Letter | Published:

Activity-dependent neuroprotective protein recruits HP1 and CHD4 to control lineage-specifying genes

Naturevolume 557pages739743 (2018) | Download Citation


De novo mutations in ADNP, which encodes activity-dependent neuroprotective protein (ADNP), have recently been found to underlie Helsmoortel–Van der Aa syndrome, a complex neurological developmental disorder that also affects several other organ functions1. ADNP is a putative transcription factor that is essential for embryonic development2. However, its precise roles in transcriptional regulation and development are not understood. Here we show that ADNP interacts with the chromatin remodeller CHD4 and the chromatin architectural protein HP1 to form a stable complex, which we refer to as ChAHP. Besides mediating complex assembly, ADNP recognizes DNA motifs that specify binding of ChAHP to euchromatin. Genetic ablation of ChAHP components in mouse embryonic stem cells results in spontaneous differentiation concomitant with premature activation of lineage-specific genes and in a failure to differentiate towards the neuronal lineage. Molecularly, ChAHP-mediated repression is fundamentally different from canonical HP1-mediated silencing: HP1 proteins, in conjunction with histone H3 lysine 9 trimethylation (H3K9me3), are thought to assemble broad heterochromatin domains that are refractory to transcription. ChAHP-mediated repression, however, acts in a locally restricted manner by establishing inaccessible chromatin around its DNA-binding sites and does not depend on H3K9me3-modified nucleosomes. Together, our results reveal that ADNP, via the recruitment of HP1 and CHD4, regulates the expression of genes that are crucial for maintaining distinct cellular states and assures accurate cell fate decisions upon external cues. Such a general role of ChAHP in governing cell fate plasticity may explain why ADNP mutations affect several organs and body functions and contribute to cancer progression1,3,4. Notably, we found that the integrity of the ChAHP complex is disrupted by nonsense mutations identified in patients with Helsmoortel–Van der Aa syndrome, and this could be rescued by aminoglycosides that suppress translation termination5. Therefore, patients might benefit from therapeutic agents that are being developed to promote ribosomal read-through of premature stop codons6,7.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    Helsmoortel, C. et al. A SWI/SNF-related autism syndrome caused by de novo mutations in ADNP. Nat. Genet. 46, 380–384 (2014).

  2. 2.

    Pinhasov, A. et al. Activity-dependent neuroprotective protein: a novel gene essential for brain formation. Brain Res. Dev. Brain Res. 144, 83–90 (2003).

  3. 3.

    Gozes, I., Yeheskel, A. & Pasmanik-Chor, M. Activity-dependent neuroprotective protein (ADNP): a case study for highly conserved chordata-specific genes shaping the brain and mutated in cancer. J. Alzheimers Dis. 45, 57–73 (2015).

  4. 4.

    Gozes, I. et al. The compassionate side of neuroscience: Tony Sermone’s undiagnosed genetic journey–ADNP mutation. J. Mol. Neurosci. 56, 751–757 (2015).

  5. 5.

    Hermann, T. Aminoglycoside antibiotics: old drugs and new therapeutic approaches. Cell. Mol. Life Sci. 64, 1841–1852 (2007).

  6. 6.

    Peltz, S. W., Morsy, M., Welch, E. M. & Jacobson, A. Ataluren as an agent for therapeutic nonsense suppression. Annu. Rev. Med. 64, 407–425 (2013).

  7. 7.

    Welch, E. M. et al. PTC124 targets genetic disorders caused by nonsense mutations. Nature 447, 87–91 (2007).

  8. 8.

    Zamostiano, R. et al. Cloning and characterization of the human activity-dependent neuroprotective protein. J. Biol. Chem. 276, 708–714 (2001).

  9. 9.

    Bassan, M. et al. Complete sequence of a novel protein containing a femtomolar-activity-dependent neuroprotective peptide. J. Neurochem. 72, 1283–1293 (1999).

  10. 10.

    Mandel, S., Rechavi, G. & Gozes, I. Activity-dependent neuroprotective protein (ADNP) differentially interacts with chromatin to regulate genes essential for embryogenesis. Dev. Biol. 303, 814–824 (2007).

  11. 11.

    Niwa, H. Mouse ES cell culture system as a model of development. Dev. Growth Differ. 52, 275–283 (2010).

  12. 12.

    Flemr, M. & Bühler, M. Single-step generation of conditional knockout mouse embryonic stem cells. Cell Reports 12, 709–716 (2015).

  13. 13.

    Molkentin, J. D. The zinc finger-containing transcription factors GATA-4, -5, and -6. Ubiquitously expressed regulators of tissue-specific gene expression. J. Biol. Chem. 275, 38949–38952 (2000).

  14. 14.

    Fujikura, J. et al. Differentiation of embryonic stem cells is induced by GATA factors. Genes Dev. 16, 784–789 (2002).

  15. 15.

    Cho, L. T. Y. et al. Conversion from mouse embryonic to extra-embryonic endoderm stem cells reveals distinct differentiation capacities of pluripotent stem cell states. Development 139, 2866–2877 (2012).

  16. 16.

    Bibel, M. et al. Differentiation of mouse embryonic stem cells into a defined neuronal lineage. Nat. Neurosci. 7, 1003–1009 (2004).

  17. 17.

    Mosch, K., Franz, H., Soeroes, S., Singh, P. B. & Fischle, W. HP1 recruits activity-dependent neuroprotective protein to H3K9me3 marked pericentromeric heterochromatin for silencing of major satellite repeats. PLoS ONE 6, e15894 (2011).

  18. 18.

    Mandel, S. & Gozes, I. Activity-dependent neuroprotective protein constitutes a novel element in the SWI/SNF chromatin remodeling complex. J. Biol. Chem. 282, 34448–34456 (2007).

  19. 19.

    Vermeulen, M. & et al. Quantitative interaction proteomics and genome-wide profiling of epigenetic histone marks and their readers. Cell 142, 967–980 (2010).

  20. 20.

    de Dieuleveult, M. et al. Genome-wide nucleosome specificity and function of chromatin remodellers in ES cells. Nature 530, 113–116 (2016).

  21. 21.

    Bannister, A. J. et al. Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410, 120–124 (2001).

  22. 22.

    Lachner, M., O’Carroll, D., Rea, S., Mechtler, K. & Jenuwein, T. Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins. Nature 410, 116–120 (2001).

  23. 23.

    Buenrostro, J. D., Giresi, P. G., Zaba, L. C., Chang, H. Y. & Greenleaf, W. J. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. Methods 10, 1213–1218 (2013).

  24. 24.

    Domcke, S. et al. Competition between DNA methylation and transcription factors determines binding of NRF1. Nature 528, 575–579 (2015).

  25. 25.

    Mohn, F. et al. Lineage-specific polycomb targets and de novo DNA methylation define restriction and potential of neuronal progenitors. Mol. Cell 30, 755–766 (2008).

  26. 26.

    Knuckles, P. et al. RNA fate determination through cotranscriptional adenosine methylation and microprocessor binding. Nat. Struct. Mol. Biol. 24, 561–569 (2017).

  27. 27.

    Jacobs, S. A. & Khorasanizadeh, S. Structure of HP1 chromodomain bound to a lysine 9-methylated histone H3 tail. Science 295, 2080–2083 (2002).

  28. 28.

    Bibel, M., Richter, J., Lacroix, E. & Barde, Y.-A. Generation of a defined and uniform population of CNS progenitors and neurons from mouse embryonic stem cells. Nat. Protocols 2, 1034–1043 (2007).

  29. 29.

    Cox, J. et al. Andromeda: a peptide search engine integrated into the MaxQuant environment. J. Proteome Res. 10, 1794–1805 (2011).

  30. 30.

    Cox, J. et al. Accurate proteome-wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol. Cell. Proteomics 13, 2513–2526 (2014).

  31. 31.

    Hubner, N. C. et al. Quantitative proteomics combined with BAC TransgeneOmics reveals in vivo protein interactions. J. Cell Biol. 189, 739–754 (2010).

  32. 32.

    Tyanova, S. et al. The Perseus computational platform for comprehensive analysis of (prote)omics data. Nat. Methods 13, 731–740 (2016).

  33. 33.

    Tusher, V. G., Tibshirani, R. & Chu, G. Significance analysis of microarrays applied to the ionizing radiation response. Proc. Natl Acad. Sci. USA 98, 5116–5121 (2001).

  34. 34.

    Schwanhäusser, B. et al. Global quantification of mammalian gene expression control. Nature 473, 337–342 (2011).

  35. 35.

    MacLean, B. et al. Skyline: an open source document editor for creating and analyzing targeted proteomics experiments. Bioinformatics 26, 966–968 (2010).

  36. 36.

    Abdulrahman, W. et al. A set of baculovirus transfer vectors for screening of affinity tags and parallel expression strategies. Anal. Biochem. 385, 383–385 (2009).

  37. 37.

    Rosenbloom, K. R. et al. The UCSC Genome Browser database: 2015 update. Nucleic Acids Res. 43, D670–D681 (2015).

  38. 38.

    Langmead, B., Trapnell, C., Pop, M. & Salzberg, S. L. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10, R25 (2009).

  39. 39.

    Gaidatzis, D., Lerch, A., Hahne, F. & Stadler, M. B. QuasR: quantification and annotation of short reads in R. Bioinformatics 31, 1130–1132 (2015).

  40. 40.

    Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139–140 (2010).

  41. 41.

    Young, M. D., Wakefield, M. J., Smyth, G. K. & Oshlack, A. Gene ontology analysis for RNA-seq: accounting for selection bias. Genome Biol. 11, R14 (2010).

  42. 42.

    Zhang, Y. et al. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 9, R137 (2008).

  43. 43.

    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).

  44. 44.

    Ramírez, F. et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44 (W1), W160–W165 (2016).

  45. 45.

    Whyte, W. A. et al. Enhancer decommissioning by LSD1 during embryonic stem cell differentiation. Nature 482, 221–225 (2012).

  46. 46.

    Tiwari, V. K. et al. A chromatin-modifying function of JNK during stem cell differentiation. Nat. Genet. 44, 94–100 (2011).

  47. 47.

    Karmodiya, K., Krebs, A. R., Oulad-Abdelghani, M., Kimura, H. & Tora, L. H3K9 and H3K14 acetylation co-occur at many gene regulatory elements, while H3K14ac marks a subset of inactive inducible promoters in mouse embryonic stem cells. BMC Genomics 13, 424 (2012).

  48. 48.

    Liu, N. et al. Recognition of H3K9 methylation by GLP is required for efficient establishment of H3K9 methylation, rapid target gene repression, and mouse viability. Genes Dev. 29, 379–393 (2015).

  49. 49.

    Mikkelsen, T. S. et al. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448, 553–560 (2007).

  50. 50.

    Vizcaíno, J. A. et al. 2016 update of the PRIDE database and its related tools. Nucleic Acids Res. 44 (D1), D447–D456 (2016).

  51. 51.

    Kwon, S. H. & Workman, J. L. The heterochromatin protein 1 (HP1) family: put away a bias toward HP1. Mol. Cells 26, 217–227 (2008).

  52. 52.

    Murzina, N., Verreault, A., Laue, E. & Stillman, B. Heterochromatin dynamics in mouse cells: interaction between chromatin assembly factor 1 and HP1 proteins. Mol. Cell 4, 529–540 (1999).

  53. 53.

    Maison, C. et al. The SUMO protease SENP7 is a critical component to ensure HP1 enrichment at pericentric heterochromatin. Nat. Struct. Mol. Biol. 19, 458–460 (2012).

  54. 54.

    Smothers, J. F. & Henikoff, S. The HP1 chromo shadow domain binds a consensus peptide pentamer. Curr. Biol. 10, 27–30 (2000).

Download references


We thank A. Peters and D. Schübeler for discussion and feedback on the manuscript, and H. Pickersgill for editing assistance. We thank Y. Shimada and N. Laschet for technical support. We would like to thank the FMI Functional Genomics facility for assistance in library construction and next generation sequencing, and J. Seebacher for discussions. This work was supported by funds from the Swiss National Science Foundation (SNF).

Reviewer information

Nature thanks P. Wade and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

Author notes

  1. These authors contributed equally: Sarah H. Carl, Anja Basters.


  1. Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland

    • Veronika Ostapcuk
    • , Fabio Mohn
    • , Sarah H. Carl
    • , Anja Basters
    • , Daniel Hess
    • , Vytautas Iesmantavicius
    • , Lisa Lampersberger
    • , Matyas Flemr
    • , Aparna Pandey
    • , Nicolas H. Thomä
    • , Joerg Betschinger
    •  & Marc Bühler
  2. University of Basel, Basel, Switzerland

    • Veronika Ostapcuk
    • , Aparna Pandey
    •  & Marc Bühler
  3. Swiss Institute of Bioinformatics, Basel, Switzerland

    • Sarah H. Carl
  4. University of Vienna, Vienna, Austria

    • Lisa Lampersberger


  1. Search for Veronika Ostapcuk in:

  2. Search for Fabio Mohn in:

  3. Search for Sarah H. Carl in:

  4. Search for Anja Basters in:

  5. Search for Daniel Hess in:

  6. Search for Vytautas Iesmantavicius in:

  7. Search for Lisa Lampersberger in:

  8. Search for Matyas Flemr in:

  9. Search for Aparna Pandey in:

  10. Search for Nicolas H. Thomä in:

  11. Search for Joerg Betschinger in:

  12. Search for Marc Bühler in:


V.O. designed and performed experiments, analysed data, generated cell lines and prepared figures. M.F. generated cell lines and advised on experimental design. D.H. and V.I. acquired and analysed mass spectrometry data. Bioinformatic and computational analysis was performed by S.H.C. and F.M. A.B. performed in vitro biochemistry experiments. V.O. and L.L. performed immunoprecipitation and mass spectrometry with patient-specific Adnp mutants, and L.L. performed ChIP experiments for the patient-specific Adnp mutants. A.P. performed ChIP experiments for the Cbx3 chromodomain mutant. F.M. generated and analysed ATAC-seq data. N.H.T. and J.B. advised on the experimental design. M.B. conceived and supervised the study, and secured funding. M.B. and F.M. wrote the manuscript. All authors discussed the results and commented on the manuscript.

Competing interests

The authors declare competing financial interests: a patent application has been filed (EP17191642.2). The Friedrich Miescher Institute for Biomedical Research (FMI) receives significant financial contributions from the Novartis Research Foundation. Published research reagents from the FMI are shared with the academic community under a Material Transfer Agreement (MTA) having terms and conditions corresponding to those of the UBMTA (Uniform Biological Material Transfer Agreement).

Corresponding author

Correspondence to Marc Bühler.

Extended data figures and tables

  1. Extended Data Fig. 1 Generation of isogenic mouse ES cell lines to interrogate protein–protein and protein–chromatin interactions.

    a, Mouse ES cells expressing the BirA biotin ligase from the Rosa26 locus were used as a parental cell line for endogenous gene tagging with the Flag-AviTag12. For a full list of mouse ES cell lines used in this study (cMB#), see Supplementary Information. b, Top, scheme depicting Flag-AviTag (not drawn to scale) insertion at the endogenous Adnp locus. Arrow indicates transcription start site. Boxes represent exons. Bottom, scheme depicting ADNP protein. Protein domains as predicted by InterPro. Nonsense (NS) and frameshift (FS) mutations found in children with Helsmoortel–Van der Aa syndrome are indicated (https://www.adnpkids.com, dated March 2017). Numbers denote amino acids. c, Sanger sequencing of AdnpFlag-AviTag/Flag-AviTag cell lines. d, Distribution of ADNP-bound genomic sites with respect to protein-coding genes. Peaks were called from ChIP–seq data acquired from three independent biological replicates (that is, three independent AdnpFlag-AviTag/Flag-AviTag mouse ES cell lines). Horizontal bars represent peaks annotated to individual categories, and vertical bars represent peaks annotated jointly to specified combinations of categories. DIG, distal intergenic; UTR, untranslated region. e, ChIP–seq profiles at two lineage-specifying gene loci that were generated from three independent AdnpFlag-AviTag/Flag-AviTag ES lines.

  2. Extended Data Fig. 2 Generation and analysis of isogenic Adnp knockout mouse ES cell lines.

    a, Scheme depicting CRISPR–Cas9 and TALEN-induced double-stranded DNA breaks to delete the Adnp open-reading frame. TSS, transcription start site. b, PCR genotyping confirming homozygous deletion of the Adnp open-reading frame in three different mouse ES cell lines used in this study. The experiment was performed twice. c, MA plot comparing fold change (FC) in gene expression for Adnp−/− versus Adnp+/+ cells (y axis) with mean mRNA abundance (x axis). Representative endoderm-specific genes are highlighted in red. Dashed red lines indicate fourfold up- or downregulation. CPM, counts per million. d, Gene Ontology enrichment analysis of genes upregulated in Adnp−/− cells. n = 3 independent biological replicates. e, Scatterplot comparing gene expression fold change upon Adnp knockout (y axis) with expression changes between extraembryonic endoderm (eXEN) and ES cells (x axis). Known key lineage markers are indicated in blue.

  3. Extended Data Fig. 3 The HP1 interactome of mouse ES cells.

    a, Isogenic mouse ES cell lines expressing endogenously tagged CHD4 and HP1 proteins. Western blot demonstrating expression of FlagAviTag-tagged proteins. The high molecular mass of CHD4 (218 kDa) does not allow discernable separation of tagged from non-tagged protein. See Supplementary Information for detailed genotype descriptions of the individual ES cell lines. For gel source data, see Supplementary Fig. 1. Experiments were performed twice. b, TAP–LC–MS/MS of endogenously FlagAviTag-tagged HP1α. Protein purification was performed in the presence of 350 mM NaCl. Parental ES cell line serves as background control. n = 3 independent biological replicates (that is, three independent Cbx5Flag-AviTag/Flag-AviTag mouse ES cell lines). c, TAP–LC–MS/MS of endogenously Flag-AviTag-tagged HP1β. Protein purification was performed in the presence of 350 mM NaCl. Parental ES cell line serves as background control. n = 3 independent biological replicates (that is, three independent Cbx1Flag-AviTag/Flag-AviTag mouse ES cell lines). d, TAP–LC–MS/MS of endogenously Flag-AviTag-tagged HP1γ. Protein purification was performed in the presence of 350 mM NaCl. Parental mouse ES cell line serves as background control. n = 3 independent biological replicates (that is, three independent Cbx1Flag-AviTag/Flag-AviTag mouse ES cell lines) e, Heat map showing the variation in co-purifying (Z-score) proteins across HP1 isoform-specific TAP–LC–MS/MS experiments. Proteins that were significantly enriched in at least one experiment (bd) were included in the analysis. ae, Validating our approach, all three HP1 isoforms co-precipitated a large number of proteins. Many of these were common to all three HP1 proteins and have previously been described51. We also observed several proteins that interacted uniquely with specific isoforms, such as the previously identified CAF-1 or SENP7 interactions with HP1α52,53. f, Heat map visualization of Pearson’s correlation coefficients for the individual HP1 isoform-specific TAP–LC–MS/MS experiments. Three independent biological replicates for HP1α and HP1γ, two biological and one technical replicate for HP1β, and three technical replicates for the parental cell line. g, iBAQ values of ADNP and CHD4 in HP1 isoform-specific TAP–LC–MS/MS experiments. Three independent biological replicates for HP1α and HP1γ, two biological and one technical replicate for HP1β. Centre value denotes the mean; error bars denote s.d. bg, Statistical analysis was performed using Perseus (see Methods). Mass spectrometry raw data are deposited with ProteomeXchange.

  4. Extended Data Fig. 4 In vitro characterization of ChAHP complex composition.

    a, Strep-tag pull-down assays with recombinant human proteins overexpressed in Hi5 insect cells, revealing that ADNP binds to both CHD4 and HP1γ, whereas CHD4 and HP1γ do not interact directly. b, SEC of the recombinant ChAHP complex. ChAHP was reconstituted from Hi5 insect cells and further purified by separation according to its molecular mass on a HiLoad Superdex 300 column. Largest fractions eluting first contain ChAHP (1), followed by ADNP–HP1γ (2) and HP1γ alone (3). c, Fractions from b were separated on SDS–PAGE and visualized by Coomassie staining. d, Pull-down analysis of Strep-tagged HP1γ (S-HP1γ) with full-length or N-terminally truncated ADNP (ΔΝ228) or CHD4. e, Pull-down analysis of Strep-tagged full-length or N-terminally truncated ADNP. be, Note that N-terminally truncated ADNP does not co-elute with CHD4 on SEC (b, c). This is confirmed by pull-down experiments (d, e), which show that ADNP lacking the first 228 amino acids is only able to bind to HP1γ but no longer to CHD4. Thus, ADNP contacts CHD4 through its N terminus. f, Pull-down analysis of His-tagged (H) full-length HP1γ, and isolated chromodomain (CD) and chromoshadow domain (CSD). Similar to other proteins containing the conserved PXVXL pentapeptide54, ADNP directly interacts with the CSD of HP1γ. This is consistent with the previously reported interaction of ADNP with HP1α17. The chromodomain of HP1γ does not bind to ADNP. Experiments in af were performed at least twice. S denotes the streptavidin tag added to the respective protein; asterisks denote a common contaminating protein in streptavidin pull-down assays. For gel source data, see Supplementary Fig. 1.

  5. Extended Data Fig. 5 HP1 occupancy at ChAHP-binding sites.

    a, Subunit occupancy at ChAHP-bound sites displayed as meta-profile integrating signal of all peaks. b, Genome browser screen shot of the Igfbp4 locus. ChIP–seq tracks represent depth-normalized read counts of averaged replicate experiments. The predicted ADNP DNA-binding motif upstream of the Igfbp4 transcription start site is shown. c, Average HP1β and HP1γ ChIP–seq enrichment on ChAHP-bound sites in wild-type cells, and average HP1β and HP1γ ChIP–seq enrichment on ChAHP-bound sites in Cbx3−/− and Cbx1−/− mouse ES cell lines, respectively. n = 2 biological replicates (that is, independently tagged mouse ES cell lines). d, Average HP1α ChIP–seq enrichment on ChAHP-bound sites in wild-type and Cbx1−/−Cbx3−/− double-knockout ES cell lines. n = 2 biological replicates. e, Histone modifications associated with heterochromatin are absent at ChAHP-bound sites. f, Histone modifications associated with active transcription are absent at ChAHP-bound sites. e, f, Histone modification profiles are displayed as meta-profile integrating signal over all peaks. g, Binding of wild-type and chromodomain mutant HP1γ to ChAHP targets (Igfbp4 and Exd1), an H3K9me3-modified region next to an L1 repeat (L1 chr4) and an inactive promoter of an unrelated gene (Hoxc5), quantified by ChIP–qPCR. Fold enrichment was normalized to an intergenic control region devoid of HP1γ and H3K9me3. Wild-type (grey) and mutant (red) HP1γ constructs were transiently transfected into HP1 triple-knockout ES cells in biological duplicates. Note the decrease of HP1γ binding at the H3K9me3-modified region (L1 chr4), whereas ChAHP targets remain unaffected in the chromodomain mutant (CDmut) that can no longer bind to H3K9me3. Black lines indicate average enrichments.

  6. Extended Data Fig. 6 Motif analysis of ADNP-bound loci.

    a, ADNP DNA-binding motifs predicted by HOMER. Frequency of occurrence and P values for motif enrichment compared to genomic background are indicated. n = 3 independent cell lines. b, Analysis of co-occurrence of the top-ten scoring ADNP DNA motifs. The bar graph shows the frequency of peaks containing the top-scoring ADNP motif and up to nine additional motifs as indicated on the x axis. Note that most peaks besides the GCCCCCTGGAG motif also contain more than five other sequence motifs out of the top-ten list.

  7. Extended Data Fig. 7 Different HP1 isoforms can functionally substitute each other.

    a, Scatterplots comparing mRNA expression changes after deletion of Adnp versus single, double or triple deletions of Cbx genes measured by RNA-seq. Green trend lines indicate a loess (locally weighted scatterplot smoothing) regression. n = 3 biological replicates (that is, three independent Adnp−/−, Cbx1−/−, Cbx3−/−, Cbx5−/−, Cbx1−/−Cbx3−/− double knockout, or Cbx1−/−Cbx3−/−Cbx5−/− triple knockout mouse ES cell lines). b, MA plot displaying fold changes in gene expression for individual Cbx knockout cell lines versus wild type. x axis denotes the mean mRNA abundance, log2(counts per million); y axis denotes the log2(fold change) between knockout and wild type. Dashed red lines indicate fourfold up- or downregulation. c, UCSC genome browser shots of three lineage-specifying genes. RNA-seq profiles normalized by library size of representative wild-type and mutant ES cell lines are shown. Experiments were performed three times. d, Gene Ontology enrichment analysis of the genes that are upregulated in Cbx1−/−Cbx3−/−Cbx5−/− triple-knockout (TKO) and Adnp−/− knockout (KO) cells (orange group of genes in a), and of the genes that are upregulated in Cbx1−/−Cbx3−/−Cbx5−/− triple-knockout but not Adnp−/− knockout cells (grey group of genes in a). See also Supplementary Table 5. n = 3 independent cell lines. e, RNA-seq library statistics showing fraction of uniquely, multi- and non-mapping reads. Note the increase in multi-mappers in the HP1 triple-knockout cells. f, Quantification of reads mapping to the major repeat classes in counts per million mappable reads. g, Quantification of reads mapping to the different LINE and LTR elements in counts per million mappable reads. All mutant cell lines were derived from the same parental mouse ES cell line through direct genome editing and are therefore isogenic.

  8. Extended Data Fig. 8 A patient-specific nonsense mutation in Adnp impairs the interaction with HP1 but not with DNA.

    a, Scheme depicting the wild-type and mutant Adnp alleles, which code for Tyr (blue) and a patient-specific premature termination codon (red) at amino acid position 718, respectively. Full-length and truncated protein products are shown on the right. Arrow indicates transcription start site. Boxes represent exons. Numbers denote amino acids. b, N-terminally Flag-AviTag-tagged ADNPPTC718 was streptavidin-purified from cells with and without aminoglycoside treatment (gentamycin or paromomycin) and subjected to LC–MS/MS analysis. ADNPPTC718-expressing cells were treated with 2 mg ml−1 gentamycin (2.9 mM) or paromomycin (3.2 mM) for 24 h. The table depicts total spectral counts, unique peptides and percentage sequence coverage (derived from Scaffold) for all ChAHP components from the different treatments. c, qRT–PCR measurement of Bmp1 and Igfbp4 mRNA levels in ES cells expressing full-length Adnp (Adnp+/+) or C-terminally truncated Adnp that interacts with CHD4 but not with HP1 (AdnpPTC718/PTC718). n = 3 biological replicates (that is, three independent RNA isolations). P values were calculated using two-tailed unpaired unequal variances t-tests. Centre value denotes the mean; error bars denote s.d. d, ChIP–qPCR enrichments for transiently transfected Flag-AviTag-tagged wild-type ADNP and ADNPPTC718 constructs on two ADNP targets, normalized to an intergenic control. Black lines indicate means. e, C-terminally Flag-AviTag-tagged ADNPPTC718 was streptavidin-purified from cells with or without gentamycin treatment (2.9 mM) and subjected to LC–MS/MS analysis. Bold letters indicate unique peptides further quantified by parallel reaction monitoring (PRM). C-terminal peptides encoded downstream of PTC718 are shown in colour. Dashed box denotes the HP1 interaction motif. f, Summed fragment intensities of five C-terminal ADNP peptides that are encoded downstream of PTC718 are shown on the left. Background proteins shown on the right serve as loading controls. Intensities were measured by PRM. Total spectrum counts were derived from Scaffold. n = 3 biological replicates.

  9. Extended Data Fig. 9 ChAHP and NuRD are distinct protein complexes.

    a, Single-step purification followed by LC–MS/MS of endogenously Flag-AviTag-tagged CHD4 and ADNP. Protein purification was performed in the presence of 350 mM NaCl. Proteins that interact predominantly with CHD4 or ADNP are indicated by UniProt names. NuRD complex components are labelled in green. n = 3 biological replicates (that is, three independent Chd4Flag-AviTag/Flag-AviTag and AdnpFlag-AviTag/Flag-AviTag ES cell lines). Statistical analysis was done with Perseus (see Methods for details). Mass spectrometry raw data are deposited with ProteomeXchange. b, SEC of nuclear protein extracts from AdnpFlag-AviTag/Flag-AviTag ES cells. Each fraction (indicated at the bottom) was resolved by SDS–PAGE and immunoblotted with the indicated antibodies. Molecular mass of individual proteins is indicated on the left. For gel source data, see Supplementary Fig. 1. Experiment was performed twice.

  10. Extended Data Fig. 10 Isogenic Cbx knockout ES cell lines.

    a, Western blot demonstrating depletion of HP1α protein in Cbx5fl/fl mouse ES cell line after treatment with 4-hydroxytamoxifen (4-OHT). b, Western blot demonstrating depletion of HP1β protein in three independent Cbx1−/− ES cell lines. c, Western blot demonstrating depletion of HP1γ protein in Cbx3fl/fl ES cell line after treatment with 4-OHT. d, Western blot demonstrating depletion of HP1β and HP1γ proteins in three independent Cbx1−/−Cbx3fl/fl double-knockout cell lines after treatment with 4-OHT. n = 3 independent 4-OHT treatments. e, qRT–PCR demonstrating depletion of Cbx5, Cbx1 and Cbx3 mRNAs in the Cbx1−/−Cbx3fl/flCbx5fl/fl triple-knockout cell line upon treatment with 4-OHT. f, Western blot demonstrating depletion of HP1γ protein in three independent Cbx1Flag-AviTag/Flag-AviTag cell lines. g, Western blot demonstrating depletion of HP1β protein in three independent Cbx3Flag-AviTag/Flag-AviTag cell lines. For gel source data shown, see Supplementary Fig. 1. Experiments were performed twice.

Supplementary information

  1. Supplementary Information

    This file contains a Supplementary Discussion, a list of PCR primers, a Supplementary Table of cell lines and Supplementary References.

  2. Reporting Summary

  3. Supplementary Figures

    This file contains uncropped gel images.

  4. Supplementary Table 1

    This file contains Supplementary Table 1 showing Adnp peak annotations.

  5. Supplementary Table 2

    This file contains Supplementary Table 2. The table shows averaged Adnp ctrl and Adnp KO expression values as TPM (transcripts per million), fold change between ctrl and Adnp KO, and Adnp ChIPseq peak association for each gene.

  6. Supplementary Table 3

    This file contains Supplementary Table 3 which shows the full list of GO terms comparing genes upregulated in Adnp KO versus control.

  7. Supplementary Table 4

    This file contains Supplementary Table 4 which shows the expression analysis for Cbx knock out ES cells.

  8. Supplementary Table 5

    This file contains Supplementary Table 5 which shows a full list of GOterms for Adnp and HP1 triple KO analysis.

  9. Supplementary Table 6

    This file contains Supplementary Table 6 which shows a sequence of TALENs or Cas9-gRNA and integration ssODNs used in this study.

About this article

Publication history




Issue Date



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.