Research Article 1607 DNA CpG hypomethylation induces heterochromatin reorganization involving the histone variant macroH2A

Yinghong Ma1, Stephanie B. Jacobs1, Laurie Jackson-Grusby2,*, Mary-Ann Mastrangelo2,‡, José A. Torres-Betancourt1, Rudolf Jaenisch2,3 and Theodore P. Rasmussen1,4,5,§ 1Center for Regenerative Biology, University of Connecticut, 1392 Storrs Road, Storrs, CT 06269-4243, USA 2Whitehead Institute for Biomedical Research, Nine Cambridge Center, Cambridge, MA 02142, USA 3Department of Biology, MIT, Cambridge, MA 02139, USA 4Department of Animal Science, University of Connecticut, Storrs, CT 06269-4040, USA 5Department of Molecular and Cell Biology, University of Connecticut, Storrs, CT 06269-3042, USA *Present address: Boston Children’s Hospital, Pathology Department, 300 Longwood Ave, Boston, MA 02115, USA ‡Present address: Section of Infectious Diseases, Department of Medicine, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA §Author for correspondence (e-mail: [email protected])

Accepted 26 January 2005 Journal of Cell Science 118, 1607-1616 Published by The Company of Biologists 2005 doi:10.1242/jcs.02291

Summary In mammalian heterochromatin, cytosine bases of CpG immunostaining in Dnmt1–/– cells becomes coincident with dinucleotides are symmetrically modified by methylation. chromocenters detected by HP1 content. MacroH2A, but Patterns of CpG methylation are maintained by the not HP1, is enriched in nuclease-resistant chromatin action of Dnmt1, the mammalian maintenance cytosine fractions extracted from Dnmt1–/– cells. Normal methyltransferase enzyme. We genetically manipulated the localization of macroH2A was restored upon levels of CpG methylation and found that extensive reintroduction of a Dnmt1 transgene into Dnmt1–/– cells. chromatin alterations occur in pericentric MacroH2A localization was also affected in T-antigen- heterochromatin. Homozygous mutations in Dnmt1 cause transformed fibroblasts subjected to the conditional severe hypomethylation of pericentric heterochromatin mutation of Dnmt1. Together, these results suggest that and concomitant chromatin reorganization involving the pericentric heterochromatin can be maintained in the histone variant macroH2A. Demethylation-induced absence of CpG methylation, but in a significantly altered alterations in macroH2A localization occur in both configuration. interphase and mitotic embryonic stem (ES) cells. Heterochromatin protein 1 (HP1) marks interphase Key words: chromatin, methylation, Dnmt1, MacroH2A, centromere, Journal of Cell Science pericentric heterochromatin (chromocenters). MacroH2A chromocenter

Introduction nucleosomes assembled at centromeres (Palmer et al., 1987). The genomes of mammals are organized as chromatin, which Pericentric nucleosomes also contain a high content of histone is a complex and dynamic assemblage of DNA and proteins. H3 that is post-translationally modified by methylation at Transcriptionally silent DNA can be roughly divided into two lysine residue 9 (me-H3-K9), a modification that is associated functional compartments – constitutive and facultative with transcriptional inactivity (Lachner et al., 2001; Maison et heterochromatin. Constitutive heterochromatin contains DNA al., 2002; Rea et al., 2000). The me-H3-K9 modification serves in a state of obligate transcriptional silence and includes as a cis-acting binding site for heterochromatin protein 1 (HP1) (Bannister et al., 2001; Jacobs and Khorasanizadeh, 2002; elements such as centromeres, pericentric DNA, telomeres Lachner et al., 2001), an interaction that serves to sequester and silenced mobile elements (LINE-1 elements, SINE interphase centromeres in discrete foci (chromocenters). In elements, disabled retroviruses, etc.). In contrast, facultative addition, regions in and around mammalian centromeres are heterochromatin is conditionally silenced. Facultative subject to heavy DNA methylation at CpG dinucleotides. In heterochromatin includes genes subject to tissue-specific short, constitutive heterochromatin contains a distinct silencing, inactive X chromosomes of female eutherian molecular signature consisting of heavy methylation of CpG mammals, imprinted genes and perinucleolar DNA. Though dinucleotides coupled with chromatin enriched in CENP-A, both classes of heterochromatin are organized into me-H3-K9 and concomitant association with HP1. nucleosomes, a growing body of evidence reveals that Facultative heterochromatin such as that of the inactive X constitutive and facultative heterochromatin differ chromosome (Xi) also contains a distinct molecular signature. substantially in their content of variant histones and post- Though numerous reports show that CpG islands of silenced translational histone modifications. X-linked genes are highly methylated, the overall levels of Constitutive heterochromatin is assembled upon mammalian methylation on Xi appear to be reduced relative to the active centromeres and surrounding pericentric sequences. The X chromosome (Xa) and autosomes (Viegas-Pequignot et al., histone H3 variant CENP-A replaces histone H3 in 1988). In addition, the chromatin of Xi is enriched in histone 1608 Journal of Cell Science 118 (8)

H3 methylated at lysine 27 (me-H3K27) (Plath et al., 2003; Southern blot analysis Silva et al., 2003) and macroH2A, a variant of canonical Southern blotting was performed by separating 10 µg of genomic histone H2A, which contains a large C-terminal extension DNA (digested to completion with indicated restriction enzymes) per (Costanzi and Pehrson, 1998; Pehrson and Fried, 1992). On a lane on 1% agarose gels. DNA was then transferred to Genescreen molecular level, MacroH2A interferes with NF-kappaB PlusTM nylon membrane (Dupont-NEN). Blots were hybridized with transcription factor binding and SWI/SNF chromatin probes derived from radiolabeled extension of random hexamers remodeling (Angelov et al., 2003). Furthermore, the C- annealed to the purified inserts of plasmids pMR150 (mouse terminal nonhistone domain (NHD) of macroH2A may be centromeric repeats) or pKS13b (Line-1 elements), or CoT1 DNA (Gibco; containing total mouse repetitive DNA). involved in ADP-ribosylation of chromatin with potential implications for transcriptional silencing (Ladurner, 2003). The crystal structure of the AF1521 protein from Antibody development Archaeoglobus fulgidus was recently determined (Allen et al., Polyclonal antibodies directed against macroH2A were prepared by 2003). This protein is similar to the NHD of macroH2A and cloning a cDNA fragment corresponding to the complete nonhistone probably binds nucleic acids, suggesting a possible role for domain (NHD) portion of a macroH2A1.2 cDNA (Rasmussen et al., macroH2A in chromatin condensation. 1999) into the plasmid pGEX-3T (Frorath et al., 1992). To do this, a Methylation of CpG dinucleotides in mice is maintained by unique PacI site was introduced at the junction of the histone and NHD coding regions in the macroH2A1.2 cDNA (Rasmussen et al., the action of Dnmt1, a maintenance DNA methyltransferase 1999), and the NHD coding region was fused to GST in a vector with that converts hemimethylated CpG dinucleotides to a thrombin cleavage site between the GST and NHD moieties (Fig. symmetrically methylated dinucleotides (Pradhan et al., 1999). 2A). Recombinant GST/NHD fusion protein was expressed in E. coli Loss of Dnmt1 results in embryonic lethality, but strain DH5α and GST-NHD fusion protein was then eluted with undifferentiated embryonic stem (ES) cells lacking Dnmt1 reduced glutathione. Alternatively, pure NHD was released from the function are fully viable (Li et al., 1992). Since ES cells are column by thrombin digestion and used to immunize a New Zealand viable in the absence of CpG demethylation, we used these white rabbit (Covance). Serum samples were assayed by western cells to investigate the molecular composition of demethylated blotting and IgG was purified from reactive serum on protein A- heterochromatin. We show that loss of DNA methylation agarose beads (Sigma). Alternatively, antibody was affinity-purified causes a significant reorganization of constitutive from serum using an affinity column assembled by covalently attaching pure NHD protein to cyanogen bromide-activated Sepharose heterochromatin involving the histone variant macroH2A. This CL-4B (Sigma). Affinity-purified antibody was eluted from this finding raises the intriguing possibility that mammalian matrix with low pH, and the eluted antibody was dialyzed in PBS. heterochromatin adopts a default state, reminiscent of The resulting antibody recognized endogenous macroH2A in total facultative heterochromatin, in response to DNA cellular protein from embryonic stem cells. In addition, the antibody demethylation. detected inactive X chromosomes in female fibroblasts, and centrosomal-associated macroH2A in undifferentiated ES cells (data not shown). These are both established patterns of macroH2A localization as determined with an independently derived antibody Materials and Methods (Costanzi and Pehrson, 1998; Rasmussen et al., 2000).

Journal of Cell Science Cell culture Male ES cell lines with genotypes Dnmt1N/N, Dnmt1S/S and Dnmt1C/C were used. The N allele of Dnmt1 consists of a targeted deletion in Cell biological methods the N-terminal coding region of Dnmt1 that causes dramatic reduction Immunofluorescence and western blotting were performed using α- of CpG methylation (Li et al., 1992). The S allele of Dnmt1 disrupts macroH2A (above), α-HP1 (Serotec, cat. no. MAC353) and CREST the domain that targets the Dnmt1 enzyme to replication forks, autoimmune serum (kindly supplied by Steven Doxsey) using resulting in a reduction of CpG methylation due to mislocalization of methods previously described (Rasmussen et al., 2000). Additional the enzyme (Lei et al., 1996; Leonhardt et al., 1992; Li et al., 1993). primary antibodies used include: α-H2B (Chemicon, cat. no. The C allele of Dnmt1 disrupts the methyltransferase domain of the AB1623); α-trimethyl-H3K27 (Upstate, cat. no. 07-449); α-trimethyl- enzyme (Lei et al., 1996). Male ES cell line B9 was derived by H3K9 (Upstate, cat. no. 07-442); α-trimethyl-H3K4 (Upstate, cat. no. introduction of a bacterial artificial chromosome bearing the wild-type 07-473). Dnmt1 gene into the Dnmt1C/C cells, which causes extensive Chromosome spreads were prepared from ES cells grown without remethylation (Biniszkiewicz et al., 2002). feeder cells in ESCM supplemented with 1000 units/ml LIF. ES cells were grown under standard conditions (on irradiated Colchicine (Sigma) was added to cells 2 hours prior to harvest at a fibroblast feeder layers) in ES cell medium (ESCM) consisting of concentration of 1 µg/ml, then cells were trypsinized, washed once, DMEM-based medium (Gibco) supplemented with non-essential and resuspended in 5 ml of 0.075 M KCl at 37°C for 12 minutes. Cells amino acids, 2 µl/l β-mercaptoethanol and 500 units/ml leukemia were then pelleted carefully at 800 rpm in a tabletop centrifuge, inhibitory factor (LIF) on gelatinized plastic tissue culture vessels. resuspended in residual supernatant and dropped onto slides to Feeders were removed by trypsinizing ES cell cultures and panning liberate chromosome spreads, followed by immediate fixation in 4% out feeder cells by incubation for 30 minutes of growth on plates not paraformaldehyde in PBS at 4°C for 5 minutes. Slides were washed treated with gelatine. ES cells were then plated for an additional in PBS and immediately subjected to immunofluorescence. passage without feeder cells in ESCM supplemented with 1000 units/ml LIF prior to trypsinization and growth on slides (for cell biological analysis) or harvest of nuclei for chromatin fractionation Chromatin fractionation (below). Nucleosomes were prepared essentially as described previously Mouse embryonic fibroblast (MEF) cell lines harboring (Thorne et al., 1998). Cells were harvested and washed in ice-cold homozygous conditional mutations in Dnmt1, methods to induce cre- PBS-butyrate (135 mM NaCl, 250 mM KCl, 8 mM Na2HPO4, 1.5 mM mediated loss of function and Y-chromosome PCR were described KH2PO4, 10 mM sodium butyrate), then resuspended in cell lysis previously (Jackson-Grusby et al., 2001). buffer (250 mM sucrose, 10 mM Tris-HCl, pH 7.4, 10 mM sodium Hypomethylation and macroH2A 1609

butyrate, 4 mM MgCl2, 0.1% Triton X-100) with protease inhibitors (1 tablet of protease inhibitor cocktail; Roche, cat. no. 1697498), and lysed with 20 strokes of a Dounce homogenizer (type B pestle). Nuclei were pelleted by centrifugation for 10 minutes at 2000 g at 4°C, then resuspended in 3 ml of wash buffer C (250 mM sucrose, 10 mM Tris-HCl, pH 7.4, 10 mM sodium butyrate, 4 mM MgCl2 and protease inhibitors) and layered over a 5 ml sucrose cushion (30% sucrose w/v in wash buffer C). Nuclei were then centrifuged at 2400 g for 5 minutes at 4°C in a swinging bucket rotor and the nuclear pellet was resuspended in 250 µl of wash buffer C. CaCl2 was added to a final concentration of 1 mM and incubated at 37°C for 8 minutes in the presence of 8 µl of micrococcal nuclease (MNase, 5 units/µl). Digestion was stopped by adding EDTA to a concentration of 10 mM. Nuclei were then Fig. 1. Genomic methylation assays for repetitive sequences. Southern blots of total DNA pelleted and resuspended in 500 µl of 0.2 mM extracted from wild-type (J1), rescued B9 cells (Dnmt1C/C cells containing a Dnmt1+ BAC) ice-cold EDTA, pH 7.0 for 1 hour. This and Dnmt1C/C ES cells, cut with HpaII (methyl sensitive) and MspI (methyl insensitive) hypotonic treatment forced nucleosomes into isoschizomeric restriction enzymes. (A) Southern analysis of equally loaded DNA using a the supernatant (extraction 1). The nuclei were probe that recognizes minor centromere repeats. (B) Identical Southern blot probed with then extracted a second time (3 hours of pKS13b, which recognizes open reading frames of Line-1 repetitive transposable diffusion), yielding a chromatin fraction elements. (C) Identical Southern blot probed with mouse CoT1 DNA, which recognizes containing micrococcal nuclease-resistant all highly repetitive sequences in the mouse genome. polynucleosomes (extraction 2). Chromatin samples were analyzed for DNA content after phenol/chloroform extraction, and protein content was analyzed by bearing the wild-type Dnmt1 gene into the Dnmt1C/C cells SDS-PAGE and western blotting using standard methods. (Biniszkiewicz et al., 2002). Southern blots were performed using the methyl-sensitive restriction enzyme HpaII, and MspI, an enzyme with the same recognition sequence as HpaII, but Results whose activity is unaffected by CpG methylation (Fig. 1). We Demethylation of repetitive DNA by Dnmt1 deletion found that Dnmt1C/C cells contained highly demethylated We assessed the levels of CpG methylation in three ES cell lines: centromeric repeats and that centromeric methylation was wild-type J1 cells, an ES cell line with a Dnmt1C/C genotype, partially restored in rescued cell line B9 (Fig. 1A). Similar

Journal of Cell Science which contains homozygous deletions in the methyltransferase analyses showed that Line-1 elements were highly demethylated portion of Dnmt1 (Lei et al., 1996), and ES cell line B9, which in Dnmt1C/C cells and partially remethylated in B9 (Fig. 1B). was derived by introduction of a bacterial artificial chromosome Finally we probed blots with CoT1 DNA (purified repetitive mouse genomic DNA) (Fig. 1C). This showed that the total repetitive DNA of Dnmt1C/C cells was highly demethylated, and Fig. 2. Production of polyclonal antibodies against macroH2A. that methylation was partially restored in B9. It is generally (A) MacroH2A consists of an N-terminal region (red) that is 64% identical to canonical H2As followed by a large C-terminal nonhistone domain (NHD) of unknown function (green). The dark green box labeled 1.2 indicates the position of one of two potential leucine zipper domains generated by alternative splicing. A PacI site was introduced into the macroH2A.2 cDNA at the junction of the histone and nonhistone domains (black triangle), allowing fusion of the nonhistone domain to GST. The GST-NHD fusion contains a thrombin cleavage site (open triangle) at the GST-NHD junction, allowing proteolytic cleavage to yield free NHD protein. (B) Recombinant NHD expression in bacterial cells. Lanes 1 and 2: Coomassie Blue-stained total bacterial protein from uninduced (lane 1) and IPTG-induced cells (lane 2) containing a GST-NHD expression plasmid. Lane 3: GST-NHD fusion protein purified from bacterial cells on a glutathione column eluted with reduced glutathione elution. Lanes 4 and 5: Purified NHD protein was eluted from a glutathione resin charged with GST-NHD by cleaving with thrombin. This allowed free NHD fusion protein to be eluted in two washes (samples of these supernatants are shown). (C) NHD protein from the same sample loaded in lane 4 was used to inoculate a rabbit that produced antibodies that recognize recombinant GST-NHD protein (lane 1) and endogenous macroH2A in total ES cell protein (lane 2) by immunoblotting. 1610 Journal of Cell Science 118 (8)

Fig. 3. Localization of macroH2A in wild-type and demethylated ES cells. (A) Representative ES cells subjected to macroH2A

Journal of Cell Science immunofluorescence (green) and DAPI staining of DNA (blue). J1 is a wild-type ES cell line. N/N, S/S and C/C denote ES cells with DnmtN/N, DnmtS/S and DnmtC/C genotypes, respectively. ES cell line B9 was derived by introduction of a Dnmt1 BAC into cell line into Dnmt1C/C cells resulting in partial remethylation of genomic CpG dinucleotides. (B) The same ES cell lines immunostained with CREST autoimmune serum, which recognizes centromeric kinetochore proteins including the histone H3 variant CENP-A. (C) Representative immunostainings of mitotic J1 and C/C ES cells using α-macroH2A and CREST serum. Note that in J1, macroH2A and kinetochore immunostaining is distinct, while in C/C cells, immunostaining is nearly coincident.

accepted that CoT1 DNA is composed of the sum total of highly Dnmt1 function in ES cells. We therefore sought to test repetitive DNA present in the mouse genome. Since the pattern whether loss of pericentric DNA methylation influenced the of bands revealed by the CoT1 DNA probe was highly similar distribution of macroH2A in Dnmt1-mutant ES cells. to the combined patterns of the centromeric and Line-1 probes, MacroH2A resembles canonical histone H2A except for the we conclude that centromeric repeats and Line-1 elements presence of a C-terminal non-histone domain (NHD) roughly constitute the bulk of murine repetitive DNA that is subject to twice the size of the histone core domain (Pehrson and Fried, CpG methylation. In summary, we conclude that Dnmt1C/C cells 1992). We purified recombinant NHD protein and used it as an contain extensively demethylated repetitive DNA and that partial immunogen to create a polyclonal antibody that detected remethylation can be achieved by reintroduction of a Dnmt1 macroH2A but not endogenous H2A in total protein extracted BAC. from ES cells (Fig. 2). We performed immunofluorescence on a panel of male ES cells bearing mutations in Dnmt1 to investigate potential Demethylation-induced chromatin remodeling in alterations of pericentric chromatin resulting from reductions pericentric DNA in levels of CpG methylation (Fig. 3A). These cells included a The variant histone macroH2A has previously been shown to wild-type line (J1), ES cells with Dnmt1N/N, Dnmt1S/S, associate with pericentric DNA in preimplantation mouse Dnmt1C/C genotypes (which all cause extensive demethylation embryos (Costanzi et al., 2000) and spermatocytes (Hoyer- of the genome), and remethylated ES cell line B9. Interphase Fender et al., 2004). The results presented in Fig. 1 confirmed nuclei of ES cells of all genotypes contained condensed regions that pericentric DNA is highly demethylated upon loss of that stained intensely with DAPI (Fig. 3A). These condensed Hypomethylation and macroH2A 1611 chromocenter-like foci correspond to the sites of interphase centromeres, and can be identified using CREST antisera, human autoimmune serum from schleroderma patients that specifically recognizes interphase kinetochore proteins (Brenner et al., 1981). Though chromocenters were intact and apparently unaffected by DNA methylation levels, we noted alterations in the appearance of chromocenter histones. We observed a dramatic increase of macroH2A immunostaining in hypomethylated chromocenters (Fig. 3A). In undifferentiated wild-type J1 ES cells, macroH2A exhibited diffuse nuclear staining; a localization pattern observed with several other fully methylated ES cell lines (data not shown). However, in Dnmt1N/N, Dnmt1S/S, Dnmt1C/C cells, we noticed intense concentrations of macroH2A that were coincident with chromocenters. We double-labeled Dnmt1C/C cells to visualize macroH2A and CREST antigens and found that concentrations of macroH2A are often adjacent to CREST signals (Fig. 3B). The pericentric concentrations of macroH2A were reversed in rescued cell line B9. We conclude that chromatin remodeling involving macroH2A occurs within interphase pericentric chromatin. Pericentric DNA is known to associate with heterochromatin protein 1 (HP1) via direct interactions that occur between HP1 and histone H3 methylated at lys 9 (Bannister et al., 2001; Jacobs et al., 2001; Lachner et al., 2001). We found that HP1 co-localizes with chromocenters in ES cells regardless of their Dnmt1 genotype (Fig. 3C). However, macroH2A becomes concentrated in demethylated chromocenters marked by HP1 immunostaining. Because chromocenters contain interphase centromeres and associated pericentric chromatin, we investigated the distribution of macroH2A on metaphase chromosomes (Fig. 4). We found that metaphase centromeres of CpG- demethylated chromosomes exhibited increased signals for macroH2A (Fig. 4B) as compared to normally methylated Journal of Cell Science chromosomes prepared from wild-type cells (Fig. 4A). In addition, we performed combined macroH2A/CREST immunofluorescence on J1 and C/C cells and identified mitotic cells. We found that the macroH2A immunostaining pattern was immediately adjacent to CREST kinetochore staining only in demethylated Dnmt1C/C cells (Fig. 4C).

Fig. 4. Representative localization of macroH2A on mitotic MacroH2A content in demethylated nuclease-resistant chromosomes. Immunodetection of macroH2A distribution on chromatin individual metaphase chromosomes from (A) J1 (wild type) ES cells C/C The above results suggested that heterochromatin acquires a and (B) C/C (Dnmt1 ) ES cells. Chromosomes were stained with DAPI (blue) to detect DNA, and with FITC (green) to detect α- substantial content of macroH2A when CpG methylation levels macroH2A antibody. Images in A and B had identical exposure are significantly reduced. We therefore conducted chromatin times. Note enrichment of macroH2A immunostaining at telocentric analyses using a cell fractionation approach based on nuclease centromeres in Dnmt1C/C cells. (C) Individual mitotic cells from J1 accessibility (Fig. 5A). We isolated nuclei from wild-type, and C/C cultures were also subjected to combined Dnmt1C/C and B9 cells, and infused them with micrococcal immunofluorescence to detect macroH2A and kinetochore proteins nuclease, which cleaves chromatin between nucleosomes. (CREST). After nearly complete nuclease digestion was achieved, nuclease activity was quenched with EDTA and soluble chromatin was forced from nuclei into the supernatant by similar efficiencies. Protein from each chromatin sample was hypotonic treatment (extraction 1). We isolated the DNA analyzed by SDS-PAGE and Coomassie Blue staining. The from extraction 1 and found that it contained primarily majority of the protein in these samples consisted of histones mononucleosomes (Fig. 5B). A second hypotonic extraction (Fig. 5C), although the pattern of non-histone auxiliary bands was enriched in nuclease-resistant chromatin as judged by the differed between nuclease-sensitive and nuclease-resistant presence of polynucleosome ladders (Fig. 5B, extraction 2). fractions. We subjected identical amounts of chromatin Chromatin from cells of each Dnmt1 genotype digested with proteins from each fraction to western blotting for HP1 and 1612 Journal of Cell Science 118 (8) respond to reduced levels of DNA methylation. We first performed a series of western blots of total protein extracted from wild-type and Dnmt1C/C to screen for major changes in histone content or modification status of histones (Fig. 6). To do this, equal amounts of protein from wild-type (J1) and DnmtC/C cells were used. We confirmed that protein loading was equal both by Coomassie Blue staining of total protein and western blots of identically loaded lanes using an antibody specific for histone H2B. Similar blots showed that macroH2A protein is slightly more abundant in Dnmt1C/C cells than in wild-type cells. We found that trimethyl modifications of histones H3 at lysines 27, 9 and 4 exhibited little or no change in overall levels in response to demethylation. We also investigated the distribution of methyl modifications at a subnuclear level through immunofluorescence analysis (Fig. 7). Trimethyl-H3K27 is a modified histone found predominantly in facultative heterochromatin (Cao et al., 2002; Plath et al., 2003). We found a finely particulate diffuse nuclear staining of trimethyl-H3K27 in both wild-type and Dnmt1C/C ES cells (Fig. 7A). A very diffuse nuclear distribution was observed for trimethyl-H3K4 (a modification enriched in euchromatin, Journal of Cell Science (Santos-Rosa et al., 2002). The Fig. 5. Analysis of demethylation-induced chromatin remodeling in demethylated ES cells. localization of trimethyl-H3K4 was (A) Outline of chromatin fractionation and analysis methods. Nuclei were isolated from ES cells and chromatin was digested in situ with micrococcal nuclease. Nucleosomes were then also unperturbed by demethylation of driven from nuclei by two successive hypotonic treatments (extractions), which liberate free DNA (Fig. 7B). Trimethyl-H3K9 is a nucleosomes into the supernatant. Thereafter, DNA and protein moieties of isolated modified histone found predominantly nucleosomes were analyzed. (B) Micrococcal nuclease-digested DNA isolated from J1, B9 and in constitutive heterochromatin, C/C cells after initial and second hypotonic extractions. (C) Protein extracted from the same especially in pericentric chromatin chromatin fractions shown in B detected by Coomassie Blue staining. (D) Western analysis of (Maison et al., 2002; Peters et al., chromatin proteins in the same preparations shown in B and C. (Loading is identical to 2001). We assessed the localization of Coomassie Blue-stained gel in C.) trimethyl-H3K9 in wild-type and Dnmt1C/C ES cells and observed little macroH2A (Fig. 5D). We found increased amounts of or no response to demethylation of DNA for this modification macroH2A in chromatin from CpG hypomethylated cell lines, (Fig. 7C). We counted cells from each experiment and and this effect was most pronounced in samples with highest determined the proportion with diffuse nuclear versus punctate resistance to micrococcal nuclease. In contrast, HP1 levels chromocenter staining (Table 1). Only macroH2A was were insensitive to the state of DNA methylation. We therefore redistributed from one class to another. conclude that loss of DNA CpG methylation leads to an increase in macroH2A in demethylated, nuclease-resistant chromatin fractions. Demethylation-induced chromatin remodeling in somatic cells Unlike somatic cells, ES cells remain viable even when CpG Analysis of post-translational histone modifications in methylation is extensively reduced (Li et al., 1992), a feature wild-type and demethylated ES cells that allowed the above studies to be performed on viable cells. Because demethylation caused alteration of macroH2A However, we wished to investigate whether loss of CpG distribution in ES cells, we performed experiments to methylation affects macroH2A in differentiated cells. To do investigate other histone features to see if they, too, would this, we conducted experiments using mouse embryonic Hypomethylation and macroH2A 1613

Table 1. Chromatin patterns in wild type and Dnmt1C/C ES cells Wild type Dnmt1C/C Chromatin component Chromatin pattern* No.† Proportion‡ No.† Proportion‡ MacroH2A Diffuse nuclear (D) 13 0.064±0.017 185 0.911±0.020 Chromocenter (C) 189 0.936±0.017 18 0.089±0.020 Total (n) 202 203 Trimethyl-H3K27 Diffuse nuclear (D) 207 1.000±0.000 203 1.000±0.000 Chromocenter (C) 0 0.000±0.000 0 0.000±0.000 Total (n) 207 203 Trimethyl-H3K9 Diffuse nuclear (D) 1 0.005±0.005 1 0.005±0.005 Chromocenter (C) 212 0.995±0.005 203 0.995±0.005 Total (n) 213 204 Trimethyl-H3K4 Diffuse nuclear (D) 198 0.990±0.007 198 0.985±0.009 Chromocenter (C) 2 0.010±0.007 3 0.015±0.009 Total (n) 200 201

Wild-type and DnmtC/C ES cells were immunostained with antibodies recognizing the listed chromatin components and greater than 200 cells were visually examined to determine the pattern of localization of chromatin components. *Two general chromatin patterns were observed: diffuse nuclear particulate distribution (D), or chromocenter-associated patterns (C). †Number of cells counted. ‡Proportions were determined by dividing the number of cells counted with a given pattern (D or C) by the total (n). Standard deviations for proportions were determined with the equation σ=√p(1–p)/n , where p=the proportion under consideration (D/n or C/n) and n is the total number of cells counted.

fibroblasts (MEFs) that harbor homozygous conditional alleles lines proved to be male, as judged by PCR to detect Y of Dnmt1 in which exons 4 and 5 are flanked by loxP sites chromosomal sequences (Jackson-Grusby et al., 2001). In (Jackson-Grusby et al., 2001). T-antigen (T-Ag)-transformed contrast, a female MEF line called X17 failed to exhibit derivatives of these MEFs remain viable after viral delivery of cre recombinase to induce recombination of the floxed Dnmt1 exons. We exposed T-Ag-transformed conditional MEF lines to virus alone (Fig. 8A,C) or virus expressing cre recombinase (Fig. 8B,D). We found that conditional loss of Dnmt1 in these MEFs led to a redistribution of the macroH2A signal similar to that observed in ES cells. Specifically, macroH2A formed foci coincident with demethylated pericentric DNA as indicated by intense DAPI and CREST staining (Fig. 6B,D). We observed this dramatic redistribution in two of four Journal of Cell Science conditional MEF lines examined. The two responsive MEF

Fig. 6. Western blot analysis of histone content in wild-type and demethylated ES cells. Equivalent amounts of cellular protein extracted from wild-type (J1) and Dnmt1C/C (c/c) ES cells were loaded onto SDS polyacrylamide gels and stained with Coomassie Blue to confirm equal loading. Equivalently loaded lanes were Fig. 7. Localization of post-translational methylation modifications transferred to membranes for western blot analyses using antibodies of histone H3 in wild-type (J1) and Dnmt1C/C (c/c) ES cells. specific for the N-terminal tail of histone H2B, macroH2A and (A) Immunostaining with an antibody specific for histone H3 modification-state-specific antibodies that recognize histone H3 trimethylated at lysine 27. (B) Immunostaining with an antibody trimethylated at lysine 27 (methyl-H3K27), lysine 9 (methyl-H3K9) specific for histone H3 trimethylated at lysine 4. (C) Immunostaining or lysine 4 (methyl-H3K4). with an antibody specific for histone H3 trimethylated at lysine 9. 1614 Journal of Cell Science 118 (8)

Fig. 8. Analysis of macroH2A distribution in T-antigen- transformed MEFs after conditional mutation of Dnmt1. (A,B) Male MEF line F11 was transfected with empty virus (A) or cre virus (B) to delete exons 4 and 5 of the Dnmt1 gene that encodes the methyltransferase core of the enzyme. Cells were fixed and subjected to combined immunofluorescence for macroH2A and kinetochore proteins (CREST) to mark chromocenters. (C,D) An independently derived cell line (F12) was subjected to similar mock (C) and cre-virus (D) transfections. F12 yielded results similar to those observed for F11. (E,F) A female MEF cell line (X17) was also subjected to mock (E) or cre-induced (F) loss of Dnmt1 function. Loss of Dnmt1 function in X17 and a second female cell line (X14, not shown) failed to perturb the localization of Journal of Cell Science maroH2A.

demethylation-induced redistribution of macroH2A (Fig. chromatin-remodeling event involving the histone variant 8E,F), although these cells contained distinctive macroH2A. macrochromatin bodies corresponding to the positions of Two possible mechanisms might explain our findings. (1) inactive X chromosomes (Costanzi and Pehrson, 1998) and Loss of DNA methylation may cause a decondensation of lacked Y chromosomes, as judged by Y-specific PCR. A heterochromatin at pericentric regions, leading to the second female MEF line containing conditional genotype for unmasking of macroH2A epitopes that are inaccessible in Dnmt1 also failed to exhibit a redistribution of macroH2A in fully-methylated cells. (2) Loss of DNA methylation may response to demethylation. Though this possible sex-dependent induce a chromatin-remodeling event that causes increased effect on chromatin dynamics is intriguing, these MEF macroH2A to become incorporated into demethylated cell lines may harbor other differences that we presently do pericentric heterochromatin. We cannot formally rule out the not understand. Nonetheless, the results indicate that first mechanism, and indeed, several results suggest that demethylation-induced chromatin reorganization involving chromatin remodeling may be involved. First, we found macroH2A can occur in somatic cells. increased macroH2A signal on demethylated centromeres of metaphase chromosomes (Fig. 4). We also found that macroH2A protein is enriched in nuclease-resistant Discussion demethylated chromatin (Fig. 5D), and that other anti-histone The results presented here show that ‘crosstalk’ exists that antibodies detect no significant differences in chromatin connects the DNA CpG methylation system with other patterns when wild-type and Dnmt1C/C cells are compared (Fig. aspects of heterochromatin structure. The constitutive 7). Together, these findings suggest that demethylation-induced heterochromatin of mammalian centromeres is ordinarily unmasking of macroH2A epitopes is not likely to account for highly methylated at CpG dinucleotides. Our results show the observations. In either case, the results indicate that a close that pericentric heterochromatin is significantly altered connection exists between DNA methylation and the structure when CpG DNA methylation is greatly reduced. Specifically, of ES cell centromeres. we find that CpG demethylation induces an apparent Heterochromatin is often classified into one of two types – Hypomethylation and macroH2A 1615

constitutive (obligately silenced, such as pericentric Biniszkiewicz, D., Gribnau, J., Ramsahoye, B., Gaudet, F., Eggan, K., heterochromatin), and facultative (contextually silenced). Humpherys, D., Mastrangelo, M. A., Jun, Z., Walter, J. and Jaenisch, Perhaps the best-studied example of facultative R. (2002). Dnmt1 overexpression causes genomic hypermethylation, loss of imprinting, and embryonic lethality. Mol. Cell Biol. 22, 2124-2135. heterochromatin is the inactive X chromosome (Xi) of female Brenner, S., Pepper, D., Berns, M. W., Tan, E. and Brinkley, B. R. (1981). placental mammals. In mice and humans, the chromatin of Xi Kinetochore structure, duplication, and distribution in mammalian cells: is characterized by its association with the untranslated Xist analysis by human autoantibodies from scleroderma patients. J. Cell Biol. RNA and the presence of abundant macroH2A protein 91, 95-102. Cao, R., Wang, L., Wang, H., Xia, L., Erdjument-Bromage, H., Tempst, (Costanzi and Pehrson, 1998). Many studies have shown that P., Jones, R. S. and Zhang, Y. (2002). Role of histone H3 lysine 27 CpG islands of X-linked genes subject to X-inactivation are methylation in Polycomb-group silencing. Science 298, 1039-1043. highly methylated. However, promoter-proximal CpG islands Costanzi, C. and Pehrson, J. R. (1998). Histone macroH2A1 is concentrated make up only a small portion of available CpG dinucleotides in the inactive X chromosome of female mammals. Nature 393, 599-601. on the X chromosome, and experiments that address global Costanzi, C., Stein, P., Worrad, D. M., Schultz, R. M. and Pehrson, J. R. (2000). Histone macroH2A1 is concentrated in the inactive X chromosome methylation levels on X chromosomes show that the inactive of female preimplantation mouse embryos. Development 127, 2283-2289. X chromosome is markedly undermethylated relative to the Frorath, B., Abney, C. C., Berthold, H., Scanarini, M. and Northemann, active X chromosome (Bernardino et al., 1996; Viegas- W. (1992). Production of recombinant rat interleukin-6 in Escherichia coli Pequignot et al., 1988). using a novel highly efficient expression vector pGEX-3T. Biotechniques 12, 558-563. Our results show that loss of CpG demethylation leads to Hoyer-Fender, S., Czirr, E., Radde, R., Turner, J. M., Mahadevaiah, S. K., the establishment of pericentric chromatin with features Pehrson, J. R. and Burgoyne, P. S. (2004). Localisation of histone reminiscent of the inactive X chromosome, namely, low levels macroH2A1.2 to the XY-body is not a response to the presence of asynapsed of CpG DNA methylation combined with a high content of chromosome axes. J. Cell Sci. 117, 189-198. macroH2A. This finding raises the interesting possibility that Jackson-Grusby, L., Beard, C., Possemato, R., Tudor, M., Fambrough, D., Csankovszki, G., Dausman, J., Lee, P., Wilson, C., Lander, E. et al. DNA methylation can act as a toggle between facultative and (2001). Loss of genomic methylation causes p53-dependent apoptosis and constitutive heterochromatin. This intriguing possibility led us epigenetic deregulation. Nat. Genet. 27, 31-39. to investigate the behavior of other chromatin constituents of Jacobs, S. A. and Khorasanizadeh, S. (2002). Structure of HP1 euchromatin and facultative and constitutive heterochromatin chromodomain bound to a lysine 9-methylated histone H3 tail. Science 295, 2080-2083. by both western blot analysis and immunofluorescence. We Jacobs, S. A., Taverna, S. D., Zhang, Y., Briggs, S. D., Li, J., Eissenberg, found that trimethyl-H3K27, which is a marker of facultative J. C., Allis, C. D. and Khorasanizadeh, S. (2001). Specificity of the HP1 heterochromatin (Cao et al., 2002; Plath et al., 2003), and chromo domain for the methylated N-terminus of histone H3. EMBO J. 20, trimethyl-H3K9, which is a marker of constitutive 5232-5241. heterochromatin (Maison et al., 2002; Peters et al., 2001), were Lachner, M., O’Carroll, D., Rea, S., Mechtler, K. and Jenuwein, T. (2001). Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins. both essentially unaffected by loss of Dnmt1 function. We also Nature 410, 116-120. found that trimethyl-H3K4, which is a marker of euchromatin Ladurner, A. G. (2003). Inactivating chromosomes: a macro domain that (Santos-Rosa et al., 2002), was unaffected by loss of Dnmt1 minimizes transcription. Mol. Cell 12, 1-3. function. These results show that macroH2A is relatively Lei, H., Oh, S. P., Okano, M., Juttermann, R., Goss, K. A., Jaenisch, R. and Li, E. (1996). De novo DNA cytosine methyltransferase activities in

Journal of Cell Science unique in its ability to respond to reductions in levels of DNA mouse embryonic stem cells. Development 122, 3195-3205. methylation. The finding that macroH2A but not trimethyl- Leonhardt, H., Page, A. W., Weier, H. U. and Bestor, T. H. (1992). A H3K27 was affected in demethylated cells suggested targeting sequence directs DNA methyltransferase to sites of DNA that additional complexities may exist in the interplay replication in mammalian nuclei. Cell 71, 865-873. between DNA methylation and facultative heterochromatin Li, E., Bestor, T. H. and Jaenisch, R. (1992). Targeted mutation of the DNA methyltransferase gene results in embryonic lethality. Cell 69, 915- organization. 926. Li, E., Beard, C., Forster, A. C., Bestor, T. H. and Jaenisch, R. (1993). DNA We thank Steve Doxsey for CREST anti-kinetochore serum, and methylation, genomic imprinting, and mammalian development. Cold Therese Doherty for proofreading. This research was supported in part Spring Harb. Symp. Quant. Biol. 58, 297-305. by the Robert Leet and Clara Guthrie Patterson Trust and grant Maison, C., Bailly, D., Peters, A. H., Quivy, J. P., Roche, D., Taddei, A., AG023687 from the National Institutes of Health, USA. Lachner, M., Jenuwein, T. and Almouzni, G. (2002). Higher-order structure in pericentric heterochromatin involves a distinct pattern of histone modification and an RNA component. Nat. Genet. 30, 329-334. Palmer, D. K., O’Day, K., Wener, M. H., Andrews, B. S. and Margolis, R. References L. (1987). A 17-kD centromere protein (CENP-A) copurifies with Allen, M. D., Buckle, A. M., Cordell, S. C., Lowe, J. and Bycroft, M. nucleosome core particles and with histones. J. Cell Biol. 104, 805-815. (2003). The crystal structure of AF1521 a protein from Archaeoglobus Pehrson, J. R. and Fried, V. A. (1992). MacroH2A, a core histone containing fulgidus with homology to the non-histone domain of macroH2A. J. Mol. a large nonhistone region. Science 257, 1398-1400. Biol. 330, 503-511. Peters, A. H., O’Carroll, D., Scherthan, H., Mechtler, K., Sauer, S., Angelov, D., Molla, A., Perche, P. Y., Hans, F., Cote, J., Khochbin, S., Schofer, C., Weipoltshammer, K., Pagani, M., Lachner, M., Kohlmaier, Bouvet, P. and Dimitrov, S. (2003). The histone variant macroH2A A. et al. (2001). Loss of the Suv39h histone methyltransferases impairs interferes with transcription factor binding and SWI/SNF nucleosome mammalian heterochromatin and genome stability. Cell 107, 323-337. remodeling. Mol. Cell 11, 1033-1041. Plath, K., Fang, J., Mlynarczyk-Evans, S. K., Cao, R., Worringer, K. A., Bannister, A. J., Zegerman, P., Partridge, J. F., Miska, E. A., Thomas, J. Wang, H., de la Cruz, C. C., Otte, A. P., Panning, B. and Zhang, Y. O., Allshire, R. C. and Kouzarides, T. (2001). Selective recognition of (2003). Role of histone H3 lysine 27 methylation in X inactivation. Science methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410, 300, 131-135. 120-124. Pradhan, S., Bacolla, A., Wells, R. D. and Roberts, R. J. (1999). Bernardino, J., Lamoliatte, E., Lombard, M., Niveleau, A., Malfoy, B., Recombinant human DNA (cytosine-5) methyltransferase. I. Expression, Dutrillaux, B. and Bourgeois, C. A. (1996). DNA methylation of the X purification, and comparison of de novo and maintenance methylation. J. chromosomes of the human female: an in situ semi-quantitative analysis. Biol. Chem. 274, 33002-33010. Chromosoma 104, 528-535. Rasmussen, T. P., Huang, T., Mastrangelo, M. A., Loring, J., Panning, 1616 Journal of Cell Science 118 (8)

B. and Jaenisch, R. (1999). Messenger RNAs encoding mouse histone (2002). Active genes are tri-methylated at K4 of histone H3. Nature 419, macroH2A1 isoforms are expressed at similar levels in male and female 407-411. cells and result from alternative splicing. Nucleic Acids Res. 27, 3685- Silva, J., Mak, W., Zvetkova, I., Appanah, R., Nesterova, T. B., Webster, 3689. Z., Peters, A. H., Jenuwein, T., Otte, A. P. and Brockdorff, N. (2003). Rasmussen, T. P., Mastrangelo, M. A., Eden, A., Pehrson, J. R. and Establishment of histone h3 methylation on the inactive X chromosome Jaenisch, R. (2000). Dynamic relocalization of histone MacroH2A1 from requires transient recruitment of Eed-Enx1 polycomb group complexes. centrosomes to inactive X chromosomes during X inactivation. J. Cell Biol. Dev. Cell 4, 481-495. 150, 1189-1198. Thorne, A. W., Cary, P. D. and Crane-Robinson, C. (1998). Extraction and Rea, S., Eisenhaber, F., O’Carroll, D., Strahl, B. D., Sun, Z. W., Schmid, separation of core histones and non-histone chromosomal proteins. In M., Opravil, S., Mechtler, K., Ponting, C. P., Allis, C. D. et al. (2000). Chromatin: A Practical Approach (ed. H. Gould), pp. 35-57. Oxford, UK: Regulation of chromatin structure by site-specific histone H3 Oxford University Press. methyltransferases. Nature 406, 593-599. Viegas-Pequignot, E., Dutrillaux, B. and Thomas, G. (1988). Inactive X Santos-Rosa, H., Schneider, R., Bannister, A. J., Sherriff, J., Bernstein, chromosome has the highest concentration of unmethylated Hha I sites. B. E., Emre, N. C., Schreiber, S. L., Mellor, J. and Kouzarides, T. Proc. Natl. Acad. Sci. USA 85, 7657-7660. Journal of Cell Science