The histone acetyl reader bromodomain-containing protein 4 (BRD4) is an important regulator of chromatin structure and transcription, yet factors modulating its activity have remained elusive. Here we describe two complementary screens for genetic and physical interactors of BRD4, which converge on the folate pathway enzyme MTHFD1 (methylenetetrahydrofolate dehydrogenase, cyclohydrolase and formyltetrahydrofolate synthetase 1). We show that a fraction of MTHFD1 resides in the nucleus, where it is recruited to distinct genomic loci by direct interaction with BRD4. Inhibition of either BRD4 or MTHFD1 results in similar changes in nuclear metabolite composition and gene expression; pharmacological inhibitors of the two pathways synergize to impair cancer cell viability in vitro and in vivo. Our finding that MTHFD1 and other metabolic enzymes are chromatin associated suggests a direct role for nuclear metabolism in the control of gene expression.
Access optionsAccess options
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Filippakopoulos, P. et al. Histone recognition and large-scale structural analysis of the human bromodomain family. Cell 149, 214–231 (2012).
Floyd, S. R. et al. The bromodomain protein Brd4 insulates chromatin from DNA damage signalling. Nature 498, 246–250 (2013).
Filippakopoulos, P. et al. Selective inhibition of BET bromodomains. Nature 468, 1067–1073 (2010).
Winter, G. E. et al. BET bromodomain proteins function as master transcription elongation factors independent of CDK9 recruitment. Mol. Cell 67, 5–18.e19 (2017).
Shi, J. & Vakoc, C. R. The mechanisms behind the therapeutic activity of BET bromodomain inhibition. Mol. Cell 54, 728–736 (2014).
Zhou, Q., Li, T. & Price, D. H. RNA polymerase II elongation control. Annu. Rev. Biochem. 81, 119–143 (2012).
Asangani, I. A. et al. Therapeutic targeting of BET bromodomain proteins in castration-resistant prostate cancer. Nature 510, 278–282 (2014).
Dawson, M. A. et al. Inhibition of BET recruitment to chromatin as an effective treatment for MLL-fusion leukaemia. Nature 478, 529–533 (2011).
Delmore, J. E. et al. BET bromodomain inhibition as a therapeutic strategy to target c-Myc. Cell 146, 904–917 (2011).
Lovén, J. et al. Selective inhibition of tumor oncogenes by disruption of super-enhancers. Cell 153, 320–334 (2013).
Shu, S. et al. Response and resistance to BET bromodomain inhibitors in triple-negative breast cancer. Nature 529, 413–417 (2016).
Zuber, J. et al. RNAi screen identifies Brd4 as a therapeutic target in acute myeloid leukaemia. Nature 478, 524–528 (2011).
Filippakopoulos, P. & Knapp, S. Targeting bromodomains: epigenetic readers of lysine acetylation. Nat. Rev. Drug Discov. 13, 337–356 (2014).
Knoechel, B. et al. An epigenetic mechanism of resistance to targeted therapy in T cell acute lymphoblastic leukemia. Nat. Genet. 46, 364–370 (2014).
Rathert, P. et al. Transcriptional plasticity promotes primary and acquired resistance to BET inhibition. Nature 525, 543–547 (2015).
Fong, C. Y. et al. BET inhibitor resistance emerges from leukaemia stem cells. Nature 525, 538–542 (2015).
Sdelci, S. et al. Mapping the chemical chromatin reactivation landscape identifies BRD4-TAF1 cross-talk. Nat. Chem. Biol. 12, 504–510 (2016).
Carette, J. E. et al. Haploid genetic screens in human cells identify host factors used by pathogens. Science 326, 1231–1235 (2009).
Tchasovnikarova, I. A. et al. Epigenetic silencing by the HUSH complex mediates position-effect variegation in human cells. Science 348, 1481–1485 (2015).
Sabari, B. R. Coactivator condensation at super-enhancers links phase separation and gene control. Science 361, eaar3958 (2018).
Cho, W. K. et al. Mediator and RNA polymerase II clusters associate in transcription-dependent condensates. Science 361, 412–415 (2018).
Ducker, G. S. & Rabinowitz, J. D. One-carbon metabolism in health and disease. Cell Metab. 25, 27–42 (2017).
Field, M. S. et al. Nuclear enrichment of folate cofactors and methylenetetrahydrofolate dehydrogenase 1 (MTHFD1) protect de novo thymidylate biosynthesis during folate deficiency. J. Biol. Chem. 289, 29642–29650 (2014).
Field, M. S., Kamynina, E. & Stover, P. J. MTHFD1 regulates nuclear de novo thymidylate biosynthesis and genome stability. Biochimie 126, 27–30 (2016).
Winter, G. E. et al. Phthalimide conjugation as a strategy for in vivo target protein degradation. Science 348, 1376–1381 (2015).
Bliss, C. I. The toxicity of poisons applied jointly. Ann. Appl. Biol. 26, 585–615 (1939).
Grabner, B. et al. Disruption of STAT3 signalling promotes KRAS-induced lung tumorigenesis. Nat. Commun. 6, 6285 (2015).
Wright, R. H. et al. ADP-ribose-derived nuclear ATP synthesis by NUDIX5 is required for chromatin remodeling. Science 352, 1221–1225 (2016).
Mews, P. et al. Acetyl-CoA synthetase regulates histone acetylation and hippocampal memory. Nature 546, 381–386 (2017).
Kamynina, E. et al. Arsenic trioxide targets MTHFD1 and SUMO-dependent nuclear de novo thymidylate biosynthesis. Proc. Natl Acad. Sci. USA 114, E2319–E2326 (2017).
MacFarlane, A. J. et al. Nuclear localization of de novo thymidylate biosynthesis pathway is required to prevent uracil accumulation in DNA. J. Biol. Chem. 286, 44015–44022 (2011).
Field, M. S., Kamynina, E., Chon, J. & Stover, P. J. Nuclear folate metabolism. Annu. Rev. Nutr. 38, 219–243 (2018).
Ginno, P. A., Burger, L., Seebacher, J., Iesmantavicius, V. & Schübeler, D. Cell cycle-resolved chromatin proteomics reveals the extent of mitotic preservation of the genomic regulatory landscape. Nat. Commun. 9, 4048 (2018).
Sudhamalla, B., Dey, D., Breski, M., Nguyen, T. & Islam, K. Site-specific azide-acetyllysine photochemistry on epigenetic readers for interactome profiling. Chem. Sci. 8, 4250–4256 (2017).
Perretta, M. & Romero, A. De novo RNA biosynthesis in isolated bone marrow nuclei. Experientia 29, 39–40 (1973).
Tan, J. L. et al. Stress from nucleotide depletion activates the transcriptional regulator HEXIM1 to suppress melanoma. Mol. Cell 62, 34–46 (2016).
Sharma, U. & Rando, O. J. Metabolic inputs into the epigenome. Cell Metab. 25, 544–558 (2017).
Li, X., Egervari, G., Wang, Y., Berger, S. L. & Lu, Z. Regulation of chromatin and gene expression by metabolic enzymes and metabolites. Nat. Rev. Mol. Cell Biol. 19, 563–578 (2018).
Boukouris, A. E., Zervopoulos, S. D. & Michelakis, E. D. Metabolic enzymes moonlighting in the nucleus: metabolic regulation of gene transcription. Trends Biochem. Sci. 41, 712–730 (2016).
van der Knaap, J. A. & Verrijzer, C. P. Undercover: gene control by metabolites and metabolic enzymes. Genes Dev. 30, 2345–2369 (2016).
Anderson, A. C. & Wright, D. L. Antifolate agents: a patent review (2010–2013). Expert Opin. Ther. Pat. 24, 687–697 (2014).
Wilson, P. M., Danenberg, P. V., Johnston, P. G., Lenz, H. J. & Ladner, R. D. Standing the test of time: targeting thymidylate biosynthesis in cancer therapy. Nat. Rev. Clin. Oncol. 11, 282–298 (2014).
Bürckstümmer, T. et al. A reversible gene trap collection empowers haploid genetics in human cells. Nat. Methods 10, 965–971 (2013).
Krzywinski, M. et al. Circos: an information aesthetic for comparative genomics. Genome Res. 19, 1639–1645 (2009).
Carette, J. E. et al. Global gene disruption in human cells to assign genes to phenotypes by deep sequencing. Nat. Biotechnol. 29, 542–546 (2011).
Ran, F. A. et al. Genome engineering using the CRISPR-Cas9 system. Nat. Protoc. 8, 2281–2308 (2013).
Kim, T. H. et al. A high-resolution map of active promoters in the human genome. Nature 436, 876–880 (2005).
Jiang, H., Lei, R., Ding, S. W. & Zhu, S. Skewer: a fast and accurate adapter trimmer for next-generation sequencing paired-end reads. BMC Bioinformatics 15, 182 (2014).
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Landt, S. G. et al. ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia. Genome Res. 22, 1813–1831 (2012).
Heinz, S. et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589 (2010).
McLean, C. Y. et al. GREAT improves functional interpretation of cis-regulatory regions. Nat. Biotechnol. 28, 495–501 (2010).
Ramírez, F. et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44, W160–W165 (2016).
Ross-Innes, C. S. et al. Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature 481, 389–393 (2012).
Trapnell, C. et al. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat. Biotechnol. 28, 511–515 (2010).
S. Sdelci is a JDRF postdoctoral fellow (no. 3-PDF-2014-206-A-N). D.L.B. is a Merck Fellow of the Damon Runyon Cancer Research Foundation (no. DRG-2196-14). Next-generation sequencing was performed by the Biomedical Sequencing Facility at the CeMM. Research in the Kubicek laboratory is supported by the Austrian Federal Ministry for Digital and Economic Affairs and the National Foundation for Research, Technology, and Development, the Austrian Science Fund (FWF; no. F4701) and the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (no. ERC-CoG-772437). Research in the Zuber laboratory was supported by the ERC (no. ERC-StG-336860 to J.Z.), a Research Fellowship of the EU (Marie Curie Actions 329492 to P.R.) and generous institutional funding from Boehringer Ingelheim. We thank all the members of the BioOptic Facility of the Research Institute of Molecular Pathology and the Institute of Molecular Biotechnology GmbH for their help with cell sorting, P. Stover (Cornell) and S. Nijman (Oxford) for kindly providing the plasmids, and A. Terenzi (Donostia International Physics Center) for advice on in vitro metabolomics.
Integrated supplementary information
a, Representative FACS panels showing REDS1, REDS2, REDS3 and REDS4 cells treated with 0.5 μM (S)-JQ1; an equal volume of DMSO was used as control. Shown are representative data from three biological replicates for each experimental condition. b, Representative cell cycle profiles evaluated by PI-staining and DNA content analysis by FACS. REDS1, REDS3 and REDS4 cells are compared to haploid WT-KBM7 (green profile). Three biological replicates were done for each experimental condition. c, Metaphase chromosome spreads in REDS clones. Nuclei were stained with DAPI; scale bar 10 μm. Three biological replicates were done for each experimental condition. d, Representation of the RFP integration site. RFP (red bar) is inserted in the antisense direction at chromosome 6 (chr6:20,520,542-20,588,419), in the first intron of the CDKAL1 gene (sense direction). e, Representative fluorescence in situ hybridization (FISH) images in REDS1 cells. RFP probe (red dots) stains the RFP insertion. DAPI (grey signal) was used to stain the nucleus. Yellow dashed lines mark nuclear perimeter. Scale bar 10 μm. The experiment was repeated three times with similar results. f, Representative live cell images of REDS1 treated with 1 μM (S)-JQ1 for 24 hours; an equal volume of DMSO was used as control. RFP expression is shown in red; Scale bar 100 μm. The experiment was repeated three times with similar results. g, Upper panel: BRD1, BRD2, BRD3, BRDT and BRD4 expression assessed by RT-PCR following shRNA-mediated downregulation of the respective protein in REDS1 cells. Two biological replicates were done for each experimental condition. Lower panel: Quantification of RFP positive cells from live images of BRD1, BRD2, BRD3, BRDT or BRD4 downregulated REDS1 cells. Two biological replicates were done for each experimental condition.
a, Representation of the gene-trap integration sites mapped on MDC1 and MTHFD1 genes. Red arrows indicate sense insertions; blue arrows indicate antisense insertions. b, Western Blot for MTHFD1 protein levels after downregulation with the indicated shRNAs in REDS3. Tubulin was used as loading control. The experiment was repeated twice with similar results. c, Validation of MTHFD1 in an alternative REDS clone described previously (Sdelci, S. et al. Mapping the chemical chromatin reactivation landscape identifies BRD4-TAF1 cross-talk. Nat Chem Biol 12, 504-10 (2016)). Quantification of RFP positive cells from live-cell imaging pictures of REDS3 cells treated with MTHFD1 shRNA. Two biological replicates were done for each experimental condition. d, Representative live-cell images of MTHFD1 knock down in REDS3 cells. RFP signal is shown in red; scale bar 100 μm.
a, Analysis of BRD4 and MTHFD1 abundance from interaction proteomics screen. b, Western blot showing GFP pull-down from HEK293T whole-cell extracts overexpressing GFP-MTHFD1 or GFP alone. Tubulin was used as loading control. The experiment was repeated twice with similar results. c, Western blot showing BRD4 pull-down from HeLa whole-cell extracts The experiment was repeated twice with similar results. d, Immunoprecipitation of MTHFD1 from the chromatin-bound fraction of HAP1 cells. BRD4 is validated as interactor by western blot and mass-spectrometry based proteomics. e, Network of Gene Ontology enrichment of all MTHFD1 interactors reveals gene expression and RNA processing as most enriched terms. Two biological replicates were performed, as shown in the western blot. Significant GO terms were selected using the Benjamini-Hochberg adjusted Fisher’s exact test p-value threshold < 1e-3. f, Primary structure of BRD4 and MTHFD1, and peptide microarray based binding evaluation of the corresponding overlapping peptide library with Flag-MTHFD1, GST-MTHFD1 and GST-BRD4. Elaboration of the effect of acetylated lysine residues (Kac) of MTHFD1 and positive control of peptide H4(2-11)K5K8 binding on GST-BRD4 binding. g, AlphaLISA for the interaction between BRD4 and an acetylated histone peptide in the presence of increasing amounts of recombinant MTHFD1. Mean ± SD from n = 2 independent samples. h, AlphaLISA assay with the indicated MTHFD1 acetylated peptides and GST-BRD4 (full length). Two replicates for each condition. i, Titration of MTHFD1 peptides unmodified and K56 acetylated for their ability to disrupt the interaction of GST-BRD4 with an acetylated histone peptide. Mean ± SD from n = 2 independent samples.
a, MTHFD1 and BRD4 constructs used in immunoprecipitation experiments. b, FLAG and GFP immunoprecipitation from HEK293 cells overexpressing the indicated constructs shows interaction of MTHFD1 with all three BRD4 isoforms. GFP immunoprecipitation from HEK293 cells overexpressing the indicated constructs shows interaction of BRD4 with full-length MTHFD1 but not with the individual domains of the protein. The experiment was repeated three times with similar results. c, GFP immunoprecipitation from HEK293 cells overexpressing the indicated constructs shows interaction of MTHFD1(K56A) and decreased interaction of MTHFD1(K56R) with the short isoform of BRD4. Similarly, the interaction is impaired in the BRD4 double bromodomain mutant. The experiment was repeated three times with similar results. Scale bar 20 μm.
a, Genomic scheme of the MTHFD1 locus and characterization of the editing events in the three knock-out cell lines. b, Propidium iodide for DNA content (x-axis) related to cell number (y-axis) indicating cell cycle distribution in HAP1 WT and MTHFD1 knock-out cells. The experiment was repeated three times with similar results. c, Cell proliferation analysis by dye dilution profiling in HAP1 WT and MTHFD1 knock-out cells. The experiment was repeated three times with similar results.
ChIP-seq results for BRD4 and MTHFD1 immunoprecipitation in HAP1 WT (WT) and MTHFD1 knock-out (KO) cells. a, Number of ChIP-seq peaks detected for each protein in WT cells. b, Amount of overlap between peaks of MTHFD1, BRD4 or H3K27ac in WT or KO cells measured by the fractional overlap, total intersection or the Jaccard index (intersection over the union of peaks). c, GREAT enrichment analysis for binding sites of BRD4 and MTHFD1 in HAP1 WT cells. The hypergeometric test and Benjamini Hochberg multiple test correction were employed. d, Volcano and scatterplots comparing binding in KO to WT cells, with significantly bound sites (FDR adjusted p-value < 0.1 and absolute log2 fold change > 1) marked in red. Differential regions were discovered with the Wald test using the DiffBind package with n = 2 biologically independent experiments. P-values were adjusted with the Benjamini-Hochberg method. e, Normalized ChIP-seq signal in differential regions bound by MTHFD1 or BRD4; f, Global differences in differentially bound MTHFD1 or BRD4 sites upon MTHFD1 KO. A two-sided Mann-Whitney U-test was used to assess significance.
a, Representative genome browser view of BRD4 and MTHFD1 binding in the H3K27ac-marked promoters of KEAP1 (left) and TFAP4 (right). All ChIP tracks were normalized to total mapped reads and the respective IgG control was subtracted from the merged replicate tracks. b, Enrichment of BRD4 and MTHFD1 ChIP signal in H3K27ac peaks. Peaks were sorted by total signal abundance and data represent merged replicates normalized to 1X genome coverage. c, Volcano and scatterplots comparing binding in dBET6- or DMSO-treated cells, with significantly bound sites (FDR adjusted p-value < 0.1 and absolute log2 fold change > 1) marked in red. Differential regions were discovered with the Wald test using the DiffBind package with n = 2 biologically independent experiments. P-values were adjusted with the Benjamini-Hochberg method. d, Normalized ChIP-seq signal in differential regions bound by MTHFD1 or BRD4 upon dBET6 treatment. e, Quantification of global differences in differentially bound MTHFD1 or BRD4 sites upon dBET6 treatment. A two-sided Mann-Whitney U-test was used to assess significance. f, Number of ChIP-seq peaks detected for each protein in WT cells. g, Amount of overlap between peaks of MTHFD1, BRD4 or H3K27ac in dBET6- or DMSO-treated cells measured by the fractional overlap, total intersection or the Jaccard index (intersection over the union of peaks).
Supplementary Fig. 8 Gene expression changes following perturbation of the BRD4 and folate pathways.
a, Gene expression levels of folate, purine and pyrimidine pathway genes as measured by RNA-seq, shown as absolute log(RPKM) values and row-wise Z scores (across all samples). b, Pearson correlation of ChIP-seq signal and basal gene expression levels for differentially expressed genes bound by BRD4 (n = 825) or MTHFD1 (n = 773) in WT HAP1 cells. c, Association of MTHFD1 and BRD4 bound loci with high basal gene expression levels in HAP1 cells. d, Amount of normalized ChIP-seq signal in genes bound by MTHFD1 and BRD4 depending on the direction of transcriptional change upon perturbation (up- or down-regulated). For panels c) and d), the number of genes differentially expressed in each condition is as described in Fig. 3f. For c) and d) boxplot boxes represent interquartile range with center on median, while whiskers represent values 1.5 times outside the respective interquartile range.
a, Heat map of relative transcription changes in K-562 cells treated with with 0.1 μM dBET6, 1 μM (S)-JQ1, 1 μM MTX, shRNAs targeting BRD4 or MTHFD1 alone or in combination. Equal amount of DMSO, or non-targeting hairpins were used as respective controls. b, Heat map matrix of relative transcription changes in A549 treated with 0.1 μM dBET6, 1 μM (S)-JQ1, 1 μM MTX, shRNAs targeting BRD4 or MTHFD1 alone or in combination. Equal amount of DMSO, or non-targeting hairpins were used as respective controls.
a, Scatter plot depicting changes in nuclear metabolite levels following BRD4 or MTHFD1 downregulation for all metabolites (upper panel) and folate-dependent metabolites (lower panel). Means from two biological replicates. r-value represents the Pearson Correlation Coefficient. b, Heat maps representing metabolite changes in the pyrimidine, purine and methionine biosynthetic pathways upon downregulation of BRD4 or MTHFD1 by shRNA. Two biological replicates were done for each experimental condition. c, Metabolite Set Enrichment Analysis (MSEA) showing a comparison of the enrichment of metabolite sets that are significant in both knockdowns (shBRD4: 1; shMTHFD1: 2), metabolite sets exclusively enriched in shMTHFD1 or metabolite sets exclusively enriched in shBRD4. The length of each bar represents fold change of enrichment compared to random expectation, colored by one-sided Fisher’s exact test p-value. d, Correlation plot depicting changes in nuclear metabolite levels following dBET1 or MTX treatment of HAP1 cells for 24 h. Means from two biological replicates. r-value represents the Pearson Correlation Coefficient. e, Heatmaps showing relative changes in folate metabolites levels in the of nuclear and cytosolic fraction of HAP1 cells treated with 1 μM Mitomycin C, Actinomycin D, Bortexomib, MTX and (S)-JQ1, 0.5 μM dBET6 or 12.5 μM Cyclohexamide for 6 hours. Equal amount of DMSO was used as control, two biological replicates were done for each experimental condition. f, LC-MS-MS peak areas for different folate metabolites in HAP1 WT and MTHFD1 knock-out cells after 2 h treatment with 13C formate. Two biological replicates were done for each experimental condition.
HAP MTHFD1 knock-out cells were transfected with constructs for the overexpression of full-length MTHFD1 either wild-type (WT) or harboring a nuclear localization (NLS) or nuclear export (NES) signal. a, Quantification of the nuclear MTHFD1 levels in the different conditions. Two biological replicates were done for each experimental condition. b, Effects of MTHFD1 reconstitution on cell proliferation at 48h and 96 hours post transfection. Two biological replicates were done for each experimental condition. c, Representative immunofluorescence microscopy images for MTHFD1 localization. Representative for two independent experiments. d, Western blot for nuclear and cytoplasmic MTHFD1 and FLAG levels in the MTHFD1 knock-out and reconstituted cells. Representative for two independent experiments. e, Quantification of nuclear (N) and cytoplasmic (C) folate levels in the different conditions. Two biological replicates were done for each experimental condition.
a, Table of IC50 values for (S)-JQ1 and MTX as reported in the Welcome Trust-Genomics of Drug Sensitivity in Cancer database (WT; http://www.cancerrxgene.org), or produced in house (right two columns), for the different cell lines. b, Determination of IC50 values for (S)-JQ1 (red) and MTX (dark red) in the indicated cell lines. Three biological replicates were done for each experimental condition (mean ± SD). c, Dose response matrices displaying cell viability of HeLa and KBM7 treated for 72 h with (S)-JQ1 and MTX alone or in combination. Means from two biological replicates. d, Deviations from Bliss additivity for drug matrices from Fig. 4c and Supplementary Fig. 6c. e, Knock-down of MTHFD1 in A549 cells followed by 72 h treatment with increasing concentrations of (S)-JQ1. Mean ± SD from n = 2 independent samples. f, Induction of caspase activity in HAP1 WT and MTHFD1 knock-out cells treated with for 72 h with (S)-JQ1 or MTX. Mean ± SD from n = 2 independent samples.
Full scans of all western blots in the manuscript.