Permalink
Cannot retrieve contributors at this time
Fetching contributors…
Cannot retrieve contributors at this time
| <?xml version="1.0" encoding="UTF-8"?><!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.1d1 20130915//EN" "JATS-archivearticle1.dtd"><article article-type="research-article" dtd-version="1.1d1" xmlns:xlink="http://www.w3.org/1999/xlink"><front><journal-meta><journal-id journal-id-type="nlm-ta">elife</journal-id><journal-id journal-id-type="hwp">eLife</journal-id><journal-id journal-id-type="publisher-id">eLife</journal-id><journal-title-group><journal-title>eLife</journal-title></journal-title-group><issn publication-format="electronic">2050-084X</issn><publisher><publisher-name>eLife Sciences Publications, Ltd</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">01776</article-id><article-id pub-id-type="doi">10.7554/eLife.01776</article-id><article-categories><subj-group subj-group-type="display-channel"><subject>Research article</subject></subj-group><subj-group subj-group-type="heading"><subject>Genes and chromosomes</subject></subj-group></article-categories><title-group><article-title>p50-associated COX-2 extragenic RNA (PACER) activates <italic>COX-2</italic> gene expression by occluding repressive NF-κB complexes</article-title></title-group><contrib-group><contrib contrib-type="author" id="author-8985"><name><surname>Krawczyk</surname><given-names>Michal</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="fn" rid="con1"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" corresp="yes" id="author-1426"><name><surname>Emerson</surname><given-names>Beverly M</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="corresp" rid="cor1">*</xref><xref ref-type="other" rid="par-1"/><xref ref-type="other" rid="par-2"/><xref ref-type="other" rid="par-3"/><xref ref-type="fn" rid="con2"/><xref ref-type="fn" rid="conf1"/></contrib><aff id="aff1"><institution content-type="dept">Regulatory Biology Laboratory</institution>, <institution>Salk Institute for Biological Studies</institution>, <addr-line><named-content content-type="city">La Jolla</named-content></addr-line>, <country>United States</country></aff></contrib-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Espinosa</surname><given-names>Joaquin M</given-names></name><role>Reviewing editor</role><aff><institution>HHMI/University of Colorado at Boulder</institution>, <country>United States</country></aff></contrib></contrib-group><author-notes><corresp id="cor1"><label>*</label>For correspondence: <email>emerson@salk.edu</email></corresp></author-notes><pub-date date-type="pub" publication-format="electronic"><day>29</day><month>04</month><year>2014</year></pub-date><pub-date pub-type="collection"><year>2014</year></pub-date><volume>3</volume><elocation-id>e01776</elocation-id><history><date date-type="received"><day>24</day><month>10</month><year>2013</year></date><date date-type="accepted"><day>11</day><month>04</month><year>2014</year></date></history><permissions><copyright-statement>© 2014, Krawczyk and Emerson</copyright-statement><copyright-year>2014</copyright-year><copyright-holder>Krawczyk and Emerson</copyright-holder><license xlink:href="http://creativecommons.org/licenses/by/3.0/"><license-p>This article is distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/3.0/">Creative Commons Attribution License</ext-link>, which permits unrestricted use and redistribution provided that the original author and source are credited.</license-p></license></permissions><self-uri content-type="pdf" xlink:href="elife01776.pdf"/><abstract><object-id pub-id-type="doi">10.7554/eLife.01776.001</object-id><p>Deregulated expression of COX-2 has been causally linked to development, progression, and outcome of several types of human cancer. We describe a novel fundamental level of transcriptional control of <italic>COX-2</italic> expression. Using primary human mammary epithelial cells and monocyte/macrophage cell lines, we show that the chromatin boundary/insulator factor CTCF establishes an open chromatin domain and induces expression of a long non-coding RNA within the upstream promoter region of COX-2. Upon induction of <italic>COX-2</italic> expression, the lncRNA associates with p50, a repressive subunit of NF-κB, and occludes it from the COX-2 promoter, potentially facilitating interaction with activation-competent NF-κB p65/p50 dimers. This enables recruitment of the p300 histone acetyltransferase, a domain-wide increase in histone acetylation and assembly of RNA Polymerase II initiation complexes. Our findings reveal an unexpected mechanism of gene control by lncRNA-mediated repressor occlusion and identify the COX-2-lncRNA, PACER, as a new potential target for COX-2-modulation in inflammation and cancer.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.001">http://dx.doi.org/10.7554/eLife.01776.001</ext-link></p></abstract><abstract abstract-type="executive-summary"><object-id pub-id-type="doi">10.7554/eLife.01776.002</object-id><title>eLife digest</title><p>To produce a protein a cell must first transcribe the DNA in a gene to make a messenger RNA molecule, which is then translated to make the protein. However, cells also produce other types of RNA molecules which do not become proteins. MicroRNAs, for example, regulate the expression of genes as proteins, while the role of other RNA molecules called long non-coding RNAs (lncRNAs) is not well understood.</p><p>Now Krawczyk and Emerson have found an lncRNA that controls a gene called <italic>COX-2</italic> that is often implicated in breast, colon, prostate, and lung cancer. This RNA molecule, which is called PACER, originates near the start of the <italic>COX-2</italic> gene, but it cannot be messenger RNA because it does not contain the instructions to make the COX-2 protein, and it is too long to be a microRNA. Further experiments showed that the newly discovered lncRNA activates the expression of the <italic>COX-2</italic> gene.</p><p>Krawczyk and Emerson found that PACER attracts enzymes that spotlight genes that need to be activated, thus increasing the transcription of these genes to make messenger RNA. Genes can also be switched on and off by various molecules binding to nearby DNA, and PACER encourages the activation of <italic>COX-2</italic> by keeping away the molecules that might switch it off.</p><p>In addition to shedding new light on the role of lncRNAs, these results suggest that PACER could be a suitable therapeutic target in cancers that involve the <italic>COX-2</italic> gene.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.002">http://dx.doi.org/10.7554/eLife.01776.002</ext-link></p></abstract><kwd-group kwd-group-type="author-keywords"><title>Author keywords</title><kwd>long non-coding RNA</kwd><kwd>Cyclooxygenase 2</kwd><kwd>NF-κB</kwd><kwd>CTCF</kwd><kwd>p300</kwd><kwd>transcription</kwd></kwd-group><kwd-group kwd-group-type="research-organism"><title>Research organism</title><kwd>human</kwd></kwd-group><funding-group><award-group id="par-1"><funding-source><institution-wrap><institution-id institution-id-type="FundRef">http://dx.doi.org/10.13039/100000002</institution-id><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>R01 CA159354</award-id><principal-award-recipient><name><surname>Emerson</surname><given-names>Beverly M</given-names></name></principal-award-recipient></award-group><award-group id="par-2"><funding-source><institution-wrap><institution>Chambers Medical Foundation</institution></institution-wrap></funding-source><principal-award-recipient><name><surname>Emerson</surname><given-names>Beverly M</given-names></name></principal-award-recipient></award-group><award-group id="par-3"><funding-source><institution-wrap><institution>GemCon Foundation</institution></institution-wrap></funding-source><principal-award-recipient><name><surname>Emerson</surname><given-names>Beverly M</given-names></name></principal-award-recipient></award-group><funding-statement>The funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.</funding-statement></funding-group><custom-meta-group><custom-meta><meta-name>elife-xml-version</meta-name><meta-value>2</meta-value></custom-meta><custom-meta specific-use="meta-only"><meta-name>Author impact statement</meta-name><meta-value>A long non-coding RNA removes the transcriptional repressor p50 to regulate recruitment of co-activator p300 and RNA Polymerase II complexes to activate the COX-2 gene in human mammary epithelial cells and macrophages.</meta-value></custom-meta></custom-meta-group></article-meta></front><body><sec id="s1" sec-type="intro"><title>Introduction</title><p>Long non-coding RNAs (lncRNAs) are a highly heterogeneous class of RNA molecules which are >200nt long and, apart from the lack of protein coding potential, share many characteristics with mRNAs in that they are transcribed by RNA polymerase II (RNAP II) and can be capped, spliced, and polyadenylated (<xref ref-type="bibr" rid="bib24">Guttman et al., 2009</xref>). LncRNAs can be located in intergenic DNA, within introns, or overlapping other genes in an antisense orientation and their expression is often tightly regulated and/or restricted to specific tissues or stages of development (<xref ref-type="bibr" rid="bib24">Guttman et al., 2009</xref>; <xref ref-type="bibr" rid="bib16">Dinger et al., 2008</xref>; <xref ref-type="bibr" rid="bib38">Mercer et al., 2008</xref>; <xref ref-type="bibr" rid="bib66">Tsai et al., 2011</xref>). Contrary to their smaller counterparts (tRNA, siRNA, miRNA etc), whose functions have been firmly established, the roles lncRNAs play in many biological phenomena are much less well understood. While examples exist of lncRNA involvement in such diverse processes as the cell cycle, cell migration and survival, metabolism, organization of subcellular compartments, among others, the predominant function of many lncRNAs appears to be regulation of gene expression (<xref ref-type="bibr" rid="bib39">Mercer et al., 2009</xref>; <xref ref-type="bibr" rid="bib69">Wang and Chang, 2011</xref>; <xref ref-type="bibr" rid="bib50">Ponting et al., 2009</xref>; <xref ref-type="bibr" rid="bib74">Wilusz et al., 2009</xref>).</p><p>LncRNAs can utilize a variety of mechanisms to regulate gene activity, where perhaps the most widespread involves interacting and recruiting repressive histone methyltransferase or demethylase complexes such as polycomb repressive complexes 1 and 2 (PRC1, PRC2), LSD1/CoREST/REST or G9a to target genes (<xref ref-type="bibr" rid="bib76">Wutz et al., 2002</xref>; <xref ref-type="bibr" rid="bib44">Nagano et al., 2008</xref>; <xref ref-type="bibr" rid="bib29">Khalil et al., 2009</xref>; <xref ref-type="bibr" rid="bib83">Zhao et al., 2010</xref>; <xref ref-type="bibr" rid="bib65">Tsai et al., 2010</xref>; <xref ref-type="bibr" rid="bib81">Yap et al., 2010</xref>). Conversely, another lncRNA was shown to interact with an activating HMT, the trithorax complex WDR5/MLL1 (<xref ref-type="bibr" rid="bib71">Wang et al., 2011</xref>). In addition, a growing number of studies implicate lncRNAs in transcriptional interference, splicing, miRNA squelching, and direct interaction with hormone receptors, transcription factors, and other RNA-binding proteins (<xref ref-type="bibr" rid="bib39">Mercer et al., 2009</xref>; <xref ref-type="bibr" rid="bib69">Wang and Chang, 2011</xref>; <xref ref-type="bibr" rid="bib50">Ponting et al., 2009</xref>; <xref ref-type="bibr" rid="bib74">Wilusz et al., 2009</xref>; <xref ref-type="bibr" rid="bib8">Carpenter et al., 2013</xref>).</p><p>Given that lncRNAs have gained increasing appreciation as key regulators of gene expression, it is not surprising that they are frequently deregulated during tumorigenesis (<xref ref-type="bibr" rid="bib72">Wapinski and Chang, 2011</xref>; <xref ref-type="bibr" rid="bib21">Gibb et al., 2011</xref>). Examples include H19, whose upregulation is detected in a variety of cancers (<xref ref-type="bibr" rid="bib18">Fellig et al., 2005</xref>; <xref ref-type="bibr" rid="bib4">Barsyte-Lovejoy et al., 2006</xref>; <xref ref-type="bibr" rid="bib26">Hibi et al., 1996</xref>; <xref ref-type="bibr" rid="bib1">Adriaenssens et al., 1998</xref>; <xref ref-type="bibr" rid="bib32">Kondo et al., 1995</xref>); HOTAIR, whose overexpression is observed in breast, prostate, and other cancers and whose loss inhibits cancer invasiveness (<xref ref-type="bibr" rid="bib23">Gupta et al., 2010</xref>); GAS5, an apoptosis regulator that is downregulated in breast cancers (<xref ref-type="bibr" rid="bib42">Mourtada-Maarabouni et al., 2009</xref>); and ANRIL, whose expression is affected by SNPs that correlate with several neoplasias and other diseases (<xref ref-type="bibr" rid="bib14">Cunnington et al., 2010</xref>). Despite the apparent functional consequences of lncRNA deregulation, very little is currently known about how expression of lncRNAs is normally modulated and what the basis is for its deregulation in cancer and other human diseases.</p><p>The CCCTC-binding factor CTCF is a highly conserved 11-zinc finger DNA binding protein. Its primary function is believed to involve formation of chromatin boundaries/insulators through chromosomal looping. However, CTCF has been implicated in a variety of other processes, including transcription, enhancer blocking, insulation, splicing, nucleolar maintenance, DNA methylation, and imprinting (<xref ref-type="bibr" rid="bib40">Merkenschlager and Odom, 2013</xref>; <xref ref-type="bibr" rid="bib48">Phillips and Corces, 2009</xref>; <xref ref-type="bibr" rid="bib45">Ohlsson et al., 2010</xref>). Genomic profiling has shown that CTCF binds to tens of thousands of genomic sites located at gene promoters, within coding sequences and in intergenic regions (<xref ref-type="bibr" rid="bib30">Kim et al., 2007</xref>; <xref ref-type="bibr" rid="bib79">Xie et al., 2007</xref>; <xref ref-type="bibr" rid="bib3">Barski et al., 2007</xref>). CTCF sites are enriched for histone variants H2A.Z and H3.3 and correlate with increased chromatin accessibility, nucleosomal depletion, presence of histone variants H2A.Z and H3.3 and histone H3K4 methylation (<xref ref-type="bibr" rid="bib40">Merkenschlager and Odom, 2013</xref>).</p><p>Cyclooxygenase 2 (COX-2, also called prostaglandin-endoperoxide synthase 2, PTGS2) is one of two COX isomers that are key enzymes in prostaglandin (PG) biosynthesis. COX enzymes catalyze conversion of arachidonic acid to PGH2 that acts as a substrate for a number of eicosanoid derivatives, such as PGE2, PGI2 and thromboxane A2, which are important mediators of many biological processes including inflammation, fever, pain, gastric and kidney function (<xref ref-type="bibr" rid="bib58">Schneider and Pozzi, 2011</xref>; <xref ref-type="bibr" rid="bib11">Chun and Surh, 2004</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2000</xref>). Unlike uniformly expressed COX-1, constitutive COX-2 expression is normally restricted to a few organs but can be induced by a variety of stimuli including cytokines, oncogenes, growth factors, and hormones (<xref ref-type="bibr" rid="bib11">Chun and Surh, 2004</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2000</xref>). Induced overexpression of COX-2 and PGs has been observed with varied frequency in a number of human cancers, including breast, colon, prostate, and lung (<xref ref-type="bibr" rid="bib15">de Moraes et al., 2007</xref>). Numerous genetic and correlation studies have documented the causal involvement of COX-2 in tumor development (<xref ref-type="bibr" rid="bib22">Gupta et al., 2007</xref>; <xref ref-type="bibr" rid="bib35">Liu et al., 2001</xref>; <xref ref-type="bibr" rid="bib41">Minn et al., 2005</xref>; <xref ref-type="bibr" rid="bib46">Oshima et al., 1996</xref>; <xref ref-type="bibr" rid="bib36">Markosyan et al., 2011</xref>).</p><p>Induction of COX-2 expression can be mediated by a number of intracellular signaling pathways in different cell types (<xref ref-type="bibr" rid="bib59">Smith et al., 2000</xref>). Indeed, involvement of PKC, Ras, and Wnt pathways through the activation of MAPK kinase family of proteins including ERK, JNK and p38 in a cell type-specific manner has been demonstrated (<xref ref-type="bibr" rid="bib20">Gauthier et al., 2005</xref>; <xref ref-type="bibr" rid="bib70">Wang et al., 2001</xref>; <xref ref-type="bibr" rid="bib53">Rodriguez-Barbero et al., 2006</xref>; <xref ref-type="bibr" rid="bib52">Ramsay et al., 2003</xref>; <xref ref-type="bibr" rid="bib82">Zhai et al., 2010</xref>; <xref ref-type="bibr" rid="bib34">Li et al., 2009</xref>). DNA binding sites for mediators of these signaling events, such as transcription factors NF-κB, AP1, CREB C/EBP, NF-IL6, MEF2, and TCF4/LEF1 have been identified in the COX-2 promoter and further studies demonstrated their functional involvement in regulating COX-2 transcription in a variety of experimental systems (<xref ref-type="bibr" rid="bib11">Chun and Surh, 2004</xref>; <xref ref-type="bibr" rid="bib15">de Moraes et al., 2007</xref>).</p><p>In this study, we describe a novel layer of <italic>COX-2</italic> transcriptional regulation in cells relevant to human tumorigenesis. Using primary human mammary epithelial cells (HMECs) and a PMA-driven human monocyte-macrophage differentiation system, we demonstrate that a nuclear antisense long non-coding RNA PACER (<bold>P</bold>50-<bold>A</bold>ssociated <bold>C</bold>OX-2 <bold>E</bold>xtragenic <bold>R</bold>NA) is expressed in the upstream region of COX-2 and functions to directly sequester the repressive NF-κB p50 subunit from the COX-2 promoter. This facilitates the recruitment of p300 histone acetyltransferase and a domain-wide increase in chromatin accessibility, as well as assembly of RNA Polymerase II pre-initiation complexes. We also show that PACER expression is controlled by two CTCF/cohesin complexes assembled in the 5′ UTR region and at a distal upstream site. CTCF/cohesin induces lncRNA expression by establishing a chromatin domain marked by increased H3K4 methylation, H4K8 acetylation and decreased H4K20 trimethylation, thus creating a permissive chromatin environment and protecting the COX-2 regulatory region from surrounding repressive chromatin. These findings offer unexpected insights into the mechanisms of gene regulation by CTCF/cohesin and by lncRNA-mediated repressor eviction.</p></sec><sec id="s2" sec-type="results"><title>Results</title><sec id="s2-1"><title>Identification of an antisense long non-coding RNA in the upstream region of the human <italic>COX-2</italic> gene</title><p>We initially hypothesized that <italic>COX-2</italic> expression might be regulated by a non-coding antisense RNA overlapping the gene or its promoter. Our search failed to identify lncRNAs within the COX-2 coding region, so we examined the region upstream of the COX-2 transcription start site. This region contains numerous putative binding sites for transcription factors but no additional annotated genes. We performed chromatin immunoprecipitation (ChIP) experiments using antibodies directed against RNA Polymerase II (RNAP II), which revealed high levels of RNAP II as far as ∼1 kb upstream of the COX-2 transcribed region. Intriguingly, we detected a peak of RNAP II binding at approximately −0.45 kb (<xref ref-type="fig" rid="fig1">Figure 1A</xref>). RNAP II molecules in this region contained C-terminal domain (CTD) phosphorylation on serine 5 as demonstrated by ChIPs with antibodies recognizing Ser-P RNAP II specifically (<xref ref-type="fig" rid="fig1">Figure 1B</xref>). These findings suggested that an additional RNAP II promoter was located ∼300 bp upstream of the COX-2 promoter, potentially driving expression of an unidentified species of extragenic RNA.<fig-group><fig id="fig1" position="float"><object-id pub-id-type="doi">10.7554/eLife.01776.003</object-id><label>Figure 1.</label><caption><title>Identification of an antisense long non-coding RNA in the upstream region of the <italic>COX-2</italic> gene.</title><p>(<bold>A</bold> and <bold>B</bold>) ChiP experiments were performed with chromatin extracts from HMEC cells with antibodies directed against RNAP II or phospho-serine 5 modified RNAP II, respectively, and analyzed by qPCR. Each histogram represents one qPCR data point (amplicon). Non-COX-2 promoter RNAP II peak is marked with an asterisk. X-axis scale is relative to COX-2 transcription start site. (<bold>C</bold>) Transcript abundance in the <italic>COX-2</italic> locus in HMECs was measured by RT-qPCR. cDNA was primed with random oligos and therefore provide no information on strand specificity. Each diamond represents one RT-qPCR data point (amplicon). Y-axis values are relative to the signal obtained with the 5′ UTR amplicon. (<bold>D</bold> through <bold>F</bold>) Strand specific RT-qPCR. cDNA synthesis was performed with two sense oligos (m2R and jR) measuring antisense transcription or three antisense oligos (jF, eF, and control −3.55F) measuring sense transcription. (<bold>G</bold>) A schematic representation of lncRNA ends mapped with RACE. (<bold>H</bold>) Relative abundance of COX-2 mRNA and non-coding RNA. (<bold>I</bold>) A reporter gene assay used to measure the strength of the ncRNA promoter. Luciferase-only construct (empty) or constructs containing indicated fragments of COX-2 upstream region were transfected into 293T cells and luciferase activities were measured 48 hr post-transfection. Activity units reflect the ratios between Firefly and control Renilla luciferases.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.003">http://dx.doi.org/10.7554/eLife.01776.003</ext-link></p></caption><graphic xlink:href="elife01776f001"/></fig><fig id="fig1s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.004</object-id><label>Figure 1—figure supplement 1.</label><caption><title>Mapping the 3′ end of the non-coding RNA transcript within the COX-2 upstream region by RACE.</title><p>Sequencing of three independent clones after amplification of 3′ ends. Oligos located at position −312 were used for initial RACE step, and resulting clones were sequenced with oligos at −347 or −462. Sequencer traces are shown on the right. Green shading indicates identical sequences.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.004">http://dx.doi.org/10.7554/eLife.01776.004</ext-link></p></caption><graphic xlink:href="elife01776fs001"/></fig><fig id="fig1s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.005</object-id><label>Figure 1—figure supplement 2.</label><caption><title>Mapping the 5′ end of the non-coding RNA transcript within the COX-2 upstream region by RACE.</title><p>Sequencing of six independent clones to map 5′ ends of the lncRNA. Nested RACE was performed using oligos at positions −900 and −700, and sequencing with oligos at −547 or −500 (f oligo).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.005">http://dx.doi.org/10.7554/eLife.01776.005</ext-link></p></caption><graphic xlink:href="elife01776fs002"/></fig><fig id="fig1s3" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.006</object-id><label>Figure 1—figure supplement 3.</label><caption><title>Entire sequence of PACER, corresponding to the longest 5′ variant mapping to nucleotides −257 to −1022.</title><p>All positions relative to transcription start site of COX-2 mRNA.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.006">http://dx.doi.org/10.7554/eLife.01776.006</ext-link></p></caption><graphic xlink:href="elife01776fs003"/></fig><fig id="fig1s4" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.007</object-id><label>Figure 1—figure supplement 4.</label><caption><title>GRO-seq data supports the existence of a long non-coding RNA species within COX-2 upstream region.</title><p>A picture from UCSC genome browser (<ext-link ext-link-type="uri" xlink:href="http://genome.ucsc.edu">genome.ucsc.edu</ext-link>) showing COX-2 (marked PTGS2) region from 1 kb downstream (left) up to 3 kb upstream (right) of COX-2 transcription start site. PACER is depicted as mapped by our RACE experiments. Sense run-on transcription is shown below the horizontal line in blue, and antisense transcription in green above the line. Position of CTCF ChIP peaks from ENCODE study in HUVEC cells and location of CpG island are shown at the bottom.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.007">http://dx.doi.org/10.7554/eLife.01776.007</ext-link></p></caption><graphic xlink:href="elife01776fs004"/></fig></fig-group></p><p>We then examined whether any transcripts were synthesized within this region using quantitative RT-PCR (RT-qPCR). Primer walking indicated that the COX-2 upstream region is transcriptionally active roughly from −0.3−0.4 kb to approximately −1.5−2 kb (<xref ref-type="fig" rid="fig1">Figure 1C</xref>). To determine the directionality of transcription, we used strand specific RT-qPCR. When cDNA synthesis was primed with either of the ‘sense’ oligos (which detect antisense transcripts), contiguous transcripts were detected extending roughly up to −0.3−0.4 kb (<xref ref-type="fig" rid="fig1">Figure 1D</xref>). By contrast, ‘anti-sense’ oligos did not produce detectable signals within the upstream region (<xref ref-type="fig" rid="fig1">Figure 1E</xref>) in contrast to positive amplification within the coding region (<xref ref-type="fig" rid="fig1">Figure 1F</xref>). These experiments document the existence of a contiguous antisense transcript originating at approximately −0.3 kb upstream of the COX-2 mRNA start site. To determine the exact 5′ and 3′ ends of the transcript, rapid amplification of cDNA ends (RACE) was employed (<xref ref-type="fig" rid="fig1">Figure 1G</xref>, <xref ref-type="fig" rid="fig1s1 fig1s2">Figure 1—figure supplements 1, 2</xref>). 3’ RACE repeatedly showed that a polyA tail is added after position −1022, consistent with the presence of a classical polyadenylation signal AAUAAA at position −999 (<xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1</xref>). Sequencing of the 5′ ends showed a variety of positions (from −257 to −332) indicating that alternative start sites are used by RNA Polymerase (<xref ref-type="fig" rid="fig1s2">Figure 1—figure supplement 2</xref>). This is consistent with the absence of a TATA box around the transcription start site of this novel RNA. Given that our qPCR analysis showed the presence of transcripts extending up to −1.5−2 kb upstream, we cannot exclude the existence of additional species of ncRNA within the COX-2 upstream region. Exploring the function of those molecules could be a subject of a future study.</p><p>A survey of a recent study utilizing strand specific Global Run-On sequencing (GRO-seq) confirmed that the upstream region of COX-2 is indeed transcribed in an antisense orientation (<xref ref-type="bibr" rid="bib19">Galbraith et al., 2013</xref>). A strong GRO-seq signal extends up to approximately −1100 bp upstream of COX-2 transcription start site, and therefore correlates with our finding of a major lncRNA species in this region (<xref ref-type="fig" rid="fig1s4">Figure 1—figure supplement 4</xref>). We used qPCR to accurately measure the relative abundance of the newly identified ncRNA and COX-2 mRNA. To do this, pairs of oligos mapping within one exon of the mRNA and within the intronless ncRNA were used for cDNA amplification and their abundance was measured relative to reference genomic DNA. The results indicate that the mRNA is present in ∼30-fold molar excess as compared to the ncRNA (<xref ref-type="fig" rid="fig1">Figure 1H</xref>).</p><p>To corroborate the finding that two transcriptional units possessing separate RNAP II promoters exist within the COX-2 locus, we used reporter gene assays (<xref ref-type="fig" rid="fig1">Figure 1I</xref>). 400 or 500 bp fragments of DNA upstream of the COX-2 mRNA start site containing the putative ncRNA promoter was fused to the Luciferase gene (constructs 2 and 3). The activities of these constructs were compared to analogous fusions containing additional sequence spanning the COX-2 mRNA promoter (constructs 4 and 5). Constructs containing the ncRNA promoter but lacking the mRNA promoter were highly active as compared to the promoterless empty plasmid. Addition of the mRNA promoter increased the activity of the constructs (up to twofold), indicating that the two regulatory elements could exhibit synergistic behavior. Nevertheless, these experiments show that a bona fide promoter driving the expression of ncRNA is present in the region ∼300–400 bp upstream of the mRNA transcription start site.</p><p>Bioinformatic analysis of the nucleotide sequence failed to support the hypothesis that the novel RNA encodes a functional protein. The longest detected Met-initiated putative ORF is 43 aa and this peptide contains no identifiable protein domains and bears no similarity to any protein in protein databases (not shown). In addition, polysome profiling indicated that the ncRNA does not co-sediment with the poly-ribosome fraction (D Henderson, personal communication). While we cannot exclude that the RNA might encode a small peptide of unknown function, we conclude with high confidence that it represents a novel species of long non-coding RNA (lncRNA). We will continue to refer to it as <bold>P</bold>50-<bold>A</bold>ssociated <bold>C</bold>OX-2 <bold>E</bold>xtragenic <bold>R</bold>NA (PACER, see below). The entire nucleotide sequence of PACER is shown in <xref ref-type="fig" rid="fig1s3">Figure 1—figure supplement 3</xref>.</p><p>During the preparation of this manuscript, Carpenter and colleagues reported that a species of lncRNA located ∼50 kb downstream of the mouse COX-2 gene affects expression of many immune-related genes both positively and negatively, likely through interactions with RNA binding proteins hnRNP A/B and hnRNP A2/B1 in murine macrophages (<xref ref-type="bibr" rid="bib8">Carpenter et al., 2013</xref>). However, COX-2 mRNA expression was neither affected by knockdown or overexpression of this lncRNA; nor in the initial screen which identified a group of lncRNAs induced by treatment with the Tlr2 ligand Pam3CSK4. Therefore, the name of this lncRNA (lincRNA-COX-2) merely reflects its genomic localization, not its functional relationship with the COX-2 gene, unlike the lncRNA reported here.</p></sec><sec id="s2-2"><title>CTCF binding demarcates an active regulatory domain in the COX-2 upstream region</title><p>Binding sites for known mediators of COX-2 activation, such as NF-κB, AP1, NF-IL6, CREB, TCF4/LEF1, cluster between the COX-2 transcription start site and approximately −1.5 kb upstream of the promoter (<xref ref-type="fig" rid="fig1">Figure 1G</xref>) (<xref ref-type="bibr" rid="bib15">de Moraes et al., 2007</xref>). This clustering, together with the presence of lncRNA suggests the existence of a specialized chromatin domain encompassing all elements involved in COX-2 expression. Upon inspection of genomic databases, we noticed that the region far upstream (-3kb to > -50 kb) of COX-2 is densely populated by repetitive DNA elements, whereas the region from +1 to roughly −3 kb upstream of the promoter is devoid of them (<xref ref-type="fig" rid="fig2">Figure 2A</xref>). We hypothesized that CTCF, the only known mammalian insulator/boundary factor, could be involved in regulation of COX-2 expression by establishing a boundary between the region containing repetitive DNA and the proximal promoter.<fig-group><fig id="fig2" position="float"><object-id pub-id-type="doi">10.7554/eLife.01776.008</object-id><label>Figure 2.</label><caption><title>CTCF/cohesin complexes bind to two sites within the <italic>COX-2</italic> locus, encompassing the promoters of mRNA and lncRNA.</title><p>(<bold>A</bold>) Genomic neighborhood of the human <italic>COX-2</italic> gene. Note the repeat-DNA-free domain extending up to approximately −3 kb upstream of the COX-2 promoter. (<bold>B</bold>) ChIP-on-chip analysis of CTCF binding within the <italic>COX-2</italic> genomic domain in human mammary epithelial cells (HMECs). Two CTCF peaks are evident at approximately −2.5 kb and the proximal promoter. Numerous other loci on the array served as negative controls (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>). Results are represented as log2ratios between hybridization signals obtained with CTCF-ChIP and input DNA samples. Each vertical line represents one probe on the array. X-axis kb scale is relative to the COX-2 transcription start site. Significant signals are marked by asterisks. (<bold>C</bold> and <bold>D</bold>) Mapping of CTCF and cohesin binding in HMECs analyzed by ChIP-qPCR primer-walking. Amplicon coordinates are relative to the COX-2 transcription start-site.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.008">http://dx.doi.org/10.7554/eLife.01776.008</ext-link></p></caption><graphic xlink:href="elife01776f002"/></fig><fig id="fig2s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.009</object-id><label>Figure 2—figure supplement 1.</label><caption><title>Selected results of CTCF ChIP-on-chip experiments.</title><p>The experiments were performed in duplicates with essentially identical results. COX-2 is located on chr1, c-myc on chr8 and INSIG2 on chr2.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.009">http://dx.doi.org/10.7554/eLife.01776.009</ext-link></p></caption><graphic xlink:href="elife01776fs005"/></fig><fig id="fig2s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.010</object-id><label>Figure 2—figure supplement 2.</label><caption><title>Identification and characterization of the 5’ UTR CTCF binding site in the <italic>COX-2</italic> gene.</title><p>(<bold>A</bold>). EMSA assays were used to map the promoter-proximal CTCF binding site. Purified recombinant CTCF was incubated with oligonucleotide probes in the COX-2 region (lanes 1–4 and 5–8) B-globin LCR site IV (lanes 9–12) and c-myc (lanes 13–16). Lanes 1, 5, 9, and 13 served as probe-only controls. Anti-CTCF antibodies were added in lanes 3, 7, 11, and 15 to produce supershifts (ss). B-globin fragment was used as cold competitor in lanes 4, 8, 12, and 16.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.010">http://dx.doi.org/10.7554/eLife.01776.010</ext-link></p></caption><graphic xlink:href="elife01776fs006"/></fig><fig id="fig2s3" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.011</object-id><label>Figure 2—figure supplement 3.</label><caption><title>Mutations in predicted CTCF sites abolish CTCF recruitment to DNA templates.</title><p>Biochemical recruitment assays using extracts prepared from cells transfected with CTCF constructs were performed with wild-type or mutant COX-2 (lanes 1–5) or c-myc (lanes 6–10) templates. A template which did not contain recognizable CTCF sites served as a negative control (lanes 11–12). Wild type and mutant sequences are shown at the bottom. IN, input; FT, flowthrough; B, bead bound DNA-associated fraction.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.011">http://dx.doi.org/10.7554/eLife.01776.011</ext-link></p></caption><graphic xlink:href="elife01776fs007"/></fig></fig-group></p><p>We addressed this possibility using ChIP-on-chip experiments. A custom genomic DNA tiling array was designed to include the entire human <italic>COX-2</italic> gene and 50 kb flanking sequence on each side. Numerous other genes were also included on the array (data not shown). Chromatin immunoprecipitation (ChIP) experiments were performed with extracts from primary human mammary epithelial cells (HMECs) using antibodies directed against CTCF. IP DNA was purified, amplified, and hybridized to DNA arrays. Before hybridization, enrichment of known genomic CTCF binding sites in IP DNA was verified by qPCR.</p><p>As expected, we found CTCF at numerous loci across the regions included on the array (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>). The locations and patterns of binding were largely similar to those seen in previously studied cells (data not shown) (<xref ref-type="bibr" rid="bib30">Kim et al., 2007</xref>; <xref ref-type="bibr" rid="bib79">Xie et al., 2007</xref>; <xref ref-type="bibr" rid="bib3">Barski et al., 2007</xref>). We found two CTCF sites in the <italic>COX-2</italic> locus, one located in the promoter region and the other at approximately −2.5 kb, precisely at the interface between the region of repetitive DNA and the putative COX-2 regulatory region (<xref ref-type="fig" rid="fig2">Figure 2A,B</xref>). Computer prediction revealed the presence of a very well-conserved CTCF recognition motif GCAGCAGAAGGGGGCAGTA at position −2508 and a weaker one GCAGCGCCTCCTTCAGCTCCA at position +65. EMSA and DNA recruitment assays provided additional evidence that CTCF indeed binds at +65bp relative to transcription start site (<xref ref-type="fig" rid="fig2s2 fig2s3">Figure 2—figure supplements 2, 3</xref>).</p><p>Previous studies have shown that the majority of CTCF-bound genomic sites are co-occupied by cohesin, a protein that has diverse roles in chromosome biology including regulating sister chromatid cohesion and DNA topologies (<xref ref-type="bibr" rid="bib47">Parelho et al., 2008</xref>; <xref ref-type="bibr" rid="bib54">Rubio et al., 2008</xref>; <xref ref-type="bibr" rid="bib62">Stedman et al., 2008</xref>; <xref ref-type="bibr" rid="bib73">Wendt et al., 2008</xref>). Together, CTCF and cohesin form a complex that regulates a variety of chromatin-related functions such as inter-chromosomal interactions, enhancer blocking and insulation (<xref ref-type="bibr" rid="bib40">Merkenschlager and Odom, 2013</xref>). We thus investigated whether cohesin and CTCF bind to the same locations within the COX-2 upstream region in HMECs, using ChIP-qPCR with antibodies recognizing CTCF and cohesin subunit Rad21 (<xref ref-type="fig" rid="fig2">Figure 2C,D</xref>). We found that CTCF and cohesin strongly bind to the −2.5-kb site, whereas the promoter site is also recognized by both proteins, albeit to a lesser extent, especially in the case of Rad21. ChIP-qPCR experiments with other known CTCF interactors, such as B23/nucleophosmin, Topoisomerase IIβ, PARP-1 and, C23/nucleolin, failed to reveal specific binding in the COX-2 upstream region (data not shown).</p><p>The relative CTCF ChIP signal strengths at the COX-promoter and at the −2.5 kb site seem to be different in ChIP-on-chip experiments as compared to ChIP-qPCR. We tend to believe that the qPCR reflects the actual situation, because of its quantitative nature, as opposed to hybridization-based detection in the case of ChIP-on-chip. The most probable cause of the discrepancy is the fact that the COX-2 promoter region contains a CpG island extending from position +245 to −525, and thus probes covering this region contained an unusually high GC content, which could skew hybridization efficiency producing aberrantly strong and wide signal peak.</p></sec><sec id="s2-3"><title>CTCF/cohesin complexes control a chromatin domain and induce expression of a long non-coding RNA and COX-2 mRNA</title><p>Given that CTCF/cohesin occupies thousands of sites across the genome, we verified their functional involvement in regulating <italic>COX-2</italic> expression. To this end, we used small molecule RNA interference to transiently ablate the expression of CTCF or cohesin in HMECs. CTCF knockdown reduced COX-2 mRNA to ∼20%, and the COX-2 protein to almost undetectable levels (<xref ref-type="fig" rid="fig3">Figure 3A,B</xref>). These effects appear to be direct, because not only CTCF mRNA and protein levels were decreased by siRNAs (<xref ref-type="fig" rid="fig3">Figure 3B,C</xref>) but also CTCF binding to both promoter and distal −2.5 kb sites in the <italic>COX-2</italic> locus (<xref ref-type="fig" rid="fig3">Figure 3D</xref>). Similar to CTCF, siRNA-mediated knockdown of the cohesin subunit SMC3 reduced COX-2 mRNA levels, albeit to a lesser extent (<xref ref-type="fig" rid="fig3">Figure 3A</xref>). Importantly, knockdown of CTCF-reduced expression of extragenic RNA to practically undetectable levels (<xref ref-type="fig" rid="fig3">Figure 3E</xref>). This means that CTCF exerts more global control over the <italic>COX-2</italic> locus by regulating both mRNA and ncRNA expression from two distinct promoters. This prompted us to investigate whether CTCF controls COX-2 expression by creating a specialized chromatin neighborhood within the putative CTCF/cohesin chromatin loop. To do this, we performed ChIP-qPCR experiments to analyze the chromatin modification status in the COX-2 promoter region. Interestingly, the effects of CTCF knockdown on several activation and repression-associated histone marks extended to ∼2.5 kb upstream. High levels of H3K4me2 and me3 showed a local maximum at approximately −0.7 kb and were reduced twofold to threefold across the entire ∼3 kb domain in the absence of CTCF (<xref ref-type="fig" rid="fig3">Figure 3F,G</xref>). H4K8 acetylation levels extended even further upstream, and were reduced significantly by CTCF knockdown (<xref ref-type="fig" rid="fig3">Figure 3H</xref>). A repressive mark H4K20me3 had a more localized pattern, with two discernible peaks at −2.5 kb and approximately −1.2 kb. This modification was increased at −2.5 kb upon CTCF knockdown (<xref ref-type="fig" rid="fig3">Figure 3I</xref>). No significant changes were observed for H3K9me3 (<xref ref-type="fig" rid="fig3">Figure 3J</xref>), or H4K12-Ac, H2AZ, and unmodified H3 (data not shown). Taken together, these experiments indicate that CTCF regulates the <italic>COX-2</italic> locus by establishing and maintaining an open chromatin domain demarcated by two CTCF/cohesin complexes; inducing extragenic transcription of PACER within this domain; and activating COX-2 mRNA expression.<fig id="fig3" position="float"><object-id pub-id-type="doi">10.7554/eLife.01776.012</object-id><label>Figure 3.</label><caption><title>CTCF and cohesin maintain COX-2 mRNA and PACER lncRNA expression by demarcating a chromatin domain that is characterized by decreased H4K20 trimethylation, increased H3K4 di- and tri- methylation and increased histone acetylation.</title><p>(<bold>A</bold>) siRNA-mediated knockdown of CTCF or a cohesin subunit SMC3 reduces COX-2 mRNA levels in HMECs. RT-qPCR was performed with RNA prepared from HMECs 72 hr after transfection with the indicated siRNAs. Signals were normalized with 18S rRNA, TBP and GAPDH genes. (<bold>B</bold>) Western blot analysis of COX-2 and CTCF levels upon siRNA-mediated knockdown of CTCF. (<bold>C</bold>) CTCF mRNA levels in control and siRNA-CTCF transfected HMECs. (<bold>D</bold>) Binding of CTCF to the COX-2 promoter and distal sites were analyzed by ChIP-qPCR in control and CTCF knockdown cells. (<bold>E</bold>) Transcription across the <italic>COX-2</italic> locus was measured by RT-qPCR in control (black diamonds) or siCTCF-transfected HMECs (grey squares). Signals were normalized as above and re-normalized using genomic DNA. (<bold>F</bold>–<bold>J</bold>) Levels of activation-associated (H3K4me2 and me3, H4K8Ac) and repression-associated (H4K20me3, H3K9me3) histone modifications were measured using ChIP-qPCR in HMECs transfected with control siRNAs (black lines) or siRNA against CTCF (grey lines). Each diamond or square denotes a qPCR amplicon. CTCF/cohesin binding sites are marked with oval shapes. X-axis scale is relative to the COX-2 transcription start site.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.012">http://dx.doi.org/10.7554/eLife.01776.012</ext-link></p></caption><graphic xlink:href="elife01776f003"/></fig></p></sec><sec id="s2-4"><title>A long non-coding antisense RNA controls COX-2 mRNA transcription</title><p>Given that the mechanism of gene activation by CTCF/cohesin is not fully understood and our observation that CTCF binding to the <italic>COX-2</italic> locus is required for both PACER and COX-2 mRNA expression, we wondered whether PACER might serve as an intermediary in activating mRNA transcription. To explore this, we first performed transient knockdown experiments using siRNA specific to the PACER sequence. Indeed, siRNA-mediated reduction of PACER decreased COX-2 mRNA levels to a comparable extent as observed with CTCF knockdowns in HMECs (<xref ref-type="fig" rid="fig4">Figure 4A</xref>) and COX-2 protein was reduced to negligible levels (<xref ref-type="fig" rid="fig4">Figure 4B</xref>). Similar results were obtained using another siRNA specific to PACER (<xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1</xref>), confirming siRNA specificity. Importantly, the effects of PACER knockdown were independent of CTCF since CTCF protein levels remained unchanged (<xref ref-type="fig" rid="fig4">Figure 4B</xref>) and CTCF binding to the <italic>COX-2</italic> locus was not reduced (<xref ref-type="fig" rid="fig4s2">Figure 4—figure supplement 2</xref>).<fig-group><fig id="fig4" position="float"><object-id pub-id-type="doi">10.7554/eLife.01776.013</object-id><label>Figure 4.</label><caption><title>The antisense PACER lncRNA is a positive regulator of COX-2 expression in human mammary epithelial cells and in monocyte-derived macrophages before and after LPS stimulation.</title><p>Levels of COX-2 mRNA (<bold>A</bold>) and protein (<bold>B</bold>) were measured by RT-qPCR and Western blotting in control-scrambled siRNA, siRNA-CTCF, and siRNA-lncRNA transfected HMECs 72 hr post transfection. (<bold>C</bold>) Schematic representation of the monocyte-macrophage system used in this study. The human monocyte cell line, U937, was differentiated to macrophages with PMA and induced to express high levels of COX-2 by LPS stimulation. (<bold>D</bold> and <bold>E</bold>) COX-2 mRNA (<bold>D</bold>) and PACER (<bold>E</bold>) levels were measured by RT-qPCR in U937 lines carrying stably integrated control shRNA or shRNA-PACER expressing constructs before and after differentiation into macrophages. Values were normalized using 18S rRNA, TBP, and ß-globin genes. (<bold>F</bold>) COX-2 protein expression analyzed by Western blotting in the same experiment. (<bold>G</bold> and <bold>H</bold>) COX-2 ncRNA and mRNA levels were measured in a time course of LPS stimulation in control U937 (black bars) and PACER knockdown U937 macrophages (white bars). Values are relative to maximal COX-2 expression in control U937 cells after 6 hr LPS stimulation. (<bold>I</bold>) COX-2 protein expression analyzed by Western blotting in the same experiment. TBP served as a control. (<bold>J</bold>) Subcellular localization of COX-2 mRNA, PACER lncRNA and controls: nuclear intronic RNA and exclusively cytoplasmic actin mRNA. Relative levels in each fraction were measured by RT-qPCR and plotted such that they add up to 100%. CH, chromatin-bound fraction, N, nucleoplasm, CE, cytoplasm.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.013">http://dx.doi.org/10.7554/eLife.01776.013</ext-link></p></caption><graphic xlink:href="elife01776f004"/></fig><fig id="fig4s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.014</object-id><label>Figure 4—figure supplement 1.</label><caption><title>Knockdown of COX-2 mRNA with siRNA targeting PACER (si+753 and si+870) or positioned outside PACER (+1049).</title><p>Numbers are relative to COX-2 transcription start site, PACER extends from position +257 to +1022. Scrambled sequence was used as CTRL. Knockdown was performed by transient transfection of siRNAs into HMEC cells. RNA was harvested 3 days post transfection.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.014">http://dx.doi.org/10.7554/eLife.01776.014</ext-link></p></caption><graphic xlink:href="elife01776fs008"/></fig><fig id="fig4s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.015</object-id><label>Figure 4—figure supplement 2.</label><caption><title>Binding of CTCF to the COX-2 promoter upon PACER knock-down in HMECs.</title><p>Binding of CTCF to the COX-2 promoter (<bold>A</bold>) and distal site (−2.5 kb, <bold>B</bold>) was assessed by ChIP.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.015">http://dx.doi.org/10.7554/eLife.01776.015</ext-link></p></caption><graphic xlink:href="elife01776fs009"/></fig></fig-group></p><p>So far, we focused our analysis of the <italic>COX-2</italic> locus in human mammary epithelial cells because these cells have been implicated in breast tumorigenesis (<xref ref-type="bibr" rid="bib12">Crawford et al., 2004</xref>). However, COX-2 is also expressed by tumor-associated macrophages (TAMs), which are known to accelerate tumor formation and metastasis in breast and other cancer types (<xref ref-type="bibr" rid="bib49">Pollard, 2004</xref>; <xref ref-type="bibr" rid="bib9">Chen and Smyth, 2011</xref>). We therefore decided to further examine lncRNA-mediated regulation of <italic>COX-2</italic> expression in the human monocyte/macrophage cell line U937, as these cells offer unique advantages as a model system to study gene regulation. U937 monocytes can be differentiated into macrophages with phorbol myristate acetate (PMA) and <italic>COX-2</italic> expression can be efficiently induced in U937 macrophages with a variety of stimuli, including LPS (schematically shown in <xref ref-type="fig" rid="fig4">Figure 4C</xref>) (<xref ref-type="bibr" rid="bib2">Arias-Negrete et al., 1995</xref>).</p><p>We first established stable U937 lines carrying short hairpin RNAs (shRNA) targeted against PACER using lentiviral vector delivery. U937-sh-lncRNA monocytes showed no detectable COX-2 mRNA or lncRNA expression, as did control monocytes carrying shRNA directed against the Luciferase gene (<xref ref-type="fig" rid="fig4">Figure 4D,E</xref>). Upon PMA-induced differentiation, both COX-2 mRNA and PACER expression were strongly induced in control U937 cells. By contrast, in PACER-knockdown cells, PMA induced lncRNA expression to much lower levels compared to control cells, and, as a result, COX-2 mRNA levels were also markedly reduced (<xref ref-type="fig" rid="fig4">Figure 4D,E</xref>). Induction of PMA-differentiated macrophages with LPS greatly up-regulated <italic>COX-2</italic> expression in control macrophages over a 6-hr time course, with the highest levels of expression reaching a plateau at ∼2 hr post induction (<xref ref-type="fig" rid="fig4">Figure 4H</xref>). In knockdown cells, the effect of LPS was severely attenuated, with COX-2 mRNA reaching ∼60% of maximal control levels 4 hr post-induction, but not abolished, consistent with the fact that PACER expression was also up-regulated during LPS treatment (<xref ref-type="fig" rid="fig4">Figure 4G</xref>). Finally, Western blotting in whole cell extracts confirmed PACER-mediated effects on COX-2 protein levels both in PMA differentiation (<xref ref-type="fig" rid="fig4">Figure 4F</xref>) and during LPS induction (<xref ref-type="fig" rid="fig4">Figure 4I</xref>). Interestingly, PACER knockdown appears to greatly reduce COX-2 protein levels upon PMA stimulation of monocytes but only transiently affect COX-2 levels upon LPS-induction. This suggests that PACER plays an essential role early in <italic>COX-2</italic> gene activation during monocyte differentiation into macrophages by controlling the critical transition from a transcriptionally inactive locus to one competent for basal transcription.</p><p>We analyzed the relative abundance of PACER within the cell using biochemical fractionation of U937 macrophages followed by qPCR. The vast majority (more than 75%) of the lncRNA was found in the nucleus, with one third being stably integrated into chromatin and the rest occupying nuclear space; the nuclear fraction increased even more upon LPS stimulation (<xref ref-type="fig" rid="fig4">Figure 4J</xref>). This distribution is very different from that of COX-2 mRNA, of which around 50% is found in the cytoplasm. This observation suggests that PACER performs its functions in the nucleus. Small amounts of PACER found in the cytoplasm likely reflect a minor leakage of nucleoplasm into the cytoplasmic fraction, as the same is observed for control intronic sequences (<xref ref-type="fig" rid="fig4">Figure 4J</xref>).</p></sec><sec id="s2-5"><title>PACER lncRNA facilitates p300 binding, chromatin opening, RNA polymerase II recruitment, and transcriptional activation</title><p>We exploited the U937 differentiation system to gain more insight into the mechanisms by which PACER regulates basal and induced <italic>COX-2</italic> transcription through chromatin modification status, transcription factor interaction, and co-activator recruitment. First, we analyzed binding of CTCF, cohesin, Sp1 and AP1 to the COX-2 upstream region upon PACER knockdown. Sp1 and AP1 have been previously implicated in mediating <italic>COX-2</italic> activation in various cell types, including macrophages (<xref ref-type="bibr" rid="bib78">Xie and Herschman, 1996</xref>; <xref ref-type="bibr" rid="bib80">Xu et al., 2000</xref>). We found that PACER knockdown did not significantly affect binding of any of these four factors to the COX-2 promoter region, apart from a modest decrease of the AP1 component c-Jun binding to distal (−0.45 kb and −0.85 kb) sites but not to the COX-2 proximal promoter (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1</xref>).</p><p>We next analyzed the effects of PACER knockdown on chromatin modifications within the COX-2 upstream domain. ChIP-qPCR experiments showed no significant changes in H3K4 trimethylation or H2A.Z association (data not shown) and a modest effect on H3K4 dimethylation (<xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2</xref>). By contrast, histone H3 acetylation was clearly affected by the loss of PACER. Levels of H3Ac in non-induced macrophages were considerably reduced at −1.25 kb and −2.5 kb in shPACER-cells and the LPS-induced increase in H3Ac was markedly attenuated compared to control macrophages (<xref ref-type="fig" rid="fig5">Figure 5A</xref>). Even more striking effects were evident for H4 acetylation. H4Ac was induced in response to LPS treatment across the upstream COX-2 domain extending to approximately −1.3 kb. Knockdown of PACER abolished H4Ac to practically background levels (<xref ref-type="fig" rid="fig5">Figure 5B</xref>). These experiments demonstrate that PACER exerts its function through modulation of chromatin acetylation rather than methylation, as has been shown previously for several lncRNAs (<xref ref-type="bibr" rid="bib76">Wutz et al., 2002</xref>; <xref ref-type="bibr" rid="bib44">Nagano et al., 2008</xref>; <xref ref-type="bibr" rid="bib29">Khalil et al., 2009</xref>; <xref ref-type="bibr" rid="bib83">Zhao et al., 2010</xref>; <xref ref-type="bibr" rid="bib65">Tsai et al., 2010</xref>; <xref ref-type="bibr" rid="bib81">Yap et al., 2010</xref>; <xref ref-type="bibr" rid="bib71">Wang et al., 2011</xref>).<fig-group><fig id="fig5" position="float"><object-id pub-id-type="doi">10.7554/eLife.01776.016</object-id><label>Figure 5.</label><caption><title>PACER facilitates recruitment of p300 HAT and RNAP II pre-initiation complexes to increase histone acetylation and induce <italic>COX-2</italic> transcription.</title><p>(<bold>A</bold>) ChIP-qPCR was used to measure levels of histone H3 acetylation in control U937 and sh-lncRNA U937 monocytes (white bars), macrophages (grey-bars) and LPS-stimulated macrophages (black bars) at the COX-2 promoter, −1.25 kb upstream and −2.5 kb upstream of the COX-2 transcription start site. (<bold>B</bold>) ChIP-qPCR analysis of histone H4 acetylation across the COX-2 upstream region in control and sh-PACER U937 macrophages before and after LPS stimulation. (<bold>C</bold>) Association of p300 with the COX-2 upstream region was analyzed by ChIP-qPCR in control (filled squares) and sh-PACER (open squares) U937 macrophages before and after LPS stimulation. (<bold>D</bold>–<bold>F</bold>) Association of bulk RNAP II (<bold>D</bold>), S2-phosphorylated (<bold>E</bold>) and S5-phosphorylated (<bold>F</bold>) RNAP II across the <italic>COX-2</italic> locus analyzed by ChIP-qPCR in control HMECs (diamonds) or PACER-knockdown (triangles) HMECs. Association of RNAP II with the <italic>COX-2</italic> locus in control or siRNA-PACER transfected HMECs was assayed using ChIP-qPCR 72 hr post transfection. X-axis scale is relative to the COX-2 transcriptional start site.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.016">http://dx.doi.org/10.7554/eLife.01776.016</ext-link></p></caption><graphic xlink:href="elife01776f005"/></fig><fig id="fig5s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.017</object-id><label>Figure 5—figure supplement 1.</label><caption><title>Binding of c-Jun to the COX-2 region upon PACER knock-down in U937 monocytes, macrophages, and LPS-stimulated macrophages analyzed by ChIP.</title><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.017">http://dx.doi.org/10.7554/eLife.01776.017</ext-link></p></caption><graphic xlink:href="elife01776fs010"/></fig><fig id="fig5s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01776.018</object-id><label>Figure 5—figure supplement 2.</label><caption><title>Levels of H3K4 dimethylation at COX-2 region upon PACER knock-down in U937 monocytes, macrophages, and LPS-stimulated macrophages analyzed by ChIP.</title><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.018">http://dx.doi.org/10.7554/eLife.01776.018</ext-link></p></caption><graphic xlink:href="elife01776fs011"/></fig></fig-group></p><p>We sought to identify histone acetyltransferase complexes that could mediate PACER function. While HAT activities have not so far been implicated in the function of known lncRNAs, the p300 histone acetyltransferase has been shown to participate in COX-2 induction in human macrophages (<xref ref-type="bibr" rid="bib77">Xiao et al., 2011</xref>; <xref ref-type="bibr" rid="bib27">Kang et al., 2006</xref>). To examine the possibility that COX-2 lncRNA might affect recruitment of p300 to the <italic>COX-2</italic> locus, we performed ChIP-qPCR experiments. p300 was absent from the COX-2 promoter in non-induced macrophages but was recruited to the promoter upon LPS stimulation (<xref ref-type="fig" rid="fig5">Figure 5C</xref>). Primer walking identified a major binding site at approximately −0.25 kb. Importantly, in shPACER knockdown macrophages, LPS-induced p300 association was severely decreased suggesting that the lncRNA regulates <italic>COX-2</italic> expression through p300 recruitment and histone acetylation.</p><p>To explore the possibility that PACER could affect (directly or indirectly), the assembly of RNA Polymerase II preinitiation complexes and/or later steps in transcriptional activation, we used ChIP-qPCR to assess the association of bulk RNAP II and RNAP II modified at serine 5 or serine 2 within its C-terminal domain across the entire <italic>COX-2</italic> locus. Upon PACER knockdown, association of RNAP II was markedly reduced at both the mRNA promoter (amplicons −0.05, 0.05) and PACER promoter (−0.25, −0.45), as well as across the <italic>COX-2</italic> gene (0 to 4 kb) and PACER gene (- 1 kb to −0.25 kb) (<xref ref-type="fig" rid="fig5">Figure 5D</xref>). Elongating/terminating S2-modified polymerase association, which is usually skewed toward the ends of transcribed regions (6 kb for mRNA, −1 kb for lncRNA in this case), was also markedly lower in KD cells (<xref ref-type="fig" rid="fig5">Figure 5E</xref>). Finally, early initiating/paused S5-modified polymerase, usually enriched within the first 50 bp of the transcribed region (0 for mRNA, −0.25 to −0.45 kb for lncRNA here) was also significantly decreased upon PACER knockdown (<xref ref-type="fig" rid="fig5">Figure 5F</xref>). These results demonstrate that PACER influences early steps in transcription initiation at the stage of formation of RNAP II preinitiation complexes, likely through p300-induced changes in promoter-associated histone acetylation.</p></sec><sec id="s2-6"><title>PACER directly interacts with the repressive NF-κB subunit p50 to occlude it from the COX-2 promoter</title><p>How does PACER recruit RNAP II initiation complexes and p300 to the COX-2 promoter? To explore the possibility that the lncRNA might directly interact with p300, we carried out RNA immunoprecipitation (RIP) experiments. Antibodies against p300 were incubated with extracts containing intact cellular RNAs, RNA was then purified from immunoprecipitates and analyzed by RT-qPCR. These experiments failed to detect the presence of lncRNA in p300 RIP samples (<xref ref-type="fig" rid="fig6">Figure 6A</xref>). Similarly, p300 was not detected in RNA pulldown experiments, where we used lncRNA fragments to isolate interacting proteins from whole cell extracts (data not shown). We therefore conclude that PACER does not directly bind to p300.<fig id="fig6" position="float"><object-id pub-id-type="doi">10.7554/eLife.01776.019</object-id><label>Figure 6.</label><caption><title>PACER controls <italic>COX-2</italic> expression through binding of the repressive NF-κB subunit p50.</title><p>(<bold>A</bold>) Direct association of p50, p65/RelA, p300, and CTCF with PACER analyzed by RNA immunoprecipitation (RIP) in U937 macrophages. Primers for amplicon −0.85 kb were used. Signal is detected only in p50 immunoprecipitates. (<bold>B</bold>) Direct association of p50 with PACER analyzed by RNA immunoprecipitation (RIP) in U937 macrophages. Sites −0.45 kb through −0.85 kb lie within the lncRNA, while site −1.25 kb is outside and shows no detectable signal. IgG immunoprecipitation is shown as a control. (<bold>C</bold>) Control RIP experiments using the same antibodies as in (<bold>A</bold>) to analyze association with U1 snRNA. (<bold>D</bold>) Association of a small NF-κB subunit p50 with the COX-2 control region was analyzed by ChIP-qPCR in control and sh-lncRNA U937 macrophages. Binding is evident at the −0.25 kb site. (<bold>E</bold> and <bold>F</bold>) Expression of COX-2 mRNA (<bold>E</bold>) and protein (<bold>F</bold>) analyzed in control U937 monocytes and in mock and LPS-stimulated U937 macrophages and upon p50 knockdown. (<bold>G</bold> and <bold>H</bold>) Expression of COX-2 mRNA (<bold>G</bold>) and protein (<bold>H</bold>) analyzed in control U937 monocytes and in mock and LPS-stimulated U937 macrophages and upon p65 knockdown (<bold>I</bold>). A model of PACER-controlled COX-2 activation involving restricting promoter interaction of repressive p50 to facilitate recruitment of p300 HAT and RNAP II pre-initiation complexes to activate mRNA transcription.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.019">http://dx.doi.org/10.7554/eLife.01776.019</ext-link></p></caption><graphic xlink:href="elife01776f006"/></fig></p><p>LPS induces several signaling pathways in macrophages, which predominantly converge on activation of NF-κB and its target genes, including <italic>COX-2</italic>. NF-κB-mediated induction of <italic>COX-2</italic> has been previously shown to involve p300 recruitment to DNA (<xref ref-type="bibr" rid="bib77">Xiao et al., 2011</xref>; <xref ref-type="bibr" rid="bib27">Kang et al., 2006</xref>). This raised the possibility that PACER could interact with a member of the NF-κB family. Indeed, RIP experiments demonstrated that p50, the small subunit of NF-κB, interacts directly with the lncRNA (<xref ref-type="fig" rid="fig6">Figure 6A,B</xref>). We also detected p50 in RNA pulldown experiments using the COX-2 lncRNA but not its inverted control (data not shown). By contrast, PACER did not interact with the NF-κB large subunit p65/RelA by RIP or with CTCF and control SNRNP70 proteins (<xref ref-type="fig" rid="fig6">Figure 6A</xref>). Binding of p50 to the lncRNA could be detected using amplicons located at multiple positions within the PACER sequence (<xref ref-type="fig" rid="fig6">Figure 6B</xref>), demonstrating that p50 interacts with full length, intact PACER. Finally, binding to PACER was specific, because p50 did not interact with control U1 snRNA (<xref ref-type="fig" rid="fig6">Figure 6C</xref>). Given these results, we investigated whether association of p50 with the COX-2 promoter was affected by knockdown of PACER using ChIP-qPCR. We detected low level binding of p50 to COX-2 at position −0.25 kb in control macrophages before and after LPS induction (<xref ref-type="fig" rid="fig6">Figure 6D</xref>). This is in agreement with previous studies demonstrating that p50 levels do not change in response to LPS (<xref ref-type="bibr" rid="bib55">Saccani et al., 2003</xref>; <xref ref-type="bibr" rid="bib56">Saccani et al., 2006</xref>). In untreated PACER-knockdown macrophages, p50 is found at the −0.25 kb site in the COX-2 promoter. However upon LPS stimulation, which normally induces PACER expression, levels of bound p50 increase dramatically when PACER is decreased (<xref ref-type="fig" rid="fig6">Figure 6D</xref>). This indicates that PACER ensures low p50 association with the COX-2 promoter during activation of the gene.</p><p>Why is it necessary to keep binding of p50 ‘in check’ during NF-κB-mediated activation of <italic>COX-2</italic> expression? p50, unlike its dimerization partner p65, lacks an activation domain, and it has been demonstrated that p50 homodimers act as repressors (<xref ref-type="bibr" rid="bib56">Saccani et al., 2006</xref>; <xref ref-type="bibr" rid="bib28">Kastenbauer and Ziegler-Heitbrock, 1999</xref>). We therefore verified whether p50 might repress <italic>COX-2</italic> transcription in the absence of stimuli that normally induce COX-2 expression. To do this, we established p50-knockdown U937 monocyte lines using lentiviral delivery and measured COX-2 levels in KD cells compared to the control U937. Indeed, knockdown of p50 led to a marked increase of COX-2 mRNA and protein expression (<xref ref-type="fig" rid="fig6">Figure 6E,F</xref>). As expected, the increase is particularly prominent in undifferentiated U937 monocytes and unstimulated monocytes, whereas upon LPS stimulation COX-2 levels are similar in wt and p50 KD macrophages. These results demonstrate that the small NF-κB subunit p50 by itself functions as a repressor of <italic>COX-2</italic> ‘basal’, non-induced transcription. To demonstrate the distinct role of p65 subunit on COX-2 expression, we performed lentiviral-mediated knockdown of p65 in U937 cells and measured COX-2 expression in U937 monocytes, macrophages, and during LPS induction (<xref ref-type="fig" rid="fig6">Figure 6G,H</xref>). Expectedly, p65 knockdown had no influence on ‘basal’ expression of COX-2 mRNA or protein in monocytes or non-induced macrophages (<xref ref-type="fig" rid="fig6">Figure 6G</xref>, left and 6H, lanes 1,2 and 6,7). Upon LPS stimulation however, COX-2 induction was severely impaired in p65 knock-down macrophages (<xref ref-type="fig" rid="fig6">Figure 6G</xref>, right, and 6H lanes 3–5 and 8–10). This indicates that the activating role of p65 is evident only during LPS induction, subsequent to the p50-mediated PACER effects at the COX-2 promoter.</p><p>Altogether, these results lead us to postulate a model in which the role of PACER lncRNA in <italic>COX-2</italic> expression is to physically interact and restrict excess p50 from binding the promoter to facilitate the exchange of repressive p50/50 homodimers for p50/p65 heterodimers, thus allowing recruitment of p300, induction of histone hyperacetylation/chromatin remodeling, and consequent assembly of RNAP II complexes competent for transcription activation (<xref ref-type="fig" rid="fig6">Figure 6G</xref>).</p></sec></sec><sec id="s3" sec-type="discussion"><title>Discussion</title><p>In recent years, there has been an explosion of interest in non-coding RNAs, resulting in the description of distinct ncRNA classes based on their length and function (<xref ref-type="bibr" rid="bib17">Esteller, 2011</xref>). While the activities of small ncRNA species have been well-characterized, the roles of long ncRNAs in gene regulation have been deciphered in only several cases. The predominant model proposes that lncRNAs serve to guide enzymatic complexes to carry out chromatin modifications at specific genes, such as HOTAIR that recruits PRC2 and LSD1 complexes to silence <italic>HOXD</italic> genes and other distant loci (<xref ref-type="bibr" rid="bib65">Tsai et al., 2010</xref>); HOTTIP that recruits MLL1 to activate transcription of <italic>HOXA</italic> genes (<xref ref-type="bibr" rid="bib71">Wang et al., 2011</xref>); and Evf2 ncRNA that recruits the transcription factor DLX2 to activate expression of the <italic>Dlx6</italic> gene (<xref ref-type="bibr" rid="bib6">Bond et al., 2009</xref>). Even in the vicinity of the <italic>COX-2</italic> locus, a recent study identified a species of lncRNA that interacts with the RNA binding proteins hnRNP A/B and hnRNP A2/B1, which might act as transcriptional regulators for a variety of immune-related genes but, interestingly, not <italic>COX-2</italic> (<xref ref-type="bibr" rid="bib8">Carpenter et al., 2013</xref>).</p><p>In our study, we describe an unexpected mechanism of gene regulation by an lncRNA (PACER) involving occlusion of a repressor complex (NF-κB p50) from promoter association which then promotes assembly of co-activators (p300 HAT) and RNAP II pre-initiation complexes. PACER directly interacts with p50, which can form both active heterodimers with p65/RelA during the normal course of NF-κB pathway activation, and inactive p50/p50 homodimers which lack transcription activation domains present in p65/RelA. The repressor role of p50 at <italic>COX-2</italic> is consistent with previous studies showing that p50 homodimers inhibit LPS-induced activation of the <italic>TNFα</italic> gene in macrophages and other cells (<xref ref-type="bibr" rid="bib56">Saccani et al., 2006</xref>; <xref ref-type="bibr" rid="bib28">Kastenbauer and Ziegler-Heitbrock, 1999</xref>). Moreover, tumor-associated macrophages, which are refractory to LPS exhibit strong overexpression of p50 and fail to induce TNFα expression despite normal levels of p65/RelA and its ability to re-localize to the nucleus upon LPS treatment (<xref ref-type="bibr" rid="bib56">Saccani et al., 2006</xref>). Importantly, the molecular mechanism by which PACER regulates COX-2 expression does not directly involve the p65/RelA because PACER does not physically interact with this subunit, and the role of p65/RelA is only evident in the induction stage of LPS-mediated COX-2 expression, while PACER mainly affects ‘basal’, unstimulated COX-2 transcription.</p><p>We also describe an additional unexpected mechanism of gene regulation by an lncRNA that involves recruitment of the histone acetyltransferase p300 to catalyze the observed increase in histone acetylation at the <italic>COX-2</italic> locus upon stimulus-induction. We postulate that a large class of activation-related lncRNAs, such as enhancer-associated RNAs (eRNAs), might employ this mode of gene regulation, utilizing NF-κB or other transcription factors as baits to recruit CBP/p300. In this regard, it is interesting to note that enhancers are often co-inhabited by CBP/p300 and lncRNAs (<xref ref-type="bibr" rid="bib31">Kim et al., 2010</xref>; <xref ref-type="bibr" rid="bib67">Visel et al., 2009</xref>). Moreover, CBP/p300 binding, extragenic non-coding transcription, H3K4me1 and increased H3K27 acetylation constitute an active enhancer signature (<xref ref-type="bibr" rid="bib67">Visel et al., 2009</xref>; <xref ref-type="bibr" rid="bib51">Rada-Iglesias et al., 2011</xref>).</p><p>Ultimately, the function of the PACER lncRNA is to facilitate assembly of RNA Polymerase II pre-initiation complexes. Upon lncRNA knockdown, the levels of elongating RNAP II and initiating S5P-modified RNAP II are significantly lower, thus affecting the ability of the transcription machinery to efficiently initiate transcription of the <italic>COX-2</italic> gene. While these effects are indirect, they nevertheless demonstrate that PACER belongs to a novel, unanticipated class of <italic>bona-fide</italic> transcriptional regulators.</p><p>Our study also describes an unexpected mode of gene regulation by CTCF. CTCF has been previously demonstrated to regulate expression of certain non-coding RNAs (<xref ref-type="bibr" rid="bib60">Sopher et al., 2011</xref>; <xref ref-type="bibr" rid="bib61">Spencer et al., 2011</xref>), but the mechanisms it utilizes remain largely unknown. In this study, we show that CTCF generates a chromatin domain that encompasses the COX-2 core promoter and the upstream regulatory region to restrict adjacent repressive heterochromatin outside of the CTCF-bracketed region. This arrangement ensures induction of the regulatory lncRNA PACER, which orchestrates recruitment of the transcriptional activator p300 and transcription-competent RNAP II complexes. This represents an unanticipated link between CTCF-regulated non-coding RNA and the assembly of RNAP II pre-initiation complexes. Notably, CTCF has been shown to interact with RNAP II and regulate its function, but at the level of transcriptional pausing and alternative splicing (<xref ref-type="bibr" rid="bib10">Chernukhin et al., 2007</xref>; <xref ref-type="bibr" rid="bib68">Wada et al., 2009</xref>), not transcription initiation.</p><p>The locations of the two chromatin-bound CTCF/cohesin complexes in the <italic>COX-2</italic> locus suggest that they might perform distinct roles in gene regulation. The −2.5 kb site alone could function in a manner similar to CTCF recognition elements found at the interface between hetero- and euchromatin, such as in the human <italic>p16/CDKN2A</italic>, <italic>FOXJ3, IGF2/H19, β-GLOBIN, TCR or HOXD</italic> loci, where CTCF establishes a chromatin boundary that protects genes from epigenetic silencing (<xref ref-type="bibr" rid="bib75">Witcher and Emerson, 2009</xref>; <xref ref-type="bibr" rid="bib13">Cuddapah et al., 2009</xref>; <xref ref-type="bibr" rid="bib7">Carabana et al., 2011</xref>) or influence from adjacent domains (<xref ref-type="bibr" rid="bib5">Bell and Felsenfeld, 2000</xref>; <xref ref-type="bibr" rid="bib25">Hark et al., 2000</xref>; <xref ref-type="bibr" rid="bib84">Zhong and Krangel, 1999</xref>; <xref ref-type="bibr" rid="bib33">Li and Stamatoyannopoulos, 1994</xref>). On the other hand, the role of CTCF complexes associated with gene promoters is much less understood. Our in vitro analysis of the COX-2 promoter showed that CTCF specifically recognizes a DNA sequence located in the transcribed region, at position +65 (<xref ref-type="fig" rid="fig2s2 fig2s3">Figure 2—figure supplements 2, 3</xref>). Similar doublets of CTCF sites, with one located several kilobases upstream and the other within the transcribed region, are found in many genes (data not shown), which could suggest a specific mode of gene regulation by CTCF. We speculate that such doublets participate in establishing a ‘promoter-proximal’ open chromatin domain, possibly through long range chromatin tethering/looping mediated by two CTCF/cohesin complexes bound to distal sites. Both the enhancer and promoter elements would be positioned inside one chromatin loop, effectively insulated by CTCF/cohesin-mediated boundaries from the surrounding environment, which could restrict the action of enhancers to the appropriate promoter.</p><p>Deciphering the mechanisms of <italic>COX-2</italic> gene regulation is of great clinical interest. COX-2 activity has been shown to be instrumental in the development of several types of cancer, including colon, breast, and prostate (<xref ref-type="bibr" rid="bib81">Yap et al., 2010</xref>). For this reason, targeting COX-2 activity with non-steroidal anti-inflammatory drugs (NSAIDs) or agents designed to specifically block COX-2 activity have been approved for therapeutic use (<xref ref-type="bibr" rid="bib57">Salinas et al., 2007</xref>). However, because these drugs have significant off-target side effects, other strategies that specifically modulate <italic>COX-2</italic> transcription could be an attractive alternative approach. For example, <italic>COX-2</italic> downregulation in early stages of tumorigenesis by interfering with PACER-induced COX-2 expression would be predicted to have therapeutic benefits. In addition, <italic>COX-2</italic> silencing has been observed in numerous analyses of tumor specimens and cancer cell lines (<xref ref-type="bibr" rid="bib43">Murata et al., 2004</xref>; <xref ref-type="bibr" rid="bib64">Toyota et al., 2000</xref>; <xref ref-type="bibr" rid="bib37">Meng et al., 2011</xref>). It is attractive to speculate that re-activating <italic>COX-2</italic> expression in advanced stages of cancer by modulating PACER expression would increase COX-2-mediated local tissue mobilization and inflammation, consequently leading to more efficient tumor clearing by immune cells.</p></sec><sec id="s4" sec-type="materials|methods"><title>Materials and methods</title><sec id="s4-1"><title>Cell culture</title><p>Early passages of human mammary epithelial cells were obtained from Martha Stampfer. Cells were grown in MEGM BulletKit (Lonza, Walkersville, MD) media at 37°C in 10% O<sub>2</sub> and 5% CO<sub>2</sub> and maintained according to procedures outlined on the Stampfer laboratory website (<ext-link ext-link-type="uri" xlink:href="http://hmec.lbl.gov/other/procs.html">http://HMEC.lbl.gov/other/procs.html</ext-link>). Human histiocytic lymphoma monocyte cell line U937 (<xref ref-type="bibr" rid="bib63">Sundstrom and Nilsson, 1976</xref>) was kindly provided by Dr Sebastien Landry. U937 cells were maintained in RPMI medium (Mediatech, Corning, Manassas, VA) supplemented with 10% fetal bovine serum and antibiotics. For macrophage differentiation, cells were seeded onto plates at 0.5 × 10<sup>6</sup>/ml and treated with 0.2 mM phorbol myristate acetate (PMA, Sigma, St. Louis, MO) for 48 hr, followed by 24 hr incubation without PMA. To induce COX-2 expression, U937-derived macrophages were treated with LPS (2 µg/ml) for 6 hr.</p></sec><sec id="s4-2"><title>ChIP</title><p>Cells in exponential growth phase (HMEC, undifferentiated U937) or differentiated (macrophages) were cross-linked with 1% formaldehyde in 5 mM Hepes pH 8.0, 10 mM NaCl, 0.1 mM EDTA, 50 µM EGTA for 10 min at RT. Cross-linking was stopped with 0.125M glycine. Adherent cells were washed two times with PBS and collected by scraping into ice-cold PBS followed by centrifugation at 300×<italic>g</italic> for 5 min. Suspension cells (U937) were washed two times and collected by centrifugation at 300×<italic>g</italic> for 5 min at 4°C. Cells were lysed by rotating for 10 min on ice in 50 mM K-Hepes pH 7.5, 140 mM NaCl, 1 mM EDTA, 10% glycerol, 0.5% NP-40, 0.25% Triton X-100, and protease inhibitors and then centrifuged at 600×<italic>g</italic> for 10 min at 4°C. Chromatin pellets were washed once in 10 mM Tris–HCl pH 8.0, 0.2M NaCl, 0.1 mM EDTA, 50 µM EGTA and resuspended in 10 mM Tris–HCl pH 8.0, 0.1% NP-40, 0.1 mM EDTA, 50 µM EGTA. Sonication (20 rounds of 10s at power setting 3.5) was performed with a Branson sonicator, followed by centrifugation to clear the chromatin solution from debris. An aliquot of chromatin corresponding to 10<sup>6</sup> cells was brought up to 1 ml with 20 mM HEPES, pH7.9, 0.2M NaCl, 2 mM EDTA, 0.1% Na-DOC, 0.5% Triton X-100, 1 mg/ml BSA, and protease inhibitors and incubated with 2 µg of antibodies overnight at 4°C. To capture immune complexes, 10 µl of protein G beads was added and incubated for 2 hr at 4°C. Beads were washed (wash 1: 20 mM HEPES, pH7.9, 0.2M NaCl, 2 mM EDTA, 0.1% Na-DOC, 0.5% Triton X-100; wash 2: same as wash 1 but with 0.4M NaCl; wash 3: 20 mM Tris–HCl, pH 8.0, 0.25M LiCl, 2 mM EDTA, 0.25% Na-DOC; wash 4: TE, 0.1% NP40), immune complexes were detached from beads by 10 min incubation at 65°C in 0.1M Tris–HCl, pH 8.0, 1% SDS, followed by adding 140 µg of Proteinase K (NEB, Ipswich, MA) and NaCl to 0.1M, and incubation for 2 hr at 42°C and then overnight at 65°C. DNA was purified by phenol/chloroform extraction and analyzed by qPCR. IP efficiencies were calculated as a ratio between qPCR signals obtained with IP material and input DNA processed in parallel with ChIP samples. The following antibodies were used in ChIP experiments: CTCF (Millipore, Billerica, MA; Cell Signalling, Danvers, MA; Santa Cruz Biotechnology, Dallas, TX), p50 (Abcam, Cambridge, MA), Rad21 (Abcam), RNAP II, H2AZ (Active Motif, Carlsbad, CA), H3K4me2, H3K4me3, H3K9me3, H4K20me3, H4K8Ac, H3Ac (Abcam), p300-Ac (Cell Signaling). qPCR primers are listed in <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1</xref>. All ChIP experiments were repeated three times and qPCR reactions were performed in triplicates.</p></sec><sec id="s4-3"><title>Subcellular fractionation</title><p>Cells were scraped into ice-cold PBS, collected by centrifugation at 300×<italic>g</italic> for 5 min at 4°C, resuspended in 5 vol of cytoplasmic extract buffer (CEB, 10 mM HEPES pH 7.9, 50 mM NaCl, 1 mM EDTA, 0.3% NP40, 2.5 mM MgCl<sub>2</sub>, 1 mM DTT, protease and RNAse inhibitors) and kept on ice for 5 min. After centrifugation at 600×<italic>g</italic> for 5 min at 4°C, the cytoplasmic fraction was collected and the nuclear pellet was resuspended in nuclear extract buffer (NEB, 10 mM HEPES pH 7.9, 500 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.1% NP40, 0.5% Triton X-100, 2.5 mM MgCl<sub>2</sub>, 1 mM DTT, protease and RNAse inhibitors), kept on ice for 10 min and centrifuged for 10 min at max speed. The nucleoplasmic fraction was collected and the pellet was resuspended in the same buffer, sonicated five times for 10 s at power 3 and cleared by centrifugation to produce the chromatin-bound fraction.</p></sec><sec id="s4-4"><title>mRNA and lncRNA quantification</title><p>RNA purification was performed with Trizol (Life Technologies, Carlsbad, CA) according to the manufacturer's protocol. cDNA was produced using the Superscript III system (Life Technologies) using random hexamers or strand-specific primers listed in <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1</xref>. Relative quantities of mRNA were calculated using standard curves generated with cDNAs containing the highest concentration of a given target. For the ncRNA, the standard curves were generated with genomic DNA. The results are presented as the mean ± s.d of three independent RNA preparations.</p></sec><sec id="s4-5"><title>ChIP-chip</title><p>ChIP DNA and reference input DNA were amplified by ligation-mediated PCR according to NimbleGen (Roche, Madison, WI) recommendations. Briefly, 1 µg of DNA was blunted with T4 DNA polymerase (NEB), double-stranded staggered oligonucleotide adaptors were ligated, and PCR amplification was performed with oligos specific to the adaptors using a mixture of Taq polymerase (NEB) and PfuTurbo (Stratagene, Agilent Technologies, Santa Clara, CA ). 4 μg of each DNA was purified and sent to NimbleGen for probe preparation and hybridization. LM-PCR amplified DNA was verified to be enriched for target sequences that were detected by ChIP.</p><p>Custom high density genomic tiling arrays contained ∼80 regions covering selected genes and 50 or 100 kb flanks on either side. Array manufacturing, probe labeling, hybridization, and signal acquisition were conducted by NimbleGen. Data were visualized using SignalMap software (NimbleGen, Roche).</p></sec><sec id="s4-6"><title>Reporter gene assays</title><p>Constructs were cloned into the pGL3basic vector (Promega, Madison, WI). Plasmids were transfected in a 20:1 ratio (pGL3 to control Renilla Luc pRL-TK vector) using Lipofectamine 2000 (Life Technologies) into HEK293T cells in 96-well dishes and assayed 24 hr post transfection using the Dual-Glo Luciferase assay system (Promega). All transfections were done in five replicates.</p></sec><sec id="s4-7"><title>Western blotting</title><p>Western blotting was performed using standard protocols using the following antibodies: CTCF (Becton Dickinson, Franklin Lakes, NJ), COX-2 (Cayman, Ann Arbor, MI), p50 (Abcam), TFIIB (Santa Cruz), actin (Sigma).</p></sec><sec id="s4-8"><title>CTCF binding site prediction</title><p>Computer-based site prediction was performed using the online tool at <ext-link ext-link-type="uri" xlink:href="http://insulatordb.uthsc.edu/">http://insulatordb.uthsc.edu/</ext-link>.</p></sec><sec id="s4-9"><title>RACE</title><p>5′ and 3′ ends of COX-2 lncRNA were mapped using Rapid Amplification of cDNA ends (RACE) kit (Roche) according to the manufacturer's recommendations. Sequences of primers used for strand-specific cDNA synthesis and RACE amplification reactions are listed in <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1</xref>. The resulting single RACE products (from five independent 5′ and three independent 3′ amplifications) were gel-purified and sequenced.</p></sec><sec id="s4-10"><title>RNA immunoprecipitation (RIP)</title><p>RIP experiments were performed using MagnaRIP RNA-Binding Protein Immunoprecipitation Kit (Millipore) according to the manufacturer's recommendations. Antibodies were purchased from Santa Cruz Biotechnology (p50, p65, p300) and Millipore (CTCF). qPCR primers are listed in <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1</xref>. All experiments were repeated three times and qPCR reactions were performed in triplicates.</p></sec><sec id="s4-11"><title>siRNA</title><p>siRNA against CTCF, custom-designed siRNA against PACER and control-scrambled siRNAs were purchased from Dharmacon (Thermo Fisher, Waltham, MA). siRNAs against SMC3 were purchased from Ambion (Life Technologies). Sequences of custom siRNA are listed in <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1</xref>. 1 × 10<sup>5</sup>/ml HMECs were transfected according to the manufacturer's protocol using DharmaFect I reagent. For PACER, four siRNA were designed and tested, from which two efficiently knocked down lncRNA expression (siPACER-753 and siPACER-870, data not shown). The siPACER-753 was most efficient and was used in all experiments. Cells were harvested 72 hr after transfection.</p></sec><sec id="s4-12"><title>shRNA</title><p>Short hairpin sequence against COX-2 lncRNA was created by linking sense siRNA 870, loop, antisense siRNA870 and RNAP II terminator sequences; and cloning into BamHI and EcoRI sites in pGreenPuro lentiviral vector (SBI, Mountain View, CA) downstream of the H1 promoter. shRNA against the Luciferase gene was used as a control. shRNAs against p50 and p65 were purchased from Sigma. U937 monocytes were transduced with lentiviral particles and selected with puromycin. Once stable lines were established, puromycin was removed at least 48 hr before harvesting or differentiation.</p></sec></sec></body><back><ack id="ack"><title>Acknowledgements</title><p>We gratefully acknowledge Dr Martha Stampfer for providing variant HMECs. This work was supported by grants from NIH (R01 CA159354), the Chambers Medical Foundation, and the GemCon Foundation to BME.</p></ack><sec sec-type="additional-information"><title>Additional information</title><fn-group content-type="competing-interest"><title>Competing interests</title><fn fn-type="conflict" id="conf1"><p>The authors declare that no competing interests exist.</p></fn></fn-group><fn-group content-type="author-contribution"><title>Author contributions</title><fn fn-type="con" id="con1"><p>MK, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con2"><p>BME, Conception and design, Analysis and interpretation of data, Drafting or revising the article</p></fn></fn-group></sec><sec sec-type="supplementary-material"><title>Additional files</title><supplementary-material id="SD1-data"><object-id pub-id-type="doi">10.7554/eLife.01776.020</object-id><label>Supplementary file 1.</label><caption><p>Sequences of oligonucleotides used in this study</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01776.020">http://dx.doi.org/10.7554/eLife.01776.020</ext-link></p></caption><media mime-subtype="docx" mimetype="application" xlink:href="elife01776s001.docx"/></supplementary-material></sec><ref-list><title>References</title><ref id="bib1"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Adriaenssens</surname><given-names>E</given-names></name><name><surname>Dumont</surname><given-names>L</given-names></name><name><surname>Lottin</surname><given-names>S</given-names></name><name><surname>Bolle</surname><given-names>D</given-names></name><name><surname>Leprêtre</surname><given-names>A</given-names></name><name><surname>Delobelle</surname><given-names>A</given-names></name><name><surname>Bouali</surname><given-names>F</given-names></name><name><surname>Dugimont</surname><given-names>T</given-names></name><name><surname>Coll</surname><given-names>J</given-names></name><name><surname>Curgy</surname><given-names>JJ</given-names></name></person-group><year>1998</year><article-title>H19 overexpression in breast adenocarcinoma stromal cells is associated with tumor values and steroid receptor status but independent of p53 and Ki-67 expression</article-title><source>The American Journal of Pathology</source><volume>153</volume><fpage>1597</fpage><lpage>1607</lpage><pub-id pub-id-type="doi">10.1016/S0002-9440(10)65748-3</pub-id></element-citation></ref><ref id="bib2"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Arias-Negrete</surname><given-names>S</given-names></name><name><surname>Keller</surname><given-names>K</given-names></name><name><surname>Chadee</surname><given-names>K</given-names></name></person-group><year>1995</year><article-title>Proinflammatory cytokines regulate cyclooxygenase-2 mRNA expression in human macrophages</article-title><source>Biochemical and Biophysical Research Communications</source><volume>208</volume><fpage>582</fpage><lpage>589</lpage><pub-id pub-id-type="doi">10.1006/bbrc.1995.1378</pub-id></element-citation></ref><ref id="bib3"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Barski</surname><given-names>A</given-names></name><name><surname>Cuddapah</surname><given-names>S</given-names></name><name><surname>Cui</surname><given-names>K</given-names></name><name><surname>Roh</surname><given-names>TY</given-names></name><name><surname>Schones</surname><given-names>DE</given-names></name><name><surname>Wang</surname><given-names>Z</given-names></name><name><surname>Wei</surname><given-names>G</given-names></name><name><surname>Chepelev</surname><given-names>I</given-names></name><name><surname>Zhao</surname><given-names>K</given-names></name></person-group><year>2007</year><article-title>High-resolution profiling of histone methylations in the human genome</article-title><source>Cell</source><volume>129</volume><fpage>823</fpage><lpage>837</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2007.05.009</pub-id></element-citation></ref><ref id="bib4"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Barsyte-Lovejoy</surname><given-names>D</given-names></name><name><surname>Lau</surname><given-names>SK</given-names></name><name><surname>Boutros</surname><given-names>PC</given-names></name><name><surname>Khosravi</surname><given-names>F</given-names></name><name><surname>Jurisica</surname><given-names>I</given-names></name><name><surname>Andrulis</surname><given-names>IL</given-names></name><name><surname>Tsao</surname><given-names>MS</given-names></name><name><surname>Penn</surname><given-names>LZ</given-names></name></person-group><year>2006</year><article-title>The c-Myc oncogene directly induces the H19 noncoding RNA by allele-specific binding to potentiate tumorigenesis</article-title><source>Cancer Research</source><volume>66</volume><fpage>5330</fpage><lpage>5337</lpage><pub-id pub-id-type="doi">10.1158/0008-5472.CAN-06-0037</pub-id></element-citation></ref><ref id="bib5"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bell</surname><given-names>AC</given-names></name><name><surname>Felsenfeld</surname><given-names>G</given-names></name></person-group><year>2000</year><article-title>Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene</article-title><source>Nature</source><volume>405</volume><fpage>482</fpage><lpage>485</lpage><pub-id pub-id-type="doi">10.1038/35013100</pub-id></element-citation></ref><ref id="bib6"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bond</surname><given-names>AM</given-names></name><name><surname>Vangompel</surname><given-names>MJ</given-names></name><name><surname>Sametsky</surname><given-names>EA</given-names></name><name><surname>Clark</surname><given-names>MF</given-names></name><name><surname>Savage</surname><given-names>JC</given-names></name><name><surname>Disterhoft</surname><given-names>JF</given-names></name><name><surname>Kohtz</surname><given-names>JD</given-names></name></person-group><year>2009</year><article-title>Balanced gene regulation by an embryonic brain ncRNA is critical for adult hippocampal GABA circuitry</article-title><source>Nature Neuroscience</source><volume>12</volume><fpage>1020</fpage><lpage>1027</lpage><pub-id pub-id-type="doi">10.1038/nn.2371</pub-id></element-citation></ref><ref id="bib7"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Carabana</surname><given-names>J</given-names></name><name><surname>Watanabe</surname><given-names>A</given-names></name><name><surname>Hao</surname><given-names>B</given-names></name><name><surname>Krangel</surname><given-names>MS</given-names></name></person-group><year>2011</year><article-title>A barrier-type insulator forms a boundary between active and inactive chromatin at the murine TCRbeta locus</article-title><source>The Journal of Immunology: official Journal of the American Association of Immunologists</source><volume>186</volume><fpage>3556</fpage><lpage>3562</lpage><pub-id pub-id-type="doi">10.4049/jimmunol.1003164</pub-id></element-citation></ref><ref id="bib8"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Carpenter</surname><given-names>S</given-names></name><name><surname>Aiello</surname><given-names>D</given-names></name><name><surname>Atianand</surname><given-names>MK</given-names></name><name><surname>Ricci</surname><given-names>EP</given-names></name><name><surname>Gandhi</surname><given-names>P</given-names></name><name><surname>Hall</surname><given-names>LL</given-names></name><name><surname>Byron</surname><given-names>M</given-names></name><name><surname>Monks</surname><given-names>B</given-names></name><name><surname>Henry-Bezy</surname><given-names>M</given-names></name><name><surname>Lawrence</surname><given-names>JB</given-names></name><name><surname>O'Neill</surname><given-names>LA</given-names></name><name><surname>Moore</surname><given-names>MJ</given-names></name><name><surname>Caffrey</surname><given-names>DR</given-names></name><name><surname>Fitzgerald</surname><given-names>KA</given-names></name></person-group><year>2013</year><article-title>A long noncoding RNA mediates both activation and repression of immune response genes</article-title><source>Science</source><volume>341</volume><fpage>789</fpage><lpage>792</lpage><pub-id pub-id-type="doi">10.1126/science.1240925</pub-id></element-citation></ref><ref id="bib9"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chen</surname><given-names>EP</given-names></name><name><surname>Smyth</surname><given-names>EM</given-names></name></person-group><year>2011</year><article-title>COX-2 and PGE2-dependent immunomodulation in breast cancer</article-title><source>Prostaglandins & Other Lipid Mediators</source><volume>96</volume><fpage>14</fpage><lpage>20</lpage><pub-id pub-id-type="doi">10.1016/j.prostaglandins.2011.08.005</pub-id></element-citation></ref><ref id="bib10"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chernukhin</surname><given-names>I</given-names></name><name><surname>Shamsuddin</surname><given-names>S</given-names></name><name><surname>Kang</surname><given-names>SY</given-names></name><name><surname>Bergström</surname><given-names>R</given-names></name><name><surname>Kwon</surname><given-names>YW</given-names></name><name><surname>Yu</surname><given-names>W</given-names></name><name><surname>Whitehead</surname><given-names>J</given-names></name><name><surname>Mukhopadhyay</surname><given-names>R</given-names></name><name><surname>Docquier</surname><given-names>F</given-names></name><name><surname>Farrar</surname><given-names>D</given-names></name><name><surname>Morrison</surname><given-names>I</given-names></name><name><surname>Vigneron</surname><given-names>M</given-names></name><name><surname>Wu</surname><given-names>SY</given-names></name><name><surname>Chiang</surname><given-names>CM</given-names></name><name><surname>Loukinov</surname><given-names>D</given-names></name><name><surname>Lobanenkov</surname><given-names>V</given-names></name><name><surname>Ohlsson</surname><given-names>R</given-names></name><name><surname>Klenova</surname><given-names>E</given-names></name></person-group><year>2007</year><article-title>CTCF interacts with and recruits the largest subunit of RNA polymerase II to CTCF target sites genome-wide</article-title><source>Molecular and Cellular Biology</source><volume>27</volume><fpage>1631</fpage><lpage>1648</lpage><pub-id pub-id-type="doi">10.1128/MCB.01993-06</pub-id></element-citation></ref><ref id="bib11"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chun</surname><given-names>KS</given-names></name><name><surname>Surh</surname><given-names>YJ</given-names></name></person-group><year>2004</year><article-title>Signal transduction pathways regulating cyclooxygenase-2 expression: potential molecular targets for chemoprevention</article-title><source>Biochemical Pharmacology</source><volume>68</volume><fpage>1089</fpage><lpage>1100</lpage><pub-id pub-id-type="doi">10.1016/j.bcp.2004.05.031</pub-id></element-citation></ref><ref id="bib12"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Crawford</surname><given-names>YG</given-names></name><name><surname>Gauthier</surname><given-names>ML</given-names></name><name><surname>Joubel</surname><given-names>A</given-names></name><name><surname>Mantei</surname><given-names>K</given-names></name><name><surname>Kozakiewicz</surname><given-names>K</given-names></name><name><surname>Afshari</surname><given-names>CA</given-names></name><name><surname>Tlsty</surname><given-names>TD</given-names></name></person-group><year>2004</year><article-title>Histologically normal human mammary epithelia with silenced p16(INK4a) overexpress COX-2, promoting a premalignant program</article-title><source>Cancer Cell</source><volume>5</volume><fpage>263</fpage><lpage>273</lpage><pub-id pub-id-type="doi">10.1016/S1535-6108(04)00023-6</pub-id></element-citation></ref><ref id="bib13"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cuddapah</surname><given-names>S</given-names></name><name><surname>Jothi</surname><given-names>R</given-names></name><name><surname>Schones</surname><given-names>DE</given-names></name><name><surname>Roh</surname><given-names>TY</given-names></name><name><surname>Cui</surname><given-names>K</given-names></name><name><surname>Zhao</surname><given-names>K</given-names></name></person-group><year>2009</year><article-title>Global analysis of the insulator binding protein CTCF in chromatin barrier regions reveals demarcation of active and repressive domains</article-title><source>Genome Research</source><volume>19</volume><fpage>24</fpage><lpage>32</lpage><pub-id pub-id-type="doi">10.1101/gr.082800.108</pub-id></element-citation></ref><ref id="bib14"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cunnington</surname><given-names>MS</given-names></name><name><surname>Santibanez Koref</surname><given-names>M</given-names></name><name><surname>Mayosi</surname><given-names>BM</given-names></name><name><surname>Burn</surname><given-names>J</given-names></name><name><surname>Keavney</surname><given-names>B</given-names></name></person-group><year>2010</year><article-title>Chromosome 9p21 SNPs associated with multiple disease Phenotypes correlate with ANRIL expression</article-title><source>PLOS Genetics</source><volume>6</volume><fpage>e1000899</fpage><pub-id pub-id-type="doi">10.1371/journal.pgen.1000899</pub-id></element-citation></ref><ref id="bib15"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>de Moraes</surname><given-names>E</given-names></name><name><surname>Dar</surname><given-names>NA</given-names></name><name><surname>de Moura Gallo</surname><given-names>CV</given-names></name><name><surname>Hainaut</surname><given-names>P</given-names></name></person-group><year>2007</year><article-title>Cross-talks between cyclooxygenase-2 and tumor suppressor protein p53: balancing life and death during inflammatory stress and carcinogenesis</article-title><source>International Journal of cancer Journal International Du Cancer</source><volume>121</volume><fpage>929</fpage><lpage>937</lpage><pub-id pub-id-type="doi">10.1002/ijc.22899</pub-id></element-citation></ref><ref id="bib16"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dinger</surname><given-names>ME</given-names></name><name><surname>Amaral</surname><given-names>PP</given-names></name><name><surname>Mercer</surname><given-names>TR</given-names></name><name><surname>Pang</surname><given-names>KC</given-names></name><name><surname>Bruce</surname><given-names>SJ</given-names></name><name><surname>Gardiner</surname><given-names>BB</given-names></name><name><surname>Askarian-Amiri</surname><given-names>ME</given-names></name><name><surname>Ru</surname><given-names>K</given-names></name><name><surname>Soldà</surname><given-names>G</given-names></name><name><surname>Simons</surname><given-names>C</given-names></name><name><surname>Sunkin</surname><given-names>SM</given-names></name><name><surname>Crowe</surname><given-names>ML</given-names></name><name><surname>Grimmond</surname><given-names>SM</given-names></name><name><surname>Perkins</surname><given-names>AC</given-names></name><name><surname>Mattick</surname><given-names>JS</given-names></name></person-group><year>2008</year><article-title>Long noncoding RNAs in mouse embryonic stem cell pluripotency and differentiation</article-title><source>Genome Research</source><volume>18</volume><fpage>1433</fpage><lpage>1445</lpage><pub-id pub-id-type="doi">10.1101/gr.078378.108</pub-id></element-citation></ref><ref id="bib17"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Esteller</surname><given-names>M</given-names></name></person-group><year>2011</year><article-title>Non-coding RNAs in human disease</article-title><source>Nature Reviews Genetics</source><volume>12</volume><fpage>861</fpage><lpage>874</lpage><pub-id pub-id-type="doi">10.1038/nrg3074</pub-id></element-citation></ref><ref id="bib18"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Fellig</surname><given-names>Y</given-names></name><name><surname>Ariel</surname><given-names>I</given-names></name><name><surname>Ohana</surname><given-names>P</given-names></name><name><surname>Schachter</surname><given-names>P</given-names></name><name><surname>Sinelnikov</surname><given-names>I</given-names></name><name><surname>Birman</surname><given-names>T</given-names></name><name><surname>Ayesh</surname><given-names>S</given-names></name><name><surname>Schneider</surname><given-names>T</given-names></name><name><surname>de Groot</surname><given-names>N</given-names></name><name><surname>Czerniak</surname><given-names>A</given-names></name><name><surname>Hochberg</surname><given-names>A</given-names></name></person-group><year>2005</year><article-title>H19 expression in hepatic metastases from a range of human carcinomas</article-title><source>Journal of Clinical Pathology</source><volume>58</volume><fpage>1064</fpage><lpage>1068</lpage><pub-id pub-id-type="doi">10.1136/jcp.2004.023648</pub-id></element-citation></ref><ref id="bib19"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Galbraith</surname><given-names>MD</given-names></name><name><surname>Allen</surname><given-names>MA</given-names></name><name><surname>Bensard</surname><given-names>CL</given-names></name><name><surname>Wang</surname><given-names>X</given-names></name><name><surname>Schwinn</surname><given-names>MK</given-names></name><name><surname>Qin</surname><given-names>B</given-names></name><name><surname>Long</surname><given-names>HW</given-names></name><name><surname>Daniels</surname><given-names>DL</given-names></name><name><surname>Hahn</surname><given-names>WC</given-names></name><name><surname>Dowell</surname><given-names>RD</given-names></name><name><surname>Espinosa</surname><given-names>JM</given-names></name></person-group><year>2013</year><article-title>HIF1A employs CDK8-mediator to stimulate RNAPII elongation in response to hypoxia</article-title><source>Cell</source><volume>153</volume><fpage>1327</fpage><lpage>1339</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2013.04.048</pub-id></element-citation></ref><ref id="bib20"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gauthier</surname><given-names>ML</given-names></name><name><surname>Pickering</surname><given-names>CR</given-names></name><name><surname>Miller</surname><given-names>CJ</given-names></name><name><surname>Fordyce</surname><given-names>CA</given-names></name><name><surname>Chew</surname><given-names>KL</given-names></name><name><surname>Berman</surname><given-names>HK</given-names></name><name><surname>Tlsty</surname><given-names>TD</given-names></name></person-group><year>2005</year><article-title>p38 regulates cyclooxygenase-2 in human mammary epithelial cells and is activated in premalignant tissue</article-title><source>Cancer Research</source><volume>65</volume><fpage>1792</fpage><lpage>1799</lpage><pub-id pub-id-type="doi">10.1158/0008-5472.CAN-04-3507</pub-id></element-citation></ref><ref id="bib21"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gibb</surname><given-names>EA</given-names></name><name><surname>Brown</surname><given-names>CJ</given-names></name><name><surname>Lam</surname><given-names>WL</given-names></name></person-group><year>2011</year><article-title>The functional role of long non-coding RNA in human carcinomas</article-title><source>Molecular Cancer</source><volume>10</volume><fpage>38</fpage><pub-id pub-id-type="doi">10.1186/1476-4598-10-38</pub-id></element-citation></ref><ref id="bib22"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gupta</surname><given-names>GP</given-names></name><name><surname>Nguyen</surname><given-names>DX</given-names></name><name><surname>Chiang</surname><given-names>AC</given-names></name><name><surname>Bos</surname><given-names>PD</given-names></name><name><surname>Kim</surname><given-names>JY</given-names></name><name><surname>Nadal</surname><given-names>C</given-names></name><name><surname>Gomis</surname><given-names>RR</given-names></name><name><surname>Manova-Todorova</surname><given-names>K</given-names></name><name><surname>Massagué</surname><given-names>J</given-names></name></person-group><year>2007</year><article-title>Mediators of vascular remodelling co-opted for sequential steps in lung metastasis</article-title><source>Nature</source><volume>446</volume><fpage>765</fpage><lpage>770</lpage><pub-id pub-id-type="doi">10.1038/nature05760</pub-id></element-citation></ref><ref id="bib23"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gupta</surname><given-names>RA</given-names></name><name><surname>Shah</surname><given-names>N</given-names></name><name><surname>Wang</surname><given-names>KC</given-names></name><name><surname>Kim</surname><given-names>J</given-names></name><name><surname>Horlings</surname><given-names>HM</given-names></name><name><surname>Wong</surname><given-names>DJ</given-names></name><name><surname>Tsai</surname><given-names>MC</given-names></name><name><surname>Hung</surname><given-names>T</given-names></name><name><surname>Argani</surname><given-names>P</given-names></name><name><surname>Rinn</surname><given-names>JL</given-names></name><name><surname>Wang</surname><given-names>Y</given-names></name><name><surname>Brzoska</surname><given-names>P</given-names></name><name><surname>Kong</surname><given-names>B</given-names></name><name><surname>Li</surname><given-names>R</given-names></name><name><surname>West</surname><given-names>RB</given-names></name><name><surname>van de Vijver</surname><given-names>MJ</given-names></name><name><surname>Sukumar</surname><given-names>S</given-names></name><name><surname>Chang</surname><given-names>HY</given-names></name></person-group><year>2010</year><article-title>Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis</article-title><source>Nature</source><volume>464</volume><fpage>1071</fpage><lpage>1076</lpage><pub-id pub-id-type="doi">10.1038/nature08975</pub-id></element-citation></ref><ref id="bib24"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Guttman</surname><given-names>M</given-names></name><name><surname>Amit</surname><given-names>I</given-names></name><name><surname>Garber</surname><given-names>M</given-names></name><name><surname>French</surname><given-names>C</given-names></name><name><surname>Lin</surname><given-names>MF</given-names></name><name><surname>Feldser</surname><given-names>D</given-names></name><name><surname>Huarte</surname><given-names>M</given-names></name><name><surname>Zuk</surname><given-names>O</given-names></name><name><surname>Carey</surname><given-names>BW</given-names></name><name><surname>Cassady</surname><given-names>JP</given-names></name><name><surname>Cabili</surname><given-names>MN</given-names></name><name><surname>Jaenisch</surname><given-names>R</given-names></name><name><surname>Mikkelsen</surname><given-names>TS</given-names></name><name><surname>Jacks</surname><given-names>T</given-names></name><name><surname>Hacohen</surname><given-names>N</given-names></name><name><surname>Bernstein</surname><given-names>BE</given-names></name><name><surname>Kellis</surname><given-names>M</given-names></name><name><surname>Regev</surname><given-names>A</given-names></name><name><surname>Rinn</surname><given-names>JL</given-names></name><name><surname>Lander</surname><given-names>ES</given-names></name></person-group><year>2009</year><article-title>Chromatin signature reveals over a thousand highly conserved large non-coding RNAs in mammals</article-title><source>Nature</source><volume>458</volume><fpage>223</fpage><lpage>227</lpage><pub-id pub-id-type="doi">10.1038/nature07672</pub-id></element-citation></ref><ref id="bib25"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hark</surname><given-names>AT</given-names></name><name><surname>Schoenherr</surname><given-names>CJ</given-names></name><name><surname>Katz</surname><given-names>DJ</given-names></name><name><surname>Ingram</surname><given-names>RS</given-names></name><name><surname>Levorse</surname><given-names>JM</given-names></name><name><surname>Tilghman</surname><given-names>SM</given-names></name></person-group><year>2000</year><article-title>CTCF mediates methylation-sensitive enhancer-blocking activity at the H19/Igf2 locus</article-title><source>Nature</source><volume>405</volume><fpage>486</fpage><lpage>489</lpage><pub-id pub-id-type="doi">10.1038/35013106</pub-id></element-citation></ref><ref id="bib26"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hibi</surname><given-names>K</given-names></name><name><surname>Nakamura</surname><given-names>H</given-names></name><name><surname>Hirai</surname><given-names>A</given-names></name><name><surname>Fujikake</surname><given-names>Y</given-names></name><name><surname>Kasai</surname><given-names>Y</given-names></name><name><surname>Akiyama</surname><given-names>S</given-names></name><name><surname>Ito</surname><given-names>K</given-names></name><name><surname>Takagi</surname><given-names>H</given-names></name></person-group><year>1996</year><article-title>Loss of H19 imprinting in esophageal cancer</article-title><source>Cancer Research</source><volume>56</volume><fpage>480</fpage><lpage>482</lpage></element-citation></ref><ref id="bib27"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kang</surname><given-names>YJ</given-names></name><name><surname>Wingerd</surname><given-names>BA</given-names></name><name><surname>Arakawa</surname><given-names>T</given-names></name><name><surname>Smith</surname><given-names>WL</given-names></name></person-group><year>2006</year><article-title>Cyclooxygenase-2 gene transcription in a macrophage model of inflammation</article-title><source>The Journal of Immunology: official Journal of the American Association of Immunologists</source><volume>177</volume><fpage>8111</fpage><lpage>8122</lpage><pub-id pub-id-type="doi">10.4049/jimmunol.177.11.8111</pub-id></element-citation></ref><ref id="bib28"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kastenbauer</surname><given-names>S</given-names></name><name><surname>Ziegler-Heitbrock</surname><given-names>HW</given-names></name></person-group><year>1999</year><article-title>NF-kappaB1 (p50) is upregulated in lipopolysaccharide tolerance and can block tumor necrosis factor gene expression</article-title><source>Infection and Immunity</source><volume>67</volume><fpage>1553</fpage><lpage>1559</lpage></element-citation></ref><ref id="bib29"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Khalil</surname><given-names>AM</given-names></name><name><surname>Guttman</surname><given-names>M</given-names></name><name><surname>Huarte</surname><given-names>M</given-names></name><name><surname>Garber</surname><given-names>M</given-names></name><name><surname>Raj</surname><given-names>A</given-names></name><name><surname>Rivea Morales</surname><given-names>D</given-names></name><name><surname>Thomas</surname><given-names>K</given-names></name><name><surname>Presser</surname><given-names>A</given-names></name><name><surname>Bernstein</surname><given-names>BE</given-names></name><name><surname>van Oudenaarden</surname><given-names>A</given-names></name><name><surname>Regev</surname><given-names>A</given-names></name><name><surname>Lander</surname><given-names>ES</given-names></name><name><surname>Rinn</surname><given-names>JL</given-names></name></person-group><year>2009</year><article-title>Many human large intergenic noncoding RNAs associate with chromatin-modifying complexes and affect gene expression</article-title><source>Proceedings of the National Academy of Sciences of the United States of America</source><volume>106</volume><fpage>11667</fpage><lpage>11672</lpage><pub-id pub-id-type="doi">10.1073/pnas.0904715106</pub-id></element-citation></ref><ref id="bib30"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kim</surname><given-names>TH</given-names></name><name><surname>Abdullaev</surname><given-names>ZK</given-names></name><name><surname>Smith</surname><given-names>AD</given-names></name><name><surname>Ching</surname><given-names>KA</given-names></name><name><surname>Loukinov</surname><given-names>DI</given-names></name><name><surname>Green</surname><given-names>RD</given-names></name><name><surname>Zhang</surname><given-names>MQ</given-names></name><name><surname>Lobanenkov</surname><given-names>VV</given-names></name><name><surname>Ren</surname><given-names>B</given-names></name></person-group><year>2007</year><article-title>Analysis of the vertebrate insulator protein CTCF-binding sites in the human genome</article-title><source>Cell</source><volume>128</volume><fpage>1231</fpage><lpage>1245</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2006.12.048</pub-id></element-citation></ref><ref id="bib31"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kim</surname><given-names>TK</given-names></name><name><surname>Hemberg</surname><given-names>M</given-names></name><name><surname>Gray</surname><given-names>JM</given-names></name><name><surname>Costa</surname><given-names>AM</given-names></name><name><surname>Bear</surname><given-names>DM</given-names></name><name><surname>Wu</surname><given-names>J</given-names></name><name><surname>Harmin</surname><given-names>DA</given-names></name><name><surname>Laptewicz</surname><given-names>M</given-names></name><name><surname>Barbara-Haley</surname><given-names>K</given-names></name><name><surname>Kuersten</surname><given-names>S</given-names></name><name><surname>Markenscoff-Papadimitriou</surname><given-names>E</given-names></name><name><surname>Kuhl</surname><given-names>D</given-names></name><name><surname>Bito</surname><given-names>H</given-names></name><name><surname>Worley</surname><given-names>PF</given-names></name><name><surname>Kreiman</surname><given-names>G</given-names></name><name><surname>Greenberg</surname><given-names>ME</given-names></name></person-group><year>2010</year><article-title>Widespread transcription at neuronal activity-regulated enhancers</article-title><source>Nature</source><volume>465</volume><fpage>182</fpage><lpage>187</lpage><pub-id pub-id-type="doi">10.1038/nature09033</pub-id></element-citation></ref><ref id="bib32"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kondo</surname><given-names>M</given-names></name><name><surname>Suzuki</surname><given-names>H</given-names></name><name><surname>Ueda</surname><given-names>R</given-names></name><name><surname>Osada</surname><given-names>H</given-names></name><name><surname>Takagi</surname><given-names>K</given-names></name><name><surname>Takahashi</surname><given-names>T</given-names></name><name><surname>Takahashi</surname><given-names>T</given-names></name></person-group><year>1995</year><article-title>Frequent loss of imprinting of the H19 gene is often associated with its overexpression in human lung cancers</article-title><source>Oncogene</source><volume>10</volume><fpage>1193</fpage><lpage>1198</lpage></element-citation></ref><ref id="bib34"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Li</surname><given-names>Q</given-names></name><name><surname>Liu</surname><given-names>N</given-names></name><name><surname>Shen</surname><given-names>B</given-names></name><name><surname>Zhou</surname><given-names>L</given-names></name><name><surname>Wang</surname><given-names>Y</given-names></name><name><surname>Wang</surname><given-names>Y</given-names></name><name><surname>Sun</surname><given-names>J</given-names></name><name><surname>Fan</surname><given-names>Z</given-names></name><name><surname>Liu</surname><given-names>RH</given-names></name></person-group><year>2009</year><article-title>Helicobacter pylori enhances cyclooxygenase 2 expression via p38MAPK/ATF-2 signaling pathway in MKN45 cells</article-title><source>Cancer Letters</source><volume>278</volume><fpage>97</fpage><lpage>103</lpage><pub-id pub-id-type="doi">10.1016/j.canlet.2008.12.032</pub-id></element-citation></ref><ref id="bib33"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Li</surname><given-names>Q</given-names></name><name><surname>Stamatoyannopoulos</surname><given-names>G</given-names></name></person-group><year>1994</year><article-title>Hypersensitive site 5 of the human beta locus control region functions as a chromatin insulator</article-title><source>Blood</source><volume>84</volume><fpage>1399</fpage><lpage>1401</lpage></element-citation></ref><ref id="bib35"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Liu</surname><given-names>CH</given-names></name><name><surname>Chang</surname><given-names>SH</given-names></name><name><surname>Narko</surname><given-names>K</given-names></name><name><surname>Trifan</surname><given-names>OC</given-names></name><name><surname>Wu</surname><given-names>MT</given-names></name><name><surname>Smith</surname><given-names>E</given-names></name><name><surname>Haudenschild</surname><given-names>C</given-names></name><name><surname>Lane</surname><given-names>TF</given-names></name><name><surname>Hla</surname><given-names>T</given-names></name></person-group><year>2001</year><article-title>Overexpression of cyclooxygenase-2 is sufficient to induce tumorigenesis in transgenic mice</article-title><source>The Journal of Biological Chemistry</source><volume>276</volume><fpage>18563</fpage><lpage>18569</lpage><pub-id pub-id-type="doi">10.1074/jbc.M010787200</pub-id></element-citation></ref><ref id="bib36"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Markosyan</surname><given-names>N</given-names></name><name><surname>Chen</surname><given-names>EP</given-names></name><name><surname>Ndong</surname><given-names>VN</given-names></name><name><surname>Yao</surname><given-names>Y</given-names></name><name><surname>Sterner</surname><given-names>CJ</given-names></name><name><surname>Chodosh</surname><given-names>LA</given-names></name><name><surname>Lawson</surname><given-names>JA</given-names></name><name><surname>Fitzgerald</surname><given-names>GA</given-names></name><name><surname>Smyth</surname><given-names>EM</given-names></name></person-group><year>2011</year><article-title>Deletion of cyclooxygenase 2 in mouse mammary epithelial cells delays breast cancer onset through augmentation of type 1 immune responses in tumors</article-title><source>Carcinogenesis</source><volume>32</volume><fpage>1441</fpage><lpage>1449</lpage><pub-id pub-id-type="doi">10.1093/carcin/bgr134</pub-id></element-citation></ref><ref id="bib37"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Meng</surname><given-names>XY</given-names></name><name><surname>Zhu</surname><given-names>ST</given-names></name><name><surname>Zong</surname><given-names>Y</given-names></name><name><surname>Wang</surname><given-names>YJ</given-names></name><name><surname>Li</surname><given-names>P</given-names></name><name><surname>Zhang</surname><given-names>ST</given-names></name></person-group><year>2011</year><article-title>Promoter hypermethylation of cyclooxygenase-2 gene in esophageal squamous cell carcinoma</article-title><source>Diseases of the Esophagus: Official Journal of the International Society for Diseases of the Esophagus/ISDE</source><volume>24</volume><fpage>444</fpage><lpage>449</lpage><pub-id pub-id-type="doi">10.1111/j.1442-2050.2010.01159.x</pub-id></element-citation></ref><ref id="bib39"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Mercer</surname><given-names>TR</given-names></name><name><surname>Dinger</surname><given-names>ME</given-names></name><name><surname>Mattick</surname><given-names>JS</given-names></name></person-group><year>2009</year><article-title>Long non-coding RNAs: insights into functions</article-title><source>Nature Reviews Genetics</source><volume>10</volume><fpage>155</fpage><lpage>159</lpage><pub-id pub-id-type="doi">10.1038/nrg2521</pub-id></element-citation></ref><ref id="bib38"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Mercer</surname><given-names>TR</given-names></name><name><surname>Dinger</surname><given-names>ME</given-names></name><name><surname>Sunkin</surname><given-names>SM</given-names></name><name><surname>Mehler</surname><given-names>MF</given-names></name><name><surname>Mattick</surname><given-names>JS</given-names></name></person-group><year>2008</year><article-title>Specific expression of long noncoding RNAs in the mouse brain</article-title><source>Proceedings of the National Academy of Sciences of the United States of America</source><volume>105</volume><fpage>716</fpage><lpage>721</lpage><pub-id pub-id-type="doi">10.1073/pnas.0706729105</pub-id></element-citation></ref><ref id="bib40"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Merkenschlager</surname><given-names>M</given-names></name><name><surname>Odom</surname><given-names>DT</given-names></name></person-group><year>2013</year><article-title>CTCF and cohesin: linking gene regulatory elements with their targets</article-title><source>Cell</source><volume>152</volume><fpage>1285</fpage><lpage>1297</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2013.02.029</pub-id></element-citation></ref><ref id="bib41"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Minn</surname><given-names>AJ</given-names></name><name><surname>Gupta</surname><given-names>GP</given-names></name><name><surname>Siegel</surname><given-names>PM</given-names></name><name><surname>Bos</surname><given-names>PD</given-names></name><name><surname>Shu</surname><given-names>W</given-names></name><name><surname>Giri</surname><given-names>DD</given-names></name><name><surname>Viale</surname><given-names>A</given-names></name><name><surname>Olshen</surname><given-names>AB</given-names></name><name><surname>Gerald</surname><given-names>WL</given-names></name><name><surname>Massagué</surname><given-names>J</given-names></name></person-group><year>2005</year><article-title>Genes that mediate breast cancer metastasis to lung</article-title><source>Nature</source><volume>436</volume><fpage>518</fpage><lpage>524</lpage><pub-id pub-id-type="doi">10.1038/nature03799</pub-id></element-citation></ref><ref id="bib42"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Mourtada-Maarabouni</surname><given-names>M</given-names></name><name><surname>Pickard</surname><given-names>MR</given-names></name><name><surname>Hedge</surname><given-names>VL</given-names></name><name><surname>Farzaneh</surname><given-names>F</given-names></name><name><surname>Williams</surname><given-names>GT</given-names></name></person-group><year>2009</year><article-title>GAS5, a non-protein-coding RNA, controls apoptosis and is downregulated in breast cancer</article-title><source>Oncogene</source><volume>28</volume><fpage>195</fpage><lpage>208</lpage><pub-id pub-id-type="doi">10.1038/onc.2008.373</pub-id></element-citation></ref><ref id="bib43"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Murata</surname><given-names>H</given-names></name><name><surname>Tsuji</surname><given-names>S</given-names></name><name><surname>Tsujii</surname><given-names>M</given-names></name><name><surname>Sakaguchi</surname><given-names>Y</given-names></name><name><surname>Fu</surname><given-names>HY</given-names></name><name><surname>Kawano</surname><given-names>S</given-names></name><name><surname>Hori</surname><given-names>M</given-names></name></person-group><year>2004</year><article-title>Promoter hypermethylation silences cyclooxygenase-2 (Cox-2) and regulates growth of human hepatocellular carcinoma cells</article-title><source>Laboratory Investigation; a Journal of Technical Methods and Pathology</source><volume>84</volume><fpage>1050</fpage><lpage>1059</lpage><pub-id pub-id-type="doi">10.1038/labinvest.3700118</pub-id></element-citation></ref><ref id="bib44"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nagano</surname><given-names>T</given-names></name><name><surname>Mitchell</surname><given-names>JA</given-names></name><name><surname>Sanz</surname><given-names>LA</given-names></name><name><surname>Pauler</surname><given-names>FM</given-names></name><name><surname>Ferguson-Smith</surname><given-names>AC</given-names></name><name><surname>Feil</surname><given-names>R</given-names></name><name><surname>Fraser</surname><given-names>P</given-names></name></person-group><year>2008</year><article-title>The Air noncoding RNA epigenetically silences transcription by targeting G9a to chromatin</article-title><source>Science</source><volume>322</volume><fpage>1717</fpage><lpage>1720</lpage><pub-id pub-id-type="doi">10.1126/science.1163802</pub-id></element-citation></ref><ref id="bib45"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ohlsson</surname><given-names>R</given-names></name><name><surname>Bartkuhn</surname><given-names>M</given-names></name><name><surname>Renkawitz</surname><given-names>R</given-names></name></person-group><year>2010</year><article-title>CTCF shapes chromatin by multiple mechanisms: the impact of 20 years of CTCF research on understanding the workings of chromatin</article-title><source>Chromosoma</source><volume>119</volume><fpage>351</fpage><lpage>360</lpage><pub-id pub-id-type="doi">10.1007/s00412-010-0262-0</pub-id></element-citation></ref><ref id="bib46"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Oshima</surname><given-names>M</given-names></name><name><surname>Dinchuk</surname><given-names>JE</given-names></name><name><surname>Kargman</surname><given-names>SL</given-names></name><name><surname>Oshima</surname><given-names>H</given-names></name><name><surname>Hancock</surname><given-names>B</given-names></name><name><surname>Kwong</surname><given-names>E</given-names></name><name><surname>Trzaskos</surname><given-names>JM</given-names></name><name><surname>Evans</surname><given-names>JF</given-names></name><name><surname>Taketo</surname><given-names>MM</given-names></name></person-group><year>1996</year><article-title>Suppression of intestinal polyposis in Apc delta716 knockout mice by inhibition of cyclooxygenase 2 (COX-2)</article-title><source>Cell</source><volume>87</volume><fpage>803</fpage><lpage>809</lpage><pub-id pub-id-type="doi">10.1016/S0092-8674(00)81988-1</pub-id></element-citation></ref><ref id="bib47"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Parelho</surname><given-names>V</given-names></name><name><surname>Hadjur</surname><given-names>S</given-names></name><name><surname>Spivakov</surname><given-names>M</given-names></name><name><surname>Leleu</surname><given-names>M</given-names></name><name><surname>Sauer</surname><given-names>S</given-names></name><name><surname>Gregson</surname><given-names>HC</given-names></name><name><surname>Jarmuz</surname><given-names>A</given-names></name><name><surname>Canzonetta</surname><given-names>C</given-names></name><name><surname>Webster</surname><given-names>Z</given-names></name><name><surname>Nesterova</surname><given-names>T</given-names></name><name><surname>Cobb</surname><given-names>BS</given-names></name><name><surname>Yokomori</surname><given-names>K</given-names></name><name><surname>Dillon</surname><given-names>N</given-names></name><name><surname>Aragon</surname><given-names>L</given-names></name><name><surname>Fisher</surname><given-names>AG</given-names></name><name><surname>Merkenschlager</surname><given-names>M</given-names></name></person-group><year>2008</year><article-title>Cohesins functionally associate with CTCF on mammalian chromosome arms</article-title><source>Cell</source><volume>132</volume><fpage>422</fpage><lpage>433</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2008.01.011</pub-id></element-citation></ref><ref id="bib48"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Phillips</surname><given-names>JE</given-names></name><name><surname>Corces</surname><given-names>VG</given-names></name></person-group><year>2009</year><article-title>CTCF: master weaver of the genome</article-title><source>Cell</source><volume>137</volume><fpage>1194</fpage><lpage>1211</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2009.06.001</pub-id></element-citation></ref><ref id="bib49"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pollard</surname><given-names>JW</given-names></name></person-group><year>2004</year><article-title>Tumour-educated macrophages promote tumour progression and metastasis</article-title><source>Nature Reviews Cancer</source><volume>4</volume><fpage>71</fpage><lpage>78</lpage><pub-id pub-id-type="doi">10.1038/nrc1256</pub-id></element-citation></ref><ref id="bib50"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ponting</surname><given-names>CP</given-names></name><name><surname>Oliver</surname><given-names>PL</given-names></name><name><surname>Reik</surname><given-names>W</given-names></name></person-group><year>2009</year><article-title>Evolution and functions of long noncoding RNAs</article-title><source>Cell</source><volume>136</volume><fpage>629</fpage><lpage>641</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2009.02.006</pub-id></element-citation></ref><ref id="bib51"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rada-Iglesias</surname><given-names>A</given-names></name><name><surname>Bajpai</surname><given-names>R</given-names></name><name><surname>Swigut</surname><given-names>T</given-names></name><name><surname>Brugmann</surname><given-names>SA</given-names></name><name><surname>Flynn</surname><given-names>RA</given-names></name><name><surname>Wysocka</surname><given-names>J</given-names></name></person-group><year>2011</year><article-title>A unique chromatin signature uncovers early developmental enhancers in humans</article-title><source>Nature</source><volume>470</volume><fpage>279</fpage><lpage>283</lpage><pub-id pub-id-type="doi">10.1038/nature09692</pub-id></element-citation></ref><ref id="bib52"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ramsay</surname><given-names>RG</given-names></name><name><surname>Ciznadija</surname><given-names>D</given-names></name><name><surname>Vanevski</surname><given-names>M</given-names></name><name><surname>Mantamadiotis</surname><given-names>T</given-names></name></person-group><year>2003</year><article-title>Transcriptional regulation of cyclo-oxygenase expression: three pillars of control</article-title><source>International Journal of Immunopathology and Pharmacology</source><volume>16</volume><fpage>59</fpage><lpage>67</lpage></element-citation></ref><ref id="bib53"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rodriguez-Barbero</surname><given-names>A</given-names></name><name><surname>Dorado</surname><given-names>F</given-names></name><name><surname>Velasco</surname><given-names>S</given-names></name><name><surname>Pandiella</surname><given-names>A</given-names></name><name><surname>Banas</surname><given-names>B</given-names></name><name><surname>López-Novoa</surname><given-names>JM</given-names></name></person-group><year>2006</year><article-title>TGF-beta1 induces COX-2 expression and PGE2 synthesis through MAPK and PI3K pathways in human mesangial cells</article-title><source>Kidney International</source><volume>70</volume><fpage>901</fpage><lpage>909</lpage><pub-id pub-id-type="doi">10.1038/sj.ki.5001626</pub-id></element-citation></ref><ref id="bib54"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rubio</surname><given-names>ED</given-names></name><name><surname>Reiss</surname><given-names>DJ</given-names></name><name><surname>Welcsh</surname><given-names>PL</given-names></name><name><surname>Disteche</surname><given-names>CM</given-names></name><name><surname>Filippova</surname><given-names>GN</given-names></name><name><surname>Baliga</surname><given-names>NS</given-names></name><name><surname>Aebersold</surname><given-names>R</given-names></name><name><surname>Ranish</surname><given-names>JA</given-names></name><name><surname>Krumm</surname><given-names>A</given-names></name></person-group><year>2008</year><article-title>CTCF physically links cohesin to chromatin</article-title><source>Proceedings of the National Academy of Sciences of the United States of America</source><volume>105</volume><fpage>8309</fpage><lpage>8314</lpage><pub-id pub-id-type="doi">10.1073/pnas.0801273105</pub-id></element-citation></ref><ref id="bib56"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Saccani</surname><given-names>A</given-names></name><name><surname>Schioppa</surname><given-names>T</given-names></name><name><surname>Porta</surname><given-names>C</given-names></name><name><surname>Biswas</surname><given-names>SK</given-names></name><name><surname>Nebuloni</surname><given-names>M</given-names></name><name><surname>Vago</surname><given-names>L</given-names></name><name><surname>Bottazzi</surname><given-names>B</given-names></name><name><surname>Colombo</surname><given-names>MP</given-names></name><name><surname>Mantovani</surname><given-names>A</given-names></name><name><surname>Sica</surname><given-names>A</given-names></name></person-group><year>2006</year><article-title>p50 nuclear factor-kappaB overexpression in tumor-associated macrophages inhibits M1 inflammatory responses and antitumor resistance</article-title><source>Cancer Research</source><volume>66</volume><fpage>11432</fpage><lpage>11440</lpage><pub-id pub-id-type="doi">10.1158/0008-5472.CAN-06-1867</pub-id></element-citation></ref><ref id="bib55"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Saccani</surname><given-names>S</given-names></name><name><surname>Pantano</surname><given-names>S</given-names></name><name><surname>Natoli</surname><given-names>G</given-names></name></person-group><year>2003</year><article-title>Modulation of NF-kappaB activity by exchange of dimers</article-title><source>Molecular Cell</source><volume>11</volume><fpage>1563</fpage><lpage>1574</lpage><pub-id pub-id-type="doi">10.1016/S1097-2765(03)00227-2</pub-id></element-citation></ref><ref id="bib57"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Salinas</surname><given-names>G</given-names></name><name><surname>Rangasetty</surname><given-names>UC</given-names></name><name><surname>Uretsky</surname><given-names>BF</given-names></name><name><surname>Birnbaum</surname><given-names>Y</given-names></name></person-group><year>2007</year><article-title>The cycloxygenase 2 (COX-2) story: it's time to explain, not inflame</article-title><source>Journal of Cardiovascular Pharmacology and Therapeutics</source><volume>12</volume><fpage>98</fpage><lpage>111</lpage><pub-id pub-id-type="doi">10.1177/1074248407301172</pub-id></element-citation></ref><ref id="bib58"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Schneider</surname><given-names>C</given-names></name><name><surname>Pozzi</surname><given-names>A</given-names></name></person-group><year>2011</year><article-title>Cyclooxygenases and lipoxygenases in cancer</article-title><source>Cancer Metastasis Reviews</source><volume>30</volume><fpage>277</fpage><lpage>294</lpage><pub-id pub-id-type="doi">10.1007/s10555-011-9310-3</pub-id></element-citation></ref><ref id="bib59"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Smith</surname><given-names>WL</given-names></name><name><surname>DeWitt</surname><given-names>DL</given-names></name><name><surname>Garavito</surname><given-names>RM</given-names></name></person-group><year>2000</year><article-title>Cyclooxygenases: structural, cellular, and molecular biology</article-title><source>Annual Review of Biochemistry</source><volume>69</volume><fpage>145</fpage><lpage>182</lpage><pub-id pub-id-type="doi">10.1146/annurev.biochem.69.1.145</pub-id></element-citation></ref><ref id="bib60"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sopher</surname><given-names>BL</given-names></name><name><surname>Ladd</surname><given-names>PD</given-names></name><name><surname>Pineda</surname><given-names>VV</given-names></name><name><surname>Libby</surname><given-names>RT</given-names></name><name><surname>Sunkin</surname><given-names>SM</given-names></name><name><surname>Hurley</surname><given-names>JB</given-names></name><name><surname>Thienes</surname><given-names>CP</given-names></name><name><surname>Gaasterland</surname><given-names>T</given-names></name><name><surname>Filippova</surname><given-names>GN</given-names></name><name><surname>La Spada</surname><given-names>AR</given-names></name></person-group><year>2011</year><article-title>CTCF regulates ataxin-7 expression through promotion of a convergently transcribed, antisense noncoding RNA</article-title><source>Neuron</source><volume>70</volume><fpage>1071</fpage><lpage>1084</lpage><pub-id pub-id-type="doi">10.1016/j.neuron.2011.05.027</pub-id></element-citation></ref><ref id="bib61"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Spencer</surname><given-names>RJ</given-names></name><name><surname>del Rosario</surname><given-names>BC</given-names></name><name><surname>Pinter</surname><given-names>SF</given-names></name><name><surname>Lessing</surname><given-names>D</given-names></name><name><surname>Sadreyev</surname><given-names>RI</given-names></name><name><surname>Lee</surname><given-names>JT</given-names></name></person-group><year>2011</year><article-title>A boundary element between Tsix and Xist binds the chromatin insulator Ctcf and contributes to initiation of X-chromosome inactivation</article-title><source>Genetics</source><volume>189</volume><fpage>441</fpage><lpage>454</lpage><pub-id pub-id-type="doi">10.1534/genetics.111.132662</pub-id></element-citation></ref><ref id="bib62"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Stedman</surname><given-names>W</given-names></name><name><surname>Kang</surname><given-names>H</given-names></name><name><surname>Lin</surname><given-names>S</given-names></name><name><surname>Kissil</surname><given-names>JL</given-names></name><name><surname>Bartolomei</surname><given-names>MS</given-names></name><name><surname>Lieberman</surname><given-names>PM</given-names></name></person-group><year>2008</year><article-title>Cohesins localize with CTCF at the KSHV latency control region and at cellular c-myc and H19/Igf2 insulators</article-title><source>The Embo Journal</source><volume>27</volume><fpage>654</fpage><lpage>666</lpage><pub-id pub-id-type="doi">10.1038/emboj.2008.1</pub-id></element-citation></ref><ref id="bib63"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sundstrom</surname><given-names>C</given-names></name><name><surname>Nilsson</surname><given-names>K</given-names></name></person-group><year>1976</year><article-title>Establishment and characterization of a human histiocytic lymphoma cell line (U-937)</article-title><source>International Journal of Cancer Journal International Du Cancer</source><volume>17</volume><fpage>565</fpage><lpage>577</lpage><pub-id pub-id-type="doi">10.1002/ijc.2910170504</pub-id></element-citation></ref><ref id="bib64"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Toyota</surname><given-names>M</given-names></name><name><surname>Shen</surname><given-names>L</given-names></name><name><surname>Ohe-Toyota</surname><given-names>M</given-names></name><name><surname>Hamilton</surname><given-names>SR</given-names></name><name><surname>Sinicrope</surname><given-names>FA</given-names></name><name><surname>Issa</surname><given-names>JP</given-names></name></person-group><year>2000</year><article-title>Aberrant methylation of the Cyclooxygenase 2 CpG island in colorectal tumors</article-title><source>Cancer Research</source><volume>60</volume><fpage>4044</fpage><lpage>4048</lpage></element-citation></ref><ref id="bib65"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Tsai</surname><given-names>MC</given-names></name><name><surname>Manor</surname><given-names>O</given-names></name><name><surname>Wan</surname><given-names>Y</given-names></name><name><surname>Mosammaparast</surname><given-names>N</given-names></name><name><surname>Wang</surname><given-names>JK</given-names></name><name><surname>Lan</surname><given-names>F</given-names></name><name><surname>Shi</surname><given-names>Y</given-names></name><name><surname>Segal</surname><given-names>E</given-names></name><name><surname>Chang</surname><given-names>HY</given-names></name></person-group><year>2010</year><article-title>Long noncoding RNA as modular scaffold of histone modification complexes</article-title><source>Science</source><volume>329</volume><fpage>689</fpage><lpage>693</lpage><pub-id pub-id-type="doi">10.1126/science.1192002</pub-id></element-citation></ref><ref id="bib66"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Tsai</surname><given-names>MC</given-names></name><name><surname>Spitale</surname><given-names>RC</given-names></name><name><surname>Chang</surname><given-names>HY</given-names></name></person-group><year>2011</year><article-title>Long intergenic noncoding RNAs: new links in cancer progression</article-title><source>Cancer Research</source><volume>71</volume><fpage>3</fpage><lpage>7</lpage><pub-id pub-id-type="doi">10.1158/0008-5472.CAN-10-2483</pub-id></element-citation></ref><ref id="bib67"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Visel</surname><given-names>A</given-names></name><name><surname>Blow</surname><given-names>MJ</given-names></name><name><surname>Li</surname><given-names>Z</given-names></name><name><surname>Zhang</surname><given-names>T</given-names></name><name><surname>Akiyama</surname><given-names>JA</given-names></name><name><surname>Holt</surname><given-names>A</given-names></name><name><surname>Plajzer-Frick</surname><given-names>I</given-names></name><name><surname>Shoukry</surname><given-names>M</given-names></name><name><surname>Wright</surname><given-names>C</given-names></name><name><surname>Chen</surname><given-names>F</given-names></name><name><surname>Afzal</surname><given-names>V</given-names></name><name><surname>Ren</surname><given-names>B</given-names></name><name><surname>Rubin</surname><given-names>EM</given-names></name><name><surname>Pennacchio</surname><given-names>LA</given-names></name></person-group><year>2009</year><article-title>ChIP-seq accurately predicts tissue-specific activity of enhancers</article-title><source>Nature</source><volume>457</volume><fpage>854</fpage><lpage>858</lpage><pub-id pub-id-type="doi">10.1038/nature07730</pub-id></element-citation></ref><ref id="bib68"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wada</surname><given-names>Y</given-names></name><name><surname>Ohta</surname><given-names>Y</given-names></name><name><surname>Xu</surname><given-names>M</given-names></name><name><surname>Tsutsumi</surname><given-names>S</given-names></name><name><surname>Minami</surname><given-names>T</given-names></name><name><surname>Inoue</surname><given-names>K</given-names></name><name><surname>Komura</surname><given-names>D</given-names></name><name><surname>Kitakami</surname><given-names>J</given-names></name><name><surname>Oshida</surname><given-names>N</given-names></name><name><surname>Papantonis</surname><given-names>A</given-names></name><name><surname>Izumi</surname><given-names>A</given-names></name><name><surname>Kobayashi</surname><given-names>M</given-names></name><name><surname>Meguro</surname><given-names>H</given-names></name><name><surname>Kanki</surname><given-names>Y</given-names></name><name><surname>Mimura</surname><given-names>I</given-names></name><name><surname>Yamamoto</surname><given-names>K</given-names></name><name><surname>Mataki</surname><given-names>C</given-names></name><name><surname>Hamakubo</surname><given-names>T</given-names></name><name><surname>Shirahige</surname><given-names>K</given-names></name><name><surname>Aburatani</surname><given-names>H</given-names></name><name><surname>Kimura</surname><given-names>H</given-names></name><name><surname>Kodama</surname><given-names>T</given-names></name><name><surname>Cook</surname><given-names>PR</given-names></name><name><surname>Ihara</surname><given-names>S</given-names></name></person-group><year>2009</year><article-title>A wave of nascent transcription on activated human genes</article-title><source>Proceedings of the National Academy of Sciences of the United States of America</source><volume>106</volume><fpage>18357</fpage><lpage>18361</lpage><pub-id pub-id-type="doi">10.1073/pnas.0902573106</pub-id></element-citation></ref><ref id="bib69"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>KC</given-names></name><name><surname>Chang</surname><given-names>HY</given-names></name></person-group><year>2011</year><article-title>Molecular mechanisms of long noncoding RNAs</article-title><source>Molecular Cell</source><volume>43</volume><fpage>904</fpage><lpage>914</lpage><pub-id pub-id-type="doi">10.1016/j.molcel.2011.08.018</pub-id></element-citation></ref><ref id="bib70"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>HQ</given-names></name><name><surname>Kim</surname><given-names>MP</given-names></name><name><surname>Tiano</surname><given-names>HF</given-names></name><name><surname>Langenbach</surname><given-names>R</given-names></name><name><surname>Smart</surname><given-names>RC</given-names></name></person-group><year>2001</year><article-title>Protein kinase C-alpha coordinately regulates cytosolic phospholipase A(2) activity and the expression of cyclooxygenase-2 through different mechanisms in mouse keratinocytes</article-title><source>Molecular Pharmacology</source><volume>59</volume><fpage>860</fpage><lpage>866</lpage></element-citation></ref><ref id="bib71"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>KC</given-names></name><name><surname>Yang</surname><given-names>YW</given-names></name><name><surname>Liu</surname><given-names>B</given-names></name><name><surname>Sanyal</surname><given-names>A</given-names></name><name><surname>Corces-Zimmerman</surname><given-names>R</given-names></name><name><surname>Chen</surname><given-names>Y</given-names></name><name><surname>Lajoie</surname><given-names>BR</given-names></name><name><surname>Protacio</surname><given-names>A</given-names></name><name><surname>Flynn</surname><given-names>RA</given-names></name><name><surname>Gupta</surname><given-names>RA</given-names></name><name><surname>Wysocka</surname><given-names>J</given-names></name><name><surname>Lei</surname><given-names>M</given-names></name><name><surname>Dekker</surname><given-names>J</given-names></name><name><surname>Helms</surname><given-names>JA</given-names></name><name><surname>Chang</surname><given-names>HY</given-names></name></person-group><year>2011</year><article-title>A long noncoding RNA maintains active chromatin to coordinate homeotic gene expression</article-title><source>Nature</source><volume>472</volume><fpage>120</fpage><lpage>124</lpage><pub-id pub-id-type="doi">10.1038/nature09819</pub-id></element-citation></ref><ref id="bib72"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wapinski</surname><given-names>O</given-names></name><name><surname>Chang</surname><given-names>HY</given-names></name></person-group><year>2011</year><article-title>Long noncoding RNAs and human disease</article-title><source>Trends in Cell Biology</source><volume>21</volume><fpage>354</fpage><lpage>361</lpage><pub-id pub-id-type="doi">10.1016/j.tcb.2011.04.001</pub-id></element-citation></ref><ref id="bib73"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wendt</surname><given-names>KS</given-names></name><name><surname>Yoshida</surname><given-names>K</given-names></name><name><surname>Itoh</surname><given-names>T</given-names></name><name><surname>Bando</surname><given-names>M</given-names></name><name><surname>Koch</surname><given-names>B</given-names></name><name><surname>Schirghuber</surname><given-names>E</given-names></name><name><surname>Tsutsumi</surname><given-names>S</given-names></name><name><surname>Nagae</surname><given-names>G</given-names></name><name><surname>Ishihara</surname><given-names>K</given-names></name><name><surname>Mishiro</surname><given-names>T</given-names></name><name><surname>Yahata</surname><given-names>K</given-names></name><name><surname>Imamoto</surname><given-names>F</given-names></name><name><surname>Aburatani</surname><given-names>H</given-names></name><name><surname>Nakao</surname><given-names>M</given-names></name><name><surname>Imamoto</surname><given-names>N</given-names></name><name><surname>Maeshima</surname><given-names>K</given-names></name><name><surname>Shirahige</surname><given-names>K</given-names></name><name><surname>Peters</surname><given-names>JM</given-names></name></person-group><year>2008</year><article-title>Cohesin mediates transcriptional insulation by CCCTC-binding factor</article-title><source>Nature</source><volume>451</volume><fpage>796</fpage><lpage>801</lpage><pub-id pub-id-type="doi">10.1038/nature06634</pub-id></element-citation></ref><ref id="bib74"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wilusz</surname><given-names>JE</given-names></name><name><surname>Sunwoo</surname><given-names>H</given-names></name><name><surname>Spector</surname><given-names>DL</given-names></name></person-group><year>2009</year><article-title>Long noncoding RNAs: functional surprises from the RNA world</article-title><source>Genes & Development</source><volume>23</volume><fpage>1494</fpage><lpage>1504</lpage><pub-id pub-id-type="doi">10.1101/gad.1800909</pub-id></element-citation></ref><ref id="bib75"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Witcher</surname><given-names>M</given-names></name><name><surname>Emerson</surname><given-names>BM</given-names></name></person-group><year>2009</year><article-title>Epigenetic silencing of the p16(INK4a) tumor suppressor is associated with loss of CTCF binding and a chromatin boundary</article-title><source>Molecular Cell</source><volume>34</volume><fpage>271</fpage><lpage>284</lpage><pub-id pub-id-type="doi">10.1016/j.molcel.2009.04.001</pub-id></element-citation></ref><ref id="bib76"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wutz</surname><given-names>A</given-names></name><name><surname>Rasmussen</surname><given-names>TP</given-names></name><name><surname>Jaenisch</surname><given-names>R</given-names></name></person-group><year>2002</year><article-title>Chromosomal silencing and localization are mediated by different domains of Xist RNA</article-title><source>Nature Genetics</source><volume>30</volume><fpage>167</fpage><lpage>174</lpage><pub-id pub-id-type="doi">10.1038/ng820</pub-id></element-citation></ref><ref id="bib77"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xiao</surname><given-names>X</given-names></name><name><surname>Shi</surname><given-names>D</given-names></name><name><surname>Liu</surname><given-names>L</given-names></name><name><surname>Wang</surname><given-names>J</given-names></name><name><surname>Xie</surname><given-names>X</given-names></name><name><surname>Kang</surname><given-names>T</given-names></name><name><surname>Deng</surname><given-names>W</given-names></name></person-group><year>2011</year><article-title>Quercetin suppresses cyclooxygenase-2 expression and angiogenesis through inactivation of P300 signaling</article-title><source>PLOS ONE</source><volume>6</volume><fpage>e22934</fpage><pub-id pub-id-type="doi">10.1371/journal.pone.0022934</pub-id></element-citation></ref><ref id="bib78"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xie</surname><given-names>W</given-names></name><name><surname>Herschman</surname><given-names>HR</given-names></name></person-group><year>1996</year><article-title>Transcriptional regulation of prostaglandin synthase 2 gene expression by platelet-derived growth factor and serum</article-title><source>The Journal of Biological Chemistry</source><volume>271</volume><fpage>31742</fpage><lpage>31748</lpage><pub-id pub-id-type="doi">10.1074/jbc.271.49.31742</pub-id></element-citation></ref><ref id="bib79"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xie</surname><given-names>X</given-names></name><name><surname>Mikkelsen</surname><given-names>TS</given-names></name><name><surname>Gnirke</surname><given-names>A</given-names></name><name><surname>Lindblad-Toh</surname><given-names>K</given-names></name><name><surname>Kellis</surname><given-names>M</given-names></name><name><surname>Lander</surname><given-names>ES</given-names></name></person-group><year>2007</year><article-title>Systematic discovery of regulatory motifs in conserved regions of the human genome, including thousands of CTCF insulator sites</article-title><source>Proceedings of the National Academy of Sciences of the United States of America</source><volume>104</volume><fpage>7145</fpage><lpage>7150</lpage><pub-id pub-id-type="doi">10.1073/pnas.0701811104</pub-id></element-citation></ref><ref id="bib80"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xu</surname><given-names>Q</given-names></name><name><surname>Ji</surname><given-names>YS</given-names></name><name><surname>Schmedtje</surname><given-names>JF</given-names><suffix>Jnr</suffix></name></person-group><year>2000</year><article-title>Sp1 increases expression of cyclooxygenase-2 in hypoxic vascular endothelium. Implications for the mechanisms of aortic aneurysm and heart failure</article-title><source>The Journal of Biological Chemistry</source><volume>275</volume><fpage>24583</fpage><lpage>24589</lpage><pub-id pub-id-type="doi">10.1074/jbc.M003894200</pub-id></element-citation></ref><ref id="bib81"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Yap</surname><given-names>KL</given-names></name><name><surname>Li</surname><given-names>S</given-names></name><name><surname>Muñoz-Cabello</surname><given-names>AM</given-names></name><name><surname>Raguz</surname><given-names>S</given-names></name><name><surname>Zeng</surname><given-names>L</given-names></name><name><surname>Mujtaba</surname><given-names>S</given-names></name><name><surname>Gil</surname><given-names>J</given-names></name><name><surname>Walsh</surname><given-names>MJ</given-names></name><name><surname>Zhou</surname><given-names>MM</given-names></name></person-group><year>2010</year><article-title>Molecular interplay of the noncoding RNA ANRIL and methylated histone H3 lysine 27 by polycomb CBX7 in transcriptional silencing of INK4a</article-title><source>Molecular Cell</source><volume>38</volume><fpage>662</fpage><lpage>674</lpage><pub-id pub-id-type="doi">10.1016/j.molcel.2010.03.021</pub-id></element-citation></ref><ref id="bib82"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhai</surname><given-names>B</given-names></name><name><surname>Yang</surname><given-names>H</given-names></name><name><surname>Mancini</surname><given-names>A</given-names></name><name><surname>He</surname><given-names>Q</given-names></name><name><surname>Antoniou</surname><given-names>J</given-names></name><name><surname>Di Battista</surname><given-names>JA</given-names></name></person-group><year>2010</year><article-title>Leukotriene B(4) BLT receptor signaling regulates the level and stability of cyclooxygenase-2 (COX-2) mRNA through restricted activation of Ras/Raf/ERK/p42 AUF1 pathway</article-title><source>The Journal of Biological Chemistry</source><volume>285</volume><fpage>23568</fpage><lpage>23580</lpage><pub-id pub-id-type="doi">10.1074/jbc.M110.107623</pub-id></element-citation></ref><ref id="bib83"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhao</surname><given-names>J</given-names></name><name><surname>Ohsumi</surname><given-names>TK</given-names></name><name><surname>Kung</surname><given-names>JT</given-names></name><name><surname>Ogawa</surname><given-names>Y</given-names></name><name><surname>Grau</surname><given-names>DJ</given-names></name><name><surname>Sarma</surname><given-names>K</given-names></name><name><surname>Song</surname><given-names>JJ</given-names></name><name><surname>Kingston</surname><given-names>RE</given-names></name><name><surname>Borowsky</surname><given-names>M</given-names></name><name><surname>Lee</surname><given-names>JT</given-names></name></person-group><year>2010</year><article-title>Genome-wide identification of polycomb-associated RNAs by RIP-seq</article-title><source>Molecular Cell</source><volume>40</volume><fpage>939</fpage><lpage>953</lpage><pub-id pub-id-type="doi">10.1016/j.molcel.2010.12.011</pub-id></element-citation></ref><ref id="bib84"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhong</surname><given-names>XP</given-names></name><name><surname>Krangel</surname><given-names>MS</given-names></name></person-group><year>1999</year><article-title>Enhancer-blocking activity within the DNase I hypersensitive site 2 to 6 region between the TCR alpha and Dad1 genes</article-title><source>The Journal of Immunology: official Journal of the American Association of Immunologists</source><volume>163</volume><fpage>295</fpage><lpage>300</lpage></element-citation></ref></ref-list></back><sub-article article-type="article-commentary" id="SA1"><front-stub><article-id pub-id-type="doi">10.7554/eLife.01776.021</article-id><title-group><article-title>Decision letter</article-title></title-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Espinosa</surname><given-names>Joaquin M</given-names></name><role>Reviewing editor</role><aff><institution>HHMI/University of Colorado at Boulder</institution>, <country>United States</country></aff></contrib></contrib-group></front-stub><body><boxed-text><p>eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see <ext-link ext-link-type="uri" xlink:href="http://elifesciences.org/review-process">review process</ext-link>). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.</p></boxed-text><p>Thank you for sending your work entitled “p50-Associated COX2 Extragenic RNA (PACER) activates <italic>COX2</italic> gene expression by sequestering repressive NFκB complexes” for consideration at <italic>eLife</italic>. Your article has been favorably evaluated by a Senior editor and 3 reviewers, one of whom is a member of our Board of Reviewing Editors.</p><p>The Reviewing editor and the other reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission:</p><p>The three reviewers agreed that the paper is exciting and interesting, as it describes a novel lncRNA involved in control of a gene locus of significant biomedical relevance by an unprecedented mechanism. The manuscript reports the characterization of the lncRNA dubbed PACER, which originates from the upstream region of the COX-2 gene in the antisense direction. It is demonstrated that PACER positively regulates COX-2 expression. Mechanistically, PACER promotes histone modifications associated with gene activation and RNAPII association with the COX2 promoter region. PACER binds to the DNA binding protein p50, and it is proposed that this binding leads to gene activation by favoring formation of active p50-p65 NFKB heterodimers, which in turn would drive COX2 transcription. The reviewers agreed that these results merit publication in a high-impact journal such as <italic>eLife</italic>, but they also concurred in that additional work is required to strengthen the main conclusions of this interesting paper.</p><p>The reviewers agreed to summarize their request for new work as follows:</p><p>1) A better characterization of PACER is needed. Key questions raised by the reviewers are: What is the size, exact sequence, relative abundance and exact intracellular distribution of PACER? Does PACER originate from its own unique promoter, is it a divergent transcript from the COX-2 promoter, or is it an eRNA? To address these issues, the reviewers propose the following work:</p><p>1.1) Full sequence data from the RACE experiments should be provided.</p><p>1.2) A Northern blot analysis and/or qPCR analysis should be performed to determine the abundance of PACER relative to COX-2 mRNA.</p><p>1.3) A better characterization of the PACER transcription start site is needed. ChIP analysis of RNAPII and chromatin modifications are provided in the manuscript, but these assays are correlative in nature and could not fully elucidate whether PACER originates from its own unique promoter, if it is a divergent transcript from the COX-2 promoter, or if it is an eRNA. This issue can be resolved with promoter reporter assays to define the DNA sequence that drives PACER transcription.</p><p>1.4) Both the RNA and RNAPII ChIP profile suggest a longer transcript than what was cloned by RACE, up to -2 kb. The authors should explain how this matches the suggested 3' end of the RNA at -1022 and show the RACE data rather than just describing it. Is there a canonical cleavage and polyadenylation signal around -1022?</p><p>1.5) Strand-specific amplification of the plus strand should include an oligo within the COX-2 gene as a positive control.</p><p>1.6) <xref ref-type="fig" rid="fig4">Figure 4J</xref> shows the subcellular localization of PACER and COX-2 as determined by cellular fractionation. The authors should perform controls of transcripts present only in the nucleus and only in the cytoplasm to account for cross-contamination of the fractions.</p><p>1.7) The reviewers wonder whether PACER has been detected by recent genome wide Global Run-On deep sequencing experiments (GRO-seq) done in epithelial cells. Publicly available GRO-seq datasets that could be mined to answer this question include Galbraith MD et al, Cell. 2013 Jun 6;153(6):1327-39 (performed in HCT116 colon cancer cells) and the various papers from the Kraus team using MCF7 breast cancer cells such as Danko CG Mol Cell. 2013 Apr 25;50(2):212-22. GRO-seq results for the COX-2 (PTGS2) locus could be displayed to further illuminate the site of PACER transcription.</p><p>2) The model of PACER action needs further experimental validation.</p><p>2.1) Whereas the repressive role of p50 is well demonstrated, the positive role of p65 NFKB in driving COX2 expression is not demonstrated in the context of PACER action. As currently depicted in <xref ref-type="fig" rid="fig6">Figure 6G</xref>, it is p65 that recruits p300 to drive histone modifications and COX2 transcription. The reviewers request that the role of p65 on PACER action be demonstrated via knockdown.</p><p>2.2) Additionally, how does PACER knockdown or CTCF knockdown affect p65 binding to the COX2 promoter? Reviewers propose that ChIP assays for p65 be performed to answer this question.</p><p>3) A better presentation and interpretation of the ChIP data is needed.</p><p>3.1) CTCF ChIPs. Regarding the ChIP-on-chip experiment, the authors should show at least one negative control and one positive control locus out of the “numerous other genes” on the chip and extend the COX-2 analysis several kb up and downstream to demonstrate the specificity of the IP. There is a discrepancy between the ChIP-on-chip and ChIP-qPCR analysis for CTCF binding sites. In the ChIP-on-chip, one sees a clear enrichment of CTCF binding at least to -500bp, which is not confirmed by the qPCR. Also, the peak for CTCF binding around TSS is much more pronounced than the one around -2.5kb. Again, this is not recapitulated in the ChIP-qPCR. In addition, the claim that Rad21 binds around the TSS along with CTCF is not substantiated by the ChIP presented in <xref ref-type="fig" rid="fig2">Fig 2D</xref>, especially if one takes into account the IgG background that differs between different amplicons. The authors should aim to explain the possible origin of these discrepancies.</p><p>3.2) Do the two described peaks overlap with a CTCF consensus motif?</p><p>3.3) It is stated in the text that H4K20me3 is reduced upon CTCF knockdown; <xref ref-type="fig" rid="fig3">Figure 3I</xref>, however, shows an increase in H4K20me3. The text reads: “This modification was reduced at both sites upon CTCF knockdown...”, whereas the data shows a clear increase of H4K20me3, although only at the distal peak at -2.5 kb. Please revise the description of this result.</p><p>3.4) Chromatin modifications are evaluated upon PACER knockdown. Although the manuscript claims “no significant changes in H3K4me2”, there are clear differences in Figure S5 (not Figure S4 as stated in the text), especially at the important promoter and -2.5 kb sites. Please revise the description of this result.</p><p>3.5) The conclusion of PACER directly recruiting RNAP II is not supported by the data shown. Instead, the changes in chromatin modification might indirectly influence RNAP II association. The text should be revised to acknowledge this.</p></body></sub-article><sub-article article-type="reply" id="SA2"><front-stub><article-id pub-id-type="doi">10.7554/eLife.01776.022</article-id><title-group><article-title>Author response</article-title></title-group></front-stub><body><p><italic>1) A better characterization of PACER is needed. Key questions raised by the reviewers are: What is the size, exact sequence, relative abundance and exact intracellular distribution of PACER? Does PACER originate from its own unique promoter, is it a divergent transcript from the COX-2 promoter, or is it an eRNA? To address these issues, the reviewers</italic> <italic>propose the following work:</italic></p><p><italic>1.1) Full sequence data from the RACE experiments should be provided</italic>.</p><p>The full sequence of PACER, as well as sequencing data (raw sequence alignments and sequencing traces) for several independent 5’ and 3’ RACE reactions are now provided in <xref ref-type="fig" rid="fig1s1 fig1s2 fig1s3">Figure 1–figure supplements 1-3</xref>.</p><p><italic>1.2) A Northern blot analysis and/or qPCR analysis should be performed to determine the abundance of PACER relative to COX-2 mRNA</italic>.</p><p>We used qPCR to accurately measure relative amounts of PACER and COX-2 mRNA in HMEC and U937 macrophages. The results for HMEC are depicted in <xref ref-type="fig" rid="fig1">Figure 1H</xref>. The abundance of PACER RNA is roughly 30 fold lower than that of mRNA, both in HMEC and macrophages. Of note, we also attempted to use Northern blotting to detect PACER expression, however those experiments were inconclusive, as the signal for PACER was at or below the level of detection. We believe this is because PACER exists as a mixture of molecules differing at their 5’ ends, as shown by RACE mapping. Because such mixtures migrate as a ‘smear’ in gels, they are difficult to detect, especially if present at very low abundance.</p><p><italic>1.3) A better characterization of the PACER transcription start site is needed. ChIP analysis of RNAPII and chromatin modifications are provided in the manuscript, but these assays are correlative in nature and could not fully elucidate whether PACER originates from its own unique promoter, if it is a divergent transcript from the COX-2 promoter, or if it is an eRNA. This issue can be resolved with promoter reporter assays to define the DNA sequence that drives PACER transcription</italic>.</p><p>We now provide additional pieces of evidence and explanation demonstrating that PACER is a nuclear lncRNA synthesized from its own RNAPII dependent promoter. We performed reporter gene assays which indicate that a separate promoter located ∼0.3-0.5kb can drive expression of PACER independently of the COX-2 mRNA promoter (<xref ref-type="fig" rid="fig1">Figure 1I</xref>). In addition, RNAPII ChIP assays have been re-formatted to allow easier visualization of a separate peak of S5P-modified RNAP II at position ∼-0.4kb (<xref ref-type="fig" rid="fig1 fig5">Figures 1AB, 5D-F</xref>).</p><p><italic>1.4) Both the RNA and RNAPII ChIP profile suggest a longer transcript than what was cloned by RACE, up to -2 kb. The authors should explain how this matches the suggested 3' end of the RNA at -1022 and show the RACE data rather than just</italic> <italic>describing it. Is there a canonical cleavage and polyadenylation signal around -1022?</italic></p><p>Indeed, the COX-2 upstream region is transcriptionally active extending to around -2kb, as suggested by our RT-qPCR analyses, RNAPII ChIP and available GRO-SEQ data. However, we believe that PACER is the most abundant well-defined message based on the GRO-SEQ results and considering that we failed to clone additional transcripts further upstream using RACE (data not shown). PACER sequence contains a well conserved polyadenylation signal at position -999 (<xref ref-type="fig" rid="fig1s1">Figure 1–figure supplement 1</xref>) and a well-defined 3’ end at -1022. Nevertheless, it is possible that additional transcripts exist further upstream. These are unlikely to function in the same manner, because we did not recover RNAs mapping to that area in p50 immunoprecipitates (e.g., <xref ref-type="fig" rid="fig6">Figure 6B</xref>, and data not shown). These issues are now also discussed in the Results section under <xref ref-type="fig" rid="fig1">Figure 1</xref>.</p><p><italic>1.5) Strand-specific amplification of the plus strand should include an oligo within the COX-2 gene as a positive control</italic>.</p><p>This control is now included as <xref ref-type="fig" rid="fig1">Figure 1F</xref>.</p><p><italic>1.6)</italic> <xref ref-type="fig" rid="fig4"><italic>Figure 4J</italic></xref> <italic>shows the subcellular localization of PACER and COX-2 as determined by cellular fractionation. The authors should perform controls of transcripts present only in the nucleus and only in the cytoplasm to account for cross-contamination of the fractions</italic>.</p><p><xref ref-type="fig" rid="fig4">Figure 4J</xref> now includes control amplification of nuclear (COX-2 intron) and cytoplasmic RNA (mRNA for actin). These results clearly indicate that PACER is predominantly nuclear, as its profile closely resembles that of the intronic RNA. Cytoplasmic RNAs are found in the nuclear fractions to some extent, likely because a fraction of mature RNA remains within the spliceosome/nuclear export complexes and co-purifies with the nuclear fractions, this is clearly visible for COX-2 mRNA but also actin mRNA. Conversely, small amounts of nuclear RNAs are found in the cytoplasmic fraction. This is likely to occur during biochemical separation of fractions, where small molecules are known to passively leak out through nuclear pores. Despite this minor cross-contamination, which is evident for a nucleus-exclusive intronic RNA, we believe our data supports the conclusion that PACER is localized to the nucleus. This discussion of this issue is included in the text under <xref ref-type="fig" rid="fig4">Figure 4</xref>. In addition, recent experiments from our laboratory indicate that PACER is not associated with cytoplasmic polysomal fractions or with mono-ribosomes (D. Henderson, personal communication), further indicating that it is confined to the nucleus. This point is mentioned in the Discussion.</p><p><italic>1.7) The reviewers wonder whether PACER has been detected by recent genome wide Global Run-On deep sequencing experiments (GRO-seq) done in epithelial cells. Publicly available GRO-seq datasets that could be mined to answer this question include Galbraith MD et al, Cell. 2013 Jun 6;153(6):1327-39 (performed in HCT116 colon cancer cells) and the various papers from the Kraus team using MCF7 breast cancer cells such as Danko CG Mol Cell. 2013 Apr 25;50(2):212-22. GRO-seq results for the COX-2 (PTGS2) locus could be displayed to further</italic> illuminate the site of PACER transcription.</p><p>Antisense transcription upstream of the COX-2 transcription start site was indeed detected in HCT116 cells in the recent study [Galbraith, M.D., et al., <italic>HIF1A employs CDK8-mediator to stimulate RNAPII elongation in response to hypoxia.</italic> Cell, 2013. 153(6): p. 1327-39.]. These signals extend up ∼1100bp upstream, therefore perfectly correlating with the presence of PACER in this region. A genome browser picture incorporating these data is now included as <xref ref-type="fig" rid="fig1s4">Figure 1—figure supplement 4</xref> and it is referenced in the main text under <xref ref-type="fig" rid="fig1">Figure 1</xref>.</p><p><italic>2) The model of PACER action needs further experimental validation</italic>.</p><p><italic>2.1) Whereas the repressive role of p50 is well demonstrated, the positive role of p65 NFKB in driving COX2 expression is not demonstrated in the context of PACER action. As currently depicted in</italic> <xref ref-type="fig" rid="fig6"><italic>Figure 6G</italic></xref><italic>, it is p65 that recruits p300 to drive histone modifications and COX2 transcription. The reviewers request that the role of p65 on PACER action be demonstrated via knockdown</italic>.</p><p>As requested by the reviewers, we performed p65 knock-down experiments in U937 monocytes and analyzed its effects on COX-2 expression in the context of PACER function (<xref ref-type="fig" rid="fig6">Figure 6GH</xref>). These results clearly indicate that the role of p65 at the COX-2 promoter is confined to the induction phase, because in non-induced macrophages the knockdown had no effects upon COX-2 expression, contrary to what we observed for p50 (<xref ref-type="fig" rid="fig6">Figure 6G</xref> vs 6E). This is to be expected because, while p50 remains bound to the COX-2 promoter in non-induced conditions, p65 is retained in the cytoplasm by IκB complexes [Gilmore, T.D., <italic>Introduction to NF-kappaB: players, pathways, perspectives.</italic> Oncogene, 2006. 25(51): p. 6680-4]. Only upon activation of the NF-κB signaling pathway p65 relocates into the nucleus and activates its target genes [Gilmore, T.D., <italic>Introduction to NF-kappaB: players, pathways, perspectives.</italic> Oncogene, 2006. 25(51): p. 6680-4]. Contrary to this, the activation role of PACER on COX-2 expression is evident both in non-induced macrophages (<xref ref-type="fig" rid="fig4">Figure 4DF</xref>) and in HMECs (<xref ref-type="fig" rid="fig4">Figure 4AB</xref>). In addition, RIP experiments suggest that PACER does not contact p65 (<xref ref-type="fig" rid="fig6">Figure 6A</xref>). We therefore believe that the role of PACER is independent of p65; as PACER occludes p50, effectively inactivating it as a repressor, and allows activators (p65 in the case of LPS stimulation in macrophages, and possibly other transcription factors in the case of HMECs) to induce RNAP II recruitment and mRNA transcription.</p><p><italic>2.2) Additionally, how does PACER knockdown or CTCF knockdown affect p65 binding to the COX2 promoter? Reviewers propose that ChIP assays for p65 be performed to answer this question</italic>.</p><p>Despite repeated efforts we were not able to demonstrate binding of p65 to the COX-2 locus. We performed these ChIP experiments before the original submission of our manuscript with no success, and repeated again in response to reviewers’ comments. We are well aware that other studies have been performed where ChIP was used to demonstrate p65 binding, including genome-wide studies in mice and humans [Xing, Y., et al., <italic>Characterization of genome-wide binding of NF-kappaB in TNFalpha-stimulated HeLa cells.</italic> Gene, 2013. 526(2): p. 142-9; Lim, C.A., et al., <italic>Genome-wide mapping of RELA(p65) binding identifies E2F1 as a transcriptional activator recruited by NF-kappaB upon TLR4 activation.</italic> Mol Cell, 2007. 27(4): p. 622-35; Schreiber, J., et al., <italic>Coordinated binding of NF-kappaB family members in the response of human cells to lipopolysaccharide.</italic> Proc Natl Acad Sci U S A, 2006. 103(15): p. 5899-904; Kaikkonen, M.U., et al., <italic>Remodeling of the enhancer landscape during macrophage activation is coupled to enhancer transcription.</italic> Mol Cell, 2013. 51(3): p. 310-25]. We attempted to perform similar analyses using the antibodies that were used by other investigators, including, but not limited to, sc-372X from Santa Cruz or Ab7970 from Abcam, all without success. We consistently observed background signals at COX-2, as well as at other regions where p65 binding was also expected, including numerous “positive control” regions that exhibited particularly strong signals for p65 in previous studies (TNFa, IL-2, CCL5 and CSF5). Since our laboratory possesses extensive expertise in ChIP and these types of experiments are performed routinely with success, we were forced to conclude that currently available antibody lots do not permit us to perform p65 ChIP experiments. However, given that the role of p65 in the context of COX-2 expression has been studied extensively [Schmedtje, J.F., Jr., et al., <italic>Hypoxia induces cyclooxygenase-2 via the NF-kappaB p65 transcription factor in human vascular endothelial cells.</italic> J Biol Chem, 1997. 272(1): p. 601-8; D'Acquisto, F., et al., <italic>Involvement of NF-kappaB in the regulation of cyclooxygenase-2 protein expression in LPS-stimulated J774 macrophages.</italic> FEBS Lett, 1997. 418(1-2): p. 175-8; Shishodia, S. and B.B. Aggarwal, <italic>Cyclooxygenase (COX)-2 inhibitor celecoxib abrogates activation of cigarette smoke-induced nuclear factor (NF)-kappaB by suppressing activation of IkappaBalpha kinase in human non-small cell lung carcinoma: correlation with suppression of cyclin D1, COX-2, and matrix metalloproteinase-9.</italic> Cancer Res, 2004. 64(14): p. 5004-12; Jang, B.C., et al., <italic>Induction of cyclooxygenase-2 in macrophages by catalase: role of NF-kappaB and PI3K signaling pathways.</italic> Biochem Biophys Res Commun, 2004. 316(2): p. 398-406] and our knock-down experiments clearly indicate a p65 role in the induction of COX-2 expression, we believe that our experiments provide sufficient support for the model of PACER action.</p><p><italic>3) A better presentation and interpretation of the ChIP data is needed</italic>.</p><p><italic>3.1) CTCF ChIPs. Regarding the ChIP-on-chip experiment, the authors should show at least one negative control and one positive control locus out of the “numerous other genes” on the chip and extend the COX-2 analysis several kb up and downstream to demonstrate the specificity of the IP. There is a discrepancy between the ChIP-on-chip and ChIP-qPCR analysis for CTCF binding sites. In the ChIP-on-chip, one sees a clear enrichment of CTCF binding at least to -500bp, which is not confirmed by the qPCR. Also, the peak for CTCF binding around TSS is much more pronounced than the one around -2.5kb. Again, this is not recapitulated in the ChIP-qPCR. In addition, the claim that Rad21 binds around the TSS along with CTCF is not substantiated by the ChIP presented in</italic> <xref ref-type="fig" rid="fig2"><italic>Fig 2D</italic></xref><italic>, especially if one takes into account the IgG background that differs between different amplicons. The authors should aim to explain the possible origin of these discrepancies</italic>.</p><p>The ChIP-on-chip data representing results at three loci, COX-2, c-myc and INSIG2 are now shown in <xref ref-type="fig" rid="fig2s1">Figure 2–figure supplement 1.</xref> Each region on the array was tiled with ∼50bp probes with approximately 5x coverage, apart from repeated DNA, which was not covered. As a result, a 50kb region contains up to 250.000 probes. Each probe that shows background signal effectively serves as a ‘negative control’ for any other probe on the array. Nevertheless, as requested, we show a gene (INSIG2) where no enrichment for CTCF is present within ∼25kb upstream or downstream of its promoter. Given the tiling density, we are confident that signals shown as peaks represent true enrichment of a DNA region in the immunoprecipitates. That said, a caveat might exist that an unusually high GC content of a region (such as a CpG island around COX-2 promoter) will cause stronger hybridization resulting in unusually high signal for relatively lowly enriched DNA. This, in our opinion, might cause the apparent discrepancy between the strength of COX-2 promoter signal on the array as compared to qPCR data. For this reason, we tend to believe that the strong ĆTCF peak extending up to ∼500bp upstream of the COX-2 promoter in ChIP-on-chip data is, in reality, weaker and represents binding of CTCF at the position +65bp, as mapped by ChIP-qPCR walking, EMSA and DNA recruitment assays (<xref ref-type="fig" rid="fig1">Figures 1C</xref>, <xref ref-type="fig" rid="fig2s2 fig2s3">Figure 2–figure supplements 2-3</xref>). These considerations are now also included in the text under <xref ref-type="fig" rid="fig2">Figure 2</xref>.</p><p>Concerning Rad21, we performed ChIP-qPCR walking several times and we have consistently seen Rad21 binding to COX-2 promoter, albeit at the lower level than at the -2.5kb site. This low level is still significantly higher than any other negative site, and any IgG signal. We calculated statistical significance levels comparing signal values at the COX-2 promoter versus two other negative sites at -1.55kb and -0.85kb and found p values equal to 0.028 and 0.033, respectively. These values are now depicted in <xref ref-type="fig" rid="fig2">Figure 2D</xref>.</p><p><italic>3.2) Do the two described peaks overlap</italic> <italic>with a CTCF consensus motif?</italic></p><p>We found CTCF binding consensus motifs located both within the -2.5kb peak (GCAGCAGAAGGGGGCAGTA at -2508bp) and at the promoter (GCAGCGCCTCCTTCAGCTCCA at +65). This information is now mentioned in the text under <xref ref-type="fig" rid="fig2">Figure 2</xref>.</p><p><italic>3.3) It is stated in the text that H4K20me3 is reduced upon CTCF knockdown;</italic> <xref ref-type="fig" rid="fig3"><italic>Figure 3I</italic></xref><italic>, however, shows an increase in H4K20me3. The text reads: “This modification was reduced at both sites upon CTCF knockdown...”, whereas the data shows a clear increase of H4K20me3, although only at the distal peak at -2.5 kb. Please revise the description of this result</italic>.</p><p>The text has been revised and now reads: “A repressive mark H4K20me3 had a more localized pattern, with two discernible peaks at -2.5kb and ∼-1.2kb. This modification was increased at -2.5kb upon CTCF knockdown (<xref ref-type="fig" rid="fig3">Figure 3I</xref>).”</p><p><italic>3.4) Chromatin modifications are evaluated upon PACER knockdown. Although the manuscript claims “no significant changes in H3K4me2”, there are clear differences in Figure S5 (not Figure S4 as stated in the text), especially at the important promoter and -2.5 kb sites. Please revise the description of this result</italic>.</p><p>We have now revised the text to mention those changes in H3K4me2. The text now reads “ChIP-qPCR experiments showed no significant changes in H3K4 trimethylation or H2AZ association (data not shown) and a modest effect on H3K4 dimethylation (<xref ref-type="fig" rid="fig5s2">Figure 5–figure supplement 2</xref>).” We have not focused our attention on these findings because in the absence of H3K4me3 changes, we found it difficult to interpret H3K4me2 differences alone. Moreover, H3K4me2 is affected at the promoter in monocytes and macrophages, while at -2.5kb the difference is only evident after LPS induction. Such distinct effects clearly need further investigation, but are very unlikely to participate in PACER-modulated regulation of COX-2 expression.</p><p><italic>3.5) The conclusion of PACER directly recruiting RNAP II is not supported by the data shown. Instead, the changes in chromatin modification might indirectly influence RNAP II association. The text should be revised to acknowledge this</italic>.</p><p>We fully agree with the notion that the effects in RNAP II recruitment are in all likelihood indirect, possibly mediated through p300-induced changes in histone acetylation. We have now revised the text to reflect this point more clearly. The text now reads: “To explore the possibility that PACER could affect (directly or indirectly) the assembly of RNA Polymerase II preinitiation complexes and/or later steps in transcriptional activation […] These results demonstrate that PACER influences early steps in transcription initiation at the stage of formation of RNAP II preinitiation complexes, likely through p300-induced changes in promoter-associated histone acetylation.” Also, in the Discussion section “Ultimately, the function of PACER lncRNA is to facilitate assembly of RNA Polymerase II pre-initiation complexes. Upon lncRNA knockdown, the levels of elongating RNAP II and initiating S5P-modified RNAP II are significantly lower, thus affecting the ability of the transcription machinery to efficiently initiate transcription of the COX-2 gene. While these effects are indirect, they nevertheless demonstrate that PACER belongs to a novel, unanticipated class of bona-fide transcriptional regulators.”</p></body></sub-article></article> |