PT. 10. Smale ST. Dimer‐specific regulatory mechanisms within the NF‐κB family of transcription factors. Immunol Rev. 2012;246(1):193-204.. CC E. 11. Oeckinghaus A, Ghosh S. The NF-kappaB family of transcription factors and its regulation. Cold Spring Harb Perspect Biol. 2009;1(4):a000034. 12. Lonard DM, Lanz RB, O’Malley BW. Nuclear receptor coregulators and human disease. Endocr Rev. 2007;28(5):575-587.. A. 13. Lempiainen JK, Niskanen EA, Vuoti KM, et al. Agonist-specific protein interactomes of glucocorticoid and androgen receptor as revealed by proximity mapping. Mol Cell Proteomics. 2017;16(8):1462-1474. 14. Childs KS, Goodbourn S. Identification of novel co‐repressor molecules for interferon regulatory Factor‐2. Nucleic Acids Res. 2003;31(12):3016-3026. 15. Koeppel M, van Heeringen SJ, Smeenk L, Navis AC, Janssen-Megens EM, Lohrum M. The novel p53 target gene IRF2BP2 participates in cell survival during the p53 stress response. Nucleic Acids Res. 2009;37(2):322-335.. 21.

(24) 16. Stadhouders R, Cico A, Stephen T, et al. Control of developmentally primed erythroid genes by combinatorial co-repressor actions. Nature communications. 2015;6:8893. 17. Yeung KT, Das S, Zhang J, et al. A novel transcription complex that selectively modulates apoptosis of breast cancer cells through regulation of FASTKD2. Mol Cell Biol. 2011;31(11):2287-2298. 18. Cruz SA, Hari A, Qin Z, et al. Loss of IRF2BP2 in microglia increases inflammation and functional deficits after focal ischemic brain injury. Frontiers in cellular neuroscience. 2017;11:201.. IP T. 19. Chen H, Keyhanian K, Zhou X, et al. IRF2BP2 reduces macrophage inflammation and susceptibility to atherosclerosis. Circ Res. 2015;117(8):671-683.. SC R. 20. Tinnikov AA, Yeung KT, Das S, Samuels HH. Identification of a novel pathway that selectively modulates apoptosis of breast cancer cells. Cancer Res. 2009;69(4):1375-1382.. 21. Paakinaho V, Kaikkonen S, Makkonen H, Benes V, Palvimo JJ. SUMOylation regulates the chromatin occupancy and anti-proliferative gene programs of glucocorticoid receptor. Nucleic Acids Res. 2014;42(3):1575-1592.. U. 22. Makkonen H, Jääskeläinen T, Rytinki MM, Palvimo JJ. Analysis of androgen receptor activity by reporter gene assays. In: Androgen action. Springer; 2011:71-80.. A. N. 23. Kaikkonen S, Paakinaho V, Sutinen P, Levonen A, Palvimo JJ. Prostaglandin 15d-PGJ2 inhibits androgen receptor signaling in prostate cancer cells. Molecular Endocrinology. 2013;27(2):212-223.. M. 24. Ikonen T, Palvimo JJ, Janne OA. Interaction between the amino- and carboxyl-terminal regions of the rat androgen receptor modulates transcriptional activity and is influenced by nuclear receptor coactivators. J Biol Chem. 1997;272(47):29821-29828.. ED. 25. Tian S, Poukka H, Palvimo JJ, Janne OA. Small ubiquitin-related modifier-1 (SUMO-1) modification of the glucocorticoid receptor. Biochem J. 2002;367(Pt 3):907-911.. PT. 26. Palvimo JJ, Reinikainen P, Ikonen T, Kallio PJ, Moilanen A, Janne OA. Mutual transcriptional interference between RelA and androgen receptor. J Biol Chem. 1996;271(39):24151-24156.. CC E. 27. ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489(7414):57. 28. Sutinen P, Malinen M, Heikkinen S, Palvimo JJ. SUMOylation modulates the transcriptional activity of androgen receptor in a target gene and pathway selective manner. Nucleic Acids Res. 2014;42(13):8310-8319.. A. 29. Heinz S, Benner C, Spann N, et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010;38(4):576589. 30. Schindelin J, Arganda-Carreras I, Frise E, et al. Fiji: An open-source platform for biological-image analysis. Nature methods. 2012;9(7):676.. 22.

(25) 31. Raskatov JA, Meier JL, Puckett JW, Yang F, Ramakrishnan P, Dervan PB. Modulation of NF-kappaBdependent gene transcription using programmable DNA minor groove binders. Proc Natl Acad Sci U S A. 2012;109(4):1023-1028. 32. Robinson JT, Thorvaldsdóttir H, Winckler W, et al. Integrative genomics viewer. Nat Biotechnol. 2011;29(1):24. 33. Bolger AM, Lohse M, Usadel B. Trimmomatic: A flexible trimmer for illumina sequence data. Bioinformatics. 2014;30(15):2114-2120.. IP T. 34. Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL. TopHat2: Accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 2013;14(4):R36.. SC R. 35. Robinson MD, McCarthy DJ, Smyth GK. edgeR: A bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26(1):139-140.. U. 36. Johnson TA, Chereji RV, Stavreva DA, Morris SA, Hager GL, Clark DJ. Conventional and pioneer modes of glucocorticoid receptor interaction with enhancer chromatin in vivo. Nucleic Acids Res. 2017;46(1):203-214.. N. 37. McLean CY, Bristor D, Hiller M, et al. GREAT improves functional interpretation of cis-regulatory regions. Nat Biotechnol. 2010;28(5):495.. M. A. 38. Newton R, Shah S, Altonsy MO, Gerber AN. Glucocorticoid and cytokine crosstalk: Feedback, feedforward, and co-regulatory interactions determine repression or resistance. J Biol Chem. 2017;292(17):7163-7172.. ED. 39. Dasgupta S, O'Malley BW. Transcriptional coregulators: Emerging roles of SRC family of coactivators in disease pathology. J Mol Endocrinol. 2014;53(2):R47-59.. PT. 40. Lonard DM, O'malley BW. Nuclear receptor coregulators: Modulators of pathology and therapeutic targets. Nature Reviews Endocrinology. 2012;8(10):598.. CC E. 41. Laity JH, Lee BM, Wright PE. Zinc finger proteins: New insights into structural and functional diversity. Curr Opin Struct Biol. 2001;11(1):39-46. 42. Saurin AJ, Borden KL, Boddy MN, Freemont PS. Does this have a familiar RING? Trends Biochem Sci. 1996;21(6):208-214. 43. Freemont PS. Ubiquitination: RING for destruction? Current Biology. 2000;10(2):R84-R87.. A. 44. Lipkowitz S, Weissman AM. RINGs of good and evil: RING finger ubiquitin ligases at the crossroads of tumour suppression and oncogenesis. Nature reviews Cancer. 2011;11(9):629. 45. Carneiro FR, Ramalho-Oliveira R, Mognol GP, Viola JP. Interferon regulatory factor 2 binding protein 2 is a new NFAT1 partner and represses its transcriptional activity. Mol Cell Biol. 2011;31(14):2889-2901. 46. Teng AC, Kuraitis D, Deeke SA, et al. IRF2BP2 is a skeletal and cardiac muscle-enriched ischemiainducible activator of VEGFA expression. The FASEB Journal. 2010;24(12):4825-4834. 23.

(26) 47. Arthur JSC, Ley SC. Mitogen-activated protein kinases in innate immunity. Nature Reviews Immunology. 2013;13(9):679. 48. Abraham SM, Lawrence T, Kleiman A, et al. Antiinflammatory effects of dexamethasone are partly dependent on induction of dual specificity phosphatase 1. J Exp Med. 2006;203(8):1883-1889. 49. Shah S, King EM, Chandrasekhar A, Newton R. Roles for the mitogen-activated protein kinase (MAPK) phosphatase, DUSP1, in feedback control of inflammatory gene expression and repression by dexamethasone. J Biol Chem. 2014;289(19):13667-13679.. IP T. 50. Lannan EA, Galliher-Beckley AJ, Scoltock AB, Cidlowski JA. Proinflammatory actions of glucocorticoids: Glucocorticoids and TNFα coregulate gene expression in vitro and in vivo. Endocrinology. 2012;153(8):3701-3712.. A. CC E. PT. ED. M. A. N. U. SC R. 51. Rosenfeld MG, Lunyak VV, Glass CK. Sensors and signals: A coactivator/corepressor/epigenetic code for integrating signal-dependent programs of transcriptional response. Genes Dev. 2006;20(11):1405-1428.. 24.

(27) IP T SC R U N A M ED PT. A. CC E. Figure 1: IRF2BP2 increases transcriptional activity of GR and NFB. (A) Schematic presentation of IRF2BP2 and its mutants used in this study. (B) COS-1 cells were co-transfected with a minimal p(GRE)4tk-luc reporter and ‘empty’ control vector (CTRL), expression vector encoding for GR (GR +/-) or increasing amounts (1, 5, 25 or 100 ng) of expression vector for wild-type or mutated IRF2BP2 as indicated. Cells were exposed to dexamethasone or vehicle (dex +/-, 100 nM) for 16 h and harvested for reporter gene assays as described in Materials and Methods. (C) COS-1 cells were co-transfected with a minimal pĸB6-tk-luc reporter together with ‘empty’ control vector (CTRL), expression vector encoding p65 (p65 +/-) or increasing amounts (1, 5, 25 or 100 ng) of expression vector for wild-type or mutated IRF2BP2 as indicated. Columns are the mean fold changes in luciferase activity of three biological replicates normalized to GR in the absence of dex (B) or p65 (C) and error bars are SDs.. 25.

(28) IP T SC R U N A. A. CC E. PT. ED. M. Figure 2: IRF2BP2 binds to GR-occupied enhancers upon glucocorticoid stimulus. ChIP-seq was used to analyze the chromatin-binding of IRF2BP2 in HEK293-GR cells treated with dex (100 nM) or vehicle (veh) for 1 h. Chromatin binding of IRF2BP2 was compared with publically available ChIP-seq signals of GR from HEK293-GR cells (dex, 100nM, 1h) and histone marks of active (H3K27ac) and repressed (H3K9me3) chromatin from HEK293 cells. (A) False-color scale heat map (intensity increases from darker to lighter color) showing IRF2BP2 ChIP-seq signals in the presence of vehicle and dex, GR ChIPseq signal in the presence of dex, and histone marks H3K27ac and H3K9me3 from ±2kb region centered at IRF2BP2 binding sites (BS). The heat map is divided between vehicle unique (black), shared (gray) and dex unique (orange) IRF2BP2-BS and the number of IRF2BP2-BS in each fraction is given. (B) Line profile of average ChIP-seq signal intensities of IRF2BP2, GR, H3K27ac and H3K9me3 at ±1kb region centered at IRF2BP2-BS fractions described in A. (C) Venn diagram showing the overlap of IRF2BP2- and GR-binding sites. (D) Distribution of IRF2BP2-BS for annotated genomic loci. IRF2BP2-BS are from veh unique (black), shared (grey) and dex unique (orange) conditions. (E) Heat map showing the hierarchical clustering of DNA-binding motifs enriched at IRF2BP2-BS. The color indicates the significance of enrichment in –log10(p-value). (F) GO molecular functions (black) and biological processes (red) of the genes associated with IRF2BP2-BS in vehicle (left) or dex (right) treatment, as mapped using GREAT.. 26.

(29) IP T SC R U. A. CC E. PT. ED. M. A. N. Figure 3: IRF2BP2 knockdown alters glucocorticoid-induced responses in gene expression of HEK293GR cells. (A) Immunoblot showing depletion of IRF2BP2. (B) Heat map of dex-regulated genes in control (siNON)- and IRF2BP2 (siIRF2BP2)-silenced HEK293-GR cells. Genes are grouped to six clusters according to their response to dex. Box plot showing the dex-induced changes (log2 fold-change (FC) (dex/veh)) in gene expression in each group. (C) Box plot of log2FC (dex/veh) of IRF2BP2-BS associated genes. IRF2BP2-BS were associated with nearest gene, which was included to analysis if it was expressed (RPKM > 5) and regulated by dex (FDR <0.01, log2FC(dex/veh) > 0.5) in corresponding siRNA treatment (siNON or siIRF2BP2). Genes associated with dex unique IRF2BP2-BS were significantly induced by dex when compared with other groups (***, p-value < 0.001 using Kruskal-Wallis test and Dunn´s post-test). Heat maps of pathway activated by dex (|log2FC| > 1, FDR < 0.01) using (D) upstream analysis or (E) diseases and functions modules of Ingenuity Pathway Analysis (IPA). Colors represent the activation Z-score associated with dex induction in siNON and siIRF2BP2 conditions. (F) Proliferation of HEK293-GR cells in siNON or siIRF2BP2 upon dex or vehicle (veh) treatment is shown as relative confluence normalized to initial measurement. G) Migration of HEK293-GR cells in siNON or siIRF2BP2 upon dex or vehicle (veh) treatment in wound healing assay is shown as percentage of wound density. F) and G) were analyzed using IncuCyte S3 live cell analysis system.. 27.

(30) IP T SC R. A. CC E. PT. ED. M. A. N. U. Figure 4: Chromatin binding of IRF2BP2 responds to dex and TNF treatments. ChIP-qPCR results of dex-induced (A-C) and TNF- induced (D) IRF2BP2 binding sites in A549 cells. The IRF2BP2 binding was measured from vehicle, dex, TNF and co-treated (dex and TNF, DT) samples. Columns represent mean ± SD of at least three biological replicates. Fold changes were calculated in reference to veh samples. Bonferroni’s multiple comparison test was used to determine the significance of fold change differences between all samples (*** p-value < 0.001, ** p-value < 0.01, * p-value < 0.05). ChIP-seq signals of IRF2BP2 in vehicle treatment (veh, black) and dex treatment (dex, orange) from HEK293-GR cells, and GR in dex treatment (red) and p65 in TNF treatment (blue) from A549 cells are shown as a reference. Numbers indicate the maximum signal, the region amplified in ChIP-qPCR is marked with green bar, and possible overlapping gene with blue bar. Genomic positions correspond to human genome hg38.. 28.

(31) IP T SC R U N A M ED. A. CC E. PT. Figure 5: IRF2BP2 modulates glucocorticoid and TNF-induced gene expression programs in A549 cells. (A) Immunoblot showing depletion of IRF2BP2 in A549 cells stimulated by vehicle (veh), dex (dex), TNF, or simultaneously with dex and TNF (DT). (B) Hierarchical clustering of genes that were differentially regulated (FDR < 0.01, log2FC > 1 or < -1) in different treatments (dex, TNF, DT) compared to veh under siNON or siIRF2BP2 condition. Change in the gene expression is shown as log2 fold-change (treatment/veh) for each treatment. (C) Number of differentially expressed genes in siNON vs siIRF2BP2 in different treatments upon IRF2BP2 silencing. (D) RT-qPCR validation of selected dex- and TNF-regulated inflammation related genes. The expression levels of each gene were measured from vehicle-, dex-, TNF and concomitantly with dex- and TNF- (DT)-treated samples. Columns represent mean ± SD of at least three biological replicates. Fold changes were calculated in reference to siNON veh samples. Bonferroni’s multiple comparison test was used to determine the significance of fold change differences between siNON and siIRF2BP2 in the corresponding treatment (*** p < 0.001, * p < 0.05). (E) Heat map of top five pathways that were differentially activated or repressed in siIRF2BP2 compared to siNON condition by IPA upstream analysis. (F) Heat map showing the selected canonical pathways that were regulated by dex, TNF or DT treatments compared to vehicle.. 29.

(32)