Epigenetic Dysregulation by Nickel Through Repressive Chromatin Domain Disruption

Epigenetic dysregulation by nickel through repressive chromatin domain disruption

Cynthia C. Josea,1, Beisi Xub,1, Lakshmanan Jagannathana, Candi Tracb, Ramya K. Mallelaa, Takamitsu Hattoric, Darson Laic, Shohei Koidec, Dustin E. Schonesb,d,2, and Suresh Cuddapaha,2

aDepartment of Environmental Medicine, New York University School of Medicine, Tuxedo, NY 10987; bDepartment of Cancer Biology, Beckman Research Institute, City of Hope, Duarte, CA 91010; dIrell and Manella Graduate School of Biological Sciences, City of Hope, Duarte, CA 91010; and cDepartment of Biochemistry and Molecular Biology, The University of Chicago, Chicago, IL 60637

Edited by Mark Groudine, Fred Hutchinson Cancer Research Center, Seattle, WA, and approved August 29, 2014 (received for review April 15, 2014)

Investigations into the genomic landscape of histone modifications a multitude of human health risks including allergic dermatitis, in heterochromatic regions have revealed histone H3 lysine 9 bronchitis, pulmonary fibrosis, pulmonary edema, diseases of dimethylation (H3K9me2) to be important for differentiation and the kidney and cardiovascular system, and lung and nasal can- maintaining cell identity. H3K9me2 is associated with gene silencing cers (15, 16). Despite the conclusive health risks of Ni com- and is organized into large repressive domains that exist in close pounds, the underlying mechanisms are not well understood proximity to active genes, indicating the importance of mainte- because their mutagenic potential is very low and does not nance of proper domain structure. Here we show that nickel, correlate with its potent toxicity (17). Recent evidence suggests a nonmutagenic environmental carcinogen, disrupted H3K9me2 that Ni could alter transcriptional regulation through epigenetic domains, resulting in the spreading of H3K9me2 into active re- alterations (18–20). gions, which was associated with gene silencing. We found weak In the transgenic gpt+ Chinese hamster V79-derived G12 cells, CCCTC-binding factor (CTCF)-binding sites and reduced CTCF bind- silencing of the bacterial xanthine-guanine phosphoribosyltransferase ing at the Ni-disrupted H3K9me2 domain boundaries, suggesting a loss of CTCF-mediated insulation function as a potential reason (gpt) transgene following Ni exposure isassociatedwithanincrease for domain disruption and spreading. We furthermore show that in H3K9me2 and a decrease in H3 and H4 acetylation at the pro- euchromatin islands, local regions of active chromatin within large moter and coding regions (13, 14, 21). Intriguingly, Ni-induced si- H3K9me2 domains, can protect genes from H3K9me2-spreading– lencing of gpt and increased H3K9me2 levels occurred only when associated gene silencing. These results have major implications the transgene was integrated near a heterochromatic region (13, in understanding H3K9me2 dynamics and the consequences of 14, 21). This suggests that Ni could induce H3K9me2 spreading. chromatin domain disruption during pathogenesis. Therefore, we reasoned that studying the effects of Ni exposure on the epigenome could provide an opportunity to understand the insulator | nickel toxicity | nickel carcinogenesis dynamics of H3K9me2 domain maintenance and the mechanistic basis of potential domain disruption and aberrant gene silencing. nvestigations into the histone modification landscape of eukary- Iotic genomes have revealed organization of the chromatin into Significance functionally distinct active and silent domains (1–4). Although active histone modifications typically display a punctate pat- Histone modifications associated with gene silencing typically tern, the silencing histone modifications such as histone H3 mark large contiguous regions of the genome forming re- lysine 9 dimethylation (H3K9me2) and H3K27me3 mark large pressive chromatin domain structures. Since the repressive regions of the genome-forming repressive domains (1, 2, 5, 6). domains exist in close proximity to active regions, maintenance Organization of active and silent chromatin regions into com- of domain structure is critically important. This study shows partments has direct implications on the establishment of gene that nickel, a nonmutagenic carcinogen, can disrupt histone H3 expression patterns and cell-type specificity (3, 7). lysine 9 dimethylation (H3K9me2) domain structures genome- Recent evidence indicates that the H3K9me2 domains are wide, resulting in spreading of H3K9me2 marks into the highly dynamic and important for differentiation and maintain- active regions, which is associated with gene silencing. Our ing cell identity (8–10). Although H3K9me2 marks large domains results suggest inhibition of DNA binding of the insulator of mostly transcriptionally silent regions of the genome, small protein CCCTC-binding factor (CTCF) at the H3K9me2 domain windows of active chromatin have been characterized within boundaries as a potential reason for H3K9me2 domain dis- the large heterochromatin domains, termed euchromatin islands ruption. These findings have major implications in under- SYSTEMS BIOLOGY (EIs) (9, 11). These EIs have been shown to be enriched for ac- standing chromatin dynamics and the consequences of tive histone modifications, DNase I hypersensitive (HS) sites, chromatin domain disruption during pathogenesis. and CCCTC-binding factor (CTCF) binding (11). Existence of H3K9me2 domains in close proximity to active genes suggests Author contributions: C.C.J., B.X., D.E.S., and S.C. designed research; C.C.J., B.X., L.J., C.T., that delimitation of these domains is critical in maintaining cel- and R.K.M. performed research; T.H., D.L., and S.K. contributed new reagents/analytic lular identity. Spreading of H3K9me2 beyond its domain bound- tools; C.C.J., B.X., L.J., D.E.S., and S.C. analyzed data; and C.C.J., B.X., D.E.S., and S.C. wrote aries is associated with gene silencing at the chicken β-globin HS4 the paper. locus (12). Moreover, discrete local changes in H3K9me2 are The authors declare no conflict of interest. associated with corresponding gene expression changes during This article is a PNAS Direct Submission. cellular differentiation (9), further emphasizing the importance of Freely available online through the PNAS open access option. maintaining the structural integrity of the H3K9me2 domains for Data deposition: The data reported in this paper have been deposited in the Gene Ex- cell-type specificity. However, how H3K9me2 domains are main- pression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo (accession no. GSE56053). tained and the consequences of aberrant domain disruption are 1C.C.J. and B.X. contributed equally to this work. poorly understood. An important clue toward understanding the 2To whom correspondence may be addressed. Email: [email protected] or H3K9me2 domains comes from studies that investigated silencing [email protected]. of a transgene following the exposure of cells to nickel (13, 14). This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. Nickel compounds are environmental carcinogens that cause 1073/pnas.1406923111/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1406923111 PNAS | October 7, 2014 | vol. 111 | no. 40 | 14631–14636 Downloaded by guest on September 26, 2021 In this study, we characterized the changes in the global pro- spreading of H3K9me2 into previously non-H3K9me2 regions files of several active and repressive histone modifications in the (Fig. 1E). Interestingly, even though the total levels of H3K27me3 Ni-exposed human lung cells. Our data showed that Ni exposure significantly increased in Ni-exposed cells (Fig. 1A), the domains of triggered H3K9me2 domain disruption, resulting in spreading H3K27me3 did not substantially change (Fig. S1 B–D). of H3K9me2 into the active regions, which was associated with We then set out to characterize all of the regions of the ge- gene silencing. We found weak CTCF-binding sites and reduced nome that displayed H3K9me2 domains that increased in size CTCF binding at the Ni-disrupted H3K9me2 domain bound- following Ni treatment, which we refer to as “spreading domains” aries. Moreover, CTCF knockdown resulted in H3K9me2 spread- (Fig. 1 F and G and SI Materials and Methods). After saturation ing. These results suggest a loss of CTCF-mediated insulation analysis of the ChIP-Seq reads to ensure sufficient coverage (SI function as a potential reason for domain disruption and spreading. Materials and Methods and Fig. S2), we called all domains in both Because epigenetic alterations are believed to be the basis of Ni- control (Ctrl) and Ni-treated (Ni) cells using RSEG software (25). induced cancers (14, 16), our finding of H3K9me2 domain dis- To validate H3K9me2 ChIP-Seq data and to optimize H3K9me2 ruption by Ni and a potential role for CTCF in maintaining domain calling, we selected several loci and performed ChIP- H3K9me2 domains will have profound implications in under- quantitative PCR (qPCR) analysis on untreated and Ni-treated standing the process of carcinogenesis. cells (Fig. 1D and SI Materials and Methods). ChIP-qPCR analysis strongly correlated with the ChIP-Seq data (Fig. S3). Fur- Results thermore, using a peptide immunoprecipitation (IP) assay, we Ni Exposure Leads to Genome-Wide H3K9me2 Domain Spreading. To evaluated the specificity and affinity of the antibody used in this investigate the Ni-induced epigenetic changes, we treated im- study for performing H3K9me2 ChIP-Seq (SI Materials and mortalized noncancerous human bronchial epithelial BEAS-2B Methods) (26). The antibody showed moderate affinity with a μ cells to a noncytotoxic concentration of 500 M NiCl2 for 72 h, submicromolar KD value and outstanding specificity with almost recapitulating the concentration used in several earlier studies no cross-reactivity to other histone modification marks tested that investigated the impact of Ni exposure on BEAS-2B cells (Fig. S4), suggesting the high quality of the antibody. (22–24). Western blotting analysis showed significant increase We identified 2,461 control domains and 1,298 Ni domains, in the total levels of the repressive marks H3K9me2 and ranging in size from 1 kb to >1 Mb (Fig. 1G). A total of 61% H3K27me3 (Fig. 1A). To identify the genes that are differentially (1,514/2,461) of the domains in untreated control cells increased expressed due to Ni exposure, we performed RNA-Seq analysis. in size upon Ni exposure, indicating spreading of these domains We then examined the levels of H3K9me2 and H3K27me3 at beyond their normal boundaries (Fig. 1G and Fig. S5). The total candidate Ni-downregulated gene promoters using chromatin number of spreading domains reduced to a third of the domains immunoprecipitation (ChIP) analysis (Fig. 1 B and C). Although detected in control cells (562/1,514) due to spreading domains we found appreciable increase in the levels of H3K9me2 (Fig. combining to form larger domains (Fig. 1G and Fig. S5). 1C), we did not find any changes in the levels of H3K27me3 (Fig. S1A). To investigate the effect of Ni exposure on the genome- Genes in H3K9me2-Spreading Domains Are Generally Downregulated. wide distribution of repressive marks, we mapped H3K9me2 and As shown in Fig. 1 B and E and Fig. S3, MAP2K3, DKK1, TAGLN, H3K27me3 in untreated and Ni-treated cells, using ChIP-Seq and CTGF were strikingly downregulated upon Ni exposure, and (Fig. 1 D and E and Fig. S1 B and C). We noted a striking this loss of gene expression was associated with increase in reorganization of H3K9me2 domains in Ni-treated cells with promoter H3K9me2 levels, consistent with the earlier studies

A B C 30 H3K9me2 D F Spreading Domain 2 1.4 ** BEAS-2B Control NiCl 1.2 25 ** > 90% in Control H3K9me2 1 20 0.8 Control 15 Control + NiCl2 H3K9me2 0.6 *** H3K27me3 Nickel 10 Nickel 0.4

mRNA levels mRNA 5 0.2 0h (Ctrl) 3d (Ni) H3K9me2 H3 0 0 MAP2K3 DKK1 TAGLN Fold enrichment/IgG MAP2K3 DKK1 TAGLN ChIP-seq H3K9me2 Prelim Domain Calling E chr6 Saturation Analysis 10 kb hg19 qPCR domain validation G All Domains Spreading Domains 7 ChIP-qPCR Primer Position 6 ** Control: 2461 Control: 1514 327 _ 5 0.6 Nickel: 1298 Nickel: 582 Ctrl RNA-seq 4 Ctrl H3K9me2 1 _ 3 Ni H3K9me2 327 _ 2 0.4 1

Ni RNA-seq Fold enrichment/IgG 0 1 _ CTGF Density 0.2 Ctrl Domain Calls Domain calling optimization 13 _ 0.0 Ctrl H3K9me2 1e+04 1e+06 1e+04 1e+06 1 _ Final Domain Calling Domain size (bp) Domain size (bp) Ni Domain Calls 13 _ 2461 Control Domains Ni H3K9me2 1298 Ni Domains 1 _ CTGF

Fig. 1. H3K9me2 domains spread upon NiCl2 treatment. (A) Western blotting analysis showing H3K9me2 and H3K27me3 levels in BEAS-2B cells exposed to 500 μM NiCl2 for 72 h. (B) qPCR analysis showing downregulation of candidate gene expression in Ni-treated cells. GAPDH was used as internal control. (C) ChIP-qPCR analysis showing increase in H3K9me2 levels at candidate downregulated gene promoters. (D) Workflow for H3K9me2 genome-wide profiling and domain calling. (E) Gene expression (RNA-Seq) and H3K9me2 (ChIP-Seq) profiles at the tumor suppressor CTGF gene locus in the untreated and Ni-treated BEAS-2B cells. In the untreated cells, depletion of H3K9me2 indicates active chromatin. Nickel exposure resulted in H3K9me2 spreading into the CTGF genic region (box), which was associated with downregulation of gene expression. The blue and red bars represent H3K9me2 domains [RSEG analysis (25)] in the control and Ni-exposed cells, respectively. (Inset) ChIP-qPCR analysis showing increase in H3K9me2 levels at CTGF promoter following Ni exposure. (F) Cartoon representing the H3K9me2 domain spreading upon Ni exposure. Spreading domains are defined as domains that are larger in Ni-treated cells, with the Ni domain overlapping at least 90% of the control domain. (G) Size distribution of H3K9me2 domains in untreated and Ni-treated cells for all domains and spreading domains. A total of 1,514 of the H3K9me2 domains in control cells (blue) increase in size after nickel exposure (red), indicating domain spreading. In addition, a number of small domains combine following Ni exposure, as indicated by the decrease in the number of domains in the Ni-exposed cells. For qPCR, statistical significance was evaluated using t test: **P < 0.001; ***P < 0.0001.

14632 | www.pnas.org/cgi/doi/10.1073/pnas.1406923111 Jose et al. Downloaded by guest on September 26, 2021 that showed association of H3K9me2 spreading with decreased revealed depletion of H3K9me2 immediately surrounding the gene expression (27–30). To examine this on a genome-wide region marked by H3K9ac (Fig. S6). We then classified EIs across scale, we stratified all of the genes into those in H3K9me2- the genome in both control and Ni-treated cells, as H3K9ac peaks spreading domains (3,930) and genes that are not in H3K9me2 in within H3K9me2 domains, resulting in 8,485 EIs in control cells either control or Ni-exposed cells (8,096) (Fig. 2A). Examination and 10,715 EIs in Ni-treated cells with 5,615 EIs in H3K9me2- of the RNA-Seq fragments per kilobase of exon per million spreading domains in Ni-treated cells. Examining H3K4me3, fragments mapped (FPKM) values for genes in each set revealed H3K9ac, and H3K9me2 profiles around the transcription start that genes in the spreading domains were significantly down- sites for genes in H3K9me2-spreading domains with and without regulated compared with genes not in H3K9me2 domains (Fig. EIs indicated that genes in EIs within spreading domains have 2B; Wilcoxon rank-sum test, P = 1.729e-13). Despite this, we also H3K4me3 in addition to H3K9ac marks and have drastically re- detected a number of genes that were not downregulated, leading duced H3K9me2 signals (Fig. 3A). Furthermore, genes within EIs us to further investigate this phenomenon. It was recently reported in spreading domains are protected from Ni silencing compared that H3K4me3 and H3K9ac marks could exist within stable with genes without EIs in spreading domains (Wilcoxon rank-sum H3K9me2 domains, forming EIs (9, 11). To investigate the po- test of FPKM distributions with P ≤ 0.01; see SI Materials and tential presence of EIs at the nondownregulated H3K9me2- Methods for details). These results suggest that EIs can function – spreading genes, we mapped the genome-wide distribution of to prevent H3K9me2-spreading associated silencing, analogous to H3K9ac in control and Ni-exposed cells using ChIP-Seq. We their characterized function in stem cell differentiation (11). To examined the ratio of H3K9ac ChIP-Seq reads in Ni-treated cells further investigate the role of EIs in preventing gene silencing, we compared with control cells in promoter regions of genes that tabulated the genomic distributions of EIs in both control and Ni- were stratified by expression (Fig. 2C). Our analysis indicated treated cells (Fig. 3B). We found that the distribution of EIs in that the upregulated H3K9me2-spreading genes were marked control cells was biased toward intergenic regions vs. promoters with higher levels of H3K9ac in promoter regions (Fig. 2C; whereas in Ni-treated cells the distribution was biased toward – promoters vs. intergenic regions (Fig. S7; Fisher’s exact test Kolmogorov Smirnov test on H3K9ac read numbers in 4-kb < windows results in P < 2.2e-16 for upregulated vs. downregulated P 2.2e-16). These results indicate that H3K9me2 spreading genes), suggesting the potential presence of EIs at these loci. preferentially disrupted EIs in the intergenic regions.

EIs Protect Genes from H3K9me2-Spreading–Associated Silencing. To Disrupted H3K9me2 Domain Boundaries Are Associated with Weaker systematically analyze the potential roles of EIs in preventing CTCF-Binding Sites. Several studies have demonstrated enrichment H3K9me2-spreading–associated gene silencing, we searched for of CTCF at heterochromatin domain boundaries (3, 31). To H3K9ac peaks (see SI Materials and Methods for details) that were investigate the potential role of CTCF in maintaining H3K9me2 unperturbed in H3K9me2-spreading domains, as exemplified by boundaries in Ni-exposed cells, we separated H3K9me2 domains into those that are maintained after Ni treatment and those that the NEO1 promoter (Fig. S6). Closer inspection of the promoter are disrupted (SI Materials and Methods). We then evaluated the best position weight matrix (PWM) score of the CTCF motif [TRANSFAC M00129 (32)] at the domain boundaries. Dis- A B rupted boundaries had a significantly lower maximum PWM H3K9me2 * score compared with the nondisrupted boundaries (Fig. 4A; t test, P = 0.0088), suggesting that stronger CTCF-binding sites could prevent H3K9me2 domain disruption and spreading. H3K9me2 To investigate this in detail, we mapped CTCF-binding sites in control and Ni-exposed cells using ChIP-Seq. Analysis of Genes not in H3K9me2 Domains the ChIP-Seq profiles at disrupted and maintained H3K9me2 Genes in spreading H3K9me2 Domains FPKM (Ni/Ctrl) boundaries confirmed the presence of stronger CTCF signals at − 0.5 0.0 0.5 1.0 1.5 the maintained boundaries (Fig. 4B). To examine if the presence

− 1.0 of the canonical CTCF motif was correlated with the retention of CTCF binding upon Ni exposure, we stratified CTCF ChIP-Seq C Genes in Spreading H3K9me2 − 1.5 peaks into two categories: (i) CTCF sites that were retained upon 276 Downregulated genes Ni exposure and (ii) CTCF sites that were lost (SI Materials and 160 Upregulated genes H3K9me2 620 genes unchanged (1/1.1 < FC < 1.1) Methods). Evaluation of the prevalence of the canonical CTCF H3K9me2 motif in these two categories showed that the retained CTCF sites * 1.6 were much more likely to have the canonical CTCF motif compared with the CTCF sites that were lost upon Ni exposure (Fig. 1.4 4C). Interestingly, de novo motif discovery analysis (SI Materials SYSTEMS BIOLOGY 1.2 and Methods) at disrupted CTCF peaks indicated the presence of 1 a CTCF motif that was previously characterized to be associated with weaker CTCF-binding compared with the canonical CTCF 0.8 H3K9ac (Ni/Ctrl) site (Fig. 4D) (33). Furthermore, evaluating the levels of

0.6 0.8 1.0 1.2 1.4 1.6 H3K9me2 at the lost and retained CTCF sites demonstrated that -2000 -1000 0 1000 2000 the lost CTCF sites do indeed have enrichment of H3K9me2 H3K9ac (Ni/Ctrl) TSS frag counts Position relative to TSS relative to retained sites (Fig. 4E). These results suggest that Ni could potentially affect DNA binding of CTCF with the effect Fig. 2. H3K9me2 spreading is associated with downregulation of gene being more prevalent at the weaker CTCF-binding sequences. expression. (A) Cartoon representing the two categories of genes used for evaluating the impact of H3K9me2 spreading. Blue arrow: genes that were Nickel Inhibits DNA Binding of CTCF in Vitro. Ni-induced reduction incorporated into H3K9me2 domains upon Ni exposure; green arrow: genes that were not located within H3K9me2 domains. (B) Comparison of the gene in CTCF binding at the sites containing weak CTCF-binding expression (FPKM) of the two categories of genes. The genes in the sequences suggests that Ni could potentially interfere with the spreading domains display the greatest downregulation. (C) The ratio of DNA-binding ability of CTCF (Fig. S8 A and B). Zinc (Zn) H3K9ac signal at genes in different expression groups indicates that genes finger structures are important targets of toxic metal com- that are upregulated despite H3K9me2 spreading have higher H3K9ac sig- pounds. Several metals have been proposed to exert toxicity nals compared with the downregulated genes. through their ability to bind Zn finger proteins (34). To examine

Jose et al. PNAS | October 7, 2014 | vol. 111 | no. 40 | 14633 Downloaded by guest on September 26, 2021 A Genes in EI of spreading H3K9me2 Genes not in EI of spreading H3K9me2 Previous studies have shown that the JmjC domain H3K9me2 Nickel 40 Nickel 3 Nickel demethylases could be inhibited by Ni, resulting in an increase in 100 the total levels of H3K9me2 (19, 36, 37). Taken together, our 20 2 50 H3K9me2 H3K9ac results suggest that, although depletion of CTCF could initiate H3K4me3 1 0 0 spreading of H3K9me2 and gene silencing, the effect is more −10k −5k 0 5k 10k −10k −5k 0 5k 10k −10k −5k 0 5k 10k TSS TSS TSS robust in Ni-exposed cells, where reduction in CTCF binding could be accompanied by a reduction in demethylase activity, 12.9 % B 14.3 % 26.3 % 20.8 % thus augmenting H3K9me2 spreading and gene silencing (Fig. 5 3.0 % F and G). Interestingly, MAP2K3 gene expression level was main-

4.2 % 3.9 % tained in CTCF knockdown cells with no changes in H3K9me2 3.1 % levels. However, it was significantly downregulated by nickel, 41.6 % 6.9 % 46.3 % accompanied by an increase in H3K9me2 (Fig. 5 F and G). This 4.6 % 50.0 % suggests that mere loss of CTCF binding is insufficient for 45.2 % Euchroma n H3K9me2 spreading into the MAP2K3 promoter and its down- Genome Euchroma n Island in Nickel Island in Control regulation and that other mechanisms such as the inhibition of

% Promoter (<=1000 bp) % Downstream (<=1000 bp) % Coding exon % 5'UTR demethylases are required. % Promoter (1000−2000 bp) % Downstream (1000−2000 bp) % Intron % 3'UTR % Promoter (2000−3000 bp) % Downstream (2000−3000 bp) % Distal intergenic Discussion In this study, we used nickel exposure to understand the dynamics Fig. 3. Euchromatin islands protect genes from H3K9me2-spreading–associated silencing. (A) H3K4me3, H3K9ac, and H3K9me2 profiles in Ni-treated of H3K9me2 domains. Ni is a nonmutagenic carcinogen. Whereas cells for genes in EIs in spreading domains and for genes not in EIs in spreading Ni affects several aspects of cellular regulation (17, 23, 38), domains. (B) Genes in EIs are protected from silencing by H3K9me2 spreading. epigenetic alterations induced by Ni play an important role in inducing gene expression changes (16, 21). Because H3K9me2 is organized into large repressive domains, Ni silencing of a trans- the potential of Ni to bind CTCF, we performed in silico mo- gene occurring only if it is positioned close to the heterochromatic + lecular docking analysis between Ni2 and CTCF (SI Materials locus (13) suggested disruption of H3K9me2 domain structure. We + and Methods). Our analysis showed the binding affinity for Ni2 used this phenomenon to understand the dynamics of H3K9me2 + to CTCF to be −24.0 kcal/mol. The Ni2 -binding affinity of the domains on a genome-wide scale. Our results show that Ni can known Ni-interacting Zn finger protein XPA was −6.93 kcal/mol, induce H3K9me2, which corresponded with the downregulation + indicating a very strong affinity of Ni2 to CTCF. of gene expression. This suggests breakdown of H3K9me2 To investigate if Ni could interfere with CTCF DNA bind- chromatin domain structure during Ni-induced gene silencing. ing in vitro, we performed electrophoretic mobility shift Local discrete changes in H3K9me2, associated with corre- assay (EMSA). As a CTCF-binding sequence, we used a DNA sponding gene expression changes during neuronal differentia- fragment containing the well-characterized CTCF-binding site tion of embryonic stem cells (9), suggested a role for H3K9me2 at the APP locus (Fig. S9A) (35). When incubated with BEAS- 2B nuclear extracts, retardation in the migration of the wild- type probe due to the formation of a protein/DNA complex was observed (Fig. 5A). To test the effect of Ni exposure on CTCF sites with motif A Disrupted Boundaries B C 2587 Disrupted Boundaries sites without motif DNA binding, we incubated the nuclear extracts with various Maintained Boundaries 314 Maintained Boundaries % sites with motif concentrations of NiCl2 before addition of the probe. In- 2587 Random Boundaries ** creasing concentrations of NiCl2 progressively affected the gel 30 retardation (Fig. 5B), suggesting concentration-dependent in- 1.0 20 0.5 hibition of CTCF/DNA complex formation by NiCl2. Because

0.0 Number of Sites

we found H3K9me2 spreading to be more prevalent in the CTCF ChIP-seq signal −1000 −500 0 500 1000 boundaries containing weak CTCF sites (Fig. 4), we next asked if of Boundary 10 Position relative to best motif match w/in 2kb of boundary

the CTCF binding is more prone to Ni-induced disruption at the Best PWM Score within 2kb 0 CTCF sitesCTCF sites weak binding sites compared with the strong sites. To answer this disrupted retained Motif in disrupted CTCF peaks question, we designed oligonucleotides encompassing two strong DE 1190 Disrupted CTCF binding sites in Nickel 2 13857 Retained CTCF binding sites in Nickel (Fig. S9 B and C)andtwoweak(Fig. S9 D and E) CTCF-binding 1 bits T A A AC 2.0 Control C T GA sites and performed EMSA analysis. Lower concentrations of 0 G A GG TT 5 10 1.5 NiCl2 could inhibit CTCF binding at the weak binding sites (Fig. 5D) compared with the strong sites (Fig. 5C), suggesting Nickel that the strength of the CTCF-binding site is inversely corre- H3K9me2 2.0 1.6 lated with the Ni-mediated inhibition of its DNA binding. 1.2

To obtain further insight into the role of CTCF at H3K9me2 CTCF Motif (Nakahashi et al.) −2000 −1000 0 1000 2000 domain boundaries, we knocked down CTCF in BEAS-2B cells CTCF (Fig. 5E). To investigate whether CTCF depletion affected gene Fig. 4. Ni-disrupted H3K9me2 domain boundaries have weak CTCF-binding expression, we examined the mRNA levels of genes located near sites. (A) PWM scores of the CTCF-binding sequence motifs at the nondisrupted the H3K9me2 domain boundaries. We observed a reduction in H3K9me2 domain boundaries (blue) are significantly higher than those of the the expression of several genes (Fig. 5F). ChIP analysis at the motifs at the Ni-disrupted boundaries (red), indicating the disruption of promoter regions of these genes showed increase in the levels boundaries with weaker CTCF motifs (t test, P = 0.0088). (B)CTCFChIP-Seq of H3K9me2 at ANXA1, DKK1, and ID3 promoters (Fig. 5G), signals are stronger at the nondisrupted boundaries (green) compared with suggesting spreading of H3K9me2 in CTCF knockdown cells. the Ni-disrupted boundaries (red). (C) CTCF-binding sites that are retained These results support a role for CTCF in maintaining H3K9me2 after Ni exposure possess canonical CTCF-binding motif whereas the sites that are lost after Ni exposure did not. (D) CTCF-binding sites that were lost upon Ni domains. We then compared the gene expression and H3K9me2 exposure possessed DNA sequence motifs that were similar to the weak alterations between CTCF knockdown cells and Ni-exposed downstream motif described by Nakahashi et al. (33) Reproduced with per- cells. Surprisingly, both the decrease in the gene expression (Fig. mission from Nakahashi et al. (33). (E) H3K9me2 signals at disrupted and 5F) and the increase in H3K9me2 levels (Fig. 5G) were larger retained CTCF-binding sites following Ni treatment. Disrupted CTCF sites have in Ni-exposed cells compared with the CTCF knockdown cells. substantially greater H3K9me2 levels before and after Ni treatment.

14634 | www.pnas.org/cgi/doi/10.1073/pnas.1406923111 Jose et al. Downloaded by guest on September 26, 2021 A B C D 2L11XD ENAH D 91PAGHRA DTL Antibodies - - - IgG CTCF 0 NiCl (μM)0 0 50 200 300 0 0 Competitors - M WT - - 2 50 100 200 300 50 100 200 300 100 50 100 200 300 50 100 200 300

1 2 3 4 5 Strong CTCF binding sites Weak CTCF binding sites

*** 1.2 7 E F * *** *** *** ** G ** * *** *** * ** * 6 1 ** *** *** ** * 5 * 0.8 * shControl shCTCFs ** 4 ** ** CTCF 0.6 Control * Control 3 ** * 0.4 CTCF KDD ** CTCF KD 2 Gene expression Ni 3 daysys H3K9me2 levels Ni 3 days 0.2 1

0 0

ID3 ID3 ANXA1 CTGF DKK1 ANXA1 CTGF DKK1 MAP2K3 MAP2K3

Fig. 5. Nickel inhibits CTCF DNA binding. (A) EMSA showing inhibition of DNA binding of CTCF by NiCl2 at APP locus. The protein/DNA complex formed (lane 1) was competed by a 100-fold excess of the unlabeled wild-type probe (WT) (lane 3), but not the mutant probe (M) (lane 2). CTCF antibody supershifted the

complex (lane 5), whereas the preimmune serum did not (lane 4). (B) NiCl2 in the EMSA reaction mix inhibited binding of CTCF to the APP CTCF-binding site in a concentration-dependent manner, as seen by the loss of the shifted band. Although higher concentrations of NiCl2 in the EMSA reaction mix were required to disrupt CTCF binding at the strong binding sites (C), CTCF binding at the weaker binding sites were disrupted by relatively lower concentrations of NiCl2 (D). Arrow indicates CTCF/DNA complex in all lanes in panels A–D.(E) Western blot analysis showing CTCF depletion in BEAS-2B cells infected with lentiviral shCTCF constructs. β-Actin was used as loading control. (F) Gene expression analysis (qPCR) of candidate genes in CTCF knockdown cells and cells treated with

500 μM NiCl2 for 72 h. GAPDH was used as internal control. mRNA levels of control cells were normalized to 1. (G) ChIP-qPCR analysis showing H3K9me2 levels at downregulated gene promoters in CTCF knockdown cells and cells treated with 500 μM NiCl2 for 72 h. Fold enrichment of H3K9me2 levels over IgG were calculated and the H3K9me2 levels of control cells were normalized to 1. For both gene expression and ChIP analysis, untreated cells and cells infected with nonspecific sequence containing shControl were used as controls for Ni treatment and CTCF knockdown, respectively. Statistical significance was evaluated using t test: *P < 0.05; **P < 0.001; ***P < 0.0001.

in fine-tuning gene expression by functioning as a regulatory suggested a role for CTCF in establishing functional expres- switch. Ni-induced spreading of H3K9me2 is reminiscent of the sion domains (3, 43, 44). Although loss of CTCF was shown to local discrete changes, suggesting that misregulation of this result in spreading of H3K27me3 at the HoxA locus (45), no regulatory switch results in altered gene expression. Further- evidence of H3K27me3 spreading was detected following CTCF more, several earlier studies have suggested ectopic spreading of knockdown in Drosophila (43, 46, 47). Therefore, the domain H3K9me2 to be an important signal for transcriptional silencing barrier activity of CTCF has remained inconclusive (43, 48). Our (27–30, 39). However, the alternate possibility of gene silencing results showing disruption of H3K9me2 domain boundaries at resulting in spreading of H3K9me2 in a subset of the spreading weak CTCF-binding sites suggest a role for CTCF in H3K9me2 domains cannot be ruled out. Nevertheless, H3K9me2 spreading domain maintenance. Previous studies have shown that Ni in- into active regions remains a significant event because this could teraction with Zn finger proteins results in inhibition of DNA play a role in stably silencing the genes, given that H3K9me2 binding and alteration of DNA-binding specificity (49, 50). In- could be a precursor to DNA methylation and long-term gene terestingly, our results suggest that nickel at noncytotoxic con- silencing (40, 41). centrations preferentially inhibits binding of CTCF to the weaker Although spreading of H3K9me2 into promoters was associ- binding sequences (Fig. 5 and Fig. S9), resulting in H3K9me2 ated with reduction in gene expression, we detected a subset of domain disruption and spreading. A recent work has indicated genes whose expression was not affected by H3K9me2 spreading. a remarkable diversity of DNA sequence motifs for CTCF SYSTEMS BIOLOGY Previous studies have characterized short H3K9me2-depleted binding, with specific motifs associated with strength of CTCF EIs within large H3K9me2 domains in both pluripotent and binding (33). Intriguingly, we identified a motif at the weak differentiated cells (11). Active modifications of H3K4me3 and CTCF-binding sites that was previously shown to be associated H3K9ac are hallmarks of EIs (11). Our results suggest that the with weaker CTCF binding (33). repressive effect of H3K9me2 could be overcome by the high Furthermore, our results show that CTCF depletion results in levels of H3K9ac at the EIs, indicating the existence of an ad- downregulation of gene expression, accompanied by an increase ditional level of regulation that potentially could overcome the in the H3K9me2 levels, which is suggestive of H3K9me2 spreading silencing effect of H3K9me2 spreading. Acetylation maintains an in CTCF knockdown cells (Fig. 5). These results suggest a role for open chromatin structure, allowing transcription factor binding CTCF in H3K9me2 domain barrier function. Interestingly, the (42). It remains to be seen if potential binding of any transcrip- levels of gene silencing (Fig. 5F) and enrichment of H3K9me2 tional activator(s) to the strongly acetylated regions has a role in (Fig. 5G) were larger in Ni-treated cells compared with the protecting the gene from epigenetic repression. CTCF knockdown cells. This suggests involvement of other Interestingly, Ni-disrupted H3K9me2 domain boundaries pre- mechanism(s) in addition to loss of CTCF binding during Ni- dominantly contained weaker CTCF-binding motifs and displayed induced gene silencing. H3K9me2 demethylation is catalyzed by lower CTCF ChIP-Seq signals compared with the nondisrupted JmjC domain histone lysine demethylases (JHDMs), which are domain boundaries. Previously, several studies, including ours, have members of the dioxygenase superfamily of enzymes containing

Jose et al. PNAS | October 7, 2014 | vol. 111 | no. 40 | 14635 Downloaded by guest on September 26, 2021 iron at its catalytic center. The demethylation of lysines by understanding pathogenesis induced by aberrant alterations to JHDMs occurs by catalyzing the generation of oxidized iron. the chromatin structure. Ni inhibits this family of enzymes by displacing the iron (19, 36, 37). Previously, loss of H3K9 demethylase LSD1 has been Materials and Methods demonstrated to result in spreading of H3K9me2 (51). Full details are in SI Materials and Methods. Therefore, inactivation of H3K9me2 demethylases could be a BEAS-2B cells were cultured in Dulbecco’s modified Eagle’s medium (Cellgro). μ contributing factor for Ni-induced H3K9me2 spreading and For Ni treatment, the cells were exposed 500 MNiCl2 (Sigma N6136) for 72 h. gene silencing. ChIP-Seq and RNA-Seq data were deposited in the Gene Expression Omnibus In conclusion, our studies demonstrate that Ni exposure can (www.ncbi.nlm.nih.gov/geo) under accession no. GSE56053. inhibit CTCF DNA binding at the weak binding sites. We show ACKNOWLEDGMENTS. We thank Drs. M. Costa and A. Barski and the members that H3K9me2 domain disruption and spreading preferentially of the D.E.S. laboratory for helpful discussions and critical reading of the manu- occurs at the boundaries containing weak CTCF-binding sites, script. This work was supported by National Institutes of Health, National Institute which is associated with downregulation of gene expression. of Environmental Health Sciences Grant R01ES023174, National Institutes of Because chromatin domain structures form an important basis Health, National Institute of Environmental Health Sciences Center of Excellence Pilot Project Grant P30ES000260 (to S.C.), and National Institutes of Health Grant for the establishment of gene expression profiles (3), the K22HL101950 (to D.E.S.). Research reported in this publication includes work per- mechanistic insights that we provide on the underlying causes formed in the Integrative Genomics Core of the City of Hope supported by the of domain disruption will have profound implications in National Cancer Institute, National Institutes of Health Award P30CA33572.

1. Barski A, et al. (2007) High-resolution profiling of histone methylations in the human 26. Nishikori S, et al. (2012) Broad ranges of affinity and specificity of anti-histone anti- genome. Cell 129(4):823–837. bodies revealed by a quantitative peptide immunoprecipitation assay. J Mol Biol 2. Bernstein BE, Meissner A, Lander ES (2007) The mammalian epigenome. Cell 128(4): 424(5):391–399. 669–681. 27. Ke Q, Davidson T, Chen H, Kluz T, Costa M (2006) Alterations of histone modifications 3. Cuddapah S, et al. (2009) Global analysis of the insulator binding protein CTCF in and transgene silencing by nickel chloride. Carcinogenesis 27(7):1481–1488. chromatin barrier regions reveals demarcation of active and repressive domains. 28. Wang C, et al. (2011) The epigenetic H3S10 phosphorylation mark is required for Genome Res 19(1):24–32. counteracting heterochromatic spreading and gene silencing in Drosophila mela- 4. Phillips-Cremins JE, Corces VG (2013) Chromatin insulators: Linking genome organi- nogaster. J Cell Sci 124(Pt 24):4309–4317. zation to cellular function. Mol Cell 50(4):461–474. 29. Wang C, et al. (2012) H3S10 phosphorylation by the JIL-1 kinase regulates H3K9 di- 5. Bickmore WA, van Steensel B (2013) Genome architecture: Domain organization of methylation and gene expression at the white locus in Drosophila. Fly (Austin) 6(2): interphase chromosomes. Cell 152(6):1270–1284. 93–97. 6. Rice JC, et al. (2003) Histone methyltransferases direct different degrees of methyl- 30. Falahi F, et al. (2013) Towards sustained silencing of HER2/neu in cancer by epigenetic ation to define distinct chromatin domains. Mol Cell 12(6):1591–1598. editing. Mol Cancer Res 11(9):1029–1039. 7. Lieberman-Aiden E, et al. (2009) Comprehensive mapping of long-range interactions 31. Van Bortle K, Corces VG (2013) The role of chromatin insulators in nuclear architec- – reveals folding principles of the human genome. Science 326(5950):289–293. ture and genome function. Curr Opin Genet Dev 23(2):212 218. 8. Chen X, et al. (2012) G9a/GLP-dependent histone H3K9me2 patterning during human 32. Matys V, et al. (2006) TRANSFAC and its module TRANSCompel: Transcriptional gene – hematopoietic stem cell lineage commitment. Genes Dev 26(22):2499–2511. regulation in eukaryotes. Nucleic Acids Res 34(Database issue):D108 D110. 9. Lienert F, et al. (2011) Genomic prevalence of heterochromatic H3K9me2 and tran- 33. Nakahashi H, et al. (2013) A genome-wide map of CTCF multivalency redefines the – scription do not discriminate pluripotent from terminally differentiated cells. PLoS CTCF code. Cell Reports 3(5):1678 1689. Genet 7(6):e1002090. 34. Hartwig A (2001) Zinc finger proteins as potential targets for toxic metal ions: Dif- – 10. Wen B, Wu H, Shinkai Y, Irizarry RA, Feinberg AP (2009) Large histone H3 lysine 9 ferential effects on structure and function. Antioxid Redox Signal 3(4):625 634. dimethylated chromatin blocks distinguish differentiated from embryonic stem cells. 35. Vostrov AA, Quitschke WW (1997) The zinc finger protein CTCF binds to the APBbeta Nat Genet 41(2):246–250. domain of the amyloid beta-protein precursor promoter. Evidence for a role in – 11. Wen B, et al. (2012) Euchromatin islands in large heterochromatin domains are en- transcriptional activation. J Biol Chem 272(52):33353 33359. 36. Giri NC, et al. (2013) Structural investigations of the nickel-induced inhibition of riched for CTCF binding and differentially DNA-methylated regions. BMC Genomics truncated constructs of the JMJD2 family of histone demethylases using X-ray ab- 13:566. sorption spectroscopy. Biochemistry 52(24):4168–4183. 12. Ma MK, Heath C, Hair A, West AG (2011) Histone crosstalk directed by H2B ubiq- 37. Chen H, et al. (2010) Nickel ions inhibit histone demethylase JMJD1A and DNA repair uitination is required for chromatin boundary integrity. PLoS Genet 7(7):e1002175. enzyme ABH2 by replacing the ferrous iron in the catalytic centers. J Biol Chem 13. Chen H, Ke Q, Kluz T, Yan Y, Costa M (2006) Nickel ions increase histone H3 lysine 9 285(10):7374–7383. dimethylation and induce transgene silencing. Mol Cell Biol 26(10):3728–3737. 38. Ding J, et al. (2009) Effects of nickel on cyclin expression, cell cycle progression and 14. Ellen TP, Kluz T, Harder ME, Xiong J, Costa M (2009) Heterochromatinization as a cell proliferation in human pulmonary cells. Cancer Epidemiol Biomarkers Prev 18(6): potential mechanism of nickel-induced carcinogenesis. Biochemistry 48(21):4626–4632. 1720–1729. 15. Cangul H, et al. (2002) Molecular mechanisms of nickel carcinogenesis. Toxicol Lett 39. Snowden AW, Gregory PD, Case CC, Pabo CO (2002) Gene-specific targeting of H3K9 127(1-3):69–75. methylation is sufficient for initiating repression in vivo. Curr Biol 12(24):2159–2166. 16. Salnikow K, Zhitkovich A (2008) Genetic and epigenetic mechanisms in metal carci- 40. Leung DC, et al. (2011) Lysine methyltransferase G9a is required for de novo DNA nogenesis and cocarcinogenesis: Nickel, arsenic, and chromium. Chem Res Toxicol methylation and the establishment, but not the maintenance, of proviral silencing. 21(1):28–44. Proc Natl Acad Sci USA 108(14):5718–5723. 17. Kasprzak KS, Sunderman FW, Jr, Salnikow K (2003) Nickel carcinogenesis. Mutat Res 41. Du J, et al. (2012) Dual binding of chromomethylase domains to H3K9me2-containing 533(1-2):67–97. nucleosomes directs DNA methylation in plants. Cell 151(1):167–180. 18. Baccarelli A, Bollati V (2009) Epigenetics and environmental chemicals. Curr Opin 42. Vettese-Dadey M, et al. (1996) Acetylation of histone H4 plays a primary role in en- – Pediatr 21(2):243 251. hancing transcription factor binding to nucleosomal DNA in vitro. EMBO J 15(10): 19. Chen H, Kluz T, Zhang R, Costa M (2010) Hypoxia and nickel inhibit histone de- 2508–2518. methylase JMJD1A and repress Spry2 expression in human bronchial epithelial BEAS- 43. Ong CT, Corces VG (2014) CTCF: An architectural protein bridging genome topology – 2B cells. Carcinogenesis 31(12):2136 2144. and function. Nat Rev Genet 15(4):234–246. 20. Golebiowski F, Kasprzak KS (2005) Inhibition of core histones acetylation by carci- 44. Handoko L, et al. (2011) CTCF-mediated functional chromatin interactome in plurip- – nogenic nickel(II). Mol Cell Biochem 279(1-2):133 139. otent cells. Nat Genet 43(7):630–638. 21. Lee YW, et al. (1995) Carcinogenic nickel silences gene expression by chromatin 45. Kim YJ, Cecchini KR, Kim TH (2011) Conserved, developmentally regulated mechanism condensation and DNA methylation: A new model for epigenetic carcinogens. Mol couples chromosomal looping and heterochromatin barrier activity at the homeobox – Cell Biol 15(5):2547 2557. gene A locus. Proc Natl Acad Sci USA 108(18):7391–7396. 22. Li Q, Suen TC, Sun H, Arita A, Costa M (2009) Nickel compounds induce apoptosis in 46. Schwartz YB, et al. (2012) Nature and function of insulator protein binding sites in the human bronchial epithelial Beas-2B cells by activation of c-Myc through ERK pathway. Drosophila genome. Genome Res 22(11):2188–2198. Toxicol Appl Pharmacol 235(2):191–198. 47. Van Bortle K, et al. (2012) Drosophila CTCF tandemly aligns with other insulator 23. Huang Y, et al. (2002) Activation of nuclear factor-kappaB and not activator protein-1 proteins at the borders of H3K27me3 domains. Genome Res 22(11):2176–2187. in cellular response to nickel compounds. Environ Health Perspect 110(Suppl 5): 48. Gaszner M, Felsenfeld G (2006) Insulators: Exploiting transcriptional and epigenetic 835–839. mechanisms. Nat Rev Genet 7(9):703–713. 24. Ding J, et al. (2006) Nickel compounds render anti-apoptotic effect to human bron- 49. Bal W, Schwerdtle T, Hartwig A (2003) Mechanism of nickel assault on the zinc finger chial epithelial Beas-2B cells by induction of cyclooxygenase-2 through an IKKbeta/ of DNA repair protein XPA. Chem Res Toxicol 16(2):242–248. p65-dependent and IKKalpha- and p50-independent pathway. J Biol Chem 281(51): 50. Asmuss M, Mullenders LH, Hartwig A (2000) Interference by toxic metal compounds 39022–39032. with isolated zinc finger DNA repair proteins. Toxicol Lett 112–113:227–231. 25. Song Q, Smith AD (2011) Identifying dispersed epigenomic domains from ChIP-Seq 51. Lan F, et al. (2007) S. pombe LSD1 homologs regulate heterochromatin propagation data. Bioinformatics 27(6):870–871. and euchromatic gene transcription. Mol Cell 26(1):89–101.

14636 | www.pnas.org/cgi/doi/10.1073/pnas.1406923111 Jose et al. Downloaded by guest on September 26, 2021