In embryonic stem cells, ZFP57/KAP1 recognize a methylated hexanucleotide to affect chromatin and DNA methylation of imprinting control regions

Abstract

The maintenance of H3K9 and DNA methylation at imprinting control regions (ICRs) during early embryogenesis is key to the regulation of imprinted genes. Here, we reveal that ZFP57, its cofactor KAP1, and associated effectors bind selectively to the H3K9me3-bearing, DNA-methylated allele of ICRs in ES cells. KAP1 deletion induces a loss of heterochromatin marks at ICRs, whereas deleting ZFP57 or DNMTs leads to ICR DNA demethylation. Accordingly, we find that ZFP57 and KAP1 associated with DNMTs and hemimethylated DNA-binding NP95. Finally, we identify the methylated TGCCGC hexanucleotide as the motif that is recognized by ZFP57 in all ICRs and in several tens of additional loci, several of which are at least ZFP57-dependently methylated in ES cells. These results significantly advance our understanding of imprinting and suggest a general mechanism for the protection of specific loci against the wave of DNA demethylation that affects the mammalian genome during early embryogenesis.

Figure 1

Altered Chromatin Status at Imprinted Control Regions in Kap1-Deleted Embryonic Stem Cells Histone 3 lysine 9 acetylation (A) and methylation (B) were assessed by ChIP/PCR in WT (sham, dark bars) and KAP1-deleted (4OH-tamoxiphen-treated, light bars) ES cells, using primers specific for incited genomic sequences. (C) ChIP/PCR analysis of control (ES, dark bars) and HA-tagged KAP1-expressing (ES + 6HAKAP1, light bars), using a HA-specific monoclonal antibody and primers for the indicated genomic sequences. Error bars indicate standard deviation; the difference between control and 4OHTam-treated cells was always significant (p < 0.01, n = 3), except for the GAPDH locus. KV, KvDMR1 ICR.

Figure 2

KAP1 ICR Recruitment in Embryonic Stem Cells Requires ZFP57 (A) ZFP57 binds to ICRs. HA-specific ChIP in ES cells expressing a HA-tagged ZFP57 protein, using primers specific for indicated genomic sequences. Error bars indicate standard deviation, and the difference between Gapdh locus and ICRs is always significant (p < 0.01, n = 3). (B) No KAP1 ICR recruitment in the absence of ZFP57. Same analysis as in (A), in control and Zfp57 knockout ES cells expressing a HA-tagged form of KAP1. Error bars indicates standard deviation, error bars indicate standard deviation, and the difference between control and Zfp57^−/− cells is always significant (p < 0.01, n = 3), except for the Gapdh locus. (C) H3K9me3 is lost at ICRs in Zfp57-defective ES cells. Same analysis as in (B), using an antibody specific for this chromatin mark. Error bars indicates standard deviation, error bars indicate standard deviation, and the difference between control and Zfp57^−/− cells is always significant (p < 0.01, n = 3), except for the Gapdh and IAP LTR loci. (D) Transcriptional deregulation of several imprinted genes in Zfp57 knockout ES cells. Scatter plot showing the gene expression data extracted from RNA-seq analysis; highlighted in red are Zpf57 and imprinted genes that are most markedly affected. Error bars indicate standard deviation. KV, KvDMR1 ICR.

Figure 3

KAP1, ZFP57, SETDB1, and HP1 Are Selectively Recruited to the Methylated Alleles of Imprinted Control Regions (A) RFLP analysis of KAP1-, ZFP57-, SETDB1- and HP1-bound material. Following ChIP with indicated antibodies recognizing endogenous proteins and ICR-specific PCR amplification, products were digested with restriction enzymes that cut only one parental allele to measure the ratio of maternal (%MAT) or paternal (%PAT) allele relative to total. H3K9me3 (which is known to bind to the methylated ICRs allele) and H3K9Ac (which is known to bind to the unmethylated ICRs allele) were analyzed as control. No Ab, no specific primary antibody. (B) Bisulfite sequencing of the KAP1- and ZFP57- bound H19, Igf2r and Snrpn ICRs. Methylated and unmethylated CpG dinuclotides are represented as open and filled circles, respectively. The input (pre-ChIP) DNA shows a 50/50 ratio of methylated and unmethylated ICR, although the KAP1- and ZFP57-immunoprecipitated (IP) material is totally or prevalently methylated. Each line corresponds to a single template DNA molecule cloned; each circle corresponds to a CpG dinucleotide. Filled circles designate the methylated cytosines; opened circles unmethylated cytosines. Differences are statistically significant (P-values < 0.05).

Figure 4

KAP1 Recruitment and H3K9 Trimethylation at Imprinting Control Regions Require DNA Methylation (A) HA-specific ChIP in control (WT) and Dnmt1/3a/3b-deleted (DNMTs TKO) ES cells expressing a HA-tagged form of KAP1 and transduced with an MLV-based retroviral vector, with primers for indicated genomic and virus-derived (MLV PBS) sequences. Error bars indicate standard deviation, and the difference between control and TKO cells is always significant (p < 0.01, n = 3), except for the GAPDH and MLV PBS. (B) Same analysis with H3K9me3-specific antibody-mediated ChIP. Histone 3 trimethylation of ICRs, but not of IAP chromatin, is abolished in the absence of DNA methylation. Error bars indicate standard deviation, and the difference between control and TKO cells is always significant (p < 0.01, n = 3), except for the GAPDH and IAP LTR. KV, KvDMR1 ICR.

Figure 5

Zfp57 Removal Leads to a Loss of DNA Methylation and ZFP57/KAP1 Interaction with NP95 and DNMTs (A) Bisulfite sequencing analysis of DNA extracted from Zfp57 knockout and WT ES cells. ICR, but not IAP, DNA methylation is lost in Zfp57 knockout ES cells. Differences between Zfp57^−/− and WT ES cells are statistically significant (p < 0.005). (B) Total protein extracts (input) or immunoprecipitates recovered with either naked (IP: beads) or antibody-coated beads with indicated antibodies (IP:KAP1 and IP:Flag) from control (Mock, left) or Flag-NP95-expressing ES cells were analyzed by western blot with antibodies, indicated on the right. (C) Same experiment in WT ES cells, using control, KAP1, and HA-specific beads for the IP and KAP1, HA, DNMT1, DNMT3A, or DNMT3B-specific antibodies for the western blot.

Figure 6

KAP1 Recognition of Imprinted Clusters Targets Imprinting Control Regions (A) ChIP-seq using a polyclonal antibody against endogenous KAP1 and ChIP-seq using an antibody against HA tagged ZFP57. Four loci are illustrated, with KAP1 binding results in green and ZFP57 in black, superimposed on published SETDB1 and H3K9me3 data in red and blue, respectively (Bilodeau et al., 2009). The ICR position is indicated in light blue, between mRNAs and ZFP57. (B) ChIP-seq data presented as (A) for nonimprinted loci predicted to bind ZFP57/KAP1. (C) Bisulfite sequencing of CpG islands located at the binding sites presented in B. Differences between Zfp57^−/− and WT ES cells are statistically significant (p < 0.001).

Figure 7

A Hexanucleotidic Motif Recognized by ZFP57 in a Methylation-Dependent Fashion (A) Logo presentation of consensus binding site deduced from the analysis of sites binding ZFP57, KAP1, and SETDB1. (B) Electrophoretic shift assay, using a recombinant MBP fusion protein encompassing the zinc fingers fragment Glu87-Ala195 from mouse ZFP57 to capture the illustrated double-stranded oligonucleotide including the consensus ZFP57 binding site (bold), in either native (unmet) or methylated (met) state. The met and unmet probes were assayed with or without a 50-fold excess of unlabeled WT met, WT unmet, mutant (mut1 and mut2), or nonspecific (NS) oligonucleotides. In addition to the methylation-dependent higher-mobility complex, a lower-mobility complex, possibly due to the binding of more than one ZFP57 protein to DNA, is also weakly detected with the unmet probe. (C) Several copies of the TGCCGC consensus are found at all known mouse and human ICRs. The KvDMR and H19 ICRs are illustrated as examples.