(A) Mass spectrometric quantification of ZBED6-enrichment using SILAC; the core sequences of the wild-type q and the mutant Q oligonucleotides used for protein capture are shown on top. ZBED6 was identified by six unique peptides in the SILAC-heavy q sample. Two peptides were simultaneously observed in the SILAC-light Q sample and were used as the basis for SILAC quantification. A representative peptide is shown where the area under the curve corresponds to the amount of ZBED6 enriched by the q probe (blue) and the Q probe (red). The average overall enrichment of ZBED6 by q was 9.0-fold (±1.2-fold, n = 2 experiments). (B) Genome organization of Zbed6 and Zc3h11a in mouse, adapted from the UCSC Genome Browser (http://genome-test.cse.ucsc.edu/). Untranslated and translated exons of Zc3h11a are indicated by black and red bars, respectively. A mammalian conservation track from the UCSC browser is included at the bottom. (C) Schematic representation of the six ZBED family members present in human and the Hermes hAT DNA transposase from the house fly. (D) Phylogenetic analysis of ZBED protein sequences from selected species. Cf (Canis familiaris; dog), Gg (Gallus gallus; chicken), Hs (Homo sapiens; human), Md (Monodelphis domestica; opossum), Mm (Mus musculus; house mouse), and Oa (Ornithorhynchus anitinus; platypus). ZBED1 and ZBED2 homologs were not found in the mouse genome, and mouse Zbed5 contains numerous mutations and stop codons efficiently destroying the reading frame. (E) EMSA showing binding of recombinant ZBED6 BED1/2 domains to wild-type (q), but not to mutant (Q) probe (left panel). Likewise, endogenous ZBED6 in nuclear extracts (NE) from C2C12 cells forms a complex with wild-type probe that is competed by q, but not Q probe (right panel). The complex is supershifted by the anti-ZBED6 antibody, and competition confirmed the specific interaction with ZBED6.

(A–J) Immunofluorescence analysis of mouse tissues stained with anti-ZBED6 or anti-ZC3H11A antibodies. (A and D) Coronal brain overview images assembled from pictures taken at 4× magnification. (B, C, E, and F) Higher magnification images of boxed areas indicated in (A and D) with anti-ZBED6 (B and C) or anti-ZC3H11A (E, F) immunopositive signals in red. (G–J) Images of anti-ZBED6 or anti-ZC3H11A antibody stainings of section from mouse thigh muscle. Nuclear localization of ZBED6 and ZC3H11A is confirmed by colabeling with DAPI (blue) in both brain and muscle as indicated by arrowheads, whereas nuclei lacking ZBED6- or ZC3H11A-positive signals are indicated by arrows. (K–N) Immunofluorescence analysis of pig muscle tissue (longissimus dorsi) stained with anti-ZBED6 (K and L) or anti-ZC3H11A (M and N). α-Bungarotoxin fused to a fluorescent dye (αBTX, green) was used to confirm identification of skeletal muscle tissue since this toxin binds cholinergic receptors in neuromuscular junctions. Scale bars indicate 480 µm in (A and D), 22 µm in (B, C, E, and F), and 14 µm in (G–N).

(A) Immunocytochemical staining of cytospins of C2C12 cells with anti-ZBED6 antibody (brown) and nuclear counterstain (blue). (B) Luciferase assays of reporter constructs containing the wild-type q or mutant Q sequence of pig IGF2 intron 3 and the pig IGF2 P3 promoter. Firefly reporter luciferase levels in relation to control Renilla luciferase level. (C) RT-PCR analysis of the Igf2 intron 3 region after chromatin immunoprecipitation (ChIP) with anti-ZBED6. (D) Igf2 mRNA expression at days 2, 3, and 6 after transfection. (E) Cell proliferation at day 3 after silencing. (F) Monolayer scratch wound healing. (G) Cell number in scratches. Assays were performed 48 h (B and C) or 72 h (E and F) after transfection. At least six transfections were performed for each siRNA except for the ChIP assay that involved three replicates (A–F). Error bars indicate s.e.m. (*p<0.05; **p<0.01; ***p<0.001).

ZBED6 repression may be released by (i) germline or somatic mutations at target sites, (ii) methylation (Me) of the CpG site in the binding motif, (iii) down-regulation of ZBED6 expression, or (iv) loss-of-function mutations in ZBED6.

(A) Multiple tissue and developmental skeletal muscle Northern blot analysis of Zbed6 and Zc3h11a; β-actin (Actb) was used as control. The Zbed6 and Zc3h11a probes apparently hybridized to the same transcript, whereas Zc3h11a also hybridized to shorter alternative transcripts lacking Zbed6. The estimated molecular weights are indicated. B, brain (whole); C, colon; H, heart; I, intestine (whole); K, kidney; Li, liver; Lu, lung; O, ovary; P, placenta (late pregnancy); S, stomach (whole); SM, skeletal muscle; Sp, spleen; Te, testis; Th, thymus; U, uterus (nonpregnant). (B) Real-time-PCR analysis of Zbed6 in several mouse tissues. Zbed6 mRNA expression was normalized using 18S-rRNA. Error bars represent standard error of the mean (s.e.m.). (C) ChIP sequencing data for the ZC3H11A/ZBED6 upstream region generated by the ENCODE project using the K-562 erythroleukemia cell line; adapted from the UCSC Genome Browser (http://genome-test.cse.ucsc.edu/). The location of RNA polymerase II binding sites (Pol2), a 5′ cap analysis gene expression tag (CAGE), binding sites for the Max, Myc, Fos, Jun, and NF-E2 transcription factors are indicated.

(A) Western blot analysis of recombinant protein (ZBED6a and ZBED6b) and C2C12 protein extracts using the polyclonal antibody raised against the ZBED6 BED domains. Two isoforms, ZBED6a and ZBED6b, with apparent molecular weights of 122 kDa and 116 kDa (calculated molecular weights 109 kDa and 104 kDa, respectively) corresponding to two predicted alternative translation start sites were detected in C2C12 protein extracts. (B) Nuclear localization of ZBED6 in C2C12 cells. Constructs containing ZBED6 BED1/2 domains fused to GFP compared with GFP alone. The constructs GFP-BED1/2x and GFP-BED1/2y contained amino acid residues 47–384 and 90–384 of the ZBED6a isoform, respectively. (C) Immunofluorescence analysis of C2C12 cells stained with anti-ZBED6 (red) and anti-nucleophosmin 1 (Npm1; green) antibodies.

(A) ChIP-sequencing peak around the quantitative trait nucleotide (QTN) position in an intron of Igf2. (B) ZBED6 binding motif in Igf2 and consensus ZBED6 motif based on peaks within 5 kb of transcription start sites (TSS) and with at least 25 overlapping fragments but excluding the binding site at Igf2. (C and D) ChIP-sequencing peaks are preferentially located in the near vicinity of (C) TSS and (D) CpG islands. (E) Gene Ontology categories highly enriched among the genes bound by ZBED6. Ontology terms are shown on the y axis; p-values for the significance of enrichment are on the x axis. (F) Overrepresented transcription factor families bound by ZBED6, including homeobox proteins (HOX), paired-like (PAX), HOX factors (MEIS), hairy and enhancer of split protein (HES), krueppel C2H2 zinc finger (KLF), Forkhead box (FOX), cbp/p300-interacting transactivator (CITED), SNO and SKI oncogenes (SNO/SKI), atonal homolog (ATOH), basic helix-loop-helix (BHLH), LIM homeobox (LHX), POU domain containing classes 3 and 4 (POU), NK transcription factor related (NKX), early B-cell factor (EBF), nuclear receptor (NR), SRY box (SOX), T-box (TBX), and Iroquois homeobox protein (IRX). The number of transcription factor genes associated with ZBED6 binding sites is indicated beside each family name. (G) Results of Ingenuity Pathways Analysis showing that putative downstream targets of ZBED6 are significantly associated with different diseases in humans. p-values for nonrandom association were calculated using the Fisher exact test, followed by a Benjamini-Hochberg correction for multiple testing. The red line indicates the 5% significance threshold.